From lists at ruby-forum.com Tue Jul 1 02:58:45 2008 From: lists at ruby-forum.com (Todd Bb) Date: Tue, 1 Jul 2008 08:58:45 +0200 Subject: [Ironruby-core] using gems In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD3F402CE@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD3F402CE@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: John, Thanks. Can I ask then how do you do a 'require'? In other words, how are you resolving to that gem? Can you point me in the right direction. Thanks. Todd -- Posted via http://www.ruby-forum.com/. From bacondarwin at googlemail.com Tue Jul 1 09:00:50 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Tue, 1 Jul 2008 14:00:50 +0100 Subject: [Ironruby-core] SVN r118 and RubySpec integration In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD3F40782@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD3F40782@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <002801c8db7a$7ebb3430$7c319c90$@com> Next thing: I changed the place that the cloned GIT repos were held and on compile I get these warnings: D:\dev\ruby\ironruby\head>rake compile (in D:/dev/ruby/ironruby/head) C:\Documents and Settings\pete/.irconfig.rb:19: warning: already initialized constant TAGS C:\Documents and Settings\pete/.irconfig.rb:20: warning: already initialized constant RUBYSPEC C:\Documents and Settings\pete/.irconfig.rb:21: warning: already initialized constant MSPEC I appreciate that this should not be a problem but could be tidied up? Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: Tuesday,01 July 01, 2008 00:59 To: ironruby-core at rubyforge.org Subject: [Ironruby-core] SVN r118 and RubySpec integration I just pushed SVN r118 out which contains RubySpec integration. There are a few things that you'll need to do: 1. (Optional) get a GitHub account (http://github.com) 2. (Optional) install GIT: a. Goto the msysgit project and download the .exe based installer: http://code.google.com/p/msysgit/ b. Make sure in the GIT setup that you click on the radio button labeled "Run Git from the Windows Command Prompt" c. Restart your command prompt 3. Navigate to my page: http://github.com/jflam to see the three GIT projects that you'll need to clone: ironruby-tags, mspec, and rubyspec. mspec and rubyspec are forks of the main projects so that we can add IronRuby specific tests and potential changes to mspec. The RubySpec committers can pull from these repositories to incorporate our changes. ironruby-tags is a project that contains the list of rubyspec specs that we currently exclude to get a clean regression baseline. 4. If you have GIT installed, create a directory to hold these projects. I use %USERPROFILE%\dev: a. in ~\dev\ironruby-tags run: git clone git://github.com/jflam/ironruby-tags.git b. in ~\dev\mspec run: git clone git://github.com/jflam/mspec.git c. in ~\dev\rubyspec run: git clone git://github.com/jflam/rubyspec.git 5. If you don't have GIT installed (and why don't you??? J) you can download the tarball using the download button in each project. Once you have those projects in place, you'll need to grab r118 from RubyForge. There's a file in \trunk called runfirst.cmd. This will enable skip verification for projects signed with our public key, and it will create a default configuration file at %USERPROFILE%\.irconfig.rb. If you followed my instructions and created %USERPROFILE%\dev\[ironruby-tags,mspec,rubyspec] you're good to go. If you didn't, you'll need to edit the paths in .irconfig.rb You can test your setup by running the new regression test baseline rake task: rake regression This should report 0 failures and 0 errors. This is the first test of this stuff outside of my machine, so please kick the tires! Thanks, -John -------------- next part -------------- An HTML attachment was scrubbed... URL: From bacondarwin at googlemail.com Tue Jul 1 09:00:50 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Tue, 1 Jul 2008 14:00:50 +0100 Subject: [Ironruby-core] SVN r118 and RubySpec integration In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD3F40782@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD3F40782@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <002d01c8db7a$9a8f1320$cfad3960$@com> First thing to note: D:\dev\ruby\ironruby\head>runfirst.cmd D:\dev\ruby\ironruby\head>sn -Vr *,31bf3856ad364e35 Microsoft (R) .NET Framework Strong Name Utility Version 3.5.21022.8 Copyright (c) Microsoft Corporation. All rights reserved. Verification entry added for assembly '*,31bf3856ad364e35' D:\dev\ruby\ironruby\head>copy default.irconfig.rb C:\Documents and Settings\pete\.irconfig.rb The syntax of the command is incorrect. The line that copies the .irconfig.rb file in runfirst.cmd needs quotes around the "to" parameter of "copy": copy default.irconfig.rb "%USERPROFILE%\.irconfig.rb" Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: Tuesday,01 July 01, 2008 00:59 To: ironruby-core at rubyforge.org Subject: [Ironruby-core] SVN r118 and RubySpec integration I just pushed SVN r118 out which contains RubySpec integration. There are a few things that you'll need to do: 1. (Optional) get a GitHub account (http://github.com) 2. (Optional) install GIT: a. Goto the msysgit project and download the .exe based installer: http://code.google.com/p/msysgit/ b. Make sure in the GIT setup that you click on the radio button labeled "Run Git from the Windows Command Prompt" c. Restart your command prompt 3. Navigate to my page: http://github.com/jflam to see the three GIT projects that you'll need to clone: ironruby-tags, mspec, and rubyspec. mspec and rubyspec are forks of the main projects so that we can add IronRuby specific tests and potential changes to mspec. The RubySpec committers can pull from these repositories to incorporate our changes. ironruby-tags is a project that contains the list of rubyspec specs that we currently exclude to get a clean regression baseline. 4. If you have GIT installed, create a directory to hold these projects. I use %USERPROFILE%\dev: a. in ~\dev\ironruby-tags run: git clone git://github.com/jflam/ironruby-tags.git b. in ~\dev\mspec run: git clone git://github.com/jflam/mspec.git c. in ~\dev\rubyspec run: git clone git://github.com/jflam/rubyspec.git 5. If you don't have GIT installed (and why don't you??? J) you can download the tarball using the download button in each project. Once you have those projects in place, you'll need to grab r118 from RubyForge. There's a file in \trunk called runfirst.cmd. This will enable skip verification for projects signed with our public key, and it will create a default configuration file at %USERPROFILE%\.irconfig.rb. If you followed my instructions and created %USERPROFILE%\dev\[ironruby-tags,mspec,rubyspec] you're good to go. If you didn't, you'll need to edit the paths in .irconfig.rb You can test your setup by running the new regression test baseline rake task: rake regression This should report 0 failures and 0 errors. This is the first test of this stuff outside of my machine, so please kick the tires! Thanks, -John -------------- next part -------------- An HTML attachment was scrubbed... URL: From bacondarwin at googlemail.com Tue Jul 1 09:15:38 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Tue, 1 Jul 2008 14:15:38 +0100 Subject: [Ironruby-core] SVN r118 and RubySpec integration In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD3F40782@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD3F40782@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <003801c8db7c$8fd57da0$af8078e0$@com> And finally when I run rake regression I get a suspiciously quick result: D:\dev\ruby\ironruby\head>rake regression (in D:/dev/ruby/ironruby/head) C:\Documents and Settings\pete/.irconfig.rb:19: warning: already initialized constant TAGS C:\Documents and Settings\pete/.irconfig.rb:20: warning: already initialized constant RUBYSPEC C:\Documents and Settings\pete/.irconfig.rb:21: warning: already initialized constant MSPEC Summary: 0 types, 0 files, 0 examples, 0 expectations, 0 failures, 0 errors Nothing seems to be getting run? If I run rake spec then it does run the tests: D:\dev\ruby\ironruby\head>rake spec (in D:/dev/ruby/ironruby/head) C:\Documents and Settings\pete/.irconfig.rb:19: warning: already initialized constant TAGS C:\Documents and Settings\pete/.irconfig.rb:20: warning: already initialized constant RUBYSPEC C:\Documents and Settings\pete/.irconfig.rb:21: warning: already initialized constant MSPEC 2199 examples, 0 failures 0 Is this what is expected? Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: Tuesday,01 July 01, 2008 00:59 To: ironruby-core at rubyforge.org Subject: [Ironruby-core] SVN r118 and RubySpec integration I just pushed SVN r118 out which contains RubySpec integration. There are a few things that you'll need to do: 1. (Optional) get a GitHub account (http://github.com) 2. (Optional) install GIT: a. Goto the msysgit project and download the .exe based installer: http://code.google.com/p/msysgit/ b. Make sure in the GIT setup that you click on the radio button labeled "Run Git from the Windows Command Prompt" c. Restart your command prompt 3. Navigate to my page: http://github.com/jflam to see the three GIT projects that you'll need to clone: ironruby-tags, mspec, and rubyspec. mspec and rubyspec are forks of the main projects so that we can add IronRuby specific tests and potential changes to mspec. The RubySpec committers can pull from these repositories to incorporate our changes. ironruby-tags is a project that contains the list of rubyspec specs that we currently exclude to get a clean regression baseline. 4. If you have GIT installed, create a directory to hold these projects. I use %USERPROFILE%\dev: a. in ~\dev\ironruby-tags run: git clone git://github.com/jflam/ironruby-tags.git b. in ~\dev\mspec run: git clone git://github.com/jflam/mspec.git c. in ~\dev\rubyspec run: git clone git://github.com/jflam/rubyspec.git 5. If you don't have GIT installed (and why don't you??? J) you can download the tarball using the download button in each project. Once you have those projects in place, you'll need to grab r118 from RubyForge. There's a file in \trunk called runfirst.cmd. This will enable skip verification for projects signed with our public key, and it will create a default configuration file at %USERPROFILE%\.irconfig.rb. If you followed my instructions and created %USERPROFILE%\dev\[ironruby-tags,mspec,rubyspec] you're good to go. If you didn't, you'll need to edit the paths in .irconfig.rb You can test your setup by running the new regression test baseline rake task: rake regression This should report 0 failures and 0 errors. This is the first test of this stuff outside of my machine, so please kick the tires! Thanks, -John -------------- next part -------------- An HTML attachment was scrubbed... URL: From jflam at microsoft.com Tue Jul 1 09:52:07 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Tue, 1 Jul 2008 06:52:07 -0700 Subject: [Ironruby-core] SVN r118 and RubySpec integration In-Reply-To: <003801c8db7c$8fd57da0$af8078e0$@com> References: <372109E149E8084D8E6C7D9CFD82E0632FD3F40782@NA-EXMSG-C115.redmond.corp.microsoft.com> <003801c8db7c$8fd57da0$af8078e0$@com> Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD3F40885@NA-EXMSG-C115.redmond.corp.microsoft.com> Hmm ... looks like an older version made it out to r118. Let me investigate. Thanks, -John > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin > Sent: Tuesday, July 01, 2008 6:16 AM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > And finally when I run rake regression I get a suspiciously quick > result: > > > > D:\dev\ruby\ironruby\head>rake regression > > (in D:/dev/ruby/ironruby/head) > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > initialized constant TAGS > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > initialized constant RUBYSPEC > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > initialized constant MSPEC > > > > Summary: > > 0 types, 0 files, 0 examples, 0 expectations, 0 failures, 0 errors > > > > Nothing seems to be getting run? If I run rake spec then it does run > the tests: > > > > D:\dev\ruby\ironruby\head>rake spec > > (in D:/dev/ruby/ironruby/head) > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > initialized constant TAGS > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > initialized constant RUBYSPEC > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > initialized constant MSPEC > > 2199 examples, 0 failures > > 0 > > > > Is this what is expected? > > > > Pete > > > > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam > (IRONRUBY) > Sent: Tuesday,01 July 01, 2008 00:59 > To: ironruby-core at rubyforge.org > Subject: [Ironruby-core] SVN r118 and RubySpec integration > > > > I just pushed SVN r118 out which contains RubySpec integration. There > are a few things that you'll need to do: > > > > 1. (Optional) get a GitHub account (http://github.com) > > 2. (Optional) install GIT: > > a. Goto the msysgit project and download the .exe based > installer: > http://code.google.com/p/msysgit/ > > b. Make sure in the GIT setup that you click on the radio button > labeled "Run Git from the Windows Command Prompt" > > c. Restart your command prompt > > 3. Navigate to my page: http://github.com/jflam to see the three > GIT projects that you'll need to clone: ironruby-tags, mspec, and > rubyspec. mspec and rubyspec are forks of the main projects so that we > can add IronRuby specific tests and potential changes to mspec. The > RubySpec committers can pull from these repositories to incorporate > our changes. ironruby-tags is a project that contains the list of > rubyspec specs that we currently exclude to get a clean regression baseline. > > 4. If you have GIT installed, create a directory to hold these > projects. I use %USERPROFILE%\dev: > > a. in ~\dev\ironruby-tags run: git clone > git://github.com/jflam/ironruby-tags.git > > b. in ~\dev\mspec run: git clone git://github.com/jflam/mspec.git > > c. in ~\dev\rubyspec run: git clone > git://github.com/jflam/rubyspec.git > > 5. If you don't have GIT installed (and why don't you??? J) you > can download the tarball using the download button in each project. > > > > Once you have those projects in place, you'll need to grab r118 from > RubyForge. There's a file in \trunk called runfirst.cmd. This will > enable skip verification for projects signed with our public key, and > it will create a default configuration file at > %USERPROFILE%\.irconfig.rb. > > > > If you followed my instructions and created > %USERPROFILE%\dev\[ironruby-tags,mspec,rubyspec] you're good to go. If > you didn't, you'll need to edit the paths in .irconfig.rb > > > > You can test your setup by running the new regression test baseline > rake > task: > > > > rake regression > > > > This should report 0 failures and 0 errors. > > > > This is the first test of this stuff outside of my machine, so please > kick the tires! > > > > Thanks, > > -John > > From jflam at microsoft.com Tue Jul 1 10:33:09 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Tue, 1 Jul 2008 07:33:09 -0700 Subject: [Ironruby-core] SVN r118 and RubySpec integration In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD3F40885@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD3F40782@NA-EXMSG-C115.redmond.corp.microsoft.com> <003801c8db7c$8fd57da0$af8078e0$@com> <372109E149E8084D8E6C7D9CFD82E0632FD3F40885@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD3F408C5@NA-EXMSG-C115.redmond.corp.microsoft.com> I just looked at the files on SVN and they're the same as mine - I'm attaching default.irconfig.rb and context.rb - can you diff to see if they're the same as what's on your machine? I've fixed the quoting problem on this end. Thanks, -John > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) > Sent: Tuesday, July 01, 2008 6:52 AM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > Hmm ... looks like an older version made it out to r118. Let me > investigate. > > Thanks, > -John > > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin > > Sent: Tuesday, July 01, 2008 6:16 AM > > To: ironruby-core at rubyforge.org > > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > > > And finally when I run rake regression I get a suspiciously quick > > result: > > > > > > > > D:\dev\ruby\ironruby\head>rake regression > > > > (in D:/dev/ruby/ironruby/head) > > > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > > initialized constant TAGS > > > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > > initialized constant RUBYSPEC > > > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > > initialized constant MSPEC > > > > > > > > Summary: > > > > 0 types, 0 files, 0 examples, 0 expectations, 0 failures, 0 errors > > > > > > > > Nothing seems to be getting run? If I run rake spec then it does run > > the tests: > > > > > > > > D:\dev\ruby\ironruby\head>rake spec > > > > (in D:/dev/ruby/ironruby/head) > > > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > > initialized constant TAGS > > > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > > initialized constant RUBYSPEC > > > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > > initialized constant MSPEC > > > > 2199 examples, 0 failures > > > > 0 > > > > > > > > Is this what is expected? > > > > > > > > Pete > > > > > > > > From: ironruby-core-bounces at rubyforge.org > > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam > > (IRONRUBY) > > Sent: Tuesday,01 July 01, 2008 00:59 > > To: ironruby-core at rubyforge.org > > Subject: [Ironruby-core] SVN r118 and RubySpec integration > > > > > > > > I just pushed SVN r118 out which contains RubySpec integration. There > > are a few things that you'll need to do: > > > > > > > > 1. (Optional) get a GitHub account (http://github.com) > > > > 2. (Optional) install GIT: > > > > a. Goto the msysgit project and download the .exe based > > installer: > > http://code.google.com/p/msysgit/ > > > > b. Make sure in the GIT setup that you click on the radio button > > labeled "Run Git from the Windows Command Prompt" > > > > c. Restart your command prompt > > > > 3. Navigate to my page: http://github.com/jflam to see the > three > > GIT projects that you'll need to clone: ironruby-tags, mspec, and > > rubyspec. mspec and rubyspec are forks of the main projects so that > we > > can add IronRuby specific tests and potential changes to mspec. The > > RubySpec committers can pull from these repositories to incorporate > > our changes. ironruby-tags is a project that contains the list of > > rubyspec specs that we currently exclude to get a clean regression > baseline. > > > > 4. If you have GIT installed, create a directory to hold these > > projects. I use %USERPROFILE%\dev: > > > > a. in ~\dev\ironruby-tags run: git clone > > git://github.com/jflam/ironruby-tags.git > > > > b. in ~\dev\mspec run: git clone > git://github.com/jflam/mspec.git > > > > c. in ~\dev\rubyspec run: git clone > > git://github.com/jflam/rubyspec.git > > > > 5. If you don't have GIT installed (and why don't you??? J) you > > can download the tarball using the download button in each project. > > > > > > > > Once you have those projects in place, you'll need to grab r118 from > > RubyForge. There's a file in \trunk called runfirst.cmd. This will > > enable skip verification for projects signed with our public key, and > > it will create a default configuration file at > > %USERPROFILE%\.irconfig.rb. > > > > > > > > If you followed my instructions and created > > %USERPROFILE%\dev\[ironruby-tags,mspec,rubyspec] you're good to go. > If > > you didn't, you'll need to edit the paths in .irconfig.rb > > > > > > > > You can test your setup by running the new regression test baseline > > rake > > task: > > > > > > > > rake regression > > > > > > > > This should report 0 failures and 0 errors. > > > > > > > > This is the first test of this stuff outside of my machine, so please > > kick the tires! > > > > > > > > Thanks, > > > > -John > > > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core -------------- next part -------------- A non-text attachment was scrubbed... Name: default.irconfig.rb Type: application/octet-stream Size: 659 bytes Desc: default.irconfig.rb URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: context.rb Type: application/octet-stream Size: 25866 bytes Desc: context.rb URL: From curth at microsoft.com Tue Jul 1 11:51:23 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Tue, 1 Jul 2008 08:51:23 -0700 Subject: [Ironruby-core] Code Review: StringElementReferenceAndSlice In-Reply-To: References: Message-ID: Looks good. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Dave Remy Sent: Montag, 30. Juni 2008 18:31 To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: StringElementReferenceAndSlice tfpt review "/shelveset:StringElementReferenceAndSlice;REDMOND\dremy" Comment : Changes for [] (alias: slice) to run clean. One small fix to not flow taint. Also, fix a small bug that was using a deprecated implicit conversion between string and MutableString. From bacondarwin at googlemail.com Tue Jul 1 11:51:45 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Tue, 1 Jul 2008 16:51:45 +0100 Subject: [Ironruby-core] SVN r118 and RubySpec integration In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD3F408C5@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD3F40782@NA-EXMSG-C115.redmond.corp.microsoft.com> <003801c8db7c$8fd57da0$af8078e0$@com> <372109E149E8084D8E6C7D9CFD82E0632FD3F40885@NA-EXMSG-C115.redmond.corp.microsoft.com> <372109E149E8084D8E6C7D9CFD82E0632FD3F408C5@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <004e01c8db92$5f0e9b00$1d2bd100$@com> Both files are the same and my entire local folder is identical to the SVN output. Pete -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: Tuesday,01 July 01, 2008 15:33 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration I just looked at the files on SVN and they're the same as mine - I'm attaching default.irconfig.rb and context.rb - can you diff to see if they're the same as what's on your machine? I've fixed the quoting problem on this end. Thanks, -John > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) > Sent: Tuesday, July 01, 2008 6:52 AM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > Hmm ... looks like an older version made it out to r118. Let me > investigate. > > Thanks, > -John > > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin > > Sent: Tuesday, July 01, 2008 6:16 AM > > To: ironruby-core at rubyforge.org > > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > > > And finally when I run rake regression I get a suspiciously quick > > result: > > > > > > > > D:\dev\ruby\ironruby\head>rake regression > > > > (in D:/dev/ruby/ironruby/head) > > > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > > initialized constant TAGS > > > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > > initialized constant RUBYSPEC > > > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > > initialized constant MSPEC > > > > > > > > Summary: > > > > 0 types, 0 files, 0 examples, 0 expectations, 0 failures, 0 errors > > > > > > > > Nothing seems to be getting run? If I run rake spec then it does > > run the tests: > > > > > > > > D:\dev\ruby\ironruby\head>rake spec > > > > (in D:/dev/ruby/ironruby/head) > > > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > > initialized constant TAGS > > > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > > initialized constant RUBYSPEC > > > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > > initialized constant MSPEC > > > > 2199 examples, 0 failures > > > > 0 > > > > > > > > Is this what is expected? > > > > > > > > Pete > > > > > > > > From: ironruby-core-bounces at rubyforge.org > > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam > > (IRONRUBY) > > Sent: Tuesday,01 July 01, 2008 00:59 > > To: ironruby-core at rubyforge.org > > Subject: [Ironruby-core] SVN r118 and RubySpec integration > > > > > > > > I just pushed SVN r118 out which contains RubySpec integration. > > There are a few things that you'll need to do: > > > > > > > > 1. (Optional) get a GitHub account (http://github.com) > > > > 2. (Optional) install GIT: > > > > a. Goto the msysgit project and download the .exe based > > installer: > > http://code.google.com/p/msysgit/ > > > > b. Make sure in the GIT setup that you click on the radio button > > labeled "Run Git from the Windows Command Prompt" > > > > c. Restart your command prompt > > > > 3. Navigate to my page: http://github.com/jflam to see the > three > > GIT projects that you'll need to clone: ironruby-tags, mspec, and > > rubyspec. mspec and rubyspec are forks of the main projects so that > we > > can add IronRuby specific tests and potential changes to mspec. The > > RubySpec committers can pull from these repositories to incorporate > > our changes. ironruby-tags is a project that contains the list of > > rubyspec specs that we currently exclude to get a clean regression > baseline. > > > > 4. If you have GIT installed, create a directory to hold these > > projects. I use %USERPROFILE%\dev: > > > > a. in ~\dev\ironruby-tags run: git clone > > git://github.com/jflam/ironruby-tags.git > > > > b. in ~\dev\mspec run: git clone > git://github.com/jflam/mspec.git > > > > c. in ~\dev\rubyspec run: git clone > > git://github.com/jflam/rubyspec.git > > > > 5. If you don't have GIT installed (and why don't you??? J) you > > can download the tarball using the download button in each project. > > > > > > > > Once you have those projects in place, you'll need to grab r118 from > > RubyForge. There's a file in \trunk called runfirst.cmd. This will > > enable skip verification for projects signed with our public key, > > and it will create a default configuration file at > > %USERPROFILE%\.irconfig.rb. > > > > > > > > If you followed my instructions and created > > %USERPROFILE%\dev\[ironruby-tags,mspec,rubyspec] you're good to go. > If > > you didn't, you'll need to edit the paths in .irconfig.rb > > > > > > > > You can test your setup by running the new regression test baseline > > rake > > task: > > > > > > > > rake regression > > > > > > > > This should report 0 failures and 0 errors. > > > > > > > > This is the first test of this stuff outside of my machine, so > > please kick the tires! > > > > > > > > Thanks, > > > > -John > > > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core From bacondarwin at googlemail.com Tue Jul 1 12:29:55 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Tue, 1 Jul 2008 17:29:55 +0100 Subject: [Ironruby-core] SVN r118 and RubySpec integration In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD3F408C5@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD3F40782@NA-EXMSG-C115.redmond.corp.microsoft.com> <003801c8db7c$8fd57da0$af8078e0$@com> <372109E149E8084D8E6C7D9CFD82E0632FD3F40885@NA-EXMSG-C115.redmond.corp.microsoft.com> <372109E149E8084D8E6C7D9CFD82E0632FD3F408C5@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <000001c8db97$b4924a90$1db6dfb0$@com> Oops. Sorry my fault. My irconfig.rb was referring to the wrong folders. (I had c: rather than d:). The line 668 in context.rb Dir["#{UserEnvironment::RUBYSPEC}/1.8/core/*"].each do |path| was never picking up the tests. Pete -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: Tuesday,01 July 01, 2008 15:33 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration I just looked at the files on SVN and they're the same as mine - I'm attaching default.irconfig.rb and context.rb - can you diff to see if they're the same as what's on your machine? I've fixed the quoting problem on this end. Thanks, -John > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) > Sent: Tuesday, July 01, 2008 6:52 AM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > Hmm ... looks like an older version made it out to r118. Let me > investigate. > > Thanks, > -John > > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin > > Sent: Tuesday, July 01, 2008 6:16 AM > > To: ironruby-core at rubyforge.org > > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > > > And finally when I run rake regression I get a suspiciously quick > > result: > > > > > > > > D:\dev\ruby\ironruby\head>rake regression > > > > (in D:/dev/ruby/ironruby/head) > > > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > > initialized constant TAGS > > > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > > initialized constant RUBYSPEC > > > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > > initialized constant MSPEC > > > > > > > > Summary: > > > > 0 types, 0 files, 0 examples, 0 expectations, 0 failures, 0 errors > > > > > > > > Nothing seems to be getting run? If I run rake spec then it does > > run the tests: > > > > > > > > D:\dev\ruby\ironruby\head>rake spec > > > > (in D:/dev/ruby/ironruby/head) > > > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > > initialized constant TAGS > > > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > > initialized constant RUBYSPEC > > > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > > initialized constant MSPEC > > > > 2199 examples, 0 failures > > > > 0 > > > > > > > > Is this what is expected? > > > > > > > > Pete > > > > > > > > From: ironruby-core-bounces at rubyforge.org > > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam > > (IRONRUBY) > > Sent: Tuesday,01 July 01, 2008 00:59 > > To: ironruby-core at rubyforge.org > > Subject: [Ironruby-core] SVN r118 and RubySpec integration > > > > > > > > I just pushed SVN r118 out which contains RubySpec integration. > > There are a few things that you'll need to do: > > > > > > > > 1. (Optional) get a GitHub account (http://github.com) > > > > 2. (Optional) install GIT: > > > > a. Goto the msysgit project and download the .exe based > > installer: > > http://code.google.com/p/msysgit/ > > > > b. Make sure in the GIT setup that you click on the radio button > > labeled "Run Git from the Windows Command Prompt" > > > > c. Restart your command prompt > > > > 3. Navigate to my page: http://github.com/jflam to see the > three > > GIT projects that you'll need to clone: ironruby-tags, mspec, and > > rubyspec. mspec and rubyspec are forks of the main projects so that > we > > can add IronRuby specific tests and potential changes to mspec. The > > RubySpec committers can pull from these repositories to incorporate > > our changes. ironruby-tags is a project that contains the list of > > rubyspec specs that we currently exclude to get a clean regression > baseline. > > > > 4. If you have GIT installed, create a directory to hold these > > projects. I use %USERPROFILE%\dev: > > > > a. in ~\dev\ironruby-tags run: git clone > > git://github.com/jflam/ironruby-tags.git > > > > b. in ~\dev\mspec run: git clone > git://github.com/jflam/mspec.git > > > > c. in ~\dev\rubyspec run: git clone > > git://github.com/jflam/rubyspec.git > > > > 5. If you don't have GIT installed (and why don't you??? J) you > > can download the tarball using the download button in each project. > > > > > > > > Once you have those projects in place, you'll need to grab r118 from > > RubyForge. There's a file in \trunk called runfirst.cmd. This will > > enable skip verification for projects signed with our public key, > > and it will create a default configuration file at > > %USERPROFILE%\.irconfig.rb. > > > > > > > > If you followed my instructions and created > > %USERPROFILE%\dev\[ironruby-tags,mspec,rubyspec] you're good to go. > If > > you didn't, you'll need to edit the paths in .irconfig.rb > > > > > > > > You can test your setup by running the new regression test baseline > > rake > > task: > > > > > > > > rake regression > > > > > > > > This should report 0 failures and 0 errors. > > > > > > > > This is the first test of this stuff outside of my machine, so > > please kick the tires! > > > > > > > > Thanks, > > > > -John > > > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core From jflam at microsoft.com Tue Jul 1 12:53:13 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Tue, 1 Jul 2008 09:53:13 -0700 Subject: [Ironruby-core] SVN r118 and RubySpec integration In-Reply-To: <000001c8db97$b4924a90$1db6dfb0$@com> References: <372109E149E8084D8E6C7D9CFD82E0632FD3F40782@NA-EXMSG-C115.redmond.corp.microsoft.com> <003801c8db7c$8fd57da0$af8078e0$@com> <372109E149E8084D8E6C7D9CFD82E0632FD3F40885@NA-EXMSG-C115.redmond.corp.microsoft.com> <372109E149E8084D8E6C7D9CFD82E0632FD3F408C5@NA-EXMSG-C115.redmond.corp.microsoft.com> <000001c8db97$b4924a90$1db6dfb0$@com> Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD3F409CC@NA-EXMSG-C115.redmond.corp.microsoft.com> Can you try out r119 that I just pushed out? It should have a test that catches the wrong folder problem that you ran into. Thanks, -John -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Tuesday, July 01, 2008 9:30 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration Oops. Sorry my fault. My irconfig.rb was referring to the wrong folders. (I had c: rather than d:). The line 668 in context.rb Dir["#{UserEnvironment::RUBYSPEC}/1.8/core/*"].each do |path| was never picking up the tests. Pete -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: Tuesday,01 July 01, 2008 15:33 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration I just looked at the files on SVN and they're the same as mine - I'm attaching default.irconfig.rb and context.rb - can you diff to see if they're the same as what's on your machine? I've fixed the quoting problem on this end. Thanks, -John > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) > Sent: Tuesday, July 01, 2008 6:52 AM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > Hmm ... looks like an older version made it out to r118. Let me > investigate. > > Thanks, > -John > > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin > > Sent: Tuesday, July 01, 2008 6:16 AM > > To: ironruby-core at rubyforge.org > > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > > > And finally when I run rake regression I get a suspiciously quick > > result: > > > > > > > > D:\dev\ruby\ironruby\head>rake regression > > > > (in D:/dev/ruby/ironruby/head) > > > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > > initialized constant TAGS > > > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > > initialized constant RUBYSPEC > > > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > > initialized constant MSPEC > > > > > > > > Summary: > > > > 0 types, 0 files, 0 examples, 0 expectations, 0 failures, 0 errors > > > > > > > > Nothing seems to be getting run? If I run rake spec then it does > > run the tests: > > > > > > > > D:\dev\ruby\ironruby\head>rake spec > > > > (in D:/dev/ruby/ironruby/head) > > > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > > initialized constant TAGS > > > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > > initialized constant RUBYSPEC > > > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > > initialized constant MSPEC > > > > 2199 examples, 0 failures > > > > 0 > > > > > > > > Is this what is expected? > > > > > > > > Pete > > > > > > > > From: ironruby-core-bounces at rubyforge.org > > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam > > (IRONRUBY) > > Sent: Tuesday,01 July 01, 2008 00:59 > > To: ironruby-core at rubyforge.org > > Subject: [Ironruby-core] SVN r118 and RubySpec integration > > > > > > > > I just pushed SVN r118 out which contains RubySpec integration. > > There are a few things that you'll need to do: > > > > > > > > 1. (Optional) get a GitHub account (http://github.com) > > > > 2. (Optional) install GIT: > > > > a. Goto the msysgit project and download the .exe based > > installer: > > http://code.google.com/p/msysgit/ > > > > b. Make sure in the GIT setup that you click on the radio button > > labeled "Run Git from the Windows Command Prompt" > > > > c. Restart your command prompt > > > > 3. Navigate to my page: http://github.com/jflam to see the > three > > GIT projects that you'll need to clone: ironruby-tags, mspec, and > > rubyspec. mspec and rubyspec are forks of the main projects so that > we > > can add IronRuby specific tests and potential changes to mspec. The > > RubySpec committers can pull from these repositories to incorporate > > our changes. ironruby-tags is a project that contains the list of > > rubyspec specs that we currently exclude to get a clean regression > baseline. > > > > 4. If you have GIT installed, create a directory to hold these > > projects. I use %USERPROFILE%\dev: > > > > a. in ~\dev\ironruby-tags run: git clone > > git://github.com/jflam/ironruby-tags.git > > > > b. in ~\dev\mspec run: git clone > git://github.com/jflam/mspec.git > > > > c. in ~\dev\rubyspec run: git clone > > git://github.com/jflam/rubyspec.git > > > > 5. If you don't have GIT installed (and why don't you??? J) you > > can download the tarball using the download button in each project. > > > > > > > > Once you have those projects in place, you'll need to grab r118 from > > RubyForge. There's a file in \trunk called runfirst.cmd. This will > > enable skip verification for projects signed with our public key, > > and it will create a default configuration file at > > %USERPROFILE%\.irconfig.rb. > > > > > > > > If you followed my instructions and created > > %USERPROFILE%\dev\[ironruby-tags,mspec,rubyspec] you're good to go. > If > > you didn't, you'll need to edit the paths in .irconfig.rb > > > > > > > > You can test your setup by running the new regression test baseline > > rake > > task: > > > > > > > > rake regression > > > > > > > > This should report 0 failures and 0 errors. > > > > > > > > This is the first test of this stuff outside of my machine, so > > please kick the tires! > > > > > > > > Thanks, > > > > -John > > > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From jirapong.nanta at gmail.com Tue Jul 1 13:45:37 2008 From: jirapong.nanta at gmail.com (Jirapong Nanta) Date: Wed, 2 Jul 2008 00:45:37 +0700 Subject: [Ironruby-core] SVN r118 and RubySpec integration In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD3F409CC@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD3F40782@NA-EXMSG-C115.redmond.corp.microsoft.com> <003801c8db7c$8fd57da0$af8078e0$@com> <372109E149E8084D8E6C7D9CFD82E0632FD3F40885@NA-EXMSG-C115.redmond.corp.microsoft.com> <372109E149E8084D8E6C7D9CFD82E0632FD3F408C5@NA-EXMSG-C115.redmond.corp.microsoft.com> <000001c8db97$b4924a90$1db6dfb0$@com> <372109E149E8084D8E6C7D9CFD82E0632FD3F409CC@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: Hi John, It was fail to run... attached file is the fixed version. Thank you, -Jirapong On Tue, Jul 1, 2008 at 11:53 PM, John Lam (IRONRUBY) wrote: > Can you try out r119 that I just pushed out? It should have a test that > catches the wrong folder problem that you ran into. > > Thanks, > -John > > > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto: > ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin > Sent: Tuesday, July 01, 2008 9:30 AM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > Oops. Sorry my fault. > My irconfig.rb was referring to the wrong folders. (I had c: rather than > d:). > The line 668 in context.rb > Dir["#{UserEnvironment::RUBYSPEC}/1.8/core/*"].each do |path| > was never picking up the tests. > Pete > > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam > (IRONRUBY) > Sent: Tuesday,01 July 01, 2008 15:33 > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > I just looked at the files on SVN and they're the same as mine - I'm > attaching default.irconfig.rb and context.rb - can you diff to see if > they're the same as what's on your machine? > > I've fixed the quoting problem on this end. > > Thanks, > -John > > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) > > Sent: Tuesday, July 01, 2008 6:52 AM > > To: ironruby-core at rubyforge.org > > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > > > Hmm ... looks like an older version made it out to r118. Let me > > investigate. > > > > Thanks, > > -John > > > > > > > -----Original Message----- > > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > > bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin > > > Sent: Tuesday, July 01, 2008 6:16 AM > > > To: ironruby-core at rubyforge.org > > > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > > > > > And finally when I run rake regression I get a suspiciously quick > > > result: > > > > > > > > > > > > D:\dev\ruby\ironruby\head>rake regression > > > > > > (in D:/dev/ruby/ironruby/head) > > > > > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > > > initialized constant TAGS > > > > > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > > > initialized constant RUBYSPEC > > > > > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > > > initialized constant MSPEC > > > > > > > > > > > > Summary: > > > > > > 0 types, 0 files, 0 examples, 0 expectations, 0 failures, 0 errors > > > > > > > > > > > > Nothing seems to be getting run? If I run rake spec then it does > > > run the tests: > > > > > > > > > > > > D:\dev\ruby\ironruby\head>rake spec > > > > > > (in D:/dev/ruby/ironruby/head) > > > > > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > > > initialized constant TAGS > > > > > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > > > initialized constant RUBYSPEC > > > > > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > > > initialized constant MSPEC > > > > > > 2199 examples, 0 failures > > > > > > 0 > > > > > > > > > > > > Is this what is expected? > > > > > > > > > > > > Pete > > > > > > > > > > > > From: ironruby-core-bounces at rubyforge.org > > > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam > > > (IRONRUBY) > > > Sent: Tuesday,01 July 01, 2008 00:59 > > > To: ironruby-core at rubyforge.org > > > Subject: [Ironruby-core] SVN r118 and RubySpec integration > > > > > > > > > > > > I just pushed SVN r118 out which contains RubySpec integration. > > > There are a few things that you'll need to do: > > > > > > > > > > > > 1. (Optional) get a GitHub account (http://github.com) > > > > > > 2. (Optional) install GIT: > > > > > > a. Goto the msysgit project and download the .exe based > > > installer: > > > http://code.google.com/p/msysgit/ > > > > > > b. Make sure in the GIT setup that you click on the radio button > > > labeled "Run Git from the Windows Command Prompt" > > > > > > c. Restart your command prompt > > > > > > 3. Navigate to my page: http://github.com/jflam to see the > > three > > > GIT projects that you'll need to clone: ironruby-tags, mspec, and > > > rubyspec. mspec and rubyspec are forks of the main projects so that > > we > > > can add IronRuby specific tests and potential changes to mspec. The > > > RubySpec committers can pull from these repositories to incorporate > > > our changes. ironruby-tags is a project that contains the list of > > > rubyspec specs that we currently exclude to get a clean regression > > baseline. > > > > > > 4. If you have GIT installed, create a directory to hold these > > > projects. I use %USERPROFILE%\dev: > > > > > > a. in ~\dev\ironruby-tags run: git clone > > > git://github.com/jflam/ironruby-tags.git > > > > > > b. in ~\dev\mspec run: git clone > > git://github.com/jflam/mspec.git > > > > > > c. in ~\dev\rubyspec run: git clone > > > git://github.com/jflam/rubyspec.git > > > > > > 5. If you don't have GIT installed (and why don't you??? J) you > > > can download the tarball using the download button in each project. > > > > > > > > > > > > Once you have those projects in place, you'll need to grab r118 from > > > RubyForge. There's a file in \trunk called runfirst.cmd. This will > > > enable skip verification for projects signed with our public key, > > > and it will create a default configuration file at > > > %USERPROFILE%\.irconfig.rb. > > > > > > > > > > > > If you followed my instructions and created > > > %USERPROFILE%\dev\[ironruby-tags,mspec,rubyspec] you're good to go. > > If > > > you didn't, you'll need to edit the paths in .irconfig.rb > > > > > > > > > > > > You can test your setup by running the new regression test baseline > > > rake > > > task: > > > > > > > > > > > > rake regression > > > > > > > > > > > > This should report 0 failures and 0 errors. > > > > > > > > > > > > This is the first test of this stuff outside of my machine, so > > > please kick the tires! > > > > > > > > > > > > Thanks, > > > > > > -John > > > > > > > > > > _______________________________________________ > > Ironruby-core mailing list > > Ironruby-core at rubyforge.org > > http://rubyforge.org/mailman/listinfo/ironruby-core > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Rakefile Type: application/octet-stream Size: 24974 bytes Desc: not available URL: From jflam at microsoft.com Tue Jul 1 13:55:40 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Tue, 1 Jul 2008 10:55:40 -0700 Subject: [Ironruby-core] SVN r118 and RubySpec integration In-Reply-To: References: <372109E149E8084D8E6C7D9CFD82E0632FD3F40782@NA-EXMSG-C115.redmond.corp.microsoft.com> <003801c8db7c$8fd57da0$af8078e0$@com> <372109E149E8084D8E6C7D9CFD82E0632FD3F40885@NA-EXMSG-C115.redmond.corp.microsoft.com> <372109E149E8084D8E6C7D9CFD82E0632FD3F408C5@NA-EXMSG-C115.redmond.corp.microsoft.com> <000001c8db97$b4924a90$1db6dfb0$@com> <372109E149E8084D8E6C7D9CFD82E0632FD3F409CC@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD3F40A93@NA-EXMSG-C115.redmond.corp.microsoft.com> Doh! Thanks for the fix! Thanks, -John From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jirapong Nanta Sent: Tuesday, July 01, 2008 10:46 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration Hi John, It was fail to run... attached file is the fixed version. Thank you, -Jirapong On Tue, Jul 1, 2008 at 11:53 PM, John Lam (IRONRUBY) > wrote: Can you try out r119 that I just pushed out? It should have a test that catches the wrong folder problem that you ran into. Thanks, -John -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Tuesday, July 01, 2008 9:30 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration Oops. Sorry my fault. My irconfig.rb was referring to the wrong folders. (I had c: rather than d:). The line 668 in context.rb Dir["#{UserEnvironment::RUBYSPEC}/1.8/core/*"].each do |path| was never picking up the tests. Pete -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: Tuesday,01 July 01, 2008 15:33 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration I just looked at the files on SVN and they're the same as mine - I'm attaching default.irconfig.rb and context.rb - can you diff to see if they're the same as what's on your machine? I've fixed the quoting problem on this end. Thanks, -John > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) > Sent: Tuesday, July 01, 2008 6:52 AM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > Hmm ... looks like an older version made it out to r118. Let me > investigate. > > Thanks, > -John > > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin > > Sent: Tuesday, July 01, 2008 6:16 AM > > To: ironruby-core at rubyforge.org > > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > > > And finally when I run rake regression I get a suspiciously quick > > result: > > > > > > > > D:\dev\ruby\ironruby\head>rake regression > > > > (in D:/dev/ruby/ironruby/head) > > > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > > initialized constant TAGS > > > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > > initialized constant RUBYSPEC > > > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > > initialized constant MSPEC > > > > > > > > Summary: > > > > 0 types, 0 files, 0 examples, 0 expectations, 0 failures, 0 errors > > > > > > > > Nothing seems to be getting run? If I run rake spec then it does > > run the tests: > > > > > > > > D:\dev\ruby\ironruby\head>rake spec > > > > (in D:/dev/ruby/ironruby/head) > > > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > > initialized constant TAGS > > > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > > initialized constant RUBYSPEC > > > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > > initialized constant MSPEC > > > > 2199 examples, 0 failures > > > > 0 > > > > > > > > Is this what is expected? > > > > > > > > Pete > > > > > > > > From: ironruby-core-bounces at rubyforge.org > > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam > > (IRONRUBY) > > Sent: Tuesday,01 July 01, 2008 00:59 > > To: ironruby-core at rubyforge.org > > Subject: [Ironruby-core] SVN r118 and RubySpec integration > > > > > > > > I just pushed SVN r118 out which contains RubySpec integration. > > There are a few things that you'll need to do: > > > > > > > > 1. (Optional) get a GitHub account (http://github.com) > > > > 2. (Optional) install GIT: > > > > a. Goto the msysgit project and download the .exe based > > installer: > > http://code.google.com/p/msysgit/ > > > > b. Make sure in the GIT setup that you click on the radio button > > labeled "Run Git from the Windows Command Prompt" > > > > c. Restart your command prompt > > > > 3. Navigate to my page: http://github.com/jflam to see the > three > > GIT projects that you'll need to clone: ironruby-tags, mspec, and > > rubyspec. mspec and rubyspec are forks of the main projects so that > we > > can add IronRuby specific tests and potential changes to mspec. The > > RubySpec committers can pull from these repositories to incorporate > > our changes. ironruby-tags is a project that contains the list of > > rubyspec specs that we currently exclude to get a clean regression > baseline. > > > > 4. If you have GIT installed, create a directory to hold these > > projects. I use %USERPROFILE%\dev: > > > > a. in ~\dev\ironruby-tags run: git clone > > git://github.com/jflam/ironruby-tags.git > > > > b. in ~\dev\mspec run: git clone > git://github.com/jflam/mspec.git > > > > c. in ~\dev\rubyspec run: git clone > > git://github.com/jflam/rubyspec.git > > > > 5. If you don't have GIT installed (and why don't you??? J) you > > can download the tarball using the download button in each project. > > > > > > > > Once you have those projects in place, you'll need to grab r118 from > > RubyForge. There's a file in \trunk called runfirst.cmd. This will > > enable skip verification for projects signed with our public key, > > and it will create a default configuration file at > > %USERPROFILE%\.irconfig.rb. > > > > > > > > If you followed my instructions and created > > %USERPROFILE%\dev\[ironruby-tags,mspec,rubyspec] you're good to go. > If > > you didn't, you'll need to edit the paths in .irconfig.rb > > > > > > > > You can test your setup by running the new regression test baseline > > rake > > task: > > > > > > > > rake regression > > > > > > > > This should report 0 failures and 0 errors. > > > > > > > > This is the first test of this stuff outside of my machine, so > > please kick the tires! > > > > > > > > Thanks, > > > > -John > > > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core -------------- next part -------------- An HTML attachment was scrubbed... URL: From jirapong.nanta at gmail.com Tue Jul 1 14:29:34 2008 From: jirapong.nanta at gmail.com (Jirapong Nanta) Date: Wed, 2 Jul 2008 01:29:34 +0700 Subject: [Ironruby-core] SVN r118 and RubySpec integration In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD3F40A93@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD3F40782@NA-EXMSG-C115.redmond.corp.microsoft.com> <003801c8db7c$8fd57da0$af8078e0$@com> <372109E149E8084D8E6C7D9CFD82E0632FD3F40885@NA-EXMSG-C115.redmond.corp.microsoft.com> <372109E149E8084D8E6C7D9CFD82E0632FD3F408C5@NA-EXMSG-C115.redmond.corp.microsoft.com> <000001c8db97$b4924a90$1db6dfb0$@com> <372109E149E8084D8E6C7D9CFD82E0632FD3F409CC@NA-EXMSG-C115.redmond.corp.microsoft.com> <372109E149E8084D8E6C7D9CFD82E0632FD3F40A93@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: next.. run got error about C:\OpenSource\ironRuby\trunk>c:\ruby\bin\ruby.exe c:\Users\v-jirapn\dev\mspec\bin\mspec-ci -fm -X "c:\Users\v-jirapn\dev \ironruby-tags" "c:\Users\v-jirapn\dev\rubyspec\1.8\core\array" c:/Users/v-jirapn/dev/mspec/lib/mspec/utils/script.rb:35:in `expand_path': couldn't find HOME environment -- expanding ` ~/.mspecrc' (ArgumentError) from c:/Users/v-jirapn/dev/mspec/lib/mspec/utils/script.rb:35:in `load' from c:/Users/v-jirapn/dev/mspec/lib/mspec/utils/script.rb:71:in `main' from c:/Users/v-jirapn/dev/mspec/bin/mspec-ci:8 Workaround by run: SET HOME=%UserProfile% then.. run: c:\ruby\bin\ruby.exe c:\Users\v-jirapn\dev\mspec\bin\mspec-ci -fm -X "c:\Users\v-jirapn\dev \ironruby-tags" "c:\Users\v-jirapn\dev\rubyspec\1.8\core\array\hash_spec.rb" Finished in 0.022000 seconds 1 file, 5 examples, 17 expectations, 0 failures, 0 errors also successful to run: rake regression Thanks, -Jirapong On Wed, Jul 2, 2008 at 12:55 AM, John Lam (IRONRUBY) wrote: > Doh! Thanks for the fix! > > > > Thanks, > > -John > > > > *From:* ironruby-core-bounces at rubyforge.org [mailto: > ironruby-core-bounces at rubyforge.org] *On Behalf Of *Jirapong Nanta > *Sent:* Tuesday, July 01, 2008 10:46 AM > > *To:* ironruby-core at rubyforge.org > *Subject:* Re: [Ironruby-core] SVN r118 and RubySpec integration > > > > Hi John, > It was fail to run... attached file is the fixed version. > > Thank you, > -Jirapong > > On Tue, Jul 1, 2008 at 11:53 PM, John Lam (IRONRUBY) > wrote: > > Can you try out r119 that I just pushed out? It should have a test that > catches the wrong folder problem that you ran into. > > > Thanks, > -John > > > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto: > ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin > > Sent: Tuesday, July 01, 2008 9:30 AM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > Oops. Sorry my fault. > My irconfig.rb was referring to the wrong folders. (I had c: rather than > d:). > The line 668 in context.rb > Dir["#{UserEnvironment::RUBYSPEC}/1.8/core/*"].each do |path| > was never picking up the tests. > Pete > > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam > (IRONRUBY) > Sent: Tuesday,01 July 01, 2008 15:33 > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > I just looked at the files on SVN and they're the same as mine - I'm > attaching default.irconfig.rb and context.rb - can you diff to see if > they're the same as what's on your machine? > > I've fixed the quoting problem on this end. > > Thanks, > -John > > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) > > Sent: Tuesday, July 01, 2008 6:52 AM > > To: ironruby-core at rubyforge.org > > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > > > Hmm ... looks like an older version made it out to r118. Let me > > investigate. > > > > Thanks, > > -John > > > > > > > -----Original Message----- > > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > > bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin > > > Sent: Tuesday, July 01, 2008 6:16 AM > > > To: ironruby-core at rubyforge.org > > > Subject: Re: [Ironruby-core] SVN r118 and RubySpec integration > > > > > > And finally when I run rake regression I get a suspiciously quick > > > result: > > > > > > > > > > > > D:\dev\ruby\ironruby\head>rake regression > > > > > > (in D:/dev/ruby/ironruby/head) > > > > > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > > > initialized constant TAGS > > > > > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > > > initialized constant RUBYSPEC > > > > > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > > > initialized constant MSPEC > > > > > > > > > > > > Summary: > > > > > > 0 types, 0 files, 0 examples, 0 expectations, 0 failures, 0 errors > > > > > > > > > > > > Nothing seems to be getting run? If I run rake spec then it does > > > run the tests: > > > > > > > > > > > > D:\dev\ruby\ironruby\head>rake spec > > > > > > (in D:/dev/ruby/ironruby/head) > > > > > > C:\Documents and Settings\pete/.irconfig.rb:19: warning: already > > > initialized constant TAGS > > > > > > C:\Documents and Settings\pete/.irconfig.rb:20: warning: already > > > initialized constant RUBYSPEC > > > > > > C:\Documents and Settings\pete/.irconfig.rb:21: warning: already > > > initialized constant MSPEC > > > > > > 2199 examples, 0 failures > > > > > > 0 > > > > > > > > > > > > Is this what is expected? > > > > > > > > > > > > Pete > > > > > > > > > > > > From: ironruby-core-bounces at rubyforge.org > > > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam > > > (IRONRUBY) > > > Sent: Tuesday,01 July 01, 2008 00:59 > > > To: ironruby-core at rubyforge.org > > > Subject: [Ironruby-core] SVN r118 and RubySpec integration > > > > > > > > > > > > I just pushed SVN r118 out which contains RubySpec integration. > > > There are a few things that you'll need to do: > > > > > > > > > > > > 1. (Optional) get a GitHub account (http://github.com) > > > > > > 2. (Optional) install GIT: > > > > > > a. Goto the msysgit project and download the .exe based > > > installer: > > > http://code.google.com/p/msysgit/ > > > > > > b. Make sure in the GIT setup that you click on the radio button > > > labeled "Run Git from the Windows Command Prompt" > > > > > > c. Restart your command prompt > > > > > > 3. Navigate to my page: http://github.com/jflam to see the > > three > > > GIT projects that you'll need to clone: ironruby-tags, mspec, and > > > rubyspec. mspec and rubyspec are forks of the main projects so that > > we > > > can add IronRuby specific tests and potential changes to mspec. The > > > RubySpec committers can pull from these repositories to incorporate > > > our changes. ironruby-tags is a project that contains the list of > > > rubyspec specs that we currently exclude to get a clean regression > > baseline. > > > > > > 4. If you have GIT installed, create a directory to hold these > > > projects. I use %USERPROFILE%\dev: > > > > > > a. in ~\dev\ironruby-tags run: git clone > > > git://github.com/jflam/ironruby-tags.git > > > > > > b. in ~\dev\mspec run: git clone > > git://github.com/jflam/mspec.git > > > > > > c. in ~\dev\rubyspec run: git clone > > > git://github.com/jflam/rubyspec.git > > > > > > 5. If you don't have GIT installed (and why don't you??? J) you > > > can download the tarball using the download button in each project. > > > > > > > > > > > > Once you have those projects in place, you'll need to grab r118 from > > > RubyForge. There's a file in \trunk called runfirst.cmd. This will > > > enable skip verification for projects signed with our public key, > > > and it will create a default configuration file at > > > %USERPROFILE%\.irconfig.rb. > > > > > > > > > > > > If you followed my instructions and created > > > %USERPROFILE%\dev\[ironruby-tags,mspec,rubyspec] you're good to go. > > If > > > you didn't, you'll need to edit the paths in .irconfig.rb > > > > > > > > > > > > You can test your setup by running the new regression test baseline > > > rake > > > task: > > > > > > > > > > > > rake regression > > > > > > > > > > > > This should report 0 failures and 0 errors. > > > > > > > > > > > > This is the first test of this stuff outside of my machine, so > > > please kick the tires! > > > > > > > > > > > > Thanks, > > > > > > -John > > > > > > > > > > _______________________________________________ > > Ironruby-core mailing list > > Ironruby-core at rubyforge.org > > http://rubyforge.org/mailman/listinfo/ironruby-core > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jflam at microsoft.com Tue Jul 1 18:31:59 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Tue, 1 Jul 2008 15:31:59 -0700 Subject: [Ironruby-core] HOME and USERPROFILE aliasing under Windows Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD3F40D14@NA-EXMSG-C115.redmond.corp.microsoft.com> MRI currently expects the HOME environment variable to be set under Windows. Various operations involving ~ and Dir will fail if HOME is not defined. However, HOME is rarely defined under Windows by default. Instead USERPROFILE is what normally represents HOME. IronRuby is considering aliasing HOME to USERPROFILE if HOME isn't already defined at startup time. Does this seem like something reasonable to do in our implementation? Thanks, -John -------------- next part -------------- An HTML attachment was scrubbed... URL: From ryan.riley at panesofglass.org Tue Jul 1 19:24:42 2008 From: ryan.riley at panesofglass.org (Ryan Riley) Date: Tue, 1 Jul 2008 18:24:42 -0500 Subject: [Ironruby-core] Is it possible to build RubyGems with IronRuby? Message-ID: I have been trying to build RubyGems with IronRuby with the latest sources over the past week, and it's getting better each time, but not quite there. Has anyone else been trying this or been successful in building RubyGems? Is this a goal of IronRuby, or will IronRuby users also need CRuby to build RubyGems and run rake tasks? The current output from running C:\Users\rriley1\dev\>ir.exe rubygems\setup.rb is: C:\Users\rriley1\dev\ironruby\src\IronRuby.Libraries\Builtins\KernelOps.cs:316:in `require': no such file to load -- rubygems (LoadError) from C:\Users\rriley1\dev\ironruby\src\Microsoft.Scripting.Core\Actions\CallSite.cs:288:in `UpdateAndExecute' from C:\Users\rriley1\dev\ironruby\src\Microsoft.Scripting.Core\Actions\UpdateDelegates.Generated.cs:41:in `Update3' from :0 from C:\Users\rriley1\dev\ironruby\src\Microsoft.Scripting.Core\Runtime\ScriptCode.cs:84:in `InvokeTarget' from C:\Users\rriley1\dev\ironruby\src\Microsoft.Scripting.Core\Runtime\ScriptCode.cs:76:in `Run' from C:\Users\rriley1\dev\ironruby\src\Microsoft.Scripting.Core\SourceUnit.cs:246:in `ExecuteProgram' -- Ryan Riley ryan.riley at panesofglass.org http://www.panesofglass.org/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From dremy at microsoft.com Tue Jul 1 20:57:50 2008 From: dremy at microsoft.com (Dave Remy) Date: Tue, 1 Jul 2008 17:57:50 -0700 Subject: [Ironruby-core] Code Review: StringGsubWithPattern Message-ID: tfpt review "/shelveset:StringGsubWithPattern;REDMOND\dremy" Comment : Changes for string.gsub with pattern ruby specs to pass. The following tests now pass: String#gsub with pattern, replacement replaces \& and \0 with the complete match String#gsub with pattern, replacement replaces \` with everything before the current match String#gsub with pattern, replacement replaces \' with everything after the current match String#gsub with pattern, replacement replaces \+ with the last paren that actually matched String#gsub with pattern, replacement treats \+ as an empty string if there was no captures String#gsub with pattern, replacement maps \\ in replacement to \ String#gsub with pattern, replacement leaves \ at the end of replacement untouched -------------- next part -------------- A non-text attachment was scrubbed... Name: StringGsubWithPattern.diff Type: application/octet-stream Size: 7932 bytes Desc: StringGsubWithPattern.diff URL: From olegtk at microsoft.com Tue Jul 1 20:58:57 2008 From: olegtk at microsoft.com (Oleg Tkachenko) Date: Tue, 1 Jul 2008 17:58:57 -0700 Subject: [Ironruby-core] Code Review: YamlRegex Message-ID: tfpt review "/shelveset:YamlRegex;REDMOND\olegtk" Comment : Fixes Ruby Regex support: implements RubyRegex.taguri() and loading regex from yaml. Implements loading literal Ruby ranges. Adds couple useful RubyUtils.Evaluate overloads (one that takes string instead of MutableString and one with no optional arguments). -- Oleg -------------- next part -------------- A non-text attachment was scrubbed... Name: YamlRegex.diff Type: application/octet-stream Size: 10472 bytes Desc: YamlRegex.diff URL: From olegtk at microsoft.com Wed Jul 2 16:01:58 2008 From: olegtk at microsoft.com (Oleg Tkachenko) Date: Wed, 2 Jul 2008 13:01:58 -0700 Subject: [Ironruby-core] Code Review: YamlRegex - Please ignore In-Reply-To: References: Message-ID: Submitted by mistake, please ignore this one. -- Oleg From Tomas.Matousek at microsoft.com Wed Jul 2 21:36:52 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Wed, 2 Jul 2008 18:36:52 -0700 Subject: [Ironruby-core] Code Review: YamlRegex In-Reply-To: References: Message-ID: I get an error when trying to review your shelveset: C:\M5\Merlin\Main>tfpt review /shelveset:YamlRegex;olegtk No shelvesets found matching YamlRegex;olegtk Tomas -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Oleg Tkachenko Sent: Tuesday, July 01, 2008 5:59 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: YamlRegex tfpt review "/shelveset:YamlRegex;REDMOND\olegtk" Comment : Fixes Ruby Regex support: implements RubyRegex.taguri() and loading regex from yaml. Implements loading literal Ruby ranges. Adds couple useful RubyUtils.Evaluate overloads (one that takes string instead of MutableString and one with no optional arguments). -- Oleg From olegtk at microsoft.com Thu Jul 3 01:27:31 2008 From: olegtk at microsoft.com (Oleg Tkachenko) Date: Wed, 2 Jul 2008 22:27:31 -0700 Subject: [Ironruby-core] Code Review: YamlRegex In-Reply-To: References: , Message-ID: I removed it, please ignore. -- Oleg ________________________________________ From: Tomas Matousek Sent: Wednesday, July 02, 2008 6:36 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: RE: Code Review: YamlRegex I get an error when trying to review your shelveset: C:\M5\Merlin\Main>tfpt review /shelveset:YamlRegex;olegtk No shelvesets found matching YamlRegex;olegtk Tomas -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Oleg Tkachenko Sent: Tuesday, July 01, 2008 5:59 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: YamlRegex tfpt review "/shelveset:YamlRegex;REDMOND\olegtk" Comment : Fixes Ruby Regex support: implements RubyRegex.taguri() and loading regex from yaml. Implements loading literal Ruby ranges. Adds couple useful RubyUtils.Evaluate overloads (one that takes string instead of MutableString and one with no optional arguments). -- Oleg From jirapong.nanta at gmail.com Thu Jul 3 14:58:22 2008 From: jirapong.nanta at gmail.com (Jirapong) Date: Fri, 4 Jul 2008 01:58:22 +0700 Subject: [Ironruby-core] HOME and USERPROFILE aliasing under Windows In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD3F40D14@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD3F40D14@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <453CD690-9221-4101-A206-64F06FCF3C02@gmail.com> same problem reported at #20475. and i agree that HOME Env should handle by that class -Jirapong On Jul 2, 2008, at 5:31 AM, "John Lam (IRONRUBY)" wrote: > MRI currently expects the HOME environment variable to be set under > Windows. Various operations involving ~ and Dir will fail if HOME is > not defined. > > > > However, HOME is rarely defined under Windows by default. Instead > USERPROFILE is what normally represents HOME. IronRuby is > considering aliasing HOME to USERPROFILE if HOME isn't already > defined at startup time. > > > > Does this seem like something reasonable to do in our implementation? > > > > Thanks, > > -John > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core From dremy at microsoft.com Thu Jul 3 20:26:44 2008 From: dremy at microsoft.com (Dave Remy) Date: Thu, 3 Jul 2008 17:26:44 -0700 Subject: [Ironruby-core] Code Review: StringGsubWithStringPatternAndBlock Message-ID: tfpt review "/shelveset:StringGsubWithStringPatternAndBlock;REDMOND\dremy" Comment : This change fixes the rest of the gsub coverage minus one (String#gsub with pattern and block raises a RuntimeError if the string is modified while substituting). We had a similar issue implementing squeeze so it looks like we will have to track inflight changes to the underlying string (frozen already does this, I wonder if there is some variant we can use). This changes fixes the following specs: String#gsub with pattern and block sets $~ for access from the block String#gsub with pattern and block restores $~ after leaving the block String#gsub with pattern and block sets $~ to MatchData of last match and nil when there's none for access from outside -------------- next part -------------- A non-text attachment was scrubbed... Name: StringGsubWithStringPatternAndBlock.diff Type: application/octet-stream Size: 4331 bytes Desc: StringGsubWithStringPatternAndBlock.diff URL: From bacondarwin at googlemail.com Fri Jul 4 14:08:02 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Fri, 4 Jul 2008 19:08:02 +0100 Subject: [Ironruby-core] Rake Regression Message-ID: <000001c8de00$e756c990$b6045cb0$@com> Is the regression supposed to pass with 100%? I get 1 fail and 20 errors (r121): D:\dev\ruby\ironruby\head>rake regression (in D:/dev/ruby/ironruby/head) array: 71 files, 485 examples, 1592 expectations, 0 failures, 0 errors bignum: 31 files, 74 examples, 209 expectations, 0 failures, 0 errors binding: 2 files, 0 examples, 0 expectations, 0 failures, 0 errors class: 6 files, 19 examples, 36 expectations, 0 failures, 0 errors comparable: 6 files, 16 examples, 90 expectations, 0 failures, 0 errors continuation: 3 files, 0 examples, 0 expectations, 0 failures, 0 errors dir: 24 files, 96 examples, 127 expectations, 0 failures, 0 errors enumerable: 23 files, 69 examples, 178 expectations, 0 failures, 0 errors env: 37 files, 47 examples, 59 expectations, 0 failures, 0 errors exception: 21 files, 39 examples, 35 expectations, 0 failures, 0 errors false: 5 files, 5 examples, 17 expectations, 0 failures, 0 errors file: 137 files, 232 examples, 355 expectations, 1 failures, 20 errors filetest: 23 files, 4 examples, 6 expectations, 0 failures, 0 errors fixnum: 33 files, 105 examples, 330 expectations, 0 failures, 0 errors float: 33 files, 51 examples, 167 expectations, 0 failures, 0 errors gc: 4 files, 0 examples, 0 expectations, 0 failures, 0 errors hash: 48 files, 177 examples, 400 expectations, 0 failures, 0 errors integer: 18 files, 32 examples, 76 expectations, 0 failures, 0 errors io: 60 files, 121 examples, 201 expectations, 0 failures, 0 errors kernel: 105 files, 304 examples, 2447 expectations, 0 failures, 0 errors marshal: 3 files, 77 examples, 93 expectations, 0 failures, 0 errors matchdata: 16 files, 16 examples, 24 expectations, 0 failures, 0 errors math: 24 files, 67 examples, 123 expectations, 0 failures, 0 errors method: 9 files, 19 examples, 41 expectations, 0 failures, 0 errors module: 64 files, 174 examples, 428 expectations, 0 failures, 0 errors nil: 10 files, 13 examples, 25 expectations, 0 failures, 0 errors numeric: 22 files, 44 examples, 109 expectations, 0 failures, 0 errors object: 9 files, 25 examples, 39 expectations, 0 failures, 0 errors objectspace: 9 files, 2 examples, 2 expectations, 0 failures, 0 errors precision: 4 files, 10 examples, 10 expectations, 0 failures, 0 errors proc: 11 files, 19 examples, 88 expectations, 0 failures, 0 errors process: 30 files, 3 examples, 6 expectations, 0 failures, 0 errors range: 18 files, 29 examples, 176 expectations, 0 failures, 0 errors regexp: 19 files, 63 examples, 177 expectations, 0 failures, 0 errors signal: 2 files, 2 examples, 2 expectations, 0 failures, 0 errors string: 68 files, 596 examples, 4262 expectations, 0 failures, 0 errors struct: 20 files, 51 examples, 106 expectations, 0 failures, 0 errors symbol: 10 files, 57 examples, 82 expectations, 0 failures, 0 errors thread: 29 files, 11 examples, 15 expectations, 0 failures, 0 errors threadgroup: 4 files, 2 examples, 8 expectations, 0 failures, 0 errors time: 51 files, 25 examples, 55 expectations, 0 failures, 0 errors true: 5 files, 5 examples, 17 expectations, 0 failures, 0 errors unboundmethod: 6 files, 20 examples, 48 expectations, 0 failures, 0 errors Summary: 43 types, 1133 files, 3206 examples, 12261 expectations, 1 failures, 20 errors -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdeville at microsoft.com Fri Jul 4 14:44:52 2008 From: jdeville at microsoft.com (Jim Deville) Date: Fri, 4 Jul 2008 11:44:52 -0700 Subject: [Ironruby-core] Rake Regression Message-ID: What version of the git repos is this? I did some fixes last night that should fix some of the failures. -----Original Message----- From: "Peter Bacon Darwin" To: "ironruby-core at rubyforge.org" Sent: 7/4/08 11:08 AM Subject: [Ironruby-core] Rake Regression Is the regression supposed to pass with 100%? I get 1 fail and 20 errors (r121): D:\dev\ruby\ironruby\head>rake regression (in D:/dev/ruby/ironruby/head) array: 71 files, 485 examples, 1592 expectations, 0 failures, 0 errors bignum: 31 files, 74 examples, 209 expectations, 0 failures, 0 errors binding: 2 files, 0 examples, 0 expectations, 0 failures, 0 errors class: 6 files, 19 examples, 36 expectations, 0 failures, 0 errors comparable: 6 files, 16 examples, 90 expectations, 0 failures, 0 errors continuation: 3 files, 0 examples, 0 expectations, 0 failures, 0 errors dir: 24 files, 96 examples, 127 expectations, 0 failures, 0 errors enumerable: 23 files, 69 examples, 178 expectations, 0 failures, 0 errors env: 37 files, 47 examples, 59 expectations, 0 failures, 0 errors exception: 21 files, 39 examples, 35 expectations, 0 failures, 0 errors false: 5 files, 5 examples, 17 expectations, 0 failures, 0 errors file: 137 files, 232 examples, 355 expectations, 1 failures, 20 errors filetest: 23 files, 4 examples, 6 expectations, 0 failures, 0 errors fixnum: 33 files, 105 examples, 330 expectations, 0 failures, 0 errors float: 33 files, 51 examples, 167 expectations, 0 failures, 0 errors gc: 4 files, 0 examples, 0 expectations, 0 failures, 0 errors hash: 48 files, 177 examples, 400 expectations, 0 failures, 0 errors integer: 18 files, 32 examples, 76 expectations, 0 failures, 0 errors io: 60 files, 121 examples, 201 expectations, 0 failures, 0 errors kernel: 105 files, 304 examples, 2447 expectations, 0 failures, 0 errors marshal: 3 files, 77 examples, 93 expectations, 0 failures, 0 errors matchdata: 16 files, 16 examples, 24 expectations, 0 failures, 0 errors math: 24 files, 67 examples, 123 expectations, 0 failures, 0 errors method: 9 files, 19 examples, 41 expectations, 0 failures, 0 errors module: 64 files, 174 examples, 428 expectations, 0 failures, 0 errors nil: 10 files, 13 examples, 25 expectations, 0 failures, 0 errors numeric: 22 files, 44 examples, 109 expectations, 0 failures, 0 errors object: 9 files, 25 examples, 39 expectations, 0 failures, 0 errors objectspace: 9 files, 2 examples, 2 expectations, 0 failures, 0 errors precision: 4 files, 10 examples, 10 expectations, 0 failures, 0 errors proc: 11 files, 19 examples, 88 expectations, 0 failures, 0 errors process: 30 files, 3 examples, 6 expectations, 0 failures, 0 errors range: 18 files, 29 examples, 176 expectations, 0 failures, 0 errors regexp: 19 files, 63 examples, 177 expectations, 0 failures, 0 errors signal: 2 files, 2 examples, 2 expectations, 0 failures, 0 errors string: 68 files, 596 examples, 4262 expectations, 0 failures, 0 errors struct: 20 files, 51 examples, 106 expectations, 0 failures, 0 errors symbol: 10 files, 57 examples, 82 expectations, 0 failures, 0 errors thread: 29 files, 11 examples, 15 expectations, 0 failures, 0 errors threadgroup: 4 files, 2 examples, 8 expectations, 0 failures, 0 errors time: 51 files, 25 examples, 55 expectations, 0 failures, 0 errors true: 5 files, 5 examples, 17 expectations, 0 failures, 0 errors unboundmethod: 6 files, 20 examples, 48 expectations, 0 failures, 0 errors Summary: 43 types, 1133 files, 3206 examples, 12261 expectations, 1 failures, 20 errors -------------- next part -------------- An HTML attachment was scrubbed... URL: From michael.letterle at gmail.com Fri Jul 4 23:01:56 2008 From: michael.letterle at gmail.com (Michael Letterle) Date: Fri, 4 Jul 2008 23:01:56 -0400 Subject: [Ironruby-core] eRubyCon 2008 Message-ID: Just wanted to make a quick plug for http://www.erubycon.com/ coming up the middle of August in Columbus, OH. It's a conference on using Ruby in the Enterprise (which I think IronRuby is well situated to be a major factor in) If you're in the USA midwest you should come on by, they're going to have a bunch of fantastic speakers including Charles Nutter, the JRuby lead. (disclaimer: I'm speaking as well) -- Michael Letterle [Polymath Prokrammer] http://blog.prokrams.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From ksunair at yahoo.com Sat Jul 5 04:19:14 2008 From: ksunair at yahoo.com (Unnikrishnan Nair) Date: Sat, 5 Jul 2008 01:19:14 -0700 (PDT) Subject: [Ironruby-core] spec question Message-ID: <658171.40545.qm@web50607.mail.re2.yahoo.com> When I was testing the File.zero? I noticed File.zero?('NUL') failed. I looked the code and it seems there is no special check there to exclude NUL as a not a valid file. I created a constant ??????? private readonly static string NUL_VALUE = "NUL"; and added a else clause in the trycreate method? as else if (path.Equals(NUL_VALUE)) ??????????????? { ??????????????????? result = null; ??????????????? } and now the NUL works fine for zero. I have two questions with this regard; 1. Can I apply these changes and push in the patch? 2. When we run the rake spec, will it include following tests on Windows? platform :not, :mswin do ??? it "returns true for /dev/null" do ????? File.zero?('/dev/null').should == true ??? end ? end It seems it does, I haven't looked at the Rubinius to see how the platform is supposed to work. Thanks, unni -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdeville at microsoft.com Sat Jul 5 16:16:22 2008 From: jdeville at microsoft.com (Jim Deville) Date: Sat, 5 Jul 2008 13:16:22 -0700 Subject: [Ironruby-core] spec question Message-ID: You should be fine making a patch. As for the test, it shouldn't run that test on Windows, but I'm not sure how the old runner works. JD -----Original Message----- From: "Unnikrishnan Nair" To: "ironruby-core at rubyforge.org" Sent: 7/5/08 1:22 AM Subject: [Ironruby-core] spec question When I was testing the File.zero? I noticed File.zero?('NUL') failed. I looked the code and it seems there is no special check there to exclude NUL as a not a valid file. I created a constant private readonly static string NUL_VALUE = "NUL"; and added a else clause in the trycreate method as else if (path.Equals(NUL_VALUE)) { result = null; } and now the NUL works fine for zero. I have two questions with this regard; 1. Can I apply these changes and push in the patch? 2. When we run the rake spec, will it include following tests on Windows? platform :not, :mswin do it "returns true for /dev/null" do File.zero?('/dev/null').should == true end end It seems it does, I haven't looked at the Rubinius to see how the platform is supposed to work. Thanks, unni -------------- next part -------------- An HTML attachment was scrubbed... URL: From lists at ruby-forum.com Sat Jul 5 22:09:55 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Sun, 6 Jul 2008 04:09:55 +0200 Subject: [Ironruby-core] Memory Leak with Certain Script Exceptions Message-ID: <18dd5199cf9965684be32e7b5c1c92eb@ruby-forum.com> Hi, I use the IronRuby scripting engine in a long running server process and came across an issue where the memory utilisation of the process started steadily rising in line with a memory leak. The server process is written in C# (no unmanaged code) and I spent a lot of time double checking and profiling the application to try ensure there were no circular references or static lists keeping hold of objects etc. Eventually I started looking at the Ruby script processing and found that when a script threw certain exceptions the memory utilisation on the process would start steadily incrementing. It's not all exceptions, syntax errors and method missing execeptions seem fine but ArgumentOutOfRange or NullReferenceExceptions exhibit the behaviour. A stack trace of such a call is shown below. I checked the behaviour using the latest version from svn (6 Jul 2008). System.ArgumentOutOfRangeException: startIndex cannot be larger than length of string. Parameter name: startIndex at System.String.InternalSubStringWithChecks(Int32 startIndex, Int32 length, Boolean fAlwaysCopy) at _stub_##44143(Closure , CallSite , CodeContext , Object , Int32 ) at Microsoft.Scripting.Utils.InvokeHelper`6.Invoke(Object arg0, Object arg1, Object arg2, Object arg3, Object arg4) at Microsoft.Scripting.Utils.ReflectedCaller.InvokeInstance(Object instance, Object[] args) at Microsoft.Scripting.Actions.ActionBinder.UpdateSiteAndExecute[T](CodeContext context, CallSite`1 site, Object[] args) at Microsoft.Scripting.Actions.UpdateDelegates.Update2[T0,T1,TRet](CallSite site, CodeContext context, T0 arg0, T1 arg1) at Initialize##44129(Closure , CodeContext ) at Microsoft.Scripting.ScriptCode.Run(CodeContext context, Boolean tryEvaluate) at Microsoft.Scripting.ScriptCode.Run(Scope scope, Boolean tryEvaluate) at Microsoft.Scripting.ScriptCode.Run(Scope scope) at Microsoft.Scripting.SourceUnit.Execute(Scope scope, ErrorSink errorSink) at Microsoft.Scripting.SourceUnit.Execute(Scope scope) at Microsoft.Scripting.Hosting.ScriptScope.Execute(String code) Regards, Aaron -- Posted via http://www.ruby-forum.com/. From ksunair at yahoo.com Sun Jul 6 02:07:19 2008 From: ksunair at yahoo.com (Unnikrishnan Nair) Date: Sat, 5 Jul 2008 23:07:19 -0700 (PDT) Subject: [Ironruby-core] spec question In-Reply-To: Message-ID: <465782.79278.qm@web50609.mail.re2.yahoo.com> Jim, ? Now that we moved to GIT, do we submit the patches through GIT? ? Thanks Unni --- On Sat, 7/5/08, Jim Deville wrote: From: Jim Deville Subject: RE: [Ironruby-core] spec question To: "ksunair at yahoo.com" , "ironruby-core at rubyforge.org" Date: Saturday, July 5, 2008, 2:16 PM You should be fine making a patch. As for the test, it shouldn't run that test on Windows, but I'm not sure how the old runner works. JD -----Original Message----- From: "Unnikrishnan Nair" To: "ironruby-core at rubyforge.org" Sent: 7/5/08 1:22 AM Subject: [Ironruby-core] spec question When I was testing the File.zero? I noticed File.zero?('NUL') failed. I looked the code and it seems there is no special check there to exclude NUL as a not a valid file. I created a constant ??????? private readonly static string NUL_VALUE = "NUL"; and added a else clause in the trycreate method? as else if (path.Equals(NUL_VALUE)) ??????????????? { ??????????????????? result = null; ??????????????? } and now the NUL works fine for zero. I have two questions with this regard; 1. Can I apply these changes and push in the patch? 2. When we run the rake spec, will it include following tests on Windows? platform :not, :mswin do ??? it "returns true for /dev/null" do ????? File.zero?('/dev/null').should == true ??? end ? end It seems it does, I haven't looked at the Rubinius to see how the platform is supposed to work. Thanks, unni -------------- next part -------------- An HTML attachment was scrubbed... URL: From Tomas.Matousek at microsoft.com Sun Jul 6 02:28:43 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Sat, 5 Jul 2008 23:28:43 -0700 Subject: [Ironruby-core] Memory Leak with Certain Script Exceptions In-Reply-To: <18dd5199cf9965684be32e7b5c1c92eb@ruby-forum.com> References: <18dd5199cf9965684be32e7b5c1c92eb@ruby-forum.com> Message-ID: What does the script look like? BTW: Circular references alone cannot cause memory leak. CRL GC handles them right. Tomas -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Aaron Clauson Sent: Saturday, July 05, 2008 7:10 PM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] Memory Leak with Certain Script Exceptions Hi, I use the IronRuby scripting engine in a long running server process and came across an issue where the memory utilisation of the process started steadily rising in line with a memory leak. The server process is written in C# (no unmanaged code) and I spent a lot of time double checking and profiling the application to try ensure there were no circular references or static lists keeping hold of objects etc. Eventually I started looking at the Ruby script processing and found that when a script threw certain exceptions the memory utilisation on the process would start steadily incrementing. It's not all exceptions, syntax errors and method missing execeptions seem fine but ArgumentOutOfRange or NullReferenceExceptions exhibit the behaviour. A stack trace of such a call is shown below. I checked the behaviour using the latest version from svn (6 Jul 2008). System.ArgumentOutOfRangeException: startIndex cannot be larger than length of string. Parameter name: startIndex at System.String.InternalSubStringWithChecks(Int32 startIndex, Int32 length, Boolean fAlwaysCopy) at _stub_##44143(Closure , CallSite , CodeContext , Object , Int32 ) at Microsoft.Scripting.Utils.InvokeHelper`6.Invoke(Object arg0, Object arg1, Object arg2, Object arg3, Object arg4) at Microsoft.Scripting.Utils.ReflectedCaller.InvokeInstance(Object instance, Object[] args) at Microsoft.Scripting.Actions.ActionBinder.UpdateSiteAndExecute[T](CodeContext context, CallSite`1 site, Object[] args) at Microsoft.Scripting.Actions.UpdateDelegates.Update2[T0,T1,TRet](CallSite site, CodeContext context, T0 arg0, T1 arg1) at Initialize##44129(Closure , CodeContext ) at Microsoft.Scripting.ScriptCode.Run(CodeContext context, Boolean tryEvaluate) at Microsoft.Scripting.ScriptCode.Run(Scope scope, Boolean tryEvaluate) at Microsoft.Scripting.ScriptCode.Run(Scope scope) at Microsoft.Scripting.SourceUnit.Execute(Scope scope, ErrorSink errorSink) at Microsoft.Scripting.SourceUnit.Execute(Scope scope) at Microsoft.Scripting.Hosting.ScriptScope.Execute(String code) Regards, Aaron -- Posted via http://www.ruby-forum.com/. _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From lists at ruby-forum.com Sun Jul 6 02:45:11 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Sun, 6 Jul 2008 08:45:11 +0200 Subject: [Ironruby-core] Memory Leak with Certain Script Exceptions In-Reply-To: References: <18dd5199cf9965684be32e7b5c1c92eb@ruby-forum.com> Message-ID: <9c209bbd19c929cfe31819fe9e2d3d9e@ruby-forum.com> Tomas Matousek wrote: > What does the script look like? Hi Tomas, The Ruby script could be a single line such as: sys.DoMethod(1, 2, 3) Where sys is an object from a custom .Net class that has been loaded and executed as: ScriptRuntime scriptRuntime = IronRuby.CreateRuntime(); ScriptScope rubyScope = scriptRuntime.CreateScope("IronRuby"); DialPlanHelper helper = new DialPlanHelper(); rubyScope.SetVariable("sys", helper); rubyScope.Execute("sys.DoMethod(1, 2, 3)"); If the DoMethod is called with the correct parameters it works fine, however if the wrong number of arguments are provided a System.ArgumentOutOfRangeException is thrown and the memory leak occurs. Regards, Aaron -- Posted via http://www.ruby-forum.com/. From jdeville at microsoft.com Sun Jul 6 06:57:34 2008 From: jdeville at microsoft.com (Jim Deville) Date: Sun, 6 Jul 2008 03:57:34 -0700 Subject: [Ironruby-core] spec question Message-ID: I'd say yes. You can commit to your local repository and use "git format-patch -1" to make the patches. Change the number to match the number of commits you have made. In addition, remember that rake spec runs the tests in tests/core, which isn't the same as the git tests. JD -----Original Message----- From: "Unnikrishnan Nair" To: "ironruby-core at rubyforge.org" Sent: 7/5/08 11:07 PM Subject: Re: [Ironruby-core] spec question Jim, Now that we moved to GIT, do we submit the patches through GIT? Thanks Unni --- On Sat, 7/5/08, Jim Deville wrote: From: Jim Deville Subject: RE: [Ironruby-core] spec question To: "ksunair at yahoo.com" , "ironruby-core at rubyforge.org" Date: Saturday, July 5, 2008, 2:16 PM You should be fine making a patch. As for the test, it shouldn't run that test on Windows, but I'm not sure how the old runner works. JD -----Original Message----- From: "Unnikrishnan Nair" To: "ironruby-core at rubyforge.org" Sent: 7/5/08 1:22 AM Subject: [Ironruby-core] spec question When I was testing the File.zero? I noticed File.zero?('NUL') failed. I looked the code and it seems there is no special check there to exclude NUL as a not a valid file. I created a constant private readonly static string NUL_VALUE = "NUL"; and added a else clause in the trycreate method as else if (path.Equals(NUL_VALUE)) { result = null; } and now the NUL works fine for zero. I have two questions with this regard; 1. Can I apply these changes and push in the patch? 2. When we run the rake spec, will it include following tests on Windows? platform :not, :mswin do it "returns true for /dev/null" do File.zero?('/dev/null').should == true end end It seems it does, I haven't looked at the Rubinius to see how the platform is supposed to work. Thanks, unni -------------- next part -------------- An HTML attachment was scrubbed... URL: From Johan.Danforth at irm.se Sun Jul 6 10:28:19 2008 From: Johan.Danforth at irm.se (Johan Danforth) Date: Sun, 6 Jul 2008 16:28:19 +0200 Subject: [Ironruby-core] rand() needs a seed? Message-ID: <12D498AE0ADB20479A22F41E113B67D302BE33CCF9@sun.irm.se> I'm a noob Ruby user, but IronRuby behaves different from other Ruby implementations that I've tried, in IronRuby (rev 121) this code: 100.times {p rand(100)} Often returns the same value for (almost) all iterations, like this: 2 2 2 2 2 2 2 2 ...and so on... Works as designed? Need to seed it or something? /Johan -------------- next part -------------- An HTML attachment was scrubbed... URL: From curth at microsoft.com Sun Jul 6 11:15:41 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Sun, 6 Jul 2008 08:15:41 -0700 Subject: [Ironruby-core] rand() needs a seed? In-Reply-To: <12D498AE0ADB20479A22F41E113B67D302BE33CCF9@sun.irm.se> References: <12D498AE0ADB20479A22F41E113B67D302BE33CCF9@sun.irm.se> Message-ID: By default, the parameterless constructor of the System.Random class (which is what we're using) uses the system time as the initial seed. Every time you ask for a random number, we create a new Random object, and in a tight loop like yours, they'll probably all get the same seed. We should probably create a single Random object the first time that one is requested and store it in the RubyContext. From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Johan Danforth Sent: Sunday, July 06, 2008 7:28 AM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] rand() needs a seed? I'm a noob Ruby user, but IronRuby behaves different from other Ruby implementations that I've tried, in IronRuby (rev 121) this code: 100.times {p rand(100)} Often returns the same value for (almost) all iterations, like this: 2 2 2 2 2 2 2 2 ...and so on... Works as designed? Need to seed it or something? /Johan -------------- next part -------------- An HTML attachment was scrubbed... URL: From Johan.Danforth at irm.se Sun Jul 6 12:50:53 2008 From: Johan.Danforth at irm.se (Johan Danforth) Date: Sun, 6 Jul 2008 18:50:53 +0200 Subject: [Ironruby-core] rand() needs a seed? In-Reply-To: References: <12D498AE0ADB20479A22F41E113B67D302BE33CCF9@sun.irm.se> Message-ID: <12D498AE0ADB20479A22F41E113B67D302BE33CCFB@sun.irm.se> Like this? private static readonly Random rndGen = new Random(); [RubyMethod("rand", RubyMethodAttributes.PrivateInstance)] [RubyMethod("rand", RubyMethodAttributes.PublicSingleton)] public static double Rand(CodeContext/*!*/ context, object self) { lock(rndGen) return rndGen.NextDouble(); } [RubyMethod("rand", RubyMethodAttributes.PrivateInstance)] [RubyMethod("rand", RubyMethodAttributes.PublicSingleton)] public static object Rand(CodeContext/*!*/ context, object self, int limit) { if (limit == 0) return Rand(context, self); lock (rndGen) return RuntimeHelpers.Int32ToObject((int)(rndGen.NextDouble() * (limit - 1))); } [RubyMethod("rand", RubyMethodAttributes.PrivateInstance)] [RubyMethod("rand", RubyMethodAttributes.PublicSingleton)] public static object Rand(CodeContext/*!*/ context, object self, double limit) { if (limit < 1) return Rand(context, self); lock (rndGen) return RuntimeHelpers.Int32ToObject((int)(rndGen.NextDouble() * (limit - 1))); } I put a lock() on each call as well, because I believe the Next-methods aren't thread safe. Works for me. /Johan From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: den 6 juli 2008 17:16 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rand() needs a seed? By default, the parameterless constructor of the System.Random class (which is what we're using) uses the system time as the initial seed. Every time you ask for a random number, we create a new Random object, and in a tight loop like yours, they'll probably all get the same seed. We should probably create a single Random object the first time that one is requested and store it in the RubyContext. From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Johan Danforth Sent: Sunday, July 06, 2008 7:28 AM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] rand() needs a seed? I'm a noob Ruby user, but IronRuby behaves different from other Ruby implementations that I've tried, in IronRuby (rev 121) this code: 100.times {p rand(100)} Often returns the same value for (almost) all iterations, like this: 2 2 2 2 2 2 2 2 ...and so on... Works as designed? Need to seed it or something? /Johan -------------- next part -------------- An HTML attachment was scrubbed... URL: From curth at microsoft.com Sun Jul 6 14:49:22 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Sun, 6 Jul 2008 11:49:22 -0700 Subject: [Ironruby-core] rand() needs a seed? In-Reply-To: <12D498AE0ADB20479A22F41E113B67D302BE33CCFB@sun.irm.se> References: <12D498AE0ADB20479A22F41E113B67D302BE33CCF9@sun.irm.se> <12D498AE0ADB20479A22F41E113B67D302BE33CCFB@sun.irm.se> Message-ID: Something like that, yes, but making it a static member of KernelOps means that we have to pay the construction cost during startup. For a function as infrequently used as "rand", it might be better to initialize it more lazily. From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Johan Danforth Sent: Sunday, July 06, 2008 9:51 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rand() needs a seed? Like this? private static readonly Random rndGen = new Random(); [RubyMethod("rand", RubyMethodAttributes.PrivateInstance)] [RubyMethod("rand", RubyMethodAttributes.PublicSingleton)] public static double Rand(CodeContext/*!*/ context, object self) { lock(rndGen) return rndGen.NextDouble(); } [RubyMethod("rand", RubyMethodAttributes.PrivateInstance)] [RubyMethod("rand", RubyMethodAttributes.PublicSingleton)] public static object Rand(CodeContext/*!*/ context, object self, int limit) { if (limit == 0) return Rand(context, self); lock (rndGen) return RuntimeHelpers.Int32ToObject((int)(rndGen.NextDouble() * (limit - 1))); } [RubyMethod("rand", RubyMethodAttributes.PrivateInstance)] [RubyMethod("rand", RubyMethodAttributes.PublicSingleton)] public static object Rand(CodeContext/*!*/ context, object self, double limit) { if (limit < 1) return Rand(context, self); lock (rndGen) return RuntimeHelpers.Int32ToObject((int)(rndGen.NextDouble() * (limit - 1))); } I put a lock() on each call as well, because I believe the Next-methods aren't thread safe. Works for me. /Johan From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: den 6 juli 2008 17:16 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rand() needs a seed? By default, the parameterless constructor of the System.Random class (which is what we're using) uses the system time as the initial seed. Every time you ask for a random number, we create a new Random object, and in a tight loop like yours, they'll probably all get the same seed. We should probably create a single Random object the first time that one is requested and store it in the RubyContext. From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Johan Danforth Sent: Sunday, July 06, 2008 7:28 AM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] rand() needs a seed? I'm a noob Ruby user, but IronRuby behaves different from other Ruby implementations that I've tried, in IronRuby (rev 121) this code: 100.times {p rand(100)} Often returns the same value for (almost) all iterations, like this: 2 2 2 2 2 2 2 2 ...and so on... Works as designed? Need to seed it or something? /Johan -------------- next part -------------- An HTML attachment was scrubbed... URL: From Tomas.Matousek at microsoft.com Sun Jul 6 18:01:30 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Sun, 6 Jul 2008 15:01:30 -0700 Subject: [Ironruby-core] rand() needs a seed? In-Reply-To: References: <12D498AE0ADB20479A22F41E113B67D302BE33CCF9@sun.irm.se> <12D498AE0ADB20479A22F41E113B67D302BE33CCFB@sun.irm.se> Message-ID: Note there is also srand() method that seeds the generator. The generator therefore needs to be associated with RubyExecutionContext. Tomas From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Sunday, July 06, 2008 11:49 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rand() needs a seed? Something like that, yes, but making it a static member of KernelOps means that we have to pay the construction cost during startup. For a function as infrequently used as "rand", it might be better to initialize it more lazily. From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Johan Danforth Sent: Sunday, July 06, 2008 9:51 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rand() needs a seed? Like this? private static readonly Random rndGen = new Random(); [RubyMethod("rand", RubyMethodAttributes.PrivateInstance)] [RubyMethod("rand", RubyMethodAttributes.PublicSingleton)] public static double Rand(CodeContext/*!*/ context, object self) { lock(rndGen) return rndGen.NextDouble(); } [RubyMethod("rand", RubyMethodAttributes.PrivateInstance)] [RubyMethod("rand", RubyMethodAttributes.PublicSingleton)] public static object Rand(CodeContext/*!*/ context, object self, int limit) { if (limit == 0) return Rand(context, self); lock (rndGen) return RuntimeHelpers.Int32ToObject((int)(rndGen.NextDouble() * (limit - 1))); } [RubyMethod("rand", RubyMethodAttributes.PrivateInstance)] [RubyMethod("rand", RubyMethodAttributes.PublicSingleton)] public static object Rand(CodeContext/*!*/ context, object self, double limit) { if (limit < 1) return Rand(context, self); lock (rndGen) return RuntimeHelpers.Int32ToObject((int)(rndGen.NextDouble() * (limit - 1))); } I put a lock() on each call as well, because I believe the Next-methods aren't thread safe. Works for me. /Johan From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: den 6 juli 2008 17:16 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rand() needs a seed? By default, the parameterless constructor of the System.Random class (which is what we're using) uses the system time as the initial seed. Every time you ask for a random number, we create a new Random object, and in a tight loop like yours, they'll probably all get the same seed. We should probably create a single Random object the first time that one is requested and store it in the RubyContext. From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Johan Danforth Sent: Sunday, July 06, 2008 7:28 AM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] rand() needs a seed? I'm a noob Ruby user, but IronRuby behaves different from other Ruby implementations that I've tried, in IronRuby (rev 121) this code: 100.times {p rand(100)} Often returns the same value for (almost) all iterations, like this: 2 2 2 2 2 2 2 2 ...and so on... Works as designed? Need to seed it or something? /Johan -------------- next part -------------- An HTML attachment was scrubbed... URL: From ksunair at yahoo.com Sun Jul 6 18:15:30 2008 From: ksunair at yahoo.com (Unnikrishnan Nair) Date: Sun, 6 Jul 2008 15:15:30 -0700 (PDT) Subject: [Ironruby-core] spec question In-Reply-To: Message-ID: <456348.93654.qm@web50601.mail.re2.yahoo.com> Jim, ? ? I put the patch on the rubyforge while I am getting used to GIT. Sorry. [#21102] Updated File/Zero method to handle NUL I am running all the spec against my sample codes I have and instead of sending one by one I will try to send couple of them together. ? Thanks. --- On Sun, 7/6/08, Jim Deville wrote: From: Jim Deville Subject: RE: [Ironruby-core] spec question To: "ksunair at yahoo.com" , "ironruby-core at rubyforge.org" Date: Sunday, July 6, 2008, 4:57 AM I'd say yes. You can commit to your local repository and use "git format-patch -1" to make the patches. Change the number to match the number of commits you have made. In addition, remember that rake spec runs the tests in tests/core, which isn't the same as the git tests. JD -----Original Message----- From: "Unnikrishnan Nair" To: "ironruby-core at rubyforge.org" Sent: 7/5/08 11:07 PM Subject: Re: [Ironruby-core] spec question Jim, ? Now that we moved to GIT, do we submit the patches through GIT? ? Thanks Unni --- On Sat, 7/5/08, Jim Deville wrote: From: Jim Deville Subject: RE: [Ironruby-core] spec question To: "ksunair at yahoo.com" , "ironruby-core at rubyforge.org" Date: Saturday, July 5, 2008, 2:16 PM You should be fine making a patch. As for the test, it shouldn't run that test on Windows, but I'm not sure how the old runner works. JD -----Original Message----- From: "Unnikrishnan Nair" To: "ironruby-core at rubyforge.org" Sent: 7/5/08 1:22 AM Subject: [Ironruby-core] spec question When I was testing the File.zero? I noticed File.zero?('NUL') failed. I looked the code and it seems there is no special check there to exclude NUL as a not a valid file. I created a constant ??????? private readonly static string NUL_VALUE = "NUL"; and added a else clause in the trycreate method? as else if (path.Equals(NUL_VALUE)) ??????????????? { ??????????????????? result = null; ??????????????? } and now the NUL works fine for zero. I have two questions with this regard; 1. Can I apply these changes and push in the patch? 2. When we run the rake spec, will it include following tests on Windows? platform :not, :mswin do ??? it "returns true for /dev/null" do ????? File.zero?('/dev/null').should == true ??? end ? end It seems it does, I haven't looked at the Rubinius to see how the platform is supposed to work. Thanks, unni -------------- next part -------------- An HTML attachment was scrubbed... URL: From curth at microsoft.com Sun Jul 6 21:03:26 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Sun, 6 Jul 2008 18:03:26 -0700 Subject: [Ironruby-core] Memory Leak with Certain Script Exceptions In-Reply-To: <9c209bbd19c929cfe31819fe9e2d3d9e@ruby-forum.com> References: <18dd5199cf9965684be32e7b5c1c92eb@ruby-forum.com> <9c209bbd19c929cfe31819fe9e2d3d9e@ruby-forum.com> Message-ID: Are you continuously evaluating new bits of Ruby code? Is each execution in its own ScriptScope? In its own ScriptRuntime? -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Aaron Clauson Sent: Saturday, July 05, 2008 11:45 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Memory Leak with Certain Script Exceptions Tomas Matousek wrote: > What does the script look like? Hi Tomas, The Ruby script could be a single line such as: sys.DoMethod(1, 2, 3) Where sys is an object from a custom .Net class that has been loaded and executed as: ScriptRuntime scriptRuntime = IronRuby.CreateRuntime(); ScriptScope rubyScope = scriptRuntime.CreateScope("IronRuby"); DialPlanHelper helper = new DialPlanHelper(); rubyScope.SetVariable("sys", helper); rubyScope.Execute("sys.DoMethod(1, 2, 3)"); If the DoMethod is called with the correct parameters it works fine, however if the wrong number of arguments are provided a System.ArgumentOutOfRangeException is thrown and the memory leak occurs. Regards, Aaron -- Posted via http://www.ruby-forum.com/. _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From ksunair at yahoo.com Sun Jul 6 23:51:26 2008 From: ksunair at yahoo.com (Unnikrishnan Nair) Date: Sun, 6 Jul 2008 20:51:26 -0700 (PDT) Subject: [Ironruby-core] spec changes Message-ID: <358328.42608.qm@web50606.mail.re2.yahoo.com> I think the following spec is wrong for FileOps.dirname ? It is ? ??? File.dirname('/////').should == '/' ? but it should be ? ??? File.dirname('/////').should == '//' ? If it is my misunderstanding, please let me know.? ? Thanks, Unni ? ? -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdeville at microsoft.com Mon Jul 7 03:32:15 2008 From: jdeville at microsoft.com (Jim Deville) Date: Mon, 7 Jul 2008 00:32:15 -0700 Subject: [Ironruby-core] spec changes In-Reply-To: <358328.42608.qm@web50606.mail.re2.yahoo.com> References: <358328.42608.qm@web50606.mail.re2.yahoo.com> Message-ID: Well, it's not as simple as that. The spec as written is for *nix paths, where ///// is root. So on one side we should move towards ///// returning the current drive. On the other hand, the notation // represents a UNC file path, which theoretically should resolve to a form of //server/share. I don't know what CRuby wants dirname to return, and I also don't know if we are going to deviate at all. I know that it currently returns //, but if you search for this function, there is some discussion about how it should behave, and I'm not sure how it resolved. For now, I'd suggest turning it into: platform_is_not :windows do File.dirname('/////').should == '/' end platform_is :windows do File.dirname('/////').should == '//' end That will keep it working on *nix. JD From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Unnikrishnan Nair Sent: Sunday, July 06, 2008 8:51 PM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] spec changes I think the following spec is wrong for FileOps.dirname It is File.dirname('/////').should == '/' but it should be File.dirname('/////').should == '//' If it is my misunderstanding, please let me know. Thanks, Unni -------------- next part -------------- An HTML attachment was scrubbed... URL: From lists at ruby-forum.com Mon Jul 7 04:23:13 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Mon, 7 Jul 2008 10:23:13 +0200 Subject: [Ironruby-core] Memory Leak with Certain Script Exceptions In-Reply-To: References: <18dd5199cf9965684be32e7b5c1c92eb@ruby-forum.com> <9c209bbd19c929cfe31819fe9e2d3d9e@ruby-forum.com> Message-ID: Curt Hagenlocher wrote: > Are you continuously evaluating new bits of Ruby code? Is each > execution in its own ScriptScope? In its own ScriptRuntime? Yes pretty much. I have a bunch of users who can all have their own little bit of script (the scripts are used as a dial plan on a SIP Proxy server). At the moment a new runtime and scope is generated on for each call received so the code per call is the equivalent of: ScriptScope rubyScope = null; try { ScriptRuntime scriptRuntime = IronRuby.CreateRuntime(); rubyScope = scriptRuntime.CreateScope("IronRuby"); DialPlanHelper helper = new DialPlanHelper(); rubyScope.SetVariable("sys", helper); rubyScope.Execute(m_dialPlanScript); } finally { rubyScope.RemoveVariable("sys"); rubyScope = null; } I previously has the ScriptRuntime as a static object common across all users thinking that that would make things quicker but since I've had the porblems with memory leaking I've swapped that to being a local variable as well. The memory leak definitely only happens when the specific exceptions are thrown. I can run 500 scripts containing an exception and increase the process memory by 50MB. Running 500 scripts that don't generate an exception does not budge the process memory. Thanks, Aaron -- Posted via http://www.ruby-forum.com/. From ksunair at yahoo.com Mon Jul 7 10:17:50 2008 From: ksunair at yahoo.com (Unnikrishnan Nair) Date: Mon, 7 Jul 2008 07:17:50 -0700 (PDT) Subject: [Ironruby-core] spec changes In-Reply-To: Message-ID: <345207.78400.qm@web50608.mail.re2.yahoo.com> Thanks Jim, I have a test program that I run in windows with all the spec against ruby and against IR. It then compare the results and there is where I found this problem. yes, you are correct with your change. I think we can go with the change suggested. By the way, did anyone compiled Ironruby on Mono? I finally got the mono installed and would like to try to compile it to run the spec on Mac this week. Do you guys consider Mac as *nix environment? Thanks, Unni --- On Mon, 7/7/08, Jim Deville wrote: From: Jim Deville Subject: RE: [Ironruby-core] spec changes To: "ksunair at yahoo.com" , "ironruby-core at rubyforge.org" Date: Monday, July 7, 2008, 1:32 AM Well, it?s not as simple as that. The spec as written is for *nix paths, where ///// is root. So on one side we should move towards ///// returning the current drive. On the other hand, the notation // represents a UNC file path, which theoretically should resolve to a form of //server/share. I don?t know what CRuby wants dirname to return, and I also don?t know if we are going to deviate at all. I know that it currently returns //, but if you search for this function, there is some discussion about how it should behave, and I?m not sure how it resolved. For now, I?d suggest turning it into: ? ??????????????? platform_is_not :windows do ??????????????????????????????? File.dirname(?/////?).should == ?/? ??????????????? end ??????????????? platform_is :windows do ??????????????????????????????? File.dirname(?/////?).should == ?//? ??????????????? end ? That will keep it working on *nix. ? JD ? From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Unnikrishnan Nair Sent: Sunday, July 06, 2008 8:51 PM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] spec changes ? I think the following spec is wrong for FileOps.dirname ? It is ? ??? File.dirname('/////').should == '/' ? but it should be ? ??? File.dirname('/////').should == '//' ? If it is my misunderstanding, please let me know.? ? Thanks, Unni ? ? ? -------------- next part -------------- An HTML attachment was scrubbed... URL: From robert.brotherus at napa.fi Mon Jul 7 10:11:27 2008 From: robert.brotherus at napa.fi (Robert Brotherus) Date: Mon, 7 Jul 2008 17:11:27 +0300 Subject: [Ironruby-core] IR SVN121 Strong Name validation failed References: Message-ID: <53194650933664488F3F0C3A0AB568326A2694@nw60.napa.fi> (I will make bug-tracker issue of this, if it turns out not to be aspect of my environment) I have been using IR113 for many weeks now and now did the jump to IR121, causing IR.exe to stop working: 1. Get IR SVN121 from SVN 2. Load solution with VisualStudio2008 and compile (compiles without errors) 3. Start debugging with Ruby.Console project as active project System.IO.FileLoadException was unhandled Message: Could not load file or assembly 'ir, Version=1.0.0.0, Culture=neutral, PublicKeyToken=31bf3856ad364e35' or one of its dependencies. Strong name validation failed. (Exception from HRESULT: 0x8013141A) Same error when starting ir.exe from command prompt. With IR SVN113 and earlier ie.exe seems to start normally. I am running on Vista-64 (dotnet x64) Robert Brotherus Software architect Napa Ltd Tammasaarenkatu 3, Helsinki FI-00180 P.O.Box 470, Helsinki FI-00181 Tel. +358 9 22 813 1 Direct. +358 9 22 813 611 GSM +358 45 11 456 02 Fax. +358 9 22 813 800 Email: Robert.Brotherus at napa.fi www.napa.fi -------------- next part -------------- An HTML attachment was scrubbed... URL: From ksunair at yahoo.com Mon Jul 7 10:32:25 2008 From: ksunair at yahoo.com (Unnikrishnan Nair) Date: Mon, 7 Jul 2008 07:32:25 -0700 (PDT) Subject: [Ironruby-core] IR SVN121 Strong Name validation failed In-Reply-To: <53194650933664488F3F0C3A0AB568326A2694@nw60.napa.fi> Message-ID: <704171.60896.qm@web50607.mail.re2.yahoo.com> Just curious, Did you run runfirst.cmd at svn\trunk before you did rake compile? Thanks, Unni --- On Mon, 7/7/08, Robert Brotherus wrote: From: Robert Brotherus Subject: [Ironruby-core] IR SVN121 Strong Name validation failed To: ironruby-core at rubyforge.org Date: Monday, July 7, 2008, 8:11 AM (I?will make bug-tracker issue of this, if it turns out not to be aspect of my environment) ? I have been using IR113 for many weeks now and now did the jump to IR121, causing IR.exe to stop working: ? 1. Get IR SVN121 from SVN 2. Load solution with VisualStudio2008 and compile (compiles without errors) 3. Start debugging with Ruby.Console project as active project ? System.IO.FileLoadException was unhandled Message: Could not load file or assembly 'ir, Version=1.0.0.0, Culture=neutral, PublicKeyToken=31bf3856ad364e35' or one of its dependencies. Strong name validation failed. (Exception from HRESULT: 0x8013141A) Same error when starting ir.exe from command prompt. ? With IR SVN113 and earlier ie.exe seems to start normally. ? I am running on Vista-64 (dotnet x64) ? Robert Brotherus Software architect Napa Ltd Tammasaarenkatu 3, Helsinki FI-00180 P.O.Box 470, Helsinki FI-00181 Tel. +358 9 22 813 1 Direct. +358 9 22 813 611 GSM +358 45 11?456 02 Fax. +358 9 22 813 800 Email: Robert.Brotherus at napa.fi www.napa.fi ?_______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core -------------- next part -------------- An HTML attachment was scrubbed... URL: From robert.brotherus at napa.fi Mon Jul 7 10:50:50 2008 From: robert.brotherus at napa.fi (Robert Brotherus) Date: Mon, 7 Jul 2008 17:50:50 +0300 Subject: [Ironruby-core] IR SVN121 Strong Name validation failed References: <704171.60896.qm@web50607.mail.re2.yahoo.com> Message-ID: <53194650933664488F3F0C3A0AB568326A269A@nw60.napa.fi> I did not do rake compile at all (nor runfirst.cmd). I have so far done all my compilations on VisualStudio2008 and had the impression that that is a valid alternative to rake compile. At least has been in case of IR SVN113. I now have run runfirst.cmd and it fixed the problem, thanks! (Still using VisualStudio2008, not rake) Robert Brotherus _____ From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Unnikrishnan Nair Sent: Monday, July 07, 2008 5:32 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] IR SVN121 Strong Name validation failed Just curious, Did you run runfirst.cmd at svn\trunk before you did rake compile? Thanks, Unni --- On Mon, 7/7/08, Robert Brotherus wrote: From: Robert Brotherus Subject: [Ironruby-core] IR SVN121 Strong Name validation failed To: ironruby-core at rubyforge.org Date: Monday, July 7, 2008, 8:11 AM (I will make bug-tracker issue of this, if it turns out not to be aspect of my environment) I have been using IR113 for many weeks now and now did the jump to IR121, causing IR.exe to stop working: 1. Get IR SVN121 from SVN 2. Load solution with VisualStudio2008 and compile (compiles without errors) 3. Start debugging with Ruby.Console project as active project System.IO.FileLoadException was unhandled Message: Could not load file or assembly 'ir, Version=1.0.0.0, Culture=neutral, PublicKeyToken=31bf3856ad364e35' or one of its dependencies. Strong name validation failed. (Exception from HRESULT: 0x8013141A) Same error when starting ir.exe from command prompt. With IR SVN113 and earlier ie.exe seems to start normally. I am running on Vista-64 (dotnet x64) Robert Brotherus Software architect Napa Ltd Tammasaarenkatu 3, Helsinki FI-00180 P.O.Box 470, Helsinki FI-00181 Tel. +358 9 22 813 1 Direct. +358 9 22 813 611 GSM +358 45 11 456 02 Fax. +358 9 22 813 800 Email: Robert.Brotherus at napa.fi www.napa.fi _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core -------------- next part -------------- An HTML attachment was scrubbed... URL: From michael.letterle at gmail.com Mon Jul 7 10:52:52 2008 From: michael.letterle at gmail.com (Michael Letterle) Date: Mon, 7 Jul 2008 10:52:52 -0400 Subject: [Ironruby-core] IR SVN121 Strong Name validation failed In-Reply-To: <53194650933664488F3F0C3A0AB568326A269A@nw60.napa.fi> References: <704171.60896.qm@web50607.mail.re2.yahoo.com> <53194650933664488F3F0C3A0AB568326A269A@nw60.napa.fi> Message-ID: runfirst.cmd registers the test key that is now included in the solution for Strong Name Verification Skipping, it doesn't matter if you're doing a rake compile or a VS compile. On Mon, Jul 7, 2008 at 10:50 AM, Robert Brotherus wrote: > I did not do *rake compile* at all (nor runfirst.cmd). > > I have so far done all my compilations on VisualStudio2008 and had the > impression that that is a valid alternative to rake compile. At least has > been in case of IR SVN113. > > I now have run runfirst.cmd and it fixed the problem, thanks! (Still using > VisualStudio2008, not rake) > > Robert Brotherus > > ------------------------------ > *From:* ironruby-core-bounces at rubyforge.org [mailto: > ironruby-core-bounces at rubyforge.org] *On Behalf Of *Unnikrishnan Nair > *Sent:* Monday, July 07, 2008 5:32 PM > *To:* ironruby-core at rubyforge.org > *Subject:* Re: [Ironruby-core] IR SVN121 Strong Name validation failed > > Just curious, Did you run runfirst.cmd at svn\trunk before you did rake > compile? > > Thanks, > > Unni > > --- On *Mon, 7/7/08, Robert Brotherus * wrote: > > From: Robert Brotherus > Subject: [Ironruby-core] IR SVN121 Strong Name validation failed > To: ironruby-core at rubyforge.org > Date: Monday, July 7, 2008, 8:11 AM > > (I will make bug-tracker issue of this, if it turns out not to be aspect > of my environment) > > I have been using IR113 for many weeks now and now did the jump to IR121, > causing IR.exe to stop working: > > 1. Get IR SVN121 from SVN > 2. Load solution with VisualStudio2008 and compile (compiles without > errors) > 3. Start debugging with Ruby.Console project as active project > > *System.IO.FileLoadException was unhandled > Message: Could not load file or assembly 'ir, Version=1.0.0.0, > Culture=neutral, PublicKeyToken=31bf3856ad364e35' or one of its > dependencies. Strong name validation failed. (Exception from HRESULT: > 0x8013141A) > * > Same error when starting ir.exe from command prompt. > > With IR SVN113 and earlier ie.exe seems to start normally. > > I am running on Vista-64 (dotnet x64) > > *Robert Brotherus > *Software architect > *Napa Ltd > *Tammasaarenkatu 3, Helsinki FI-00180 > P.O.Box 470, Helsinki FI-00181 > > Tel. +358 9 22 813 1 > Direct. +358 9 22 813 611 > GSM +358 45 11 456 02 > Fax. +358 9 22 813 800 > > Email: *Robert.Brotherus at napa.fi* > *www.napa.fi* > > > _______________________________________________ > Ironruby-core mailing listIronruby-core at rubyforge.orghttp://rubyforge.org/mailman/listinfo/ironruby-core > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > -- Michael Letterle [Polymath Prokrammer] http://blog.prokrams.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From jirapong.nanta at gmail.com Mon Jul 7 12:16:50 2008 From: jirapong.nanta at gmail.com (Jirapong Nanta) Date: Mon, 7 Jul 2008 23:16:50 +0700 Subject: [Ironruby-core] spec changes In-Reply-To: <345207.78400.qm@web50608.mail.re2.yahoo.com> References: <345207.78400.qm@web50608.mail.re2.yahoo.com> Message-ID: Hi Unni, Did you check this out? - http://sparcs.kaist.ac.kr/~tinuviel/download/IronRuby/ - http://flanders.co.nz/2008/02/05/compiling-mono-and-ironruby-on-osx-leopard/ Hope this helps, -Jirapong 2008/7/7 Unnikrishnan Nair : > Thanks Jim, I have a test program that I run in windows with all the spec > against ruby and against IR. It then compare the results and there is where > I found this problem. yes, you are correct with your change. I think we can > go with the change suggested. > > By the way, did anyone compiled Ironruby on Mono? I finally got the mono > installed and would like to try to compile it to run the spec on Mac this > week. Do you guys consider Mac as *nix environment? > > Thanks, > > Unni > > --- On *Mon, 7/7/08, Jim Deville * wrote: > > From: Jim Deville > Subject: RE: [Ironruby-core] spec changes > To: "ksunair at yahoo.com" , "ironruby-core at rubyforge.org" > > Date: Monday, July 7, 2008, 1:32 AM > > > Well, it's not as simple as that. The spec as written is for *nix paths, > where ///// is root. So on one side we should move towards ///// returning > the current drive. On the other hand, the notation // represents a UNC file > path, which theoretically should resolve to a form of //server/share. I > don't know what CRuby wants dirname to return, and I also don't know if we > are going to deviate at all. I know that it currently returns //, but if you > search for this function, there is some discussion about how it should > behave, and I'm not sure how it resolved. For now, I'd suggest turning it > into: > > > > platform_is_not :windows do > > File.dirname('/////').should == '/' > > end > > platform_is :windows do > > File.dirname('/////').should == '//' > > end > > > > That will keep it working on *nix. > > > > JD > > > > *From:* ironruby-core-bounces at rubyforge.org [mailto: > ironruby-core-bounces at rubyforge.org] *On Behalf Of *Unnikrishnan Nair > *Sent:* Sunday, July 06, 2008 8:51 PM > *To:* ironruby-core at rubyforge.org > *Subject:* [Ironruby-core] spec changes > > > > I think the following spec is wrong for FileOps.dirname > > > > It is > > > > File.dirname('/////').should == '/' > > > > but it should be > > > > File.dirname('/////').should == '//' > > > > If it is my misunderstanding, please let me know. > > > > Thanks, > > Unni > > > > > > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jflam at microsoft.com Mon Jul 7 13:46:50 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Mon, 7 Jul 2008 10:46:50 -0700 Subject: [Ironruby-core] mspec breaking change Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD4004121@NA-EXMSG-C115.redmond.corp.microsoft.com> The mspec team checked in a change on June 30 that broke all of our RubySpec tests. The -X option has been removed which is the option that lets us tell the specs where the tags directory is. I've spent the morning figuring out how to work around this and have something more or less working now modulo a few new bugs that have now shown up. Thanks, -John -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdeville at microsoft.com Mon Jul 7 13:46:59 2008 From: jdeville at microsoft.com (Jim Deville) Date: Mon, 7 Jul 2008 10:46:59 -0700 Subject: [Ironruby-core] spec changes In-Reply-To: <345207.78400.qm@web50608.mail.re2.yahoo.com> References: <345207.78400.qm@web50608.mail.re2.yahoo.com> Message-ID: Glad the spec stuff worked. I have tried to compile IR on Mono, but there are some patches that need to be applied first, I'm not sure where those patches are. I also ran into a problem on Mac with filename and folder casing not matching the csproj file. I would consider Mac as a *nix platform. It has all the stuff I look for (mainly bash, and the UNIX system underneath. JD From: Unnikrishnan Nair [mailto:ksunair at yahoo.com] Sent: Monday, July 07, 2008 7:18 AM To: ironruby-core at rubyforge.org; Jim Deville Subject: RE: [Ironruby-core] spec changes Thanks Jim, I have a test program that I run in windows with all the spec against ruby and against IR. It then compare the results and there is where I found this problem. yes, you are correct with your change. I think we can go with the change suggested. By the way, did anyone compiled Ironruby on Mono? I finally got the mono installed and would like to try to compile it to run the spec on Mac this week. Do you guys consider Mac as *nix environment? Thanks, Unni --- On Mon, 7/7/08, Jim Deville wrote: From: Jim Deville Subject: RE: [Ironruby-core] spec changes To: "ksunair at yahoo.com" , "ironruby-core at rubyforge.org" Date: Monday, July 7, 2008, 1:32 AM Well, it?s not as simple as that. The spec as written is for *nix paths, where ///// is root. So on one side we should move towards ///// returning the current drive. On the other hand, the notation // represents a UNC file path, which theoretically should resolve to a form of //server/share. I don?t know what CRuby wants dirname to return, and I also don?t know if we are going to deviate at all. I know that it currently returns //, but if you search for this function, there is some discussion about how it should behave, and I?m not sure how it resolved. For now, I?d suggest turning it into: platform_is_not :windows do File.dirname(?/////?).should == ?/? end platform_is :windows do File.dirname(?/////?).should == ?//? end That will keep it working on *nix. JD From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Unnikrishnan Nair Sent: Sunday, July 06, 2008 8:51 PM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] spec changes I think the following spec is wrong for FileOps.dirname It is File.dirname('/////').should == '/' but it should be File.dirname('/////').should == '//' If it is my misunderstanding, please let me know. Thanks, Unni -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdeville at microsoft.com Mon Jul 7 13:50:51 2008 From: jdeville at microsoft.com (Jim Deville) Date: Mon, 7 Jul 2008 10:50:51 -0700 Subject: [Ironruby-core] mspec breaking change In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD4004121@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD4004121@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: Sorry for not dealing with this earlier. I saw this after I committed on Friday. I wanted to get to it this weekend, but I've been dealing with a sick kid. I was planning on doing it first thing. As a side note, does it seem right to ask why this wasn't announced on the Rubyspec list? It's a pretty big change in my opinion that should have been discussed on the mailing list. JD From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: Monday, July 07, 2008 10:47 AM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] mspec breaking change The mspec team checked in a change on June 30 that broke all of our RubySpec tests. The -X option has been removed which is the option that lets us tell the specs where the tags directory is. I've spent the morning figuring out how to work around this and have something more or less working now modulo a few new bugs that have now shown up. Thanks, -John -------------- next part -------------- An HTML attachment was scrubbed... URL: From jflam at microsoft.com Mon Jul 7 13:55:26 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Mon, 7 Jul 2008 10:55:26 -0700 Subject: [Ironruby-core] mspec breaking change In-Reply-To: References: <372109E149E8084D8E6C7D9CFD82E0632FD4004121@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD4004134@NA-EXMSG-C115.redmond.corp.microsoft.com> Jim Deville: > As a side note, does it seem right to ask why this wasn't announced on > the Rubyspec list? It's a pretty big change in my opinion that should > have been discussed on the mailing list. Jim - can you act as our liason to the mspec team? Thanks, -John From ksunair at yahoo.com Mon Jul 7 14:05:44 2008 From: ksunair at yahoo.com (Unnikrishnan Nair) Date: Mon, 7 Jul 2008 11:05:44 -0700 (PDT) Subject: [Ironruby-core] spec changes In-Reply-To: Message-ID: <537238.43146.qm@web50609.mail.re2.yahoo.com> Thanks, just to make sure what ever? changes I make I want to test it both in Windows and *nix world. Unni --- On Mon, 7/7/08, Jim Deville wrote: From: Jim Deville Subject: RE: [Ironruby-core] spec changes To: "ksunair at yahoo.com" , "ironruby-core at rubyforge.org" Date: Monday, July 7, 2008, 11:46 AM Glad the spec stuff worked. ? I have tried to compile IR on Mono, but there are some patches that need to be applied first, I?m not sure where those patches are. I also ran into a problem on Mac with filename and folder casing not matching the csproj file. I would consider Mac as a *nix platform. It has all the stuff I look for (mainly bash, and the UNIX system underneath. ? JD ? From: Unnikrishnan Nair [mailto:ksunair at yahoo.com] Sent: Monday, July 07, 2008 7:18 AM To: ironruby-core at rubyforge.org; Jim Deville Subject: RE: [Ironruby-core] spec changes ? Thanks Jim, I have a test program that I run in windows with all the spec against ruby and against IR. It then compare the results and there is where I found this problem. yes, you are correct with your change. I think we can go with the change suggested. By the way, did anyone compiled Ironruby on Mono? I finally got the mono installed and would like to try to compile it to run the spec on Mac this week. Do you guys consider Mac as *nix environment? Thanks, Unni --- On Mon, 7/7/08, Jim Deville wrote: From: Jim Deville Subject: RE: [Ironruby-core] spec changes To: "ksunair at yahoo.com" , "ironruby-core at rubyforge.org" Date: Monday, July 7, 2008, 1:32 AM Well, it?s not as simple as that. The spec as written is for *nix paths, where ///// is root. So on one side we should move towards ///// returning the current drive. On the other hand, the notation // represents a UNC file path, which theoretically should resolve to a form of //server/share. I don?t know what CRuby wants dirname to return, and I also don?t know if we are going to deviate at all. I know that it currently returns //, but if you search for this function, there is some discussion about how it should behave, and I?m not sure how it resolved. For now, I?d suggest turning it into: ? ??????????????? platform_is_not :windows do ??????????????????????????????? File.dirname(?/////?).should == ?/? ??????????????? end ??????????????? platform_is :windows do ??????????????????????????????? File.dirname(?/////?).should == ?//? ??????????????? end ? That will keep it working on *nix. ? JD ? From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Unnikrishnan Nair Sent: Sunday, July 06, 2008 8:51 PM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] spec changes ? I think the following spec is wrong for FileOps.dirname ? It is ? ??? File.dirname('/////').should == '/' ? but it should be ? ??? File.dirname('/////').should == '//' ? If it is my misunderstanding, please let me know.? ? Thanks, Unni ? ? ? ? -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdeville at microsoft.com Mon Jul 7 14:13:09 2008 From: jdeville at microsoft.com (Jim Deville) Date: Mon, 7 Jul 2008 11:13:09 -0700 Subject: [Ironruby-core] mspec breaking change In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD4004134@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD4004121@NA-EXMSG-C115.redmond.corp.microsoft.com> <372109E149E8084D8E6C7D9CFD82E0632FD4004134@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: Sure. I can do it. I just wanted to make sure that I wasn't over-reacting. JD > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) > Sent: Monday, July 07, 2008 10:55 AM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] mspec breaking change > > Jim Deville: > > > As a side note, does it seem right to ask why this wasn't announced > on > > the Rubyspec list? It's a pretty big change in my opinion that should > > have been discussed on the mailing list. > > Jim - can you act as our liason to the mspec team? > > Thanks, > -John > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core From jdeville at microsoft.com Mon Jul 7 14:23:59 2008 From: jdeville at microsoft.com (Jim Deville) Date: Mon, 7 Jul 2008 11:23:59 -0700 Subject: [Ironruby-core] Ruby as a standard Message-ID: http://www.infoq.com/news/2008/07/rubykaigi "Matz plans to hand in the standard to the ISO (International Organization for Standardization), however, a concrete date has not yet been determined-only that it will "at least take a couple of years for standardization"." So this should make the spec issue easier. JD -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdeville at microsoft.com Mon Jul 7 15:44:59 2008 From: jdeville at microsoft.com (Jim Deville) Date: Mon, 7 Jul 2008 12:44:59 -0700 Subject: [Ironruby-core] mspec breaking change In-Reply-To: References: <372109E149E8084D8E6C7D9CFD82E0632FD4004121@NA-EXMSG-C115.redmond.corp.microsoft.com> <372109E149E8084D8E6C7D9CFD82E0632FD4004134@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: Have you commited the stuff to make it run yet? JD > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of Jim Deville > Sent: Monday, July 07, 2008 11:13 AM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] mspec breaking change > > Sure. I can do it. I just wanted to make sure that I wasn't over- > reacting. > > JD > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) > > Sent: Monday, July 07, 2008 10:55 AM > > To: ironruby-core at rubyforge.org > > Subject: Re: [Ironruby-core] mspec breaking change > > > > Jim Deville: > > > > > As a side note, does it seem right to ask why this wasn't announced > > on > > > the Rubyspec list? It's a pretty big change in my opinion that > should > > > have been discussed on the mailing list. > > > > Jim - can you act as our liason to the mspec team? > > > > Thanks, > > -John > > > > _______________________________________________ > > Ironruby-core mailing list > > Ironruby-core at rubyforge.org > > http://rubyforge.org/mailman/listinfo/ironruby-core > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core From michael.letterle at gmail.com Mon Jul 7 16:21:26 2008 From: michael.letterle at gmail.com (Michael Letterle) Date: Mon, 7 Jul 2008 16:21:26 -0400 Subject: [Ironruby-core] mspec breaking change In-Reply-To: References: <372109E149E8084D8E6C7D9CFD82E0632FD4004121@NA-EXMSG-C115.redmond.corp.microsoft.com> <372109E149E8084D8E6C7D9CFD82E0632FD4004134@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: Just a reminder that the rubyspec guys are on IRC, irc.freenode.net/#rubyspec, might be better to hop in there and ask what's up before putting it out on a public mailing list. On Mon, Jul 7, 2008 at 3:44 PM, Jim Deville wrote: > Have you commited the stuff to make it run yet? > > > JD > > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > bounces at rubyforge.org] On Behalf Of Jim Deville > > Sent: Monday, July 07, 2008 11:13 AM > > To: ironruby-core at rubyforge.org > > Subject: Re: [Ironruby-core] mspec breaking change > > > > Sure. I can do it. I just wanted to make sure that I wasn't over- > > reacting. > > > > JD > > > > > -----Original Message----- > > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > > bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) > > > Sent: Monday, July 07, 2008 10:55 AM > > > To: ironruby-core at rubyforge.org > > > Subject: Re: [Ironruby-core] mspec breaking change > > > > > > Jim Deville: > > > > > > > As a side note, does it seem right to ask why this wasn't announced > > > on > > > > the Rubyspec list? It's a pretty big change in my opinion that > > should > > > > have been discussed on the mailing list. > > > > > > Jim - can you act as our liason to the mspec team? > > > > > > Thanks, > > > -John > > > > > > _______________________________________________ > > > Ironruby-core mailing list > > > Ironruby-core at rubyforge.org > > > http://rubyforge.org/mailman/listinfo/ironruby-core > > > > _______________________________________________ > > Ironruby-core mailing list > > Ironruby-core at rubyforge.org > > http://rubyforge.org/mailman/listinfo/ironruby-core > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > -- Michael Letterle [Polymath Prokrammer] http://blog.prokrams.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdeville at microsoft.com Mon Jul 7 16:25:46 2008 From: jdeville at microsoft.com (Jim Deville) Date: Mon, 7 Jul 2008 13:25:46 -0700 Subject: [Ironruby-core] mspec breaking change In-Reply-To: References: <372109E149E8084D8E6C7D9CFD82E0632FD4004121@NA-EXMSG-C115.redmond.corp.microsoft.com> <372109E149E8084D8E6C7D9CFD82E0632FD4004134@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: Already brought it up on the rubyspec mailing list. I avoid IRC because I get too distracted. JD From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Michael Letterle Sent: Monday, July 07, 2008 1:21 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] mspec breaking change Just a reminder that the rubyspec guys are on IRC, irc.freenode.net/#rubyspec, might be better to hop in there and ask what's up before putting it out on a public mailing list. On Mon, Jul 7, 2008 at 3:44 PM, Jim Deville > wrote: Have you commited the stuff to make it run yet? JD > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of Jim Deville > Sent: Monday, July 07, 2008 11:13 AM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] mspec breaking change > > Sure. I can do it. I just wanted to make sure that I wasn't over- > reacting. > > JD > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) > > Sent: Monday, July 07, 2008 10:55 AM > > To: ironruby-core at rubyforge.org > > Subject: Re: [Ironruby-core] mspec breaking change > > > > Jim Deville: > > > > > As a side note, does it seem right to ask why this wasn't announced > > on > > > the Rubyspec list? It's a pretty big change in my opinion that > should > > > have been discussed on the mailing list. > > > > Jim - can you act as our liason to the mspec team? > > > > Thanks, > > -John > > > > _______________________________________________ > > Ironruby-core mailing list > > Ironruby-core at rubyforge.org > > http://rubyforge.org/mailman/listinfo/ironruby-core > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core -- Michael Letterle [Polymath Prokrammer] http://blog.prokrams.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From olegtk at microsoft.com Mon Jul 7 20:14:19 2008 From: olegtk at microsoft.com (Oleg Tkachenko) Date: Mon, 7 Jul 2008 17:14:19 -0700 Subject: [Ironruby-core] Code Review: MoreYamlStuff Message-ID: tfpt review "/shelveset:MoreYamlStuff;REDMOND\olegtk" Comment : Implements taguri() for regexps and structs. Moves accessor for code context to IConstructor level so all static constructors receiving IConstructor can use it; and makes it getter method for interface consistency. Implements loading of ranges expressed using Ruby native range syntax (a..b|a...b) Implements loading regular expressions. Implements loading structs. Adds RubyUtils.CreateRubyObject(context, RubyClass, attribute hash) method, which deserializes Ruby object from a mapping containing attributre values. Implements loading objects of an arbitrary (but known) class. -- Oleg -------------- next part -------------- A non-text attachment was scrubbed... Name: MoreYamlStuff.diff Type: application/octet-stream Size: 21038 bytes Desc: MoreYamlStuff.diff URL: From dremy at microsoft.com Mon Jul 7 20:29:21 2008 From: dremy at microsoft.com (Dave Remy) Date: Mon, 7 Jul 2008 17:29:21 -0700 Subject: [Ironruby-core] Code Review: StringGsubEach Message-ID: tfpt review "/shelveset:StringGsubEach;REDMOND\dremy" Comment : Changes to for string.gsub and string.each (and each_line) to run clean. The most significant change is to track version in MutableString. This version is bumped on any mutation and functions that iterate (each, each_line, gsub) check the version before and after to assure there has been no version change during the iteration. After these changes all gsub, each, and each_line specs run clean (no excludes). Note that although the specs run clean the each behavior does not match MRI. The spec test contains a new line in the iterating string ("hello\nworld") and MRI does throw a runtime exception if this string is iterated. However if there is no new line in the string MRI does not throw an exception if the underlying string is mutated. This seems inconsistent but worth noting. String#gsub with pattern and block sets $~ for access from the block String#gsub with pattern and block restores $~ after leaving the block String#gsub with pattern and block sets $~ to MatchData of last match and nil when there's none for access from outside String#gsub with pattern and block raises a RuntimeError if the string is modified while substituting String#each raises a RuntimeError if the string is modified while substituting String#each_line raises a RuntimeError if the string is modified while substituting -------------- next part -------------- A non-text attachment was scrubbed... Name: StringGsubEach.diff Type: application/octet-stream Size: 19089 bytes Desc: StringGsubEach.diff URL: From ksunair at yahoo.com Tue Jul 8 01:21:49 2008 From: ksunair at yahoo.com (Unnikrishnan Nair) Date: Mon, 7 Jul 2008 22:21:49 -0700 (PDT) Subject: [Ironruby-core] Mono build update.. Message-ID: <293714.17190.qm@web50606.mail.re2.yahoo.com> Finally I got mono to build successfully.I checkout 121 from rubyforge and tried to do rake compile and, it failed on the following line; path_to_config = ENV['USERPROFILE'] + '/.irconfig.rb' I remember seeing some mail about Home and Userprofile setting from John and others sometime back I am not sure it is related to the same. Regardless, I thought I will send this to the group so that if someone already came across this problem, they could help me without spending more of my time. If not, I will look into it tomorrow morning. Thanks, Unni From curth at microsoft.com Tue Jul 8 01:59:59 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Mon, 7 Jul 2008 22:59:59 -0700 Subject: [Ironruby-core] Code Review: MoreYamlStuff In-Reply-To: References: Message-ID: Looks good. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Oleg Tkachenko Sent: Monday, July 07, 2008 5:14 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: MoreYamlStuff tfpt review "/shelveset:MoreYamlStuff;REDMOND\olegtk" Comment : Implements taguri() for regexps and structs. Moves accessor for code context to IConstructor level so all static constructors receiving IConstructor can use it; and makes it getter method for interface consistency. Implements loading of ranges expressed using Ruby native range syntax (a..b|a...b) Implements loading regular expressions. Implements loading structs. Adds RubyUtils.CreateRubyObject(context, RubyClass, attribute hash) method, which deserializes Ruby object from a mapping containing attributre values. Implements loading objects of an arbitrary (but known) class. -- Oleg From curth at microsoft.com Tue Jul 8 02:01:04 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Mon, 7 Jul 2008 23:01:04 -0700 Subject: [Ironruby-core] Code Review: StringGsubEach In-Reply-To: References: Message-ID: Looks good. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Dave Remy Sent: Monday, July 07, 2008 5:29 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: StringGsubEach tfpt review "/shelveset:StringGsubEach;REDMOND\dremy" Comment : Changes to for string.gsub and string.each (and each_line) to run clean. The most significant change is to track version in MutableString. This version is bumped on any mutation and functions that iterate (each, each_line, gsub) check the version before and after to assure there has been no version change during the iteration. After these changes all gsub, each, and each_line specs run clean (no excludes). Note that although the specs run clean the each behavior does not match MRI. The spec test contains a new line in the iterating string ("hello\nworld") and MRI does throw a runtime exception if this string is iterated. However if there is no new line in the string MRI does not throw an exception if the underlying string is mutated. This seems inconsistent but worth noting. String#gsub with pattern and block sets $~ for access from the block String#gsub with pattern and block restores $~ after leaving the block String#gsub with pattern and block sets $~ to MatchData of last match and nil when there's none for access from outside String#gsub with pattern and block raises a RuntimeError if the string is modified while substituting String#each raises a RuntimeError if the string is modified while substituting String#each_line raises a RuntimeError if the string is modified while substituting From raymond.roestenburg at gmail.com Tue Jul 8 03:31:19 2008 From: raymond.roestenburg at gmail.com (Raymond Roestenburg) Date: Tue, 8 Jul 2008 09:31:19 +0200 Subject: [Ironruby-core] newbie questions Message-ID: <729ba3df0807080031g60d21470mf769998db255fdb8@mail.gmail.com> Hi, Just some newbie questions: I have seen quite a few tftp shelvesets coming by in the mailing list. Is this shelvesets in TFS server, is that publicly available or is this only for the core team? At what times is this replicated to the svn server? Kind Regards Raymond Roestenburg -------------- next part -------------- An HTML attachment was scrubbed... URL: From sanxiyn at gmail.com Tue Jul 8 04:54:52 2008 From: sanxiyn at gmail.com (Sanghyeon Seo) Date: Tue, 8 Jul 2008 17:54:52 +0900 Subject: [Ironruby-core] newbie questions In-Reply-To: <729ba3df0807080031g60d21470mf769998db255fdb8@mail.gmail.com> References: <729ba3df0807080031g60d21470mf769998db255fdb8@mail.gmail.com> Message-ID: <5b0248170807080154s63e3e004o2faca66bfdf1ab38@mail.gmail.com> 2008/7/8 Raymond Roestenburg : > I have seen quite a few tftp shelvesets coming by in the mailing list. Is > this shelvesets in TFS server, is that publicly available or is this only > for the core team? It is not availble to the public. Patch files are attached to review mails though. > At what times is this replicated to the svn server? When they are ready. -- Seo Sanghyeon From robert.brotherus at napa.fi Tue Jul 8 12:04:19 2008 From: robert.brotherus at napa.fi (Robert Brotherus) Date: Tue, 8 Jul 2008 19:04:19 +0300 Subject: [Ironruby-core] Duplicate definition of Action & other classes References: Message-ID: <53194650933664488F3F0C3A0AB568326A2776@nw60.napa.fi> In my ongoing switch from IR SVN 113 to IR SVN 121, another issue has emerged. It seems that multiple new classes have been added to Microsoft.Scripting.Core.dll that have identical name with corresponding classes in System.Core.dll (v. 3.5). Below is my code (that is complicing and running ok against IR SVN 113) and the build errors against IR SVN 121. I can rather easily make a work-around for this case but still wondering if it is possible to make the code work as it is now and why such duplicate classes have been added? Robert Brotherus Software architect Napa Ltd Email: Robert.Brotherus at napa.fi www.napa.fi ------------------------------------------------------------------ public T ExecuteStr(string code) { return Engine.CreateScriptSourceFromString(code).Execute(Engine.CreateScope( )); } public Action RunTests // scriptDir, testMethod <-- ERROR CS0433 HERE { get { ExecuteFile("Test/TestResults.rb"); return ExecuteStr>("TestRunner.method(:run)"); } } ... ScriptService.SharedService.RunTests(ScriptService.ScriptDir, app.ArgValue("test")); ... ------ Build started: Project: gui, Configuration: Debug Any CPU ------ warning CS1685: The predefined type 'System.Linq.Expressions.Expression' is defined in multiple assemblies in the global alias; using definition from 'c:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll' warning CS1685: The predefined type 'System.Linq.Expressions.Expression' is defined in multiple assemblies in the global alias; using definition from 'c:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll' ... warning CS1685: The predefined type 'System.Linq.Expressions.ElementInit' is defined in multiple assemblies in the global alias; using definition from 'c:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll' C:\work\napa.net\gui\ScriptService.cs(113,16): error CS0433: The type 'System.Action' exists in both 'c:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll' and 'c:\programs\IronRuby\trunk\build\debug\svn121\Microsoft.Scripting.Core. dll' c:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll: (Related file) c:\programs\IronRuby\trunk\build\debug\svn121\Microsoft.Scripting.Core.d ll: (Related file) Compile complete -- 1 errors, 21 warnings From curth at microsoft.com Tue Jul 8 12:34:31 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Tue, 8 Jul 2008 09:34:31 -0700 Subject: [Ironruby-core] Duplicate definition of Action & other classes In-Reply-To: <53194650933664488F3F0C3A0AB568326A2776@nw60.napa.fi> References: <53194650933664488F3F0C3A0AB568326A2776@nw60.napa.fi> Message-ID: There's going to be some merging of classes as parts of the DLR move into the next version of the CLR. The alternative to the pain you're feeling would be for us to branch the DLR, which we're not ready to do yet. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Robert Brotherus Sent: Tuesday, July 08, 2008 9:04 AM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] Duplicate definition of Action & other classes In my ongoing switch from IR SVN 113 to IR SVN 121, another issue has emerged. It seems that multiple new classes have been added to Microsoft.Scripting.Core.dll that have identical name with corresponding classes in System.Core.dll (v. 3.5). Below is my code (that is complicing and running ok against IR SVN 113) and the build errors against IR SVN 121. I can rather easily make a work-around for this case but still wondering if it is possible to make the code work as it is now and why such duplicate classes have been added? Robert Brotherus Software architect Napa Ltd Email: Robert.Brotherus at napa.fi www.napa.fi ------------------------------------------------------------------ public T ExecuteStr(string code) { return Engine.CreateScriptSourceFromString(code).Execute(Engine.CreateScope( )); } public Action RunTests // scriptDir, testMethod <-- ERROR CS0433 HERE { get { ExecuteFile("Test/TestResults.rb"); return ExecuteStr>("TestRunner.method(:run)"); } } ... ScriptService.SharedService.RunTests(ScriptService.ScriptDir, app.ArgValue("test")); ... ------ Build started: Project: gui, Configuration: Debug Any CPU ------ warning CS1685: The predefined type 'System.Linq.Expressions.Expression' is defined in multiple assemblies in the global alias; using definition from 'c:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll' warning CS1685: The predefined type 'System.Linq.Expressions.Expression' is defined in multiple assemblies in the global alias; using definition from 'c:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll' ... warning CS1685: The predefined type 'System.Linq.Expressions.ElementInit' is defined in multiple assemblies in the global alias; using definition from 'c:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll' C:\work\napa.net\gui\ScriptService.cs(113,16): error CS0433: The type 'System.Action' exists in both 'c:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll' and 'c:\programs\IronRuby\trunk\build\debug\svn121\Microsoft.Scripting.Core. dll' c:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll: (Related file) c:\programs\IronRuby\trunk\build\debug\svn121\Microsoft.Scripting.Core.d ll: (Related file) Compile complete -- 1 errors, 21 warnings _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From blowmage at gmail.com Tue Jul 8 12:42:31 2008 From: blowmage at gmail.com (Mike Moore) Date: Tue, 8 Jul 2008 10:42:31 -0600 Subject: [Ironruby-core] Duplicate definition of Action & other classes In-Reply-To: References: <53194650933664488F3F0C3A0AB568326A2776@nw60.napa.fi> Message-ID: Microsoft is already starting the process of merging the DLR into the CLR? Is the DLR considered "done"? On Tue, Jul 8, 2008 at 10:34 AM, Curt Hagenlocher wrote: > There's going to be some merging of classes as parts of the DLR move into > the next version of the CLR. The alternative to the pain you're feeling > would be for us to branch the DLR, which we're not ready to do yet. > > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto: > ironruby-core-bounces at rubyforge.org] On Behalf Of Robert Brotherus > Sent: Tuesday, July 08, 2008 9:04 AM > To: ironruby-core at rubyforge.org > Subject: [Ironruby-core] Duplicate definition of Action & other classes > > In my ongoing switch from IR SVN 113 to IR SVN 121, another issue has > emerged. > > It seems that multiple new classes have been added to > Microsoft.Scripting.Core.dll that have identical name with corresponding > classes in System.Core.dll (v. 3.5). Below is my code (that is > complicing and running ok against IR SVN 113) and the build errors > against IR SVN 121. > > I can rather easily make a work-around for this case but still wondering > if it is possible to make the code work as it is now and why such > duplicate classes have been added? > > Robert Brotherus > Software architect > Napa Ltd > Email: Robert.Brotherus at napa.fi > www.napa.fi > > ------------------------------------------------------------------ > > public T ExecuteStr(string code) > { > return > Engine.CreateScriptSourceFromString(code).Execute(Engine.CreateScope( > )); > } > > public Action RunTests // scriptDir, testMethod > <-- ERROR CS0433 HERE > { > get > { > ExecuteFile("Test/TestResults.rb"); > return ExecuteStr string>>("TestRunner.method(:run)"); > } > } > > ... > > ScriptService.SharedService.RunTests(ScriptService.ScriptDir, > app.ArgValue("test")); > > ... > > ------ Build started: Project: gui, Configuration: Debug Any CPU ------ > > warning CS1685: The predefined type 'System.Linq.Expressions.Expression' > is defined in multiple assemblies in the global alias; using definition > from 'c:\Program Files (x86)\Reference > Assemblies\Microsoft\Framework\v3.5\System.Core.dll' > > warning CS1685: The predefined type 'System.Linq.Expressions.Expression' > is defined in multiple assemblies in the global alias; using definition > from 'c:\Program Files (x86)\Reference > Assemblies\Microsoft\Framework\v3.5\System.Core.dll' > > ... > > warning CS1685: The predefined type > 'System.Linq.Expressions.ElementInit' is defined in multiple assemblies > in the global alias; using definition from 'c:\Program Files > (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll' > > C:\work\napa.net\gui\ScriptService.cs(113,16): error CS0433: The type > 'System.Action' exists in both 'c:\Program Files (x86)\Reference > Assemblies\Microsoft\Framework\v3.5\System.Core.dll' and > 'c:\programs\IronRuby\trunk\build\debug\svn121\Microsoft.Scripting.Core. > dll' > > c:\Program Files (x86)\Reference > Assemblies\Microsoft\Framework\v3.5\System.Core.dll: (Related file) > > c:\programs\IronRuby\trunk\build\debug\svn121\Microsoft.Scripting.Core.d > ll: (Related file) > > Compile complete -- 1 errors, 21 warnings > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > -------------- next part -------------- An HTML attachment was scrubbed... URL: From curth at microsoft.com Tue Jul 8 12:46:36 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Tue, 8 Jul 2008 09:46:36 -0700 Subject: [Ironruby-core] Duplicate definition of Action & other classes In-Reply-To: References: <53194650933664488F3F0C3A0AB568326A2776@nw60.napa.fi> Message-ID: It's "parts of the DLR", not the whole thing. And no, the DLR is not done. From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Mike Moore Sent: Tuesday, July 08, 2008 9:43 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Duplicate definition of Action & other classes Microsoft is already starting the process of merging the DLR into the CLR? Is the DLR considered "done"? On Tue, Jul 8, 2008 at 10:34 AM, Curt Hagenlocher > wrote: There's going to be some merging of classes as parts of the DLR move into the next version of the CLR. The alternative to the pain you're feeling would be for us to branch the DLR, which we're not ready to do yet. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Robert Brotherus Sent: Tuesday, July 08, 2008 9:04 AM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] Duplicate definition of Action & other classes In my ongoing switch from IR SVN 113 to IR SVN 121, another issue has emerged. It seems that multiple new classes have been added to Microsoft.Scripting.Core.dll that have identical name with corresponding classes in System.Core.dll (v. 3.5). Below is my code (that is complicing and running ok against IR SVN 113) and the build errors against IR SVN 121. I can rather easily make a work-around for this case but still wondering if it is possible to make the code work as it is now and why such duplicate classes have been added? Robert Brotherus Software architect Napa Ltd Email: Robert.Brotherus at napa.fi www.napa.fi ------------------------------------------------------------------ public T ExecuteStr(string code) { return Engine.CreateScriptSourceFromString(code).Execute(Engine.CreateScope( )); } public Action RunTests // scriptDir, testMethod <-- ERROR CS0433 HERE { get { ExecuteFile("Test/TestResults.rb"); return ExecuteStr>("TestRunner.method(:run)"); } } ... ScriptService.SharedService.RunTests(ScriptService.ScriptDir, app.ArgValue("test")); ... ------ Build started: Project: gui, Configuration: Debug Any CPU ------ warning CS1685: The predefined type 'System.Linq.Expressions.Expression' is defined in multiple assemblies in the global alias; using definition from 'c:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll' warning CS1685: The predefined type 'System.Linq.Expressions.Expression' is defined in multiple assemblies in the global alias; using definition from 'c:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll' ... warning CS1685: The predefined type 'System.Linq.Expressions.ElementInit' is defined in multiple assemblies in the global alias; using definition from 'c:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll' C:\work\napa.net\gui\ScriptService.cs(113,16): error CS0433: The type 'System.Action' exists in both 'c:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll' and 'c:\programs\IronRuby\trunk\build\debug\svn121\Microsoft.Scripting.Core. dll' c:\Program Files (x86)\Reference Assemblies\Microsoft\Framework\v3.5\System.Core.dll: (Related file) c:\programs\IronRuby\trunk\build\debug\svn121\Microsoft.Scripting.Core.d ll: (Related file) Compile complete -- 1 errors, 21 warnings _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core -------------- next part -------------- An HTML attachment was scrubbed... URL: From jirapong.nanta at gmail.com Tue Jul 8 15:47:18 2008 From: jirapong.nanta at gmail.com (Jirapong) Date: Wed, 9 Jul 2008 02:47:18 +0700 Subject: [Ironruby-core] Mono build update.. Message-ID: Hi, do you build on Mac? the UserProfile is where your home directory. you might try, before compile USERPROFILE=[your home dir] export USERPROFILE Hope this help, -Jirapong On Jul 8, 2008, at 12:21 PM, Unnikrishnan Nair wrote: > Finally I got mono to build successfully.I checkout 121 from > rubyforge and tried to do rake compile and, it failed on the > following line; > > path_to_config = ENV['USERPROFILE'] + '/.irconfig.rb' > > I remember seeing some mail about Home and Userprofile setting from > John and others sometime back I am not sure it is related to the > same. Regardless, I thought I will send this to the group so that if > someone already came across this problem, they could help me without > spending more of my time. If not, I will look into it tomorrow > morning. > > Thanks, > Unni > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core From jdeville at microsoft.com Tue Jul 8 15:57:32 2008 From: jdeville at microsoft.com (Jim Deville) Date: Tue, 8 Jul 2008 12:57:32 -0700 Subject: [Ironruby-core] Mono build update.. In-Reply-To: References: Message-ID: Technically, that can be set automatically with: ENV['USERPROFILE'] ||= ENV['HOME'] (from ruby) or USERPROFILE = $HOME (from shell) It should be fixed next time we do a push. I've changed the Rakefile and context.rb to rely on ENV['HOME'] and set ENV['HOME'] ||= ENV['USERPROFILE'] at the beginning of each of those files. JD > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of Jirapong > Sent: Tuesday, July 08, 2008 12:47 PM > To: ksunair at yahoo.com, ironruby-core at rubyforge.org; ironruby- > core at rubyforge.org> > Cc: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] Mono build update.. > > Hi, > do you build on Mac? the UserProfile is where your home directory. > > you might try, before compile > USERPROFILE=[your home dir] > export USERPROFILE > > Hope this help, > -Jirapong > > On Jul 8, 2008, at 12:21 PM, Unnikrishnan Nair > wrote: > > > Finally I got mono to build successfully.I checkout 121 from > > rubyforge and tried to do rake compile and, it failed on the > > following line; > > > > path_to_config = ENV['USERPROFILE'] + '/.irconfig.rb' > > > > I remember seeing some mail about Home and Userprofile setting from > > John and others sometime back I am not sure it is related to the > > same. Regardless, I thought I will send this to the group so that if > > someone already came across this problem, they could help me without > > spending more of my time. If not, I will look into it tomorrow > > morning. > > > > Thanks, > > Unni > > _______________________________________________ > > Ironruby-core mailing list > > Ironruby-core at rubyforge.org > > http://rubyforge.org/mailman/listinfo/ironruby-core > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core From jflam at microsoft.com Tue Jul 8 18:44:03 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Tue, 8 Jul 2008 15:44:03 -0700 Subject: [Ironruby-core] FW: MindTouch Wik.is Upgrade Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD40AFCC1@NA-EXMSG-C115.redmond.corp.microsoft.com> FYI Thanks, -John From: MindTouch [mailto:MindTouch at mail.vresp.com] Sent: Tuesday, July 08, 2008 2:32 PM To: John Lam (IRONRUBY) Subject: MindTouch Wik.is Upgrade [https://na3.salesforce.com/servlet/servlet.ImageServer?id=01550000000DvzJ&oid=00D500000007V1n] Wik.is will be undergoing an upgrade on Saturday July 12, 2008 between the times of 12:00pm and 8:00pm Pacific Standard Time. During this time there will be intermittent downtime to your wik.is account. Please pass this along to your users. Thank you for your patience during this upgrade. Sincerely, The MindTouch Team [https://na3.salesforce.com/servlet/servlet.ImageServer?id=01550000000CiLj&oid=00D500000007V1n] ________________________________ If you no longer wish to receive these emails, please reply to this message with "Unsubscribe" in the subject line or simply click on the following link: Unsubscribe ________________________________ MindTouch 555 W Beech St Ste 501 San Diego, California 92101 Read the VerticalResponse marketing policy. [http://img.pcdn.vresp.com/pwrby_vr_logo_120.gif] [http://cts.vresp.com/o.gif?d439fff537/d7e3ed46ef/0336aa5] -------------- next part -------------- An HTML attachment was scrubbed... URL: From jflam at microsoft.com Wed Jul 9 00:32:26 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Tue, 8 Jul 2008 21:32:26 -0700 Subject: [Ironruby-core] Code Review: fixes-2 Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD40AFDC8@NA-EXMSG-C115.redmond.corp.microsoft.com> Ruby only. All local tests pass. tfpt review "/shelveset:fixes-2;REDMOND\jflam" Comment : - Added a ChildProcessExitStatus global variable ($?) which contains the exit code of a sub-process (used by system, backtick, exec etc) - Added Glob.cs to Ruby/Ruby/Builtins. This contains Ruby glob functionality that previously resided in the libraries (FileOps.cs and Dir.cs), but was moved so that we can do glob expansions for globs passed in as command-line parameters to the Ruby script. We now expand parameters that look like glob expansions before adding them to ARGV. - Changed RUBY_INSTALL_NAME to ironruby from ruby in rbconfig.rb - Added a Process::Status class - Moved File.CanonicalizePath to from FileOps.cs to Glob.cs since this will be used in multiple places in the future. - Added a new algorithm for determining what the user's HOME path is. Ruby uses HOME for several library methods such as Dir.chdir(). The new algorithm tries HOME, HOMEDRIVE + HOMEPATH, USERPROFILE, and the Personal special folder in that sequence. This is the same algorithm that is used in Ruby 1.9. - Added implementation of `, exec and system to KernelOps.cs - Enabled mspec / rubyspec to run by cleaning up default.mspec.rb, runfirst.cmd. Enables support for specifying the class *and* method when running tests / regressions / baselining. -------------- next part -------------- A non-text attachment was scrubbed... Name: fixes-2.diff Type: application/octet-stream Size: 139823 bytes Desc: fixes-2.diff URL: From jdeville at microsoft.com Wed Jul 9 01:20:30 2008 From: jdeville at microsoft.com (Jim Deville) Date: Tue, 8 Jul 2008 22:20:30 -0700 Subject: [Ironruby-core] Code Review: fixes-2 In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD40AFDC8@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD40AFDC8@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: I've included Srivatsn since he will be covering for me when I'm on paternity leave. Context.rb: None of the changes in context are using the -B option, so they aren't using the default.mspec, unless, perhaps, you have it in the default search locations for mspec. I've been looking more, and running via mspec, instead of the direct mspec-{ci,tag,run} commands also can affect which specs run. Some specs are guarded for if the runner is MSpec or RSpec, using that env. Variable (see mspec/lib/mspec/guards/runner.rb). I'd also expect more config to move into mspec, so we probably want to keep using it. I also wonder why you removed the common command that removed repetition when preparing the command line. Default.mspec.rb: I purposely left core and lang active in default.mspec to allow one command to run all passing tests. Do we want that changed? Target should also stay in to allow mspec to be run without requiring the -t target. Rakefile: I don't see the point in changing path_to_ir to iruby, isn't path_to_ir already a string? Runfirst.cmd: I thought it made sense to keep the config file both in ~, any reason why not? > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) > Sent: Tuesday, July 08, 2008 9:32 PM > To: IronRuby External Code Reviewers > Cc: ironruby-core at rubyforge.org > Subject: [Ironruby-core] Code Review: fixes-2 > > Ruby only. All local tests pass. > > tfpt review "/shelveset:fixes-2;REDMOND\jflam" > > Comment : > - Added a ChildProcessExitStatus global variable ($?) which > contains the exit code of a sub-process (used by system, backtick, exec > etc) > - Added Glob.cs to Ruby/Ruby/Builtins. This contains Ruby glob > functionality that previously resided in the libraries (FileOps.cs and > Dir.cs), but was moved so that we can do glob expansions for globs > passed in as command-line parameters to the Ruby script. We now expand > parameters that look like glob expansions before adding them to ARGV. > - Changed RUBY_INSTALL_NAME to ironruby from ruby in rbconfig.rb > - Added a Process::Status class > - Moved File.CanonicalizePath to from FileOps.cs to Glob.cs since > this will be used in multiple places in the future. > - Added a new algorithm for determining what the user's HOME path > is. Ruby uses HOME for several library methods such as Dir.chdir(). The > new algorithm tries HOME, HOMEDRIVE + HOMEPATH, USERPROFILE, and the > Personal special folder in that sequence. This is the same algorithm > that is used in Ruby 1.9. > - Added implementation of `, exec and system to KernelOps.cs > - Enabled mspec / rubyspec to run by cleaning up > default.mspec.rb, runfirst.cmd. Enables support for specifying the > class *and* method when running tests / regressions / baselining. > From jflam at microsoft.com Wed Jul 9 01:30:53 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Tue, 8 Jul 2008 22:30:53 -0700 Subject: [Ironruby-core] Code Review: fixes-2 In-Reply-To: References: <372109E149E8084D8E6C7D9CFD82E0632FD40AFDC8@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD40AFDE2@NA-EXMSG-C115.redmond.corp.microsoft.com> Comments inline: Thanks, -John > -----Original Message----- > From: Jim Deville > Sent: Tuesday, July 08, 2008 10:21 PM > To: ironruby-core at rubyforge.org; John Lam (IRONRUBY); Oleg Tkachenko; > Tomas Matousek > Cc: Srivatsn Narayanan > Subject: RE: Code Review: fixes-2 > > I've included Srivatsn since he will be covering for me when I'm on > paternity leave. > > Context.rb: > None of the changes in context are using the -B option, so they aren't > using the default.mspec, unless, perhaps, you have it in the default > search locations for mspec. Thanks for the catch. > I've been looking more, and running via mspec, instead of the direct > mspec-{ci,tag,run} commands also can affect which specs run. Some specs > are guarded for if the runner is MSpec or RSpec, using that env. > Variable (see mspec/lib/mspec/guards/runner.rb). I'd also expect more > config to move into mspec, so we probably want to keep using it. Not sure what this means? I had trouble getting the target stuff working in the past, which is why I'm using the direct commands. > I also wonder why you removed the common command that removed > repetition when preparing the command line. This wound up making it much more difficult to tell at a glance what was going on - there was magic inside of that method. I prefer to see the entire command at a glance which is what the original solution did. > Default.mspec.rb: > I purposely left core and lang active in default.mspec to allow one > command to run all passing tests. Do we want that changed? Target > should also stay in to allow mspec to be run without requiring the -t > target. I don't think that enabling core at this point makes sense since none of the test running infrastructure lets you selectively 'run' a particular core test like we do with the libs. I think we should think some more about this before we enable the core specs. > Rakefile: > I don't see the point in changing path_to_ir to iruby, isn't path_to_ir > already a string? iruby is a local variable, path_to_ir is a method call. iruby will be used in the future to pass a -X:Interpret flag to run the specs under Tomas' beauty new interpreter. > Runfirst.cmd: > I thought it made sense to keep the config file both in ~, any reason > why not? Because the relative paths inside of default.mspec wouldn't line up otherwise. Besides, the default locations of the rubyspec, mspec, and ironruby-tags projects all assume a dev subdirectory under home. From lists at ruby-forum.com Wed Jul 9 01:39:43 2008 From: lists at ruby-forum.com (Web Reservoir) Date: Wed, 9 Jul 2008 07:39:43 +0200 Subject: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions Message-ID: <7ca58dc067dc9a67c58ac102feed5325@ruby-forum.com> Hi, May i know, how many months are we still far from IronRuby 1.0...? I understand Asp.Net MVC preview 4 will support IronRuby. What type of support will it be... Beta 1.0 or just a simple working solution with not much benefits..? Will there be enough guidelines/tutorials on.. how to make IronRuby work with Asp.Net MVC 4..? Thanks -- Posted via http://www.ruby-forum.com/. From jdeville at microsoft.com Wed Jul 9 01:41:26 2008 From: jdeville at microsoft.com (Jim Deville) Date: Tue, 8 Jul 2008 22:41:26 -0700 Subject: [Ironruby-core] Code Review: fixes-2 In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD40AFDE2@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD40AFDC8@NA-EXMSG-C115.redmond.corp.microsoft.com> <372109E149E8084D8E6C7D9CFD82E0632FD40AFDE2@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: > -----Original Message----- > From: John Lam (IRONRUBY) > Sent: Tuesday, July 08, 2008 10:31 PM > To: Jim Deville; ironruby-core at rubyforge.org; Oleg Tkachenko; Tomas > Matousek > Cc: Srivatsn Narayanan > Subject: RE: Code Review: fixes-2 > > Comments inline: > > Thanks, > -John > > > > -----Original Message----- > > From: Jim Deville > > Sent: Tuesday, July 08, 2008 10:21 PM > > To: ironruby-core at rubyforge.org; John Lam (IRONRUBY); Oleg Tkachenko; > > Tomas Matousek > > Cc: Srivatsn Narayanan > > Subject: RE: Code Review: fixes-2 > > > > I've included Srivatsn since he will be covering for me when I'm on > > paternity leave. > > > > Context.rb: > > None of the changes in context are using the -B option, so they > aren't > > using the default.mspec, unless, perhaps, you have it in the default > > search locations for mspec. > > Thanks for the catch. > > > I've been looking more, and running via mspec, instead of the direct > > mspec-{ci,tag,run} commands also can affect which specs run. Some > specs > > are guarded for if the runner is MSpec or RSpec, using that env. > > Variable (see mspec/lib/mspec/guards/runner.rb). I'd also expect more > > config to move into mspec, so we probably want to keep using it. > > Not sure what this means? I had trouble getting the target stuff > working in the past, which is why I'm using the direct commands. > Mspec sets the environment variable MSPEC_RUNNER to equal 1. There is a guard that prevents specs from running if MSPEC_RUNNER is set, or maybe vice versa. The point is to be able to disable specs that may be affected by the RSpec framework. In addition, the logic for running parallel is in mspec. Finally, I expect that future configuration might be moved to the mspec script. The -t option should work fine if you pass it a full path. I've only had problems passing things with the shortcuts, due to path issues. In addition, putting it back in the default.mspec file should make it a moot point, that's how we were running it yesterday. If you are seeing problems with the default.mspec file, you may need to make sure that the file ends in .mspec, the runner checks for this. > > I also wonder why you removed the common command that removed > > repetition when preparing the command line. > > This wound up making it much more difficult to tell at a glance what > was going on - there was magic inside of that method. I prefer to see > the entire command at a glance which is what the original solution did. > Fair enough. I was just trying to avoid changing in 4 places, that's part of why I ended up with a half done refactor. > > Default.mspec.rb: > > I purposely left core and lang active in default.mspec to allow one > > command to run all passing tests. Do we want that changed? > Target > > should also stay in to allow mspec to be run without requiring the -t > > target. > > I don't think that enabling core at this point makes sense since none > of the test running infrastructure lets you selectively 'run' a > particular core test like we do with the libs. I think we should think > some more about this before we enable the core specs. I think you mean lang. Core is the std lib tests. Lang is selectively runnable via the mspec:lang task, which I use due to more configurability (method and runner selection). > > > Rakefile: > > I don't see the point in changing path_to_ir to iruby, isn't > path_to_ir > > already a string? > > iruby is a local variable, path_to_ir is a method call. iruby will be > used in the future to pass a -X:Interpret flag to run the specs under > Tomas' beauty new interpreter. Sounds good. I just wanted to know. > > > Runfirst.cmd: > > I thought it made sense to keep the config file both in ~, any reason > > why not? > > Because the relative paths inside of default.mspec wouldn't line up > otherwise. Besides, the default locations of the rubyspec, mspec, and > ironruby-tags projects all assume a dev subdirectory under home. > Ok. From Shri.Borde at microsoft.com Wed Jul 9 02:26:53 2008 From: Shri.Borde at microsoft.com (Shri Borde) Date: Tue, 8 Jul 2008 23:26:53 -0700 Subject: [Ironruby-core] Code Review: StringGsubEach In-Reply-To: References: Message-ID: <50B69702CA6E6D4E849D30CD4989AB8EE90FEF5C36@DF-GRTDANE-MSG.exchange.corp.microsoft.com> Why isn't MutableString.version called MutableString.Version? That would be the consistent naming convention, right? The current versioning implementation is not thread-safe. A thread-safe implementation would require using InterlockedCompareAndExchange to update version, right? If we are deferring nailing of the threading story, do we have a bug to track the issue so that we can revisit this? Similarly, what is the semantics of freezing? The MutableStringOps methods like Append call MutableStringOps.RequireNotFrozenAndIncrVersion and then do the update without a lock. Since atomicity is not guaranteed, there could be race with another thread freezing the object half way through the update which could result in weird behavior (for eg, if the update is a multiple step operation). This change does not have to be blocked on this issue, but if we can decide how we want thread-safety to work, we can start acting on it instead of building up debt in the library. Should incrementing the version check for overflow? If not, a comment would be useful to make the intent obvious that overflow would cause an issue only in extreme cases that will surely never happen for real since most updates will be pretty quick. OTOH, if any of the updates runs user code, then overflow is a real issue, though still quite unlikely. Can MutableStringOps.RequireNotFrozenAndIncrVersion and MutableStringOps.RequireNoVersionChange be methods on MutableString? Inspecting/updating the version outside of MutableString seems to break the encapsulation pretty badly. Thanks, Shri -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Monday, July 07, 2008 11:01 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Looks good. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Dave Remy Sent: Monday, July 07, 2008 5:29 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: StringGsubEach tfpt review "/shelveset:StringGsubEach;REDMOND\dremy" Comment : Changes to for string.gsub and string.each (and each_line) to run clean. The most significant change is to track version in MutableString. This version is bumped on any mutation and functions that iterate (each, each_line, gsub) check the version before and after to assure there has been no version change during the iteration. After these changes all gsub, each, and each_line specs run clean (no excludes). Note that although the specs run clean the each behavior does not match MRI. The spec test contains a new line in the iterating string ("hello\nworld") and MRI does throw a runtime exception if this string is iterated. However if there is no new line in the string MRI does not throw an exception if the underlying string is mutated. This seems inconsistent but worth noting. String#gsub with pattern and block sets $~ for access from the block String#gsub with pattern and block restores $~ after leaving the block String#gsub with pattern and block sets $~ to MatchData of last match and nil when there's none for access from outside String#gsub with pattern and block raises a RuntimeError if the string is modified while substituting String#each raises a RuntimeError if the string is modified while substituting String#each_line raises a RuntimeError if the string is modified while substituting _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From lists at ruby-forum.com Wed Jul 9 07:54:21 2008 From: lists at ruby-forum.com (Rahil Kumar) Date: Wed, 9 Jul 2008 13:54:21 +0200 Subject: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions In-Reply-To: <7ca58dc067dc9a67c58ac102feed5325@ruby-forum.com> References: <7ca58dc067dc9a67c58ac102feed5325@ruby-forum.com> Message-ID: Web Reservoir wrote: > Hi, > > May i know, how many months are we still far from IronRuby 1.0...? It was july 2008 previously. It seems its delayed. I am also waiting for a confirmed date. > > I understand Asp.Net MVC preview 4 will support IronRuby. Asp.Net MVC 4 is scheduled to be in this month only, perhaps mid july. I am not aware about any do's / dont's with this preview. I am sailing in the same boat like you. -- Posted via http://www.ruby-forum.com/. From lists at ruby-forum.com Wed Jul 9 08:05:54 2008 From: lists at ruby-forum.com (Rahil Kumar) Date: Wed, 9 Jul 2008 14:05:54 +0200 Subject: [Ironruby-core] Any Plans for a StarterKit in IronRuby on CodePlex. In-Reply-To: References: <5ba19d9113f99f91b343762ff6c29501@ruby-forum.com> <24e27e560802070114q18f58e25kf959e09b2983e204@mail.gmail.com> <157026466f342eb44dd122dee6d45c2a@ruby-forum.com> <3e6dd551ddcff00ce3d29e653d1d1be9@ruby-forum.com> Message-ID: <91fdfb15fc6a5d67b657d4d513f6ba3d@ruby-forum.com> Hi, It would be great to have a small and working starter kit with Asp.Net MVC with IronRuby, from one of those currently available starter kits for C# and Vb.net. Since Asp.Net MVC preview 4 supports IronRuby, there is nothing wrong in expecting some working solutions to learn more. -- Posted via http://www.ruby-forum.com/. From lists at ruby-forum.com Wed Jul 9 08:17:12 2008 From: lists at ruby-forum.com (Web Reservoir) Date: Wed, 9 Jul 2008 14:17:12 +0200 Subject: [Ironruby-core] Any Plans for a StarterKit in IronRuby on CodePlex. In-Reply-To: <91fdfb15fc6a5d67b657d4d513f6ba3d@ruby-forum.com> References: <5ba19d9113f99f91b343762ff6c29501@ruby-forum.com> <24e27e560802070114q18f58e25kf959e09b2983e204@mail.gmail.com> <157026466f342eb44dd122dee6d45c2a@ruby-forum.com> <3e6dd551ddcff00ce3d29e653d1d1be9@ruby-forum.com> <91fdfb15fc6a5d67b657d4d513f6ba3d@ruby-forum.com> Message-ID: Rahil Kumar wrote: > Hi, > > It would be great to have a small and working starter kit with Asp.Net > MVC with IronRuby, from one of those currently available starter kits > for C# and Vb.net. ---------------------- Hi, Nice to see the old popular thread emerging once again, and perhaps well in time. I would suggest the most popular Classified Starter Kit, which is running almost everywhere on the net in C# and Vb versions. It would be great to have a kit like this, specially with Asp.Net MVC in IronRuby. Lets hope that the prayers are heard by the masters. -- Posted via http://www.ruby-forum.com/. From ben2004uk at googlemail.com Wed Jul 9 08:23:33 2008 From: ben2004uk at googlemail.com (Ben Hall) Date: Wed, 9 Jul 2008 13:23:33 +0100 Subject: [Ironruby-core] Any Plans for a StarterKit in IronRuby on CodePlex. In-Reply-To: <91fdfb15fc6a5d67b657d4d513f6ba3d@ruby-forum.com> References: <5ba19d9113f99f91b343762ff6c29501@ruby-forum.com> <24e27e560802070114q18f58e25kf959e09b2983e204@mail.gmail.com> <157026466f342eb44dd122dee6d45c2a@ruby-forum.com> <3e6dd551ddcff00ce3d29e653d1d1be9@ruby-forum.com> <91fdfb15fc6a5d67b657d4d513f6ba3d@ruby-forum.com> Message-ID: The potential problem to this would be that the starter kits are based on webforms, as such porting them to MVC and IronRuby would be a big task without much benefit? Personally, I'm not a great fan of massive starter kit projects as they trend to take too long to get your head around. Plus they are pretty boring I imagine to implement. I would much prefer to see simple well focused examples of how to use IronRuby + MVC in the form of blog posts or articles. This might include a comparison of how it would be done using C#. Then, when you wanted to find out certain bits of information you could easily search for articles instead of a massive codebase which might not include what your after, and even if it does - without an explication. Ben Blog.BenHall.me.uk On Wed, Jul 9, 2008 at 1:05 PM, Rahil Kumar wrote: > Hi, > > It would be great to have a small and working starter kit with Asp.Net > MVC with IronRuby, from one of those currently available starter kits > for C# and Vb.net. > > Since Asp.Net MVC preview 4 supports IronRuby, there is nothing wrong in > expecting some working solutions to learn more. > -- > Posted via http://www.ruby-forum.com/. > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > From curth at microsoft.com Wed Jul 9 09:10:03 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Wed, 9 Jul 2008 06:10:03 -0700 Subject: [Ironruby-core] Code Review: StringGsubEach In-Reply-To: <50B69702CA6E6D4E849D30CD4989AB8EE90FEF5C36@DF-GRTDANE-MSG.exchange.corp.microsoft.com> References: <50B69702CA6E6D4E849D30CD4989AB8EE90FEF5C36@DF-GRTDANE-MSG.exchange.corp.microsoft.com> Message-ID: Our current threading story is to ensure that shared state is protected, but we haven't really done any work to protect individual objects from threading issues. The version implementation is "thread safe enough" because the absolute value of the version isn't important, only the fact that it's changed. With an initial value of 1, simultaneous updates from two threads could result in a value of 2 instead of 3 -- but this still tells the iterator that a change has occurred. In any event, I'm pretty sure that the subsequent mutation of the string object itself isn't thread-safe at all. I agree with moving the helper functions onto the MutableString class itself and then making the version private. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Shri Borde Sent: Tuesday, July 08, 2008 11:27 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Why isn't MutableString.version called MutableString.Version? That would be the consistent naming convention, right? The current versioning implementation is not thread-safe. A thread-safe implementation would require using InterlockedCompareAndExchange to update version, right? If we are deferring nailing of the threading story, do we have a bug to track the issue so that we can revisit this? Similarly, what is the semantics of freezing? The MutableStringOps methods like Append call MutableStringOps.RequireNotFrozenAndIncrVersion and then do the update without a lock. Since atomicity is not guaranteed, there could be race with another thread freezing the object half way through the update which could result in weird behavior (for eg, if the update is a multiple step operation). This change does not have to be blocked on this issue, but if we can decide how we want thread-safety to work, we can start acting on it instead of building up debt in the library. Should incrementing the version check for overflow? If not, a comment would be useful to make the intent obvious that overflow would cause an issue only in extreme cases that will surely never happen for real since most updates will be pretty quick. OTOH, if any of the updates runs user code, then overflow is a real issue, though still quite unlikely. Can MutableStringOps.RequireNotFrozenAndIncrVersion and MutableStringOps.RequireNoVersionChange be methods on MutableString? Inspecting/updating the version outside of MutableString seems to break the encapsulation pretty badly. Thanks, Shri -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Monday, July 07, 2008 11:01 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Looks good. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Dave Remy Sent: Monday, July 07, 2008 5:29 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: StringGsubEach tfpt review "/shelveset:StringGsubEach;REDMOND\dremy" Comment : Changes to for string.gsub and string.each (and each_line) to run clean. The most significant change is to track version in MutableString. This version is bumped on any mutation and functions that iterate (each, each_line, gsub) check the version before and after to assure there has been no version change during the iteration. After these changes all gsub, each, and each_line specs run clean (no excludes). Note that although the specs run clean the each behavior does not match MRI. The spec test contains a new line in the iterating string ("hello\nworld") and MRI does throw a runtime exception if this string is iterated. However if there is no new line in the string MRI does not throw an exception if the underlying string is mutated. This seems inconsistent but worth noting. String#gsub with pattern and block sets $~ for access from the block String#gsub with pattern and block restores $~ after leaving the block String#gsub with pattern and block sets $~ to MatchData of last match and nil when there's none for access from outside String#gsub with pattern and block raises a RuntimeError if the string is modified while substituting String#each raises a RuntimeError if the string is modified while substituting String#each_line raises a RuntimeError if the string is modified while substituting _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From curth at microsoft.com Wed Jul 9 09:26:00 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Wed, 9 Jul 2008 06:26:00 -0700 Subject: [Ironruby-core] Code Review: StringGsubEach In-Reply-To: References: <50B69702CA6E6D4E849D30CD4989AB8EE90FEF5C36@DF-GRTDANE-MSG.exchange.corp.microsoft.com> Message-ID: Hmm... now that I think about it, it's really only "thread safe enough" if you restrict yourself to 2 threads. With three threads performing simultaneous access, I can work out a sequence that breaks. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Wednesday, July 09, 2008 6:10 AM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Our current threading story is to ensure that shared state is protected, but we haven't really done any work to protect individual objects from threading issues. The version implementation is "thread safe enough" because the absolute value of the version isn't important, only the fact that it's changed. With an initial value of 1, simultaneous updates from two threads could result in a value of 2 instead of 3 -- but this still tells the iterator that a change has occurred. In any event, I'm pretty sure that the subsequent mutation of the string object itself isn't thread-safe at all. I agree with moving the helper functions onto the MutableString class itself and then making the version private. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Shri Borde Sent: Tuesday, July 08, 2008 11:27 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Why isn't MutableString.version called MutableString.Version? That would be the consistent naming convention, right? The current versioning implementation is not thread-safe. A thread-safe implementation would require using InterlockedCompareAndExchange to update version, right? If we are deferring nailing of the threading story, do we have a bug to track the issue so that we can revisit this? Similarly, what is the semantics of freezing? The MutableStringOps methods like Append call MutableStringOps.RequireNotFrozenAndIncrVersion and then do the update without a lock. Since atomicity is not guaranteed, there could be race with another thread freezing the object half way through the update which could result in weird behavior (for eg, if the update is a multiple step operation). This change does not have to be blocked on this issue, but if we can decide how we want thread-safety to work, we can start acting on it instead of building up debt in the library. Should incrementing the version check for overflow? If not, a comment would be useful to make the intent obvious that overflow would cause an issue only in extreme cases that will surely never happen for real since most updates will be pretty quick. OTOH, if any of the updates runs user code, then overflow is a real issue, though still quite unlikely. Can MutableStringOps.RequireNotFrozenAndIncrVersion and MutableStringOps.RequireNoVersionChange be methods on MutableString? Inspecting/updating the version outside of MutableString seems to break the encapsulation pretty badly. Thanks, Shri -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Monday, July 07, 2008 11:01 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Looks good. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Dave Remy Sent: Monday, July 07, 2008 5:29 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: StringGsubEach tfpt review "/shelveset:StringGsubEach;REDMOND\dremy" Comment : Changes to for string.gsub and string.each (and each_line) to run clean. The most significant change is to track version in MutableString. This version is bumped on any mutation and functions that iterate (each, each_line, gsub) check the version before and after to assure there has been no version change during the iteration. After these changes all gsub, each, and each_line specs run clean (no excludes). Note that although the specs run clean the each behavior does not match MRI. The spec test contains a new line in the iterating string ("hello\nworld") and MRI does throw a runtime exception if this string is iterated. However if there is no new line in the string MRI does not throw an exception if the underlying string is mutated. This seems inconsistent but worth noting. String#gsub with pattern and block sets $~ for access from the block String#gsub with pattern and block restores $~ after leaving the block String#gsub with pattern and block sets $~ to MatchData of last match and nil when there's none for access from outside String#gsub with pattern and block raises a RuntimeError if the string is modified while substituting String#each raises a RuntimeError if the string is modified while substituting String#each_line raises a RuntimeError if the string is modified while substituting _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From jflam at microsoft.com Wed Jul 9 11:39:22 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Wed, 9 Jul 2008 08:39:22 -0700 Subject: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions In-Reply-To: References: <7ca58dc067dc9a67c58ac102feed5325@ruby-forum.com> Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD40AFED2@NA-EXMSG-C115.redmond.corp.microsoft.com> The ASP.NET MVC questions should be asked on the ASP.NET MVC forums. The current plan is to ship IronRuby early next year. Thanks, -John -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Rahil Kumar Sent: Wednesday, July 09, 2008 4:54 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions Web Reservoir wrote: > Hi, > > May i know, how many months are we still far from IronRuby 1.0...? It was july 2008 previously. It seems its delayed. I am also waiting for a confirmed date. > > I understand Asp.Net MVC preview 4 will support IronRuby. Asp.Net MVC 4 is scheduled to be in this month only, perhaps mid july. I am not aware about any do's / dont's with this preview. I am sailing in the same boat like you. -- Posted via http://www.ruby-forum.com/. _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From sanxiyn at gmail.com Wed Jul 9 12:26:07 2008 From: sanxiyn at gmail.com (Sanghyeon Seo) Date: Thu, 10 Jul 2008 01:26:07 +0900 Subject: [Ironruby-core] Patch to find_executable Message-ID: <5b0248170807090926x9754a8et4c543c48b41db16@mail.gmail.com> IronRuby's Rakefile and context.rb calls UserEnvironment.find_executable 'ruby.exe', which is hopelessly wrong. (No 'ruby.exe' on Linux, etc.) I wrote a patch to correct this. http://sparcs.kaist.ac.kr/~tinuviel/download/IronRuby/patch-find-executable Please apply. -- Seo Sanghyeon From Tomas.Matousek at microsoft.com Wed Jul 9 12:45:44 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Wed, 9 Jul 2008 09:45:44 -0700 Subject: [Ironruby-core] Code Review: StringGsubEach In-Reply-To: References: <50B69702CA6E6D4E849D30CD4989AB8EE90FEF5C36@DF-GRTDANE-MSG.exchange.corp.microsoft.com> Message-ID: I don't think MutableString needs to be thread safe. User should ensure that a string instance is not shared among threads or use synchronization primitives of 'thread' library to access it. It would maybe make sense to make frozen string thread-safe since it is supposed to be read-only. Like Dictionary is not thread safe. It does increment its version like this: this.version++ We should ensure that runtime structures (RubyExecutionContext, RubyModule, ...) are thead-safe (which they are not now, we need to fix that). Other than that any mutable structure is thread-unsafe unless specified otherwise (e.g. Queue is thread-safe). RW overflow: I thought we compile in "checked" mode, so that all operations are checked for overflow/underflow unless marked unchecked. I've just double-checked and apparently we don't. I think we should flip the bit. RE encapsulation - I agree. The version advancing and checks should be on MutableString. String freezing also needs some improvements, so let's file a bug for both. Tomas -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Wednesday, July 09, 2008 6:26 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: StringGsubEach Hmm... now that I think about it, it's really only "thread safe enough" if you restrict yourself to 2 threads. With three threads performing simultaneous access, I can work out a sequence that breaks. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Wednesday, July 09, 2008 6:10 AM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Our current threading story is to ensure that shared state is protected, but we haven't really done any work to protect individual objects from threading issues. The version implementation is "thread safe enough" because the absolute value of the version isn't important, only the fact that it's changed. With an initial value of 1, simultaneous updates from two threads could result in a value of 2 instead of 3 -- but this still tells the iterator that a change has occurred. In any event, I'm pretty sure that the subsequent mutation of the string object itself isn't thread-safe at all. I agree with moving the helper functions onto the MutableString class itself and then making the version private. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Shri Borde Sent: Tuesday, July 08, 2008 11:27 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Why isn't MutableString.version called MutableString.Version? That would be the consistent naming convention, right? The current versioning implementation is not thread-safe. A thread-safe implementation would require using InterlockedCompareAndExchange to update version, right? If we are deferring nailing of the threading story, do we have a bug to track the issue so that we can revisit this? Similarly, what is the semantics of freezing? The MutableStringOps methods like Append call MutableStringOps.RequireNotFrozenAndIncrVersion and then do the update without a lock. Since atomicity is not guaranteed, there could be race with another thread freezing the object half way through the update which could result in weird behavior (for eg, if the update is a multiple step operation). This change does not have to be blocked on this issue, but if we can decide how we want thread-safety to work, we can start acting on it instead of building up debt in the library. Should incrementing the version check for overflow? If not, a comment would be useful to make the intent obvious that overflow would cause an issue only in extreme cases that will surely never happen for real since most updates will be pretty quick. OTOH, if any of the updates runs user code, then overflow is a real issue, though still quite unlikely. Can MutableStringOps.RequireNotFrozenAndIncrVersion and MutableStringOps.RequireNoVersionChange be methods on MutableString? Inspecting/updating the version outside of MutableString seems to break the encapsulation pretty badly. Thanks, Shri -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Monday, July 07, 2008 11:01 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Looks good. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Dave Remy Sent: Monday, July 07, 2008 5:29 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: StringGsubEach tfpt review "/shelveset:StringGsubEach;REDMOND\dremy" Comment : Changes to for string.gsub and string.each (and each_line) to run clean. The most significant change is to track version in MutableString. This version is bumped on any mutation and functions that iterate (each, each_line, gsub) check the version before and after to assure there has been no version change during the iteration. After these changes all gsub, each, and each_line specs run clean (no excludes). Note that although the specs run clean the each behavior does not match MRI. The spec test contains a new line in the iterating string ("hello\nworld") and MRI does throw a runtime exception if this string is iterated. However if there is no new line in the string MRI does not throw an exception if the underlying string is mutated. This seems inconsistent but worth noting. String#gsub with pattern and block sets $~ for access from the block String#gsub with pattern and block restores $~ after leaving the block String#gsub with pattern and block sets $~ to MatchData of last match and nil when there's none for access from outside String#gsub with pattern and block raises a RuntimeError if the string is modified while substituting String#each raises a RuntimeError if the string is modified while substituting String#each_line raises a RuntimeError if the string is modified while substituting _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From Shri.Borde at microsoft.com Wed Jul 9 12:52:10 2008 From: Shri.Borde at microsoft.com (Shri Borde) Date: Wed, 9 Jul 2008 09:52:10 -0700 Subject: [Ironruby-core] Code Review: StringGsubEach In-Reply-To: References: <50B69702CA6E6D4E849D30CD4989AB8EE90FEF5C36@DF-GRTDANE-MSG.exchange.corp.microsoft.com> Message-ID: <50B69702CA6E6D4E849D30CD4989AB8EE90FEF5CC6@DF-GRTDANE-MSG.exchange.corp.microsoft.com> If MRI strings are not meant to be thread-safe, we don't have to worry about it. Has anyone confirmed that the intent is to move the synchronization responsibility to user code? Note that Python lists guarantee thread-safety and so IronPython had to do work to support that. Regarding Curt's comment, I think the code is broken even with 2 threads (if we care about thread-safety). If you have a multi-processor machine, and the version field and the string data are on separate cache lines, then when the first thread on the first processor does a mutating action, only the string data cache line might get written to main memory. The second thread on the second processor could pick up the mutated data but not the new version number. Beware of hand-rolling your own synchronization primitives. Unless you have read all the ugly details of memory models, you are very likely to get it wrong. See http://msdn.microsoft.com/en-us/magazine/cc163744.aspx. Just use a lock. Thanks, Shri -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Tomas Matousek Sent: Wednesday, July 09, 2008 9:46 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: StringGsubEach I don't think MutableString needs to be thread safe. User should ensure that a string instance is not shared among threads or use synchronization primitives of 'thread' library to access it. It would maybe make sense to make frozen string thread-safe since it is supposed to be read-only. Like Dictionary is not thread safe. It does increment its version like this: this.version++ We should ensure that runtime structures (RubyExecutionContext, RubyModule, ...) are thead-safe (which they are not now, we need to fix that). Other than that any mutable structure is thread-unsafe unless specified otherwise (e.g. Queue is thread-safe). RW overflow: I thought we compile in "checked" mode, so that all operations are checked for overflow/underflow unless marked unchecked. I've just double-checked and apparently we don't. I think we should flip the bit. RE encapsulation - I agree. The version advancing and checks should be on MutableString. String freezing also needs some improvements, so let's file a bug for both. Tomas -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Wednesday, July 09, 2008 6:26 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: StringGsubEach Hmm... now that I think about it, it's really only "thread safe enough" if you restrict yourself to 2 threads. With three threads performing simultaneous access, I can work out a sequence that breaks. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Wednesday, July 09, 2008 6:10 AM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Our current threading story is to ensure that shared state is protected, but we haven't really done any work to protect individual objects from threading issues. The version implementation is "thread safe enough" because the absolute value of the version isn't important, only the fact that it's changed. With an initial value of 1, simultaneous updates from two threads could result in a value of 2 instead of 3 -- but this still tells the iterator that a change has occurred. In any event, I'm pretty sure that the subsequent mutation of the string object itself isn't thread-safe at all. I agree with moving the helper functions onto the MutableString class itself and then making the version private. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Shri Borde Sent: Tuesday, July 08, 2008 11:27 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Why isn't MutableString.version called MutableString.Version? That would be the consistent naming convention, right? The current versioning implementation is not thread-safe. A thread-safe implementation would require using InterlockedCompareAndExchange to update version, right? If we are deferring nailing of the threading story, do we have a bug to track the issue so that we can revisit this? Similarly, what is the semantics of freezing? The MutableStringOps methods like Append call MutableStringOps.RequireNotFrozenAndIncrVersion and then do the update without a lock. Since atomicity is not guaranteed, there could be race with another thread freezing the object half way through the update which could result in weird behavior (for eg, if the update is a multiple step operation). This change does not have to be blocked on this issue, but if we can decide how we want thread-safety to work, we can start acting on it instead of building up debt in the library. Should incrementing the version check for overflow? If not, a comment would be useful to make the intent obvious that overflow would cause an issue only in extreme cases that will surely never happen for real since most updates will be pretty quick. OTOH, if any of the updates runs user code, then overflow is a real issue, though still quite unlikely. Can MutableStringOps.RequireNotFrozenAndIncrVersion and MutableStringOps.RequireNoVersionChange be methods on MutableString? Inspecting/updating the version outside of MutableString seems to break the encapsulation pretty badly. Thanks, Shri -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Monday, July 07, 2008 11:01 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Looks good. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Dave Remy Sent: Monday, July 07, 2008 5:29 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: StringGsubEach tfpt review "/shelveset:StringGsubEach;REDMOND\dremy" Comment : Changes to for string.gsub and string.each (and each_line) to run clean. The most significant change is to track version in MutableString. This version is bumped on any mutation and functions that iterate (each, each_line, gsub) check the version before and after to assure there has been no version change during the iteration. After these changes all gsub, each, and each_line specs run clean (no excludes). Note that although the specs run clean the each behavior does not match MRI. The spec test contains a new line in the iterating string ("hello\nworld") and MRI does throw a runtime exception if this string is iterated. However if there is no new line in the string MRI does not throw an exception if the underlying string is mutated. This seems inconsistent but worth noting. String#gsub with pattern and block sets $~ for access from the block String#gsub with pattern and block restores $~ after leaving the block String#gsub with pattern and block sets $~ to MatchData of last match and nil when there's none for access from outside String#gsub with pattern and block raises a RuntimeError if the string is modified while substituting String#each raises a RuntimeError if the string is modified while substituting String#each_line raises a RuntimeError if the string is modified while substituting _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From curth at microsoft.com Wed Jul 9 13:06:02 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Wed, 9 Jul 2008 10:06:02 -0700 Subject: [Ironruby-core] Code Review: StringGsubEach In-Reply-To: References: <50B69702CA6E6D4E849D30CD4989AB8EE90FEF5C36@DF-GRTDANE-MSG.exchange.corp.microsoft.com> Message-ID: If we use Interlocked.Increment and Interlocked.Decrement, it will automatically wrap the value around without throwing an exception. I don't generally think about cache coherency issues, so Shri is absolutely right. I assume a lock statement would generate the appropriate memory barriers? -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Tomas Matousek Sent: Wednesday, July 09, 2008 9:46 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: StringGsubEach I don't think MutableString needs to be thread safe. User should ensure that a string instance is not shared among threads or use synchronization primitives of 'thread' library to access it. It would maybe make sense to make frozen string thread-safe since it is supposed to be read-only. Like Dictionary is not thread safe. It does increment its version like this: this.version++ We should ensure that runtime structures (RubyExecutionContext, RubyModule, ...) are thead-safe (which they are not now, we need to fix that). Other than that any mutable structure is thread-unsafe unless specified otherwise (e.g. Queue is thread-safe). RW overflow: I thought we compile in "checked" mode, so that all operations are checked for overflow/underflow unless marked unchecked. I've just double-checked and apparently we don't. I think we should flip the bit. RE encapsulation - I agree. The version advancing and checks should be on MutableString. String freezing also needs some improvements, so let's file a bug for both. Tomas -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Wednesday, July 09, 2008 6:26 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: StringGsubEach Hmm... now that I think about it, it's really only "thread safe enough" if you restrict yourself to 2 threads. With three threads performing simultaneous access, I can work out a sequence that breaks. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Wednesday, July 09, 2008 6:10 AM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Our current threading story is to ensure that shared state is protected, but we haven't really done any work to protect individual objects from threading issues. The version implementation is "thread safe enough" because the absolute value of the version isn't important, only the fact that it's changed. With an initial value of 1, simultaneous updates from two threads could result in a value of 2 instead of 3 -- but this still tells the iterator that a change has occurred. In any event, I'm pretty sure that the subsequent mutation of the string object itself isn't thread-safe at all. I agree with moving the helper functions onto the MutableString class itself and then making the version private. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Shri Borde Sent: Tuesday, July 08, 2008 11:27 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Why isn't MutableString.version called MutableString.Version? That would be the consistent naming convention, right? The current versioning implementation is not thread-safe. A thread-safe implementation would require using InterlockedCompareAndExchange to update version, right? If we are deferring nailing of the threading story, do we have a bug to track the issue so that we can revisit this? Similarly, what is the semantics of freezing? The MutableStringOps methods like Append call MutableStringOps.RequireNotFrozenAndIncrVersion and then do the update without a lock. Since atomicity is not guaranteed, there could be race with another thread freezing the object half way through the update which could result in weird behavior (for eg, if the update is a multiple step operation). This change does not have to be blocked on this issue, but if we can decide how we want thread-safety to work, we can start acting on it instead of building up debt in the library. Should incrementing the version check for overflow? If not, a comment would be useful to make the intent obvious that overflow would cause an issue only in extreme cases that will surely never happen for real since most updates will be pretty quick. OTOH, if any of the updates runs user code, then overflow is a real issue, though still quite unlikely. Can MutableStringOps.RequireNotFrozenAndIncrVersion and MutableStringOps.RequireNoVersionChange be methods on MutableString? Inspecting/updating the version outside of MutableString seems to break the encapsulation pretty badly. Thanks, Shri -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Monday, July 07, 2008 11:01 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Looks good. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Dave Remy Sent: Monday, July 07, 2008 5:29 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: StringGsubEach tfpt review "/shelveset:StringGsubEach;REDMOND\dremy" Comment : Changes to for string.gsub and string.each (and each_line) to run clean. The most significant change is to track version in MutableString. This version is bumped on any mutation and functions that iterate (each, each_line, gsub) check the version before and after to assure there has been no version change during the iteration. After these changes all gsub, each, and each_line specs run clean (no excludes). Note that although the specs run clean the each behavior does not match MRI. The spec test contains a new line in the iterating string ("hello\nworld") and MRI does throw a runtime exception if this string is iterated. However if there is no new line in the string MRI does not throw an exception if the underlying string is mutated. This seems inconsistent but worth noting. String#gsub with pattern and block sets $~ for access from the block String#gsub with pattern and block restores $~ after leaving the block String#gsub with pattern and block sets $~ to MatchData of last match and nil when there's none for access from outside String#gsub with pattern and block raises a RuntimeError if the string is modified while substituting String#each raises a RuntimeError if the string is modified while substituting String#each_line raises a RuntimeError if the string is modified while substituting _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From curth at microsoft.com Wed Jul 9 13:16:50 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Wed, 9 Jul 2008 10:16:50 -0700 Subject: [Ironruby-core] Code Review: StringGsubEach In-Reply-To: <50B69702CA6E6D4E849D30CD4989AB8EE90FEF5CC6@DF-GRTDANE-MSG.exchange.corp.microsoft.com> References: <50B69702CA6E6D4E849D30CD4989AB8EE90FEF5C36@DF-GRTDANE-MSG.exchange.corp.microsoft.com> <50B69702CA6E6D4E849D30CD4989AB8EE90FEF5CC6@DF-GRTDANE-MSG.exchange.corp.microsoft.com> Message-ID: Another good reference by the same author is at http://msdn.microsoft.com/en-us/magazine/cc163715.aspx -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Shri Borde Sent: Wednesday, July 09, 2008 9:52 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: StringGsubEach If MRI strings are not meant to be thread-safe, we don't have to worry about it. Has anyone confirmed that the intent is to move the synchronization responsibility to user code? Note that Python lists guarantee thread-safety and so IronPython had to do work to support that. Regarding Curt's comment, I think the code is broken even with 2 threads (if we care about thread-safety). If you have a multi-processor machine, and the version field and the string data are on separate cache lines, then when the first thread on the first processor does a mutating action, only the string data cache line might get written to main memory. The second thread on the second processor could pick up the mutated data but not the new version number. Beware of hand-rolling your own synchronization primitives. Unless you have read all the ugly details of memory models, you are very likely to get it wrong. See http://msdn.microsoft.com/en-us/magazine/cc163744.aspx. Just use a lock. Thanks, Shri -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Tomas Matousek Sent: Wednesday, July 09, 2008 9:46 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: StringGsubEach I don't think MutableString needs to be thread safe. User should ensure that a string instance is not shared among threads or use synchronization primitives of 'thread' library to access it. It would maybe make sense to make frozen string thread-safe since it is supposed to be read-only. Like Dictionary is not thread safe. It does increment its version like this: this.version++ We should ensure that runtime structures (RubyExecutionContext, RubyModule, ...) are thead-safe (which they are not now, we need to fix that). Other than that any mutable structure is thread-unsafe unless specified otherwise (e.g. Queue is thread-safe). RW overflow: I thought we compile in "checked" mode, so that all operations are checked for overflow/underflow unless marked unchecked. I've just double-checked and apparently we don't. I think we should flip the bit. RE encapsulation - I agree. The version advancing and checks should be on MutableString. String freezing also needs some improvements, so let's file a bug for both. Tomas -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Wednesday, July 09, 2008 6:26 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: StringGsubEach Hmm... now that I think about it, it's really only "thread safe enough" if you restrict yourself to 2 threads. With three threads performing simultaneous access, I can work out a sequence that breaks. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Wednesday, July 09, 2008 6:10 AM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Our current threading story is to ensure that shared state is protected, but we haven't really done any work to protect individual objects from threading issues. The version implementation is "thread safe enough" because the absolute value of the version isn't important, only the fact that it's changed. With an initial value of 1, simultaneous updates from two threads could result in a value of 2 instead of 3 -- but this still tells the iterator that a change has occurred. In any event, I'm pretty sure that the subsequent mutation of the string object itself isn't thread-safe at all. I agree with moving the helper functions onto the MutableString class itself and then making the version private. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Shri Borde Sent: Tuesday, July 08, 2008 11:27 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Why isn't MutableString.version called MutableString.Version? That would be the consistent naming convention, right? The current versioning implementation is not thread-safe. A thread-safe implementation would require using InterlockedCompareAndExchange to update version, right? If we are deferring nailing of the threading story, do we have a bug to track the issue so that we can revisit this? Similarly, what is the semantics of freezing? The MutableStringOps methods like Append call MutableStringOps.RequireNotFrozenAndIncrVersion and then do the update without a lock. Since atomicity is not guaranteed, there could be race with another thread freezing the object half way through the update which could result in weird behavior (for eg, if the update is a multiple step operation). This change does not have to be blocked on this issue, but if we can decide how we want thread-safety to work, we can start acting on it instead of building up debt in the library. Should incrementing the version check for overflow? If not, a comment would be useful to make the intent obvious that overflow would cause an issue only in extreme cases that will surely never happen for real since most updates will be pretty quick. OTOH, if any of the updates runs user code, then overflow is a real issue, though still quite unlikely. Can MutableStringOps.RequireNotFrozenAndIncrVersion and MutableStringOps.RequireNoVersionChange be methods on MutableString? Inspecting/updating the version outside of MutableString seems to break the encapsulation pretty badly. Thanks, Shri -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Monday, July 07, 2008 11:01 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers Subject: Re: [Ironruby-core] Code Review: StringGsubEach Looks good. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Dave Remy Sent: Monday, July 07, 2008 5:29 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: StringGsubEach tfpt review "/shelveset:StringGsubEach;REDMOND\dremy" Comment : Changes to for string.gsub and string.each (and each_line) to run clean. The most significant change is to track version in MutableString. This version is bumped on any mutation and functions that iterate (each, each_line, gsub) check the version before and after to assure there has been no version change during the iteration. After these changes all gsub, each, and each_line specs run clean (no excludes). Note that although the specs run clean the each behavior does not match MRI. The spec test contains a new line in the iterating string ("hello\nworld") and MRI does throw a runtime exception if this string is iterated. However if there is no new line in the string MRI does not throw an exception if the underlying string is mutated. This seems inconsistent but worth noting. String#gsub with pattern and block sets $~ for access from the block String#gsub with pattern and block restores $~ after leaving the block String#gsub with pattern and block sets $~ to MatchData of last match and nil when there's none for access from outside String#gsub with pattern and block raises a RuntimeError if the string is modified while substituting String#each raises a RuntimeError if the string is modified while substituting String#each_line raises a RuntimeError if the string is modified while substituting _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From jdeville at microsoft.com Wed Jul 9 16:21:04 2008 From: jdeville at microsoft.com (Jim Deville) Date: Wed, 9 Jul 2008 13:21:04 -0700 Subject: [Ironruby-core] RubySpec changes Message-ID: I'm syncing up the copy of Rubyspec in TFS/SVN with the copy in Git. I'd like to get a code review of the changes. Eventually this will be automated, but for now, I'm attaching them directly. Question: Does the attached format work? It's the format output by git when you run git format-patch. It's basically a diff, but it's different enough that I want to know. I can change to diff's if needed. ~~~~ JD http://blog.jredville.com -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-adding-bignum-spec-for-bignum-divide-by-float-0.0.patch Type: application/octet-stream Size: 1080 bytes Desc: 0001-adding-bignum-spec-for-bignum-divide-by-float-0.0.patch URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 0002-added-guards-for-ironruby-to-bignum-specs-converted.patch Type: application/octet-stream Size: 3772 bytes Desc: 0002-added-guards-for-ironruby-to-bignum-specs-converted.patch URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 0003-added-guards-for-continuation.patch Type: application/octet-stream Size: 1132 bytes Desc: 0003-added-guards-for-continuation.patch URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 0004-modifications-to-dir-specs-to-work-on-windows.patch Type: application/octet-stream Size: 12699 bytes Desc: 0004-modifications-to-dir-specs-to-work-on-windows.patch URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 0005-changed-wording-for-some-enumerable-each-specs.patch Type: application/octet-stream Size: 1015 bytes Desc: 0005-changed-wording-for-some-enumerable-each-specs.patch URL: From bacondarwin at googlemail.com Wed Jul 9 16:35:50 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Wed, 9 Jul 2008 21:35:50 +0100 Subject: [Ironruby-core] RubySpec changes In-Reply-To: References: Message-ID: <008201c8e203$617f96b0$247ec410$@com> Why not fork the RubySpec project on GitHub, apply your patches and point us at your forked GIT repository? That way we can easily see the changes using the tools on GitHub. Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Wednesday,09 July 09, 2008 21:21 To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers; Srivatsn Narayanan Subject: [Ironruby-core] RubySpec changes I'm syncing up the copy of Rubyspec in TFS/SVN with the copy in Git. I'd like to get a code review of the changes. Eventually this will be automated, but for now, I'm attaching them directly. Question: Does the attached format work? It's the format output by git when you run git format-patch. It's basically a diff, but it's different enough that I want to know. I can change to diff's if needed. ~~~~ JD http://blog.jredville.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From Tomas.Matousek at microsoft.com Wed Jul 9 22:09:28 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Wed, 9 Jul 2008 19:09:28 -0700 Subject: [Ironruby-core] Code Review: BlockDispatch7 Message-ID: tfpt review "/shelveset:BlockDispatch7;REDMOND\tomat" Improves dispatch to blocks. Previously, DynamicSites were used to adapt call site's arguments to the signature of the target block. In block invocation no resolution needs to be performed, which makes it different from method invocation. The dynamic behavior is only in the arguments to parameters transformation. In usual case it is straightforward though. Only if splatting/unsplatting is used (and in some other special cases) there are various checks that need to be performed to shuffle the arguments right. Although dynamic sites could help here in some cases (by caching by a shape of the target block signatures) the usual cases are rather slowed down by the overhead. In an optimal case w/o any splatting/unsplatting and without polymorphic sites kicking in at least one comparison and 2 delegate calls needs to be done. This change replaces dynamic site dispatch by a virtual dispatch optimized for 0-4 parameters. Each block is associated with a block dispatcher (a subclass of BlockDispatcher abstract class, previously RubyBlockInfo) that corresponds to its signature. The specialized dispatchers implement virtual Invoke methods for 0...4 and N parameters w/ and w/o splatting. The call site uses one of those Invoke methods (based on its arguments) and calls it. The dispatcher holds on a delegate that points to the block method. The delegate is called by Invoke methods with transformed arguments. In the optimal case (e.g. 1.times {|x| puts x}) the cost of block yield is a virtual method dispatch and a delegate call. Besides no runtime-generated stubs are needed which improves startup time. Using the dispatchers also enables to move some previously generated code into RubyOps and therefore decreases the amount of generated code even more. Blocks are still IDOs to provide good interop. This change also made the rules much simpler. TODO: There is some work to be done to optimize some paths thru dispatchers. Will need to run some micro-benchmarks for block dispatch to see where we should do better. Also, some parts of the code seem to be good candidates for source code generation, but I haven't opted for that for now since it was easier to write it by hand (there are many exemptions to the "rules" of the block dispatch, so even if the code looks like it could be generated at the first glance the generator would actually get more complicated to handle all such cases). I've let this in TODO bucket. Tomas -------------- next part -------------- A non-text attachment was scrubbed... Name: BlockDispatch7.diff Type: application/octet-stream Size: 318183 bytes Desc: BlockDispatch7.diff URL: From ivan at flanders.co.nz Wed Jul 9 22:19:28 2008 From: ivan at flanders.co.nz (Ivan Porto Carrero) Date: Thu, 10 Jul 2008 14:19:28 +1200 Subject: [Ironruby-core] WPF philosophy Message-ID: Hi I don't really know if this is an appropriate discussion but here it goes. Ruby and Xaml at first sight a DSL is very appropriate to get rid of the xml syntax. I'm all for that :) However the price you pay is that at that moment only developers can change the design of the application. Blend will happily refuse to load it for example which makes it a bad idea for projects that are developed by a designer and a developer IMHO. What I do see happening is and what I'll use in my chapter is smallish additions, control instantiation and code generation dsl's at functional places in the application. With a DSL you also loose skinning support and the lot. I would be very happy to be proven wrong though :). There are a lot of other places where DSL's really work like setting up timers, threads and those things. Furthermore because xaml controls exist in the statically typed part of the .NET universe it's pretty hard to have controls written in Dynamic languages. I started the Dynamic script control that allows you to inject a DLR based control in a XAML tree. I didn't have time to fully fix it but it does work to a certain degree. One of the problems I have there is that I could not get to the controls inside the dyn script control anymore, the reason is unknown to me at this time. I put that on the back burner for now as I have to make more progress on my book but intend to fully finish that control when my book is finished if nobody else has done one by then or you tell me you will have something like that when IR goes 1.0. I think it would be great to be able to use DLR controls with similar paradigms as their CLR cousins. Then another avenue I've been thinking about is a couple of rake tasks that will take a control tree and spit out the CLR based control with stub event handlers etc so that a designer can still change it but the real implementation of the event handlers occurs in a monkey patched version of that control. This should give you the best of both worlds because you don't need to venture into C# while retaining full tooling support. A downside to this approach is that you will always have a compilation step after you make a change to the design of a control. This is the approach the guys of rucola (a rails like framework for rubycocoa) are taking with interface builder. While blend is no interface builder there are some similarities. A last approach I have in mind but I tend to think that this is a bad one :) A XAML template engine. I think its a bad idea because you lose everything designer support, databinding and dsl niceness. As a conclusion I think I would probably favor a mixture of approach 1-3 with approach 3 being the predominant one and approach 1-2 when needed. Thougths? Thanks Ivan -------------- next part -------------- An HTML attachment was scrubbed... URL: From jirapong.nanta at gmail.com Thu Jul 10 02:17:25 2008 From: jirapong.nanta at gmail.com (Jirapong Nanta) Date: Thu, 10 Jul 2008 13:17:25 +0700 Subject: [Ironruby-core] FW: MindTouch Wik.is Upgrade In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD40AFCC1@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD40AFCC1@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: Hello John, website show - reference to undefined name 'syntax' ..for all block. Thanks, -Jirapong On Wed, Jul 9, 2008 at 5:44 AM, John Lam (IRONRUBY) wrote: > FYI > > > > Thanks, > > -John > > > > *From:* MindTouch [mailto:MindTouch at mail.vresp.com] > *Sent:* Tuesday, July 08, 2008 2:32 PM > *To:* John Lam (IRONRUBY) > *Subject:* MindTouch Wik.is Upgrade > > > > > > Wik.is will be undergoing an upgrade on Saturday July 12, 2008 between the > times of 12:00pm and 8:00pm Pacific Standard Time. During this time there > will be intermittent downtime to your wik.is account. Please pass this > along to your users. > > Thank you for your patience during this upgrade. > > > > Sincerely, > > The MindTouch Team > > > > > > > > > > ------------------------------ > > If you no longer wish to receive these emails, please reply to this message > with "Unsubscribe" in the subject line or simply click on the following > link: Unsubscribe > ------------------------------ > > MindTouch > 555 W Beech St > Ste 501 > San Diego, California 92101 > > Read the > VerticalResponse marketing policy. > > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From stefan.dobrev at gmail.com Thu Jul 10 04:24:02 2008 From: stefan.dobrev at gmail.com (Stefan Dobrev) Date: Thu, 10 Jul 2008 11:24:02 +0300 Subject: [Ironruby-core] WPF philosophy In-Reply-To: References: Message-ID: <928de89c0807100124n6e04a2d6g96f5e9fdf3e2cc19@mail.gmail.com> I have also thought of similar ideas. However mine were more code centric, not XAML. For example if you want to create a DependecyProperty or RoutedEvent you have to write a lot of code in a static language like DependecyProperty.Register(...) and you end up with something like 20 lines of code including all the property changed, coerce, validate handlers. This can be a lot simpler in a dynamic language like ruby when you can have approach similar to rails' active record: has_many: or belongs_to: I think we should try to avoid going away from XAML, because this is our common language with designers. They live in tools like Blend and even Visual Studio is cumbersome for some of them. What we can have there is some kind of ruby markup extension that will allow us to do dynamic dispatch or something similar. P.S.: You can check a simple "still buggy" tool I have developed DLR Pad(XAML Pad with DLR scripting). Regards, Stefan 2008/7/10 Ivan Porto Carrero : > Hi > > I don't really know if this is an appropriate discussion but here it goes. > > Ruby and Xaml at first sight a DSL is very appropriate to get rid of the > xml syntax. I'm all for that :) However the price you pay is that at that > moment only developers can change the design of the application. Blend will > happily refuse to load it for example which makes it a bad idea for projects > that are developed by a designer and a developer IMHO. What I do see > happening is and what I'll use in my chapter is smallish additions, control > instantiation and code generation dsl's at functional places in the > application. > With a DSL you also loose skinning support and the lot. I would be very > happy to be proven wrong though :). > There are a lot of other places where DSL's really work like setting up > timers, threads and those things. > > Furthermore because xaml controls exist in the statically typed part of the > .NET universe it's pretty hard to have controls written in Dynamic > languages. I started the Dynamic script control that allows you to inject a > DLR based control in a XAML tree. I didn't have time to fully fix it but it > does work to a certain degree. One of the problems I have there is that I > could not get to the controls inside the dyn script control anymore, the > reason is unknown to me at this time. I put that on the back burner for now > as I have to make more progress on my book but intend to fully finish that > control when my book is finished if nobody else has done one by then or you > tell me you will have something like that when IR goes 1.0. I think it > would be great to be able to use DLR controls with similar paradigms as > their CLR cousins. > > Then another avenue I've been thinking about is a couple of rake tasks that > will take a control tree and spit out the CLR based control with stub event > handlers etc so that a designer can still change it but the real > implementation of the event handlers occurs in a monkey patched version of > that control. This should give you the best of both worlds because you don't > need to venture into C# while retaining full tooling support. A downside to > this approach is that you will always have a compilation step after you make > a change to the design of a control. This is the approach the guys of > rucola (a rails like framework for rubycocoa) are taking with interface > builder. While blend is no interface builder there are some similarities. > > A last approach I have in mind but I tend to think that this is a bad one > :) A XAML template engine. I think its a bad idea because you lose > everything designer support, databinding and dsl niceness. > > As a conclusion I think I would probably favor a mixture of approach 1-3 > with approach 3 being the predominant one and approach 1-2 when needed. > Thougths? > > Thanks > Ivan > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From robert.brotherus at napa.fi Thu Jul 10 05:19:20 2008 From: robert.brotherus at napa.fi (Robert Brotherus) Date: Thu, 10 Jul 2008 12:19:20 +0300 Subject: [Ironruby-core] Accessing CLR-class methods/properties/fields References: <928de89c0807100124n6e04a2d6g96f5e9fdf3e2cc19@mail.gmail.com> Message-ID: <53194650933664488F3F0C3A0AB568326A28CE@nw60.napa.fi> Consider following C# program (ScriptService.SharedService provides access to IronRuby and is unimportant for understanding the case): ---------------------------------------- using System; using System.Diagnostics; using Fi.Napa.Gui; public class A { public string SomeField = "field"; public string SomeProperty { get { return "prop"; } } public string SomeMethod() { return "method"; } } class Program { static void Main(string[] args) { ScriptService.SharedService.ExecuteStr(@" def test(obj) Debug::print 'In Ruby' Debug::print obj.some_method Debug::print obj.SomeMethod Debug::print obj.send(:some_method) Debug::print obj.send(:SomeMethod) Debug::print obj.some_property Debug::print obj.SomeProperty Debug::print obj.send(:some_property) Debug::print obj.send(:SomeProperty) Debug::print obj.SomeField Debug::print obj.some_field Debug::print obj.send(:SomeField) Debug::print obj.send(:some_field) end "); Action test = ScriptService.SharedService.ExecuteStr>("method(:test)"); try { test(new A()); // Call the Ruby "test"-function with instance of A as parameter } catch (Exception ex) { Debug.Print(ex.ToString()); } } } ---------------------------------------- Output of the program is: In Ruby method method method method prop prop prop prop field field A first chance exception of type 'System.ArgumentException' occurred in Unknown Module. System.ArgumentException: wrong number or type of arguments for `SomeField' at _stub_$27##27(Closure , CallSite , CodeContext , Object , RubyArray ) at _stub_MatchCaller(Object , CallSite , Object[] ) at System.Scripting.Actions.CallSite`1.UpdateAndExecute(Object[] args) in C:\programs\IronRuby\trunk\src\Microsoft.Scripting.Core\Actions\CallSite .cs:line 288 ... at System.Void(Object)(Object[] , Object ) at Program.Main(String[] args) in C:\DATA\IronRubyIssueX\IronRubyIssueX\Program.cs:line 38 ---------------------------------------- This indicates that CLR-class fields (unlike CLR-methods and CLR-properties) cannot be accessed with ruby:s send-method although they are directly accessible as ruby-methods: Debug::print obj.SomeField # ok Debug::print obj.some_field # ok Debug::print obj.send(:SomeField) # Exception: wrong number of arguments for method SomeField Debug::print obj.send(:some_field) # Exception: wrong number of arguments for method some_field I would expect that in final IronRuby 1.0 CLR-interoperability the syntaxes a.x and a.send(:x) would always work the same, regardless of a and x. I assume this is the intention of IR team as well? Also I am intrigued by the type of the exception: "wrong number or type of arguments for `SomeField'". Does that indicate that the fields *could* still be accessed with the send-method with some additional arguments...? Robert Brotherus Software architect Napa Ltd Tammasaarenkatu 3, Helsinki FI-00180 P.O.Box 470, Helsinki FI-00181 Tel. +358 9 22 813 1 Direct. +358 9 22 813 611 GSM +358 45 11 456 02 Fax. +358 9 22 813 800 Email: Robert.Brotherus at napa.fi www.napa.fi From lists at ruby-forum.com Thu Jul 10 07:19:48 2008 From: lists at ruby-forum.com (Web Reservoir) Date: Thu, 10 Jul 2008 13:19:48 +0200 Subject: [Ironruby-core] Rails for Asp.Net developers, 3 chapters to download Message-ID: Hi, This is just for Information. Rails for .Net Developers, is a new Book coming in November 2008. You can down load 3 chapters for free here... http://rubyindia.wordpress.com/2008/07/10/rails-for-aspnet-developers-read-and-download-3-chapters/ Thanks -- Posted via http://www.ruby-forum.com/. From jdeville at microsoft.com Thu Jul 10 15:44:42 2008 From: jdeville at microsoft.com (Jim Deville) Date: Thu, 10 Jul 2008 12:44:42 -0700 Subject: [Ironruby-core] RubySpec changes In-Reply-To: <008201c8e203$617f96b0$247ec410$@com> References: <008201c8e203$617f96b0$247ec410$@com> Message-ID: Good idea. I'm going to start doing that, but here's an update on what I've done while I move my get repo's origin. I've already sent some of the out, but I resent all just to be safe. JD From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Wednesday, July 09, 2008 1:36 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] RubySpec changes Why not fork the RubySpec project on GitHub, apply your patches and point us at your forked GIT repository? That way we can easily see the changes using the tools on GitHub. Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Wednesday,09 July 09, 2008 21:21 To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers; Srivatsn Narayanan Subject: [Ironruby-core] RubySpec changes I'm syncing up the copy of Rubyspec in TFS/SVN with the copy in Git. I'd like to get a code review of the changes. Eventually this will be automated, but for now, I'm attaching them directly. Question: Does the attached format work? It's the format output by git when you run git format-patch. It's basically a diff, but it's different enough that I want to know. I can change to diff's if needed. ~~~~ JD http://blog.jredville.com -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 0001-adding-bignum-spec-for-bignum-divide-by-float-0.0.patch Type: application/octet-stream Size: 1080 bytes Desc: 0001-adding-bignum-spec-for-bignum-divide-by-float-0.0.patch URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 0002-added-guards-for-ironruby-to-bignum-specs-converted.patch Type: application/octet-stream Size: 3772 bytes Desc: 0002-added-guards-for-ironruby-to-bignum-specs-converted.patch URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 0003-added-guards-for-continuation.patch Type: application/octet-stream Size: 1132 bytes Desc: 0003-added-guards-for-continuation.patch URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 0004-modifications-to-dir-specs-to-work-on-windows.patch Type: application/octet-stream Size: 12699 bytes Desc: 0004-modifications-to-dir-specs-to-work-on-windows.patch URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 0005-changed-wording-for-some-enumerable-each-specs.patch Type: application/octet-stream Size: 1015 bytes Desc: 0005-changed-wording-for-some-enumerable-each-specs.patch URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 0006-modified-env-specs-for-windows.patch Type: application/octet-stream Size: 1265 bytes Desc: 0006-modified-env-specs-for-windows.patch URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 0007-uncommented-script_error_spec-added-case-insensitiv.patch Type: application/octet-stream Size: 2020 bytes Desc: 0007-uncommented-script_error_spec-added-case-insensitiv.patch URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 0008-removed-extra-puts-in-file-open_spec.rb.patch Type: application/octet-stream Size: 763 bytes Desc: 0008-removed-extra-puts-in-file-open_spec.rb.patch URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 0009-modified-hash-spec-guards-to-include-IronRuby.patch Type: application/octet-stream Size: 2638 bytes Desc: 0009-modified-hash-spec-guards-to-include-IronRuby.patch URL: From curth at microsoft.com Thu Jul 10 16:09:47 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Thu, 10 Jul 2008 13:09:47 -0700 Subject: [Ironruby-core] Code Review: BlockDispatch7 In-Reply-To: References: Message-ID: There doesn't generally appear to be any consistency about comparing the block parameter to null. For instance, in RangeOps, only StepFixnum checks; StepString, StepNumeric and StepObject do not. I know that the situation predates the current changeset, but it might be a good time to go through and fix it. The test for a non-empty range in RangeOps.StepFixnum does not correctly handle ExcludeEnd. In method Enumerable.Find, there's now an extraneous assignment of "item" to "result". In Proc.cs, there are a bunch of newly-added Call methods inside an #if UNUSED. Are these for future use? The rethrow inside ThreadOps.CreateThread will take down the process. Is that what's desired? I wouldn't swear to understanding all of the changes to the compiler, but things otherwise look good to me. I particularly like the new pattern for Block.Yield; the "out" parameter will make it much harder to forget to check for a jump. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Tomas Matousek Sent: Wednesday, July 09, 2008 7:09 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: BlockDispatch7 tfpt review "/shelveset:BlockDispatch7;REDMOND\tomat" Improves dispatch to blocks. Previously, DynamicSites were used to adapt call site's arguments to the signature of the target block. In block invocation no resolution needs to be performed, which makes it different from method invocation. The dynamic behavior is only in the arguments to parameters transformation. In usual case it is straightforward though. Only if splatting/unsplatting is used (and in some other special cases) there are various checks that need to be performed to shuffle the arguments right. Although dynamic sites could help here in some cases (by caching by a shape of the target block signatures) the usual cases are rather slowed down by the overhead. In an optimal case w/o any splatting/unsplatting and without polymorphic sites kicking in at least one comparison and 2 delegate calls needs to be done. This change replaces dynamic site dispatch by a virtual dispatch optimized for 0-4 parameters. Each block is associated with a block dispatcher (a subclass of BlockDispatcher abstract class, previously RubyBlockInfo) that corresponds to its signature. The specialized dispatchers implement virtual Invoke methods for 0...4 and N parameters w/ and w/o splatting. The call site uses one of those Invoke methods (based on its arguments) and calls it. The dispatcher holds on a delegate that points to the block method. The delegate is called by Invoke methods with transformed arguments. In the optimal case (e.g. 1.times {|x| puts x}) the cost of block yield is a virtual method dispatch and a delegate call. Besides no runtime-generated stubs are needed which improves startup time. Using the dispatchers also enables to move some previously generated code into RubyOps and therefore decreases the amount of generated code even more. Blocks are still IDOs to provide good interop. This change also made the rules much simpler. TODO: There is some work to be done to optimize some paths thru dispatchers. Will need to run some micro-benchmarks for block dispatch to see where we should do better. Also, some parts of the code seem to be good candidates for source code generation, but I haven't opted for that for now since it was easier to write it by hand (there are many exemptions to the "rules" of the block dispatch, so even if the code looks like it could be generated at the first glance the generator would actually get more complicated to handle all such cases). I've let this in TODO bucket. Tomas From Tomas.Matousek at microsoft.com Thu Jul 10 17:20:58 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Thu, 10 Jul 2008 14:20:58 -0700 Subject: [Ironruby-core] Code Review: BlockDispatch7 In-Reply-To: References: Message-ID: StepString, StepNumeric, StepFixnum check for the block in the loop. Note that the exception is not thrown if no item is visited by the enumerator. But you're right there are methods that handle the block incorrectly. We need to write some tests that pass nil to all library methods that use blocks to cover all cases. The assignment "result = item" in Enumerable.Find isn't in fact redundant. It assigns to the closed-over variable "result". "return RuntimeFlowControl.BlockBreak(selfBlock, item);" doesn't return from the Find method, it returns from the lambda defined within the method :) Yes, I'll revisit the Call methods on the next pass. I've added rethrow to the Thread.CreateThread since it is imo better to kill the process rather than silently swallow the exception (I hit this when something wrong happened in the thread and I didn't know what because the exception has been swallowed). Tomas -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Thursday, July 10, 2008 1:10 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: BlockDispatch7 There doesn't generally appear to be any consistency about comparing the block parameter to null. For instance, in RangeOps, only StepFixnum checks; StepString, StepNumeric and StepObject do not. I know that the situation predates the current changeset, but it might be a good time to go through and fix it. The test for a non-empty range in RangeOps.StepFixnum does not correctly handle ExcludeEnd. In method Enumerable.Find, there's now an extraneous assignment of "item" to "result". In Proc.cs, there are a bunch of newly-added Call methods inside an #if UNUSED. Are these for future use? The rethrow inside ThreadOps.CreateThread will take down the process. Is that what's desired? I wouldn't swear to understanding all of the changes to the compiler, but things otherwise look good to me. I particularly like the new pattern for Block.Yield; the "out" parameter will make it much harder to forget to check for a jump. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Tomas Matousek Sent: Wednesday, July 09, 2008 7:09 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: BlockDispatch7 tfpt review "/shelveset:BlockDispatch7;REDMOND\tomat" Improves dispatch to blocks. Previously, DynamicSites were used to adapt call site's arguments to the signature of the target block. In block invocation no resolution needs to be performed, which makes it different from method invocation. The dynamic behavior is only in the arguments to parameters transformation. In usual case it is straightforward though. Only if splatting/unsplatting is used (and in some other special cases) there are various checks that need to be performed to shuffle the arguments right. Although dynamic sites could help here in some cases (by caching by a shape of the target block signatures) the usual cases are rather slowed down by the overhead. In an optimal case w/o any splatting/unsplatting and without polymorphic sites kicking in at least one comparison and 2 delegate calls needs to be done. This change replaces dynamic site dispatch by a virtual dispatch optimized for 0-4 parameters. Each block is associated with a block dispatcher (a subclass of BlockDispatcher abstract class, previously RubyBlockInfo) that corresponds to its signature. The specialized dispatchers implement virtual Invoke methods for 0...4 and N parameters w/ and w/o splatting. The call site uses one of those Invoke methods (based on its arguments) and calls it. The dispatcher holds on a delegate that points to the block method. The delegate is called by Invoke methods with transformed arguments. In the optimal case (e.g. 1.times {|x| puts x}) the cost of block yield is a virtual method dispatch and a delegate call. Besides no runtime-generated stubs are needed which improves startup time. Using the dispatchers also enables to move some previously generated code into RubyOps and therefore decreases the amount of generated code even more. Blocks are still IDOs to provide good interop. This change also made the rules much simpler. TODO: There is some work to be done to optimize some paths thru dispatchers. Will need to run some micro-benchmarks for block dispatch to see where we should do better. Also, some parts of the code seem to be good candidates for source code generation, but I haven't opted for that for now since it was easier to write it by hand (there are many exemptions to the "rules" of the block dispatch, so even if the code looks like it could be generated at the first glance the generator would actually get more complicated to handle all such cases). I've let this in TODO bucket. Tomas _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From jdeville at microsoft.com Thu Jul 10 17:24:04 2008 From: jdeville at microsoft.com (Jim Deville) Date: Thu, 10 Jul 2008 14:24:04 -0700 Subject: [Ironruby-core] RubySpec changes In-Reply-To: References: <008201c8e203$617f96b0$247ec410$@com> Message-ID: I went over these face to face with John, but for those who want, I've also now pushed ( and will in the future) these to my github fork: http://github.com/jredville/rubyspec. You can check them out there. JD From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Thursday, July 10, 2008 12:45 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers; Srivatsn Narayanan Subject: Re: [Ironruby-core] RubySpec changes Good idea. I'm going to start doing that, but here's an update on what I've done while I move my get repo's origin. I've already sent some of the out, but I resent all just to be safe. JD From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Wednesday, July 09, 2008 1:36 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] RubySpec changes Why not fork the RubySpec project on GitHub, apply your patches and point us at your forked GIT repository? That way we can easily see the changes using the tools on GitHub. Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Wednesday,09 July 09, 2008 21:21 To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers; Srivatsn Narayanan Subject: [Ironruby-core] RubySpec changes I'm syncing up the copy of Rubyspec in TFS/SVN with the copy in Git. I'd like to get a code review of the changes. Eventually this will be automated, but for now, I'm attaching them directly. Question: Does the attached format work? It's the format output by git when you run git format-patch. It's basically a diff, but it's different enough that I want to know. I can change to diff's if needed. ~~~~ JD http://blog.jredville.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From curth at microsoft.com Thu Jul 10 17:41:05 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Thu, 10 Jul 2008 14:41:05 -0700 Subject: [Ironruby-core] Code Review: BlockDispatch7 In-Reply-To: References: Message-ID: In principle I agree with you, but killing the process on an unhandled thread exception is an incompatibility with MRI: irb(main):001:0> Thread.start { raise 'foo' } => # irb(main):002:0> -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Tomas Matousek Sent: Thursday, July 10, 2008 2:21 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: BlockDispatch7 StepString, StepNumeric, StepFixnum check for the block in the loop. Note that the exception is not thrown if no item is visited by the enumerator. But you're right there are methods that handle the block incorrectly. We need to write some tests that pass nil to all library methods that use blocks to cover all cases. The assignment "result = item" in Enumerable.Find isn't in fact redundant. It assigns to the closed-over variable "result". "return RuntimeFlowControl.BlockBreak(selfBlock, item);" doesn't return from the Find method, it returns from the lambda defined within the method :) Yes, I'll revisit the Call methods on the next pass. I've added rethrow to the Thread.CreateThread since it is imo better to kill the process rather than silently swallow the exception (I hit this when something wrong happened in the thread and I didn't know what because the exception has been swallowed). Tomas -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Thursday, July 10, 2008 1:10 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: BlockDispatch7 There doesn't generally appear to be any consistency about comparing the block parameter to null. For instance, in RangeOps, only StepFixnum checks; StepString, StepNumeric and StepObject do not. I know that the situation predates the current changeset, but it might be a good time to go through and fix it. The test for a non-empty range in RangeOps.StepFixnum does not correctly handle ExcludeEnd. In method Enumerable.Find, there's now an extraneous assignment of "item" to "result". In Proc.cs, there are a bunch of newly-added Call methods inside an #if UNUSED. Are these for future use? The rethrow inside ThreadOps.CreateThread will take down the process. Is that what's desired? I wouldn't swear to understanding all of the changes to the compiler, but things otherwise look good to me. I particularly like the new pattern for Block.Yield; the "out" parameter will make it much harder to forget to check for a jump. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Tomas Matousek Sent: Wednesday, July 09, 2008 7:09 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: BlockDispatch7 tfpt review "/shelveset:BlockDispatch7;REDMOND\tomat" Improves dispatch to blocks. Previously, DynamicSites were used to adapt call site's arguments to the signature of the target block. In block invocation no resolution needs to be performed, which makes it different from method invocation. The dynamic behavior is only in the arguments to parameters transformation. In usual case it is straightforward though. Only if splatting/unsplatting is used (and in some other special cases) there are various checks that need to be performed to shuffle the arguments right. Although dynamic sites could help here in some cases (by caching by a shape of the target block signatures) the usual cases are rather slowed down by the overhead. In an optimal case w/o any splatting/unsplatting and without polymorphic sites kicking in at least one comparison and 2 delegate calls needs to be done. This change replaces dynamic site dispatch by a virtual dispatch optimized for 0-4 parameters. Each block is associated with a block dispatcher (a subclass of BlockDispatcher abstract class, previously RubyBlockInfo) that corresponds to its signature. The specialized dispatchers implement virtual Invoke methods for 0...4 and N parameters w/ and w/o splatting. The call site uses one of those Invoke methods (based on its arguments) and calls it. The dispatcher holds on a delegate that points to the block method. The delegate is called by Invoke methods with transformed arguments. In the optimal case (e.g. 1.times {|x| puts x}) the cost of block yield is a virtual method dispatch and a delegate call. Besides no runtime-generated stubs are needed which improves startup time. Using the dispatchers also enables to move some previously generated code into RubyOps and therefore decreases the amount of generated code even more. Blocks are still IDOs to provide good interop. This change also made the rules much simpler. TODO: There is some work to be done to optimize some paths thru dispatchers. Will need to run some micro-benchmarks for block dispatch to see where we should do better. Also, some parts of the code seem to be good candidates for source code generation, but I haven't opted for that for now since it was easier to write it by hand (there are many exemptions to the "rules" of the block dispatch, so even if the code looks like it could be generated at the first glance the generator would actually get more complicated to handle all such cases). I've let this in TODO bucket. Tomas _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From dremy at microsoft.com Thu Jul 10 21:21:06 2008 From: dremy at microsoft.com (Dave Remy) Date: Thu, 10 Jul 2008 18:21:06 -0700 Subject: [Ironruby-core] Code Review: StringGsubVersion Message-ID: tfpt review "/shelveset:StringGsubVersion;REDMOND\dremy" Comment : This change fixes all of the gsub, each, and each_line specs. Added aditional tests to gsub spec. The excluded gsub tests are only failing on unimplemented string methods (lstrip!, rstrip!) Implemented simple mutablestring version that is used to assure that mutation hasn't occurred during an iteration. tring#gsub with pattern and block sets $~ for access from the block String#gsub with pattern and block restores $~ after leaving the block String#gsub with pattern and block sets $~ to MatchData of last match and nil when there's none for access from outside String#gsub with pattern and block raises a RuntimeError if the string is modified while substituting This is a revision on a previous shelfset I submitted for these changes. Based on feedback I did a few things differently. I didn't do anything special with version it is just a simple property implementation at this point. Shri's points about potential multi-threaded issues and overflow are well taken but it sounds like we will pick this up as requirements dictate. I did revise the way version gets incremented. Rather than creating a single function that both requires frozen and increments (didn't someone say sometime that function should only do one thing ;)) I think this was a good thing, I ended up putting the Version++ in the logical place close to where a mutation actually occurs. This combined with putting tests for each mutation method caught some scenarios that would have failed. -------------- next part -------------- A non-text attachment was scrubbed... Name: StringGsubVersion.diff Type: application/octet-stream Size: 20202 bytes Desc: StringGsubVersion.diff URL: From Tomas.Matousek at microsoft.com Thu Jul 10 21:46:24 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Thu, 10 Jul 2008 18:46:24 -0700 Subject: [Ironruby-core] Code Review: StringGsubVersion In-Reply-To: References: Message-ID: I think it would be better to increment the version inside MutableString class (not in -Ops). I can do that transition when I'll revisit frozen/tainted flags on MutableString. So go ahead with your change. Tomas -----Original Message----- From: Dave Remy Sent: Thursday, July 10, 2008 6:21 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: Code Review: StringGsubVersion tfpt review "/shelveset:StringGsubVersion;REDMOND\dremy" Comment : This change fixes all of the gsub, each, and each_line specs. Added aditional tests to gsub spec. The excluded gsub tests are only failing on unimplemented string methods (lstrip!, rstrip!) Implemented simple mutablestring version that is used to assure that mutation hasn't occurred during an iteration. tring#gsub with pattern and block sets $~ for access from the block String#gsub with pattern and block restores $~ after leaving the block String#gsub with pattern and block sets $~ to MatchData of last match and nil when there's none for access from outside String#gsub with pattern and block raises a RuntimeError if the string is modified while substituting This is a revision on a previous shelfset I submitted for these changes. Based on feedback I did a few things differently. I didn't do anything special with version it is just a simple property implementation at this point. Shri's points about potential multi-threaded issues and overflow are well taken but it sounds like we will pick this up as requirements dictate. I did revise the way version gets incremented. Rather than creating a single function that both requires frozen and increments (didn't someone say sometime that function should only do one thing ;)) I think this was a good thing, I ended up putting the Version++ in the logical place close to where a mutation actually occurs. This combined with putting tests for each mutation method caught some scenarios that would have failed. From dremy at microsoft.com Thu Jul 10 21:50:54 2008 From: dremy at microsoft.com (Dave Remy) Date: Thu, 10 Jul 2008 18:50:54 -0700 Subject: [Ironruby-core] Code Review: StringGsubVersion In-Reply-To: References: Message-ID: Ok, cool. Sounds good. Are you thinking of a method in MutableString like IncrementVersion()? Or something? -----Original Message----- From: Tomas Matousek Sent: Thursday, July 10, 2008 6:46 PM To: Dave Remy; IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: RE: Code Review: StringGsubVersion I think it would be better to increment the version inside MutableString class (not in -Ops). I can do that transition when I'll revisit frozen/tainted flags on MutableString. So go ahead with your change. Tomas -----Original Message----- From: Dave Remy Sent: Thursday, July 10, 2008 6:21 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: Code Review: StringGsubVersion tfpt review "/shelveset:StringGsubVersion;REDMOND\dremy" Comment : This change fixes all of the gsub, each, and each_line specs. Added aditional tests to gsub spec. The excluded gsub tests are only failing on unimplemented string methods (lstrip!, rstrip!) Implemented simple mutablestring version that is used to assure that mutation hasn't occurred during an iteration. tring#gsub with pattern and block sets $~ for access from the block String#gsub with pattern and block restores $~ after leaving the block String#gsub with pattern and block sets $~ to MatchData of last match and nil when there's none for access from outside String#gsub with pattern and block raises a RuntimeError if the string is modified while substituting This is a revision on a previous shelfset I submitted for these changes. Based on feedback I did a few things differently. I didn't do anything special with version it is just a simple property implementation at this point. Shri's points about potential multi-threaded issues and overflow are well taken but it sounds like we will pick this up as requirements dictate. I did revise the way version gets incremented. Rather than creating a single function that both requires frozen and increments (didn't someone say sometime that function should only do one thing ;)) I think this was a good thing, I ended up putting the Version++ in the logical place close to where a mutation actually occurs. This combined with putting tests for each mutation method caught some scenarios that would have failed. From Tomas.Matousek at microsoft.com Thu Jul 10 21:52:13 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Thu, 10 Jul 2008 18:52:13 -0700 Subject: [Ironruby-core] Code Review: StringGsubVersion In-Reply-To: References: Message-ID: The version updates would be internal detail of MutableString. You would be able to ask for the current version from other classes, but not change it. The operations on MutableString will do. Tomas -----Original Message----- From: Dave Remy Sent: Thursday, July 10, 2008 6:51 PM To: Tomas Matousek; IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: RE: Code Review: StringGsubVersion Ok, cool. Sounds good. Are you thinking of a method in MutableString like IncrementVersion()? Or something? -----Original Message----- From: Tomas Matousek Sent: Thursday, July 10, 2008 6:46 PM To: Dave Remy; IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: RE: Code Review: StringGsubVersion I think it would be better to increment the version inside MutableString class (not in -Ops). I can do that transition when I'll revisit frozen/tainted flags on MutableString. So go ahead with your change. Tomas -----Original Message----- From: Dave Remy Sent: Thursday, July 10, 2008 6:21 PM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: Code Review: StringGsubVersion tfpt review "/shelveset:StringGsubVersion;REDMOND\dremy" Comment : This change fixes all of the gsub, each, and each_line specs. Added aditional tests to gsub spec. The excluded gsub tests are only failing on unimplemented string methods (lstrip!, rstrip!) Implemented simple mutablestring version that is used to assure that mutation hasn't occurred during an iteration. tring#gsub with pattern and block sets $~ for access from the block String#gsub with pattern and block restores $~ after leaving the block String#gsub with pattern and block sets $~ to MatchData of last match and nil when there's none for access from outside String#gsub with pattern and block raises a RuntimeError if the string is modified while substituting This is a revision on a previous shelfset I submitted for these changes. Based on feedback I did a few things differently. I didn't do anything special with version it is just a simple property implementation at this point. Shri's points about potential multi-threaded issues and overflow are well taken but it sounds like we will pick this up as requirements dictate. I did revise the way version gets incremented. Rather than creating a single function that both requires frozen and increments (didn't someone say sometime that function should only do one thing ;)) I think this was a good thing, I ended up putting the Version++ in the logical place close to where a mutation actually occurs. This combined with putting tests for each mutation method caught some scenarios that would have failed. From jdeville at microsoft.com Fri Jul 11 02:17:11 2008 From: jdeville at microsoft.com (Jim Deville) Date: Thu, 10 Jul 2008 23:17:11 -0700 Subject: [Ironruby-core] More spec review Message-ID: I've commited some more stuff to my Rubyspec fork, I'd like to get some reviews before I merge them to Johns. The commit numbers are: 123084 fb9a920 b7d69ef ddbab38 a2a3c4b cc48d366 That's the last 6 commits I've made. Thanks. ~~~~ JD http://blog.jredville.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From bacondarwin at googlemail.com Fri Jul 11 15:39:48 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Fri, 11 Jul 2008 20:39:48 +0100 Subject: [Ironruby-core] More spec review In-Reply-To: References: Message-ID: <003e01c8e38d$e27b3b80$a771b280$@com> Hi Jim, Great stuff. The more I use Git and GitHub the more I love it. I started to write comments, in GitHub, against the commits but then thought better of it. I don't know what happens to them. Sorry if some of these comments are pedantic. Feel free to ignore them. Pete Commit 12308400 - There is a typo in the description of the it "yields while decreaing[sic] self until it less than ceil for a Float endpoint" description - Should you also consider testing downto and upto towards negative floats since Ruby has different semantics for ceil than some other languages? E.g. 3.downto(-1.3) should produce [3, 2, 1, 0, -1] rather than [3,2,1,0,-1,-2]. Commit fb9a920 - Do you need to remove the constant that is created inside it "should perform top level evaluations from inside a block" to prevent leakage of memory and/or semantic? Commit cc48d366 - There is a typo in the description of the it "includes the curret[sic] directory" description. - My version of MRI Ruby does not have a RUBY_ENGINE constant. If this is the case does the whole it clause need to be wrapped in a compliant_on() block? - There is a typo in the description of the it "allow substitutionof[sic] literal regexps" description. From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Friday,11 July 11, 2008 07:17 To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers; Srivatsn Narayanan Subject: [Ironruby-core] More spec review I've commited some more stuff to my Rubyspec fork, I'd like to get some reviews before I merge them to Johns. The commit numbers are: 123084 fb9a920 b7d69ef ddbab38 a2a3c4b cc48d366 That's the last 6 commits I've made. Thanks. ~~~~ JD http://blog.jredville.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From Jimmy.Schementi at microsoft.com Fri Jul 11 17:02:58 2008 From: Jimmy.Schementi at microsoft.com (Jimmy Schementi) Date: Fri, 11 Jul 2008 14:02:58 -0700 Subject: [Ironruby-core] More spec review In-Reply-To: <003e01c8e38d$e27b3b80$a771b280$@com> References: <003e01c8e38d$e27b3b80$a771b280$@com> Message-ID: <5283CA0A4168DF4FBBD71AE9ECA5A32845D45E2522@NA-EXMSG-C116.redmond.corp.microsoft.com> Peter, Yes, git is awesome. We're new to it as well, so we're all learning about to use it most effectively. One awesome github feature is that you can comment on specific commits from the website. Putting any conversation about commits right on that page is an awesome way to "archive" those types of communications; the more context with the code the better =) Plus, those comments will always follow that change, so when specific changes get merged back into John's ruby-spec fork, or even the master ruby-spec repo, those comments follow it. =) So, does everyone agree that any comments associated with a git commit (eg. code-reviews, etc) should be done on github itself? Of course, it would be extra helpful to send out the link to the commit to ironruby-core =) ~Jimmy From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Friday, July 11, 2008 12:40 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] More spec review Hi Jim, Great stuff. The more I use Git and GitHub the more I love it. I started to write comments, in GitHub, against the commits but then thought better of it. I don't know what happens to them. Sorry if some of these comments are pedantic. Feel free to ignore them. Pete Commit 12308400 - There is a typo in the description of the it "yields while decreaing[sic] self until it less than ceil for a Float endpoint" description - Should you also consider testing downto and upto towards negative floats since Ruby has different semantics for ceil than some other languages? E.g. 3.downto(-1.3) should produce [3, 2, 1, 0, -1] rather than [3,2,1,0,-1,-2]. Commit fb9a920 - Do you need to remove the constant that is created inside it "should perform top level evaluations from inside a block" to prevent leakage of memory and/or semantic? Commit cc48d366 - There is a typo in the description of the it "includes the curret[sic] directory" description. - My version of MRI Ruby does not have a RUBY_ENGINE constant. If this is the case does the whole it clause need to be wrapped in a compliant_on() block? - There is a typo in the description of the it "allow substitutionof[sic] literal regexps" description. From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Friday,11 July 11, 2008 07:17 To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers; Srivatsn Narayanan Subject: [Ironruby-core] More spec review I've commited some more stuff to my Rubyspec fork, I'd like to get some reviews before I merge them to Johns. The commit numbers are: 123084 fb9a920 b7d69ef ddbab38 a2a3c4b cc48d366 That's the last 6 commits I've made. Thanks. ~~~~ JD http://blog.jredville.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdeville at microsoft.com Fri Jul 11 17:16:37 2008 From: jdeville at microsoft.com (Jim Deville) Date: Fri, 11 Jul 2008 14:16:37 -0700 Subject: [Ironruby-core] More spec review In-Reply-To: <003e01c8e38d$e27b3b80$a771b280$@com> References: <003e01c8e38d$e27b3b80$a771b280$@com> Message-ID: inline Changes commited as: d3d88c3 JD From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Friday, July 11, 2008 12:40 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] More spec review Hi Jim, Great stuff. The more I use Git and GitHub the more I love it. I started to write comments, in GitHub, against the commits but then thought better of it. I don't know what happens to them. Sorry if some of these comments are pedantic. Feel free to ignore them. Pete Commit 12308400 - There is a typo in the description of the it "yields while decreaing[sic] self until it less than ceil for a Float endpoint" description fixed - Should you also consider testing downto and upto towards negative floats since Ruby has different semantics for ceil than some other languages? E.g. 3.downto(-1.3) should produce [3, 2, 1, 0, -1] rather than [3,2,1,0,-1,-2]. Good catch. Tests added. Commit fb9a920 - Do you need to remove the constant that is created inside it "should perform top level evaluations from inside a block" to prevent leakage of memory and/or semantic? done Commit cc48d366 - There is a typo in the description of the it "includes the curret[sic] directory" description. - My version of MRI Ruby does not have a RUBY_ENGINE constant. If this is the case does the whole it clause need to be wrapped in a compliant_on() block? - There is a typo in the description of the it "allow substitutionof[sic] literal regexps" description. Typo's fixed. On #rubyspec Brian suggested we leave it as is to add an incentive to MRI. If it bugs you we can do a version guard. From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Friday,11 July 11, 2008 07:17 To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers; Srivatsn Narayanan Subject: [Ironruby-core] More spec review I've commited some more stuff to my Rubyspec fork, I'd like to get some reviews before I merge them to Johns. The commit numbers are: 123084 fb9a920 b7d69ef ddbab38 a2a3c4b cc48d366 That's the last 6 commits I've made. Thanks. ~~~~ JD http://blog.jredville.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdeville at microsoft.com Fri Jul 11 17:47:52 2008 From: jdeville at microsoft.com (Jim Deville) Date: Fri, 11 Jul 2008 14:47:52 -0700 Subject: [Ironruby-core] MSpec in SVN/TFS closing down Message-ID: Now that I've ported the MSpec changes to Git, I'll be getting ready to delete the version in the SVN/TFS repository. Before I do that I wanted to make sure there aren't any outstanding patches that need to be incorporated. I know of one from Dave Remy, and I believe there is one on the Rubyforge site. Are there any others? Thanks, ~~~~ JD http://blog.jredville.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From lists at ruby-forum.com Fri Jul 11 22:12:01 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Sat, 12 Jul 2008 04:12:01 +0200 Subject: [Ironruby-core] IronRuby and Strong Name Verification Skipping Message-ID: Hi All, I was wondering if anyone could shed any light on how strong names, assembly signing and verification skipping is working with IronRuby now? After checking out the latest version of IronRuby from svn I was able to build ok but when running any application that used IronRuby (including the ir console) I would get assembly load errors normally to do with the fact that a Microsoft.Scripting could not be found, the error related to the fact that a signed version of the assembly could not be found. After a quick look around I found the SIGNED compilation option and switched that off but that didn't help. I then tried to turn off assembly signing in for each of the IronRuby assemblies in VS.Net 2008. That didn't work but it seems to be a bug in VS.Net becuase even though I unticked the option to sign the assemblies the build would fail if I moved the key that VS.Net supposedly shouldn't have been looking for. So I then resorted to editing the project files to remove all options relating to strong naming and that worked. I was able to get applications to load the assemblies and run correctly. That was about a month ago and I have since re-checked out the IronRuby code and thought it would be better to get this string naming thing sorted out. Initially I hit all the same problems and then happended upon the command below in runfirst.cmd: sn -Vr *,31bf3856ad364e35 After running that command all the apps that use IronRuby (including the ir console) work fine with the default signed assemblies. But if you have to turn off assembly strong name verification to get the IronRuby assemblies to load what's the point of signing them in the first place? Regards, Aaron -- Posted via http://www.ruby-forum.com/. From michael.letterle at gmail.com Fri Jul 11 22:25:53 2008 From: michael.letterle at gmail.com (Michael Letterle) Date: Fri, 11 Jul 2008 22:25:53 -0400 Subject: [Ironruby-core] IronRuby and Strong Name Verification Skipping In-Reply-To: References: Message-ID: The key that is in SVN is only a test key. If you notice, the solutions are set up to DELAY sign, not FULLY sign. Any delay signed assemblies have to be registered for verification skipping. This is a common development technique to ensure the build scripts and everything work correctly, without having to put your true key out in the wild. When Microsoft officially releases IR, they will fully sign it using a true private key. On Fri, Jul 11, 2008 at 10:12 PM, Aaron Clauson wrote: > Hi All, > > I was wondering if anyone could shed any light on how strong names, > assembly signing and verification skipping is working with IronRuby now? > > After checking out the latest version of IronRuby from svn I was able to > build ok but when running any application that used IronRuby (including > the ir console) I would get assembly load errors normally to do with the > fact that a Microsoft.Scripting could not be found, the error related to > the fact that a signed version of the assembly could not be found. > > After a quick look around I found the SIGNED compilation option and > switched that off but that didn't help. I then tried to turn off > assembly signing in for each of the IronRuby assemblies in VS.Net 2008. > That didn't work but it seems to be a bug in VS.Net becuase even though > I unticked the option to sign the assemblies the build would fail if I > moved the key that VS.Net supposedly shouldn't have been looking for. So > I then resorted to editing the project files to remove all options > relating to strong naming and that worked. I was able to get > applications to load the assemblies and run correctly. > > That was about a month ago and I have since re-checked out the IronRuby > code and thought it would be better to get this string naming thing > sorted out. Initially I hit all the same problems and then happended > upon the command below in runfirst.cmd: > > sn -Vr *,31bf3856ad364e35 > > After running that command all the apps that use IronRuby (including the > ir console) work fine with the default signed assemblies. But if you > have to turn off assembly strong name verification to get the IronRuby > assemblies to load what's the point of signing them in the first place? > > Regards, > > Aaron > -- > Posted via http://www.ruby-forum.com/. > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > -- Michael Letterle [Polymath Prokrammer] http://blog.prokrams.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From lists at ruby-forum.com Sat Jul 12 08:49:23 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Sat, 12 Jul 2008 14:49:23 +0200 Subject: [Ironruby-core] IronRuby and Strong Name Verification Skipping In-Reply-To: References: Message-ID: Michael Letterle wrote: > The key that is in SVN is only a test key. If you notice, the solutions > are > set up to DELAY sign, not FULLY sign. Any delay signed assemblies have > to > be registered for verification skipping. This is a common development > technique to ensure the build scripts and everything work correctly, > without > having to put your true key out in the wild. > > When Microsoft officially releases IR, they will fully sign it using a > true > private key. > > On Fri, Jul 11, 2008 at 10:12 PM, Aaron Clauson Perfect. Thanks for the explanation. -- Posted via http://www.ruby-forum.com/. From lists at ruby-forum.com Sat Jul 12 08:59:52 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Sat, 12 Jul 2008 14:59:52 +0200 Subject: [Ironruby-core] Memory Usage Tests Message-ID: <85f3d2ddd732e815a141ef49d0cc9f5d@ruby-forum.com> Hi All, In order to track down some high memory usage problems I have in a server process that is using the IronRuby engine I have run a test using a single line script and have encountered an issue that looks suspiciously like a memory leak. I'm scouring through the IronRuby code commenting out a few static lists as I find them to see if they make any difference but to date haven't come across anything useful. The test I'm running is: static void Main(string[] args) { ScriptRuntime scriptRuntime = IronRuby.CreateRuntime(); for (int index = 0; index < 1000; index++) { ScriptScope rubyScope = scriptRuntime.CreateScope("IronRuby"); rubyScope.Execute("print 'execute\n'"); } Console.WriteLine("finished"); Console.ReadLine(); } Executing this three times on 3 separate runs the resultant process had memory allocations of the folowing amounts as shown in the Windows Task Manager: 31,940k 31,936k 31,936k Executing the loop 10,000 times on 3 separate runs resulted in memory usage of: 133,648k 134,036k 133,808k Changing to static void Main(string[] args) { for (int index = 0; index < 1000; index++) { Console.WriteLine("execute"); } Console.WriteLine("finished"); Console.ReadLine(); } For 1,000 iterations memory usage was 8,402k. For 10,000 iterations memory usage was 8,704k. For 1,000,000 iterations memory usage was 9,348. The reason I'm looking into this is that I am having to jeep a close eye on my server process and re-start it every 3 or 4 days when the memory usage gets up to 300 to 400MB. Regards, Aaron -- Posted via http://www.ruby-forum.com/. From bacondarwin at googlemail.com Sat Jul 12 09:59:33 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Sat, 12 Jul 2008 14:59:33 +0100 Subject: [Ironruby-core] Memory Usage Tests In-Reply-To: <85f3d2ddd732e815a141ef49d0cc9f5d@ruby-forum.com> References: <85f3d2ddd732e815a141ef49d0cc9f5d@ruby-forum.com> Message-ID: <002401c8e427$852a52a0$8f7ef7e0$@com> Additional info... If you move the scriptRuntime.CreateScope method outside the loop then there is no increase in memory. Pete -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Aaron Clauson Sent: Saturday,12 July 12, 2008 14:00 To: ironruby-core at rubyforge.org Subject: [Ironruby-core] Memory Usage Tests Hi All, In order to track down some high memory usage problems I have in a server process that is using the IronRuby engine I have run a test using a single line script and have encountered an issue that looks suspiciously like a memory leak. I'm scouring through the IronRuby code commenting out a few static lists as I find them to see if they make any difference but to date haven't come across anything useful. The test I'm running is: static void Main(string[] args) { ScriptRuntime scriptRuntime = IronRuby.CreateRuntime(); for (int index = 0; index < 1000; index++) { ScriptScope rubyScope = scriptRuntime.CreateScope("IronRuby"); rubyScope.Execute("print 'execute\n'"); } Console.WriteLine("finished"); Console.ReadLine(); } Executing this three times on 3 separate runs the resultant process had memory allocations of the folowing amounts as shown in the Windows Task Manager: 31,940k 31,936k 31,936k Executing the loop 10,000 times on 3 separate runs resulted in memory usage of: 133,648k 134,036k 133,808k Changing to static void Main(string[] args) { for (int index = 0; index < 1000; index++) { Console.WriteLine("execute"); } Console.WriteLine("finished"); Console.ReadLine(); } For 1,000 iterations memory usage was 8,402k. For 10,000 iterations memory usage was 8,704k. For 1,000,000 iterations memory usage was 9,348. The reason I'm looking into this is that I am having to jeep a close eye on my server process and re-start it every 3 or 4 days when the memory usage gets up to 300 to 400MB. Regards, Aaron -- Posted via http://www.ruby-forum.com/. _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From bacondarwin at googlemail.com Sat Jul 12 10:24:27 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Sat, 12 Jul 2008 15:24:27 +0100 Subject: [Ironruby-core] Memory Usage Tests In-Reply-To: <85f3d2ddd732e815a141ef49d0cc9f5d@ruby-forum.com> References: <85f3d2ddd732e815a141ef49d0cc9f5d@ruby-forum.com> Message-ID: <002501c8e42a$fee77480$fcb65d80$@com> Also... If you only have the CreateScope method and not the Execute method in the loop then there is no memory increase. With both in the loop you get a linear increase in memory usage, even if you regularly force a Garbage Collection, of approximately 10K per iteration. Pete -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Aaron Clauson Sent: Saturday,12 July 12, 2008 14:00 To: ironruby-core at rubyforge.org Subject: [Ironruby-core] Memory Usage Tests Hi All, In order to track down some high memory usage problems I have in a server process that is using the IronRuby engine I have run a test using a single line script and have encountered an issue that looks suspiciously like a memory leak. I'm scouring through the IronRuby code commenting out a few static lists as I find them to see if they make any difference but to date haven't come across anything useful. The test I'm running is: static void Main(string[] args) { ScriptRuntime scriptRuntime = IronRuby.CreateRuntime(); for (int index = 0; index < 1000; index++) { ScriptScope rubyScope = scriptRuntime.CreateScope("IronRuby"); rubyScope.Execute("print 'execute\n'"); } Console.WriteLine("finished"); Console.ReadLine(); } Executing this three times on 3 separate runs the resultant process had memory allocations of the folowing amounts as shown in the Windows Task Manager: 31,940k 31,936k 31,936k Executing the loop 10,000 times on 3 separate runs resulted in memory usage of: 133,648k 134,036k 133,808k Changing to static void Main(string[] args) { for (int index = 0; index < 1000; index++) { Console.WriteLine("execute"); } Console.WriteLine("finished"); Console.ReadLine(); } For 1,000 iterations memory usage was 8,402k. For 10,000 iterations memory usage was 8,704k. For 1,000,000 iterations memory usage was 9,348. The reason I'm looking into this is that I am having to jeep a close eye on my server process and re-start it every 3 or 4 days when the memory usage gets up to 300 to 400MB. Regards, Aaron -- Posted via http://www.ruby-forum.com/. _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From lists at ruby-forum.com Sat Jul 12 10:30:04 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Sat, 12 Jul 2008 16:30:04 +0200 Subject: [Ironruby-core] Memory Usage Tests In-Reply-To: <002401c8e427$852a52a0$8f7ef7e0$@com> References: <85f3d2ddd732e815a141ef49d0cc9f5d@ruby-forum.com> <002401c8e427$852a52a0$8f7ef7e0$@com> Message-ID: Peter Bacon Darwin wrote: > Additional info... > If you move the scriptRuntime.CreateScope method outside the loop then > there > is no increase in memory. But from my understanding of script scopes I need a new scope when using local variables. I can have multiple scripts executing on different threads with different local variables so as far as I'm aware creating a new scope each time is the only way to allow that? static void Main(string[] args) { ScriptRuntime scriptRuntime = IronRuby.CreateRuntime(); for (int index = 0; index < 1000; index++) { ScriptScope rubyScope = scriptRuntime.CreateScope("IronRuby"); rubyScope.SetVariable("local1", "hello"); rubyScope.SetVariable("local2", "world"); rubyScope.Execute("print \"#{local1} #{local2}\n\""); } Console.WriteLine("finished"); Console.ReadLine(); } Regards, Aaron -- Posted via http://www.ruby-forum.com/. From lists at ruby-forum.com Sat Jul 12 10:32:54 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Sat, 12 Jul 2008 16:32:54 +0200 Subject: [Ironruby-core] Memory Usage Tests In-Reply-To: <002501c8e42a$fee77480$fcb65d80$@com> References: <85f3d2ddd732e815a141ef49d0cc9f5d@ruby-forum.com> <002501c8e42a$fee77480$fcb65d80$@com> Message-ID: Peter Bacon Darwin wrote: > Also... > If you only have the CreateScope method and not the Execute method in > the > loop then there is no memory increase. > With both in the loop you get a linear increase in memory usage, even if > you > regularly force a Garbage Collection, of approximately 10K per > iteration. > Pete Hi Pete, If I take the Execute method out of the loop then I don't get any significant increase in memory usage. However as per my post above I think I do need the CreateScope and Execute calls in the loop to allow for multiple simultaneous script executions. Regards, Aaron -- Posted via http://www.ruby-forum.com/. From lists at ruby-forum.com Sat Jul 12 11:53:06 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Sat, 12 Jul 2008 17:53:06 +0200 Subject: [Ironruby-core] Memory Usage Tests In-Reply-To: References: <85f3d2ddd732e815a141ef49d0cc9f5d@ruby-forum.com> <002501c8e42a$fee77480$fcb65d80$@com> Message-ID: Looks like keeping a limit on the number of ScriptScope's being used is the best approach for a long running process which makes sense. The code below does not have a memory usage problem. class Program { private static Queue m_scriptScopes = new Queue(); private static AutoResetEvent m_scriptScopeAvailable = new AutoResetEvent(false); static void Main(string[] args) { ScriptRuntime m_scriptRuntime = IronRuby.CreateRuntime(); m_scriptScopes.Enqueue(m_scriptRuntime.CreateScope("IronRuby")); m_scriptScopes.Enqueue(m_scriptRuntime.CreateScope("IronRuby")); m_scriptScopes.Enqueue(m_scriptRuntime.CreateScope("IronRuby")); for (int index = 0; index < 10000; index++) { ScriptScope rubyScope = null; while (rubyScope == null) { lock (m_scriptScopes) { if (m_scriptScopes.Count > 0) { rubyScope = m_scriptScopes.Dequeue(); } else { m_scriptScopeAvailable.Reset(); } } if (rubyScope == null) { m_scriptScopeAvailable.WaitOne(); } } ThreadPool.QueueUserWorkItem(new WaitCallback(RunRubyScript), new object[] {rubyScope, index}); } Console.ReadLine(); } private static void RunRubyScript(object state) { ScriptScope rubyScope = (ScriptScope)((object[])state)[0]; int count = (int)((object[])state)[1]; rubyScope.ClearVariables(); rubyScope.SetVariable("local1", count); rubyScope.SetVariable("local2", count); rubyScope.Execute("print \"#{local1} - #{local2}\n\""); lock (m_scriptScopes) { m_scriptScopes.Enqueue(rubyScope); m_scriptScopeAvailable.Set(); } } } Regards, Aaron -- Posted via http://www.ruby-forum.com/. From bacondarwin at googlemail.com Sat Jul 12 12:44:24 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Sat, 12 Jul 2008 17:44:24 +0100 Subject: [Ironruby-core] Memory Usage Tests In-Reply-To: References: <85f3d2ddd732e815a141ef49d0cc9f5d@ruby-forum.com> <002501c8e42a$fee77480$fcb65d80$@com> Message-ID: <002901c8e43e$8bc68e50$a353aaf0$@com> Hi Aaron, Sorry I wasn't trying to suggest a solution for your problem. I was just trying to narrow down where the increase in memory is coming from. I suspect that the cause is initializing or calling into the core libraries but I can't isolate that bit. Pete -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Aaron Clauson Sent: Saturday,12 July 12, 2008 15:33 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Memory Usage Tests Peter Bacon Darwin wrote: > Also... > If you only have the CreateScope method and not the Execute method in > the > loop then there is no memory increase. > With both in the loop you get a linear increase in memory usage, even if > you > regularly force a Garbage Collection, of approximately 10K per > iteration. > Pete Hi Pete, If I take the Execute method out of the loop then I don't get any significant increase in memory usage. However as per my post above I think I do need the CreateScope and Execute calls in the loop to allow for multiple simultaneous script executions. Regards, Aaron -- Posted via http://www.ruby-forum.com/. _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From Tomas.Matousek at microsoft.com Sat Jul 12 13:08:57 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Sat, 12 Jul 2008 10:08:57 -0700 Subject: [Ironruby-core] Memory Usage Tests In-Reply-To: <002901c8e43e$8bc68e50$a353aaf0$@com> References: <85f3d2ddd732e815a141ef49d0cc9f5d@ruby-forum.com> <002501c8e42a$fee77480$fcb65d80$@com> <002901c8e43e$8bc68e50$a353aaf0$@com> Message-ID: I'm working on working set issues right now. It should be better soon. Will look at this one especially. Thanks, Tomas -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Saturday, July 12, 2008 9:44 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Memory Usage Tests Hi Aaron, Sorry I wasn't trying to suggest a solution for your problem. I was just trying to narrow down where the increase in memory is coming from. I suspect that the cause is initializing or calling into the core libraries but I can't isolate that bit. Pete -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Aaron Clauson Sent: Saturday,12 July 12, 2008 15:33 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Memory Usage Tests Peter Bacon Darwin wrote: > Also... > If you only have the CreateScope method and not the Execute method in > the > loop then there is no memory increase. > With both in the loop you get a linear increase in memory usage, even if > you > regularly force a Garbage Collection, of approximately 10K per > iteration. > Pete Hi Pete, If I take the Execute method out of the loop then I don't get any significant increase in memory usage. However as per my post above I think I do need the CreateScope and Execute calls in the loop to allow for multiple simultaneous script executions. Regards, Aaron -- Posted via http://www.ruby-forum.com/. _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From lists at ruby-forum.com Sat Jul 12 19:15:36 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Sun, 13 Jul 2008 01:15:36 +0200 Subject: [Ironruby-core] Memory Usage Tests In-Reply-To: <002901c8e43e$8bc68e50$a353aaf0$@com> References: <85f3d2ddd732e815a141ef49d0cc9f5d@ruby-forum.com> <002501c8e42a$fee77480$fcb65d80$@com> <002901c8e43e$8bc68e50$a353aaf0$@com> Message-ID: Peter Bacon Darwin wrote: > Hi Aaron, > Sorry I wasn't trying to suggest a solution for your problem. I was > just > trying to narrow down where the increase in memory is coming from. > I suspect that the cause is initializing or calling into the core > libraries > but I can't isolate that bit. > Pete Hi Pete, Thanks for the suggestion in the first place! I realise IronRuby is still under heavy development so expect to have to do the odd workaround. Tomas thanks for taking a look. Regards, Aaron -- Posted via http://www.ruby-forum.com/. From lists at ruby-forum.com Sat Jul 12 19:33:10 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Sun, 13 Jul 2008 01:33:10 +0200 Subject: [Ironruby-core] Memory Usage Tests In-Reply-To: References: <85f3d2ddd732e815a141ef49d0cc9f5d@ruby-forum.com> <002501c8e42a$fee77480$fcb65d80$@com> <002901c8e43e$8bc68e50$a353aaf0$@com> Message-ID: <1a06056678d8bab2391f047012d89068@ruby-forum.com> Tomas Matousek wrote: > I'm working on working set issues right now. It should be better soon. > Will look at this one especially. > > Thanks, > Tomas Hi Tomas, I didn't quite nail down where the high memory usage was coming from either but I did get to a stage where I could determine it was happening after the RubyScope was being created. So up until the call to InvokeTarget in the ScriptCode everything was fine. That obersvation is probably not particularly useful since all the magic seems to happen when InvokeTarget is called. Another observation is that in the SymbolTable class in Microsoft.Scripting.Core there are only objects being put onto the two static dictionaries. Nothing is ever taken off them. I would have thought the SymbolTable would need to somehow know to remove objects after the ScriptScope that puts them on has diesd. Regards, Aaron -- Posted via http://www.ruby-forum.com/. From bacondarwin at googlemail.com Sun Jul 13 16:45:32 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Sun, 13 Jul 2008 21:45:32 +0100 Subject: [Ironruby-core] Adding methods to Kernel Message-ID: <000b01c8e529$662bb760$32832620$@com> I am still having problems adding the BigDecimal method into the Kernel module. As I understand it the correct method is to create a new class that extends Kernel and adds the methods as both Private-Instance and Public-Singleton. This is the code: namespace Ruby.StandardLibrary.BigDecimal { [RubyModule(Extends = typeof(Kernel))] public static class KernelOps { [RubyMethod("BigDecimal", RubyMethodAttributes.PrivateInstance)] [RubyMethod("BigDecimal", RubyMethodAttributes.PublicSingleton)] public static BigDecimal/*!*/ CreateBigDecimal(CodeContext/*!*/ context, [NotNull]MutableString/*!*/ value, [Optional]int n) { return BigDecimalOps.CreateBigDecimal(value, n); } [RubyMethod("BigDecimal", RubyMethodAttributes.PrivateInstance)] [RubyMethod("BigDecimal", RubyMethodAttributes.PublicSingleton)] public static BigDecimal/*!*/ CreateBigDecimal(CodeContext/*!*/ context, object value, [Optional]int n) { return BigDecimalOps.CreateBigDecimal(context, value, n); } } } This appears to create the methods (you can see them in Kernel.methods) but it is not possible to call them. You always get an ArgumentError thrown with "incorrect type or number of parameters" as a message. I have been digging through the code but I can't work out why this is happening. One thing that seems a little worrying but not necessarily wrong is that it appears that the target (self) has a RubyClass of type Ruby.Builtins. BuiltinsLibraryInitializer, which seems rather bizarre. Currently I am able to get around this by defining the methods in Ruby itself, I changed the bigdecimal.rb file in the libs folder to: load_assembly 'IronRuby.Libraries', 'Ruby.StandardLibrary.BigDecimal' class Object def BigDecimal(v, n = 0) BigDecimal.new(v,n) end end Any ideas? Pete -------------- next part -------------- An HTML attachment was scrubbed... URL: From curth at microsoft.com Sun Jul 13 16:47:51 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Sun, 13 Jul 2008 13:47:51 -0700 Subject: [Ironruby-core] Adding methods to Kernel In-Reply-To: <000b01c8e529$662bb760$32832620$@com> References: <000b01c8e529$662bb760$32832620$@com> Message-ID: You need to declare a "self" parameter, even though it will be unused. From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Sunday, July 13, 2008 1:46 PM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] Adding methods to Kernel I am still having problems adding the BigDecimal method into the Kernel module. As I understand it the correct method is to create a new class that extends Kernel and adds the methods as both Private-Instance and Public-Singleton. This is the code: namespace Ruby.StandardLibrary.BigDecimal { [RubyModule(Extends = typeof(Kernel))] public static class KernelOps { [RubyMethod("BigDecimal", RubyMethodAttributes.PrivateInstance)] [RubyMethod("BigDecimal", RubyMethodAttributes.PublicSingleton)] public static BigDecimal/*!*/ CreateBigDecimal(CodeContext/*!*/ context, [NotNull]MutableString/*!*/ value, [Optional]int n) { return BigDecimalOps.CreateBigDecimal(value, n); } [RubyMethod("BigDecimal", RubyMethodAttributes.PrivateInstance)] [RubyMethod("BigDecimal", RubyMethodAttributes.PublicSingleton)] public static BigDecimal/*!*/ CreateBigDecimal(CodeContext/*!*/ context, object value, [Optional]int n) { return BigDecimalOps.CreateBigDecimal(context, value, n); } } } This appears to create the methods (you can see them in Kernel.methods) but it is not possible to call them. You always get an ArgumentError thrown with "incorrect type or number of parameters" as a message. I have been digging through the code but I can't work out why this is happening. One thing that seems a little worrying but not necessarily wrong is that it appears that the target (self) has a RubyClass of type Ruby.Builtins. BuiltinsLibraryInitializer, which seems rather bizarre. Currently I am able to get around this by defining the methods in Ruby itself, I changed the bigdecimal.rb file in the libs folder to: load_assembly 'IronRuby.Libraries', 'Ruby.StandardLibrary.BigDecimal' class Object def BigDecimal(v, n = 0) BigDecimal.new(v,n) end end Any ideas? Pete -------------- next part -------------- An HTML attachment was scrubbed... URL: From bacondarwin at googlemail.com Sun Jul 13 17:15:36 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Sun, 13 Jul 2008 22:15:36 +0100 Subject: [Ironruby-core] Adding methods to Kernel In-Reply-To: References: <000b01c8e529$662bb760$32832620$@com> Message-ID: <002001c8e52d$99028f70$cb07ae50$@com> Magic thanks! I knew you needed this for Singleton methods but the fact that it was a private instance method as well threw me. Now I can start hitting the rubyspec tests for BigDecimal. Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Sunday,13 July 13, 2008 21:48 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Adding methods to Kernel You need to declare a "self" parameter, even though it will be unused. From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Sunday, July 13, 2008 1:46 PM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] Adding methods to Kernel I am still having problems adding the BigDecimal method into the Kernel module. As I understand it the correct method is to create a new class that extends Kernel and adds the methods as both Private-Instance and Public-Singleton. This is the code: namespace Ruby.StandardLibrary.BigDecimal { [RubyModule(Extends = typeof(Kernel))] public static class KernelOps { [RubyMethod("BigDecimal", RubyMethodAttributes.PrivateInstance)] [RubyMethod("BigDecimal", RubyMethodAttributes.PublicSingleton)] public static BigDecimal/*!*/ CreateBigDecimal(CodeContext/*!*/ context, [NotNull]MutableString/*!*/ value, [Optional]int n) { return BigDecimalOps.CreateBigDecimal(value, n); } [RubyMethod("BigDecimal", RubyMethodAttributes.PrivateInstance)] [RubyMethod("BigDecimal", RubyMethodAttributes.PublicSingleton)] public static BigDecimal/*!*/ CreateBigDecimal(CodeContext/*!*/ context, object value, [Optional]int n) { return BigDecimalOps.CreateBigDecimal(context, value, n); } } } This appears to create the methods (you can see them in Kernel.methods) but it is not possible to call them. You always get an ArgumentError thrown with "incorrect type or number of parameters" as a message. I have been digging through the code but I can't work out why this is happening. One thing that seems a little worrying but not necessarily wrong is that it appears that the target (self) has a RubyClass of type Ruby.Builtins. BuiltinsLibraryInitializer, which seems rather bizarre. Currently I am able to get around this by defining the methods in Ruby itself, I changed the bigdecimal.rb file in the libs folder to: load_assembly 'IronRuby.Libraries', 'Ruby.StandardLibrary.BigDecimal' class Object def BigDecimal(v, n = 0) BigDecimal.new(v,n) end end Any ideas? Pete -------------- next part -------------- An HTML attachment was scrubbed... URL: From olegtk at microsoft.com Mon Jul 14 02:20:11 2008 From: olegtk at microsoft.com (Oleg Tkachenko) Date: Sun, 13 Jul 2008 23:20:11 -0700 Subject: [Ironruby-core] MutableString encoding issue Message-ID: Stumbled on this when testing yaml. >>> "\204" => "?" While irb(main):005:0>"\204" => "\204" I believe Ruby string can hold arbitrary byte values, but as we are storing content as a string we are obviously losing all values that cannot be represented in default encoding. Tomas, what do you think? -- Oleg -------------- next part -------------- An HTML attachment was scrubbed... URL: From lists at ruby-forum.com Mon Jul 14 07:50:13 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Mon, 14 Jul 2008 13:50:13 +0200 Subject: [Ironruby-core] Using Local Variable in Silverlight Message-ID: <7d3ae7368a80a972216fb719e3400131@ruby-forum.com> Hi All, I'm getting an exception when I try and use the code snippet below in Silverlight. The snippet works ok with a normal .Net app. What I am trying to do is use a Ruby script running in my Silverlight app to call a method that updates the text on a text block. public delegate void LogMessageDelegate(string message); public partial class Page : UserControl { ... private void RunScript(object sender, System.Windows.RoutedEventArgs e) { ScriptHelper scriptHelper = new ScriptHelper(WriteDebugMessage); ScriptScope scriptScope = IronRuby.CreateRuntime().CreateScope("IronRuby"); scriptScope.SetVariable("sys", scriptHelper); scriptScope.Execute("sys.Log(\"hello world\")"); } private void WriteDebugMessage(string message) { m_scriptDebugText.Text = message; // m_scriptDebugText is a text block. } } public class ScriptHelper { private event LogMessageDelegate m_logMessage; public ScriptHelper(LogMessageDelegate logMessage) { m_logMessage = logMessage; } public void Log(string message) { m_logMessage(message); } } The exception I get back is: "Ruby.Runtime.RubyTopLevelScope.TopMethodMissing(Microsoft.Scripting.Runtime.CodeContext, System.Object, Ruby.Runtime.BlockParam, Microsoft.Scripting.SymbolId, System.Object[])" Is there a better approach to interacting between Silverlight UI elements and the Ruby script? I'm not using any HTML as the whole interface is built of SilverLight controls. Regards, Aaron -- Posted via http://www.ruby-forum.com/. From michael.letterle at gmail.com Mon Jul 14 09:21:09 2008 From: michael.letterle at gmail.com (Michael Letterle) Date: Mon, 14 Jul 2008 09:21:09 -0400 Subject: [Ironruby-core] MutableString encoding issue In-Reply-To: References: Message-ID: This was a known issue a while back, it's the reason the Zlib library didn't work well with binary files. I'm fairly certain there was work being done on making String be backed by a byte array... and in fact I thought this was already done. On Mon, Jul 14, 2008 at 2:20 AM, Oleg Tkachenko wrote: > Stumbled on this when testing yaml. > > > > >>> "\204" > > => "?" > > > > While > > > > irb(main):005:0>"\204" > > => "\204" > > > > I believe Ruby string can hold arbitrary byte values, but as we are storing > content as a string we are obviously losing all values that cannot be > represented in default encoding. Tomas, what do you think? > > > > -- > > Oleg > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > -- Michael Letterle [Polymath Prokrammer] http://blog.prokrams.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From curth at microsoft.com Mon Jul 14 09:27:14 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Mon, 14 Jul 2008 06:27:14 -0700 Subject: [Ironruby-core] MutableString encoding issue In-Reply-To: References: Message-ID: MutableString can have one of three internal representations, depending on how it was last used. One of these is a byte array. This particular problem may be in the scanner or parser and not in the actual string class, as we don't otherwise have a problem storing the character: >>> $s = "\204" => "?" >>> $s[0] => 63 >>> $s[0] = 132 => 132 >>> $s => "\204" >>> From: Michael Letterle [mailto:michael.letterle at gmail.com] Sent: Monday, July 14, 2008 6:21 AM To: ironruby-core at rubyforge.org Cc: IronRuby Team Subject: Re: [Ironruby-core] MutableString encoding issue This was a known issue a while back, it's the reason the Zlib library didn't work well with binary files. I'm fairly certain there was work being done on making String be backed by a byte array... and in fact I thought this was already done. On Mon, Jul 14, 2008 at 2:20 AM, Oleg Tkachenko > wrote: Stumbled on this when testing yaml. >>> "\204" => "?" While irb(main):005:0>"\204" => "\204" I believe Ruby string can hold arbitrary byte values, but as we are storing content as a string we are obviously losing all values that cannot be represented in default encoding. Tomas, what do you think? -- Oleg _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core -- Michael Letterle [Polymath Prokrammer] http://blog.prokrams.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From Tomas.Matousek at microsoft.com Mon Jul 14 12:13:55 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Mon, 14 Jul 2008 09:13:55 -0700 Subject: [Ironruby-core] MutableString encoding issue In-Reply-To: References: Message-ID: I think it's a bug :). Could you file it? If it's something that's blocking you I can look at it asap. Tomas From: Oleg Tkachenko Sent: Sunday, July 13, 2008 11:20 PM To: IronRuby Team Cc: ironruby-core at rubyforge.org Subject: MutableString encoding issue Stumbled on this when testing yaml. >>> "\204" => "?" While irb(main):005:0>"\204" => "\204" I believe Ruby string can hold arbitrary byte values, but as we are storing content as a string we are obviously losing all values that cannot be represented in default encoding. Tomas, what do you think? -- Oleg -------------- next part -------------- An HTML attachment was scrubbed... URL: From Tomas.Matousek at microsoft.com Mon Jul 14 12:17:03 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Mon, 14 Jul 2008 09:17:03 -0700 Subject: [Ironruby-core] Using Local Variable in Silverlight In-Reply-To: <7d3ae7368a80a972216fb719e3400131@ruby-forum.com> References: <7d3ae7368a80a972216fb719e3400131@ruby-forum.com> Message-ID: What's the exception type and message? Could you please file this as a bug? It seems it should just work. Tomas -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Aaron Clauson Sent: Monday, July 14, 2008 4:50 AM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] Using Local Variable in Silverlight Hi All, I'm getting an exception when I try and use the code snippet below in Silverlight. The snippet works ok with a normal .Net app. What I am trying to do is use a Ruby script running in my Silverlight app to call a method that updates the text on a text block. public delegate void LogMessageDelegate(string message); public partial class Page : UserControl { ... private void RunScript(object sender, System.Windows.RoutedEventArgs e) { ScriptHelper scriptHelper = new ScriptHelper(WriteDebugMessage); ScriptScope scriptScope = IronRuby.CreateRuntime().CreateScope("IronRuby"); scriptScope.SetVariable("sys", scriptHelper); scriptScope.Execute("sys.Log(\"hello world\")"); } private void WriteDebugMessage(string message) { m_scriptDebugText.Text = message; // m_scriptDebugText is a text block. } } public class ScriptHelper { private event LogMessageDelegate m_logMessage; public ScriptHelper(LogMessageDelegate logMessage) { m_logMessage = logMessage; } public void Log(string message) { m_logMessage(message); } } The exception I get back is: "Ruby.Runtime.RubyTopLevelScope.TopMethodMissing(Microsoft.Scripting.Runtime.CodeContext, System.Object, Ruby.Runtime.BlockParam, Microsoft.Scripting.SymbolId, System.Object[])" Is there a better approach to interacting between Silverlight UI elements and the Ruby script? I'm not using any HTML as the whole interface is built of SilverLight controls. Regards, Aaron -- Posted via http://www.ruby-forum.com/. _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From olegtk at microsoft.com Mon Jul 14 12:17:42 2008 From: olegtk at microsoft.com (Oleg Tkachenko) Date: Mon, 14 Jul 2008 09:17:42 -0700 Subject: [Ironruby-core] MutableString encoding issue In-Reply-To: References: Message-ID: This problem is probably StringContent.ToByteArray(). It uses Encoding.GetBytes(string) which obeys .NET encoding semantics and by default replaces any nonconvertible characters to '?'. And then MutableStringOps.Dump() is using it to create string representation. We could make StringContent.ToByteArray() not replacing nonconvertible characters by using EncodingFallback. BinaryContent.ToString()/ToStringBuilder() also has the same issue. -- Oleg From: Curt Hagenlocher Sent: Monday, July 14, 2008 6:27 AM To: Michael Letterle; ironruby-core at rubyforge.org Cc: IronRuby Team Subject: RE: [Ironruby-core] MutableString encoding issue MutableString can have one of three internal representations, depending on how it was last used. One of these is a byte array. This particular problem may be in the scanner or parser and not in the actual string class, as we don't otherwise have a problem storing the character: >>> $s = "\204" => "?" >>> $s[0] => 63 >>> $s[0] = 132 => 132 >>> $s => "\204" >>> From: Michael Letterle [mailto:michael.letterle at gmail.com] Sent: Monday, July 14, 2008 6:21 AM To: ironruby-core at rubyforge.org Cc: IronRuby Team Subject: Re: [Ironruby-core] MutableString encoding issue This was a known issue a while back, it's the reason the Zlib library didn't work well with binary files. I'm fairly certain there was work being done on making String be backed by a byte array... and in fact I thought this was already done. On Mon, Jul 14, 2008 at 2:20 AM, Oleg Tkachenko > wrote: Stumbled on this when testing yaml. >>> "\204" => "?" While irb(main):005:0>"\204" => "\204" I believe Ruby string can hold arbitrary byte values, but as we are storing content as a string we are obviously losing all values that cannot be represented in default encoding. Tomas, what do you think? -- Oleg _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core -- Michael Letterle [Polymath Prokrammer] http://blog.prokrams.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From olegtk at microsoft.com Mon Jul 14 12:20:51 2008 From: olegtk at microsoft.com (Oleg Tkachenko) Date: Mon, 14 Jul 2008 09:20:51 -0700 Subject: [Ironruby-core] MutableString encoding issue In-Reply-To: References: Message-ID: Sure. Unfortunately this one blocks our yaml impl passing MRI's test_yaml.rb. -- Oleg From: Tomas Matousek Sent: Monday, July 14, 2008 9:14 AM To: Oleg Tkachenko; IronRuby Team Cc: ironruby-core at rubyforge.org Subject: RE: MutableString encoding issue I think it's a bug :). Could you file it? If it's something that's blocking you I can look at it asap. Tomas From: Oleg Tkachenko Sent: Sunday, July 13, 2008 11:20 PM To: IronRuby Team Cc: ironruby-core at rubyforge.org Subject: MutableString encoding issue Stumbled on this when testing yaml. >>> "\204" => "?" While irb(main):005:0>"\204" => "\204" I believe Ruby string can hold arbitrary byte values, but as we are storing content as a string we are obviously losing all values that cannot be represented in default encoding. Tomas, what do you think? -- Oleg -------------- next part -------------- An HTML attachment was scrubbed... URL: From Tomas.Matousek at microsoft.com Mon Jul 14 21:28:58 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Mon, 14 Jul 2008 18:28:58 -0700 Subject: [Ironruby-core] Code Review: rubyopt Message-ID: tfpt review "/shelveset:rubyopt;REDMOND\tomat" Misc Ruby optimizations. - Optimizes method lookup: 1) Adds cache for unsuccessful CLR member lookup. CLR types are read-only and so the knowledge of a member lookup failure could be stored for the next lookup. This saves about 10% of allocated memory since unsuccessful lookups in reflection allocate some MethodInfo arrays. 2) Stores successfully resolved CLR members in the module method table. We don't need to create a RubyMethodGroupInfo each time we find CLR member and also if we find it once it is available the next time as well (unless it is removed/modified). 3) When a new method is added into a module/class and it doesn't hide a method of the same name that was already used in dynamic sites, the sites don't need to be invalidated. - Adds some perf counters. - Fixes message in method_missing. - Refactors RubyBinder a bit, especially super-call action. Tomas -------------- next part -------------- A non-text attachment was scrubbed... Name: rubyopt.diff Type: application/octet-stream Size: 50586 bytes Desc: rubyopt.diff URL: From jdeville at microsoft.com Tue Jul 15 01:22:11 2008 From: jdeville at microsoft.com (Jim Deville) Date: Mon, 14 Jul 2008 22:22:11 -0700 Subject: [Ironruby-core] Code Review: Spec changes Message-ID: tfpt review "/shelveset:rubyspec2;jdeville" * Changed all invocations of mspec to use the mspec command with Ruby instead of the mspec-ci or mspec-run command with IronRuby. This will allow us to take advantage of future options added to the mspec runner, and possibly add parallel process support if it gets added to mspec. * Changed the custom runners for baseline and tag to mspec ci and mspec tag. No reason to invent the wheel, and this is how SNAP will be running. This has the side effect of changing the tags to fails from fail * Added the -G critical option to all runners. Critical tags will be used to mark IronRuby bugs that block the runner. This is based on a discussion on #rubyspec about how other implementations are tracking bugs. The tag file is how things should be excluded unless they are bugs with the reference implementation (CRuby). Thus instead of the ironruby_bug guard, I have moved to a critical tag, which can be regenerated from this file. * Added a method to generate the tags from an existing critical_tags.txt file. Currently this file is the output of Powershell's Get-Child-Items -r -i *.txt | select-string critical, with some modification to remove new lines. Suggestions for options for automatically generating this file are welcome. * Added a rake task (regen_tags) to generate the critical tags. Github commits: IronRuby-tags: 90e33e19 Added a critical tags file to regenerate tags from. c5f6359 Rebaselined the specs. Mspec: 18db1cf Added critical tag to mspec ci Reordered tmp helper to respect ENV variables before attempting to guess. Rubyspec: 1e57876 Removed ironruby_bug guards due to the change to critical tags. -------------- next part -------------- An HTML attachment was scrubbed... URL: From lists at ruby-forum.com Tue Jul 15 04:48:37 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Tue, 15 Jul 2008 10:48:37 +0200 Subject: [Ironruby-core] Using Local Variable in Silverlight In-Reply-To: References: <7d3ae7368a80a972216fb719e3400131@ruby-forum.com> Message-ID: <9ce13c74637f918e4345427cc212db74@ruby-forum.com> Tomas Matousek wrote: > What's the exception type and message? The exception type is MethodAccessException and the message is: "Ruby.Runtime.RubyTopLevelScope.TopMethodMissing(Microsoft.Scripting.Runtime.CodeContext, System.Object, Ruby.Runtime.BlockParam, Microsoft.Scripting.SymbolId, System.Object[])" > Could you please file this as a bug? I submitted a bug report on the RubyForge site but is hasn't shown up. Hopefully that's just to do with it ebing moderated. The original code snippet works if changed to: ScriptScope scriptScope = IronRuby.CreateRuntime().CreateScope("IronPython"); or ScriptScope scriptScope = IronRuby.CreateRuntime().CreateScope("jscript"); Regards, Aaron -- Posted via http://www.ruby-forum.com/. From Tomas.Matousek at microsoft.com Tue Jul 15 14:23:42 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Tue, 15 Jul 2008 11:23:42 -0700 Subject: [Ironruby-core] Code Review: TokenizerFixes Message-ID: tfpt review "/shelveset:TokenizerFixes;REDMOND\tomat" Refactors string terminators in Ruby tokenizer a bit. Removes unnecesary calls to newtok() method that allocated a new StringBuilder per each token. Fixes handling of \e in strings. Removes uses of obsolete method TryStatementBuilder::SkipIf. Removes TokenizerBuffer's dependency on SourceUnitReader. Moves the responsibility for line mapping to the tokenizer. Tomas -------------- next part -------------- A non-text attachment was scrubbed... Name: TokenizerFixes.diff Type: application/octet-stream Size: 33701 bytes Desc: TokenizerFixes.diff URL: From jdeville at microsoft.com Tue Jul 15 16:44:35 2008 From: jdeville at microsoft.com (Jim Deville) Date: Tue, 15 Jul 2008 13:44:35 -0700 Subject: [Ironruby-core] Code Review: unni1 Message-ID: tfpt review "/shelveset:unni1;REDMOND\jdeville" Comment : Patch from Unnikrishnan Nair to properly handle File.zero? for 'NUL' -------------- next part -------------- A non-text attachment was scrubbed... Name: unni1.diff Type: application/octet-stream Size: 1335 bytes Desc: unni1.diff URL: From jdeville at microsoft.com Tue Jul 15 17:31:13 2008 From: jdeville at microsoft.com (Jim Deville) Date: Tue, 15 Jul 2008 14:31:13 -0700 Subject: [Ironruby-core] Code Review: unni1 In-Reply-To: References: Message-ID: Reviewed F2F with John JD > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of Jim Deville > Sent: Tuesday, July 15, 2008 1:45 PM > To: IronRuby External Code Reviewers > Cc: ironruby-core at rubyforge.org > Subject: [Ironruby-core] Code Review: unni1 > > tfpt review "/shelveset:unni1;REDMOND\jdeville" > Comment : > Patch from Unnikrishnan Nair to properly handle File.zero? for 'NUL' > From jdeville at microsoft.com Tue Jul 15 17:31:28 2008 From: jdeville at microsoft.com (Jim Deville) Date: Tue, 15 Jul 2008 14:31:28 -0700 Subject: [Ironruby-core] Code Review: Spec changes In-Reply-To: References: Message-ID: Reviewed F2F with John. JD From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Monday, July 14, 2008 10:22 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers; Srivatsn Narayanan Subject: [Ironruby-core] Code Review: Spec changes tfpt review "/shelveset:rubyspec2;jdeville" * Changed all invocations of mspec to use the mspec command with Ruby instead of the mspec-ci or mspec-run command with IronRuby. This will allow us to take advantage of future options added to the mspec runner, and possibly add parallel process support if it gets added to mspec. * Changed the custom runners for baseline and tag to mspec ci and mspec tag. No reason to invent the wheel, and this is how SNAP will be running. This has the side effect of changing the tags to fails from fail * Added the -G critical option to all runners. Critical tags will be used to mark IronRuby bugs that block the runner. This is based on a discussion on #rubyspec about how other implementations are tracking bugs. The tag file is how things should be excluded unless they are bugs with the reference implementation (CRuby). Thus instead of the ironruby_bug guard, I have moved to a critical tag, which can be regenerated from this file. * Added a method to generate the tags from an existing critical_tags.txt file. Currently this file is the output of Powershell's Get-Child-Items -r -i *.txt | select-string critical, with some modification to remove new lines. Suggestions for options for automatically generating this file are welcome. * Added a rake task (regen_tags) to generate the critical tags. Github commits: IronRuby-tags: 90e33e19 Added a critical tags file to regenerate tags from. c5f6359 Rebaselined the specs. Mspec: 18db1cf Added critical tag to mspec ci Reordered tmp helper to respect ENV variables before attempting to guess. Rubyspec: 1e57876 Removed ironruby_bug guards due to the change to critical tags. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Jimmy.Schementi at microsoft.com Tue Jul 15 17:48:37 2008 From: Jimmy.Schementi at microsoft.com (Jimmy Schementi) Date: Tue, 15 Jul 2008 14:48:37 -0700 Subject: [Ironruby-core] Code Review: Spec changes In-Reply-To: References: , Message-ID: <5283CA0A4168DF4FBBD71AE9ECA5A32845D41AFA78@NA-EXMSG-C116.redmond.corp.microsoft.com> Can you send links to github instead of just the commit sha1? ________________________________ From: ironruby-core-bounces at rubyforge.org [ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville [jdeville at microsoft.com] Sent: Tuesday, July 15, 2008 2:31 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers; Srivatsn Narayanan Subject: Re: [Ironruby-core] Code Review: Spec changes Reviewed F2F with John. JD From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Monday, July 14, 2008 10:22 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers; Srivatsn Narayanan Subject: [Ironruby-core] Code Review: Spec changes tfpt review ?/shelveset:rubyspec2;jdeville? ? Changed all invocations of mspec to use the mspec command with Ruby instead of the mspec-ci or mspec-run command with IronRuby. This will allow us to take advantage of future options added to the mspec runner, and possibly add parallel process support if it gets added to mspec. ? Changed the custom runners for baseline and tag to mspec ci and mspec tag. No reason to invent the wheel, and this is how SNAP will be running. This has the side effect of changing the tags to fails from fail ? Added the -G critical option to all runners. Critical tags will be used to mark IronRuby bugs that block the runner. This is based on a discussion on #rubyspec about how other implementations are tracking bugs. The tag file is how things should be excluded unless they are bugs with the reference implementation (CRuby). Thus instead of the ironruby_bug guard, I have moved to a critical tag, which can be regenerated from this file. ? Added a method to generate the tags from an existing critical_tags.txt file. Currently this file is the output of Powershell?s Get-Child-Items -r -i *.txt | select-string critical, with some modification to remove new lines. Suggestions for options for automatically generating this file are welcome. ? Added a rake task (regen_tags) to generate the critical tags. Github commits: IronRuby-tags: 90e33e19 Added a critical tags file to regenerate tags from. c5f6359 Rebaselined the specs. Mspec: 18db1cf Added critical tag to mspec ci Reordered tmp helper to respect ENV variables before attempting to guess. Rubyspec: 1e57876 Removed ironruby_bug guards due to the change to critical tags. -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdeville at microsoft.com Tue Jul 15 17:51:23 2008 From: jdeville at microsoft.com (Jim Deville) Date: Tue, 15 Jul 2008 14:51:23 -0700 Subject: [Ironruby-core] Code Review: unni1 In-Reply-To: References: Message-ID: While testing this manually I discovered that CRuby treats nul and NUL as zero length files, while this patch only works for NUL. Unni, can you fix that and resubmit? Thanks, JD > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of Jim Deville > Sent: Tuesday, July 15, 2008 2:31 PM > To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers > Subject: Re: [Ironruby-core] Code Review: unni1 > > Reviewed F2F with John > > > JD > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > bounces at rubyforge.org] On Behalf Of Jim Deville > > Sent: Tuesday, July 15, 2008 1:45 PM > > To: IronRuby External Code Reviewers > > Cc: ironruby-core at rubyforge.org > > Subject: [Ironruby-core] Code Review: unni1 > > > > tfpt review "/shelveset:unni1;REDMOND\jdeville" > > Comment : > > Patch from Unnikrishnan Nair to properly handle File.zero? for > 'NUL' > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core From jdeville at microsoft.com Tue Jul 15 17:51:48 2008 From: jdeville at microsoft.com (Jim Deville) Date: Tue, 15 Jul 2008 14:51:48 -0700 Subject: [Ironruby-core] Code Review: Spec changes In-Reply-To: <5283CA0A4168DF4FBBD71AE9ECA5A32845D41AFA78@NA-EXMSG-C116.redmond.corp.microsoft.com> References: , <5283CA0A4168DF4FBBD71AE9ECA5A32845D41AFA78@NA-EXMSG-C116.redmond.corp.microsoft.com> Message-ID: I remembered that shortly after I sent this mail last night. One of these times I will get it. JD From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jimmy Schementi Sent: Tuesday, July 15, 2008 2:49 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers; Srivatsn Narayanan Subject: Re: [Ironruby-core] Code Review: Spec changes Can you send links to github instead of just the commit sha1? ________________________________ From: ironruby-core-bounces at rubyforge.org [ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville [jdeville at microsoft.com] Sent: Tuesday, July 15, 2008 2:31 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers; Srivatsn Narayanan Subject: Re: [Ironruby-core] Code Review: Spec changes Reviewed F2F with John. JD From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Monday, July 14, 2008 10:22 PM To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers; Srivatsn Narayanan Subject: [Ironruby-core] Code Review: Spec changes tfpt review "/shelveset:rubyspec2;jdeville" * Changed all invocations of mspec to use the mspec command with Ruby instead of the mspec-ci or mspec-run command with IronRuby. This will allow us to take advantage of future options added to the mspec runner, and possibly add parallel process support if it gets added to mspec. * Changed the custom runners for baseline and tag to mspec ci and mspec tag. No reason to invent the wheel, and this is how SNAP will be running. This has the side effect of changing the tags to fails from fail * Added the -G critical option to all runners. Critical tags will be used to mark IronRuby bugs that block the runner. This is based on a discussion on #rubyspec about how other implementations are tracking bugs. The tag file is how things should be excluded unless they are bugs with the reference implementation (CRuby). Thus instead of the ironruby_bug guard, I have moved to a critical tag, which can be regenerated from this file. * Added a method to generate the tags from an existing critical_tags.txt file. Currently this file is the output of Powershell's Get-Child-Items -r -i *.txt | select-string critical, with some modification to remove new lines. Suggestions for options for automatically generating this file are welcome. * Added a rake task (regen_tags) to generate the critical tags. Github commits: IronRuby-tags: 90e33e19 Added a critical tags file to regenerate tags from. c5f6359 Rebaselined the specs. Mspec: 18db1cf Added critical tag to mspec ci Reordered tmp helper to respect ENV variables before attempting to guess. Rubyspec: 1e57876 Removed ironruby_bug guards due to the change to critical tags. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ksunair at yahoo.com Wed Jul 16 12:34:40 2008 From: ksunair at yahoo.com (Unnikrishnan Nair) Date: Wed, 16 Jul 2008 09:34:40 -0700 (PDT) Subject: [Ironruby-core] Code Review: unni1 In-Reply-To: Message-ID: <962846.3315.qm@web50601.mail.re2.yahoo.com> yes, I will do that. Thanks. Unni --- On Tue, 7/15/08, Jim Deville wrote: From: Jim Deville Subject: Re: [Ironruby-core] Code Review: unni1 To: "ironruby-core at rubyforge.org" , "IronRuby External Code Reviewers" Date: Tuesday, July 15, 2008, 3:51 PM While testing this manually I discovered that CRuby treats nul and NUL as zero length files, while this patch only works for NUL. Unni, can you fix that and resubmit? Thanks, JD > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of Jim Deville > Sent: Tuesday, July 15, 2008 2:31 PM > To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers > Subject: Re: [Ironruby-core] Code Review: unni1 > > Reviewed F2F with John > > > JD > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > bounces at rubyforge.org] On Behalf Of Jim Deville > > Sent: Tuesday, July 15, 2008 1:45 PM > > To: IronRuby External Code Reviewers > > Cc: ironruby-core at rubyforge.org > > Subject: [Ironruby-core] Code Review: unni1 > > > > tfpt review "/shelveset:unni1;REDMOND\jdeville" > > Comment : > > Patch from Unnikrishnan Nair to properly handle File.zero? for > 'NUL' > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdeville at microsoft.com Wed Jul 16 20:57:44 2008 From: jdeville at microsoft.com (Jim Deville) Date: Wed, 16 Jul 2008 17:57:44 -0700 Subject: [Ironruby-core] Code Review: rubyspec3 Message-ID: tfpt review "/shelveset:rubyspec3;REDMOND\jdeville" Comment : removed duplication in *_all_core methods in context. Made changes to allow command line options to be passed to the target runner. Added rake tasks to setup running under CRuby, and PEverify. Made standard run go under -X:Interpret. -------------- next part -------------- A non-text attachment was scrubbed... Name: rubyspec3.diff Type: application/octet-stream Size: 17788 bytes Desc: rubyspec3.diff URL: From ksunair at yahoo.com Wed Jul 16 22:51:49 2008 From: ksunair at yahoo.com (Unnikrishnan Nair) Date: Wed, 16 Jul 2008 19:51:49 -0700 (PDT) Subject: [Ironruby-core] Code Review: unni1 In-Reply-To: <962846.3315.qm@web50601.mail.re2.yahoo.com> Message-ID: <182694.73032.qm@web50605.mail.re2.yahoo.com> Jim, ? As per your comments, I have modified the code and ran against Matz Ruby and verified the results. Yes you are correct 'nul' is same as 'NUL' as well. I have updated the patch as well. ? Just curious, how come there is no representation of IronRuby in Lonestar Ruby conference? ? Thanks. Unni --- On Wed, 7/16/08, Unnikrishnan Nair wrote: From: Unnikrishnan Nair Subject: Re: [Ironruby-core] Code Review: unni1 To: ironruby-core at rubyforge.org Date: Wednesday, July 16, 2008, 10:34 AM yes, I will do that. Thanks. Unni --- On Tue, 7/15/08, Jim Deville wrote: From: Jim Deville Subject: Re: [Ironruby-core] Code Review: unni1 To: "ironruby-core at rubyforge.org" , "IronRuby External Code Reviewers" Date: Tuesday, July 15, 2008, 3:51 PM While testing this manually I discovered that CRuby treats nul and NUL as zero length files, while this patch only works for NUL. Unni, can you fix that and resubmit? Thanks, JD > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of Jim Deville > Sent: Tuesday, July 15, 2008 2:31 PM > To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers > Subject: Re: [Ironruby-core] Code Review: unni1 > > Reviewed F2F with John > > > JD > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > bounces at rubyforge.org] On Behalf Of Jim Deville > > Sent: Tuesday, July 15, 2008 1:45 PM > > To: IronRuby External Code Reviewers > > Cc: ironruby-core at rubyforge.org > > Subject: [Ironruby-core] Code Review: unni1 > > > > tfpt review "/shelveset:unni1;REDMOND\jdeville" > > Comment : > > Patch from Unnikrishnan Nair to properly handle File.zero? for > 'NUL' > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core_______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdeville at microsoft.com Wed Jul 16 23:10:44 2008 From: jdeville at microsoft.com (Jim Deville) Date: Wed, 16 Jul 2008 20:10:44 -0700 Subject: [Ironruby-core] Code Review: unni1 In-Reply-To: <182694.73032.qm@web50605.mail.re2.yahoo.com> References: <962846.3315.qm@web50601.mail.re2.yahoo.com> <182694.73032.qm@web50605.mail.re2.yahoo.com> Message-ID: Cool, I'll get that integrated into the repo. I don't know why we didn't send someone to Lonestar. Probably just a case of limited resources. JD From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Unnikrishnan Nair Sent: Wednesday, July 16, 2008 7:52 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: unni1 Jim, As per your comments, I have modified the code and ran against Matz Ruby and verified the results. Yes you are correct 'nul' is same as 'NUL' as well. I have updated the patch as well. Just curious, how come there is no representation of IronRuby in Lonestar Ruby conference? Thanks. Unni --- On Wed, 7/16/08, Unnikrishnan Nair wrote: From: Unnikrishnan Nair Subject: Re: [Ironruby-core] Code Review: unni1 To: ironruby-core at rubyforge.org Date: Wednesday, July 16, 2008, 10:34 AM yes, I will do that. Thanks. Unni --- On Tue, 7/15/08, Jim Deville wrote: From: Jim Deville Subject: Re: [Ironruby-core] Code Review: unni1 To: "ironruby-core at rubyforge.org" , "IronRuby External Code Reviewers" Date: Tuesday, July 15, 2008, 3:51 PM While testing this manually I discovered that CRuby treats nul and NUL as zero length files, while this patch only works for NUL. Unni, can you fix that and resubmit? Thanks, JD > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of Jim Deville > Sent: Tuesday, July 15, 2008 2:31 PM > To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers > Subject: Re: [Ironruby-core] Code Review: unni1 > > Reviewed F2F with John > > > JD > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > bounces at rubyforge.org] On Behalf Of Jim Deville > > Sent: Tuesday, July 15, 2008 1:45 PM > > To: IronRuby External Code Reviewers > > Cc: ironruby-core at rubyforge.org > > Subject: [Ironruby-core] Code Review: unni1 > > > > tfpt review "/shelveset:unni1;REDMOND\jdeville" > > Comment : > > Patch from Unnikrishnan Nair to properly handle File.zero? for > 'NUL' > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core -------------- next part -------------- An HTML attachment was scrubbed... URL: From jflam at microsoft.com Thu Jul 17 11:05:56 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Thu, 17 Jul 2008 08:05:56 -0700 Subject: [Ironruby-core] Code Review: unni1 In-Reply-To: <182694.73032.qm@web50605.mail.re2.yahoo.com> References: <962846.3315.qm@web50601.mail.re2.yahoo.com> <182694.73032.qm@web50605.mail.re2.yahoo.com> Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD4206158@NA-EXMSG-C115.redmond.corp.microsoft.com> As much as we would love to attend more Ruby conferences (and there are a LOT of Ruby conferences!), we really don't have the bandwidth. So we're being very careful about where we go. We'll be at RubyConf this year. Thanks, -John From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Unnikrishnan Nair Sent: Wednesday, July 16, 2008 10:52 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: unni1 Jim, As per your comments, I have modified the code and ran against Matz Ruby and verified the results. Yes you are correct 'nul' is same as 'NUL' as well. I have updated the patch as well. Just curious, how come there is no representation of IronRuby in Lonestar Ruby conference? Thanks. Unni --- On Wed, 7/16/08, Unnikrishnan Nair wrote: From: Unnikrishnan Nair Subject: Re: [Ironruby-core] Code Review: unni1 To: ironruby-core at rubyforge.org Date: Wednesday, July 16, 2008, 10:34 AM yes, I will do that. Thanks. Unni --- On Tue, 7/15/08, Jim Deville wrote: From: Jim Deville Subject: Re: [Ironruby-core] Code Review: unni1 To: "ironruby-core at rubyforge.org" , "IronRuby External Code Reviewers" Date: Tuesday, July 15, 2008, 3:51 PM While testing this manually I discovered that CRuby treats nul and NUL as zero length files, while this patch only works for NUL. Unni, can you fix that and resubmit? Thanks, JD > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of Jim Deville > Sent: Tuesday, July 15, 2008 2:31 PM > To: ironruby-core at rubyforge.org; IronRuby External Code Reviewers > Subject: Re: [Ironruby-core] Code Review: unni1 > > Reviewed F2F with John > > > JD > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > > bounces at rubyforge.org] On Behalf Of Jim Deville > > Sent: Tuesday, July 15, 2008 1:45 PM > > To: IronRuby External Code Reviewers > > Cc: ironruby-core at rubyforge.org > > Subject: [Ironruby-core] Code Review: unni1 > > > > tfpt review "/shelveset:unni1;REDMOND\jdeville" > > Comment : > > Patch from Unnikrishnan Nair to properly handle File.zero? for > 'NUL' > > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core -------------- next part -------------- An HTML attachment was scrubbed... URL: From ksunair at yahoo.com Thu Jul 17 11:16:47 2008 From: ksunair at yahoo.com (Unnikrishnan Nair) Date: Thu, 17 Jul 2008 08:16:47 -0700 (PDT) Subject: [Ironruby-core] Code Review: unni1 In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD4206158@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <38596.26432.qm@web50602.mail.re2.yahoo.com> I was really looking forward to meet you guys here at Texas, well next time. Matz is giving the key note. Thanks. Unni --- On Thu, 7/17/08, John Lam (IRONRUBY) wrote: > From: John Lam (IRONRUBY) > Subject: RE: [Ironruby-core] Code Review: unni1 > To: "ksunair at yahoo.com" , "ironruby-core at rubyforge.org" > Date: Thursday, July 17, 2008, 9:05 AM > As much as we would love to attend more Ruby conferences > (and there are a LOT of Ruby conferences!), we really > don't have the bandwidth. So we're being very > careful about where we go. > > We'll be at RubyConf this year. > > Thanks, > -John > > > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of > Unnikrishnan Nair > Sent: Wednesday, July 16, 2008 10:52 PM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] Code Review: unni1 > > > Jim, > > > > As per your comments, I have modified the code and ran > against Matz Ruby and verified the results. Yes you are > correct 'nul' is same as 'NUL' as well. I > have updated the patch as well. > > > > Just curious, how come there is no representation of > IronRuby in Lonestar Ruby conference? > > > > Thanks. > > Unni > > --- On Wed, 7/16/08, Unnikrishnan Nair > wrote: > From: Unnikrishnan Nair > Subject: Re: [Ironruby-core] Code Review: unni1 > To: ironruby-core at rubyforge.org > Date: Wednesday, July 16, 2008, 10:34 AM > > yes, I will do that. > > Thanks. > > Unni > > --- On Tue, 7/15/08, Jim Deville > wrote: > From: Jim Deville > Subject: Re: [Ironruby-core] Code Review: unni1 > To: "ironruby-core at rubyforge.org" > , "IronRuby > External Code Reviewers" > Date: Tuesday, July 15, 2008, 3:51 PM > > While testing this manually I discovered that CRuby treats > nul and NUL as zero > > length files, while this patch only works for NUL. Unni, > can you fix that and > > resubmit? > > > > Thanks, > > > > > > JD > > > > > > > -----Original Message----- > > > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core- > > > bounces at rubyforge.org] On Behalf Of Jim Deville > > > Sent: Tuesday, July 15, 2008 2:31 PM > > > To: ironruby-core at rubyforge.org; IronRuby External > Code Reviewers > > > Subject: Re: [Ironruby-core] Code Review: unni1 > > > > > > Reviewed F2F with John > > > > > > > > > JD > > > > > > > -----Original Message----- > > > > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core- > > > > bounces at rubyforge.org] On Behalf Of Jim Deville > > > > Sent: Tuesday, July 15, 2008 1:45 PM > > > > To: IronRuby External Code Reviewers > > > > Cc: ironruby-core at rubyforge.org > > > > Subject: [Ironruby-core] Code Review: unni1 > > > > > > > > tfpt review > "/shelveset:unni1;REDMOND\jdeville" > > > > Comment : > > > > Patch from Unnikrishnan Nair to properly handle > File.zero? for > > > 'NUL' > > > > > > > > > > _______________________________________________ > > > Ironruby-core mailing list > > > Ironruby-core at rubyforge.org > > > http://rubyforge.org/mailman/listinfo/ironruby-core > > > > _______________________________________________ > > Ironruby-core mailing list > > Ironruby-core at rubyforge.org > > http://rubyforge.org/mailman/listinfo/ironruby-core > > > _______________________________________________ > > Ironruby-core mailing list > > Ironruby-core at rubyforge.org > > http://rubyforge.org/mailman/listinfo/ironruby-core From Tomas.Matousek at microsoft.com Thu Jul 17 21:26:50 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Thu, 17 Jul 2008 18:26:50 -0700 Subject: [Ironruby-core] Using Local Variable in Silverlight In-Reply-To: <9ce13c74637f918e4345427cc212db74@ruby-forum.com> References: <7d3ae7368a80a972216fb719e3400131@ruby-forum.com> <9ce13c74637f918e4345427cc212db74@ruby-forum.com> Message-ID: Oh, I know what this is. Could you verify that the method is internal in the bits you're using? The method should be public and it is in the current code. Tomas -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Aaron Clauson Sent: Tuesday, July 15, 2008 1:49 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Using Local Variable in Silverlight Tomas Matousek wrote: > What's the exception type and message? The exception type is MethodAccessException and the message is: "Ruby.Runtime.RubyTopLevelScope.TopMethodMissing(Microsoft.Scripting.Runtime.CodeContext, System.Object, Ruby.Runtime.BlockParam, Microsoft.Scripting.SymbolId, System.Object[])" > Could you please file this as a bug? I submitted a bug report on the RubyForge site but is hasn't shown up. Hopefully that's just to do with it ebing moderated. The original code snippet works if changed to: ScriptScope scriptScope = IronRuby.CreateRuntime().CreateScope("IronPython"); or ScriptScope scriptScope = IronRuby.CreateRuntime().CreateScope("jscript"); Regards, Aaron -- Posted via http://www.ruby-forum.com/. _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From lists at ruby-forum.com Fri Jul 18 01:32:05 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Fri, 18 Jul 2008 07:32:05 +0200 Subject: [Ironruby-core] Using Local Variable in Silverlight In-Reply-To: References: <7d3ae7368a80a972216fb719e3400131@ruby-forum.com> <9ce13c74637f918e4345427cc212db74@ruby-forum.com> Message-ID: <06014d114bf89a8f0e2bcf7d2a79fafe@ruby-forum.com> Tomas Matousek wrote: > Oh, I know what this is. Could you verify that the method is internal in > the bits you're using? > The method should be public and it is in the current code. > > Tomas Hi Tomas, Do you mean the CreateScope method, i.e. IronRuby.CreateRuntime().CreateScope("IronRuby")? I was using the IronRuby version from the Silverlight DLR SDK on http://www.codeplex.com/sdlsdk release id 14254. I don't think the method CreateScope could have been internal as I wouldn't have been able to call it? I did attempt to build the IronRuby code from svn with the SILVERLIGHT pre-compilation option sent but got a fair few errors so tried the SDK version instead. I'll give building from SVN another attempt. Regards, Aaron -- Posted via http://www.ruby-forum.com/. From lists at ruby-forum.com Fri Jul 18 01:40:19 2008 From: lists at ruby-forum.com (Web Reservoir) Date: Fri, 18 Jul 2008 07:40:19 +0200 Subject: [Ironruby-core] IronPython 2.0 Road Map Plans announced. Message-ID: <1de3c3111ae25279cb96442e89ad9f9e@ruby-forum.com> Hi, Have a look here, for detailed information on IronPython 2.0 Plans http://devhawk.net/2008/07/17/IronPython+Post+20+Roadmap.aspx It gives,details about its current status and Road Map Plans for version 2.0 Hope we can get such Road Map Plans for IronRuby 1.0 too. Thanks -- Posted via http://www.ruby-forum.com/. From Tomas.Matousek at microsoft.com Fri Jul 18 02:17:49 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Thu, 17 Jul 2008 23:17:49 -0700 Subject: [Ironruby-core] Using Local Variable in Silverlight In-Reply-To: <06014d114bf89a8f0e2bcf7d2a79fafe@ruby-forum.com> References: <7d3ae7368a80a972216fb719e3400131@ruby-forum.com> <9ce13c74637f918e4345427cc212db74@ruby-forum.com> <06014d114bf89a8f0e2bcf7d2a79fafe@ruby-forum.com> Message-ID: No, I mean: Ruby.Runtime.RubyTopLevelScope.TopMethodMissing. I'm not sure you can build successfully against Silverlight. There might have been some changes in Silverlight that required us to change DLR/IronRuby code since the last public release. Tomas -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Aaron Clauson Sent: Thursday, July 17, 2008 10:32 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Using Local Variable in Silverlight Tomas Matousek wrote: > Oh, I know what this is. Could you verify that the method is internal in > the bits you're using? > The method should be public and it is in the current code. > > Tomas Hi Tomas, Do you mean the CreateScope method, i.e. IronRuby.CreateRuntime().CreateScope("IronRuby")? I was using the IronRuby version from the Silverlight DLR SDK on http://www.codeplex.com/sdlsdk release id 14254. I don't think the method CreateScope could have been internal as I wouldn't have been able to call it? I did attempt to build the IronRuby code from svn with the SILVERLIGHT pre-compilation option sent but got a fair few errors so tried the SDK version instead. I'll give building from SVN another attempt. Regards, Aaron -- Posted via http://www.ruby-forum.com/. _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From lists at ruby-forum.com Fri Jul 18 04:42:43 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Fri, 18 Jul 2008 10:42:43 +0200 Subject: [Ironruby-core] Using Local Variable in Silverlight In-Reply-To: References: <7d3ae7368a80a972216fb719e3400131@ruby-forum.com> <9ce13c74637f918e4345427cc212db74@ruby-forum.com> <06014d114bf89a8f0e2bcf7d2a79fafe@ruby-forum.com> Message-ID: Tomas Matousek wrote: > No, I mean: Ruby.Runtime.RubyTopLevelScope.TopMethodMissing. > > I'm not sure you can build successfully against Silverlight. There might > have been some changes in Silverlight that required us to change > DLR/IronRuby code since the last public release. > > Tomas Hi Tomas, I've checked and you are correct Ruby.Runtime.RubyTopLevelScope.TopMethodMissing is static internal . Any idea when the next IronRuby DLR release is likely to be? I can get by with using JScript within Silverlight for the time being. Thanks, Aaron -- Posted via http://www.ruby-forum.com/. From Johan.Danforth at irm.se Fri Jul 18 05:30:53 2008 From: Johan.Danforth at irm.se (Johan Danforth) Date: Fri, 18 Jul 2008 11:30:53 +0200 Subject: [Ironruby-core] IronRyby VS integration? In-Reply-To: References: <7d3ae7368a80a972216fb719e3400131@ruby-forum.com> <9ce13c74637f918e4345427cc212db74@ruby-forum.com> <06014d114bf89a8f0e2bcf7d2a79fafe@ruby-forum.com> Message-ID: <12D498AE0ADB20479A22F41E113B67D302BE33CE25@sun.irm.se> DevHawk wrote in his blog (http://devhawk.net/2008/07/17/IronPython+Post+20+Roadmap.aspx): "Besides IPy 2.1 and 3.0, the other major thing we're working on is Visual Studio integration for IronPython ... ... (it would be nice if IronRuby could leverage our efforts down the road). Like everything else we do in this group, we'll be publishing the VS Integration source code up on CodePlex as early and often as we can." Are there any plans for a similar VS integration for IronRuby? /Johan From jflam at microsoft.com Fri Jul 18 09:23:35 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Fri, 18 Jul 2008 06:23:35 -0700 Subject: [Ironruby-core] IronRyby VS integration? In-Reply-To: <12D498AE0ADB20479A22F41E113B67D302BE33CE25@sun.irm.se> References: <7d3ae7368a80a972216fb719e3400131@ruby-forum.com> <9ce13c74637f918e4345427cc212db74@ruby-forum.com> <06014d114bf89a8f0e2bcf7d2a79fafe@ruby-forum.com> <12D498AE0ADB20479A22F41E113B67D302BE33CE25@sun.irm.se> Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD42065EF@NA-EXMSG-C115.redmond.corp.microsoft.com> Yes. > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of Johan Danforth > Sent: Friday, July 18, 2008 5:31 AM > To: ironruby-core at rubyforge.org > Subject: [Ironruby-core] IronRyby VS integration? > > DevHawk wrote in his blog > (http://devhawk.net/2008/07/17/IronPython+Post+20+Roadmap.aspx): > > "Besides IPy 2.1 and 3.0, the other major thing we're working on is > Visual Studio integration for IronPython ... ... (it would be > nice if IronRuby could leverage our efforts down the road). Like > everything else we do in this group, we'll be publishing the VS > Integration source code up on CodePlex as early and often as we can." > > Are there any plans for a similar VS integration for IronRuby? > > /Johan > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core From olegtk at microsoft.com Fri Jul 18 12:19:20 2008 From: olegtk at microsoft.com (Oleg Tkachenko) Date: Fri, 18 Jul 2008 09:19:20 -0700 Subject: [Ironruby-core] IronRyby VS integration? In-Reply-To: <12D498AE0ADB20479A22F41E113B67D302BE33CE25@sun.irm.se> References: <7d3ae7368a80a972216fb719e3400131@ruby-forum.com> <9ce13c74637f918e4345427cc212db74@ruby-forum.com> <06014d114bf89a8f0e2bcf7d2a79fafe@ruby-forum.com> <12D498AE0ADB20479A22F41E113B67D302BE33CE25@sun.irm.se> Message-ID: Yes and as we are in a planning stage, we are extremely eager to hear what do you guys think about IronRuby VS integration. What features you need, how does it need to look and feel? Don't hesitate to express your thoughts on this subject. We want to make it useful and so need to know what do you want and expect. VS integration is going to be open source like IronRuby itself and I also hope we will be able to accept contributions into it too. -- Oleg -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Johan Danforth Sent: Friday, July 18, 2008 2:31 AM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] IronRyby VS integration? DevHawk wrote in his blog (http://devhawk.net/2008/07/17/IronPython+Post+20+Roadmap.aspx): "Besides IPy 2.1 and 3.0, the other major thing we're working on is Visual Studio integration for IronPython ... ... (it would be nice if IronRuby could leverage our efforts down the road). Like everything else we do in this group, we'll be publishing the VS Integration source code up on CodePlex as early and often as we can." Are there any plans for a similar VS integration for IronRuby? /Johan _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From Johan.Danforth at irm.se Sat Jul 19 07:11:39 2008 From: Johan.Danforth at irm.se (Johan Danforth) Date: Sat, 19 Jul 2008 13:11:39 +0200 Subject: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD40AFED2@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <7ca58dc067dc9a67c58ac102feed5325@ruby-forum.com> <372109E149E8084D8E6C7D9CFD82E0632FD40AFED2@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <12D498AE0ADB20479A22F41E113B67D302BE33CE2D@sun.irm.se> The question has been asked on the MVC forum, but Levi (I think it was) wrote: "What John is saying is that you can download the ZIP that allows IronRuby to run in ASP MVC and that the project will run on top of Preview 4 bits" I like to work with both ASP.NET MVC and IronRuby, so I would love to get my hands on that ZIP ;) /Johan -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: den 9 juli 2008 17:39 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions The ASP.NET MVC questions should be asked on the ASP.NET MVC forums. The current plan is to ship IronRuby early next year. Thanks, -John -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Rahil Kumar Sent: Wednesday, July 09, 2008 4:54 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions Web Reservoir wrote: > Hi, > > May i know, how many months are we still far from IronRuby 1.0...? It was july 2008 previously. It seems its delayed. I am also waiting for a confirmed date. > > I understand Asp.Net MVC preview 4 will support IronRuby. Asp.Net MVC 4 is scheduled to be in this month only, perhaps mid july. I am not aware about any do's / dont's with this preview. I am sailing in the same boat like you. -- Posted via http://www.ruby-forum.com/. _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From Johan.Danforth at irm.se Sat Jul 19 07:15:02 2008 From: Johan.Danforth at irm.se (Johan Danforth) Date: Sat, 19 Jul 2008 13:15:02 +0200 Subject: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions In-Reply-To: <12D498AE0ADB20479A22F41E113B67D302BE33CE2D@sun.irm.se> References: <7ca58dc067dc9a67c58ac102feed5325@ruby-forum.com> <372109E149E8084D8E6C7D9CFD82E0632FD40AFED2@NA-EXMSG-C115.redmond.corp.microsoft.com> <12D498AE0ADB20479A22F41E113B67D302BE33CE2D@sun.irm.se> Message-ID: <12D498AE0ADB20479A22F41E113B67D302BE33CE2F@sun.irm.se> And a minute later I found a zip file on your blog entry (http://www.iunknown.com/2008/06/ironruby-and-aspnet-mvc.html) John. I'll have a look at that at once. Sorry for being too trigger happy... :/ /Johan -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Johan Danforth Sent: den 19 juli 2008 13:12 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions The question has been asked on the MVC forum, but Levi (I think it was) wrote: "What John is saying is that you can download the ZIP that allows IronRuby to run in ASP MVC and that the project will run on top of Preview 4 bits" I like to work with both ASP.NET MVC and IronRuby, so I would love to get my hands on that ZIP ;) /Johan -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: den 9 juli 2008 17:39 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions The ASP.NET MVC questions should be asked on the ASP.NET MVC forums. The current plan is to ship IronRuby early next year. Thanks, -John -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Rahil Kumar Sent: Wednesday, July 09, 2008 4:54 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions Web Reservoir wrote: > Hi, > > May i know, how many months are we still far from IronRuby 1.0...? It was july 2008 previously. It seems its delayed. I am also waiting for a confirmed date. > > I understand Asp.Net MVC preview 4 will support IronRuby. Asp.Net MVC 4 is scheduled to be in this month only, perhaps mid july. I am not aware about any do's / dont's with this preview. I am sailing in the same boat like you. -- Posted via http://www.ruby-forum.com/. _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From bacondarwin at googlemail.com Sat Jul 19 15:02:55 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Sat, 19 Jul 2008 20:02:55 +0100 Subject: [Ironruby-core] Multiple overloads VS Inline casting Message-ID: <001901c8e9d2$0ebeac00$2c3c0400$@com> I have noticed in the libraries that are often methods that take multiple parameters and each parameter can take a number of different types. A simple example, BigDecimal#add method has two parameters: a.add(b,n), where a is BigDecimal, b needs to be compatible with BigDecimal and n needs to be compatible with Fixnum. In other words it is possible that b and n can be one of [NotNull]BigDecimal/*!*/, [NotNull]BigInteger/*!*/, int, double and object. This leads to an implementation question. Initially it seemed sensible to abstract out the conversion of these types - except perhaps the best case ([NotNull]BigDecimal/*!*/ and int) - into a helper method and then have just: public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, BigDecimal/*!*/ b, int n) ... and public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, object b, object n) ... In the second method you convert the parameters to the correct types and pass them back to the first method. This is basically what happens in the current implementation of many of the Socket library methods. This provides a single place of conversion, which is particularly helpful given the BigDecimal class has lots of methods that take this form that will all use this conversion. It makes the code readable and doesn't create masses of overloads for the class. After thinking about this for a bit I started to wonder about performance. This style of method does not allow performance tricks in the DLR to benefit the code. If you have the type lookup inside the methods then that gets called every single time those methods are called. The other implementation idea then is to create method overloads for all the permutations of the parameter types. This leads to a massive increase in methods but could provide better performance at run time; It may happen already, or it certainly could in the future, that the DLR is able to calculate the method required from the calling code (i.e. Ruby code) and generate IL that will call that method directly during execution. Therefore the type lookup only needs to happen once. This would make a huge difference in some of the BigDecimal functionality where you may well be doing complex calculations and expecting a reasonable level of performance. The trouble with this is that it is a right pain to code up and leaves lots of room for programmer error. Also I imagine that this will lead to even longer start-up times for IronRuby, which I suspect is going to be the major performance issue going forward, what with all the code generation and JIT compilation. Any one got thoughts on this? Pete -------------- next part -------------- An HTML attachment was scrubbed... URL: From Tomas.Matousek at microsoft.com Sat Jul 19 15:49:57 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Sat, 19 Jul 2008 12:49:57 -0700 Subject: [Ironruby-core] Multiple overloads VS Inline casting In-Reply-To: <001901c8e9d2$0ebeac00$2c3c0400$@com> References: <001901c8e9d2$0ebeac00$2c3c0400$@com> Message-ID: We are well aware of this issue and have a plan to solve it: 1) You define overloads for type combinations that you want to optimize. I'm going to improve class initializers to have a less overhead. Even now though the initialization doesn't contribute significantly to startup time. 2) For the rest of the parameter type combinations whose performance is not critical you define a strongly typed overload(s) that can accept all of them. The conversions should happen automatically in the binder. There will be some well defined conversions + a set of attributes to customize them. The goal is not to perform type conversions imperatively in library code if possible. The predefined conversions will follow Ruby conversion protocols like to_i, to_s, etc. If a particular method doesn't support the default conversion (Ruby library methods are inconsistent on what conversions are used) or a there are multiple applicable conversions to chose from, attributes would be available to specify the right one. Tomas From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Saturday, July 19, 2008 12:03 PM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] Multiple overloads VS Inline casting I have noticed in the libraries that are often methods that take multiple parameters and each parameter can take a number of different types. A simple example, BigDecimal#add method has two parameters: a.add(b,n), where a is BigDecimal, b needs to be compatible with BigDecimal and n needs to be compatible with Fixnum. In other words it is possible that b and n can be one of [NotNull]BigDecimal/*!*/, [NotNull]BigInteger/*!*/, int, double and object. This leads to an implementation question. Initially it seemed sensible to abstract out the conversion of these types - except perhaps the best case ([NotNull]BigDecimal/*!*/ and int) - into a helper method and then have just: public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, BigDecimal/*!*/ b, int n) ... and public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, object b, object n) ... In the second method you convert the parameters to the correct types and pass them back to the first method. This is basically what happens in the current implementation of many of the Socket library methods. This provides a single place of conversion, which is particularly helpful given the BigDecimal class has lots of methods that take this form that will all use this conversion. It makes the code readable and doesn't create masses of overloads for the class. After thinking about this for a bit I started to wonder about performance. This style of method does not allow performance tricks in the DLR to benefit the code. If you have the type lookup inside the methods then that gets called every single time those methods are called. The other implementation idea then is to create method overloads for all the permutations of the parameter types. This leads to a massive increase in methods but could provide better performance at run time; It may happen already, or it certainly could in the future, that the DLR is able to calculate the method required from the calling code (i.e. Ruby code) and generate IL that will call that method directly during execution. Therefore the type lookup only needs to happen once. This would make a huge difference in some of the BigDecimal functionality where you may well be doing complex calculations and expecting a reasonable level of performance. The trouble with this is that it is a right pain to code up and leaves lots of room for programmer error. Also I imagine that this will lead to even longer start-up times for IronRuby, which I suspect is going to be the major performance issue going forward, what with all the code generation and JIT compilation. Any one got thoughts on this? Pete -------------- next part -------------- An HTML attachment was scrubbed... URL: From bacondarwin at googlemail.com Sat Jul 19 16:25:52 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Sat, 19 Jul 2008 21:25:52 +0100 Subject: [Ironruby-core] Multiple overloads VS Inline casting In-Reply-To: References: <001901c8e9d2$0ebeac00$2c3c0400$@com> Message-ID: <002401c8e9dd$a53d6e90$efb84bb0$@com> So in the example I gave, what would be the preferred set of overloads? Obviously from (1) there would be: public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, BigDecimal/*!*/ b, int n) ... but what form should the "strongly typed overload" that you describe in (2) take? Cheers, Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Tomas Matousek Sent: Saturday,19 July 19, 2008 20:50 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Multiple overloads VS Inline casting We are well aware of this issue and have a plan to solve it: 1) You define overloads for type combinations that you want to optimize. I'm going to improve class initializers to have a less overhead. Even now though the initialization doesn't contribute significantly to startup time. 2) For the rest of the parameter type combinations whose performance is not critical you define a strongly typed overload(s) that can accept all of them. The conversions should happen automatically in the binder. There will be some well defined conversions + a set of attributes to customize them. The goal is not to perform type conversions imperatively in library code if possible. The predefined conversions will follow Ruby conversion protocols like to_i, to_s, etc. If a particular method doesn't support the default conversion (Ruby library methods are inconsistent on what conversions are used) or a there are multiple applicable conversions to chose from, attributes would be available to specify the right one. Tomas From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Saturday, July 19, 2008 12:03 PM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] Multiple overloads VS Inline casting I have noticed in the libraries that are often methods that take multiple parameters and each parameter can take a number of different types. A simple example, BigDecimal#add method has two parameters: a.add(b,n), where a is BigDecimal, b needs to be compatible with BigDecimal and n needs to be compatible with Fixnum. In other words it is possible that b and n can be one of [NotNull]BigDecimal/*!*/, [NotNull]BigInteger/*!*/, int, double and object. This leads to an implementation question. Initially it seemed sensible to abstract out the conversion of these types - except perhaps the best case ([NotNull]BigDecimal/*!*/ and int) - into a helper method and then have just: public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, BigDecimal/*!*/ b, int n) ... and public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, object b, object n) ... In the second method you convert the parameters to the correct types and pass them back to the first method. This is basically what happens in the current implementation of many of the Socket library methods. This provides a single place of conversion, which is particularly helpful given the BigDecimal class has lots of methods that take this form that will all use this conversion. It makes the code readable and doesn't create masses of overloads for the class. After thinking about this for a bit I started to wonder about performance. This style of method does not allow performance tricks in the DLR to benefit the code. If you have the type lookup inside the methods then that gets called every single time those methods are called. The other implementation idea then is to create method overloads for all the permutations of the parameter types. This leads to a massive increase in methods but could provide better performance at run time; It may happen already, or it certainly could in the future, that the DLR is able to calculate the method required from the calling code (i.e. Ruby code) and generate IL that will call that method directly during execution. Therefore the type lookup only needs to happen once. This would make a huge difference in some of the BigDecimal functionality where you may well be doing complex calculations and expecting a reasonable level of performance. The trouble with this is that it is a right pain to code up and leaves lots of room for programmer error. Also I imagine that this will lead to even longer start-up times for IronRuby, which I suspect is going to be the major performance issue going forward, what with all the code generation and JIT compilation. Any one got thoughts on this? Pete -------------- next part -------------- An HTML attachment was scrubbed... URL: From Tomas.Matousek at microsoft.com Sat Jul 19 18:09:54 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Sat, 19 Jul 2008 15:09:54 -0700 Subject: [Ironruby-core] Multiple overloads VS Inline casting In-Reply-To: <002401c8e9dd$a53d6e90$efb84bb0$@com> References: <001901c8e9d2$0ebeac00$2c3c0400$@com> <002401c8e9dd$a53d6e90$efb84bb0$@com> Message-ID: If there is an implicit conversion from whatever types you need to convert to BigDecimal then this single overload should be sufficient: public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, BigDecimal/*!*/ b, int n) The second overload you use now public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, object b, object n) does conversions and call the first overload, right? In future the conversions will be described by attributes on the first overload and the second overload won't be necessary. It also depends how we define implicit conversions for BigDecimal. If an arbitrary method has a BigDecimal parameter should it be possible to pass Fixnum? If that is a common case (maybe with some exemptions) it might be reasonable to define an implicit conversion from Fixnum to BigDecimal. Then the single overload would handle all cases. Otherwise you would need overloads from all types of "b" parameter that this method can take. Also, if you want to prevent allocation of BigDecimal for "b" parameter when Fixnum is passed in (which might be a good optimization) you need public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, int b, int n) overload. Note that the implicit conversions and protocol attributes are not supported right now. It will take some time to do so. For now public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, BigDecimal/*!*/ b, int n) public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, object b, object n) overloads with manual conversions might be good enough. If you want to optimize add overloads for int and double, unless the only thing they do is to allocate BigDecimal and call the first one. Then it doesn't make sense to optimize them. Tomas From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Saturday, July 19, 2008 1:26 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Multiple overloads VS Inline casting So in the example I gave, what would be the preferred set of overloads? Obviously from (1) there would be: public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, BigDecimal/*!*/ b, int n) ... but what form should the "strongly typed overload" that you describe in (2) take? Cheers, Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Tomas Matousek Sent: Saturday,19 July 19, 2008 20:50 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Multiple overloads VS Inline casting We are well aware of this issue and have a plan to solve it: 1) You define overloads for type combinations that you want to optimize. I'm going to improve class initializers to have a less overhead. Even now though the initialization doesn't contribute significantly to startup time. 2) For the rest of the parameter type combinations whose performance is not critical you define a strongly typed overload(s) that can accept all of them. The conversions should happen automatically in the binder. There will be some well defined conversions + a set of attributes to customize them. The goal is not to perform type conversions imperatively in library code if possible. The predefined conversions will follow Ruby conversion protocols like to_i, to_s, etc. If a particular method doesn't support the default conversion (Ruby library methods are inconsistent on what conversions are used) or a there are multiple applicable conversions to chose from, attributes would be available to specify the right one. Tomas From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Saturday, July 19, 2008 12:03 PM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] Multiple overloads VS Inline casting I have noticed in the libraries that are often methods that take multiple parameters and each parameter can take a number of different types. A simple example, BigDecimal#add method has two parameters: a.add(b,n), where a is BigDecimal, b needs to be compatible with BigDecimal and n needs to be compatible with Fixnum. In other words it is possible that b and n can be one of [NotNull]BigDecimal/*!*/, [NotNull]BigInteger/*!*/, int, double and object. This leads to an implementation question. Initially it seemed sensible to abstract out the conversion of these types - except perhaps the best case ([NotNull]BigDecimal/*!*/ and int) - into a helper method and then have just: public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, BigDecimal/*!*/ b, int n) ... and public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, object b, object n) ... In the second method you convert the parameters to the correct types and pass them back to the first method. This is basically what happens in the current implementation of many of the Socket library methods. This provides a single place of conversion, which is particularly helpful given the BigDecimal class has lots of methods that take this form that will all use this conversion. It makes the code readable and doesn't create masses of overloads for the class. After thinking about this for a bit I started to wonder about performance. This style of method does not allow performance tricks in the DLR to benefit the code. If you have the type lookup inside the methods then that gets called every single time those methods are called. The other implementation idea then is to create method overloads for all the permutations of the parameter types. This leads to a massive increase in methods but could provide better performance at run time; It may happen already, or it certainly could in the future, that the DLR is able to calculate the method required from the calling code (i.e. Ruby code) and generate IL that will call that method directly during execution. Therefore the type lookup only needs to happen once. This would make a huge difference in some of the BigDecimal functionality where you may well be doing complex calculations and expecting a reasonable level of performance. The trouble with this is that it is a right pain to code up and leaves lots of room for programmer error. Also I imagine that this will lead to even longer start-up times for IronRuby, which I suspect is going to be the major performance issue going forward, what with all the code generation and JIT compilation. Any one got thoughts on this? Pete -------------- next part -------------- An HTML attachment was scrubbed... URL: From bacondarwin at googlemail.com Sun Jul 20 04:24:07 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Sun, 20 Jul 2008 09:24:07 +0100 Subject: [Ironruby-core] Multiple overloads VS Inline casting In-Reply-To: References: <001901c8e9d2$0ebeac00$2c3c0400$@com> <002401c8e9dd$a53d6e90$efb84bb0$@com> Message-ID: <000601c8ea41$fbe63100$f3b29300$@com> Thanks Tomas. From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Tomas Matousek Sent: Saturday,19 July 19, 2008 23:10 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Multiple overloads VS Inline casting If there is an implicit conversion from whatever types you need to convert to BigDecimal then this single overload should be sufficient: public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, BigDecimal/*!*/ b, int n) The second overload you use now public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, object b, object n) does conversions and call the first overload, right? In future the conversions will be described by attributes on the first overload and the second overload won't be necessary. It also depends how we define implicit conversions for BigDecimal. If an arbitrary method has a BigDecimal parameter should it be possible to pass Fixnum? If that is a common case (maybe with some exemptions) it might be reasonable to define an implicit conversion from Fixnum to BigDecimal. Then the single overload would handle all cases. Otherwise you would need overloads from all types of "b" parameter that this method can take. Also, if you want to prevent allocation of BigDecimal for "b" parameter when Fixnum is passed in (which might be a good optimization) you need public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, int b, int n) overload. Note that the implicit conversions and protocol attributes are not supported right now. It will take some time to do so. For now public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, BigDecimal/*!*/ b, int n) public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, object b, object n) overloads with manual conversions might be good enough. If you want to optimize add overloads for int and double, unless the only thing they do is to allocate BigDecimal and call the first one. Then it doesn't make sense to optimize them. Tomas From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Saturday, July 19, 2008 1:26 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Multiple overloads VS Inline casting So in the example I gave, what would be the preferred set of overloads? Obviously from (1) there would be: public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, BigDecimal/*!*/ b, int n) ... but what form should the "strongly typed overload" that you describe in (2) take? Cheers, Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Tomas Matousek Sent: Saturday,19 July 19, 2008 20:50 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Multiple overloads VS Inline casting We are well aware of this issue and have a plan to solve it: 1) You define overloads for type combinations that you want to optimize. I'm going to improve class initializers to have a less overhead. Even now though the initialization doesn't contribute significantly to startup time. 2) For the rest of the parameter type combinations whose performance is not critical you define a strongly typed overload(s) that can accept all of them. The conversions should happen automatically in the binder. There will be some well defined conversions + a set of attributes to customize them. The goal is not to perform type conversions imperatively in library code if possible. The predefined conversions will follow Ruby conversion protocols like to_i, to_s, etc. If a particular method doesn't support the default conversion (Ruby library methods are inconsistent on what conversions are used) or a there are multiple applicable conversions to chose from, attributes would be available to specify the right one. Tomas From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Saturday, July 19, 2008 12:03 PM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] Multiple overloads VS Inline casting I have noticed in the libraries that are often methods that take multiple parameters and each parameter can take a number of different types. A simple example, BigDecimal#add method has two parameters: a.add(b,n), where a is BigDecimal, b needs to be compatible with BigDecimal and n needs to be compatible with Fixnum. In other words it is possible that b and n can be one of [NotNull]BigDecimal/*!*/, [NotNull]BigInteger/*!*/, int, double and object. This leads to an implementation question. Initially it seemed sensible to abstract out the conversion of these types - except perhaps the best case ([NotNull]BigDecimal/*!*/ and int) - into a helper method and then have just: public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, BigDecimal/*!*/ b, int n) ... and public static BigDecimal/*!*/ Add(CodeContext/*!*/ context, BigDecimal/*!*/ self, object b, object n) ... In the second method you convert the parameters to the correct types and pass them back to the first method. This is basically what happens in the current implementation of many of the Socket library methods. This provides a single place of conversion, which is particularly helpful given the BigDecimal class has lots of methods that take this form that will all use this conversion. It makes the code readable and doesn't create masses of overloads for the class. After thinking about this for a bit I started to wonder about performance. This style of method does not allow performance tricks in the DLR to benefit the code. If you have the type lookup inside the methods then that gets called every single time those methods are called. The other implementation idea then is to create method overloads for all the permutations of the parameter types. This leads to a massive increase in methods but could provide better performance at run time; It may happen already, or it certainly could in the future, that the DLR is able to calculate the method required from the calling code (i.e. Ruby code) and generate IL that will call that method directly during execution. Therefore the type lookup only needs to happen once. This would make a huge difference in some of the BigDecimal functionality where you may well be doing complex calculations and expecting a reasonable level of performance. The trouble with this is that it is a right pain to code up and leaves lots of room for programmer error. Also I imagine that this will lead to even longer start-up times for IronRuby, which I suspect is going to be the major performance issue going forward, what with all the code generation and JIT compilation. Any one got thoughts on this? Pete -------------- next part -------------- An HTML attachment was scrubbed... URL: From ben2004uk at googlemail.com Sun Jul 20 06:07:16 2008 From: ben2004uk at googlemail.com (Ben Hall) Date: Sun, 20 Jul 2008 11:07:16 +0100 Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages Message-ID: Hi everyone, Last night I uploaded an initial release of an application I've been working on - IronEditor. This is a simple application designed to make it easier to start coding against the DLR based languages, out of the box it supports IronPython and IronRuby. The reason why I decided to build this is because Visual Studio Integration for the languages is a long way off and playing around creating code to for the languages is painful via the provided console applications. As such, the aim of the application is to provide a very lightweight way to edit and execute code, great while learning the languages and giving demos. Blog Post: http://blog.benhall.me.uk/2008/07/announcing-ironeditor-editor-for.html Download: http://www.codeplex.com/IronEditor Build: 1.0.0.46 This is a release early with not too many advanced features but it can execute code! :) Over the next few weeks and months, I will be adding more features and implementing additional parts of the Hosting API. I will also try to blog about the actual implementation of IronEditor. Any feedback or feature suggestions you have on the application would be great! Thanks Ben Blog.BenHall.me.uk Microsoft MVP (C#) From bacondarwin at googlemail.com Sun Jul 20 08:13:05 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Sun, 20 Jul 2008 13:13:05 +0100 Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages In-Reply-To: References: Message-ID: <000e01c8ea61$f888c7a0$e99a56e0$@com> Don't mean to be a party-pooper - this project looks cool and could provide a really good alternative to VS for a range of developers. But how about writing a plug-in for SharpDevelop instead? This would give you a head start on a lot of the base-line functionality and let you concentrate on the Iron... stuff. Pete -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Ben Hall Sent: Sunday,20 July 20, 2008 11:07 To: ironruby-core at rubyforge.org Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages Hi everyone, Last night I uploaded an initial release of an application I've been working on - IronEditor. This is a simple application designed to make it easier to start coding against the DLR based languages, out of the box it supports IronPython and IronRuby. The reason why I decided to build this is because Visual Studio Integration for the languages is a long way off and playing around creating code to for the languages is painful via the provided console applications. As such, the aim of the application is to provide a very lightweight way to edit and execute code, great while learning the languages and giving demos. Blog Post: http://blog.benhall.me.uk/2008/07/announcing-ironeditor-editor-for.html Download: http://www.codeplex.com/IronEditor Build: 1.0.0.46 This is a release early with not too many advanced features but it can execute code! :) Over the next few weeks and months, I will be adding more features and implementing additional parts of the Hosting API. I will also try to blog about the actual implementation of IronEditor. Any feedback or feature suggestions you have on the application would be great! Thanks Ben Blog.BenHall.me.uk Microsoft MVP (C#) _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From ben2004uk at googlemail.com Sun Jul 20 11:04:46 2008 From: ben2004uk at googlemail.com (Ben Hall) Date: Sun, 20 Jul 2008 16:04:46 +0100 Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages In-Reply-To: <000e01c8ea61$f888c7a0$e99a56e0$@com> References: <000e01c8ea61$f888c7a0$e99a56e0$@com> Message-ID: Hi Peter, It's definitely a valid suggestion! The reason why I didn't go down the route of plugging into SharpDevelop \ MonoDevelop is because I wanted to provide something quick, lightweight and easy to start using without having to install too many additional items. I want the application to be focused on the DLR and dynamic languages, not sure hacked on the side. If I want to write a quick Ruby script, I don't want to wait 10 minutes for the IDE to load and create a new solution etc - I just want a way to write the code, maybe debug, maybe unit test and execute. Hopefully, IronEditor will provide that support. I'm not 100% confident you could easily fit a dynamic language into an IDE developed with statically typed languages in mind without either changing a lot of the core IDE functionality, or not taking full advantage of the language itself. It will be interesting to see how the team copes with VS integration. Maybe I am going down the wrong route, if there is a strong feeling for having this integrated instead of a standalone application then I will take a much closer look. Thanks for the comment. Ben On Sun, Jul 20, 2008 at 1:13 PM, Peter Bacon Darwin wrote: > Don't mean to be a party-pooper - this project looks cool and could provide > a really good alternative to VS for a range of developers. > > But how about writing a plug-in for SharpDevelop instead? This would give > you a head start on a lot of the base-line functionality and let you > concentrate on the Iron... stuff. > > Pete > > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Ben Hall > Sent: Sunday,20 July 20, 2008 11:07 > To: ironruby-core at rubyforge.org > Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, > IronPython and other DLR languages > > Hi everyone, > > Last night I uploaded an initial release of an application I've been > working on - IronEditor. This is a simple application designed to make > it easier to start coding against the DLR based languages, out of the > box it supports IronPython and IronRuby. > > The reason why I decided to build this is because Visual Studio > Integration for the languages is a long way off and playing around > creating code to for the languages is painful via the provided console > applications. As such, the aim of the application is to provide a very > lightweight way to edit and execute code, great while learning the > languages and giving demos. > > Blog Post: > http://blog.benhall.me.uk/2008/07/announcing-ironeditor-editor-for.html > > Download: http://www.codeplex.com/IronEditor > Build: 1.0.0.46 > > This is a release early with not too many advanced features but it can > execute code! :) Over the next few weeks and months, I will be adding > more features and implementing additional parts of the Hosting API. I > will also try to blog about the actual implementation of IronEditor. > > Any feedback or feature suggestions you have on the application would be > great! > > Thanks > > Ben > Blog.BenHall.me.uk > Microsoft MVP (C#) > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > From info at counterforensics.net Sun Jul 20 14:42:59 2008 From: info at counterforensics.net (Michael F. Graham) Date: Sun, 20 Jul 2008 11:42:59 -0700 Subject: [Ironruby-core] Ironruby-core Digest, Vol 11, Issue 44 In-Reply-To: References: Message-ID: <3055C528CE1D401F9CDCBBFF7DDB3EB4@cracker01> Sapphire Steel Software has a version of their IDE that supports Iron Ruby.More details can be found here: http://www.sapphiresteel.com/Ruby-In-Steel-For-IronRuby I have been very pleased with Ruby in Steel as my development environment and will be experimenting with their Iron Ruby IDE once I finish building my eHawala financial transaction tool kit. Mike Graham ------------------------------ Message: 2 Date: Sun, 20 Jul 2008 11:07:16 +0100 From: "Ben Hall" Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages To: ironruby-core at rubyforge.org Message-ID: Content-Type: text/plain; charset=ISO-8859-1 Hi everyone, Last night I uploaded an initial release of an application I've been working on - IronEditor. This is a simple application designed to make it easier to start coding against the DLR based languages, out of the box it supports IronPython and IronRuby. The reason why I decided to build this is because Visual Studio Integration for the languages is a long way off and playing around creating code to for the languages is painful via the provided console applications. As such, the aim of the application is to provide a very lightweight way to edit and execute code, great while learning the languages and giving demos. Blog Post: http://blog.benhall.me.uk/2008/07/announcing-ironeditor-editor-for.html Download: http://www.codeplex.com/IronEditor Build: 1.0.0.46 This is a release early with not too many advanced features but it can execute code! :) Over the next few weeks and months, I will be adding more features and implementing additional parts of the Hosting API. I will also try to blog about the actual implementation of IronEditor. Any feedback or feature suggestions you have on the application would be great! Thanks Ben Blog.BenHall.me.uk Microsoft MVP (C#) From lists at ruby-forum.com Sun Jul 20 16:21:16 2008 From: lists at ruby-forum.com (Huw Collingbourne) Date: Sun, 20 Jul 2008 22:21:16 +0200 Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages In-Reply-To: References: <000e01c8ea61$f888c7a0$e99a56e0$@com> Message-ID: <707bfd532c1ec8f2e82c0fbb2a0913f7@ruby-forum.com> Ben Hall wrote: > I'm not 100% confident you could easily fit a dynamic language into an > IDE developed with statically typed languages in mind without either > changing a lot of the core IDE functionality, or not taking full > advantage of the language itself. It can be done ;-) Standard Ruby (with all the bells and whistles - editing/debugging/intellisense etc.): http://www.sapphiresteel.com/Ruby-In-Steel-Developer-Overview IronRuby (editing/design - more stuff coming later): http://www.sapphiresteel.com/Ruby-In-Steel-For-IronRuby Good luck with the editor. There will always be some people who prefer an IDE and others who prefer an editor. I wish you well. all the best Huw -- Posted via http://www.ruby-forum.com/. From olegtk at microsoft.com Mon Jul 21 00:53:08 2008 From: olegtk at microsoft.com (Oleg Tkachenko) Date: Sun, 20 Jul 2008 21:53:08 -0700 Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages In-Reply-To: References: Message-ID: This is cool! There is definitely a need for a lightweight cross platform IronRuby/IronPython (or even any DLR language) editor. I wish it was based on WPF, but then it won't run on Mono :( Let's coordinate our efforts? -- Oleg -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Ben Hall Sent: Sunday, July 20, 2008 3:07 AM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages Hi everyone, Last night I uploaded an initial release of an application I've been working on - IronEditor. This is a simple application designed to make it easier to start coding against the DLR based languages, out of the box it supports IronPython and IronRuby. The reason why I decided to build this is because Visual Studio Integration for the languages is a long way off and playing around creating code to for the languages is painful via the provided console applications. As such, the aim of the application is to provide a very lightweight way to edit and execute code, great while learning the languages and giving demos. Blog Post: http://blog.benhall.me.uk/2008/07/announcing-ironeditor-editor-for.html Download: http://www.codeplex.com/IronEditor Build: 1.0.0.46 This is a release early with not too many advanced features but it can execute code! :) Over the next few weeks and months, I will be adding more features and implementing additional parts of the Hosting API. I will also try to blog about the actual implementation of IronEditor. Any feedback or feature suggestions you have on the application would be great! Thanks Ben Blog.BenHall.me.uk Microsoft MVP (C#) _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From stefan.dobrev at gmail.com Mon Jul 21 01:17:45 2008 From: stefan.dobrev at gmail.com (Stefan Dobrev) Date: Mon, 21 Jul 2008 08:17:45 +0300 Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages In-Reply-To: References: Message-ID: <928de89c0807202217l41c10183l73d6a11dcdc3aaa2@mail.gmail.com> If it is Silverlight API based it "should" run on Mono as well. 2008/7/21 Oleg Tkachenko : > This is cool! There is definitely a need for a lightweight cross platform > IronRuby/IronPython (or even any DLR language) editor. > I wish it was based on WPF, but then it won't run on Mono :( > Let's coordinate our efforts? > > -- > Oleg > > > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto: > ironruby-core-bounces at rubyforge.org] On Behalf Of Ben Hall > Sent: Sunday, July 20, 2008 3:07 AM > To: ironruby-core at rubyforge.org > Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, > IronPython and other DLR languages > > Hi everyone, > > Last night I uploaded an initial release of an application I've been > working on - IronEditor. This is a simple application designed to make > it easier to start coding against the DLR based languages, out of the > box it supports IronPython and IronRuby. > > The reason why I decided to build this is because Visual Studio > Integration for the languages is a long way off and playing around > creating code to for the languages is painful via the provided console > applications. As such, the aim of the application is to provide a very > lightweight way to edit and execute code, great while learning the > languages and giving demos. > > Blog Post: > http://blog.benhall.me.uk/2008/07/announcing-ironeditor-editor-for.html > > Download: http://www.codeplex.com/IronEditor > Build: 1.0.0.46 > > This is a release early with not too many advanced features but it can > execute code! :) Over the next few weeks and months, I will be adding > more features and implementing additional parts of the Hosting API. I > will also try to blog about the actual implementation of IronEditor. > > Any feedback or feature suggestions you have on the application would be > great! > > Thanks > > Ben > Blog.BenHall.me.uk > Microsoft MVP (C#) > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > -------------- next part -------------- An HTML attachment was scrubbed... URL: From lists at ruby-forum.com Mon Jul 21 03:21:20 2008 From: lists at ruby-forum.com (Softmind Technology) Date: Mon, 21 Jul 2008 09:21:20 +0200 Subject: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions In-Reply-To: <12D498AE0ADB20479A22F41E113B67D302BE33CE2F@sun.irm.se> References: <7ca58dc067dc9a67c58ac102feed5325@ruby-forum.com> <372109E149E8084D8E6C7D9CFD82E0632FD40AFED2@NA-EXMSG-C115.redmond.corp.microsoft.com> <12D498AE0ADB20479A22F41E113B67D302BE33CE2D@sun.irm.se> <12D498AE0ADB20479A22F41E113B67D302BE33CE2F@sun.irm.se> Message-ID: More Information on:- Asp.Net MVC with IronRuby Just in case, you would like to have more details and a working sample its here.... http://haacked.com/archive/2008/07/20/ironruby-aspnetmvc-prototype.aspx Download a working code from the above link. Hope this helps. BTW... does any one have any links for IronRuby with Active Record/DataMapper/ SubSonic / Linqtosql i'm rather more interested in that. Thanks -- Posted via http://www.ruby-forum.com/. From ivan at flanders.co.nz Mon Jul 21 03:57:35 2008 From: ivan at flanders.co.nz (Ivan Porto Carrero) Date: Mon, 21 Jul 2008 09:57:35 +0200 Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages In-Reply-To: <928de89c0807202217l41c10183l73d6a11dcdc3aaa2@mail.gmail.com> References: <928de89c0807202217l41c10183l73d6a11dcdc3aaa2@mail.gmail.com> Message-ID: But that isn't at this time is it? AFAIK mono currently only supports Silverlight 1.0 or am I wrong in my assumption? On Mon, Jul 21, 2008 at 7:17 AM, Stefan Dobrev wrote: > If it is Silverlight API based it "should" run on Mono as well. > > 2008/7/21 Oleg Tkachenko : > > This is cool! There is definitely a need for a lightweight cross platform >> IronRuby/IronPython (or even any DLR language) editor. >> I wish it was based on WPF, but then it won't run on Mono :( >> Let's coordinate our efforts? >> >> -- >> Oleg >> >> >> -----Original Message----- >> From: ironruby-core-bounces at rubyforge.org [mailto: >> ironruby-core-bounces at rubyforge.org] On Behalf Of Ben Hall >> Sent: Sunday, July 20, 2008 3:07 AM >> To: ironruby-core at rubyforge.org >> Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, >> IronPython and other DLR languages >> >> Hi everyone, >> >> Last night I uploaded an initial release of an application I've been >> working on - IronEditor. This is a simple application designed to make >> it easier to start coding against the DLR based languages, out of the >> box it supports IronPython and IronRuby. >> >> The reason why I decided to build this is because Visual Studio >> Integration for the languages is a long way off and playing around >> creating code to for the languages is painful via the provided console >> applications. As such, the aim of the application is to provide a very >> lightweight way to edit and execute code, great while learning the >> languages and giving demos. >> >> Blog Post: >> http://blog.benhall.me.uk/2008/07/announcing-ironeditor-editor-for.html >> >> Download: http://www.codeplex.com/IronEditor >> Build: 1.0.0.46 >> >> This is a release early with not too many advanced features but it can >> execute code! :) Over the next few weeks and months, I will be adding >> more features and implementing additional parts of the Hosting API. I >> will also try to blog about the actual implementation of IronEditor. >> >> Any feedback or feature suggestions you have on the application would be >> great! >> >> Thanks >> >> Ben >> Blog.BenHall.me.uk >> Microsoft MVP (C#) >> _______________________________________________ >> Ironruby-core mailing list >> Ironruby-core at rubyforge.org >> http://rubyforge.org/mailman/listinfo/ironruby-core >> >> _______________________________________________ >> Ironruby-core mailing list >> Ironruby-core at rubyforge.org >> http://rubyforge.org/mailman/listinfo/ironruby-core >> > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Johan.Danforth at irm.se Mon Jul 21 04:39:20 2008 From: Johan.Danforth at irm.se (Johan Danforth) Date: Mon, 21 Jul 2008 10:39:20 +0200 Subject: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions In-Reply-To: <12D498AE0ADB20479A22F41E113B67D302BE33CE2F@sun.irm.se> References: <7ca58dc067dc9a67c58ac102feed5325@ruby-forum.com> <372109E149E8084D8E6C7D9CFD82E0632FD40AFED2@NA-EXMSG-C115.redmond.corp.microsoft.com> <12D498AE0ADB20479A22F41E113B67D302BE33CE2D@sun.irm.se> <12D498AE0ADB20479A22F41E113B67D302BE33CE2F@sun.irm.se> Message-ID: <12D498AE0ADB20479A22F41E113B67D302BE33CE4B@sun.irm.se> Phil Haack just posted a blog entry with a working IronRuby + ASP.NET MVC sample :) http://haacked.com/archive/2008/07/20/ironruby-aspnetmvc-prototype.aspx /Johan -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Johan Danforth Sent: den 19 juli 2008 13:15 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions And a minute later I found a zip file on your blog entry (http://www.iunknown.com/2008/06/ironruby-and-aspnet-mvc.html) John. I'll have a look at that at once. Sorry for being too trigger happy... :/ /Johan -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Johan Danforth Sent: den 19 juli 2008 13:12 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions The question has been asked on the MVC forum, but Levi (I think it was) wrote: "What John is saying is that you can download the ZIP that allows IronRuby to run in ASP MVC and that the project will run on top of Preview 4 bits" I like to work with both ASP.NET MVC and IronRuby, so I would love to get my hands on that ZIP ;) /Johan -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: den 9 juli 2008 17:39 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions The ASP.NET MVC questions should be asked on the ASP.NET MVC forums. The current plan is to ship IronRuby early next year. Thanks, -John -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Rahil Kumar Sent: Wednesday, July 09, 2008 4:54 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions Web Reservoir wrote: > Hi, > > May i know, how many months are we still far from IronRuby 1.0...? It was july 2008 previously. It seems its delayed. I am also waiting for a confirmed date. > > I understand Asp.Net MVC preview 4 will support IronRuby. Asp.Net MVC 4 is scheduled to be in this month only, perhaps mid july. I am not aware about any do's / dont's with this preview. I am sailing in the same boat like you. -- Posted via http://www.ruby-forum.com/. _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From ben2004uk at googlemail.com Mon Jul 21 05:43:58 2008 From: ben2004uk at googlemail.com (Ben Hall) Date: Mon, 21 Jul 2008 10:43:58 +0100 Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages In-Reply-To: References: <928de89c0807202217l41c10183l73d6a11dcdc3aaa2@mail.gmail.com> Message-ID: Hi guys, Thanks for the comments. DevHawk (Harry Pierson) mentioned on twitter that he would have liked to see the application developed onto of WPF. It would have added coolness (3D error syntax checking?), but there is the added problem of Mono support which is why I went for WinForms. The Engine is separated from the UI, so WPF could be added at a later date, well almost - I think a bit too much logic is in the UI at the moment :( As for Silverlight, yes this would be cool (more cool than WPF!) however Moonlight's current "support for Silverlight 2.0 is in pre-Alpha stage". I would love to have a Silverlight UI on top of this (more than WPF) but I wanted to get WinForms app out the way (which I think is more useful if not as cool). Maybe when the WinForms app has the core functionality in place, I can look at doing something similar for Silverlight. Any other comments, suggestions or feature requests please let me know. Thanks Ben On Mon, Jul 21, 2008 at 8:57 AM, Ivan Porto Carrero wrote: > But that isn't at this time is it? AFAIK mono currently only supports > Silverlight 1.0 or am I wrong in my assumption? > > On Mon, Jul 21, 2008 at 7:17 AM, Stefan Dobrev > wrote: >> >> If it is Silverlight API based it "should" run on Mono as well. >> >> 2008/7/21 Oleg Tkachenko : >>> >>> This is cool! There is definitely a need for a lightweight cross platform >>> IronRuby/IronPython (or even any DLR language) editor. >>> I wish it was based on WPF, but then it won't run on Mono :( >>> Let's coordinate our efforts? >>> >>> -- >>> Oleg >>> >>> >>> -----Original Message----- >>> From: ironruby-core-bounces at rubyforge.org >>> [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Ben Hall >>> Sent: Sunday, July 20, 2008 3:07 AM >>> To: ironruby-core at rubyforge.org >>> Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, >>> IronPython and other DLR languages >>> >>> Hi everyone, >>> >>> Last night I uploaded an initial release of an application I've been >>> working on - IronEditor. This is a simple application designed to make >>> it easier to start coding against the DLR based languages, out of the >>> box it supports IronPython and IronRuby. >>> >>> The reason why I decided to build this is because Visual Studio >>> Integration for the languages is a long way off and playing around >>> creating code to for the languages is painful via the provided console >>> applications. As such, the aim of the application is to provide a very >>> lightweight way to edit and execute code, great while learning the >>> languages and giving demos. >>> >>> Blog Post: >>> http://blog.benhall.me.uk/2008/07/announcing-ironeditor-editor-for.html >>> >>> Download: http://www.codeplex.com/IronEditor >>> Build: 1.0.0.46 >>> >>> This is a release early with not too many advanced features but it can >>> execute code! :) Over the next few weeks and months, I will be adding >>> more features and implementing additional parts of the Hosting API. I >>> will also try to blog about the actual implementation of IronEditor. >>> >>> Any feedback or feature suggestions you have on the application would be >>> great! >>> >>> Thanks >>> >>> Ben >>> Blog.BenHall.me.uk >>> Microsoft MVP (C#) >>> _______________________________________________ >>> Ironruby-core mailing list >>> Ironruby-core at rubyforge.org >>> http://rubyforge.org/mailman/listinfo/ironruby-core >>> >>> _______________________________________________ >>> Ironruby-core mailing list >>> Ironruby-core at rubyforge.org >>> http://rubyforge.org/mailman/listinfo/ironruby-core >> >> >> _______________________________________________ >> Ironruby-core mailing list >> Ironruby-core at rubyforge.org >> http://rubyforge.org/mailman/listinfo/ironruby-core >> > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > From lists at ruby-forum.com Mon Jul 21 06:34:04 2008 From: lists at ruby-forum.com (Softmind Technology) Date: Mon, 21 Jul 2008 12:34:04 +0200 Subject: [Ironruby-core] IronRuby 1.0 and Asp.Net MVC preview 4 Questions In-Reply-To: <12D498AE0ADB20479A22F41E113B67D302BE33CE4B@sun.irm.se> References: <7ca58dc067dc9a67c58ac102feed5325@ruby-forum.com> <372109E149E8084D8E6C7D9CFD82E0632FD40AFED2@NA-EXMSG-C115.redmond.corp.microsoft.com> <12D498AE0ADB20479A22F41E113B67D302BE33CE2D@sun.irm.se> <12D498AE0ADB20479A22F41E113B67D302BE33CE2F@sun.irm.se> <12D498AE0ADB20479A22F41E113B67D302BE33CE4B@sun.irm.se> Message-ID: <6c223bc6351a8540eb5a61d992cc4f26@ruby-forum.com> Johan Danforth wrote: > http://haacked.com/archive/2008/07/20/ironruby-aspnetmvc-prototype.aspx > > /Johan Hi, It seems, its the same link, i have posted above. Thanks -- Posted via http://www.ruby-forum.com/. From lists at ruby-forum.com Mon Jul 21 08:11:00 2008 From: lists at ruby-forum.com (Aaron Clauson) Date: Mon, 21 Jul 2008 14:11:00 +0200 Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages In-Reply-To: References: <928de89c0807202217l41c10183l73d6a11dcdc3aaa2@mail.gmail.com> Message-ID: Ben Hall wrote: > As for Silverlight, yes this would be cool (more cool than WPF!) > however Moonlight's current "support for Silverlight 2.0 is in > pre-Alpha stage". I would love to have a Silverlight UI on top of this > (more than WPF) but I wanted to get WinForms app out the way (which I > think is more useful if not as cool). Maybe when the WinForms app has > the core functionality in place, I can look at doing something similar > for Silverlight. Hi Ben, I'd put in a big vote for an IronRuby editor done up in Silverlight. I've got a bunch of 500 or so users who are struggling to get a grasp on IronRuby who would thank you as well, we get lots of Ruby usage questions on our forums http://www.mysipswitch.com/forum/viewforum.php?f=6 (that link is also for a free, open source project so no plugging is involved here). Any kind of syntax highlighting and/or test execution would definitely help people who aren't that used to scripting let alone Ruby to get a handle on things. Regards, Aaron -- Posted via http://www.ruby-forum.com/. From lists at ruby-forum.com Mon Jul 21 09:30:54 2008 From: lists at ruby-forum.com (Softmind Technology) Date: Mon, 21 Jul 2008 15:30:54 +0200 Subject: [Ironruby-core] Good Efforts by CodeThinked.com for IronRuby Message-ID: <307d7076337c4bc359b430d77247252c@ruby-forum.com> Hi, I would suggest to keep an eye on http://www.codethinked.com/ A very good series of blogs for " IronRuby " are shaping yup there. My special thanks to the blogger for his contribution and special focus on IronRuby. SoftMind -- Posted via http://www.ruby-forum.com/. From Jimmy.Schementi at microsoft.com Mon Jul 21 10:58:16 2008 From: Jimmy.Schementi at microsoft.com (Jimmy Schementi) Date: Mon, 21 Jul 2008 07:58:16 -0700 Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages In-Reply-To: References: <928de89c0807202217l41c10183l73d6a11dcdc3aaa2@mail.gmail.com> Message-ID: <1EA05042-0B3C-4B8F-ABC1-E972EFE88350@microsoft.com> Yep, that's true. Moonlight today only has basic support for sl2 stuff. However, if based on silverlight apis then eventually moonlight will run it. And, a moonlight app can be run just like any desktop app on Linux. On Jul 21, 2008, at 3:57 AM, "Ivan Porto Carrero" > wrote: But that isn't at this time is it? AFAIK mono currently only supports Silverlight 1.0 or am I wrong in my assumption? On Mon, Jul 21, 2008 at 7:17 AM, Stefan Dobrev <stefan.dobrev at gmail.com> wrote: If it is Silverlight API based it "should" run on Mono as well. 2008/7/21 Oleg Tkachenko <olegtk at microsoft.com>: This is cool! There is definitely a need for a lightweight cross platform IronRuby/IronPython (or even any DLR language) editor. I wish it was based on WPF, but then it won't run on Mono :( Let's coordinate our efforts? -- Oleg -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Ben Hall Sent: Sunday, July 20, 2008 3:07 AM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages Hi everyone, Last night I uploaded an initial release of an application I've been working on - IronEditor. This is a simple application designed to make it easier to start coding against the DLR based languages, out of the box it supports IronPython and IronRuby. The reason why I decided to build this is because Visual Studio Integration for the languages is a long way off and playing around creating code to for the languages is painful via the provided console applications. As such, the aim of the application is to provide a very lightweight way to edit and execute code, great while learning the languages and giving demos. Blog Post: http://blog.benhall.me.uk/2008/07/announcing-ironeditor-editor-for.html Download: http://www.codeplex.com/IronEditor Build: 1.0.0.46 This is a release early with not too many advanced features but it can execute code! :) Over the next few weeks and months, I will be adding more features and implementing additional parts of the Hosting API. I will also try to blog about the actual implementation of IronEditor. Any feedback or feature suggestions you have on the application would be great! Thanks Ben Blog.BenHall.me.uk Microsoft MVP (C#) _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core -------------- next part -------------- An HTML attachment was scrubbed... URL: From ben2004uk at googlemail.com Mon Jul 21 11:00:03 2008 From: ben2004uk at googlemail.com (Ben Hall) Date: Mon, 21 Jul 2008 16:00:03 +0100 Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages In-Reply-To: <1EA05042-0B3C-4B8F-ABC1-E972EFE88350@microsoft.com> References: <928de89c0807202217l41c10183l73d6a11dcdc3aaa2@mail.gmail.com> <1EA05042-0B3C-4B8F-ABC1-E972EFE88350@microsoft.com> Message-ID: It's a shame Silverlight doesn't offer the same flexibility. Well, I say that. I have had Silverlight working in this way before so maybe I could get it to work. On Mon, Jul 21, 2008 at 3:58 PM, Jimmy Schementi wrote: > Yep, that's true. Moonlight today only has basic support for sl2 stuff. > However, if based on silverlight apis then eventually moonlight will run it. > And, a moonlight app can be run just like any desktop app on Linux. > > On Jul 21, 2008, at 3:57 AM, "Ivan Porto Carrero" > wrote: > > But that isn't at this time is it? AFAIK mono currently only supports > Silverlight 1.0 or am I wrong in my assumption? > > On Mon, Jul 21, 2008 at 7:17 AM, Stefan Dobrev > wrote: >> >> If it is Silverlight API based it "should" run on Mono as well. >> >> 2008/7/21 Oleg Tkachenko : >>> >>> This is cool! There is definitely a need for a lightweight cross platform >>> IronRuby/IronPython (or even any DLR language) editor. >>> I wish it was based on WPF, but then it won't run on Mono :( >>> Let's coordinate our efforts? >>> >>> -- >>> Oleg >>> >>> >>> -----Original Message----- >>> From: ironruby-core-bounces at rubyforge.org >>> [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Ben Hall >>> Sent: Sunday, July 20, 2008 3:07 AM >>> To: ironruby-core at rubyforge.org >>> Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, >>> IronPython and other DLR languages >>> >>> Hi everyone, >>> >>> Last night I uploaded an initial release of an application I've been >>> working on - IronEditor. This is a simple application designed to make >>> it easier to start coding against the DLR based languages, out of the >>> box it supports IronPython and IronRuby. >>> >>> The reason why I decided to build this is because Visual Studio >>> Integration for the languages is a long way off and playing around >>> creating code to for the languages is painful via the provided console >>> applications. As such, the aim of the application is to provide a very >>> lightweight way to edit and execute code, great while learning the >>> languages and giving demos. >>> >>> Blog Post: >>> http://blog.benhall.me.uk/2008/07/announcing-ironeditor-editor-for.html >>> >>> Download: http://www.codeplex.com/IronEditor >>> Build: 1.0.0.46 >>> >>> This is a release early with not too many advanced features but it can >>> execute code! :) Over the next few weeks and months, I will be adding >>> more features and implementing additional parts of the Hosting API. I >>> will also try to blog about the actual implementation of IronEditor. >>> >>> Any feedback or feature suggestions you have on the application would be >>> great! >>> >>> Thanks >>> >>> Ben >>> Blog.BenHall.me.uk >>> Microsoft MVP (C#) >>> _______________________________________________ >>> Ironruby-core mailing list >>> Ironruby-core at rubyforge.org >>> http://rubyforge.org/mailman/listinfo/ironruby-core >>> >>> _______________________________________________ >>> Ironruby-core mailing list >>> Ironruby-core at rubyforge.org >>> http://rubyforge.org/mailman/listinfo/ironruby-core >> >> >> _______________________________________________ >> Ironruby-core mailing list >> Ironruby-core at rubyforge.org >> http://rubyforge.org/mailman/listinfo/ironruby-core >> > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > From stefan.dobrev at gmail.com Mon Jul 21 14:12:16 2008 From: stefan.dobrev at gmail.com (Stefan Dobrev) Date: Mon, 21 Jul 2008 21:12:16 +0300 Subject: [Ironruby-core] Fwd: [IronPython] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages In-Reply-To: <928de89c0807211109o20db5955l7db817d47eb3848e@mail.gmail.com> References: <928de89c0807211109o20db5955l7db817d47eb3848e@mail.gmail.com> Message-ID: <928de89c0807211112s1ae48e3cjf5af28cf6973ca92@mail.gmail.com> I have looked at the code. Some neat stuff. However my idea is that we can leverage as much as we can from the DLR platform itself. For example we can create our own editor and use the code sense and token infrastructure for intellisense and syntax highlighting. I have a working version of that implemented in WPF using its RichTextBox. Unfortunately the RichTextBox DOM (FlowDocument) is too heavy for code editor and runs poorly. We need a lightweight version of RichTextBox for code editor - our custom control. It should also run on Silverlight as well. Expression Blend and PowerShell v2 have implemented similar controls in WPF for their code editing (Blend - XAML; PowerShell - cmdlet/script editing). So it can be done. Then we can build a lightweight IDE (like TextMate) for both Silverlight and WPF with a transperant project model that works both on web and locally. All this seems to be a perfect project for showcasing DLR API - architecture and services as well as Silverlight / WPF. Comments? Regards, Stefan 2008/7/20 Ben Hall : Hi everyone, > > Last night I uploaded an initial release of an application I've been > working on - IronEditor. This is a simple application designed to make > it easier to start coding against the DLR based languages, out of the > box it supports IronPython and IronRuby. > > The reason why I decided to build this is because Visual Studio > Integration for the languages is a long way off and playing around > creating code to for the languages is painful via the provided console > applications. As such, the aim of the application is to provide a very > lightweight way to edit and execute code, great while learning the > languages and giving demos. > > Blog Post: > http://blog.benhall.me.uk/2008/07/announcing-ironeditor-editor-for.html > > Download: http://www.codeplex.com/IronEditor > Build: 1.0.0.46 > > This is a release early with not too many advanced features but it can > execute code! :) Over the next few weeks and months, I will be adding > more features and implementing additional parts of the Hosting API. I > will also try to blog about the actual implementation of IronEditor. > > Any feedback or feature suggestions you have on the application would be > great! > > Thanks > > Ben > Blog.BenHall.me.uk > Microsoft MVP (C#) > _______________________________________________ > Users mailing list > Users at lists.ironpython.com > http://lists.ironpython.com/listinfo.cgi/users-ironpython.com > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ksunair at yahoo.com Mon Jul 21 16:31:43 2008 From: ksunair at yahoo.com (Unnikrishnan Nair) Date: Mon, 21 Jul 2008 13:31:43 -0700 (PDT) Subject: [Ironruby-core] Good Efforts by CodeThinked.com for IronRuby In-Reply-To: <307d7076337c4bc359b430d77247252c@ruby-forum.com> Message-ID: <813003.44368.qm@web50602.mail.re2.yahoo.com> Sorry if it is duplicate, getting some starge errors on outlook. Also sorry for blowing my own horn :) ? I have been writing some blogs for my friends who are C# developers on introduction to IronRuby with C#. It is at http://www.myfavoritemovies.us/Ironruby It is for beginners, especially for people coming from C#. I thought I will throw it out as well. If any of you have any comments or suggestions, I will be more than happy incorporate them there. ? Thanks, Unni --- On Mon, 7/21/08, Softmind Technology wrote: From: Softmind Technology Subject: [Ironruby-core] Good Efforts by CodeThinked.com for IronRuby To: ironruby-core at rubyforge.org Date: Monday, July 21, 2008, 7:30 AM Hi, I would suggest to keep an eye on http://www.codethinked.com/ A very good series of blogs for " IronRuby " are shaping yup there. My special thanks to the blogger for his contribution and special focus on IronRuby. SoftMind -- Posted via http://www.ruby-forum.com/. _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core -------------- next part -------------- An HTML attachment was scrubbed... URL: From ben2004uk at googlemail.com Mon Jul 21 19:59:14 2008 From: ben2004uk at googlemail.com (Ben Hall) Date: Tue, 22 Jul 2008 00:59:14 +0100 Subject: [Ironruby-core] Announcing IronEditor - An Editor for IronRuby, IronPython and other DLR languages In-Reply-To: References: <928de89c0807202217l41c10183l73d6a11dcdc3aaa2@mail.gmail.com> Message-ID: Hi Aaron Clauson Sounds like you have some interesting requirements. Could you please ping me offline? Email Blog {at} BenHall .me .uk I would be interested to hear what your have to say and your scenarios in more depth. Cheers Ben On Mon, Jul 21, 2008 at 1:11 PM, Aaron Clauson wrote: > Ben Hall wrote: >> As for Silverlight, yes this would be cool (more cool than WPF!) >> however Moonlight's current "support for Silverlight 2.0 is in >> pre-Alpha stage". I would love to have a Silverlight UI on top of this >> (more than WPF) but I wanted to get WinForms app out the way (which I >> think is more useful if not as cool). Maybe when the WinForms app has >> the core functionality in place, I can look at doing something similar >> for Silverlight. > > Hi Ben, > > I'd put in a big vote for an IronRuby editor done up in Silverlight. > I've got a bunch of 500 or so users who are struggling to get a grasp on > IronRuby who would thank you as well, we get lots of Ruby usage > questions on our forums > http://www.mysipswitch.com/forum/viewforum.php?f=6 (that link is also > for a free, open source project so no plugging is involved here). Any > kind of syntax highlighting and/or test execution would definitely help > people who aren't that used to scripting let alone Ruby to get a handle > on things. > > Regards, > > Aaron > > -- > Posted via http://www.ruby-forum.com/. > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > From lists at ruby-forum.com Tue Jul 22 03:08:05 2008 From: lists at ruby-forum.com (Web Reservoir) Date: Tue, 22 Jul 2008 09:08:05 +0200 Subject: [Ironruby-core] IronRuby 1.0 Road Map Plans Message-ID: Hi, Mr. John Lam shall be discussing IronRuby Road Map Plans at Oscon 2008 on 24th Thursday. Information:- IronRuby: The Road to 1.0 4:30pm Thursday, 07/24/2008 Location: Portland 251 Speaker:- John Lam (Microsoft) A year ago, we shipped the first drop of the IronRuby source code at OSCON. In September, we released our project on RubyForge and began accepting contributions. A few months after that, the OSI certified the Microsoft Public License as an official open source license. Come to this talk to get an update on where we are today, and what we have to do to get to 1.0. More details and other Ruby details at this link... http://rubyworld.wordpress.com/2008/07/22/huge-list-of-rubyrails-topics-to-be-discussed-at-oscon2008/ Just for your information Thanks -- Posted via http://www.ruby-forum.com/. From ivan at flanders.co.nz Tue Jul 22 04:15:37 2008 From: ivan at flanders.co.nz (Ivan Porto Carrero) Date: Tue, 22 Jul 2008 10:15:37 +0200 Subject: [Ironruby-core] somebody up for an IM session on IronRuby and WPF? Message-ID: Hi I'd like to talk some ideas through about WPF and IronRuby is there anybody up for an IM session to quickly exchange some thoughts? Email me off list if you are :) Cheers Ivan -------------- next part -------------- An HTML attachment was scrubbed... URL: From Tomas.Matousek at microsoft.com Tue Jul 22 14:36:57 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Tue, 22 Jul 2008 11:36:57 -0700 Subject: [Ironruby-core] Code Review: License2 Message-ID: tfpt review "/shelveset:License2;REDMOND\tomat" Modifies GPPG to take a header of parser.y file and copy it to the generated parser file. Adds license statements to generators. Tomas -------------- next part -------------- A non-text attachment was scrubbed... Name: License2.diff Type: application/octet-stream Size: 127032 bytes Desc: License2.diff URL: From olegtk at microsoft.com Wed Jul 23 04:35:29 2008 From: olegtk at microsoft.com (Oleg Tkachenko) Date: Wed, 23 Jul 2008 01:35:29 -0700 Subject: [Ironruby-core] Code Review: YamlFixes Message-ID: tfpt review "/shelveset:YamlFixes;REDMOND\olegtk" Comment : Removes obsolete YamlTest project from the solution. Fixes DateTime serialization and loading of timestamps. Fixes nil serialization. Implements YAML::add_domain_type() - user controlled parsing of domain types. Fixes loading binary data in Ruby context. Implements loading dates - instantiates Date class (from date.rb curently) instead of DateTime. Adds "require 'date'" into yaml.rb to be compatible with MRI (it makes loading yaml much slower currently - need to implement Date class). Fixes bug with serializing a hash with a nil as a key. Implements loading abbreviated types (looks like obsoleted yaml 1.0 only feature, but MRI tests it. http://yaml.org/spec/history/2002-10-31.html#preview-family). Splits SafeConstructor.ConstructYamlTimestamp into two methods so RubyConstructor can override Date construction. Fixes Time.at() - the argument must be a double type. Fixes Time.local() and Time.utc() - usec argument is actually microseconds, not milliseconds. -- Oleg -------------- next part -------------- A non-text attachment was scrubbed... Name: YamlFixes.diff Type: application/octet-stream Size: 31168 bytes Desc: YamlFixes.diff URL: From curth at microsoft.com Wed Jul 23 12:11:10 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Wed, 23 Jul 2008 09:11:10 -0700 Subject: [Ironruby-core] Code Review: YamlFixes In-Reply-To: References: Message-ID: Microseconds are being truncated to milliseconds instead being rounded. The least painful way to fix this is to .AddMilliseconds(1) to the result if usec >= 500. (I might be okay with the truncation if it's acknowledged in a comment.) There's a typo in a comment added to BaseConstructor.cs. I'm not qualified to comment on the more YAML-specific stuff. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Oleg Tkachenko Sent: Wednesday, July 23, 2008 1:35 AM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: YamlFixes tfpt review "/shelveset:YamlFixes;REDMOND\olegtk" Comment : Removes obsolete YamlTest project from the solution. Fixes DateTime serialization and loading of timestamps. Fixes nil serialization. Implements YAML::add_domain_type() - user controlled parsing of domain types. Fixes loading binary data in Ruby context. Implements loading dates - instantiates Date class (from date.rb curently) instead of DateTime. Adds "require 'date'" into yaml.rb to be compatible with MRI (it makes loading yaml much slower currently - need to implement Date class). Fixes bug with serializing a hash with a nil as a key. Implements loading abbreviated types (looks like obsoleted yaml 1.0 only feature, but MRI tests it. http://yaml.org/spec/history/2002-10-31.html#preview-family). Splits SafeConstructor.ConstructYamlTimestamp into two methods so RubyConstructor can override Date construction. Fixes Time.at() - the argument must be a double type. Fixes Time.local() and Time.utc() - usec argument is actually microseconds, not milliseconds. -- Oleg From Tomas.Matousek at microsoft.com Wed Jul 23 12:30:59 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Wed, 23 Jul 2008 09:30:59 -0700 Subject: [Ironruby-core] Code Review: YamlFixes In-Reply-To: References: Message-ID: 1) add_domain_type should probably be split into 2 overloads - one for Regex and other for object converted to a string since the implementation is different. 2) Do not use dynamic sites for block invocation (RubyConstructor.cs): return _Block.Invoke(ctor.GetContext(), _block, MutableString.Create(tag), ctor.ConstructPrimitive(node)); Use _Block.Yield. 3) Representer.cs: why do we compare to two different "Null" objects? if (key == BaseConstructor.NullObjectKey) { key = null; } else { key = BaseSymbolDictionary.ObjToNull(key); } Other than that looks good. Tomas -----Original Message----- From: Oleg Tkachenko Sent: Wednesday, July 23, 2008 1:35 AM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: Code Review: YamlFixes tfpt review "/shelveset:YamlFixes;REDMOND\olegtk" Comment : Removes obsolete YamlTest project from the solution. Fixes DateTime serialization and loading of timestamps. Fixes nil serialization. Implements YAML::add_domain_type() - user controlled parsing of domain types. Fixes loading binary data in Ruby context. Implements loading dates - instantiates Date class (from date.rb curently) instead of DateTime. Adds "require 'date'" into yaml.rb to be compatible with MRI (it makes loading yaml much slower currently - need to implement Date class). Fixes bug with serializing a hash with a nil as a key. Implements loading abbreviated types (looks like obsoleted yaml 1.0 only feature, but MRI tests it. http://yaml.org/spec/history/2002-10-31.html#preview-family). Splits SafeConstructor.ConstructYamlTimestamp into two methods so RubyConstructor can override Date construction. Fixes Time.at() - the argument must be a double type. Fixes Time.local() and Time.utc() - usec argument is actually microseconds, not milliseconds. -- Oleg From olegtk at microsoft.com Wed Jul 23 18:51:59 2008 From: olegtk at microsoft.com (Oleg Tkachenko) Date: Wed, 23 Jul 2008 15:51:59 -0700 Subject: [Ironruby-core] Code Review: YamlFixes In-Reply-To: References: Message-ID: Thanks for comments, guys! 3) some old rudimentary stuff, will clean it out. -- Oleg -----Original Message----- From: Tomas Matousek Sent: Wednesday, July 23, 2008 9:31 AM To: Oleg Tkachenko; IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: RE: Code Review: YamlFixes 1) add_domain_type should probably be split into 2 overloads - one for Regex and other for object converted to a string since the implementation is different. 2) Do not use dynamic sites for block invocation (RubyConstructor.cs): return _Block.Invoke(ctor.GetContext(), _block, MutableString.Create(tag), ctor.ConstructPrimitive(node)); Use _Block.Yield. 3) Representer.cs: why do we compare to two different "Null" objects? if (key == BaseConstructor.NullObjectKey) { key = null; } else { key = BaseSymbolDictionary.ObjToNull(key); } Other than that looks good. Tomas -----Original Message----- From: Oleg Tkachenko Sent: Wednesday, July 23, 2008 1:35 AM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: Code Review: YamlFixes tfpt review "/shelveset:YamlFixes;REDMOND\olegtk" Comment : Removes obsolete YamlTest project from the solution. Fixes DateTime serialization and loading of timestamps. Fixes nil serialization. Implements YAML::add_domain_type() - user controlled parsing of domain types. Fixes loading binary data in Ruby context. Implements loading dates - instantiates Date class (from date.rb curently) instead of DateTime. Adds "require 'date'" into yaml.rb to be compatible with MRI (it makes loading yaml much slower currently - need to implement Date class). Fixes bug with serializing a hash with a nil as a key. Implements loading abbreviated types (looks like obsoleted yaml 1.0 only feature, but MRI tests it. http://yaml.org/spec/history/2002-10-31.html#preview-family). Splits SafeConstructor.ConstructYamlTimestamp into two methods so RubyConstructor can override Date construction. Fixes Time.at() - the argument must be a double type. Fixes Time.local() and Time.utc() - usec argument is actually microseconds, not milliseconds. -- Oleg From lists at ruby-forum.com Thu Jul 24 00:04:10 2008 From: lists at ruby-forum.com (Web Reservoir) Date: Thu, 24 Jul 2008 06:04:10 +0200 Subject: [Ironruby-core] Good Efforts by CodeThinked.com for IronRuby In-Reply-To: <813003.44368.qm@web50602.mail.re2.yahoo.com> References: <307d7076337c4bc359b430d77247252c@ruby-forum.com> <813003.44368.qm@web50602.mail.re2.yahoo.com> Message-ID: Unni Nair wrote: > http://www.myfavoritemovies.us/Ironruby Your link seems to be dead. -- Posted via http://www.ruby-forum.com/. From olegtk at microsoft.com Thu Jul 24 02:49:11 2008 From: olegtk at microsoft.com (Oleg Tkachenko) Date: Wed, 23 Jul 2008 23:49:11 -0700 Subject: [Ironruby-core] Code Review: YamlFixes In-Reply-To: References: Message-ID: Well, actually we don't need to truncate or round altogether as DateTime has 0.1 microsecond precision. Here is a better way: return new DateTime(year, month, day, hour, minute, second, DateTimeKind.Utc).AddTicks(microsecond * 10); If that looks ok, I'm going to submit the shelveset. -- Oleg -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Wednesday, July 23, 2008 9:11 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: YamlFixes Microseconds are being truncated to milliseconds instead being rounded. The least painful way to fix this is to .AddMilliseconds(1) to the result if usec >= 500. (I might be okay with the truncation if it's acknowledged in a comment.) There's a typo in a comment added to BaseConstructor.cs. I'm not qualified to comment on the more YAML-specific stuff. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Oleg Tkachenko Sent: Wednesday, July 23, 2008 1:35 AM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: YamlFixes tfpt review "/shelveset:YamlFixes;REDMOND\olegtk" Comment : Removes obsolete YamlTest project from the solution. Fixes DateTime serialization and loading of timestamps. Fixes nil serialization. Implements YAML::add_domain_type() - user controlled parsing of domain types. Fixes loading binary data in Ruby context. Implements loading dates - instantiates Date class (from date.rb curently) instead of DateTime. Adds "require 'date'" into yaml.rb to be compatible with MRI (it makes loading yaml much slower currently - need to implement Date class). Fixes bug with serializing a hash with a nil as a key. Implements loading abbreviated types (looks like obsoleted yaml 1.0 only feature, but MRI tests it. http://yaml.org/spec/history/2002-10-31.html#preview-family). Splits SafeConstructor.ConstructYamlTimestamp into two methods so RubyConstructor can override Date construction. Fixes Time.at() - the argument must be a double type. Fixes Time.local() and Time.utc() - usec argument is actually microseconds, not milliseconds. -- Oleg _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From curth at microsoft.com Thu Jul 24 02:55:26 2008 From: curth at microsoft.com (Curt Hagenlocher) Date: Wed, 23 Jul 2008 23:55:26 -0700 Subject: [Ironruby-core] Code Review: YamlFixes In-Reply-To: References: Message-ID: Ah, right. Looks good. AddTicks is a reasonably cheap operations so there's probably no compelling reason to special-case (microsecond == 0). -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Oleg Tkachenko Sent: Wednesday, July 23, 2008 11:49 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: YamlFixes Well, actually we don't need to truncate or round altogether as DateTime has 0.1 microsecond precision. Here is a better way: return new DateTime(year, month, day, hour, minute, second, DateTimeKind.Utc).AddTicks(microsecond * 10); If that looks ok, I'm going to submit the shelveset. -- Oleg -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Curt Hagenlocher Sent: Wednesday, July 23, 2008 9:11 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] Code Review: YamlFixes Microseconds are being truncated to milliseconds instead being rounded. The least painful way to fix this is to .AddMilliseconds(1) to the result if usec >= 500. (I might be okay with the truncation if it's acknowledged in a comment.) There's a typo in a comment added to BaseConstructor.cs. I'm not qualified to comment on the more YAML-specific stuff. -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Oleg Tkachenko Sent: Wednesday, July 23, 2008 1:35 AM To: IronRuby External Code Reviewers Cc: ironruby-core at rubyforge.org Subject: [Ironruby-core] Code Review: YamlFixes tfpt review "/shelveset:YamlFixes;REDMOND\olegtk" Comment : Removes obsolete YamlTest project from the solution. Fixes DateTime serialization and loading of timestamps. Fixes nil serialization. Implements YAML::add_domain_type() - user controlled parsing of domain types. Fixes loading binary data in Ruby context. Implements loading dates - instantiates Date class (from date.rb curently) instead of DateTime. Adds "require 'date'" into yaml.rb to be compatible with MRI (it makes loading yaml much slower currently - need to implement Date class). Fixes bug with serializing a hash with a nil as a key. Implements loading abbreviated types (looks like obsoleted yaml 1.0 only feature, but MRI tests it. http://yaml.org/spec/history/2002-10-31.html#preview-family). Splits SafeConstructor.ConstructYamlTimestamp into two methods so RubyConstructor can override Date construction. Fixes Time.at() - the argument must be a double type. Fixes Time.local() and Time.utc() - usec argument is actually microseconds, not milliseconds. -- Oleg _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From ivan at flanders.co.nz Thu Jul 24 15:15:39 2008 From: ivan at flanders.co.nz (Ivan Porto Carrero) Date: Thu, 24 Jul 2008 21:15:39 +0200 Subject: [Ironruby-core] inspect in console Message-ID: Hi, Did the -i switch break on the ir command ? I create a file test.rb with the following content: myvar = "hello world" puts myvar and then type in ir -i test.rb That yields: (loosely translated I can't find the file .....\-i ) mscorlib:0:in `WinIOError': Kan bestand C:\projects\lumpr\src\Sylvester.DesktopEdition\Sylvester.IronRuby\-i niet vinden . (System::IO::FileNotFoundException) from mscorlib:0:in `Init' from mscorlib:0:in `.ctor' from c:\tools\IronRuby\src\microsoft.scripting.core\PlatformAdaptationLayer.cs:187:in `OpenInputFileStream' from c:\tools\IronRuby\src\microsoft.scripting.core\SourceFileContentProvider.cs:62:in `GetStream' from c:\tools\IronRuby\src\microsoft.scripting.core\SourceFileContentProvider.cs:45:in `GetStream' from c:\tools\IronRuby\src\microsoft.scripting.core\runtime\languageboundtextcontentprovider.cs:38:in `GetReader ' from c:\tools\IronRuby\src\microsoft.scripting.core\sourceunit.cs:102:in `GetReader' from c:\tools\IronRuby\src\microsoft.scripting.core\sourceunit.cs:212:in `Compile' from c:\tools\IronRuby\src\microsoft.scripting.core\sourceunit.cs:246:in `ExecuteProgram' -------------- next part -------------- An HTML attachment was scrubbed... URL: From ksunair at yahoo.com Thu Jul 24 16:15:12 2008 From: ksunair at yahoo.com (Unnikrishnan Nair) Date: Thu, 24 Jul 2008 13:15:12 -0700 (PDT) Subject: [Ironruby-core] Good Efforts by CodeThinked.com for IronRuby In-Reply-To: Message-ID: <359160.56698.qm@web50607.mail.re2.yahoo.com> Try this one; http://myfavoritemovies.us/ironruby/ --- On Wed, 7/23/08, Web Reservoir wrote: From: Web Reservoir Subject: Re: [Ironruby-core] Good Efforts by CodeThinked.com for IronRuby To: ironruby-core at rubyforge.org Date: Wednesday, July 23, 2008, 10:04 PM Unni Nair wrote: > http://www.myfavoritemovies.us/Ironruby Your link seems to be dead. -- Posted via http://www.ruby-forum.com/. _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core -------------- next part -------------- An HTML attachment was scrubbed... URL: From Tomas.Matousek at microsoft.com Thu Jul 24 18:47:35 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Thu, 24 Jul 2008 15:47:35 -0700 Subject: [Ironruby-core] inspect in console In-Reply-To: References: Message-ID: -i was Python specific option, MRI uses this option for a different feature (that we don?t support yet). Tomas From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Ivan Porto Carrero Sent: Thursday, July 24, 2008 12:16 PM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] inspect in console Hi, Did the -i switch break on the ir command ? I create a file test.rb with the following content: myvar = "hello world" puts myvar and then type in ir -i test.rb That yields: (loosely translated I can't find the file .....\-i ) mscorlib:0:in `WinIOError': Kan bestand C:\projects\lumpr\src\Sylvester.DesktopEdition\Sylvester.IronRuby\-i niet vinden . (System::IO::FileNotFoundException) from mscorlib:0:in `Init' from mscorlib:0:in `.ctor' from c:\tools\IronRuby\src\microsoft.scripting.core\PlatformAdaptationLayer.cs:187:in `OpenInputFileStream' from c:\tools\IronRuby\src\microsoft.scripting.core\SourceFileContentProvider.cs:62:in `GetStream' from c:\tools\IronRuby\src\microsoft.scripting.core\SourceFileContentProvider.cs:45:in `GetStream' from c:\tools\IronRuby\src\microsoft.scripting.core\runtime\languageboundtextcontentprovider.cs:38:in `GetReader ' from c:\tools\IronRuby\src\microsoft.scripting.core\sourceunit.cs:102:in `GetReader' from c:\tools\IronRuby\src\microsoft.scripting.core\sourceunit.cs:212:in `Compile' from c:\tools\IronRuby\src\microsoft.scripting.core\sourceunit.cs:246:in `ExecuteProgram' -------------- next part -------------- An HTML attachment was scrubbed... URL: From stefan.dobrev at gmail.com Fri Jul 25 02:38:50 2008 From: stefan.dobrev at gmail.com (Stefan Dobrev) Date: Fri, 25 Jul 2008 09:38:50 +0300 Subject: [Ironruby-core] DLR IDE Message-ID: <928de89c0807242338s59aadf34uf29936c844d0c30f@mail.gmail.com> Hi all, Wouldn't be nice if we all create a DLR IDE. I have being thinking for this for quite some times recently. As usually happens someone else also has been thinking of this and actually blogged about it here. Saying: --- So what's my point? I think all of these projects are great and kudos to the people that built them. It takes a lot of time and effort above and beyond just regular work hours. I have been there myself, my hats off to you folks! But, there are 8 versions of the interactive console and a few versions of a basic code editor. I know it may be a dream, but it would be great to collaborate with these people and write out a simple set of requirements for what a great DLR console and code editor would be. And then as a virtual team, implement it. After all, to a large degree, it will be how well supported the language is from a tools perspective that will really determine the rate of adoption. And right now, the tools (or IDE) experience for Dynamic Languages on .NET is severely lacking to the point of having several people independently developing their own tooling. In this post I only pointed out a handful of these tools and I know there are others, but I was really targeting web-based IDE's. Maybe that is an opportunity? Or is it a pipe dream? --- So let's come together and build a difference. I have made the first step and created a project on codeplex - http://codeplex.com/dlride. I have also written some basic requirements for the project in the wiki. Let we all collaborate on them. For everyone who want to take a part, please feel free to contact me, so I can give you permissions for the project on codeplex. Regards, Stefan -------------- next part -------------- An HTML attachment was scrubbed... URL: From lists at ruby-forum.com Fri Jul 25 03:27:57 2008 From: lists at ruby-forum.com (Shin guey Wong) Date: Fri, 25 Jul 2008 09:27:57 +0200 Subject: [Ironruby-core] Can we compile a ruby src code into .exe and run it? Message-ID: <240dcf37467ab48e09ee258d98365d3d@ruby-forum.com> Hi, For current ironruby, can we use it to compile a ruby source code into exe and release it to other person to run it? If it is possible, the other person need to install .net framework + anything else? Thanks in advance. sgwong -- Posted via http://www.ruby-forum.com/. From m.david at xmlhacker.com Fri Jul 25 03:34:32 2008 From: m.david at xmlhacker.com (M. David Peterson) Date: Fri, 25 Jul 2008 01:34:32 -0600 Subject: [Ironruby-core] Can we compile a ruby src code into .exe and run it? In-Reply-To: <240dcf37467ab48e09ee258d98365d3d@ruby-forum.com> References: <240dcf37467ab48e09ee258d98365d3d@ruby-forum.com> Message-ID: On Fri, 25 Jul 2008 01:27:57 -0600, Shin guey Wong wrote: > For current ironruby, can we use it to compile a ruby source code into > exe and release it to other person to run it? No. You can with Ruby.NET, but as far as I understand things correctly, support from IronRuby, if it happens, will come after the 1.0 release. -- /M:D M. David Peterson Co-Founder & Chief Architect, 3rd&Urban, LLC Email: m.david at 3rdandUrban.com | m.david at amp.fm Mobile: (206) 999-0588 http://3rdandUrban.com | http://amp.fm | http://www.oreillynet.com/pub/au/2354 | http://news.oreilly.com/m-david-peterson/ From lists at ruby-forum.com Fri Jul 25 04:29:05 2008 From: lists at ruby-forum.com (Web Reservoir) Date: Fri, 25 Jul 2008 10:29:05 +0200 Subject: [Ironruby-core] Quick IronRuby Links to Book Mark for Ever at RubyWorld. Message-ID: <8506ceb071aa3c632623efdf20919714@ruby-forum.com> Hi, I found this nice blog on RubyWorld and would like to share here. They cover quick links for IronRuby. http://rubyworld.wordpress.com/2008/07/25/quick-ironruby-links-to-bookmark-for-ever/ Pl. share more links, if you find any good link missing from the list, its a request by th e blogger at RubyWorld. Thanks -- Posted via http://www.ruby-forum.com/. From ivan at flanders.co.nz Fri Jul 25 05:32:55 2008 From: ivan at flanders.co.nz (Ivan Porto Carrero) Date: Fri, 25 Jul 2008 11:32:55 +0200 Subject: [Ironruby-core] binary release Message-ID: Hi I've noticed that the -I switch for passing in search paths has also gone away. In the binary release John talks about a batteries included approach. Does that mean all we have to do is extract it to a folder, add the bin folder to the PATH variable and off we go or are there still some switches to pass in when we want to use rubygems etc? Thanks Ivan -------------- next part -------------- An HTML attachment was scrubbed... URL: From ben2004uk at googlemail.com Fri Jul 25 05:33:46 2008 From: ben2004uk at googlemail.com (Ben Hall) Date: Fri, 25 Jul 2008 10:33:46 +0100 Subject: [Ironruby-core] DLR IDE In-Reply-To: <928de89c0807242338s59aadf34uf29936c844d0c30f@mail.gmail.com> References: <928de89c0807242338s59aadf34uf29936c844d0c30f@mail.gmail.com> Message-ID: Hi Stefan, An interesting approach - you want people to collaborate so you start a new project? When writing the editors, people have lots of different reasons for doing so - not everyone wants to create an IDE, some people just want to create a cool sample to learn Silverlight \ DLR which they then release. Some people don't want to collaborate with others, that is not their aim and you can't blame people for that. Personally, since releasing IronEditor I have had loads of people sending me sample code, feature requests, and some other really really interesting stuff. People are collaborating behind the scenes to improve the application, personally I have a list of other people I want to contact about improving the application and working together. Same with Michael Foord's TryPython, lots of people provided feedback after he went live. Collaboration isn't just code. Good luck with your IDE, but open source projects are not as easy as they sound, I look forward to seeing how it goes. I tried to make this point on your post, however your spam filter wasn't displaying any images - makes it difficult to make a comment then :) On a side note, if people want to help with IronEditor - contact me offline. Thanks for everyone who has already. Thanks Ben Blog.BenHall.me.uk On Fri, Jul 25, 2008 at 7:38 AM, Stefan Dobrev wrote: > Hi all, > > Wouldn't be nice if we all create a DLR IDE. I have being thinking for this > for quite some times recently. As usually happens someone else also has been > thinking of this and actually blogged about it here. Saying: > --- > > So what's my point? I think all of these projects are great and kudos to > the people that built them. It takes a lot of time and effort above and > beyond just regular work hours. I have been there myself, my hats off to you > folks! But, there are 8 versions of the interactive console and a few > versions of a basic code editor. I know it may be a dream, but it would be > great to collaborate with these people and write out a simple set of > requirements for what a great DLR console and code editor would be. And > then as a virtual team, implement it. > > > > After all, to a large degree, it will be how well supported the language is > from a tools perspective that will really determine the rate of adoption. > And right now, the tools (or IDE) experience for Dynamic Languages on .NET > is severely lacking to the point of having several people independently > developing their own tooling. In this post I only pointed out a handful of > these tools and I know there are others, but I was really targeting > web-based IDE's. Maybe that is an opportunity? Or is it a pipe dream? > > --- > > So let's come together and build a difference. > > I have made the first step and created a project on codeplex - > http://codeplex.com/dlride. I have also written some basic requirements for > the project in the wiki. Let we all collaborate on them. > > For everyone who want to take a part, please feel free to contact me, so I > can give you permissions for the project on codeplex. > > Regards, > Stefan > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > From ben2004uk at googlemail.com Fri Jul 25 08:32:18 2008 From: ben2004uk at googlemail.com (Ben Hall) Date: Fri, 25 Jul 2008 13:32:18 +0100 Subject: [Ironruby-core] [IronPython] DLR IDE In-Reply-To: <1986630732.2885041216987012204.JavaMail.root@ganesh.brynmawr.edu> References: <1168836735.2884381216986070126.JavaMail.root@ganesh.brynmawr.edu> <1986630732.2885041216987012204.JavaMail.root@ganesh.brynmawr.edu> Message-ID: Hi Douglas, Integration into Microsoft Robotics Developer Studio for IronPython is something which is very interesting. It sounds like you guys are doing some interesting stuff, maybe a bit out of the scope which I had planned for IronEditor. I take it as your using this as an educational exercise, you will want to write everything from scatch? Maybe we should share ideas about how you go about implementing syntax highlighting etc What do you think about using IronEditor (or similar) as the base and for me to allow some more extendability to allow you to customise the IDE Part (Editor + Output)? It would be interesting to chat with you about this offline. Ben On Fri, Jul 25, 2008 at 12:56 PM, Douglas Blank wrote: > Stefan, Ben, and all, > > It looks like there may be a few choices to choose from! We are also working on an IDE, but it is part of a larger project. Our goal is to create an educational framework for teaching ideas in computer science. It is also open source, but no doubt differs from your initial ideas---but we could all work together as well. Our main goals: > > - crossplatform (Mac, Windows, Linux) > - multi-programming language (DLR languages: initially Python, Ruby, and Scheme) > - i18n support > - IDE with color syntax, and emacs-like extensibility/use > - designed for ease-of-use, but powerful > - easy to install and run > - collect of libraries that are fast and easy > > Our IDE is part of the Pyjama Project. The IDE you can think of it as IDLE for .NET/Mono. One of the specific goals is to allow easy connections to Microsoft Robotics Developer Studio, but it has a much larger scope, too. We are putting together some libraries that will be very useful in interactive environments. Of course, these libraries will be able to be used by any .NET/Mono system. These include: > > - 2d graphics (see Zelle's Python graphics.py and Guzdial's Media Computation) > - 3d graphics and games (like pygame) > - artificial intelligence (see http://pyrorobotics.org) > > We'll be announcing some more details about the Pyjama Project, but would love to collaborate with others in whatever ways possible. Perhaps initially a single mailing list to discuss the details would be useful? > > Count us in! > > -Doug > > ----- "Ben Hall" wrote: > >> Hi Stefan, >> >> An interesting approach - you want people to collaborate so you start >> a new project? >> >> When writing the editors, people have lots of different reasons for >> doing so - not everyone wants to create an IDE, some people just want >> to create a cool sample to learn Silverlight \ DLR which they then >> release. Some people don't want to collaborate with others, that is >> not their aim and you can't blame people for that. >> >> Personally, since releasing IronEditor I have had loads of people >> sending me sample code, feature requests, and some other really >> really >> interesting stuff. People are collaborating behind the scenes to >> improve the application, personally I have a list of other people I >> want to contact about improving the application and working together. >> Same with Michael Foord's TryPython, lots of people provided feedback >> after he went live. Collaboration isn't just code. >> >> Good luck with your IDE, but open source projects are not as easy as >> they sound, I look forward to seeing how it goes. I tried to make >> this point on your post, however your spam filter wasn't displaying >> any images - makes it difficult to make a comment then :) >> >> On a side note, if people want to help with IronEditor - contact me >> offline. Thanks for everyone who has already. >> >> Thanks >> >> Ben >> Blog.BenHall.me.uk >> >> >> On Fri, Jul 25, 2008 at 7:38 AM, Stefan Dobrev >> wrote: >> > Hi all, >> > >> > Wouldn't be nice if we all create a DLR IDE. I have being thinking >> for this >> > for quite some times recently. As usually happens someone else also >> has been >> > thinking of this and actually blogged about it here. Saying: >> > --- >> > >> > So what's my point? I think all of these projects are great and >> kudos to >> > the people that built them. It takes a lot of time and effort above >> and >> > beyond just regular work hours. I have been there myself, my hats >> off to you >> > folks! But, there are 8 versions of the interactive console and a >> few >> > versions of a basic code editor. I know it may be a dream, but it >> would be >> > great to collaborate with these people and write out a simple set >> of >> > requirements for what a great DLR console and code editor would be. >> And >> > then as a virtual team, implement it. >> > >> > >> > >> > After all, to a large degree, it will be how well supported the >> language is >> > from a tools perspective that will really determine the rate of >> adoption. >> > And right now, the tools (or IDE) experience for Dynamic Languages >> on .NET >> > is severely lacking to the point of having several people >> independently >> > developing their own tooling. In this post I only pointed out a >> handful of >> > these tools and I know there are others, but I was really targeting >> > web-based IDE's. Maybe that is an opportunity? Or is it a pipe >> dream? >> > >> > --- >> > >> > So let's come together and build a difference. >> > >> > I have made the first step and created a project on codeplex - >> > http://codeplex.com/dlride. I have also written some basic >> requirements for >> > the project in the wiki. Let we all collaborate on them. >> > >> > For everyone who want to take a part, please feel free to contact >> me, so I >> > can give you permissions for the project on codeplex. >> > >> > Regards, >> > Stefan >> > >> > _______________________________________________ >> > Ironruby-core mailing list >> > Ironruby-core at rubyforge.org >> > http://rubyforge.org/mailman/listinfo/ironruby-core >> > >> > >> _______________________________________________ >> Users mailing list >> Users at lists.ironpython.com >> http://lists.ironpython.com/listinfo.cgi/users-ironpython.com > _______________________________________________ > Users mailing list > Users at lists.ironpython.com > http://lists.ironpython.com/listinfo.cgi/users-ironpython.com > From jflam at microsoft.com Fri Jul 25 12:16:23 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Fri, 25 Jul 2008 09:16:23 -0700 Subject: [Ironruby-core] binary release In-Reply-To: References: Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD4357ABA@NA-EXMSG-C115.redmond.corp.microsoft.com> The ?I switch is still there. We?re still delegating to ir.cmd to run IronRuby today. We?ll fix some of these problems now that we have a fixed relative path layout to our libs directory. We?ll be updating this drop to reflect this soon (and to do a few other things that we missed this time around like enabling RubyGems support out of the box). Thanks, -John From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Ivan Porto Carrero Sent: Friday, July 25, 2008 2:33 AM To: ironruby-core at rubyforge.org Subject: [Ironruby-core] binary release Hi I've noticed that the -I switch for passing in search paths has also gone away. In the binary release John talks about a batteries included approach. Does that mean all we have to do is extract it to a folder, add the bin folder to the PATH variable and off we go or are there still some switches to pass in when we want to use rubygems etc? Thanks Ivan -------------- next part -------------- An HTML attachment was scrubbed... URL: From jgbailey at gmail.com Fri Jul 25 14:28:01 2008 From: jgbailey at gmail.com (Justin Bailey) Date: Fri, 25 Jul 2008 11:28:01 -0700 Subject: [Ironruby-core] binary release In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD4357ABA@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD4357ABA@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: On Fri, Jul 25, 2008 at 9:16 AM, John Lam (IRONRUBY) wrote: > The ?I switch is still there. We're still delegating to ir.cmd to run > IronRuby today. We'll fix some of these problems now that we have a fixed > relative path layout to our libs directory. We'll be updating this drop to > reflect this soon (and to do a few other things that we missed this time > around like enabling RubyGems support out of the box). > Will we see regular binary releases? I'm behind a particularly draconian proxy and SVN won't work. HTTP headers that SVN uses are blocked. Being able to get regular binaries or even zipped source would be so nice! Justin From jflam at microsoft.com Fri Jul 25 16:13:29 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Fri, 25 Jul 2008 13:13:29 -0700 Subject: [Ironruby-core] binary release In-Reply-To: References: <372109E149E8084D8E6C7D9CFD82E0632FD4357ABA@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <38843188-D8BE-4CDE-BFA3-A92312BF8408@microsoft.com> Life will get better for you when we move to github for our sources as they support tarball gen. And yes we will do more regular binary drops. Sent from my phone On Jul 25, 2008, at 11:30 AM, "Justin Bailey" wrote: > On Fri, Jul 25, 2008 at 9:16 AM, John Lam (IRONRUBY) > wrote: >> The ?I switch is still there. We're still delegating to ir.cmd to >> run >> IronRuby today. We'll fix some of these problems now that we have a >> fixed >> relative path layout to our libs directory. We'll be updating this >> drop to >> reflect this soon (and to do a few other things that we missed this >> time >> around like enabling RubyGems support out of the box). >> > > Will we see regular binary releases? I'm behind a particularly > draconian proxy and SVN won't work. HTTP headers that SVN uses are > blocked. Being able to get regular binaries or even zipped source > would be so nice! > > Justin > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > From jdeville at microsoft.com Fri Jul 25 16:57:54 2008 From: jdeville at microsoft.com (Jim Deville) Date: Fri, 25 Jul 2008 13:57:54 -0700 Subject: [Ironruby-core] binary release In-Reply-To: <38843188-D8BE-4CDE-BFA3-A92312BF8408@microsoft.com> References: <372109E149E8084D8E6C7D9CFD82E0632FD4357ABA@NA-EXMSG-C115.redmond.corp.microsoft.com> , <38843188-D8BE-4CDE-BFA3-A92312BF8408@microsoft.com> Message-ID: I can look into syncing svn to my github account. If it's easy I'll do it and let you know. JD ________________________________________ From: ironruby-core-bounces at rubyforge.org [ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) [jflam at microsoft.com] Sent: Friday, July 25, 2008 1:13 PM To: ironruby-core at rubyforge.org Cc: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] binary release Life will get better for you when we move to github for our sources as they support tarball gen. And yes we will do more regular binary drops. Sent from my phone On Jul 25, 2008, at 11:30 AM, "Justin Bailey" wrote: > On Fri, Jul 25, 2008 at 9:16 AM, John Lam (IRONRUBY) > wrote: >> The ?I switch is still there. We're still delegating to ir.cmd to >> run >> IronRuby today. We'll fix some of these problems now that we have a >> fixed >> relative path layout to our libs directory. We'll be updating this >> drop to >> reflect this soon (and to do a few other things that we missed this >> time >> around like enabling RubyGems support out of the box). >> > > Will we see regular binary releases? I'm behind a particularly > draconian proxy and SVN won't work. HTTP headers that SVN uses are > blocked. Being able to get regular binaries or even zipped source > would be so nice! > > Justin > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From bacondarwin at googlemail.com Fri Jul 25 17:11:05 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Fri, 25 Jul 2008 22:11:05 +0100 Subject: [Ironruby-core] binary release In-Reply-To: References: <372109E149E8084D8E6C7D9CFD82E0632FD4357ABA@NA-EXMSG-C115.redmond.corp.microsoft.com> , <38843188-D8BE-4CDE-BFA3-A92312BF8408@microsoft.com> Message-ID: <005b01c8ee9a$f5068a80$df139f80$@com> I have a mirror of the IronRuby svn at GitHub: http://github.com/petebacondarwin/ironruby. I will manually track the svn repository on the "master" branch of this git repository for the time being. (The git-svn doesn't work too well on windows at the moment). I have been using it to track my implementation of the BigDecimal library, which you are welcome to take a look at in the "bigdecimal" branch. Checkout the cool network graph to see what I mean: http://github.com/petebacondarwin/ironruby/network. You can pull zipped up sources from github. E.g. The master branch (which is currently a copy of SVN r127) can be pulled from: http://github.com/petebacondarwin/ironruby/tarball/master. Hope that helps, Pete -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Friday,25 July 25, 2008 21:58 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] binary release I can look into syncing svn to my github account. If it's easy I'll do it and let you know. JD ________________________________________ From: ironruby-core-bounces at rubyforge.org [ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) [jflam at microsoft.com] Sent: Friday, July 25, 2008 1:13 PM To: ironruby-core at rubyforge.org Cc: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] binary release Life will get better for you when we move to github for our sources as they support tarball gen. And yes we will do more regular binary drops. Sent from my phone On Jul 25, 2008, at 11:30 AM, "Justin Bailey" wrote: > On Fri, Jul 25, 2008 at 9:16 AM, John Lam (IRONRUBY) > wrote: >> The -I switch is still there. We're still delegating to ir.cmd to >> run >> IronRuby today. We'll fix some of these problems now that we have a >> fixed >> relative path layout to our libs directory. We'll be updating this >> drop to >> reflect this soon (and to do a few other things that we missed this >> time >> around like enabling RubyGems support out of the box). >> > > Will we see regular binary releases? I'm behind a particularly > draconian proxy and SVN won't work. HTTP headers that SVN uses are > blocked. Being able to get regular binaries or even zipped source > would be so nice! > > Justin > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From jdeville at microsoft.com Fri Jul 25 17:14:29 2008 From: jdeville at microsoft.com (Jim Deville) Date: Fri, 25 Jul 2008 14:14:29 -0700 Subject: [Ironruby-core] binary release In-Reply-To: <005b01c8ee9a$f5068a80$df139f80$@com> References: <372109E149E8084D8E6C7D9CFD82E0632FD4357ABA@NA-EXMSG-C115.redmond.corp.microsoft.com> , <38843188-D8BE-4CDE-BFA3-A92312BF8408@microsoft.com> , <005b01c8ee9a$f5068a80$df139f80$@com> Message-ID: Cool, I'm going to let you handle it. BTW, I'll probably be pulling your BigDecimal stuff tonight. Thanks for the contribution. JD ________________________________________ From: ironruby-core-bounces at rubyforge.org [ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin [bacondarwin at googlemail.com] Sent: Friday, July 25, 2008 2:11 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] binary release I have a mirror of the IronRuby svn at GitHub: http://github.com/petebacondarwin/ironruby. I will manually track the svn repository on the "master" branch of this git repository for the time being. (The git-svn doesn't work too well on windows at the moment). I have been using it to track my implementation of the BigDecimal library, which you are welcome to take a look at in the "bigdecimal" branch. Checkout the cool network graph to see what I mean: http://github.com/petebacondarwin/ironruby/network. You can pull zipped up sources from github. E.g. The master branch (which is currently a copy of SVN r127) can be pulled from: http://github.com/petebacondarwin/ironruby/tarball/master. Hope that helps, Pete -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Friday,25 July 25, 2008 21:58 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] binary release I can look into syncing svn to my github account. If it's easy I'll do it and let you know. JD ________________________________________ From: ironruby-core-bounces at rubyforge.org [ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) [jflam at microsoft.com] Sent: Friday, July 25, 2008 1:13 PM To: ironruby-core at rubyforge.org Cc: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] binary release Life will get better for you when we move to github for our sources as they support tarball gen. And yes we will do more regular binary drops. Sent from my phone On Jul 25, 2008, at 11:30 AM, "Justin Bailey" wrote: > On Fri, Jul 25, 2008 at 9:16 AM, John Lam (IRONRUBY) > wrote: >> The -I switch is still there. We're still delegating to ir.cmd to >> run >> IronRuby today. We'll fix some of these problems now that we have a >> fixed >> relative path layout to our libs directory. We'll be updating this >> drop to >> reflect this soon (and to do a few other things that we missed this >> time >> around like enabling RubyGems support out of the box). >> > > Will we see regular binary releases? I'm behind a particularly > draconian proxy and SVN won't work. HTTP headers that SVN uses are > blocked. Being able to get regular binaries or even zipped source > would be so nice! > > Justin > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From jgbailey at gmail.com Fri Jul 25 17:15:53 2008 From: jgbailey at gmail.com (Justin Bailey) Date: Fri, 25 Jul 2008 14:15:53 -0700 Subject: [Ironruby-core] binary release In-Reply-To: <005b01c8ee9a$f5068a80$df139f80$@com> References: <372109E149E8084D8E6C7D9CFD82E0632FD4357ABA@NA-EXMSG-C115.redmond.corp.microsoft.com> <38843188-D8BE-4CDE-BFA3-A92312BF8408@microsoft.com> <005b01c8ee9a$f5068a80$df139f80$@com> Message-ID: That is great! Thanks! On Fri, Jul 25, 2008 at 2:11 PM, Peter Bacon Darwin wrote: > I have a mirror of the IronRuby svn at GitHub: > http://github.com/petebacondarwin/ironruby. > I will manually track the svn repository on the "master" branch of this git > repository for the time being. (The git-svn doesn't work too well on windows > at the moment). > I have been using it to track my implementation of the BigDecimal library, > which you are welcome to take a look at in the "bigdecimal" branch. > Checkout the cool network graph to see what I mean: > http://github.com/petebacondarwin/ironruby/network. > You can pull zipped up sources from github. E.g. The master branch (which > is currently a copy of SVN r127) can be pulled from: > http://github.com/petebacondarwin/ironruby/tarball/master. > Hope that helps, > Pete > > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville > Sent: Friday,25 July 25, 2008 21:58 > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] binary release > > I can look into syncing svn to my github account. If it's easy I'll do it > and let you know. > > JD > ________________________________________ > From: ironruby-core-bounces at rubyforge.org > [ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) > [jflam at microsoft.com] > Sent: Friday, July 25, 2008 1:13 PM > To: ironruby-core at rubyforge.org > Cc: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] binary release > > Life will get better for you when we move to github for our sources as > they support tarball gen. And yes we will do more regular binary drops. > > Sent from my phone > > On Jul 25, 2008, at 11:30 AM, "Justin Bailey" > wrote: > >> On Fri, Jul 25, 2008 at 9:16 AM, John Lam (IRONRUBY) >> wrote: >>> The -I switch is still there. We're still delegating to ir.cmd to >>> run >>> IronRuby today. We'll fix some of these problems now that we have a >>> fixed >>> relative path layout to our libs directory. We'll be updating this >>> drop to >>> reflect this soon (and to do a few other things that we missed this >>> time >>> around like enabling RubyGems support out of the box). >>> >> >> Will we see regular binary releases? I'm behind a particularly >> draconian proxy and SVN won't work. HTTP headers that SVN uses are >> blocked. Being able to get regular binaries or even zipped source >> would be so nice! >> >> Justin >> _______________________________________________ >> Ironruby-core mailing list >> Ironruby-core at rubyforge.org >> http://rubyforge.org/mailman/listinfo/ironruby-core >> > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > From fuzzyman at voidspace.org.uk Fri Jul 25 15:12:52 2008 From: fuzzyman at voidspace.org.uk (Michael Foord) Date: Fri, 25 Jul 2008 20:12:52 +0100 Subject: [Ironruby-core] DLR IDE In-Reply-To: References: <928de89c0807242338s59aadf34uf29936c844d0c30f@mail.gmail.com> Message-ID: <488A25B4.8090409@voidspace.org.uk> Ben Hall wrote: > Hi Stefan, > > An interesting approach - you want people to collaborate so you start > a new project? > > Hehe. That's a computing classic. "There are too many protocols, we need to invent a new one so that everyone will use that one..." Michael > When writing the editors, people have lots of different reasons for > doing so - not everyone wants to create an IDE, some people just want > to create a cool sample to learn Silverlight \ DLR which they then > release. Some people don't want to collaborate with others, that is > not their aim and you can't blame people for that. > > Personally, since releasing IronEditor I have had loads of people > sending me sample code, feature requests, and some other really really > interesting stuff. People are collaborating behind the scenes to > improve the application, personally I have a list of other people I > want to contact about improving the application and working together. > Same with Michael Foord's TryPython, lots of people provided feedback > after he went live. Collaboration isn't just code. > > Good luck with your IDE, but open source projects are not as easy as > they sound, I look forward to seeing how it goes. I tried to make > this point on your post, however your spam filter wasn't displaying > any images - makes it difficult to make a comment then :) > > On a side note, if people want to help with IronEditor - contact me > offline. Thanks for everyone who has already. > > Thanks > > Ben > Blog.BenHall.me.uk > > > On Fri, Jul 25, 2008 at 7:38 AM, Stefan Dobrev wrote: > >> Hi all, >> >> Wouldn't be nice if we all create a DLR IDE. I have being thinking for this >> for quite some times recently. As usually happens someone else also has been >> thinking of this and actually blogged about it here. Saying: >> --- >> >> So what's my point? I think all of these projects are great and kudos to >> the people that built them. It takes a lot of time and effort above and >> beyond just regular work hours. I have been there myself, my hats off to you >> folks! But, there are 8 versions of the interactive console and a few >> versions of a basic code editor. I know it may be a dream, but it would be >> great to collaborate with these people and write out a simple set of >> requirements for what a great DLR console and code editor would be. And >> then as a virtual team, implement it. >> >> >> >> After all, to a large degree, it will be how well supported the language is >> from a tools perspective that will really determine the rate of adoption. >> And right now, the tools (or IDE) experience for Dynamic Languages on .NET >> is severely lacking to the point of having several people independently >> developing their own tooling. In this post I only pointed out a handful of >> these tools and I know there are others, but I was really targeting >> web-based IDE's. Maybe that is an opportunity? Or is it a pipe dream? >> >> --- >> >> So let's come together and build a difference. >> >> I have made the first step and created a project on codeplex - >> http://codeplex.com/dlride. I have also written some basic requirements for >> the project in the wiki. Let we all collaborate on them. >> >> For everyone who want to take a part, please feel free to contact me, so I >> can give you permissions for the project on codeplex. >> >> Regards, >> Stefan >> >> _______________________________________________ >> Ironruby-core mailing list >> Ironruby-core at rubyforge.org >> http://rubyforge.org/mailman/listinfo/ironruby-core >> >> >> > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > -- http://www.ironpythoninaction.com/ http://www.voidspace.org.uk/ http://www.trypython.org/ http://www.ironpython.info/ http://www.theotherdelia.co.uk/ http://www.resolverhacks.net/ From Tomas.Matousek at microsoft.com Fri Jul 25 23:31:34 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Fri, 25 Jul 2008 20:31:34 -0700 Subject: [Ironruby-core] Code Review: mo6 Message-ID: tfpt review "/shelveset:mo6;REDMOND\tomat" Preparation for meta-objects support in IronRuby. Replaces OldInvokeMemberAction by Ruby specific RubyCallAction. Uses this new action for all Ruby call sites. ArgumentKind.Block is not needed anymore in DLR, so it could be removed (this shelveset doesn't touch DLR though). Refactors RubyBinder and rule generators. Adds meta-object classes for some Ruby classes (they are not used yet). Tomas -------------- next part -------------- A non-text attachment was scrubbed... Name: mo6.diff Type: application/octet-stream Size: 184634 bytes Desc: mo6.diff URL: From bacondarwin at googlemail.com Sat Jul 26 01:39:54 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Sat, 26 Jul 2008 06:39:54 +0100 Subject: [Ironruby-core] binary release In-Reply-To: References: <372109E149E8084D8E6C7D9CFD82E0632FD4357ABA@NA-EXMSG-C115.redmond.corp.microsoft.com> , <38843188-D8BE-4CDE-BFA3-A92312BF8408@microsoft.com> , <005b01c8ee9a$f5068a80$df139f80$@com> Message-ID: <000c01c8eee2$098cef70$1ca6ce50$@com> Just be aware that this BigDecimal is not quite there on division, remainder/modulos and sqrt at the moment. Pete -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Friday,25 July 25, 2008 22:14 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] binary release Cool, I'm going to let you handle it. BTW, I'll probably be pulling your BigDecimal stuff tonight. Thanks for the contribution. JD ________________________________________ From: ironruby-core-bounces at rubyforge.org [ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin [bacondarwin at googlemail.com] Sent: Friday, July 25, 2008 2:11 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] binary release I have a mirror of the IronRuby svn at GitHub: http://github.com/petebacondarwin/ironruby. I will manually track the svn repository on the "master" branch of this git repository for the time being. (The git-svn doesn't work too well on windows at the moment). I have been using it to track my implementation of the BigDecimal library, which you are welcome to take a look at in the "bigdecimal" branch. Checkout the cool network graph to see what I mean: http://github.com/petebacondarwin/ironruby/network. You can pull zipped up sources from github. E.g. The master branch (which is currently a copy of SVN r127) can be pulled from: http://github.com/petebacondarwin/ironruby/tarball/master. Hope that helps, Pete -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Friday,25 July 25, 2008 21:58 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] binary release I can look into syncing svn to my github account. If it's easy I'll do it and let you know. JD ________________________________________ From: ironruby-core-bounces at rubyforge.org [ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) [jflam at microsoft.com] Sent: Friday, July 25, 2008 1:13 PM To: ironruby-core at rubyforge.org Cc: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] binary release Life will get better for you when we move to github for our sources as they support tarball gen. And yes we will do more regular binary drops. Sent from my phone On Jul 25, 2008, at 11:30 AM, "Justin Bailey" wrote: > On Fri, Jul 25, 2008 at 9:16 AM, John Lam (IRONRUBY) > wrote: >> The -I switch is still there. We're still delegating to ir.cmd to >> run >> IronRuby today. We'll fix some of these problems now that we have a >> fixed >> relative path layout to our libs directory. We'll be updating this >> drop to >> reflect this soon (and to do a few other things that we missed this >> time >> around like enabling RubyGems support out of the box). >> > > Will we see regular binary releases? I'm behind a particularly > draconian proxy and SVN won't work. HTTP headers that SVN uses are > blocked. Being able to get regular binaries or even zipped source > would be so nice! > > Justin > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From lists at ruby-forum.com Sat Jul 26 08:40:07 2008 From: lists at ruby-forum.com (Web Reservoir) Date: Sat, 26 Jul 2008 14:40:07 +0200 Subject: [Ironruby-core] A Very huge list of Ruby, Rails, IronRuby Books Message-ID: A Very huge list of Ruby, Rails, IronRuby Books arriving in 2008/2009. Have a look at this blog by RubyWorld. Its worth to bookmark this. http://rubyworld.wordpress.com/2008/07/26/wow-very-huge-list-of-upcoming-rubyrails-books-in-2008/ Hope many would appreciate this. Thanks -- Posted via http://www.ruby-forum.com/. From lists at ruby-forum.com Sat Jul 26 10:41:32 2008 From: lists at ruby-forum.com (N. F.) Date: Sat, 26 Jul 2008 16:41:32 +0200 Subject: [Ironruby-core] IronRuby and MarkLogic In-Reply-To: References: <372109E149E8084D8E6C7D9CFD82E06329DB623698@NA-EXMSG-C115.redmond.corp.microsoft.com> <75f192a57ebd31511ac864d1101e95aa@ruby-forum.com> <918705E903F4714CB713D89AB5F1857D738E8C6137@NA-EXMSG-C116.redmond.corp.microsoft.com> <13f5089658d136717f882cf693cf5a89@ruby-forum.com> <918705E903F4714CB713D89AB5F1857D738E8C6586@NA-EXMSG-C116.redmond.corp.microsoft.com> <83fbe5c2780512469056ee889bd5de5f@ruby-forum.com> <918705E903F4714CB713D89AB5F1857D73A79302E2@NA-EXMSG-C116.redmond.corp.microsoft.com> Message-ID: <69872013390c5d11594627ea8654f941@ruby-forum.com> Hi, This problem reported here and fixed by John Messerly is back in the IronRuby Alpha release at http://rubyforge.org/projects/ironruby. So, the patch by John never made it "officially"? Thanks! N. F. wrote: > John Messerly wrote: >> Thanks for the ping. I think the simple fix actually works fine here. >> Let me know if it works for you & I'll get it checked in. > > Hi John...it worked great! Thanks so very much...I went ahead and pulled > the record from the MarkLogic server and all looks good. Here is the > sample code (very basic...just to show it works) in case anyone may > care: > > require 'System, Version=2.0.0.0, Culture=neutral, > PublicKeyToken=b77a5c561934e089' > require "c:\\IronRuby\\build\\release\\MarklogicXcc.dll" > # all xcc dlls have to be in the same dir as ironruby (except for > MarklogicXcc.dll) > > uri = System::Uri.new('xcc://adm:adm at localhost:9003') > content_source = > Marklogic::Xcc::ContentSourceFactory::NewContentSource(uri) > session = content_source::NewSession() > query = 'let $r := //Record[@UID eq "200608199-060"] return $r' > request = session.NewAdhocQuery(query) > resultSequence = session.SubmitRequest(request) > results = resultSequence.AsString("/n") > puts results > > -- Nivaldo -- Posted via http://www.ruby-forum.com/. From ben2004uk at googlemail.com Sat Jul 26 17:21:07 2008 From: ben2004uk at googlemail.com (Ben Hall) Date: Sat, 26 Jul 2008 22:21:07 +0100 Subject: [Ironruby-core] binary release In-Reply-To: <000c01c8eee2$098cef70$1ca6ce50$@com> References: <372109E149E8084D8E6C7D9CFD82E0632FD4357ABA@NA-EXMSG-C115.redmond.corp.microsoft.com> <38843188-D8BE-4CDE-BFA3-A92312BF8408@microsoft.com> <005b01c8ee9a$f5068a80$df139f80$@com> <000c01c8eee2$098cef70$1ca6ce50$@com> Message-ID: Out of interest, when are we going to have sync'ed releases of the DLR, IronRuby and IronPython?? Would be really useful for IronEditor ;) On Sat, Jul 26, 2008 at 6:39 AM, Peter Bacon Darwin wrote: > Just be aware that this BigDecimal is not quite there on division, > remainder/modulos and sqrt at the moment. > Pete > > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville > Sent: Friday,25 July 25, 2008 22:14 > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] binary release > > Cool, I'm going to let you handle it. BTW, I'll probably be pulling your > BigDecimal stuff tonight. Thanks for the contribution. > > JD > ________________________________________ > From: ironruby-core-bounces at rubyforge.org > [ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin > [bacondarwin at googlemail.com] > Sent: Friday, July 25, 2008 2:11 PM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] binary release > > I have a mirror of the IronRuby svn at GitHub: > http://github.com/petebacondarwin/ironruby. > I will manually track the svn repository on the "master" branch of this git > repository for the time being. (The git-svn doesn't work too well on windows > at the moment). > I have been using it to track my implementation of the BigDecimal library, > which you are welcome to take a look at in the "bigdecimal" branch. > Checkout the cool network graph to see what I mean: > http://github.com/petebacondarwin/ironruby/network. > You can pull zipped up sources from github. E.g. The master branch (which > is currently a copy of SVN r127) can be pulled from: > http://github.com/petebacondarwin/ironruby/tarball/master. > Hope that helps, > Pete > > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville > Sent: Friday,25 July 25, 2008 21:58 > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] binary release > > I can look into syncing svn to my github account. If it's easy I'll do it > and let you know. > > JD > ________________________________________ > From: ironruby-core-bounces at rubyforge.org > [ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) > [jflam at microsoft.com] > Sent: Friday, July 25, 2008 1:13 PM > To: ironruby-core at rubyforge.org > Cc: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] binary release > > Life will get better for you when we move to github for our sources as > they support tarball gen. And yes we will do more regular binary drops. > > Sent from my phone > > On Jul 25, 2008, at 11:30 AM, "Justin Bailey" > wrote: > >> On Fri, Jul 25, 2008 at 9:16 AM, John Lam (IRONRUBY) >> wrote: >>> The -I switch is still there. We're still delegating to ir.cmd to >>> run >>> IronRuby today. We'll fix some of these problems now that we have a >>> fixed >>> relative path layout to our libs directory. We'll be updating this >>> drop to >>> reflect this soon (and to do a few other things that we missed this >>> time >>> around like enabling RubyGems support out of the box). >>> >> >> Will we see regular binary releases? I'm behind a particularly >> draconian proxy and SVN won't work. HTTP headers that SVN uses are >> blocked. Being able to get regular binaries or even zipped source >> would be so nice! >> >> Justin >> _______________________________________________ >> Ironruby-core mailing list >> Ironruby-core at rubyforge.org >> http://rubyforge.org/mailman/listinfo/ironruby-core >> > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > From fuzzyman at voidspace.org.uk Sat Jul 26 17:25:14 2008 From: fuzzyman at voidspace.org.uk (Michael Foord) Date: Sat, 26 Jul 2008 22:25:14 +0100 Subject: [Ironruby-core] binary release In-Reply-To: References: <372109E149E8084D8E6C7D9CFD82E0632FD4357ABA@NA-EXMSG-C115.redmond.corp.microsoft.com> <38843188-D8BE-4CDE-BFA3-A92312BF8408@microsoft.com> <005b01c8ee9a$f5068a80$df139f80$@com> <000c01c8eee2$098cef70$1ca6ce50$@com> Message-ID: <488B963A.9010605@voidspace.org.uk> Ben Hall wrote: > Out of interest, when are we going to have sync'ed releases of the > DLR, IronRuby and IronPython?? Would be really useful for IronEditor > ;) > > As IronPython releases much less frequently, perhaps you could sync a binary release of IronRuby at the same time as the next IronPython release. Michael > On Sat, Jul 26, 2008 at 6:39 AM, Peter Bacon Darwin > wrote: > >> Just be aware that this BigDecimal is not quite there on division, >> remainder/modulos and sqrt at the moment. >> Pete >> >> -----Original Message----- >> From: ironruby-core-bounces at rubyforge.org >> [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville >> Sent: Friday,25 July 25, 2008 22:14 >> To: ironruby-core at rubyforge.org >> Subject: Re: [Ironruby-core] binary release >> >> Cool, I'm going to let you handle it. BTW, I'll probably be pulling your >> BigDecimal stuff tonight. Thanks for the contribution. >> >> JD >> ________________________________________ >> From: ironruby-core-bounces at rubyforge.org >> [ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin >> [bacondarwin at googlemail.com] >> Sent: Friday, July 25, 2008 2:11 PM >> To: ironruby-core at rubyforge.org >> Subject: Re: [Ironruby-core] binary release >> >> I have a mirror of the IronRuby svn at GitHub: >> http://github.com/petebacondarwin/ironruby. >> I will manually track the svn repository on the "master" branch of this git >> repository for the time being. (The git-svn doesn't work too well on windows >> at the moment). >> I have been using it to track my implementation of the BigDecimal library, >> which you are welcome to take a look at in the "bigdecimal" branch. >> Checkout the cool network graph to see what I mean: >> http://github.com/petebacondarwin/ironruby/network. >> You can pull zipped up sources from github. E.g. The master branch (which >> is currently a copy of SVN r127) can be pulled from: >> http://github.com/petebacondarwin/ironruby/tarball/master. >> Hope that helps, >> Pete >> >> -----Original Message----- >> From: ironruby-core-bounces at rubyforge.org >> [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville >> Sent: Friday,25 July 25, 2008 21:58 >> To: ironruby-core at rubyforge.org >> Subject: Re: [Ironruby-core] binary release >> >> I can look into syncing svn to my github account. If it's easy I'll do it >> and let you know. >> >> JD >> ________________________________________ >> From: ironruby-core-bounces at rubyforge.org >> [ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) >> [jflam at microsoft.com] >> Sent: Friday, July 25, 2008 1:13 PM >> To: ironruby-core at rubyforge.org >> Cc: ironruby-core at rubyforge.org >> Subject: Re: [Ironruby-core] binary release >> >> Life will get better for you when we move to github for our sources as >> they support tarball gen. And yes we will do more regular binary drops. >> >> Sent from my phone >> >> On Jul 25, 2008, at 11:30 AM, "Justin Bailey" >> wrote: >> >> >>> On Fri, Jul 25, 2008 at 9:16 AM, John Lam (IRONRUBY) >>> wrote: >>> >>>> The -I switch is still there. We're still delegating to ir.cmd to >>>> run >>>> IronRuby today. We'll fix some of these problems now that we have a >>>> fixed >>>> relative path layout to our libs directory. We'll be updating this >>>> drop to >>>> reflect this soon (and to do a few other things that we missed this >>>> time >>>> around like enabling RubyGems support out of the box). >>>> >>>> >>> Will we see regular binary releases? I'm behind a particularly >>> draconian proxy and SVN won't work. HTTP headers that SVN uses are >>> blocked. Being able to get regular binaries or even zipped source >>> would be so nice! >>> >>> Justin >>> _______________________________________________ >>> Ironruby-core mailing list >>> Ironruby-core at rubyforge.org >>> http://rubyforge.org/mailman/listinfo/ironruby-core >>> >>> >> _______________________________________________ >> Ironruby-core mailing list >> Ironruby-core at rubyforge.org >> http://rubyforge.org/mailman/listinfo/ironruby-core >> _______________________________________________ >> Ironruby-core mailing list >> Ironruby-core at rubyforge.org >> http://rubyforge.org/mailman/listinfo/ironruby-core >> >> _______________________________________________ >> Ironruby-core mailing list >> Ironruby-core at rubyforge.org >> http://rubyforge.org/mailman/listinfo/ironruby-core >> >> _______________________________________________ >> Ironruby-core mailing list >> Ironruby-core at rubyforge.org >> http://rubyforge.org/mailman/listinfo/ironruby-core >> >> _______________________________________________ >> Ironruby-core mailing list >> Ironruby-core at rubyforge.org >> http://rubyforge.org/mailman/listinfo/ironruby-core >> >> > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > -- http://www.ironpythoninaction.com/ http://www.voidspace.org.uk/ http://www.trypython.org/ http://www.ironpython.info/ http://www.theotherdelia.co.uk/ http://www.resolverhacks.net/ From Jimmy.Schementi at microsoft.com Sat Jul 26 17:34:45 2008 From: Jimmy.Schementi at microsoft.com (Jimmy Schementi) Date: Sat, 26 Jul 2008 14:34:45 -0700 Subject: [Ironruby-core] binary release In-Reply-To: References: <372109E149E8084D8E6C7D9CFD82E0632FD4357ABA@NA-EXMSG-C115.redmond.corp.microsoft.com> <38843188-D8BE-4CDE-BFA3-A92312BF8408@microsoft.com> <005b01c8ee9a$f5068a80$df139f80$@com> <000c01c8eee2$098cef70$1ca6ce50$@com> Message-ID: <5283CA0A4168DF4FBBD71AE9ECA5A32845D48A7D4E@NA-EXMSG-C116.redmond.corp.microsoft.com> The DLR is going to stabilize at some point in the fall (I forget the exact date, but probably around PDC) and have its own codeplex project. At that time, the language repositories will start to share the version of the DLR on codeplex. In the shorter term, I'm updating the codeplex.com/sdlsdk bits shortly, so I'd suggest just taking that and then changing around things to build against the desktop assemblies. Peter has done this already, so if you have questions he's your man. Daily builds for all these things are definitely something else that'll have to come shortly. There is a light at the end of the tunnel =) ~js > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of Ben Hall > Sent: Saturday, July 26, 2008 2:21 PM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] binary release > > Out of interest, when are we going to have sync'ed releases of the > DLR, IronRuby and IronPython?? Would be really useful for IronEditor > ;) > > On Sat, Jul 26, 2008 at 6:39 AM, Peter Bacon Darwin > wrote: > > Just be aware that this BigDecimal is not quite there on division, > > remainder/modulos and sqrt at the moment. > > Pete > > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org > > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville > > Sent: Friday,25 July 25, 2008 22:14 > > To: ironruby-core at rubyforge.org > > Subject: Re: [Ironruby-core] binary release > > > > Cool, I'm going to let you handle it. BTW, I'll probably be pulling > your > > BigDecimal stuff tonight. Thanks for the contribution. > > > > JD > > ________________________________________ > > From: ironruby-core-bounces at rubyforge.org > > [ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin > > [bacondarwin at googlemail.com] > > Sent: Friday, July 25, 2008 2:11 PM > > To: ironruby-core at rubyforge.org > > Subject: Re: [Ironruby-core] binary release > > > > I have a mirror of the IronRuby svn at GitHub: > > http://github.com/petebacondarwin/ironruby. > > I will manually track the svn repository on the "master" branch of > this git > > repository for the time being. (The git-svn doesn't work too well on > windows > > at the moment). > > I have been using it to track my implementation of the BigDecimal > library, > > which you are welcome to take a look at in the "bigdecimal" branch. > > Checkout the cool network graph to see what I mean: > > http://github.com/petebacondarwin/ironruby/network. > > You can pull zipped up sources from github. E.g. The master branch > (which > > is currently a copy of SVN r127) can be pulled from: > > http://github.com/petebacondarwin/ironruby/tarball/master. > > Hope that helps, > > Pete > > > > -----Original Message----- > > From: ironruby-core-bounces at rubyforge.org > > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville > > Sent: Friday,25 July 25, 2008 21:58 > > To: ironruby-core at rubyforge.org > > Subject: Re: [Ironruby-core] binary release > > > > I can look into syncing svn to my github account. If it's easy I'll > do it > > and let you know. > > > > JD > > ________________________________________ > > From: ironruby-core-bounces at rubyforge.org > > [ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam > (IRONRUBY) > > [jflam at microsoft.com] > > Sent: Friday, July 25, 2008 1:13 PM > > To: ironruby-core at rubyforge.org > > Cc: ironruby-core at rubyforge.org > > Subject: Re: [Ironruby-core] binary release > > > > Life will get better for you when we move to github for our sources > as > > they support tarball gen. And yes we will do more regular binary > drops. > > > > Sent from my phone > > > > On Jul 25, 2008, at 11:30 AM, "Justin Bailey" > > wrote: > > > >> On Fri, Jul 25, 2008 at 9:16 AM, John Lam (IRONRUBY) > >> wrote: > >>> The -I switch is still there. We're still delegating to ir.cmd to > >>> run > >>> IronRuby today. We'll fix some of these problems now that we have a > >>> fixed > >>> relative path layout to our libs directory. We'll be updating this > >>> drop to > >>> reflect this soon (and to do a few other things that we missed this > >>> time > >>> around like enabling RubyGems support out of the box). > >>> > >> > >> Will we see regular binary releases? I'm behind a particularly > >> draconian proxy and SVN won't work. HTTP headers that SVN uses are > >> blocked. Being able to get regular binaries or even zipped source > >> would be so nice! > >> > >> Justin > >> _______________________________________________ > >> Ironruby-core mailing list > >> Ironruby-core at rubyforge.org > >> http://rubyforge.org/mailman/listinfo/ironruby-core > >> > > _______________________________________________ > > Ironruby-core mailing list > > Ironruby-core at rubyforge.org > > http://rubyforge.org/mailman/listinfo/ironruby-core > > _______________________________________________ > > Ironruby-core mailing list > > Ironruby-core at rubyforge.org > > http://rubyforge.org/mailman/listinfo/ironruby-core > > > > _______________________________________________ > > Ironruby-core mailing list > > Ironruby-core at rubyforge.org > > http://rubyforge.org/mailman/listinfo/ironruby-core > > > > _______________________________________________ > > Ironruby-core mailing list > > Ironruby-core at rubyforge.org > > http://rubyforge.org/mailman/listinfo/ironruby-core > > > > _______________________________________________ > > Ironruby-core mailing list > > Ironruby-core at rubyforge.org > > http://rubyforge.org/mailman/listinfo/ironruby-core > > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core From bacondarwin at googlemail.com Sun Jul 27 10:47:57 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Sun, 27 Jul 2008 15:47:57 +0100 Subject: [Ironruby-core] rake mspec:... broken in r127 Message-ID: <000501c8eff7$c3738eb0$4a5aac10$@com> After shifting up to 127 I can't get run the rubyspecs any more. I fixed up the .irconfig.rb and default.mspec files correctly (by the way I had to explicitly specify the MRI binary file in .irconfig.rb since it didn't seem to be able to find it any more). I have also pulled the latest mspec, rubyspec and ironruby-tags from jflam's GitHub account. (I also tried pulling in Jim Deville's versions too). Now when I run a rubyspec I get: D:\dev\ruby\ironruby\r127>rake mspec:core Fixnum (in D:/dev/ruby/ironruby/r127) d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\callsite.cs:2 88:in `UpdateAndExecute': No such file or directory - No such file or directory - d:/dev/ruby/ironruby/r127/tests/ironruby/u til/and (Errno::ENOENT) from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\updatedelegat es.generated.cs:34:in `Update2' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\runtime\ScriptCode.cs :79:in `Run' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\sourceunit.cs:246:in `ExecuteProgram' rake aborted! Command failed with status (1): [d:/ruby/bin/ruby.exe d:/dev/ruby/ironruby/...] D:/dev/ruby/ironruby/r127/rakefile:409:in `invoke_mspec' (See full trace by running task with --trace) I guess there is something inside Rakefile or nearby that is breaking this? Unfortunately I can't get onto the SVN repository to search the log history to see what changed (Bring on Git, I say). Any ideas? Cheers Pete -------------- next part -------------- An HTML attachment was scrubbed... URL: From bacondarwin at googlemail.com Sun Jul 27 13:58:08 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Sun, 27 Jul 2008 18:58:08 +0100 Subject: [Ironruby-core] rake mspec:... broken in r127 Message-ID: <000a01c8f012$54f26b30$fed74190$@com> Further to this. The problem seems to have been introduced between r121 and r122. Pete From: Peter Bacon Darwin [mailto:bacondarwin at googlemail.com] Sent: Sunday,27 July 27, 2008 15:48 To: 'ironruby-core at rubyforge.org' Subject: rake mspec:... broken in r127 After shifting up to 127 I can't get run the rubyspecs any more. I fixed up the .irconfig.rb and default.mspec files correctly (by the way I had to explicitly specify the MRI binary file in .irconfig.rb since it didn't seem to be able to find it any more). I have also pulled the latest mspec, rubyspec and ironruby-tags from jflam's GitHub account. (I also tried pulling in Jim Deville's versions too). Now when I run a rubyspec I get: D:\dev\ruby\ironruby\r127>rake mspec:core Fixnum (in D:/dev/ruby/ironruby/r127) d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\callsite.cs:2 88:in `UpdateAndExecute': No such file or directory - No such file or directory - d:/dev/ruby/ironruby/r127/tests/ironruby/u til/and (Errno::ENOENT) from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\updatedelegat es.generated.cs:34:in `Update2' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\runtime\ScriptCode.cs :79:in `Run' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\sourceunit.cs:246:in `ExecuteProgram' rake aborted! Command failed with status (1): [d:/ruby/bin/ruby.exe d:/dev/ruby/ironruby/...] D:/dev/ruby/ironruby/r127/rakefile:409:in `invoke_mspec' (See full trace by running task with --trace) I guess there is something inside Rakefile or nearby that is breaking this? Unfortunately I can't get onto the SVN repository to search the log history to see what changed (Bring on Git, I say). Any ideas? Cheers Pete -------------- next part -------------- An HTML attachment was scrubbed... URL: From jflam at microsoft.com Sun Jul 27 14:36:18 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Sun, 27 Jul 2008 11:36:18 -0700 Subject: [Ironruby-core] rake mspec:... broken in r127 In-Reply-To: <000a01c8f012$54f26b30$fed74190$@com> References: <000a01c8f012$54f26b30$fed74190$@com> Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD4357F3A@NA-EXMSG-C115.redmond.corp.microsoft.com> Thanks for the bug report. Let me investigate - likely will be tomorrow since I am catching up on family stuff after 2 weeks on the road. Thanks, -John From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Sunday, July 27, 2008 10:58 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 Further to this. The problem seems to have been introduced between r121 and r122. Pete From: Peter Bacon Darwin [mailto:bacondarwin at googlemail.com] Sent: Sunday,27 July 27, 2008 15:48 To: 'ironruby-core at rubyforge.org' Subject: rake mspec:... broken in r127 After shifting up to 127 I can't get run the rubyspecs any more. I fixed up the .irconfig.rb and default.mspec files correctly (by the way I had to explicitly specify the MRI binary file in .irconfig.rb since it didn't seem to be able to find it any more). I have also pulled the latest mspec, rubyspec and ironruby-tags from jflam's GitHub account. (I also tried pulling in Jim Deville's versions too). Now when I run a rubyspec I get: D:\dev\ruby\ironruby\r127>rake mspec:core Fixnum (in D:/dev/ruby/ironruby/r127) d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\callsite.cs:288:in `UpdateAndExecute': No such file or directory - No such file or directory - d:/dev/ruby/ironruby/r127/tests/ironruby/u til/and (Errno::ENOENT) from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\updatedelegates.generated.cs:34:in `Update2' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\runtime\ScriptCode.cs:79:in `Run' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\sourceunit.cs:246:in `ExecuteProgram' rake aborted! Command failed with status (1): [d:/ruby/bin/ruby.exe d:/dev/ruby/ironruby/...] D:/dev/ruby/ironruby/r127/rakefile:409:in `invoke_mspec' (See full trace by running task with --trace) I guess there is something inside Rakefile or nearby that is breaking this? Unfortunately I can't get onto the SVN repository to search the log history to see what changed (Bring on Git, I say). Any ideas? Cheers Pete -------------- next part -------------- An HTML attachment was scrubbed... URL: From bacondarwin at googlemail.com Mon Jul 28 01:59:04 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Mon, 28 Jul 2008 06:59:04 +0100 Subject: [Ironruby-core] rake mspec:... broken in r127 In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD4357F3A@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <000a01c8f012$54f26b30$fed74190$@com> <372109E149E8084D8E6C7D9CFD82E0632FD4357F3A@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <000001c8f077$0b7e9a40$227bcec0$@com> I think I have solved the problem. See the attached patch. There are two issues in the Rakefile: 1) UserEnvironment::MRI just holds the path two the base of MRI not the path to the binary. I have added a new constant MRI_BINARY and modified the places where this is used. 2) The -B option on mspec takes a path to the config file. On my machine this is in "C:\Documents and Settings\pete\default.mspec". Since this has spaces in it you have to put quotes round the path. I have added this but I am not sure what effect this would have on Unix based systems and also whether it ought to be done for all the path parameters. Cheers, Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: Sunday,27 July 27, 2008 19:36 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 Thanks for the bug report. Let me investigate - likely will be tomorrow since I am catching up on family stuff after 2 weeks on the road. Thanks, -John From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Sunday, July 27, 2008 10:58 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 Further to this. The problem seems to have been introduced between r121 and r122. Pete From: Peter Bacon Darwin [mailto:bacondarwin at googlemail.com] Sent: Sunday,27 July 27, 2008 15:48 To: 'ironruby-core at rubyforge.org' Subject: rake mspec:... broken in r127 After shifting up to 127 I can't get run the rubyspecs any more. I fixed up the .irconfig.rb and default.mspec files correctly (by the way I had to explicitly specify the MRI binary file in .irconfig.rb since it didn't seem to be able to find it any more). I have also pulled the latest mspec, rubyspec and ironruby-tags from jflam's GitHub account. (I also tried pulling in Jim Deville's versions too). Now when I run a rubyspec I get: D:\dev\ruby\ironruby\r127>rake mspec:core Fixnum (in D:/dev/ruby/ironruby/r127) d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\callsite.cs:2 88:in `UpdateAndExecute': No such file or directory - No such file or directory - d:/dev/ruby/ironruby/r127/tests/ironruby/u til/and (Errno::ENOENT) from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\updatedelegat es.generated.cs:34:in `Update2' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\runtime\ScriptCode.cs :79:in `Run' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\sourceunit.cs:246:in `ExecuteProgram' rake aborted! Command failed with status (1): [d:/ruby/bin/ruby.exe d:/dev/ruby/ironruby/...] D:/dev/ruby/ironruby/r127/rakefile:409:in `invoke_mspec' (See full trace by running task with --trace) I guess there is something inside Rakefile or nearby that is breaking this? Unfortunately I can't get onto the SVN repository to search the log history to see what changed (Bring on Git, I say). Any ideas? Cheers Pete -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: mspec.patch Type: application/octet-stream Size: 1513 bytes Desc: not available URL: From bacondarwin at googlemail.com Mon Jul 28 03:24:19 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Mon, 28 Jul 2008 08:24:19 +0100 Subject: [Ironruby-core] rake mspec:... broken in r127 In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD4357F3A@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <000a01c8f012$54f26b30$fed74190$@com> <372109E149E8084D8E6C7D9CFD82E0632FD4357F3A@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <000601c8f082$f4bcfca0$de36f5e0$@com> Also my runfirst.cmd looks like this: sn -Vr *,31bf3856ad364e35 copy default.irconfig.rb "%USERPROFILE%\.irconfig.rb" copy default.mspec.rb "%USERPROFILE%\dev\default.mspec" Personally it seems that it would be better for default.mspec to be placed directly under USERPROFILE. E.g. sn -Vr *,31bf3856ad364e35 copy default.irconfig.rb "%USERPROFILE%\.irconfig.rb" copy default.mspec.rb "%USERPROFILE%\default.mspec" Especially since I don't keep my other stuff inside %USERPROFILE%\dev. Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: Sunday,27 July 27, 2008 19:36 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 Thanks for the bug report. Let me investigate - likely will be tomorrow since I am catching up on family stuff after 2 weeks on the road. Thanks, -John From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Sunday, July 27, 2008 10:58 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 Further to this. The problem seems to have been introduced between r121 and r122. Pete From: Peter Bacon Darwin [mailto:bacondarwin at googlemail.com] Sent: Sunday,27 July 27, 2008 15:48 To: 'ironruby-core at rubyforge.org' Subject: rake mspec:... broken in r127 After shifting up to 127 I can't get run the rubyspecs any more. I fixed up the .irconfig.rb and default.mspec files correctly (by the way I had to explicitly specify the MRI binary file in .irconfig.rb since it didn't seem to be able to find it any more). I have also pulled the latest mspec, rubyspec and ironruby-tags from jflam's GitHub account. (I also tried pulling in Jim Deville's versions too). Now when I run a rubyspec I get: D:\dev\ruby\ironruby\r127>rake mspec:core Fixnum (in D:/dev/ruby/ironruby/r127) d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\callsite.cs:2 88:in `UpdateAndExecute': No such file or directory - No such file or directory - d:/dev/ruby/ironruby/r127/tests/ironruby/u til/and (Errno::ENOENT) from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\updatedelegat es.generated.cs:34:in `Update2' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\runtime\ScriptCode.cs :79:in `Run' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\sourceunit.cs:246:in `ExecuteProgram' rake aborted! Command failed with status (1): [d:/ruby/bin/ruby.exe d:/dev/ruby/ironruby/...] D:/dev/ruby/ironruby/r127/rakefile:409:in `invoke_mspec' (See full trace by running task with --trace) I guess there is something inside Rakefile or nearby that is breaking this? Unfortunately I can't get onto the SVN repository to search the log history to see what changed (Bring on Git, I say). Any ideas? Cheers Pete -------------- next part -------------- An HTML attachment was scrubbed... URL: From bacondarwin at googlemail.com Mon Jul 28 09:07:17 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Mon, 28 Jul 2008 14:07:17 +0100 Subject: [Ironruby-core] $HOME Message-ID: <002d01c8f0b2$ddd8b170$998a1450$@com> By the way, this is how $HOME is setup in mspec (on Windows): # Code to setup HOME directory correctly on Windows # This duplicates Ruby 1.9 semantics for defining HOME platform_is :windows do if ENV['HOME'] ENV['HOME'] = ENV['HOME'].tr '\\', '/' elsif ENV['HOMEDIR'] && ENV['HOMEDRIVE'] ENV['HOME'] = File.join(ENV['HOMEDRIVE'], ENV['HOMEDIR']) elsif ENV['HOMEDIR'] ENV['HOME'] = ENV['HOMEDIR'] elsif ENV['HOMEDRIVE'] ENV['HOME'] = ENV['HOMEDRIVE'] elsif ENV['USERPROFILE'] ENV['HOME'] = ENV['USERPROFILE'] else puts "No suitable HOME environment found. This means that all of HOME, HOMEDIR, HOMEDRIVE, and USERPROFILE are not set" exit 1 end end -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdeville at microsoft.com Mon Jul 28 10:43:54 2008 From: jdeville at microsoft.com (Jim Deville) Date: Mon, 28 Jul 2008 07:43:54 -0700 Subject: [Ironruby-core] rake mspec:... broken in r127 In-Reply-To: <000001c8f077$0b7e9a40$227bcec0$@com> References: <000a01c8f012$54f26b30$fed74190$@com> <372109E149E8084D8E6C7D9CFD82E0632FD4357F3A@NA-EXMSG-C115.redmond.corp.microsoft.com>, <000001c8f077$0b7e9a40$227bcec0$@com> Message-ID: If you look at how UserEnvironment sets up the MRI constant (if it's missing) you'll see that it should point to the binary. We should update the default .irconfig.rb, and change the comment instead. JD ________________________________________ From: ironruby-core-bounces at rubyforge.org [ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin [bacondarwin at googlemail.com] Sent: Sunday, July 27, 2008 10:59 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 I think I have solved the problem. See the attached patch. There are two issues in the Rakefile: 1) UserEnvironment::MRI just holds the path two the base of MRI not the path to the binary. I have added a new constant MRI_BINARY and modified the places where this is used. 2) The ?B option on mspec takes a path to the config file. On my machine this is in ?C:\Documents and Settings\pete\default.mspec?. Since this has spaces in it you have to put quotes round the path. I have added this but I am not sure what effect this would have on Unix based systems and also whether it ought to be done for all the path parameters. Cheers, Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: Sunday,27 July 27, 2008 19:36 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 Thanks for the bug report. Let me investigate ? likely will be tomorrow since I am catching up on family stuff after 2 weeks on the road. Thanks, -John From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Sunday, July 27, 2008 10:58 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 Further to this. The problem seems to have been introduced between r121 and r122. Pete From: Peter Bacon Darwin [mailto:bacondarwin at googlemail.com] Sent: Sunday,27 July 27, 2008 15:48 To: 'ironruby-core at rubyforge.org' Subject: rake mspec:... broken in r127 After shifting up to 127 I can?t get run the rubyspecs any more. I fixed up the .irconfig.rb and default.mspec files correctly (by the way I had to explicitly specify the MRI binary file in .irconfig.rb since it didn?t seem to be able to find it any more). I have also pulled the latest mspec, rubyspec and ironruby-tags from jflam?s GitHub account. (I also tried pulling in Jim Deville?s versions too). Now when I run a rubyspec I get: D:\dev\ruby\ironruby\r127>rake mspec:core Fixnum (in D:/dev/ruby/ironruby/r127) d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\callsite.cs:288:in `UpdateAndExecute': No such file or directory - No such file or directory - d:/dev/ruby/ironruby/r127/tests/ironruby/u til/and (Errno::ENOENT) from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\updatedelegates.generated.cs:34:in `Update2' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\runtime\ScriptCode.cs:79:in `Run' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\sourceunit.cs:246:in `ExecuteProgram' rake aborted! Command failed with status (1): [d:/ruby/bin/ruby.exe d:/dev/ruby/ironruby/...] D:/dev/ruby/ironruby/r127/rakefile:409:in `invoke_mspec' (See full trace by running task with --trace) I guess there is something inside Rakefile or nearby that is breaking this? Unfortunately I can?t get onto the SVN repository to search the log history to see what changed (Bring on Git, I say). Any ideas? Cheers Pete From bacondarwin at googlemail.com Mon Jul 28 12:48:20 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Mon, 28 Jul 2008 17:48:20 +0100 Subject: [Ironruby-core] rake mspec:... broken in r127 In-Reply-To: References: <000a01c8f012$54f26b30$fed74190$@com> <372109E149E8084D8E6C7D9CFD82E0632FD4357F3A@NA-EXMSG-C115.redmond.corp.microsoft.com>, <000001c8f077$0b7e9a40$227bcec0$@com> Message-ID: <000001c8f0d1$bf443ad0$3dccb070$@com> I don't think it does (r127): unless defined?(UserEnvironment::MRI) ruby_exe_paths = UserEnvironment.find_executable 'ruby.exe' if ruby_exe_paths.length == 1 UserEnvironment.const_set(:MRI, Pathname.new(ruby_exe_paths.first + '\..\\')) else raise ArgumentError.new("Found more than one version of ruby.exe on your path #{ruby_exe_paths.join(', ')}") end end ... def self.find_executable(executable) executable.downcase! result = [] search_path = ENV['PATH'].split(';') search_path.each do |dir| path = dir.gsub '\\', '/' Dir[path + '/*.exe'].each do |file| file_path = Pathname.new(file) result << file_path.dirname if file_path.basename.downcase == executable end end result end -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Monday,28 July 28, 2008 15:44 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 If you look at how UserEnvironment sets up the MRI constant (if it's missing) you'll see that it should point to the binary. We should update the default .irconfig.rb, and change the comment instead. JD ________________________________________ From: ironruby-core-bounces at rubyforge.org [ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin [bacondarwin at googlemail.com] Sent: Sunday, July 27, 2008 10:59 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 I think I have solved the problem. See the attached patch. There are two issues in the Rakefile: 1) UserEnvironment::MRI just holds the path two the base of MRI not the path to the binary. I have added a new constant MRI_BINARY and modified the places where this is used. 2) The -B option on mspec takes a path to the config file. On my machine this is in "C:\Documents and Settings\pete\default.mspec". Since this has spaces in it you have to put quotes round the path. I have added this but I am not sure what effect this would have on Unix based systems and also whether it ought to be done for all the path parameters. Cheers, Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: Sunday,27 July 27, 2008 19:36 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 Thanks for the bug report. Let me investigate - likely will be tomorrow since I am catching up on family stuff after 2 weeks on the road. Thanks, -John From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Sunday, July 27, 2008 10:58 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 Further to this. The problem seems to have been introduced between r121 and r122. Pete From: Peter Bacon Darwin [mailto:bacondarwin at googlemail.com] Sent: Sunday,27 July 27, 2008 15:48 To: 'ironruby-core at rubyforge.org' Subject: rake mspec:... broken in r127 After shifting up to 127 I can't get run the rubyspecs any more. I fixed up the .irconfig.rb and default.mspec files correctly (by the way I had to explicitly specify the MRI binary file in .irconfig.rb since it didn't seem to be able to find it any more). I have also pulled the latest mspec, rubyspec and ironruby-tags from jflam's GitHub account. (I also tried pulling in Jim Deville's versions too). Now when I run a rubyspec I get: D:\dev\ruby\ironruby\r127>rake mspec:core Fixnum (in D:/dev/ruby/ironruby/r127) d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\callsite.cs:2 88:in `UpdateAndExecute': No such file or directory - No such file or directory - d:/dev/ruby/ironruby/r127/tests/ironruby/u til/and (Errno::ENOENT) from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\updatedelegat es.generated.cs:34:in `Update2' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\runtime\ScriptCode.cs :79:in `Run' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\sourceunit.cs:246:in `ExecuteProgram' rake aborted! Command failed with status (1): [d:/ruby/bin/ruby.exe d:/dev/ruby/ironruby/...] D:/dev/ruby/ironruby/r127/rakefile:409:in `invoke_mspec' (See full trace by running task with --trace) I guess there is something inside Rakefile or nearby that is breaking this? Unfortunately I can't get onto the SVN repository to search the log history to see what changed (Bring on Git, I say). Any ideas? Cheers Pete _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From jdeville at microsoft.com Mon Jul 28 13:57:18 2008 From: jdeville at microsoft.com (Jim Deville) Date: Mon, 28 Jul 2008 10:57:18 -0700 Subject: [Ironruby-core] rake mspec:... broken in r127 In-Reply-To: <000001c8f0d1$bf443ad0$3dccb070$@com> References: <000a01c8f012$54f26b30$fed74190$@com> <372109E149E8084D8E6C7D9CFD82E0632FD4357F3A@NA-EXMSG-C115.redmond.corp.microsoft.com>, <000001c8f077$0b7e9a40$227bcec0$@com> , <000001c8f0d1$bf443ad0$3dccb070$@com> Message-ID: Looks like I've been using it wrong. Sorry about that. JD ________________________________________ From: ironruby-core-bounces at rubyforge.org [ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin [bacondarwin at googlemail.com] Sent: Monday, July 28, 2008 9:48 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 I don't think it does (r127): unless defined?(UserEnvironment::MRI) ruby_exe_paths = UserEnvironment.find_executable 'ruby.exe' if ruby_exe_paths.length == 1 UserEnvironment.const_set(:MRI, Pathname.new(ruby_exe_paths.first + '\..\\')) else raise ArgumentError.new("Found more than one version of ruby.exe on your path #{ruby_exe_paths.join(', ')}") end end ... def self.find_executable(executable) executable.downcase! result = [] search_path = ENV['PATH'].split(';') search_path.each do |dir| path = dir.gsub '\\', '/' Dir[path + '/*.exe'].each do |file| file_path = Pathname.new(file) result << file_path.dirname if file_path.basename.downcase == executable end end result end -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Jim Deville Sent: Monday,28 July 28, 2008 15:44 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 If you look at how UserEnvironment sets up the MRI constant (if it's missing) you'll see that it should point to the binary. We should update the default .irconfig.rb, and change the comment instead. JD ________________________________________ From: ironruby-core-bounces at rubyforge.org [ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin [bacondarwin at googlemail.com] Sent: Sunday, July 27, 2008 10:59 PM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 I think I have solved the problem. See the attached patch. There are two issues in the Rakefile: 1) UserEnvironment::MRI just holds the path two the base of MRI not the path to the binary. I have added a new constant MRI_BINARY and modified the places where this is used. 2) The -B option on mspec takes a path to the config file. On my machine this is in "C:\Documents and Settings\pete\default.mspec". Since this has spaces in it you have to put quotes round the path. I have added this but I am not sure what effect this would have on Unix based systems and also whether it ought to be done for all the path parameters. Cheers, Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: Sunday,27 July 27, 2008 19:36 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 Thanks for the bug report. Let me investigate - likely will be tomorrow since I am catching up on family stuff after 2 weeks on the road. Thanks, -John From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin Sent: Sunday, July 27, 2008 10:58 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 Further to this. The problem seems to have been introduced between r121 and r122. Pete From: Peter Bacon Darwin [mailto:bacondarwin at googlemail.com] Sent: Sunday,27 July 27, 2008 15:48 To: 'ironruby-core at rubyforge.org' Subject: rake mspec:... broken in r127 After shifting up to 127 I can't get run the rubyspecs any more. I fixed up the .irconfig.rb and default.mspec files correctly (by the way I had to explicitly specify the MRI binary file in .irconfig.rb since it didn't seem to be able to find it any more). I have also pulled the latest mspec, rubyspec and ironruby-tags from jflam's GitHub account. (I also tried pulling in Jim Deville's versions too). Now when I run a rubyspec I get: D:\dev\ruby\ironruby\r127>rake mspec:core Fixnum (in D:/dev/ruby/ironruby/r127) d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\callsite.cs:2 88:in `UpdateAndExecute': No such file or directory - No such file or directory - d:/dev/ruby/ironruby/r127/tests/ironruby/u til/and (Errno::ENOENT) from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\updatedelegat es.generated.cs:34:in `Update2' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\runtime\ScriptCode.cs :79:in `Run' from d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\sourceunit.cs:246:in `ExecuteProgram' rake aborted! Command failed with status (1): [d:/ruby/bin/ruby.exe d:/dev/ruby/ironruby/...] D:/dev/ruby/ironruby/r127/rakefile:409:in `invoke_mspec' (See full trace by running task with --trace) I guess there is something inside Rakefile or nearby that is breaking this? Unfortunately I can't get onto the SVN repository to search the log history to see what changed (Bring on Git, I say). Any ideas? Cheers Pete _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From dotnetguy at gmail.com Mon Jul 28 18:17:48 2008 From: dotnetguy at gmail.com (Brad Wilson) Date: Mon, 28 Jul 2008 15:17:48 -0700 Subject: [Ironruby-core] $HOME In-Reply-To: <002d01c8f0b2$ddd8b170$998a1450$@com> References: <002d01c8f0b2$ddd8b170$998a1450$@com> Message-ID: <6bbcce480807281517m769580cep87395dc378f97481@mail.gmail.com> That's wrong. The order should be: HOME USERPROFILE HOMEDRIVE + HOMEDIR Using HOMEDIR without HOMEDRIVE (and the opposite) are bad ideas. On Mon, Jul 28, 2008 at 6:07 AM, Peter Bacon Darwin < bacondarwin at googlemail.com> wrote: > By the way, this is how $HOME is setup in mspec (on Windows): > > > > # Code to setup HOME directory correctly on Windows > > # This duplicates Ruby 1.9 semantics for defining HOME > > platform_is :windows do > > if ENV['HOME'] > > ENV['HOME'] = ENV['HOME'].tr '\\', '/' > > elsif ENV['HOMEDIR'] && ENV['HOMEDRIVE'] > > ENV['HOME'] = File.join(ENV['HOMEDRIVE'], ENV['HOMEDIR']) > > elsif ENV['HOMEDIR'] > > ENV['HOME'] = ENV['HOMEDIR'] > > elsif ENV['HOMEDRIVE'] > > ENV['HOME'] = ENV['HOMEDRIVE'] > > elsif ENV['USERPROFILE'] > > ENV['HOME'] = ENV['USERPROFILE'] > > else > > puts "No suitable HOME environment found. This means that all of > HOME, HOMEDIR, HOMEDRIVE, and USERPROFILE are not set" > > exit 1 > > end > > end > > _______________________________________________ > Ironruby-core mailing list > Ironruby-core at rubyforge.org > http://rubyforge.org/mailman/listinfo/ironruby-core > > -- http://bradwilson.typepad.com/ http://social.zune.net/member/AgilistBrad http://www.flickr.com/photos/dotnetguy/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From bacondarwin at googlemail.com Tue Jul 29 03:48:04 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Tue, 29 Jul 2008 08:48:04 +0100 Subject: [Ironruby-core] $HOME In-Reply-To: <6bbcce480807281517m769580cep87395dc378f97481@mail.gmail.com> References: <002d01c8f0b2$ddd8b170$998a1450$@com> <6bbcce480807281517m769580cep87395dc378f97481@mail.gmail.com> Message-ID: <000601c8f14f$711d11a0$535734e0$@com> So the question is, ?Given that this is in mspec and it claims to reproduce Ruby 1.9 semantic, who do we send the bug report to?? From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Brad Wilson Sent: Monday,28 July 28, 2008 23:18 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] $HOME That's wrong. The order should be: HOME USERPROFILE HOMEDRIVE + HOMEDIR Using HOMEDIR without HOMEDRIVE (and the opposite) are bad ideas. On Mon, Jul 28, 2008 at 6:07 AM, Peter Bacon Darwin wrote: By the way, this is how $HOME is setup in mspec (on Windows): # Code to setup HOME directory correctly on Windows # This duplicates Ruby 1.9 semantics for defining HOME platform_is :windows do if ENV['HOME'] ENV['HOME'] = ENV['HOME'].tr '\\', '/' elsif ENV['HOMEDIR'] && ENV['HOMEDRIVE'] ENV['HOME'] = File.join(ENV['HOMEDRIVE'], ENV['HOMEDIR']) elsif ENV['HOMEDIR'] ENV['HOME'] = ENV['HOMEDIR'] elsif ENV['HOMEDRIVE'] ENV['HOME'] = ENV['HOMEDRIVE'] elsif ENV['USERPROFILE'] ENV['HOME'] = ENV['USERPROFILE'] else puts "No suitable HOME environment found. This means that all of HOME, HOMEDIR, HOMEDRIVE, and USERPROFILE are not set" exit 1 end end -------------- next part -------------- An HTML attachment was scrubbed... URL: From bacondarwin at googlemail.com Tue Jul 29 03:57:45 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Tue, 29 Jul 2008 08:57:45 +0100 Subject: [Ironruby-core] $HOME In-Reply-To: <6bbcce480807281517m769580cep87395dc378f97481@mail.gmail.com> References: <002d01c8f0b2$ddd8b170$998a1450$@com> <6bbcce480807281517m769580cep87395dc378f97481@mail.gmail.com> Message-ID: <000b01c8f150$ca98d330$5fca7990$@com> Ruby 1.9 actually uses this order: HOME HOMEDRIVE + HOMEPATH HOMEPATH USERPROFILE SHGetSpecialFolderLocation(NULL, CSIDL_PERSONAL, &pidl) This seems reasonable. I suspect mspec needs tweaking. Pete From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Brad Wilson Sent: Monday,28 July 28, 2008 23:18 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] $HOME That's wrong. The order should be: HOME USERPROFILE HOMEDRIVE + HOMEDIR Using HOMEDIR without HOMEDRIVE (and the opposite) are bad ideas. On Mon, Jul 28, 2008 at 6:07 AM, Peter Bacon Darwin wrote: By the way, this is how $HOME is setup in mspec (on Windows): # Code to setup HOME directory correctly on Windows # This duplicates Ruby 1.9 semantics for defining HOME platform_is :windows do if ENV['HOME'] ENV['HOME'] = ENV['HOME'].tr '\\', '/' elsif ENV['HOMEDIR'] && ENV['HOMEDRIVE'] ENV['HOME'] = File.join(ENV['HOMEDRIVE'], ENV['HOMEDIR']) elsif ENV['HOMEDIR'] ENV['HOME'] = ENV['HOMEDIR'] elsif ENV['HOMEDRIVE'] ENV['HOME'] = ENV['HOMEDRIVE'] elsif ENV['USERPROFILE'] ENV['HOME'] = ENV['USERPROFILE'] else puts "No suitable HOME environment found. This means that all of HOME, HOMEDIR, HOMEDRIVE, and USERPROFILE are not set" exit 1 end end _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core -- http://bradwilson.typepad.com/ http://social.zune.net/member/AgilistBrad http://www.flickr.com/photos/dotnetguy/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From jflam at microsoft.com Tue Jul 29 13:13:38 2008 From: jflam at microsoft.com (John Lam (IRONRUBY)) Date: Tue, 29 Jul 2008 10:13:38 -0700 Subject: [Ironruby-core] rake mspec:... broken in r127 In-Reply-To: <000001c8f077$0b7e9a40$227bcec0$@com> References: <000a01c8f012$54f26b30$fed74190$@com> <372109E149E8084D8E6C7D9CFD82E0632FD4357F3A@NA-EXMSG-C115.redmond.corp.microsoft.com> <000001c8f077$0b7e9a40$227bcec0$@com> Message-ID: <372109E149E8084D8E6C7D9CFD82E0632FD43F2D97@NA-EXMSG-C115.redmond.corp.microsoft.com> Pete - Applying your patch now. However, in your patch I'm assuming that the part where you have a reference to RUBY_BINARY should be MRI_BINARY? Also fixing quotes around the path issue. Thanks, -John > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin > Sent: Sunday, July 27, 2008 10:59 PM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] rake mspec:... broken in r127 > > I think I have solved the problem. See the attached patch. There are > two issues in the Rakefile: > > 1) UserEnvironment::MRI just holds the path two the base of MRI > not > the path to the binary. I have added a new constant MRI_BINARY and > modified the places where this is used. > > 2) The -B option on mspec takes a path to the config file. On my > machine this is in "C:\Documents and Settings\pete\default.mspec". > Since this has spaces in it you have to put quotes round the path. I > have added this but I am not sure what effect this would have on Unix > based systems and also whether it ought to be done for all the path > parameters. > > Cheers, > > Pete > > > > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam > (IRONRUBY) > Sent: Sunday,27 July 27, 2008 19:36 > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] rake mspec:... broken in r127 > > > > Thanks for the bug report. Let me investigate - likely will be > tomorrow since I am catching up on family stuff after 2 weeks on the road. > > > > Thanks, > > -John > > > > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon > Darwin > Sent: Sunday, July 27, 2008 10:58 AM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] rake mspec:... broken in r127 > > > > Further to this. The problem seems to have been introduced between > r121 and r122. > > Pete > > > > From: Peter Bacon Darwin [mailto:bacondarwin at googlemail.com] > Sent: Sunday,27 July 27, 2008 15:48 > To: 'ironruby-core at rubyforge.org' > Subject: rake mspec:... broken in r127 > > > > After shifting up to 127 I can't get run the rubyspecs any more. > > I fixed up the .irconfig.rb and default.mspec files correctly (by the > way I had to explicitly specify the MRI binary file in .irconfig.rb > since it didn't seem to be able to find it any more). > > I have also pulled the latest mspec, rubyspec and ironruby-tags from > jflam's GitHub account. (I also tried pulling in Jim Deville's > versions too). > > > > Now when I run a rubyspec I get: > > > > D:\dev\ruby\ironruby\r127>rake mspec:core Fixnum > > (in D:/dev/ruby/ironruby/r127) > > d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\callsit > e > . > cs:288:in `UpdateAndExecute': No such file or directory - No such file > or directory - d:/dev/ruby/ironruby/r127/tests/ironruby/u > > til/and (Errno::ENOENT) > > from > d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\updated > e > l > egates.generated.cs:34:in `Update2' > > from > d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\runtime\ScriptC > o > d > e.cs:79:in `Run' > > from > d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\sourceunit.cs:2 > 4 > 6 > :in `ExecuteProgram' > > rake aborted! > > Command failed with status (1): [d:/ruby/bin/ruby.exe > d:/dev/ruby/ironruby/...] > > D:/dev/ruby/ironruby/r127/rakefile:409:in `invoke_mspec' > > (See full trace by running task with --trace) > > > > I guess there is something inside Rakefile or nearby that is breaking > this? Unfortunately I can't get onto the SVN repository to search the > log history to see what changed (Bring on Git, I say). > > > > Any ideas? > > Cheers > > Pete From bacondarwin at googlemail.com Tue Jul 29 13:36:26 2008 From: bacondarwin at googlemail.com (Peter Bacon Darwin) Date: Tue, 29 Jul 2008 18:36:26 +0100 Subject: [Ironruby-core] rake mspec:... broken in r127 In-Reply-To: <372109E149E8084D8E6C7D9CFD82E0632FD43F2D97@NA-EXMSG-C115.redmond.corp.microsoft.com> References: <000a01c8f012$54f26b30$fed74190$@com> <372109E149E8084D8E6C7D9CFD82E0632FD4357F3A@NA-EXMSG-C115.redmond.corp.microsoft.com> <000001c8f077$0b7e9a40$227bcec0$@com> <372109E149E8084D8E6C7D9CFD82E0632FD43F2D97@NA-EXMSG-C115.redmond.corp.microsoft.com> Message-ID: <003001c8f1a1$a1c91e00$e55b5a00$@com> Oops. Yes you are absolutely correct. Also, by the way. I noticed that mspec has gone to 1.4.1 (and latest RubySpec requires this) but the versions I can get working are the forks on your github account that are at 1.1.1 still. Moreover I noticed that you guys have set up a new github account called ironruby. Is this the one we should be aiming to work from in the future? What's the plan. Thanks, Pete -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: Tuesday,29 July 29, 2008 18:14 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 Pete - Applying your patch now. However, in your patch I'm assuming that the part where you have a reference to RUBY_BINARY should be MRI_BINARY? Also fixing quotes around the path issue. Thanks, -John > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin > Sent: Sunday, July 27, 2008 10:59 PM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] rake mspec:... broken in r127 > > I think I have solved the problem. See the attached patch. There are > two issues in the Rakefile: > > 1) UserEnvironment::MRI just holds the path two the base of MRI > not > the path to the binary. I have added a new constant MRI_BINARY and > modified the places where this is used. > > 2) The -B option on mspec takes a path to the config file. On my > machine this is in "C:\Documents and Settings\pete\default.mspec". > Since this has spaces in it you have to put quotes round the path. I > have added this but I am not sure what effect this would have on Unix > based systems and also whether it ought to be done for all the path > parameters. > > Cheers, > > Pete > > > > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam > (IRONRUBY) > Sent: Sunday,27 July 27, 2008 19:36 > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] rake mspec:... broken in r127 > > > > Thanks for the bug report. Let me investigate - likely will be > tomorrow since I am catching up on family stuff after 2 weeks on the road. > > > > Thanks, > > -John > > > > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon > Darwin > Sent: Sunday, July 27, 2008 10:58 AM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] rake mspec:... broken in r127 > > > > Further to this. The problem seems to have been introduced between > r121 and r122. > > Pete > > > > From: Peter Bacon Darwin [mailto:bacondarwin at googlemail.com] > Sent: Sunday,27 July 27, 2008 15:48 > To: 'ironruby-core at rubyforge.org' > Subject: rake mspec:... broken in r127 > > > > After shifting up to 127 I can't get run the rubyspecs any more. > > I fixed up the .irconfig.rb and default.mspec files correctly (by the > way I had to explicitly specify the MRI binary file in .irconfig.rb > since it didn't seem to be able to find it any more). > > I have also pulled the latest mspec, rubyspec and ironruby-tags from > jflam's GitHub account. (I also tried pulling in Jim Deville's > versions too). > > > > Now when I run a rubyspec I get: > > > > D:\dev\ruby\ironruby\r127>rake mspec:core Fixnum > > (in D:/dev/ruby/ironruby/r127) > > d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\callsit > e > . > cs:288:in `UpdateAndExecute': No such file or directory - No such file > or directory - d:/dev/ruby/ironruby/r127/tests/ironruby/u > > til/and (Errno::ENOENT) > > from > d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\updated > e > l > egates.generated.cs:34:in `Update2' > > from > d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\runtime\ScriptC > o > d > e.cs:79:in `Run' > > from > d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\sourceunit.cs:2 > 4 > 6 > :in `ExecuteProgram' > > rake aborted! > > Command failed with status (1): [d:/ruby/bin/ruby.exe > d:/dev/ruby/ironruby/...] > > D:/dev/ruby/ironruby/r127/rakefile:409:in `invoke_mspec' > > (See full trace by running task with --trace) > > > > I guess there is something inside Rakefile or nearby that is breaking > this? Unfortunately I can't get onto the SVN repository to search the > log history to see what changed (Bring on Git, I say). > > > > Any ideas? > > Cheers > > Pete _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From jdeville at microsoft.com Tue Jul 29 14:07:38 2008 From: jdeville at microsoft.com (Jim Deville) Date: Tue, 29 Jul 2008 11:07:38 -0700 Subject: [Ironruby-core] rake mspec:... broken in r127 In-Reply-To: <003001c8f1a1$a1c91e00$e55b5a00$@com> References: <000a01c8f012$54f26b30$fed74190$@com> <372109E149E8084D8E6C7D9CFD82E0632FD4357F3A@NA-EXMSG-C115.redmond.corp.microsoft.com> <000001c8f077$0b7e9a40$227bcec0$@com> <372109E149E8084D8E6C7D9CFD82E0632FD43F2D97@NA-EXMSG-C115.redmond.corp.microsoft.com>, <003001c8f1a1$a1c91e00$e55b5a00$@com> Message-ID: The ironruby account will the authoritative account in the future. It is up to date at 1.4, and I had it working before our release on Thursday. JD ________________________________________ From: ironruby-core-bounces at rubyforge.org [ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin [bacondarwin at googlemail.com] Sent: Tuesday, July 29, 2008 10:36 AM To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 Oops. Yes you are absolutely correct. Also, by the way. I noticed that mspec has gone to 1.4.1 (and latest RubySpec requires this) but the versions I can get working are the forks on your github account that are at 1.1.1 still. Moreover I noticed that you guys have set up a new github account called ironruby. Is this the one we should be aiming to work from in the future? What's the plan. Thanks, Pete -----Original Message----- From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam (IRONRUBY) Sent: Tuesday,29 July 29, 2008 18:14 To: ironruby-core at rubyforge.org Subject: Re: [Ironruby-core] rake mspec:... broken in r127 Pete - Applying your patch now. However, in your patch I'm assuming that the part where you have a reference to RUBY_BINARY should be MRI_BINARY? Also fixing quotes around the path issue. Thanks, -John > -----Original Message----- > From: ironruby-core-bounces at rubyforge.org [mailto:ironruby-core- > bounces at rubyforge.org] On Behalf Of Peter Bacon Darwin > Sent: Sunday, July 27, 2008 10:59 PM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] rake mspec:... broken in r127 > > I think I have solved the problem. See the attached patch. There are > two issues in the Rakefile: > > 1) UserEnvironment::MRI just holds the path two the base of MRI > not > the path to the binary. I have added a new constant MRI_BINARY and > modified the places where this is used. > > 2) The -B option on mspec takes a path to the config file. On my > machine this is in "C:\Documents and Settings\pete\default.mspec". > Since this has spaces in it you have to put quotes round the path. I > have added this but I am not sure what effect this would have on Unix > based systems and also whether it ought to be done for all the path > parameters. > > Cheers, > > Pete > > > > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of John Lam > (IRONRUBY) > Sent: Sunday,27 July 27, 2008 19:36 > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] rake mspec:... broken in r127 > > > > Thanks for the bug report. Let me investigate - likely will be > tomorrow since I am catching up on family stuff after 2 weeks on the road. > > > > Thanks, > > -John > > > > From: ironruby-core-bounces at rubyforge.org > [mailto:ironruby-core-bounces at rubyforge.org] On Behalf Of Peter Bacon > Darwin > Sent: Sunday, July 27, 2008 10:58 AM > To: ironruby-core at rubyforge.org > Subject: Re: [Ironruby-core] rake mspec:... broken in r127 > > > > Further to this. The problem seems to have been introduced between > r121 and r122. > > Pete > > > > From: Peter Bacon Darwin [mailto:bacondarwin at googlemail.com] > Sent: Sunday,27 July 27, 2008 15:48 > To: 'ironruby-core at rubyforge.org' > Subject: rake mspec:... broken in r127 > > > > After shifting up to 127 I can't get run the rubyspecs any more. > > I fixed up the .irconfig.rb and default.mspec files correctly (by the > way I had to explicitly specify the MRI binary file in .irconfig.rb > since it didn't seem to be able to find it any more). > > I have also pulled the latest mspec, rubyspec and ironruby-tags from > jflam's GitHub account. (I also tried pulling in Jim Deville's > versions too). > > > > Now when I run a rubyspec I get: > > > > D:\dev\ruby\ironruby\r127>rake mspec:core Fixnum > > (in D:/dev/ruby/ironruby/r127) > > d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\callsit > e > . > cs:288:in `UpdateAndExecute': No such file or directory - No such file > or directory - d:/dev/ruby/ironruby/r127/tests/ironruby/u > > til/and (Errno::ENOENT) > > from > d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\actions\updated > e > l > egates.generated.cs:34:in `Update2' > > from > d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\runtime\ScriptC > o > d > e.cs:79:in `Run' > > from > d:\dev\ruby\ironruby\r127\src\microsoft.scripting.core\sourceunit.cs:2 > 4 > 6 > :in `ExecuteProgram' > > rake aborted! > > Command failed with status (1): [d:/ruby/bin/ruby.exe > d:/dev/ruby/ironruby/...] > > D:/dev/ruby/ironruby/r127/rakefile:409:in `invoke_mspec' > > (See full trace by running task with --trace) > > > > I guess there is something inside Rakefile or nearby that is breaking > this? Unfortunately I can't get onto the SVN repository to search the > log history to see what changed (Bring on Git, I say). > > > > Any ideas? > > Cheers > > Pete _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core _______________________________________________ Ironruby-core mailing list Ironruby-core at rubyforge.org http://rubyforge.org/mailman/listinfo/ironruby-core From ivan at flanders.co.nz Wed Jul 30 16:01:01 2008 From: ivan at flanders.co.nz (Ivan Porto Carrero) Date: Wed, 30 Jul 2008 22:01:01 +0200 Subject: [Ironruby-core] Sails, WPF/Silverlight the rails way Message-ID: Hi all, I worked on a way to do wpf/development the rails way. It's heavily inspired by rucola and its interaction with interface builder. I created and will finish this framework when there is more possible with databinding between DLR languages and XAML. you can find it on github. http://github.com/casualjim/sails/tree/master If people think this is a great idea and want to pitch in all help is welcome of course :) If you hate it please let me know and why. The readme: Sails > ===== > > Sails is a framework inspired by the Rails and rucola frameworks. It offers > a rails-like way of developing applications > with IronRuby and Windows Presentation Foundation (WPF). > This framework uses the pattern Model - ViewModel - View - Controller > (M-VM-V-C). It should be able to run on both WPF > and Silverlight. > The idea is that the views can be created using a design tool like Blend > for example and just save that xaml as is. The > designer should not need to use anything else than drag and drop to create > a GUI design. The behaviors are then added to > the view by using predefined behaviors in the framework or by defining your > own behavior. > The framework then generates a proxy for the view which will be used to > transparently wire up the commands in the > behaviors to controller actions. > > You should be able to define a controller a little like this (doesn't work > yet and will most likely change in the > future): > > class MyController < Sails::Controllers::Base > > view_object :some_model, :refresh => :refresh_some_model, > :refresh_interval => 2.minutes > > view_action :some_action, :target => :my_button, :action => > :some_action_implementation > > def refresh_some_model > # code here > end > > def some_action_implementation > # code here > end > > end > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Tomas.Matousek at microsoft.com Thu Jul 31 13:34:41 2008 From: Tomas.Matousek at microsoft.com (Tomas Matousek) Date: Thu, 31 Jul 2008 10:34:41 -0700 Subject: [Ironruby-core] Code Review: MOX4 Message-ID: tfpt review "/shelveset:MOX4;REDMOND\tomat" Next step towards meta-objects in Ruby. Ruby: - Removes dependency on RuleBuilder from all rule generators in Ruby. The generators use a MetaObjectBuilder that stores restriction expression, condition expression, result expression and temp variables defined in the expressions. MetaObjectBuilder can produce both RuleBuilder's Test and Target, and MetaObject's Expression and Restrictions. Currently it is called only to produce rules, but as soon as we have MetaObjects for all Ruby dynamic objects we can start producing MetaObjects w/o changes to the rule generators. DLR (Microsoft.Scripting): - Removes ArgumentKind.Block - it's not needed anymore. - Adds RuleBuilder.AddTemporary(). Allows to add variables on the rule after the tree is generated. This is especially useful for variables that are defined in test expression and used in target expression. Without this method it's not possible to generate a tree with such variables first and then create the rule. - Removes a dependency on RuleBilder from BindingTarget.MakeExpression. Only CodeContext expression is needed there. Tomas -------------- next part -------------- A non-text attachment was scrubbed... Name: MOX4.diff Type: application/octet-stream Size: 108659 bytes Desc: MOX4.diff URL: From sanxiyn at gmail.com Thu Jul 31 18:15:17 2008 From: sanxiyn at gmail.com (Sanghyeon Seo) Date: Fri, 1 Aug 2008 07:15:17 +0900 Subject: [Ironruby-core] Patch to find_executable In-Reply-To: <5b0248170807090926x9754a8et4c543c48b41db16@mail.gmail.com> References: <5b0248170807090926x9754a8et4c543c48b41db16@mail.gmail.com> Message-ID: <5b0248170807311515g7069958dx5935b42fb4cccd58@mail.gmail.com> 2008/7/10 Sanghyeon Seo : > IronRuby's Rakefile and context.rb calls > UserEnvironment.find_executable 'ruby.exe', > which is hopelessly wrong. (No 'ruby.exe' on Linux, etc.) I wrote a > patch to correct > this. > > http://sparcs.kaist.ac.kr/~tinuviel/download/IronRuby/patch-find-executable > > Please apply. Ping? -- Seo Sanghyeon