From william.full.moon at gmail.com Fri Dec 1 07:12:54 2006 From: william.full.moon at gmail.com (* William) Date: Fri, 1 Dec 2006 23:12:54 +1100 Subject: [Nitro] Stroustrup speaks -- words to consider In-Reply-To: <55c107bf0611300456w5f802979r4e1eeeff48df1ae7@mail.gmail.com> Message-ID: <021f01c71542$1e412880$0201a8c0@ghostgum> Aloha, Alas it is so. Naturally "most" companies implement several of these activities with a memo and never actually do something. Or as so often happens water down engineering with "cherry picking" side-stepping the difficult bits. Pointy haired folk (by definition) don't do extreme things like "Get your best people into Six Sigma. Then you can give them [stock] options." (Jack Welch) The connection with Stroustrup is that he's saying the same thing I'm saying now. 'Management' makes good processes and good tools like C++ or Ruby or Six Sigma, or TQM, yes even Democracy work. I once read a paper form the Netherlands by a fellow called Jootsen, don't recall when it was written (apologies to Jootsen). When I too some percentages from the numbers it turned out that about 60% of improvement projects did not complete or not succeed. I'm not surprised that ".. companies that used six sigma have trailed the S&P". Some of you may have been to training courses. Only to return to work and have people "resist" doing thing differently, implementing the very training your employer just paid too much money for. May be not. In the case of something like TQM, Six Sigma, or BPR, etc you need to look at the places that really committed to the changes. Say like Japan, 3M, GE, Harley Davison. If I want to win a race, do I consider all the runners and what they do? Or is it better to look at the top three first and beat their results?! Wikipedia ... HYPERLINK "http://en.wikipedia.org/wiki/Six_sigma"http://en.wikipedia.org/wiki/Six_sig ma Recent reports HYPERLINK "BLOCKED::http://ct.bnet.com/clicks?t=20548487-0a60b3c4a458fa58d234fc1bee5c5 150-bf&s=5&fs=0"Six Sigma Still Pays Off At Motorola It may surprise those who have come to know Motorola for its cool cell phones, but the company's more lasting contribution to the world is something decidedly more wonkish: the quality-improvement process called Six Sigma. Source: HYPERLINK "BLOCKED::http://ct.bnet.com/clicks?t=20548488-0a60b3c4a458fa58d234fc1bee5c5 150-bf&s=5&fs=0"BusinessWeek Online I've seen some very cool stuff written in C++, and about 9 x more rubbish code. I think the Dilbert quote can apply as well to C++ or anti-biotics if you look at the press on super-bugs these days. Cheers all, Will. _____ From: Dimitri Aivaliotis [mailto:aglarond at gmail.com] Sent: Thursday, 30 November 2006 23:57 To: william.full.moon at gmail.com; General discussion about Nitro Subject: Re: [Nitro] FW: Stroustrup speaks -- words to consider Importance: Low On 11/30/06, * William wrote: I was reading one of those things about six sigma ... in lay terms six sigma looks for 99.99% correctness. HYPERLINK "http://www.dilbert.com/comics/dilbert/archive/dilbert-20061126.html"six sigma[1], you say? Sorry, couldn't resist. :) - Dimitri [1] HYPERLINK "http://www.dilbert.com/comics/dilbert/archive/dilbert-20061126.html"http:// www.dilbert.com/comics/dilbert/archive/dilbert-20061126.html -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.430 / Virus Database: 268.14.19/556 - Release Date: 28-Nov-2006 15:22 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061201/6682b0ef/attachment.html From noe.rubinstein at gmail.com Sat Dec 2 05:35:56 2006 From: noe.rubinstein at gmail.com (=?UTF-8?Q?No=C3=A9_Rubinstein?=) Date: Sat, 2 Dec 2006 11:35:56 +0100 Subject: [Nitro] where is the documentation? In-Reply-To: <20061130153604.GB15604@oxyliquit.de> References: <017c01c71470$dd915390$0201a8c0@ghostgum> <20061130114928.GA14218@oxyliquit.de> <20061130153604.GB15604@oxyliquit.de> Message-ID: mmmh, is it possible to get the screencast in a non-swf format ? If not, is there a screencast "scenario" available, for modifying it for 0.41 and trying to remake a screencast ? I would maybe take the time to do that. No? Rubinstein. 2006/11/30, Jonathan Buch : > Hi, > > > As a matter of fact, I can't find the videos! > > Thanks to Reid Thompson I now have the three videos. > > For the record, the three screencasts from George will probably only > work with Nitro 0.31 so I tagged the tip accordingly. > > URL again: http://oxyliquit.de/tip/35 > > If you know of more screencasts or just did one yourself yesterday, > please tell me! :P > > Jo > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > From john at oxyliquit.de Sat Dec 2 07:27:36 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Sat, 02 Dec 2006 13:27:36 +0100 Subject: [Nitro] where is the documentation? In-Reply-To: References: <017c01c71470$dd915390$0201a8c0@ghostgum> <20061130114928.GA14218@oxyliquit.de> <20061130153604.GB15604@oxyliquit.de> Message-ID: Hi, > mmmh, is it possible to get the screencast in a non-swf format ? If > not, is there a screencast "scenario" available, for modifying it for > 0.41 and trying to remake a screencast ? > I would maybe take the time to do that. Maybe one could try to 'extract' the video from the flash file? Don't know if that works. :P I find the first video from George highly informative for newbies. And it is even almost 0.40 ready. I just watched it and found only 1 thing which is different in 0.40. That is: `form_for()`, for which the information about the 'translation' is here: http://oxyliquit.de/question/84 . Of course there might be more areas (maybe the admin stuff, which I never used) which could differ. So if you find the time to recreate that video using Nitro 0.40 this would be highly appriciated. Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From james.britt at gmail.com Sat Dec 2 10:47:14 2006 From: james.britt at gmail.com (James Britt) Date: Sat, 02 Dec 2006 07:47:14 -0800 Subject: [Nitro] where is the documentation? In-Reply-To: References: <017c01c71470$dd915390$0201a8c0@ghostgum> <20061130114928.GA14218@oxyliquit.de> <20061130153604.GB15604@oxyliquit.de> Message-ID: <4571A002.20803@gmail.com> Jonathan Buch wrote: > Hi, > >> mmmh, is it possible to get the screencast in a non-swf format ? If >> not, is there a screencast "scenario" available, for modifying it for >> 0.41 and trying to remake a screencast ? >> I would maybe take the time to do that. > > Maybe one could try to 'extract' the video from the flash file? Don't > know if that works. :P VLC (VideoLAN) plays Flash videos. It will also stream whatever it can play, and it can stream to a file, in one of several formats. So it works as a trans coder. -- James Britt "People want simple stories." From noe.rubinstein at gmail.com Sat Dec 2 12:57:30 2006 From: noe.rubinstein at gmail.com (=?UTF-8?Q?No=C3=A9_Rubinstein?=) Date: Sat, 2 Dec 2006 18:57:30 +0100 Subject: [Nitro] [Patch] easy use without run.rb Message-ID: coi rodo I thought it would be better than, if no run.rb (nor any application_file) is found, Nitro just set the templates dir to "." and ran Nitro.start, possibly generating a run.rb. This makes ultra-easy the use of Nitro like a PHP-and-co. And "hello world" is even simpler (echo hello world > index.html ; nitro ) Here is a quick patch. Maybe not the better way to implement it. -------------- next part -------------- A non-text attachment was scrubbed... Name: patch.diff Type: text/x-diff Size: 764 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061202/9a633e57/attachment.bin From noe.rubinstein at gmail.com Sat Dec 2 13:34:46 2006 From: noe.rubinstein at gmail.com (=?UTF-8?Q?No=C3=A9_Rubinstein?=) Date: Sat, 2 Dec 2006 19:34:46 +0100 Subject: [Nitro] [Patch] easy use without run.rb In-Reply-To: References: Message-ID: Sorry. Maybe a darcs-generated patch is better ? Here it is. No? Rubinstein 2006/12/2, No? Rubinstein : > coi rodo > > I thought it would be better than, if no run.rb (nor any > application_file) is found, Nitro just set the templates dir to "." > and ran Nitro.start, possibly generating a run.rb. > This makes ultra-easy the use of Nitro like a PHP-and-co. And "hello > world" is even simpler (echo hello world > index.html ; nitro ) > > Here is a quick patch. Maybe not the better way to implement it. > > > -------------- next part -------------- A non-text attachment was scrubbed... Name: patch-generated-by-darcs.diff Type: text/x-diff Size: 858 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061202/badafd48/attachment.bin From george.moschovitis at gmail.com Sun Dec 3 04:38:21 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sun, 3 Dec 2006 11:38:21 +0200 Subject: [Nitro] [Patch] easy use without run.rb In-Reply-To: References: Message-ID: thanks, -g. On 12/2/06, No? Rubinstein wrote: > Sorry. > Maybe a darcs-generated patch is better ? > Here it is. > > No? Rubinstein > > 2006/12/2, No? Rubinstein : > > coi rodo > > > > I thought it would be better than, if no run.rb (nor any > > application_file) is found, Nitro just set the templates dir to "." > > and ran Nitro.start, possibly generating a run.rb. > > This makes ultra-easy the use of Nitro like a PHP-and-co. And "hello > > world" is even simpler (echo hello world > index.html ; nitro ) > > > > Here is a quick patch. Maybe not the better way to implement it. > > > > > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > > -- http://blog.gmosx.com http://nitroproject.org From george.moschovitis at gmail.com Sun Dec 3 04:39:25 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sun, 3 Dec 2006 11:39:25 +0200 Subject: [Nitro] [Patch] easy use without run.rb In-Reply-To: References: Message-ID: better do this: darcs send -o bundle tar cvfj bundle.tgz bundle and email the bundle.tgz file to the list (add [PATCH] to the subject of the email) thanks, George. On 12/2/06, No? Rubinstein wrote: > Sorry. > Maybe a darcs-generated patch is better ? > Here it is. > > No? Rubinstein > > 2006/12/2, No? Rubinstein : > > coi rodo > > > > I thought it would be better than, if no run.rb (nor any > > application_file) is found, Nitro just set the templates dir to "." > > and ran Nitro.start, possibly generating a run.rb. > > This makes ultra-easy the use of Nitro like a PHP-and-co. And "hello > > world" is even simpler (echo hello world > index.html ; nitro ) > > > > Here is a quick patch. Maybe not the better way to implement it. > > > > > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > > -- http://blog.gmosx.com http://nitroproject.org From noe.rubinstein at gmail.com Sun Dec 3 06:12:38 2006 From: noe.rubinstein at gmail.com (=?UTF-8?Q?No=C3=A9_Rubinstein?=) Date: Sun, 3 Dec 2006 12:12:38 +0100 Subject: [Nitro] [PATCH] easy use without run.rb Message-ID: This time the file if good :^p Truly sorry for these problems, next time I'll read a little more doc before sending insanities ;) Thanks for the explanation. (this is a patch wich makes nitro generate, when no run.rb nor any application_file is found, a run.rb setting the templates root to '.' so Nitro can be ultra-easily used like PHP-and-co. This makes the first web page with nitro very easy : "echo hello world > index.xhtml ; nitro". However, this is maybe not the better way to implement it.) No? Rubinstein -------------- next part -------------- A non-text attachment was scrubbed... Name: easy_wo_run_rb.patch.tar.gz Type: application/x-gzip Size: 1850 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061203/67818fbb/attachment.gz From john at oxyliquit.de Mon Dec 4 07:32:47 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Mon, 04 Dec 2006 13:32:47 +0100 Subject: [Nitro] [PATCH] Important DoS vulnerability fix in cgi.rb Message-ID: Hi, attached patch fixes this. Also, minor testcase fix for tc_cgi.rb so it doesn't fail anymore. If you're paranoid, please apply this patch at once to feel cozy. :P Additional notes from me: The vulnerability is again in the multipart parsing. This wasn't mentioned in the article below. The multipart parsing didn't escape the boundaries of the parts and so any regex could be used there. An additional check for validating the end part of the multipart also has been added. Jo If you haven't heard already: ------------------------------------------------------------------------ Another vulnerability has been discovered in the CGI library (cgi.rb) that ships with Ruby which could be used by a malicious user to create a denial of service attack (DoS). This vulnerability is open to the public as JVN#84798830. Please note that the previous patch () does not fix this problem. Impact ------ A specific HTTP request for any web application using cgi.rb causes CPU consumption on the machine on which the web application is running. Many such requests result in a denial of service. Vulnerable versions ------------------- * 1.8 series 1.8.5 and all prior versions * Development version (1.9 series) All versions before 2006-12-04 Solution -------- * 1.8 series Please upgrade to 1.8.5-p2. (4519151 bytes, md5sum: a3517a224716f79b14196adda3e88057) Please note that a package that corrects this weakness may already be available through your package management software. * Development version (1.9 series) Please update your Ruby to a version after 2006-12-04. Article --------- http://www.ruby-lang.org/en/news/2006/12/04/another-dos-vulnerability-in-cgi-library/ -- Feel the love http://pinkjuice.com/pics/ruby.png -------------- next part -------------- A non-text attachment was scrubbed... Name: dosfix2_cgi.patch.tar.bz2 Type: application/bzip2 Size: 14123 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061204/df2ecfeb/attachment-0002.bin -------------- next part -------------- A non-text attachment was scrubbed... Name: fix_tc_cgi.patch.tar.bz2 Type: application/bzip2 Size: 13770 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061204/df2ecfeb/attachment-0003.bin From george.moschovitis at gmail.com Mon Dec 4 15:23:43 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Mon, 4 Dec 2006 22:23:43 +0200 Subject: [Nitro] [PATCH] Important DoS vulnerability fix in cgi.rb In-Reply-To: References: Message-ID: Argh, this is nasty! thanks, -g. On 12/4/06, Jonathan Buch wrote: > Hi, > > attached patch fixes this. > > Also, minor testcase fix for tc_cgi.rb so it doesn't fail anymore. > > If you're paranoid, please apply this patch at once to feel cozy. :P > > Additional notes from me: > > The vulnerability is again in the multipart parsing. This wasn't > mentioned in the article below. > The multipart parsing didn't escape the boundaries of the parts and so > any regex could be used there. An additional check for validating the > end part of the multipart also has been added. > > Jo > > If you haven't heard already: > > ------------------------------------------------------------------------ > > Another vulnerability has been discovered in the CGI library (cgi.rb) > that ships with Ruby which could be used by a malicious user to create a > denial of service attack (DoS). > > This vulnerability is open to the public as JVN#84798830. > > Please note that the previous patch > () > does not fix this problem. > > Impact > ------ > A specific HTTP request for any web application using cgi.rb causes CPU > consumption on the machine on which the web application is running. Many > such requests result in a denial of service. > > Vulnerable versions > ------------------- > * 1.8 series > 1.8.5 and all prior versions > * Development version (1.9 series) > All versions before 2006-12-04 > > Solution > -------- > * 1.8 series > Please upgrade to 1.8.5-p2. > > (4519151 bytes, md5sum: a3517a224716f79b14196adda3e88057) > Please note that a package that corrects this weakness may already be > available through your package management software. > * Development version (1.9 series) > Please update your Ruby to a version after 2006-12-04. > > Article > --------- > http://www.ruby-lang.org/en/news/2006/12/04/another-dos-vulnerability-in-cgi-library/ > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > > -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From george.moschovitis at gmail.com Wed Dec 6 04:29:14 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Wed, 6 Dec 2006 11:29:14 +0200 Subject: [Nitro] nitro error pages Message-ID: Dear devs, I would like to improve how nitro handles error pages expecially in live mode. Anyone has any related ideas suggestions that would like to share before I start? -g. -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From john at oxyliquit.de Wed Dec 6 06:49:14 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 06 Dec 2006 12:49:14 +0100 Subject: [Nitro] nitro error pages In-Reply-To: References: Message-ID: Hi, > I would like to improve how nitro handles error pages especially in > live mode. Anyone has any related ideas suggestions that would like to > share before I start? Yes, two slightly unrelated suggestions: * please write something about caching * please release 0.41 bugfix release before putting your mind to new cool stuff. :) I think there was a thread on error handling before.. Ah yeah, how it's done in IOWA, very neat and flexible approach: http://rubyforge.org/pipermail/nitro-general/2006-October/006157.html This probably would be like a lookup table, if one of the rules matches the thrown exception. If it matches, it uses the given (maybe custom) handler. That handler would probably have to be given the occured exception. The handler might be a class responding to .render_error. I'm not sure if the error handling has to be so fine-grained that controllers can own a custom..... But that certanly could be done quite easily... render.rb, Nitro::Render#render rescue Object => ex if handler = lookup_error_handler(@controller, ex) handler.render_error(ex) else log_error(e1, path, false) print '(error)' end end def lookup_error_handler(controller, ex) if controller has custom error_handler controller.get_error_handler(ex) # custom else Nitro::Render.get_error_handler(ex) # global `setting` end end class CustomErrorHandler def handles_error?(ex); ex.is_a?(ActionError); end def handle_error(ex) raise ArgumentError unless handles_error?(ex) redirect_to '/argument_error_page.html' end end class Nitro::Render setting :error_handlers, :default => [RenderExit => SwallowErrorHandler] end class MyController def self.error_handlers [ArgumentError => CustomErrorHandler] end end or something similar along those lines. I like the idea and it's pretty flexible and easy to implement. What'cha think? Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From george.moschovitis at gmail.com Wed Dec 6 07:07:51 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Wed, 6 Dec 2006 14:07:51 +0200 Subject: [Nitro] nitro error pages In-Reply-To: References: Message-ID: > * please write something about caching please remind me the oxyliquit link for the question. > * please release 0.41 bugfix release ok will do :) -g. -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From john at oxyliquit.de Wed Dec 6 07:25:09 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 06 Dec 2006 13:25:09 +0100 Subject: [Nitro] nitro error pages In-Reply-To: References: Message-ID: Hi, >> * please write something about caching > > please remind me the oxyliquit link for the question. Aye, here you go. http://oxyliquit.de/question/17 >> * please release 0.41 bugfix release > > ok will do :) Thx! ^_^ Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From john at oxyliquit.de Wed Dec 6 07:30:55 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 06 Dec 2006 13:30:55 +0100 Subject: [Nitro] nitro error pages In-Reply-To: References: Message-ID: Hi, OT, but..... have you realized how FAST the ML is again? It's like .... sw3333333t. ^___^ Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From john at oxyliquit.de Wed Dec 6 09:10:10 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 06 Dec 2006 15:10:10 +0100 Subject: [Nitro] [BUG] Can't create new member on nitroproject.org Message-ID: George, [15:03] membership creation for the wiki is erroring out My bro said this was mentioned already, no luck finding the bug? Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From george.moschovitis at gmail.com Wed Dec 6 09:20:17 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Wed, 6 Dec 2006 16:20:17 +0200 Subject: [Nitro] [BUG] Can't create new member on nitroproject.org In-Reply-To: References: Message-ID: let me work a bit on getting the 0.41 release out, then I will try to give the np.org site some lovin. -g. On 12/6/06, Jonathan Buch wrote: > George, > > [15:03] membership creation for the wiki is erroring out > > My bro said this was mentioned already, no luck finding the bug? > > Jo > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From lpruszynski at gmail.com Wed Dec 6 10:57:34 2006 From: lpruszynski at gmail.com (Les Pruszynski) Date: Wed, 6 Dec 2006 15:57:34 +0000 Subject: [Nitro] [BUG] Can't create new member on nitroproject.org Message-ID: <404FF377-69ED-4B4E-8579-36AFB130A067@gmail.com> I was able to subscribe after 3rd attempt. Great to hear that 0.41 is on the way. By the way, can I download nitro from the repository? I am unable to locate it. Also will we see more documentation with the next release? Many thanks for your hard work, Les From john at oxyliquit.de Wed Dec 6 11:26:16 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 06 Dec 2006 17:26:16 +0100 Subject: [Nitro] [BUG] Can't create new member on nitroproject.org In-Reply-To: <404FF377-69ED-4B4E-8579-36AFB130A067@gmail.com> References: <404FF377-69ED-4B4E-8579-36AFB130A067@gmail.com> Message-ID: Hi, > I was able to subscribe after 3rd attempt. glad you could join! :P > By the way, can I download nitro from the repository? I am unable to > locate it. The nitro repository is located on http://repo.nitroproject.org On how to use darcs for nitro development: http://oxyliquit.de/tip/18 > Also will we see more documentation with the next release? The next release is a pure bugfix/security release. The amount of documentation is quite unrelated to the release of any nitro version anyhow. :P Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From lasso at lassoweb.se Wed Dec 6 12:23:22 2006 From: lasso at lassoweb.se (Lars Olsson) Date: Wed, 06 Dec 2006 18:23:22 +0100 Subject: [Nitro] Redirecting to error action/method Message-ID: <4576FC8A.6050100@lassoweb.se> Hi list! In Nitro 0.31 and earlier all errors raised was automatically redirected to the error action/method . In 0.40, this doesn't work. My error method never get called, some kind of default error page appears instead. Any hints on how to fix this? Sincerely /lasso -- ________________________________________ Lars Olsson lasso at lassoweb.se http://www.lassoweb.se/ From fabian at fabian-buch.de Wed Dec 6 12:46:17 2006 From: fabian at fabian-buch.de (Fabian Buch) Date: Wed, 6 Dec 2006 18:46:17 +0100 Subject: [Nitro] [ANN] LoxParts - sharing Nitro parts Message-ID: <6D6076BE-0434-4132-A691-1F27E78E2208@fabian-buch.de> hi all I just released a beta of LoxParts.de (http://loxparts.de/). In other systems you'd maybe call them Plug-ins, but in Nitro they are called "parts". These parts can be dropped into your Nitro Web- Application and extend your Application with fully functional app parts. You could for example add a Blog to your current Application by adding a Blog Part. Lox Parts provides such Parts that aren't in Nitro's standard distribution. And registered user can add their own parts. Just written a CMS in Nitro? Why not share it as part -> http://loxparts.de The Nitro community didn't decide for a name of small, non-part extensions (plug-ins). Once they were called "components" in a discussion. These can be shared via LoxParts too, of course. How to make such Parts, that can be shared on LoxParts, can be read in a tip I created recently on Oxyliquit: http://oxyliquit.de/tip/36 LoxParts is still in "beta", so keep that in mind when using it and please tell me what you think, featurerequest and bugreports. Have fun, Fabian -- Nitro Q&A: http://oxyliquit.de LoxParts: http://loxparts.de Blog: http://blog.fabian-buch.de From john at oxyliquit.de Wed Dec 6 13:00:23 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 06 Dec 2006 19:00:23 +0100 Subject: [Nitro] Redirecting to error action/method In-Reply-To: <4576FC8A.6050100@lassoweb.se> References: <4576FC8A.6050100@lassoweb.se> Message-ID: Hi, > In Nitro 0.31 and earlier all errors raised was automatically redirected > to the error action/method . In 0.40, this doesn't work. My error method > never get called, some kind of default error page appears instead. Any > hints on how to fix this? This still happens with Oxy which also has a custom error page. I don't have an error action, only a `error.xhtml` in my template root dir though. Maybe that does the trick? Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From lasso at lassoweb.se Wed Dec 6 13:44:27 2006 From: lasso at lassoweb.se (Lars Olsson) Date: Wed, 06 Dec 2006 19:44:27 +0100 Subject: [Nitro] Redirecting to error action/method In-Reply-To: References: <4576FC8A.6050100@lassoweb.se> Message-ID: <45770F8B.5080206@lassoweb.se> Thanks Jonathan! It was actually a coding error on my part that caused the strange effect I saw. WHY WHY WHY is Nitro::NoActionError renamed to Nitro::ActionError WITHOUT A BIG SIGN declaring the change in the release notes??? Now I hade to go through the darcs changelog to find out...Important API changes need to be documented better!!! /lasso Jonathan Buch skrev: > Hi, > >> In Nitro 0.31 and earlier all errors raised was automatically redirected >> to the error action/method . In 0.40, this doesn't work. My error method >> never get called, some kind of default error page appears instead. Any >> hints on how to fix this? > > This still happens with Oxy which also has a custom error page. > > I don't have an error action, only a `error.xhtml` in my template root dir > though. Maybe that does the trick? > > Jo From fabian at fabian-buch.de Wed Dec 6 13:54:21 2006 From: fabian at fabian-buch.de (Fabian Buch) Date: Wed, 6 Dec 2006 19:54:21 +0100 Subject: [Nitro] [ANN] LoxAuth released to public Message-ID: <2E620CBE-81BB-47FD-B659-AC8709EA3A44@fabian-buch.de> LoxAuth 0.6.1 is the first Part released on LoxParts (http:// loxparts.de/part/LoxAuth). It's the Auth system behind Oxyliquit and also LoxParts itself. How LoxAuth works can be read in its README or on Oxyliquit Tip 37 (http://oxyliquit.de/tip/37). Fabian -- Nitro Q&A: http://oxyliquit.de LoxParts: http://loxparts.de Blog: http://blog.fabian-buch.de From localuser at gmail.com Wed Dec 6 14:07:39 2006 From: localuser at gmail.com (Ed Chang) Date: Wed, 6 Dec 2006 13:07:39 -0600 Subject: [Nitro] [BUG] Can't create new member on nitroproject.org In-Reply-To: References: Message-ID: <106a0270612061107s5c8aaebseba44a77dcc00c90@mail.gmail.com> I had problems registering at nitroproject.org because my email address had a non alpha-numeric character that was not url encoded. For example, I registered as a new Nitro Project user with an email like: somedumbguy+nitroproject at gmail.com I then received a verification email with the following hyperlink to complete the registration process: http://www.nitroproject.org/users/create?t=NIIJ8KUhkSeyw;e=somedumbguy+nitroproject at gmail.com If I look at the URL in the web browser it shows that I'm currently viewing a web page at: http://www.nitroproject.org/users/create?t=NIIJ8KUhkSeyw;e=somedumbguy+nitroproject at gmail.com However, the form input fields show the following: I'm guessing that the verification token and email do not match and the frustrated user is directed to a stark error page. Hope that help, Ed On 12/6/06, George Moschovitis wrote: > let me work a bit on getting the 0.41 release out, then I will try to > give the np.org site some lovin. > > -g. > > On 12/6/06, Jonathan Buch wrote: > > George, > > > > [15:03] membership creation for the wiki is erroring out > > > > My bro said this was mentioned already, no luck finding the bug? > > > > Jo > > > > -- > > Feel the love > > http://pinkjuice.com/pics/ruby.png > > _______________________________________________ > > Nitro-general mailing list > > Nitro-general at rubyforge.org > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > -- > http://cull.gr > http://www.joy.gr > http://blog.gmosx.com > http://nitroproject.org > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > From john at oxyliquit.de Wed Dec 6 14:08:30 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 06 Dec 2006 20:08:30 +0100 Subject: [Nitro] Redirecting to error action/method In-Reply-To: <45770F8B.5080206@lassoweb.se> References: <4576FC8A.6050100@lassoweb.se> <45770F8B.5080206@lassoweb.se> Message-ID: Hi, > Thanks Jonathan! It was actually a coding error on my part that caused > the strange effect I saw. > > > WHY WHY WHY is Nitro::NoActionError renamed to Nitro::ActionError > WITHOUT A BIG SIGN declaring the change in the release notes??? Now I > hade to go through the darcs changelog to find out...Important API > changes need to be documented better!!! > Ayyy... I feel partly responsible for that I guess. ^^; But actually neither George nor I had forseen this internal error be used outside of dispatcher.rb and render.rb I guess.... Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From lasso at lassoweb.se Wed Dec 6 14:42:46 2006 From: lasso at lassoweb.se (Lars Olsson) Date: Wed, 06 Dec 2006 20:42:46 +0100 Subject: [Nitro] Redirecting to error action/method In-Reply-To: References: <4576FC8A.6050100@lassoweb.se> <45770F8B.5080206@lassoweb.se> Message-ID: <45771D36.3010503@lassoweb.se> Hi! I actually use it in two kinds of situations: 1. Custom error handler def error error, path = @context.rendering_errors[0] if error.kind_of?(Nitro::ActionError) # Tell user page doesn't exist else # Tell user an internal error occurred # Additional actions (logging etc) end end 2. A simple way to "hide" pages from unauthorized users def some_action if some_condition # Allow user to see page else raise Nitro::ActionError.new end end Kindly /lasso Jonathan Buch skrev: > Hi, > >> Thanks Jonathan! It was actually a coding error on my part that caused >> the strange effect I saw. >> >> >> WHY WHY WHY is Nitro::NoActionError renamed to Nitro::ActionError >> WITHOUT A BIG SIGN declaring the change in the release notes??? Now I >> hade to go through the darcs changelog to find out...Important API >> changes need to be documented better!!! >> > > Ayyy... I feel partly responsible for that I guess. ^^; > > But actually neither George nor I had forseen this internal error be > used outside of dispatcher.rb and render.rb I guess.... > > Jo From john at oxyliquit.de Wed Dec 6 15:07:45 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 06 Dec 2006 21:07:45 +0100 Subject: [Nitro] [BUG] Can't create new member on nitroproject.org In-Reply-To: <106a0270612061107s5c8aaebseba44a77dcc00c90@mail.gmail.com> References: <106a0270612061107s5c8aaebseba44a77dcc00c90@mail.gmail.com> Message-ID: Hi. > http://www.nitroproject.org/users/create?t=NIIJ8KUhkSeyw;e=somedumbguy+nitroproject at gmail.com > Oooh.. this is valuable information, for Oxyliqut as well. The Cgi parsing of parameters unescapes '+' to ' ' for (historical?) reasons. So one better escapes that stuff in case someone uses 'extension' addresses with + .... Anyway, thx for that info. :) Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From john at oxyliquit.de Wed Dec 6 15:13:57 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 06 Dec 2006 21:13:57 +0100 Subject: [Nitro] Redirecting to error action/method In-Reply-To: <45771D36.3010503@lassoweb.se> References: <4576FC8A.6050100@lassoweb.se> <45770F8B.5080206@lassoweb.se> <45771D36.3010503@lassoweb.se> Message-ID: Hi, > I actually use it in two kinds of situations: > > 1. Custom error handler > 2. A simple way to "hide" pages from unauthorized users I hope you hacks there is going to vanish when George gets around implementing my error handling idea. ;D Well, or at least something similar (I will pester George more about this, since I like the idea). :P Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From george.moschovitis at gmail.com Thu Dec 7 04:44:25 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 7 Dec 2006 11:44:25 +0200 Subject: [Nitro] [BUG] Can't create new member on nitroproject.org In-Reply-To: <404FF377-69ED-4B4E-8579-36AFB130A067@gmail.com> References: <404FF377-69ED-4B4E-8579-36AFB130A067@gmail.com> Message-ID: > Great to hear that 0.41 is on the way. I have updated the repo with an almost final version of 0.41.0. there are 2 test cases that don't pass. Jonathan could you also have a look at them? -g. -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From george.moschovitis at gmail.com Thu Dec 7 04:45:31 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 7 Dec 2006 11:45:31 +0200 Subject: [Nitro] [ANN] LoxParts - sharing Nitro parts In-Reply-To: <6D6076BE-0434-4132-A691-1F27E78E2208@fabian-buch.de> References: <6D6076BE-0434-4132-A691-1F27E78E2208@fabian-buch.de> Message-ID: thats nice :) congrats! -g. On 12/6/06, Fabian Buch wrote: > hi all > > I just released a beta of LoxParts.de (http://loxparts.de/). > > In other systems you'd maybe call them Plug-ins, but in Nitro they > are called "parts". These parts can be dropped into your Nitro Web- > Application and extend your Application with fully functional app > parts. You could for example add a Blog to your current Application > by adding a Blog Part. > > Lox Parts provides such Parts that aren't in Nitro's standard > distribution. And registered user can add their own parts. Just > written a CMS in Nitro? Why not share it as part -> http://loxparts.de > > The Nitro community didn't decide for a name of small, non-part > extensions (plug-ins). Once they were called "components" in a > discussion. These can be shared via LoxParts too, of course. > > How to make such Parts, that can be shared on LoxParts, can be read > in a tip I created recently on Oxyliquit: http://oxyliquit.de/tip/36 > > LoxParts is still in "beta", so keep that in mind when using it and > please tell me what you think, featurerequest and bugreports. > > Have fun, > Fabian > > > -- > Nitro Q&A: http://oxyliquit.de > LoxParts: http://loxparts.de > Blog: http://blog.fabian-buch.de > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From george.moschovitis at gmail.com Thu Dec 7 04:55:25 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 7 Dec 2006 11:55:25 +0200 Subject: [Nitro] Redirecting to error action/method In-Reply-To: References: <4576FC8A.6050100@lassoweb.se> <45770F8B.5080206@lassoweb.se> <45771D36.3010503@lassoweb.se> Message-ID: Will do it for 0.42.0, I need this feature urgently for my own projects too :) -g. On 12/6/06, Jonathan Buch wrote: > Hi, > > > I actually use it in two kinds of situations: > > > > 1. Custom error handler > > 2. A simple way to "hide" pages from unauthorized users > > I hope you hacks there is going to vanish when George gets around > implementing my error handling idea. ;D > Well, or at least something similar (I will pester George more about > this, since I like the idea). :P > > Jo > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From john at oxyliquit.de Thu Dec 7 05:58:11 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Thu, 7 Dec 2006 11:58:11 +0100 Subject: [Nitro] Redirecting to error action/method In-Reply-To: References: <4576FC8A.6050100@lassoweb.se> <45770F8B.5080206@lassoweb.se> <45771D36.3010503@lassoweb.se> Message-ID: <20061207105811.GA25931@oxyliquit.de> Hi, > Will do it for 0.42.0, I need this feature urgently for my own projects too :) lol, don't feel pressured, feel free to think of another way to do this. ;) Jo From fabian at fabian-buch.de Thu Dec 7 08:58:52 2006 From: fabian at fabian-buch.de (Fabian Buch) Date: Thu, 7 Dec 2006 14:58:52 +0100 Subject: [Nitro] [ANN] LoxParts - sharing Nitro parts In-Reply-To: References: <6D6076BE-0434-4132-A691-1F27E78E2208@fabian-buch.de> Message-ID: how about making Nitroproject.org's blog a little livelier and post a small ann there too? ;) Fabian -- Nitro Q&A: http://oxyliquit.de LoxParts: http://loxparts.de Blog: http://blog.fabian-buch.de From george.moschovitis at gmail.com Thu Dec 7 11:08:45 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 7 Dec 2006 18:08:45 +0200 Subject: [Nitro] [ANN] LoxParts - sharing Nitro parts In-Reply-To: References: <6D6076BE-0434-4132-A691-1F27E78E2208@fabian-buch.de> Message-ID: will do ;-) -g. On 12/7/06, Fabian Buch wrote: > how about making Nitroproject.org's blog a little livelier and post a > small ann there too? ;) > > Fabian > > > -- > Nitro Q&A: http://oxyliquit.de > LoxParts: http://loxparts.de > Blog: http://blog.fabian-buch.de > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From john at oxyliquit.de Thu Dec 7 13:25:25 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Thu, 07 Dec 2006 19:25:25 +0100 Subject: [Nitro] [PATCH] important (and not so much) bugfixes Message-ID: Hi, * taggable patch seemed not to be in * speed patch (halves startup time of nitro) * mysql transactions (don't seem to hurt MyISAM) * logging change (George: pleaaaseeee, with sugar on top ;)) * readme from og still claimed to be able to speak Oracle * STI lookup for classes was broken (with tc attached) Enjoy! George, will have a look at the failing tcs, not sure which ones, since my Open4 seems to stall some times and you removed the Reapfiles so I can't run the testcases properly. Will look for another way to run them. Jo Fri Nov 17 11:23:46 CET 2006 Jonathan Buch * Fix bug in taggable, add Tag#tagged module Taggable many_to_many Tag end class User; is Taggable; end class Category; is Taggable; end those two classes share the _SAME_ many_to_many relation object. Worked around that by specifying it in Taggable.included. TODO: check for more possible errors of that kind. Wed Nov 22 13:26:24 CET 2006 Jonathan Buch * speed patch for attributes.rb (2 times faster nitro startup) Tue Nov 28 19:01:02 CET 2006 Jonathan Buch * Add mysql transactions (only does something with InnoDB tables) Wed Dec 6 18:04:14 CET 2006 Jonathan Buch * Change logging configuration use NITRO_MODE for Configuration.mode, CONFIGURATION_mode is deprecated. use NITRO_LOG env for specifying the log destination. Thu Dec 7 19:15:39 CET 2006 Jonathan Buch * README og fix, STI fix, STI tc addition -- Feel the love http://pinkjuice.com/pics/ruby.png -------------- next part -------------- A non-text attachment was scrubbed... Name: rdmeogfixsti.patch.tar.bz2 Type: application/bzip2 Size: 16215 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061207/4e669d4f/attachment.bin From lasso at lassoweb.se Thu Dec 7 13:49:15 2006 From: lasso at lassoweb.se (Lars Olsson) Date: Thu, 07 Dec 2006 19:49:15 +0100 Subject: [Nitro] action_name attribute Message-ID: <4578622B.20107@lassoweb.se> Hi! Just a quick question. Nitro::Context includes Nitro::Render which exposes the action_name (represents "The name of the currently executing action") attribute. However, this attribute always seems to return nil whenever I call it. I also cannot find any evidence that @context.action_name ever get touched (at least not in render.rb). Is it a bug? A feature? Or a misunderstanding on my part? Kindly /lasso -- ________________________________________ Lars Olsson lasso at lassoweb.se http://www.lassoweb.se/ From john at oxyliquit.de Thu Dec 7 14:57:05 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Thu, 07 Dec 2006 20:57:05 +0100 Subject: [Nitro] action_name attribute In-Reply-To: <4578622B.20107@lassoweb.se> References: <4578622B.20107@lassoweb.se> Message-ID: Hi, > Just a quick question. Nitro::Context includes Nitro::Render which > exposes the action_name (represents "The name of the currently executing > action") attribute. However, this attribute always seems to return nil > whenever I call it. I also cannot find any evidence that > @context.action_name ever get touched (at least not in render.rb). Is it > a bug? A feature? Or a misunderstanding on my part? Might be a bug. The compiler should set this attribute in compile_action() which it doesn't. Might be due to the parameter reorganizing. Try using the @action instance variable. George, can you verify that the documentation (first part) above compile_action() ist still like the implementation is meant to be? No need to hurry, this can go into 0.42, but it definitly should be revised. Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From john at oxyliquit.de Thu Dec 7 16:00:22 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Thu, 07 Dec 2006 22:00:22 +0100 Subject: [Nitro] [PATCH] Oracle adapter preview Message-ID: Hi, there have been 2 people asking for that on irc, might as well try to implement it. What works: * basic query/exec * creating/dropping tables * automatic sequence generation Note that this is by far not production ready, just something hacked together in a few hours. So, anyone else, who'd be using this besides the two people (who are just evaluating Nitro and Rails if they can use it)? Jo -- Feel the love http://pinkjuice.com/pics/ruby.png -------------- next part -------------- A non-text attachment was scrubbed... Name: oracle1.patch.tar.bz2 Type: application/bzip2 Size: 19105 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061207/9a701b92/attachment.bin From lasso at lassoweb.se Thu Dec 7 18:31:59 2006 From: lasso at lassoweb.se (Lars Olsson) Date: Fri, 08 Dec 2006 00:31:59 +0100 Subject: [Nitro] Nitro Configuration object Message-ID: <4578A46F.7000305@lassoweb.se> Last request for tonight... Could someone please explain to me how to use the Nitro::Configuration object. It is supposed to replace configuration options in Runner, Server etc, but I cannot find a clue on how it is used...Examples including adapter and mode settings would be most welcome :) Kindly /lasso -- ________________________________________ Lars Olsson lasso at lassoweb.se http://www.lassoweb.se/ From george.moschovitis at gmail.com Fri Dec 8 03:43:46 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Fri, 8 Dec 2006 10:43:46 +0200 Subject: [Nitro] [PATCH] important (and not so much) bugfixes In-Reply-To: References: Message-ID: > Reapfiles so I can't run the testcases properly. to run the tests form nitro dir: ruby script/test.rb (runs all tests) ruby script/test.rb og (runs og tests, you can use nitro, og, glue, gen here) -g. > Will look for another way to run them. > > Jo > > Fri Nov 17 11:23:46 CET 2006 Jonathan Buch > * Fix bug in taggable, add Tag#tagged > module Taggable > many_to_many Tag > end > class User; is Taggable; end > class Category; is Taggable; end > > those two classes share the _SAME_ many_to_many relation object. > > Worked around that by specifying it in Taggable.included. > > TODO: check for more possible errors of that kind. > > Wed Nov 22 13:26:24 CET 2006 Jonathan Buch > * speed patch for attributes.rb (2 times faster nitro startup) > > Tue Nov 28 19:01:02 CET 2006 Jonathan Buch > * Add mysql transactions (only does something with InnoDB tables) > > Wed Dec 6 18:04:14 CET 2006 Jonathan Buch > * Change logging configuration > use NITRO_MODE for Configuration.mode, CONFIGURATION_mode is deprecated. > use NITRO_LOG env for specifying the log destination. > > Thu Dec 7 19:15:39 CET 2006 Jonathan Buch > * README og fix, STI fix, STI tc addition > > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > > -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From george.moschovitis at gmail.com Fri Dec 8 03:47:57 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Fri, 8 Dec 2006 10:47:57 +0200 Subject: [Nitro] action_name attribute In-Reply-To: References: <4578622B.20107@lassoweb.se> Message-ID: this looks like a bug. strange though, I think I am using action name in some of my projects. -g. On 12/7/06, Jonathan Buch wrote: > Hi, > > > Just a quick question. Nitro::Context includes Nitro::Render which > > exposes the action_name (represents "The name of the currently executing > > action") attribute. However, this attribute always seems to return nil > > whenever I call it. I also cannot find any evidence that > > @context.action_name ever get touched (at least not in render.rb). Is it > > a bug? A feature? Or a misunderstanding on my part? > > Might be a bug. The compiler should set this attribute in compile_action() > which it doesn't. > Might be due to the parameter reorganizing. Try using the @action instance > variable. > > George, can you verify that the documentation (first part) above > compile_action() ist still like the implementation is meant to be? > > No need to hurry, this can go into 0.42, but it definitly should be revised. > > Jo > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From george.moschovitis at gmail.com Fri Dec 8 03:48:34 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Fri, 8 Dec 2006 10:48:34 +0200 Subject: [Nitro] [PATCH] Oracle adapter preview In-Reply-To: References: Message-ID: thanks for this skeleton :) -g. On 12/7/06, Jonathan Buch wrote: > Hi, > > there have been 2 people asking for that on irc, might as well > try to implement it. > > What works: > > * basic query/exec > * creating/dropping tables > * automatic sequence generation > > Note that this is by far not production ready, just something > hacked together in a few hours. > > So, anyone else, who'd be using this besides the two people (who > are just evaluating Nitro and Rails if they can use it)? > > Jo > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > > -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From john at oxyliquit.de Fri Dec 8 04:13:19 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Fri, 8 Dec 2006 10:13:19 +0100 Subject: [Nitro] [PATCH] important (and not so much) bugfixes In-Reply-To: References: Message-ID: <20061208091319.GA13611@oxyliquit.de> Hi, > to run the tests form nitro dir: > > ruby script/test.rb > (runs all tests) > > ruby script/test.rb og > (runs og tests, you can use nitro, og, glue, gen here) ah yes, I do know that. :) But my version of that file uses Open4 to present nicely formatted testcase results. I think this was done by manveru. It really looks nice, but Open4 has the tendency to stall sometimes on my computer... But anyway, I will have a look at the tcs today . Jo From john at oxyliquit.de Fri Dec 8 06:15:35 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Fri, 08 Dec 2006 12:15:35 +0100 Subject: [Nitro] [PATCH] important (and not so much) bugfixes In-Reply-To: References: Message-ID: Hi, attached tc fixes. The only testcase which failed was the tc_reldelete.rb one. It relied on my assumption, that relation item is updated automatically, which it isn't. This is the 'uglyness' introduced by not .saving a relation after adding it to the set/parent. Since I forgot about that I spent again quite some time tracking that down. :P And, no testcase is especially tailored to such situations, so we just get bitten by that. :P Reminds me, I forgot to answer on the 'please elaborate' questions on that topic when it came up the last time... Anyway, I just fixed the testcase, we should perhaps think about a more extensive way to test that for the next releases however. Jo -- Feel the love http://pinkjuice.com/pics/ruby.png -------------- next part -------------- A non-text attachment was scrubbed... Name: reldeltcfix.patch.tar.bz2 Type: application/bzip2 Size: 14555 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061208/16b57da9/attachment-0001.bin From john at oxyliquit.de Fri Dec 8 06:17:00 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Fri, 08 Dec 2006 12:17:00 +0100 Subject: [Nitro] [PATCH] important (and not so much) bugfixes In-Reply-To: References: Message-ID: Oh, yes, forgot to say: All testcases are green now with the last patch (and the taggable patch). Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From lasso at lassoweb.se Fri Dec 8 06:11:00 2006 From: lasso at lassoweb.se (Lars Olsson) Date: Fri, 8 Dec 2006 11:11:00 -0000 (UTC) Subject: [Nitro] action_name attribute In-Reply-To: References: <4578622B.20107@lassoweb.se> Message-ID: <44215.192.176.230.1.1165576260.squirrel@webmail.lassoweb.se> Hi! Yes, it's a bit strange. If I'm in an action (or template) self.action_name works, but @context.action_name doesn't. Poke, poke...How about some tips on the Nitro::Configuration object? Kindly /lasso On Fri, December 8, 2006 08:47, George Moschovitis wrote: > this looks like a bug. strange though, I think I am using action name in > some of my projects. > > -g. > > > On 12/7/06, Jonathan Buch wrote: > >> Hi, >> >> >>> Just a quick question. Nitro::Context includes Nitro::Render which >>> exposes the action_name (represents "The name of the currently >>> executing action") attribute. However, this attribute always seems to >>> return nil whenever I call it. I also cannot find any evidence that >>> @context.action_name ever get touched (at least not in render.rb). Is >>> it a bug? A feature? Or a misunderstanding on my part? >> >> Might be a bug. The compiler should set this attribute in >> compile_action() which it doesn't. Might be due to the parameter >> reorganizing. Try using the @action instance variable. >> >> George, can you verify that the documentation (first part) above >> compile_action() ist still like the implementation is meant to be? >> >> No need to hurry, this can go into 0.42, but it definitly should be >> revised. >> >> Jo >> >> >> -- >> Feel the love >> http://pinkjuice.com/pics/ruby.png >> _______________________________________________ >> Nitro-general mailing list >> Nitro-general at rubyforge.org >> http://rubyforge.org/mailman/listinfo/nitro-general >> >> > > > -- > http://cull.gr > http://www.joy.gr > http://blog.gmosx.com > http://nitroproject.org > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > From george.moschovitis at gmail.com Fri Dec 8 06:54:29 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Fri, 8 Dec 2006 13:54:29 +0200 Subject: [Nitro] [PATCH] important (and not so much) bugfixes In-Reply-To: References: Message-ID: Ok, thanks :) I will apply your changes and 0.41.0 is ready for release. I will do it on monday (perhaps we will find and fix some more problems over the w/e) regards, -g. On 12/8/06, Jonathan Buch wrote: > Oh, yes, forgot to say: > > All testcases are green now with the last patch (and the > taggable patch). > > Jo > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From george.moschovitis at gmail.com Fri Dec 8 06:55:52 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Fri, 8 Dec 2006 13:55:52 +0200 Subject: [Nitro] action_name attribute In-Reply-To: <44215.192.176.230.1.1165576260.squirrel@webmail.lassoweb.se> References: <4578622B.20107@lassoweb.se> <44215.192.176.230.1.1165576260.squirrel@webmail.lassoweb.se> Message-ID: ah, you use @context.action_name... interesting... > Poke, poke...How about some tips on the Nitro::Configuration object? i think i have some useful rdoc comments for this in the source code. -g. -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From john at oxyliquit.de Fri Dec 8 08:06:05 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Fri, 08 Dec 2006 14:06:05 +0100 Subject: [Nitro] [PATCH] template fix Message-ID: Hi, bug report from Rayman: and it didn't eval te #{} attached patch should fix that. Jo -- Feel the love http://pinkjuice.com/pics/ruby.png -------------- next part -------------- A non-text attachment was scrubbed... Name: tmpltefix.patch.tar.bz2 Type: application/bzip2 Size: 14151 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061208/15a80ed7/attachment-0001.bin From john at oxyliquit.de Fri Dec 8 08:10:15 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Fri, 08 Dec 2006 14:10:15 +0100 Subject: [Nitro] Nitro Configuration object In-Reply-To: <4578A46F.7000305@lassoweb.se> References: <4578A46F.7000305@lassoweb.se> Message-ID: Hi, > Could someone please explain to me how to use the Nitro::Configuration > object. It is supposed to replace configuration options in Runner, > Server etc, but I cannot find a clue on how it is used...Examples > including adapter and mode settings would be most welcome :) usage of Configuration (how I use it, anyway): * put all kinds of stuff in conf/(debug|live).rb example: Og.manager_options = { ... } Nitro::Compiler.mixin_get_parameters = false * call Configuration.load in run.rb right before Og.run/Nitro.run Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From vikingtux at gmail.com Fri Dec 8 08:26:30 2006 From: vikingtux at gmail.com (Alexandre Gravem) Date: Fri, 8 Dec 2006 10:26:30 -0300 Subject: [Nitro] [ANN] LoxAuth released to public In-Reply-To: <2E620CBE-81BB-47FD-B659-AC8709EA3A44@fabian-buch.de> References: <2E620CBE-81BB-47FD-B659-AC8709EA3A44@fabian-buch.de> Message-ID: <40b05ebe0612080526t1cedf937m9ab91bd6d07ecc38@mail.gmail.com> > LoxAuth 0.6.1 is the first Part released on LoxParts (http:// > loxparts.de/part/LoxAuth). It's the Auth system behind Oxyliquit and > also LoxParts itself. > > How LoxAuth works can be read in its README or on Oxyliquit Tip 37 > (http://oxyliquit.de/tip/37). > > Fabian As Oxyliquit this site is GREAT!!! I hope it helps once again the Nitro comunity to grow. A. Gravem -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061208/a562c8b0/attachment.html From george.moschovitis at gmail.com Fri Dec 8 08:32:53 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Fri, 8 Dec 2006 15:32:53 +0200 Subject: [Nitro] [PATCH] template fix In-Reply-To: References: Message-ID: thanks :) -g. On 12/8/06, Jonathan Buch wrote: > Hi, > > bug report from Rayman: > > > and it didn't eval te #{} > > attached patch should fix that. > > Jo > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > > -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From john at oxyliquit.de Fri Dec 8 09:00:02 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Fri, 08 Dec 2006 15:00:02 +0100 Subject: [Nitro] [PATCH] important (and not so much) bugfixes In-Reply-To: References: Message-ID: On Fri, 08 Dec 2006 12:54:29 +0100, George Moschovitis wrote: > Ok, thanks :) > I will apply your changes and 0.41.0 is ready for release. I will do > it on monday (perhaps we will find and fix some more problems over the > w/e) don't apply the speed patch, it prevents things to work in certain circumstances. Sorry, Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From george.moschovitis at gmail.com Fri Dec 8 13:08:22 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Fri, 8 Dec 2006 20:08:22 +0200 Subject: [Nitro] [PATCH] important (and not so much) bugfixes In-Reply-To: References: Message-ID: ok... -g. On 12/8/06, Jonathan Buch wrote: > On Fri, 08 Dec 2006 12:54:29 +0100, George Moschovitis wrote: > > > Ok, thanks :) > > I will apply your changes and 0.41.0 is ready for release. I will do > > it on monday (perhaps we will find and fix some more problems over the > > w/e) > > don't apply the speed patch, it prevents things to work in certain > circumstances. > > Sorry, > > Jo > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From george.moschovitis at gmail.com Sat Dec 9 03:09:14 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sat, 9 Dec 2006 10:09:14 +0200 Subject: [Nitro] [ANN] LoxParts - sharing Nitro parts In-Reply-To: References: <6D6076BE-0434-4132-A691-1F27E78E2208@fabian-buch.de> Message-ID: took a bit longer but here you are now :) -g. On 12/7/06, Fabian Buch wrote: > how about making Nitroproject.org's blog a little livelier and post a > small ann there too? ;) > > Fabian > > > -- > Nitro Q&A: http://oxyliquit.de > LoxParts: http://loxparts.de > Blog: http://blog.fabian-buch.de > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From fabian at fabian-buch.de Sat Dec 9 04:49:24 2006 From: fabian at fabian-buch.de (Fabian Buch) Date: Sat, 9 Dec 2006 10:49:24 +0100 Subject: [Nitro] [ANN] LoxParts - sharing Nitro parts In-Reply-To: References: <6D6076BE-0434-4132-A691-1F27E78E2208@fabian-buch.de> Message-ID: <17F3AB28-24A5-4B85-BAC6-A181D0326B17@fabian-buch.de> Am 09.12.2006 um 09:09 schrieb George Moschovitis: > took a bit longer but here you are now :) no problem, cool, thanks :) Fabian -- Nitro Q&A: http://oxyliquit.de LoxParts: http://loxparts.de Blog: http://blog.fabian-buch.de From manveru at weez-int.com Mon Dec 11 05:56:57 2006 From: manveru at weez-int.com (Michael Fellinger) Date: Mon, 11 Dec 2006 19:56:57 +0900 Subject: [Nitro] [PATCH] template fix In-Reply-To: References: Message-ID: <200612111956.57623.manveru@weez-int.com> On Friday 08 December 2006 22:06, Jonathan Buch wrote: > Hi, > > bug report from Rayman: > > > and it didn't eval te #{} > > attached patch should fix that. > > Jo * use the proper symbol for postgresql-adapter in the og/test/CONFIG.rb * rewrite of some gsub!s in template.rb, this should speed up the whole process. Also introduced SuperSafe(TM) delimter by using the MD5-hash of rand() (only once on startup, so no big deal) - this should make templating a whole lot more secure * added some tests for template.rb - still not enough... ^manveru -------------- next part -------------- A non-text attachment was scrubbed... Name: bundle.tar.gz Type: application/x-tgz Size: 3042 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061211/ba9e2575/attachment.bin From george.moschovitis at gmail.com Mon Dec 11 07:07:12 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Mon, 11 Dec 2006 14:07:12 +0200 Subject: [Nitro] [PATCH] template fix In-Reply-To: <200612111956.57623.manveru@weez-int.com> References: <200612111956.57623.manveru@weez-int.com> Message-ID: Thanks, Jonathan did you sent me some fixes for tc_reldel ant tc_controller? I have some tc errors still. -g. On 12/11/06, Michael Fellinger wrote: > On Friday 08 December 2006 22:06, Jonathan Buch wrote: > > Hi, > > > > bug report from Rayman: > > > > > > and it didn't eval te #{} > > > > attached patch should fix that. > > > > Jo > > * use the proper symbol for postgresql-adapter in the og/test/CONFIG.rb > * rewrite of some gsub!s in template.rb, this should speed up the whole > process. Also introduced SuperSafe(TM) delimter by using the MD5-hash of > rand() (only once on startup, so no big deal) - this should make > templating a whole lot more secure > * added some tests for template.rb - still not enough... > > ^manveru > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > > -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From john at oxyliquit.de Mon Dec 11 08:12:09 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Mon, 11 Dec 2006 14:12:09 +0100 Subject: [Nitro] [PATCH] template fix In-Reply-To: References: <200612111956.57623.manveru@weez-int.com> Message-ID: Hi, > Jonathan did you sent me some fixes for tc_reldel ant tc_controller? > I have some tc errors still. Re: [Nitro] [PATCH] important (and not so much) bugfixes It's in that thread, I guess you overlooked it because I didn't open an extra thread for it. File's called 'reldeltcfix.patch`. Can't find a failing tc in tc_controller though. Guess we're ready to release 0.41 now? When you do it today, you are actually within the promised time! ;D Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From george.moschovitis at gmail.com Mon Dec 11 11:37:21 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Mon, 11 Dec 2006 18:37:21 +0200 Subject: [Nitro] [PATCH] template fix In-Reply-To: References: <200612111956.57623.manveru@weez-int.com> Message-ID: > Guess we're ready to release 0.41 now? When you do it today, you are > actually within the promised time! ;D Yeap, we are ready. Will release tommorow though. I want to do some final testing first. Thanks for your great work. -g. -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From john at oxyliquit.de Mon Dec 11 14:25:36 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Mon, 11 Dec 2006 20:25:36 +0100 Subject: [Nitro] [PATCH] template fix In-Reply-To: References: <200612111956.57623.manveru@weez-int.com> Message-ID: aaand... you were right to wait! :P > Yeap, we are ready. Will release tommorow though. I want to do some > final testing first. Thanks for your great work. I messed up while testing my NITRO_LOG patch. Attached is a fix to make it work. (The logger.rb was not correctly getting valid arguments.) Jo -- Feel the love http://pinkjuice.com/pics/ruby.png -------------- next part -------------- A non-text attachment was scrubbed... Name: logfix.patch.tar.bz2 Type: application/bzip2 Size: 2323 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061211/62740692/attachment.bin From george.moschovitis at gmail.com Tue Dec 12 01:16:15 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Tue, 12 Dec 2006 08:16:15 +0200 Subject: [Nitro] [PATCH] template fix In-Reply-To: References: <200612111956.57623.manveru@weez-int.com> Message-ID: thanks! -g. On 12/11/06, Jonathan Buch wrote: > aaand... > > you were right to wait! :P > > > Yeap, we are ready. Will release tommorow though. I want to do some > > final testing first. Thanks for your great work. > > I messed up while testing my NITRO_LOG patch. > > Attached is a fix to make it work. (The logger.rb was not correctly > getting valid arguments.) > > Jo > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > > -- http://cull.gr http://www.joy.gr http://blog.gmosx.com http://nitroproject.org From george.moschovitis at gmail.com Tue Dec 12 05:56:08 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Tue, 12 Dec 2006 12:56:08 +0200 Subject: [Nitro] [ANN] Nitro + Og 0.41.0 Maintenance release Message-ID: Hello everyone, New versions of Nitro (Web Application Framework) and Og (Object Relational Mapping) were released: Homepage: http://www.nitroproject.org Download: http://rubyforge.org/projects/ nitro Mailing List: http://rubyforge.org/mailman/l istinfo/nitro-general IRC: #nitro on irc.freenode.net Documentation: http://oxyliquit.de Extra parts: http://loxparts.de This is a bug fix release. As it fixes some important bugs of the previous release, including a DOS vulnurability you are strongly advised to update your version. However, you will also find a couple of new features. Special thanks for this release fly to Jonathan/Fabian Buch. Most notable changes: * Extended entity .finder method can now handle relations. Post.find_by_title_and_forum_name(title,forumName) class Forum property :name, String has_many :posts, Post end class Post property :title, String property :message, String belongs_to :forum, Forum end 'forum' is the :forum from belongs_to, 'name' is a property from the relations. It creates a SQL subquery to find the correct forum_oid. * Added a simple call/answer mechanism that may be useful in some cases. Will be improved in future versions. Nitro provides everything you need to develop professional Web applications using Ruby and Javascript. Nitro redefines Rapid Application Development by providing a clean, yet efficient API, a layer of domain specific languages implemented on top of Ruby and the most powerful and elegant object relational mapping solution available everywhere. have fun, Nitro Team -- http://blog.gmosx.com http://nitroproject.org http://cull.gr http://www.joy.gr From john at oxyliquit.de Wed Dec 13 06:18:38 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 13 Dec 2006 12:18:38 +0100 Subject: [Nitro] Roadmap 0.41 Message-ID: Hi, George, we gotta choose a little roadmappy for 0.42 to not fall into disarray. ;) What I'd like to see: * Implement Kirby again Well, not so much kirby as just a way for storing plaintext stuff. That would include creating a plaintext store as opposed to sqlstore, right? And I'd rather see this done by you (george) and not by my grubby little fingers. ;) Attached are a few patches with bugfixes + more oracle stuff. (Anyone who gets bitten by STI (except Rayman)? Apply attached patches and report back.) Have a nice day, Jo -- Feel the love http://pinkjuice.com/pics/ruby.png -------------- next part -------------- A non-text attachment was scrubbed... Name: more_patches.patch.tar.bz2 Type: application/bzip2 Size: 5670 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061213/211af6a1/attachment.bin From john at oxyliquit.de Wed Dec 13 06:29:43 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 13 Dec 2006 12:29:43 +0100 Subject: [Nitro] Roadmap 0.42 In-Reply-To: References: Message-ID: Damn, that had to meant 0.42. Sorry for the disturbance. :P Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From john at oxyliquit.de Wed Dec 13 07:46:22 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 13 Dec 2006 13:46:22 +0100 Subject: [Nitro] 0.41 with speed patch Message-ID: George, so you _DID_ include my speed patch (mixed together with the log patch). I really hope that noone hits the wall with that one, I can't guarantee anything though. this in attribute.rb (glue): if klass && !serializable_attributes.empty? has to say unless serializable_attributes.empty? I really do hope you tested the first enough that it works under all circumstances.... Anyone tried 0.41 with his app yet, with success? Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From george.moschovitis at gmail.com Wed Dec 13 08:38:41 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Wed, 13 Dec 2006 15:38:41 +0200 Subject: [Nitro] Roadmap 0.41 In-Reply-To: References: Message-ID: > George, we gotta choose a little roadmappy for 0.42 to not fall > into disarray. ;) Ok, I will post my proposed roadmap tommorow. we can prioritize/merge our ideas then. -g. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Wed Dec 13 08:39:50 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Wed, 13 Dec 2006 15:39:50 +0200 Subject: [Nitro] 0.41 with speed patch In-Reply-To: References: Message-ID: Hm, I thought I hade removed that patch. perhaps this patch got included with another bundle of yours :( This creates not problems for my webapps though. thanks for mentioning this. -g. On 12/13/06, Jonathan Buch wrote: > George, > > so you _DID_ include my speed patch (mixed together with the log patch). > I really hope that noone hits the wall with that one, I can't guarantee > anything though. > > this in attribute.rb (glue): > > if klass && !serializable_attributes.empty? > > has to say > > unless serializable_attributes.empty? > > I really do hope you tested the first enough that it works under all > circumstances.... > > Anyone tried 0.41 with his app yet, with success? > > Jo > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From dan at tastapod.com Wed Dec 13 16:52:14 2006 From: dan at tastapod.com (Dan North) Date: Wed, 13 Dec 2006 21:52:14 +0000 Subject: [Nitro] [ANN] Nitro + Og 0.41.0 Maintenance release In-Reply-To: References: Message-ID: <4580760E.7090602@tastapod.com> Hi George. I'm still getting an (error) with this: http://www.nitroproject.org/videos/nitro4.html I wouldn't want that to be people's first impression of nitro. Cheers, Dan George Moschovitis wrote: > Hello everyone, > > New versions of Nitro (Web Application Framework) and Og (Object > Relational Mapping) were released: > > Homepage: http://www.nitroproject.org > Download: http://rubyforge.org/projects/ nitro > Mailing List: http://rubyforge.org/mailman/l istinfo/nitro-general > IRC: #nitro on irc.freenode.net > Documentation: http://oxyliquit.de > Extra parts: http://loxparts.de > > This is a bug fix release. As it fixes some important bugs of the > previous release, including a DOS vulnurability you are strongly > advised to update your version. However, you will also find > a couple of new features. > > Special thanks for this release fly to Jonathan/Fabian Buch. > > Most notable changes: > > * Extended entity .finder method can now handle relations. > > Post.find_by_title_and_forum_name(title,forumName) > > class Forum > property :name, String > has_many :posts, Post > end > > class Post > property :title, String > property :message, String > belongs_to :forum, Forum > end > > 'forum' is the :forum from belongs_to, 'name' is a property from the > relations. It creates a SQL subquery to find the correct forum_oid. > > * Added a simple call/answer mechanism that may be useful in some > cases. Will be improved in future versions. > > > > Nitro provides everything you need to develop professional Web > applications using Ruby and Javascript. > > Nitro redefines Rapid Application Development by providing a > clean, yet efficient API, a layer of domain specific languages > implemented on top of Ruby and the most powerful and elegant > object relational mapping solution available everywhere. > > > have fun, > Nitro Team > > From george.moschovitis at gmail.com Thu Dec 14 04:38:19 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 14 Dec 2006 11:38:19 +0200 Subject: [Nitro] [ANN] Nitro + Og 0.41.0 Maintenance release In-Reply-To: <4580760E.7090602@tastapod.com> References: <4580760E.7090602@tastapod.com> Message-ID: Ok, this fixed now... -g. On 12/13/06, Dan North wrote: > Hi George. > > I'm still getting an (error) with this: > http://www.nitroproject.org/videos/nitro4.html > > I wouldn't want that to be people's first impression of nitro. > > Cheers, > Dan > > George Moschovitis wrote: > > Hello everyone, > > > > New versions of Nitro (Web Application Framework) and Og (Object > > Relational Mapping) were released: > > > > Homepage: http://www.nitroproject.org > > Download: http://rubyforge.org/projects/ nitro > > Mailing List: http://rubyforge.org/mailman/l istinfo/nitro-general > > IRC: #nitro on irc.freenode.net > > Documentation: http://oxyliquit.de > > Extra parts: http://loxparts.de > > > > This is a bug fix release. As it fixes some important bugs of the > > previous release, including a DOS vulnurability you are strongly > > advised to update your version. However, you will also find > > a couple of new features. > > > > Special thanks for this release fly to Jonathan/Fabian Buch. > > > > Most notable changes: > > > > * Extended entity .finder method can now handle relations. > > > > Post.find_by_title_and_forum_name(title,forumName) > > > > class Forum > > property :name, String > > has_many :posts, Post > > end > > > > class Post > > property :title, String > > property :message, String > > belongs_to :forum, Forum > > end > > > > 'forum' is the :forum from belongs_to, 'name' is a property from the > > relations. It creates a SQL subquery to find the correct forum_oid. > > > > * Added a simple call/answer mechanism that may be useful in some > > cases. Will be improved in future versions. > > > > > > > > Nitro provides everything you need to develop professional Web > > applications using Ruby and Javascript. > > > > Nitro redefines Rapid Application Development by providing a > > clean, yet efficient API, a layer of domain specific languages > > implemented on top of Ruby and the most powerful and elegant > > object relational mapping solution available everywhere. > > > > > > have fun, > > Nitro Team > > > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Thu Dec 14 04:42:10 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 14 Dec 2006 11:42:10 +0200 Subject: [Nitro] 0.42.0 Roadmap Message-ID: Here is what I would like to have included in 0.42.0. Due to serious time constraints I will only plan to work on the features that I need for my apps and bug fixes. The proposed KirbyBase code does not fall in this category, so this will have to either wait for a future release or someone else work on this. I would like to hear more opinions on what should be included in the next version. Even better I would like to see some patches preferably with new cool features. regards, George. PS: Btw, I have already started with the improved errors/error pages infrastructure, more or less as proposed by jonathan. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From john at oxyliquit.de Thu Dec 14 05:12:30 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Thu, 14 Dec 2006 11:12:30 +0100 Subject: [Nitro] 0.42.0 Roadmap In-Reply-To: References: Message-ID: Lets compress this: Roadmap 0.42 * Bugfixes * Stability * feature: Error page infrastructure * feature: Oracle Adapter (I hope to work on this) * feature: separation of get/post parameters * verify: @context.action_name working? * docs: Nitro::Configuration, how to use * docs: Caching (session,og,template), how to use Anyone has further feature requests? Jo PS: I have some other items on my todo list, so if somebody wants to make him/herself useful... (mostly Og) :P -- Feel the love http://pinkjuice.com/pics/ruby.png From lpruszynski at gmail.com Thu Dec 14 05:13:41 2006 From: lpruszynski at gmail.com (Les Pruszynski) Date: Thu, 14 Dec 2006 10:13:41 +0000 Subject: [Nitro] Can't start app with nitro Message-ID: <88E76B4F-0378-45F4-A53C-881DD480E846@gmail.com> Hello, I'm quite new to nitro and hopefully you'll be able to clarify a few things for me. I tried to find out the answers for myself by so far not so good. 1. A new way to start nitro application is nitro command but unfortunately that does not work for me. I get the following error: xxx:~/Projects/Ruby/nitro/examples/blog xxx$ nitro app.rb:1:in `require': no such file to load -- facet/time/stamp (LoadError) from app.rb:1 Obviously the command cannot find a path to a file. Do you also get the same error as I? 2 . When trying out a blog example from the examples folder I am unable to get to the admin part. The documentation says that the command require 'part/admin' will create admin automatically, but it doesn't. Have I misunderstood the documentation? In fact none of the examples with require 'part/admin' create the part. Any help appreciated. Many thanks, Les From john at oxyliquit.de Thu Dec 14 05:48:49 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Thu, 14 Dec 2006 11:48:49 +0100 Subject: [Nitro] Can't start app with nitro In-Reply-To: <88E76B4F-0378-45F4-A53C-881DD480E846@gmail.com> References: <88E76B4F-0378-45F4-A53C-881DD480E846@gmail.com> Message-ID: Hi, I'll try to help with the first question, can't comment on the second as I have no experience with the admin interface. > 1. A new way to start nitro application is nitro command but > unfortunately that does not work for me. I get the following error: > xxx:~/Projects/Ruby/nitro/examples/blog xxx$ nitro > app.rb:1:in `require': no such file to load -- facet/time/stamp > (LoadError) > from app.rb:1 > Obviously the command cannot find a path to a file. > Do you also get the same error as I? This sounds like you don't have rubygems required. http://rubygems.org/read/chapter/3#page70 When you use the RUBYOPT method, you'll never worry about rubygems again. Also make sure that facets 1.4.5 is installed (and that it's the only facets available, don't use newer versions). If you have another facets version installed and want to keep it or you don't want to use the RUBYOPT method for rubygems, use this as 'header' in any of your run.rb/app.rb files: ## app.rb require 'rubygems' # a specific facets version require_gem 'facets', '= 1.4.5' require 'facets' require 'nitro' require 'og' __END__ Hope that helps, Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From lpruszynski at gmail.com Thu Dec 14 07:24:55 2006 From: lpruszynski at gmail.com (Les Pruszynski) Date: Thu, 14 Dec 2006 12:24:55 +0000 Subject: [Nitro] Can't start app with nitro Message-ID: <34E0AB74-306B-4425-BD77-AE7D5E8294DB@gmail.com> Hello John, Thank you for trying to help me, but I think I have all the necessary gems installed. Se below: xxx:~/Projects/Ruby/nitro/examples/blog xxx$ gem list --local *** LOCAL GEMS *** actionmailer (1.2.5) Service layer for easy email delivery and testing. actionpack (1.12.5) Web-flow and rendering framework putting the VC in MVC. actionwebservice (1.1.6) Web service support for Action Pack. activerecord (1.14.4) Implements the ActiveRecord pattern for ORM. activesupport (1.3.1) Support and utility classes used by the Rails framework. daemons (1.0.3, 0.4.2) A toolkit to create and control daemons in different ways facets (1.4.5) Exceptional collection of extensions and additions. gem_plugin (0.2.1) A plugin system based only on rubygems that uses dependencies only gen (0.41.0) A simple code generation system glue (0.41.0) Utility methods and classes for Nitro + Og mongrel (0.3.13.4) A small fast HTTP library and server that runs Rails, Camping, and Nitro apps. mongrel_cluster (0.2.1) Mongrel plugin that provides commands and Capistrano tasks for managing multiple Mongrel processes. nitro (0.41.0) Everything you need to create Web 2.0 applications with Ruby and Javascript og (0.41.0) State of the art object-relational mapping system rails (1.1.6) Web-application framework with template engine, control-flow layer, and ORM. rake (0.7.1) Ruby based make-like utility. RedCloth (3.0.4) RedCloth is a module for using Textile and Markdown in Ruby. Textile and Markdown are text formats. A very simple text format. Another stab at making readable text that can be converted to HTML. ruby-breakpoint (0.5.1, 0.5.0) ruby-breakpoint lets you inspect and modify state at run time. rubygems-update (0.9.0) RubyGems Update GEM sources (0.0.1) This package provides download sources for remote gem installation syntax (1.0.0) Syntax is Ruby library for performing simple syntax highlighting. wee (0.10.0) Wee is a framework for building highly dynamic web applications. So I'm not sure what the problem is. Les From john at oxyliquit.de Thu Dec 14 07:31:56 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Thu, 14 Dec 2006 13:31:56 +0100 Subject: [Nitro] Can't start app with nitro In-Reply-To: <34E0AB74-306B-4425-BD77-AE7D5E8294DB@gmail.com> References: <34E0AB74-306B-4425-BD77-AE7D5E8294DB@gmail.com> Message-ID: Hi, your list of installed gems looks good, did you try my 'header' approach and it still produces the same error message? Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From lpruszynski at gmail.com Thu Dec 14 07:30:58 2006 From: lpruszynski at gmail.com (Les Pruszynski) Date: Thu, 14 Dec 2006 12:30:58 +0000 Subject: [Nitro] Can't start app with nitro Message-ID: <70D09985-C691-4D9D-9582-F6EBC3D6274A@gmail.com> Hello John, I think I wasn't clear first time round. I can indeed launch the examples applications via ruby -rubygems app.rb. That works OK. However, the new command nitro that is supposed to do the same thing does not work for me. I can live without it but was just curious whether anybody else has got the same problem with the nitro command. Les From fabian at fabian-buch.de Thu Dec 14 08:51:18 2006 From: fabian at fabian-buch.de (Fabian Buch) Date: Thu, 14 Dec 2006 14:51:18 +0100 Subject: [Nitro] [PATCH] Webfile Controls Message-ID: <032CD1BD-F05E-479E-8EC7-5A2BF8C19B1E@fabian-buch.de> It makes no sense to use text controls for files. Attached patch makes webfile use file control. Fabian -------------- next part -------------- A non-text attachment was scrubbed... Name: webfile_controls.patch.gz Type: application/x-gzip Size: 2435 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061214/27325799/attachment-0001.gz -------------- next part -------------- -- Nitro Q&A: http://oxyliquit.de LoxParts: http://loxparts.de Blog: http://blog.fabian-buch.de From Reid.Thompson at ateb.com Thu Dec 14 08:49:36 2006 From: Reid.Thompson at ateb.com (Reid Thompson) Date: Thu, 14 Dec 2006 08:49:36 -0500 Subject: [Nitro] [ANN] Nitro + Og 0.41.0 Maintenance release In-Reply-To: References: <4580760E.7090602@tastapod.com> Message-ID: <1166104176.32529.47.camel@localhost> On Thu, 2006-12-14 at 11:38 +0200, George Moschovitis wrote: > Ok, this fixed now... > > -g. Broken again. so is http://www.nitroproject.org/videos -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061214/f45961d7/attachment.html From george.moschovitis at gmail.com Thu Dec 14 12:36:31 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 14 Dec 2006 19:36:31 +0200 Subject: [Nitro] [ANN] Nitro + Og 0.41.0 Maintenance release In-Reply-To: <1166104176.32529.47.camel@localhost> References: <4580760E.7090602@tastapod.com> <1166104176.32529.47.camel@localhost> Message-ID: http://www.nitroproject.org/documentation check the links... there is no link calles videos in the site... -g. On 12/14/06, Reid Thompson wrote: > > > > On Thu, 2006-12-14 at 11:38 +0200, George Moschovitis wrote: > > Ok, this fixed now... > > > > -g. > > Broken again. > > so is http://www.nitroproject.org/videos > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Thu Dec 14 12:38:13 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 14 Dec 2006 19:38:13 +0200 Subject: [Nitro] [PATCH] Webfile Controls In-Reply-To: <032CD1BD-F05E-479E-8EC7-5A2BF8C19B1E@fabian-buch.de> References: <032CD1BD-F05E-479E-8EC7-5A2BF8C19B1E@fabian-buch.de> Message-ID: thanks :) -g. On 12/14/06, Fabian Buch wrote: > It makes no sense to use text controls for files. > Attached patch makes webfile use file control. > > Fabian > > > > > > > > > -- > Nitro Q&A: http://oxyliquit.de > LoxParts: http://loxparts.de > Blog: http://blog.fabian-buch.de > > > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From fabian at fabian-buch.de Fri Dec 15 06:18:41 2006 From: fabian at fabian-buch.de (Fabian Buch) Date: Fri, 15 Dec 2006 12:18:41 +0100 Subject: [Nitro] [PATCH] FeedHelper Atom with html content Message-ID: Fri Dec 15 12:06:19 CET 2006 Fabian Buch * FeedHelper Atom with html content Atom can contain html, but only if marked as that. This patch sets the markup type of content to be always html (doesn't hurt if non- markup text is provided). With this change it looks much nicer in many FeedReaders. ATTENTION: changes API: provide markuped content to FeedHelper, it makes no sense that the FeedHelper calls the markup() method, since not everyone uses RedCloth for his/her markup (e.g. Oxy uses BlueCloth). Note: Tests pass and Oxyliquit uses this patch since its redesign. Fabian -- Nitro Q&A: http://oxyliquit.de LoxParts: http://loxparts.de Blog: http://blog.fabian-buch.de From fabian at fabian-buch.de Fri Dec 15 10:46:34 2006 From: fabian at fabian-buch.de (Fabian Buch) Date: Fri, 15 Dec 2006 16:46:34 +0100 Subject: [Nitro] [ANN] LoxGallery 0.1.0 released on loxparts.de Message-ID: <3C31817F-F81C-4211-B56E-BCD0A86E780E@fabian-buch.de> I just released LoxGallery 0.1.0 (http://loxparts.de/part/LoxGallery) This is the initial release of LoxGallery. I hope it's mostly bugfree, but if you find any, please report as well as feature requests. Any comment is welcome! For installation instructions see the README file. LoxGallery, mainly developed by Fabian Buch (fabian at fabian-buch.de) is a small photo gallery webapp developed with Nitro which can be used as standalone gallery webapp or as part of other Nitro webapps. Copyright (c) 2006 Fabian Buch under MIT license. See LICENSE file for details. The famous PHP "Gallery" webapp is just too big for me. I like it a little cleaner, with less features, but customized for my needs and maybe also suitable for others, especially if combined with other Nitro applications, like LoxBlog. There's something from Gallery though that I stole and customized it only a little: it's API. Even though it's not perfect, there are lots of apps out there that use it. With some customization (e.g. added Tag functionality) I added it to my LoxGallery and changed Zach Wily's iPhotoToGallery (http://zwily.com/iphoto/) to use my slightly incompatible API for my needs. Attached to LoxGallery is iPhotoToLoxGallery which is an iPhoto Exporter. It's sources are quite big (zipped 29 MB), so if you need them, just mail me. Example Galleries: http://fabian-buch.com/gallery http://vanessa-und-fabian.de/gallery Both styles are included in this Part. Feel free use and alter them for your needs. Fabian -- Nitro Q&A: http://oxyliquit.de LoxParts: http://loxparts.de Blog: http://blog.fabian-buch.de From george.moschovitis at gmail.com Fri Dec 15 11:33:30 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Fri, 15 Dec 2006 18:33:30 +0200 Subject: [Nitro] [ANN] LoxGallery 0.1.0 released on loxparts.de In-Reply-To: <3C31817F-F81C-4211-B56E-BCD0A86E780E@fabian-buch.de> References: <3C31817F-F81C-4211-B56E-BCD0A86E780E@fabian-buch.de> Message-ID: nice, thanks -g. On 12/15/06, Fabian Buch wrote: > I just released LoxGallery 0.1.0 (http://loxparts.de/part/LoxGallery) > > This is the initial release of LoxGallery. I hope it's mostly > bugfree, but if you find any, please report as well as feature > requests. Any comment is welcome! > > For installation instructions see the README file. > > LoxGallery, mainly developed by Fabian Buch (fabian at fabian-buch.de) is a > small photo gallery webapp developed with Nitro which can be used as > standalone gallery webapp or as part of other Nitro webapps. > > Copyright (c) 2006 Fabian Buch under MIT license. > See LICENSE file for details. > > The famous PHP "Gallery" webapp is just too big for me. I like it a > little cleaner, with less features, but customized for my needs and > maybe also suitable for others, especially if combined with other > Nitro applications, like LoxBlog. > > There's something from Gallery though that I stole and customized it > only a little: it's API. Even though it's not perfect, there are lots > of apps out there that use it. With some customization (e.g. added > Tag functionality) I added it to my LoxGallery and changed Zach > Wily's iPhotoToGallery (http://zwily.com/iphoto/) to use my slightly > incompatible API for my needs. > Attached to LoxGallery is iPhotoToLoxGallery which is an iPhoto > Exporter. It's sources are quite big (zipped 29 MB), so if you need > them, just mail me. > > Example Galleries: > http://fabian-buch.com/gallery > http://vanessa-und-fabian.de/gallery > > Both styles are included in this Part. Feel free use and alter them > for your needs. > > Fabian > > > > > -- > Nitro Q&A: http://oxyliquit.de > LoxParts: http://loxparts.de > Blog: http://blog.fabian-buch.de > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Sat Dec 16 05:56:12 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sat, 16 Dec 2006 12:56:12 +0200 Subject: [Nitro] [PATCH] FeedHelper Atom with html content In-Reply-To: References: Message-ID: there is no patch attached! -g. On 12/15/06, Fabian Buch wrote: > Fri Dec 15 12:06:19 CET 2006 Fabian Buch > * FeedHelper Atom with html content > Atom can contain html, but only if marked as that. This patch sets > the markup type of content to be always html (doesn't hurt if non- > markup > text is provided). With this change it looks much nicer in many > FeedReaders. > ATTENTION: changes API: provide markuped content to FeedHelper, it > makes no > sense that the FeedHelper calls the markup() method, since not > everyone > uses RedCloth for his/her markup (e.g. Oxy uses BlueCloth). > > Note: Tests pass and Oxyliquit uses this patch since its redesign. > > Fabian > > > > > -- > Nitro Q&A: http://oxyliquit.de > LoxParts: http://loxparts.de > Blog: http://blog.fabian-buch.de > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From fabian at fabian-buch.de Sat Dec 16 06:06:23 2006 From: fabian at fabian-buch.de (Fabian Buch) Date: Sat, 16 Dec 2006 12:06:23 +0100 Subject: [Nitro] [PATCH] FeedHelper Atom with html content In-Reply-To: References: Message-ID: <103886EF-24EC-4C22-83CD-6CBEA1CE149E@fabian-buch.de> oops, now: -------------- next part -------------- A non-text attachment was scrubbed... Name: feedhelper_atom_html.patch.gz Type: application/x-gzip Size: 2695 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061216/0722d12d/attachment.gz -------------- next part -------------- Am 16.12.2006 um 11:56 schrieb George Moschovitis: > there is no patch attached! > > -g. -- Nitro Q&A: http://oxyliquit.de LoxParts: http://loxparts.de Blog: http://blog.fabian-buch.de From george.moschovitis at gmail.com Sat Dec 16 06:20:09 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sat, 16 Dec 2006 13:20:09 +0200 Subject: [Nitro] [PATCH] FeedHelper Atom with html content In-Reply-To: <103886EF-24EC-4C22-83CD-6CBEA1CE149E@fabian-buch.de> References: <103886EF-24EC-4C22-83CD-6CBEA1CE149E@fabian-buch.de> Message-ID: Ok, thanks. -g. On 12/16/06, Fabian Buch wrote: > oops, now: > > > > > > Am 16.12.2006 um 11:56 schrieb George Moschovitis: > > > there is no patch attached! > > > > -g. > > > > > > -- > Nitro Q&A: http://oxyliquit.de > LoxParts: http://loxparts.de > Blog: http://blog.fabian-buch.de > > > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From lpruszynski at gmail.com Sun Dec 17 00:08:44 2006 From: lpruszynski at gmail.com (Les Pruszynski) Date: Sun, 17 Dec 2006 05:08:44 +0000 Subject: [Nitro] Problems downloading LoxGallery Message-ID: Hello Fabian, When trying to download LoxGallery from: http://loxparts.de/part/LoxGallery what gets downloaded is LoxAuth. Can you doublecheck the path to it? Les From george.moschovitis at gmail.com Sun Dec 17 03:53:20 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sun, 17 Dec 2006 10:53:20 +0200 Subject: [Nitro] Nitro dir structure Message-ID: Dear devs, I would like to change the structure of the nitro directory a bit. I would like to remove the /nitro/proto directory. I would like to make /nitro the protoype directory. Ie the nitro directory would be a webapp itself (a hello world webapp). It will contain an app.rb/rub.rb file that would start a simple web app (welcome to nitro + links to info). what do you think? -g. PS: btw, I am working on some of the proposed features for 0.42.0 Would you like me to sync with the public repo daily (so that more people have the chance to comment early on the implementation) or obly commit final, working code? -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From noe.rubinstein at gmail.com Sun Dec 17 03:55:18 2006 From: noe.rubinstein at gmail.com (=?UTF-8?Q?No=C3=A9_Rubinstein?=) Date: Sun, 17 Dec 2006 09:55:18 +0100 Subject: [Nitro] [PATCH] Separate get_params and post_params Message-ID: This patch separates GET params and POST params. request.post_params # now returns POST params request.get_params # now returns GET params request.params # Compatibility method returning the result of post_params.merge(get_params) From noe.rubinstein at gmail.com Sun Dec 17 04:21:57 2006 From: noe.rubinstein at gmail.com (=?UTF-8?Q?No=C3=A9_Rubinstein?=) Date: Sun, 17 Dec 2006 10:21:57 +0100 Subject: [Nitro] [PATCH] Separate get_params and post_params In-Reply-To: References: Message-ID: Oooops, forgotten the patch itself. 2006/12/17, No? Rubinstein : > This patch separates GET params and POST params. > request.post_params # now returns POST params > request.get_params # now returns GET params > request.params # Compatibility method returning the result of > post_params.merge(get_params) > -------------- next part -------------- A non-text attachment was scrubbed... Name: separate_post_and_get_params.bz2 Type: application/x-bzip2 Size: 2673 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061217/189fd7c9/attachment.bz2 From fabian at fabian-buch.de Sun Dec 17 04:32:08 2006 From: fabian at fabian-buch.de (Fabian Buch) Date: Sun, 17 Dec 2006 10:32:08 +0100 Subject: [Nitro] Problems downloading LoxGallery In-Reply-To: References: Message-ID: <804F2010-B806-494A-B2E0-DF7CF406CF11@fabian-buch.de> Am 17.12.2006 um 06:08 schrieb Les Pruszynski: > Hello Fabian, > When trying to download LoxGallery from: > http://loxparts.de/part/LoxGallery > what gets downloaded is LoxAuth. > Can you doublecheck the path to it? > Les Thanks for reporting. It wasn't a wrong path. It was me not selecting the right file when I uploaded it. I'm truly sorry. Be sure to download the file from the Germany (slow) mirror for now, since the mirrors will take a while to have the just newly uploaded file. Sorry again, and for this mistake I'll implement tag-search into LoxGallery later today and release a new version :). Fabian -- Nitro Q&A: http://oxyliquit.de LoxParts: http://loxparts.de Blog: http://blog.fabian-buch.de From george.moschovitis at gmail.com Sun Dec 17 04:35:27 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sun, 17 Dec 2006 11:35:27 +0200 Subject: [Nitro] [PATCH] Separate get_params and post_params In-Reply-To: References: Message-ID: Interesting, I will have a look at this... -g. On 12/17/06, No? Rubinstein wrote: > This patch separates GET params and POST params. > request.post_params # now returns POST params > request.get_params # now returns GET params > request.params # Compatibility method returning the result of > post_params.merge(get_params) > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From fabian at fabian-buch.de Sun Dec 17 04:37:23 2006 From: fabian at fabian-buch.de (Fabian Buch) Date: Sun, 17 Dec 2006 10:37:23 +0100 Subject: [Nitro] Nitro dir structure In-Reply-To: References: Message-ID: <00B3C7A5-BE72-4E1A-8F76-64CD029AC4B7@fabian-buch.de> Am 17.12.2006 um 09:53 schrieb George Moschovitis: > I would like to change the structure of the nitro directory a bit. I > would like to remove > the /nitro/proto directory. I would like to make /nitro the protoype > directory. Ie the nitro directory would be a webapp itself (a hello > world webapp). It will contain an app.rb/rub.rb file that would start > a simple web app (welcome to nitro + links to info). Two thoughts: - where do you put the error.xhtml then? - why not make a part of proto and put it into the part/ directory? just a thought Fabian -- Nitro Q&A: http://oxyliquit.de LoxParts: http://loxparts.de Blog: http://blog.fabian-buch.de From noe.rubinstein at gmail.com Sun Dec 17 05:01:07 2006 From: noe.rubinstein at gmail.com (=?UTF-8?Q?No=C3=A9_Rubinstein?=) Date: Sun, 17 Dec 2006 11:01:07 +0100 Subject: [Nitro] [PATCH] Separate get_params and post_params In-Reply-To: References: Message-ID: Same but with post precedence other get to prevent from hacking. (thanks manv) 2006/12/17, George Moschovitis : > Interesting, > > I will have a look at this... > > -g. > > On 12/17/06, No? Rubinstein wrote: > > This patch separates GET params and POST params. > > request.post_params # now returns POST params > > request.get_params # now returns GET params > > request.params # Compatibility method returning the result of > > post_params.merge(get_params) > > _______________________________________________ > > Nitro-general mailing list > > Nitro-general at rubyforge.org > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > -- > http://blog.gmosx.com > http://cull.gr > http://www.joy.gr > http://nitroproject.org > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -------------- next part -------------- A non-text attachment was scrubbed... Name: post_precedence_over_get.bz2 Type: application/x-bzip2 Size: 2704 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061217/639ac75e/attachment.bz2 From noe.rubinstein at gmail.com Sun Dec 17 05:28:17 2006 From: noe.rubinstein at gmail.com (=?UTF-8?Q?No=C3=A9_Rubinstein?=) Date: Sun, 17 Dec 2006 11:28:17 +0100 Subject: [Nitro] [PATCH] Separate get_params and post_params In-Reply-To: References: Message-ID: Same but params and params= are now post_params and post_params= if method is POST, get_params and get_params= otherwise. (thanks Kashia) Sorry for the flood. 2006/12/17, No? Rubinstein : > Same but with post precedence other get to prevent from hacking. (thanks manv) > > 2006/12/17, George Moschovitis : > > Interesting, > > > > I will have a look at this... > > > > -g. > > > > On 12/17/06, No? Rubinstein wrote: > > > This patch separates GET params and POST params. > > > request.post_params # now returns POST params > > > request.get_params # now returns GET params > > > request.params # Compatibility method returning the result of > > > post_params.merge(get_params) > > > _______________________________________________ > > > Nitro-general mailing list > > > Nitro-general at rubyforge.org > > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > > > > > -- > > http://blog.gmosx.com > > http://cull.gr > > http://www.joy.gr > > http://nitroproject.org > > _______________________________________________ > > Nitro-general mailing list > > Nitro-general at rubyforge.org > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > -------------- next part -------------- A non-text attachment was scrubbed... Name: params_is_method_params.bz2 Type: application/x-bzip2 Size: 2792 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061217/2f6d3fa9/attachment-0001.bz2 From john at oxyliquit.de Sun Dec 17 05:38:07 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Sun, 17 Dec 2006 11:38:07 +0100 Subject: [Nitro] Nitro dir structure In-Reply-To: References: Message-ID: Hi, > I would like to change the structure of the nitro directory a bit. I > would like to remove > the /nitro/proto directory. I would like to make /nitro the protoype > directory. Ie the nitro directory would be a webapp itself (a hello > world webapp). It will contain an app.rb/run.rb file that would start > a simple web app (welcome to nitro + links to info). > > what do you think? Not so sure about that... Can you explain the use case? What we would have then, is an incompatibility with gen (what would it copy, the whole nitro tree?) So, what would this 'standard help app' be used for? You can't copy it to modify it and make your own app with it (without having lots of baggage attached). It is not self contained (in case you wanted that) in case it uses Og, and there's still glue. If it's about 'ease of first use' for a newbie, then I would also argue against it, gen is easy enough (gem is used as the premier install method anyway so it gets installed just like that). * side note: One could perhaps think of merging the gen functionality to the `nitro` command (`nitro --gen [-g] app test_app`) so only a single command is used for all nitro commands... So, about helping new users, documentation is all around more helpful than a 'self running nitro dir' for most kinds of people (imo). > PS: btw, I am working on some of the proposed features for 0.42.0 > Would you like me to sync with the public repo daily (so that more > people have the chance to comment early on the implementation) or obly > commit final, working code? I think I would prefer you just pushing the code you have, it's the dev repo after all. :) This will prevent any merge issues, in case someone works in the same area. Someone not wanting those patches can just chose to not pull them thanks to the power of darcs. Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From george.moschovitis at gmail.com Sun Dec 17 06:09:51 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sun, 17 Dec 2006 13:09:51 +0200 Subject: [Nitro] Nitro dir structure In-Reply-To: <00B3C7A5-BE72-4E1A-8F76-64CD029AC4B7@fabian-buch.de> References: <00B3C7A5-BE72-4E1A-8F76-64CD029AC4B7@fabian-buch.de> Message-ID: > Two thoughts: > - where do you put the error.xhtml then? i put it in: nitro/template static files from proto go to: nitro/public > - why not make a part of proto and put it into the part/ directory? well this is an alternative, but I think nitro as a webapp is cooler? what do others think? -g. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Sun Dec 17 06:12:41 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sun, 17 Dec 2006 13:12:41 +0200 Subject: [Nitro] Nitro dir structure In-Reply-To: References: Message-ID: > * side note: One could perhaps think of merging the gen functionality > to the `nitro` command (`nitro --gen [-g] app test_app`) so only a > single command is used for all nitro commands... yeah, i want to do that. nitro --gen could copy part of the nitro directory, or even the whole directory. When I deploy my apps on the live server I copy the whole nitro framework in a lib directory inside the webapp directory. This way, each web app can use a different version of the nitro framework. > I think I would prefer you just pushing the code you have, it's the dev > repo after all. :) ok, will push some stuff tommorow then ;-) -g. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From john at oxyliquit.de Sun Dec 17 08:56:28 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Sun, 17 Dec 2006 14:56:28 +0100 Subject: [Nitro] Nitro dir structure In-Reply-To: References: Message-ID: Hi, > yeah, i want to do that. nitro --gen could copy part of the nitro > directory, or even the whole directory. When I deploy my apps on the > live server I copy the whole nitro framework in a lib directory inside > the webapp directory. This way, each web app can use a different > version of the nitro framework. One of the things I love about Nitro is, that the standard is to have very few files in the project directory. I remember rails creating lotsa files there... IMO a little too much baggage, even if it's more convenient in certain deployment scenarios... I see that this can have advantages... can we let the proto dir (as a kind of minimal 'how a nitro app looks like') alone and push the functionality of putting the whole nitro app into a lib/ dir into --gen ? nitro --gen --include-nitro --include-og newproject This would of course add glue as well. It maybe could create a run.rb with $:.unshift(File.dirname(__FILE__)+'/lib')... How about it? >> I think I would prefer you just pushing the code you have, it's the dev >> repo after all. :) > > ok, will push some stuff tommorow then ;-) nice :) Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From lists at zarac.com Sun Dec 17 11:51:04 2006 From: lists at zarac.com (Ric Turley) Date: Sun, 17 Dec 2006 09:51:04 -0700 Subject: [Nitro] [BUG] Can't create new member on nitroproject.org In-Reply-To: Message-ID: I can't get registered. Entered email address, got confirmation email and entered token, but get a page with only '(error)' after entering name and password: http://www.nitroproject.org/users/create?e=ric%40zarac.com&t=NI2bNK5BD%2F2KQ Thanks! Ric Turley > From: "George Moschovitis" > Reply-To: General discussion about Nitro > Date: Thu, 7 Dec 2006 11:44:25 +0200 > To: "General discussion about Nitro" > Subject: Re: [Nitro] [BUG] Can't create new member on nitroproject.org > >> Great to hear that 0.41 is on the way. > > I have updated the repo with an almost final version of 0.41.0. there > are 2 test cases that don't pass. Jonathan could you also have a look > at them? > > -g. > > -- > http://cull.gr > http://www.joy.gr > http://blog.gmosx.com > http://nitroproject.org > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > From fabian at fabian-buch.de Sun Dec 17 12:33:16 2006 From: fabian at fabian-buch.de (Fabian Buch) Date: Sun, 17 Dec 2006 18:33:16 +0100 Subject: [Nitro] [ANN] LoxGallery 0.2.0 released Message-ID: notable changes: Tag-Search download: http://loxparts.de/part/LoxGallery Fabian -- Nitro Q&A: http://oxyliquit.de LoxParts: http://loxparts.de Blog: http://blog.fabian-buch.de From transfire at gmail.com Sun Dec 17 14:08:47 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Sun, 17 Dec 2006 19:08:47 -0000 Subject: [Nitro] Nitro dir structure In-Reply-To: References: Message-ID: <1166382527.366073.65930@80g2000cwy.googlegroups.com> George Moschovitis wrote: > Dear devs, > > I would like to change the structure of the nitro directory a bit. I > would like to remove > the /nitro/proto directory. I would like to make /nitro the protoype > directory. Ie the nitro directory would be a webapp itself (a hello > world webapp). It will contain an app.rb/rub.rb file that would start > a simple web app (welcome to nitro + links to info). > > what do you think? slightly aside but it may be helpful in this consideration. for all my projects i have moved to a model of website over repository. ie. my project layout looks like this: myproject/ index.html js/ css/ repo/ ProjectInfo bin/ lib/ pkg/ and so on. the great thing about this is that I can upload my website and everying goes with it repository and all. (and is why I asked Tom Copeland to implement rsync on rubyforge. super yea that he did!) the only downside is that my webpages aren't part of my repo's version control, but that's not that big a deal. so perhaps this pattern would work for nitro too? but in nitro's case the distribution (and versioning) actually includes this website container as the starter app? in other words maybe the proto app should exist over the nitro system rather than under or parallel to it. just a thought. (hmm... i see maybe some other advantages too... the repo dir (or whatever you want to call it) could be soft linked to a shared location and the proto app could be a "reconfigurable" app, ie. a gen-based web app. yes, i'm a dreamer ;-) t. From john at oxyliquit.de Sun Dec 17 15:27:46 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Sun, 17 Dec 2006 21:27:46 +0100 Subject: [Nitro] Nitro dir structure In-Reply-To: <1166382527.366073.65930@80g2000cwy.googlegroups.com> References: <1166382527.366073.65930@80g2000cwy.googlegroups.com> Message-ID: Hi, > so perhaps this pattern would work for nitro too? but in nitro's case > the distribution (and versioning) actually includes this website > container as the starter app? in other words maybe the proto app should > exist over the nitro system rather than under or parallel to it. just a > thought. This layout would imply that the whole tree (app source) would be accessable from the outside (browser) if you don't deny access to that folder by using an Apache .htaccess file or other configuration options provided by the webserver. > (hmm... i see maybe some other advantages too... the repo dir (or > whatever you want to call it) could be soft linked to a shared location > and the proto app could be a "reconfigurable" app, ie. a gen-based web > app. yes, i'm a dreamer ;-) Well, as Nitro (meaning the nitro app) is practically your server (even if cgi based), the application can reside anywhere. The only thing the webserver (any proxy, mongrel, fcgi client, etc) has to know (besides the nitro app server) is the path to the public files. So the public/ folder is completely decoupled from the nitro app. Your proposed folder layout is doable, but it's 'upside down' IMO. Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From george.moschovitis at gmail.com Mon Dec 18 06:17:56 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Mon, 18 Dec 2006 13:17:56 +0200 Subject: [Nitro] [PATCH] Separate get_params and post_params In-Reply-To: References: Message-ID: this patch is not complete. there are more fixes needed to make this work. btw, shouldnt we merge the post and get parameters as you originally suggested? -g. On 12/17/06, No? Rubinstein wrote: > Same but params and params= are now post_params and post_params= if > method is POST, get_params and get_params= otherwise. (thanks Kashia) > Sorry for the flood. > > 2006/12/17, No? Rubinstein : > > Same but with post precedence other get to prevent from hacking. (thanks manv) > > > > 2006/12/17, George Moschovitis : > > > Interesting, > > > > > > I will have a look at this... > > > > > > -g. > > > > > > On 12/17/06, No? Rubinstein wrote: > > > > This patch separates GET params and POST params. > > > > request.post_params # now returns POST params > > > > request.get_params # now returns GET params > > > > request.params # Compatibility method returning the result of > > > > post_params.merge(get_params) > > > > _______________________________________________ > > > > Nitro-general mailing list > > > > Nitro-general at rubyforge.org > > > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > > > > > > > > > -- > > > http://blog.gmosx.com > > > http://cull.gr > > > http://www.joy.gr > > > http://nitroproject.org > > > _______________________________________________ > > > Nitro-general mailing list > > > Nitro-general at rubyforge.org > > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Mon Dec 18 06:19:31 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Mon, 18 Dec 2006 13:19:31 +0200 Subject: [Nitro] [PATCH] Separate get_params and post_params In-Reply-To: References: Message-ID: btw, why is it useful to separate get/post params? if u use them through params (which in your latest implementation just selects from one collection or the other anyway)? -g. On 12/18/06, George Moschovitis wrote: > this patch is not complete. there are more fixes needed to make this work. > btw, shouldnt we merge the post and get parameters as you originally suggested? > > -g. > > On 12/17/06, No? Rubinstein wrote: > > Same but params and params= are now post_params and post_params= if > > method is POST, get_params and get_params= otherwise. (thanks Kashia) > > Sorry for the flood. > > > > 2006/12/17, No? Rubinstein : > > > Same but with post precedence other get to prevent from hacking. (thanks manv) > > > > > > 2006/12/17, George Moschovitis : > > > > Interesting, > > > > > > > > I will have a look at this... > > > > > > > > -g. > > > > > > > > On 12/17/06, No? Rubinstein wrote: > > > > > This patch separates GET params and POST params. > > > > > request.post_params # now returns POST params > > > > > request.get_params # now returns GET params > > > > > request.params # Compatibility method returning the result of > > > > > post_params.merge(get_params) > > > > > _______________________________________________ > > > > > Nitro-general mailing list > > > > > Nitro-general at rubyforge.org > > > > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > > > > > > > > > > > > > -- > > > > http://blog.gmosx.com > > > > http://cull.gr > > > > http://www.joy.gr > > > > http://nitroproject.org > > > > _______________________________________________ > > > > Nitro-general mailing list > > > > Nitro-general at rubyforge.org > > > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > > > > > > > > > > _______________________________________________ > > Nitro-general mailing list > > Nitro-general at rubyforge.org > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > > > > > -- > http://blog.gmosx.com > http://cull.gr > http://www.joy.gr > http://nitroproject.org > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From wes.sheldahl at gmail.com Mon Dec 18 09:02:08 2006 From: wes.sheldahl at gmail.com (Wes Sheldahl) Date: Mon, 18 Dec 2006 09:02:08 -0500 Subject: [Nitro] Nitro dir structure In-Reply-To: References: <1166382527.366073.65930@80g2000cwy.googlegroups.com> Message-ID: On 12/17/06, Jonathan Buch wrote: > > Hi, > > > so perhaps this pattern would work for nitro too? but in nitro's case > > the distribution (and versioning) actually includes this website > > container as the starter app? in other words maybe the proto app should > > exist over the nitro system rather than under or parallel to it. just a > > thought. > > This layout would imply that the whole tree (app source) would be > accessable from the outside (browser) if you don't deny access to that > folder by using an Apache .htaccess file or other configuration options > provided by the webserver. This looks like a strong argument against that sort of directory structure. Applications in general should be secure by default. For web applications, this means generally not putting config files (which often have passwords) and library code directly in the DocumentRoot where browsers can access them directly. Making the DocumentRoot a folder callled something like 'public' is a great way remind you (the developer) that whatever you put in there is public; the private parts of your app need to be elsewhere. Yes the .htaccess solution would work, except for when you forget to do it. :-) -- Wes Sheldahl wes.sheldahl at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061218/1d99119c/attachment.html From vikingtux at gmail.com Mon Dec 18 10:15:25 2006 From: vikingtux at gmail.com (Alexandre Gravem) Date: Mon, 18 Dec 2006 13:15:25 -0200 Subject: [Nitro] Nitro dir structure In-Reply-To: References: Message-ID: <40b05ebe0612180715n2173f47j132fe06197562435@mail.gmail.com> > > I see that this can have advantages... can we let the proto dir (as a kind > of minimal 'how a nitro app looks like') alone and push the functionality > of putting the whole nitro app into a lib/ dir into --gen ? > > nitro --gen --include-nitro --include-og newproject > > This would of course add glue as well. It maybe could create a run.rbwith > $:.unshift(File.dirname(__FILE__)+'/lib')... How about it? I liked this idea very much!! In fact thats kind of what i am doing in my projects. A. Gravem -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061218/17b325a2/attachment.html From john at oxyliquit.de Mon Dec 18 10:29:16 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Mon, 18 Dec 2006 16:29:16 +0100 Subject: [Nitro] Nitro dir structure In-Reply-To: References: <1166382527.366073.65930@80g2000cwy.googlegroups.com> Message-ID: Hi, >> > over the nitro system rather than under or parallel to it. > This looks like a strong argument against that sort of directory structure. > Applications in general should be secure by default. For web applications, > this means generally not putting config files (which often have passwords) > and library code directly in the DocumentRoot where browsers can access them > directly. Making the DocumentRoot a folder callled something like 'public' > is a great way remind you (the developer) that whatever you put in there is > public; the private parts of your app need to be elsewhere. Yes the > .htaccess solution would work, except for when you forget to do it. :-) exactly, this is why I think that over as well as parallel is basically not a good idea to make it the 'standard'. :) (Welcome to Nitro btw, seeing this is your first post here. :P) Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From william.full.moon at gmail.com Mon Dec 18 10:42:40 2006 From: william.full.moon at gmail.com (* William) Date: Tue, 19 Dec 2006 02:42:40 +1100 Subject: [Nitro] Nitro dir structure In-Reply-To: Message-ID: <00fa01c722bb$32e6be40$0201a8c0@ghostgum> Hi everyone ... I have two observations on this discussion. One concerns this repost, ... >> This looks like a strong argument against that sort of directory structure. Applications in >> general should be secure by default. For Yes the first rule on any web-application, intra-net or extra-net and database application must be "allow for no harm". Secondus ... I wonder if things are becoming "opportunities" instead of "designs"? By this I am pondering the intention behind some of the discussion. To me it makes some sense to take one or two steps back, away from what exists Now -- to imagine what is Desired. Here is an example from an application developer perspective. I would like a framework that allows "me" to have several APP-s that might share some core technology. Within an APP-lication I will want to share application specifics (such as user-data, etc.). At the same time I may -- in fact as far as I can tell, I would ALWAYS want to have some application specific code divorced from content or customisation. And also, always I want some form of version control over both the web site (meaning multiple applications across many frameworks) and for each application (under any particular framework, e.g. Nitro, Rails, Plone, PHP, other(s)) To me that leaves me with a couple of choices. SERVER -- FRAMEWORK (e.g. php) -- php doesn't work like this unfortunately ~sigh~ -- FRAMEWORK (e.g. nitro) -- core -- parts -- config ... other stuff ... -- FRAMEWORK (e.g. rails) -- MODULE ONE -- parts -- config -- APPLICATION ONE -- (sharing/include from module one) -- APPLICATION TWO -- (sharing/include from module one) -- (sharing/include from module two) -- parts -- config -- MODULE TWO -- (library of stuff) -- parts -- config -- APPLICATION THREE -- (sharing/include from module two) -- APPLICATION FOUR -- (sharing/include from module one) -- (sharing/include from module two) I realise that looks very messy. Off-hand the closest thing I can liken it to is something like PEARs, where you have re-usable modules that are on a par with an application. In fact an application might use more than one module (see Application 2 and application 4). It is a little general. In my experience one can always implement a degenerate case of a sophisticated structure -- it is virtually impossible to provide a sophisticated structure from a degenerate implementation. So it is worth considering a really "where do I need to get to" organisation as soon as you may. There are several other comments to add/consider. Such as a heterogeneous approach to how data is maintained -- Something like a CSS library for user-specific layouts might be kept 'under' and shared via a module. Yet an application could opt to re-use the code: and maintain data within the owner's application context. That would be the purpose of those areas called "config". One more real-world spanner. In many cases one is faced with several clients using more or less the same application. Such that what I've depicted above is more like "Instance One", "Instance Two" instead of "application one", etc. I invite you to think about those ideas please. While the example itself is very weak, the idea is to enforce and encourage independence, reuse and intra-dependence. Happy Christmas and . . . Kiearr'wo, Will. MBA, B.Math, STI, NLP MP _My_Bright_Ideas___________________________________________________abn:77425 560110_ "You must BE the change you want." ~ Gandhi Success Coaching http://adroit-process.blogspot.com ph/sms: 041-476-1217 (?2006) Information proprietary and confidential intended for direct recipient(s) and mutually agreed correspondents. Kiearr'wo -- "Everything is beautiful, sweet, delicious, happy, good". The usual greeting & departure of the Ya-idt'midtung people. ________________________________ From: nitro-general-bounces at rubyforge.org [mailto:nitro-general-bounces at rubyforge.org] On Behalf Of Wes Sheldahl Sent: Tuesday, 19 December 2006 01:02 To: General discussion about Nitro Subject: Re: [Nitro] Nitro dir structure Importance: Low On 12/17/06, Jonathan Buch wrote: Hi, > so perhaps this pattern would work for nitro too? but in nitro's case > the distribution (and versioning) actually includes this website > container as the starter app? in other words maybe the proto app should > exist over the nitro system rather than under or parallel to it. just a > thought. This layout would imply that the whole tree (app source) would be accessable from the outside (browser) if you don't deny access to that folder by using an Apache .htaccess file or other configuration options provided by the webserver. This looks like a strong argument against that sort of directory structure. Applications in general should be secure by default. For web applications, this means generally not putting config files (which often have passwords) and library code directly in the DocumentRoot where browsers can access them directly. Making the DocumentRoot a folder callled something like 'public' is a great way remind you (the developer) that whatever you put in there is public; the private parts of your app need to be elsewhere. Yes the .htaccess solution would work, except for when you forget to do it. :-) -- Wes Sheldahl wes.sheldahl at gmail.com -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.432 / Virus Database: 268.15.20/588 - Release Date: 15-Dec-2006 10:02 From transfire at gmail.com Mon Dec 18 10:44:27 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Mon, 18 Dec 2006 15:44:27 -0000 Subject: [Nitro] Nitro dir structure In-Reply-To: References: <1166382527.366073.65930@80g2000cwy.googlegroups.com> Message-ID: <1166456667.513326.73720@73g2000cwn.googlegroups.com> Wes Sheldahl wrote: > On 12/17/06, Jonathan Buch wrote: > > > > Hi, > > > > > so perhaps this pattern would work for nitro too? but in nitro's case > > > the distribution (and versioning) actually includes this website > > > container as the starter app? in other words maybe the proto app should > > > exist over the nitro system rather than under or parallel to it. just a > > > thought. > > > > This layout would imply that the whole tree (app source) would be > > accessable from the outside (browser) if you don't deny access to that > > folder by using an Apache .htaccess file or other configuration options > > provided by the webserver. > > > This looks like a strong argument against that sort of directory structure. > Applications in general should be secure by default. For web applications, > this means generally not putting config files (which often have passwords) > and library code directly in the DocumentRoot where browsers can access them > directly. Making the DocumentRoot a folder callled something like 'public' > is a great way remind you (the developer) that whatever you put in there is > public; the private parts of your app need to be elsewhere. Yes the > .htaccess solution would work, except for when you forget to do it. :-) i think its a valid consideration, but i don't think it's "strong" for a couple of reasons. first, doesn't nitro do url rewriting? and wouldn't that in itself effect access? in anycase, using a 'public' subdir doesn't neccessarily make the files above it secure --indeed you may be getting a false sense of security. so exomplict access controls can be a good thing. so i would suggest considering the merits of this layout on other grounds first. using .htaccess or other means to lock down access to a 'private' section isn't difficult, it's probably a good thing for it's explictness (think enterprise quality) and i imagine it can be handled automatically when using 'gen' in most cases. t. From m.fellinger at gmail.com Mon Dec 18 11:07:55 2006 From: m.fellinger at gmail.com (Michael Fellinger) Date: Tue, 19 Dec 2006 01:07:55 +0900 Subject: [Nitro] [PATCH] support for :store => :psql Message-ID: <9c00d3e00612180807jb9d74f2q4d72a0c1c12a771c@mail.gmail.com> one thing that constantly comes up again is that the name of the store has changed to :postgresql... due to popular demand we are able to support :psql again :) -------------- next part -------------- A non-text attachment was scrubbed... Name: bundle.tar.gz Type: application/x-gzip Size: 934 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061219/3787fa96/attachment.gz From john at oxyliquit.de Mon Dec 18 11:33:19 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Mon, 18 Dec 2006 17:33:19 +0100 Subject: [Nitro] Nitro dir structure In-Reply-To: <1166456667.513326.73720@73g2000cwn.googlegroups.com> References: <1166382527.366073.65930@80g2000cwy.googlegroups.com> <1166456667.513326.73720@73g2000cwn.googlegroups.com> Message-ID: Hi. > i think its a valid consideration, but i don't think it's "strong" for > a couple of reasons. > first, doesn't nitro do url rewriting? and wouldn't that in itself > effect access? Yes, but nitro is one level too deep. The server (webrick, apache, nginx) provides the access to the public directory to serve those files (since static) as fast as it can. Not sure on how deep your knowledge on Nitro internals is, but Nitro itself does not serve static files. So any request which comes through to nitro runs through the router (if a route for that 'url' exists) or gets mapped to a function inside a controller. So this whole process is quite 'costly' so it's better that nitro doesn't serve static files. > in anycase, using a 'public' subdir doesn't neccessarily make the files > above it secure --indeed you may be getting a false sense of security. I would argue that making the application files unreachable from 'outside' by 'standard methods' does make things more secure by default. > so explicit access controls > can be a good thing. so i would suggest considering the merits of this > layout on other grounds first. using .htaccess or other means to lock > down access to a 'private' section isn't difficult, it's probably a > good thing for it's explictness (think enterprise quality) and i > imagine it can be handled automatically when using 'gen' in most cases. The .haccess itself wouldn't be a problem to generate. But I'm not sure on how to do the same with a nginx config or in webrick or even mongrel. In any case, putting the public directory below the nitro app is actually better to control than right besides (which would require more rules from the server to prevent access). Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From wyhaines at gmail.com Mon Dec 18 12:10:49 2006 From: wyhaines at gmail.com (Kirk Haines) Date: Mon, 18 Dec 2006 10:10:49 -0700 Subject: [Nitro] Nitro dir structure In-Reply-To: References: <1166382527.366073.65930@80g2000cwy.googlegroups.com> <1166456667.513326.73720@73g2000cwn.googlegroups.com> Message-ID: On 12/18/06, Jonathan Buch wrote: > > in anycase, using a 'public' subdir doesn't neccessarily make the files > > above it secure --indeed you may be getting a false sense of security. > > I would argue that making the application files unreachable from 'outside' > by 'standard methods' does make things more secure by default. I completely agree on this point. Long ago, in the bad old days, IOWA apps lived in the web server docroot. Sure, one could alter the web server configuration to make sure that things that people should not have access to (raw code, configs, etc...) were kept away from them, but all sorts of fun threatened to break loose if one forgot to do that, or made a mistake with that and didn't catch it. Suddenly things that shouldn't be public were. Not good. Now I use a structure that is much more secure by default. The app is outside of the docroot. The webservers don't care, and by default one isn't risking revealing details best left unrevealed. If there is something in the application directory that I want to expose in the docroot, I can explicitly choose to do that via symlinks or webserver configuration. That is the rarer situation, so it makes sense to me that it is the situation that requires extra work, instead of the other way around, where extra work may be required just to make sure there is some basic security in place. > The .haccess itself wouldn't be a problem to generate. But I'm not sure > on how to do the same with a nginx config or in webrick or even mongrel. > > In any case, putting the public directory below the nitro app is actually > better to control than right besides (which would require more rules from > the server to prevent access). Exactly. .htaccess is only relevant to a single web server. Nitro supports multiple web servers, so standard deployment structures and security considerations should be designed to work as much as possible across multiple web servers without any changes to the app. Kirk Haines From george.moschovitis at gmail.com Mon Dec 18 13:51:30 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Mon, 18 Dec 2006 20:51:30 +0200 Subject: [Nitro] Nitro dir structure In-Reply-To: <40b05ebe0612180715n2173f47j132fe06197562435@mail.gmail.com> References: <40b05ebe0612180715n2173f47j132fe06197562435@mail.gmail.com> Message-ID: Yeah, I have a script that does that, perhaps we can integrate this in gen. -g. On 12/18/06, Alexandre Gravem wrote: > > > I see that this can have advantages... can we let the proto dir (as a kind > > of minimal 'how a nitro app looks like') alone and push the functionality > > of putting the whole nitro app into a lib/ dir into --gen ? > > > > nitro --gen --include-nitro --include-og newproject > > > > This would of course add glue as well. It maybe could create a run.rb > with > > $:.unshift(File.dirname (__FILE__)+'/lib')... How about it? > > I liked this idea very much!! In fact thats kind of what i am doing in my > projects. > > A. Gravem > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From john at oxyliquit.de Mon Dec 18 14:01:50 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Mon, 18 Dec 2006 20:01:50 +0100 Subject: [Nitro] Nitro dir structure In-Reply-To: References: <40b05ebe0612180715n2173f47j132fe06197562435@mail.gmail.com> Message-ID: Hi, > Yeah, I have a script that does that, perhaps we can integrate this in gen. or, like proposed, remove gen, make nitro more powerful. :P Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From george.moschovitis at gmail.com Mon Dec 18 14:03:00 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Mon, 18 Dec 2006 21:03:00 +0200 Subject: [Nitro] Nitro Prototype helpers Message-ID: Dear devs, does anyone use the nitro Prorotype javascript helpers? I would like to remove them. It seems to me they are not really useful. You can use JQuery to do whatever a (rails/rjs) system does only better. What do you think? -g. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From john at oxyliquit.de Mon Dec 18 14:33:05 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Mon, 18 Dec 2006 20:33:05 +0100 Subject: [Nitro] Nitro Prototype helpers In-Reply-To: References: Message-ID: Hi, > does anyone use the nitro Prorotype javascript helpers? I would like > to remove them. It seems to me they are not really useful. You can use > JQuery to do whatever a (rails/rjs) system does only better. could you add the jquery, but only deprecate the prototype helpers? class MyController < Nitro::Controller helper :javascript_prototype end We can remove those then in a few iterations after people who still use prototype switched to jquery. I removed javascript from Oxy just because it creates more load... If I had a better system I'd probably still using it. :P Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From george.moschovitis at gmail.com Mon Dec 18 14:45:25 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Mon, 18 Dec 2006 21:45:25 +0200 Subject: [Nitro] Nitro Prototype helpers In-Reply-To: References: Message-ID: there is no need to add any helpers for jquery ;-) it is as simple as it gets, adding an extra layer of ruby above jquery would only complicate things. Jquery may not be the best design but it works very well with the current nitro design (this is not perfect either ;-)). -g. On 12/18/06, Jonathan Buch wrote: > Hi, > > > does anyone use the nitro Prorotype javascript helpers? I would like > > to remove them. It seems to me they are not really useful. You can use > > JQuery to do whatever a (rails/rjs) system does only better. > > could you add the jquery, but only deprecate the prototype helpers? > > class MyController < Nitro::Controller > helper :javascript_prototype > end > > We can remove those then in a few iterations after people who still use > prototype switched to jquery. > > I removed javascript from Oxy just because it creates more load... If I > had a better system I'd probably still using it. :P > > Jo > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From john at oxyliquit.de Tue Dec 19 03:36:24 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Tue, 19 Dec 2006 09:36:24 +0100 Subject: [Nitro] Nitro Prototype helpers In-Reply-To: References: Message-ID: <20061219083623.GA5273@oxyliquit.de> Hi, > there is no need to add any helpers for jquery ;-) it is as simple as > it gets, adding an extra layer of ruby above jquery would only > complicate things. Jquery may not be the best design but it works very > well with the current nitro design (this is not perfect either ;-)). Ah, this is good. :P Then I propose to add a big 'will be deprecated' mark in the next release and remove it after that. Jo From george.moschovitis at gmail.com Tue Dec 19 04:37:44 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Tue, 19 Dec 2006 11:37:44 +0200 Subject: [Nitro] Nitro Prototype helpers In-Reply-To: <20061219083623.GA5273@oxyliquit.de> References: <20061219083623.GA5273@oxyliquit.de> Message-ID: Ok. On 12/19/06, Jonathan Buch wrote: > Hi, > > there is no need to add any helpers for jquery ;-) it is as simple as > > it gets, adding an extra layer of ruby above jquery would only > > complicate things. Jquery may not be the best design but it works very > > well with the current nitro design (this is not perfect either ;-)). > > Ah, this is good. :P Then I propose to add a big 'will be deprecated' > mark in the next release and remove it after that. > > Jo > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Tue Dec 19 17:11:29 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Wed, 20 Dec 2006 00:11:29 +0200 Subject: [Nitro] Facets to add Message-ID: Tom, here are some nice ideas to add to Facets: http://errtheblog.com/post/44 -g. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From transfire at gmail.com Wed Dec 20 00:44:19 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Wed, 20 Dec 2006 05:44:19 -0000 Subject: [Nitro] Facets to add In-Reply-To: References: Message-ID: <1166593459.792266.43990@80g2000cwy.googlegroups.com> George Moschovitis wrote: > Tom, > > here are some nice ideas to add to Facets: > > http://errtheblog.com/post/44 thanks. i'll have a look (at a quick glance i think facets has most of these). btw, facets got a little write-up: http://www.oreillynet.com/ruby/blog/2006/12/save_time_and_effort_with_face.html t. From george.moschovitis at gmail.com Wed Dec 20 01:36:08 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Wed, 20 Dec 2006 08:36:08 +0200 Subject: [Nitro] Facets to add In-Reply-To: <1166593459.792266.43990@80g2000cwy.googlegroups.com> References: <1166593459.792266.43990@80g2000cwy.googlegroups.com> Message-ID: > http://www.oreillynet.com/ruby/blog/2006/12/save_time_and_effort_with_face.html thats great :) -g. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From m.fellinger at gmail.com Wed Dec 20 04:20:16 2006 From: m.fellinger at gmail.com (Michael Fellinger) Date: Wed, 20 Dec 2006 18:20:16 +0900 Subject: [Nitro] Nitro Prototype helpers In-Reply-To: <20061219083623.GA5273@oxyliquit.de> References: <20061219083623.GA5273@oxyliquit.de> Message-ID: <9c00d3e00612200120pa698afaj7663f1b48bf32ec8@mail.gmail.com> On 12/19/06, Jonathan Buch wrote: > Hi, > > there is no need to add any helpers for jquery ;-) it is as simple as > > it gets, adding an extra layer of ruby above jquery would only > > complicate things. Jquery may not be the best design but it works very > > well with the current nitro design (this is not perfect either ;-)). > > Ah, this is good. :P Then I propose to add a big 'will be deprecated' > mark in the next release and remove it after that. Do we need to deprecate the prototype-helpers? i would imagine that one can use either... also avoiding changes. I mean, the helpers are not broken or anything, so why remove the functionality? it is just that one has to use either one, since they conflict with each other. how about aliasing the current :javascript helper to be :prototype and add the :jquery helper as well? not so sure about additional helpers which build upon these... are there any? i think i remember something... gotta investigate :) ^manveru > > Jo > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > From george.moschovitis at gmail.com Wed Dec 20 08:18:35 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Wed, 20 Dec 2006 15:18:35 +0200 Subject: [Nitro] Nitro Prototype helpers In-Reply-To: <9c00d3e00612200120pa698afaj7663f1b48bf32ec8@mail.gmail.com> References: <20061219083623.GA5273@oxyliquit.de> <9c00d3e00612200120pa698afaj7663f1b48bf32ec8@mail.gmail.com> Message-ID: I would like to simplify the framework, and in the process remove some code that is not really useful and is hard to maintain. I don't think that the prototype helpers are that useful. Perhaps we can provide them as nitro plugins (or nitro components). Anw, this is food for thoughts... -g. On 12/20/06, Michael Fellinger wrote: > On 12/19/06, Jonathan Buch wrote: > > Hi, > > > there is no need to add any helpers for jquery ;-) it is as simple as > > > it gets, adding an extra layer of ruby above jquery would only > > > complicate things. Jquery may not be the best design but it works very > > > well with the current nitro design (this is not perfect either ;-)). > > > > Ah, this is good. :P Then I propose to add a big 'will be deprecated' > > mark in the next release and remove it after that. > > Do we need to deprecate the prototype-helpers? > i would imagine that one can use either... also avoiding changes. > I mean, the helpers are not broken or anything, so why remove the functionality? > it is just that one has to use either one, since they conflict with each other. > > how about aliasing the current :javascript helper to be :prototype and > add the :jquery helper as well? > > not so sure about additional helpers which build upon these... are > there any? i think i remember something... gotta investigate :) > > ^manveru > > > > > Jo > > > > _______________________________________________ > > Nitro-general mailing list > > Nitro-general at rubyforge.org > > http://rubyforge.org/mailman/listinfo/nitro-general > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From lists at zarac.com Wed Dec 20 11:26:52 2006 From: lists at zarac.com (Ric Turley) Date: Wed, 20 Dec 2006 09:26:52 -0700 Subject: [Nitro] [BUG] Can't create new member on nitroproject.org In-Reply-To: Message-ID: Let me try asking a different way. I am trying to get a handle on what Nitro can do, but much of the documentation (like the Nitro Guide) is in the wiki which denies me access to most pages until I register. My attempts to register are met with the error below. How do I get access to the documentation in the wiki? Thanks Ric Turley > From: Ric Turley > Reply-To: General discussion about Nitro > Date: Sun, 17 Dec 2006 09:51:04 -0700 > To: General discussion about Nitro > Subject: Re: [Nitro] [BUG] Can't create new member on nitroproject.org > > I can't get registered. Entered email address, got confirmation email and > entered token, but get a page with only '(error)' after entering name and > password: > http://www.nitroproject.org/users/create?e=ric%40zarac.com&t=NI2bNK5BD%2F2KQ > > Thanks! > Ric Turley > > >> From: "George Moschovitis" >> Reply-To: General discussion about Nitro >> Date: Thu, 7 Dec 2006 11:44:25 +0200 >> To: "General discussion about Nitro" >> Subject: Re: [Nitro] [BUG] Can't create new member on nitroproject.org >> >>> Great to hear that 0.41 is on the way. >> >> I have updated the repo with an almost final version of 0.41.0. there >> are 2 test cases that don't pass. Jonathan could you also have a look >> at them? >> >> -g. >> >> -- >> http://cull.gr >> http://www.joy.gr >> http://blog.gmosx.com >> http://nitroproject.org >> _______________________________________________ >> Nitro-general mailing list >> Nitro-general at rubyforge.org >> http://rubyforge.org/mailman/listinfo/nitro-general >> >> > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > From dan at tastapod.com Wed Dec 20 12:14:23 2006 From: dan at tastapod.com (Dan North) Date: Wed, 20 Dec 2006 17:14:23 +0000 Subject: [Nitro] Nitro dir structure In-Reply-To: References: <40b05ebe0612180715n2173f47j132fe06197562435@mail.gmail.com> Message-ID: <45896F6F.30705@tastapod.com> One tiny suggestion: Rather than --gen, why not have the second param be a command, like subversion or apachectl? So you'd have nitro start --opts myapp, nitro stop --opts myapp, nitro gen --opts myapp, etc. The "gen" isn't an option to nitro as such - it's an operating mode. Cheers, Dan Jonathan Buch wrote: > Hi, > > >> Yeah, I have a script that does that, perhaps we can integrate this in gen. >> > > or, like proposed, remove gen, make nitro more powerful. :P > > Jo > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061220/a83bfe32/attachment.html From vikingtux at gmail.com Wed Dec 20 15:11:22 2006 From: vikingtux at gmail.com (Alexandre Gravem) Date: Wed, 20 Dec 2006 18:11:22 -0200 Subject: [Nitro] Nitro Prototype helpers In-Reply-To: References: <20061219083623.GA5273@oxyliquit.de> <9c00d3e00612200120pa698afaj7663f1b48bf32ec8@mail.gmail.com> Message-ID: <40b05ebe0612201211u2197d5f1maab3d25f40e23fc3@mail.gmail.com> I agree that Prototype helpers aren't that useful and jQuery should be the "default" js library for Nitro. And it's so easy to create a plugins and get the helpers, that one who miss them can easily create the plugin and post it in LoxParts ;-) On 12/20/06, George Moschovitis wrote: > > I would like to simplify the framework, and in the process remove some > code that is not really useful and is hard to maintain. I don't think > that the prototype helpers are that useful. Perhaps we can provide > them as nitro plugins (or nitro components). Anw, this is food for > thoughts... > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061220/87bf4c7e/attachment.html From george.moschovitis at gmail.com Thu Dec 21 03:06:07 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 21 Dec 2006 10:06:07 +0200 Subject: [Nitro] [BUG] Can't create new member on nitroproject.org In-Reply-To: References: Message-ID: I thought the wiki was open for reading. Anyway, I will have a look at this today and get back to you. thanks for your patience, George. On 12/20/06, Ric Turley wrote: > Let me try asking a different way. > > I am trying to get a handle on what Nitro can do, but much of the > documentation (like the Nitro Guide) is in the wiki which denies me access > to most pages until I register. My attempts to register are met with the > error below. > > How do I get access to the documentation in the wiki? > > Thanks > Ric Turley > > > > From: Ric Turley > > Reply-To: General discussion about Nitro > > Date: Sun, 17 Dec 2006 09:51:04 -0700 > > To: General discussion about Nitro > > Subject: Re: [Nitro] [BUG] Can't create new member on nitroproject.org > > > > I can't get registered. Entered email address, got confirmation email and > > entered token, but get a page with only '(error)' after entering name and > > password: > > http://www.nitroproject.org/users/create?e=ric%40zarac.com&t=NI2bNK5BD%2F2KQ > > > > Thanks! > > Ric Turley > > > > > >> From: "George Moschovitis" > >> Reply-To: General discussion about Nitro > >> Date: Thu, 7 Dec 2006 11:44:25 +0200 > >> To: "General discussion about Nitro" > >> Subject: Re: [Nitro] [BUG] Can't create new member on nitroproject.org > >> > >>> Great to hear that 0.41 is on the way. > >> > >> I have updated the repo with an almost final version of 0.41.0. there > >> are 2 test cases that don't pass. Jonathan could you also have a look > >> at them? > >> > >> -g. > >> > >> -- > >> http://cull.gr > >> http://www.joy.gr > >> http://blog.gmosx.com > >> http://nitroproject.org > >> _______________________________________________ > >> Nitro-general mailing list > >> Nitro-general at rubyforge.org > >> http://rubyforge.org/mailman/listinfo/nitro-general > >> > >> > > > > _______________________________________________ > > Nitro-general mailing list > > Nitro-general at rubyforge.org > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Thu Dec 21 03:15:09 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 21 Dec 2006 10:15:09 +0200 Subject: [Nitro] Repo update Message-ID: Dear devs, the latest repo update contains a change that may require a minor change to make your apps run. I have made the Template.transformation stage to be just a stage in the transformation pipeline instead of beeing applied by default on every action. This way nitro now can support pluggable template engines. This change was suggested by manveru. regards, George. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Thu Dec 21 03:36:08 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 21 Dec 2006 10:36:08 +0200 Subject: [Nitro] Nitro Real World example Message-ID: Dear devs, as promised yesterday on #nitro, I am releasing a real world example of using Nitro. You can find the code in my latest blog entry: http://blog.gmosx.com/2006/12/21/real-world-nitro-example I hope this will be useful for you. regards, George. -- http://blog.gmosx.com http://cull.gr http://nitroproject.org http://www.joy.gr From john at oxyliquit.de Thu Dec 21 06:10:14 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Thu, 21 Dec 2006 12:10:14 +0100 Subject: [Nitro] [PATCH] Separate get_params and post_params In-Reply-To: References: Message-ID: Hi, > btw, why is it useful to separate get/post params? if u use them > through params (which in your latest implementation just selects from > one collection or the other anyway)? you may not remember it, but this idea appeared in the params thread and you asked me (or others) to implement this. :P The idea was the following: When using the POST method, also provide access to the GET parameters. This is now done as well kinda, but has drawbacks, as parameters can get overwritten. So I proposed the following (though I think it was in pseudo code): * GET * save GET params in get_params * point params to get_params * point params= to get_params= * POST * save POST params in post_params * save GET params in get_params * use get_params.update(post_params) as params (post precendence) * point params= to post_params (maybe merged params here as well?) Btw, I'm really happy that Noe asked where he could help a little around Nitro, also to 'get down to code' as I understood. That's why I gave him his little thing, and I'd be quite happy if I can get him to do more slav... eh, I mean little enhancements. ;D Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From john at oxyliquit.de Thu Dec 21 06:10:15 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Thu, 21 Dec 2006 12:10:15 +0100 Subject: [Nitro] Nitro dir structure In-Reply-To: <45896F6F.30705@tastapod.com> References: <40b05ebe0612180715n2173f47j132fe06197562435@mail.gmail.com> <45896F6F.30705@tastapod.com> Message-ID: Hi, > One tiny suggestion: > > Rather than --gen, why not have the second param be a command, like > subversion or apachectl? > > So you'd have nitro start --opts myapp, nitro stop --opts myapp, nitro > gen --opts myapp, etc. > > The "gen" isn't an option to nitro as such - it's an operating mode. Oh, yes, I like that even better. :) Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From john at oxyliquit.de Thu Dec 21 06:10:15 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Thu, 21 Dec 2006 12:10:15 +0100 Subject: [Nitro] [BUG] Can't create new member on nitroproject.org In-Reply-To: References: Message-ID: Hi, > Let me try asking a different way. > > I am trying to get a handle on what Nitro can do, but much of the > documentation (like the Nitro Guide) is in the wiki which denies me access > to most pages until I register. My attempts to register are met with the > error below. > > How do I get access to the documentation in the wiki? there's a good and a bad message now. The good one is, that you're not being left out from anything. The pages which require registering are actually empty and lead to a form where you can fill the page. That's why it asks for registration. Bad message is now, that there is even less documentation than you expected. George, can you make it visible (make the links gray) when the pages behind are not available/empty? Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From wes.sheldahl at gmail.com Thu Dec 21 08:20:17 2006 From: wes.sheldahl at gmail.com (Wes Sheldahl) Date: Thu, 21 Dec 2006 08:20:17 -0500 Subject: [Nitro] [BUG] Can't create new member on nitroproject.org In-Reply-To: References: Message-ID: On 12/21/06, Jonathan Buch wrote: > > Hi, > > > Let me try asking a different way. > > > > I am trying to get a handle on what Nitro can do, but much of the > > documentation (like the Nitro Guide) is in the wiki which denies me > access > > to most pages until I register. My attempts to register are met with the > > error below. > > > > How do I get access to the documentation in the wiki? > > there's a good and a bad message now. The good one is, that you're not > being > left out from anything. The pages which require registering are actually > empty and lead to a form where you can fill the page. That's why it asks > for registration. > Bad message is now, that there is even less documentation than you > expected. > > George, can you make it visible (make the links gray) when the pages > behind > are not available/empty? > > Jo Changing the link color would be good. Also, if it's not too much work, it might be helpful if the registration page said something to the effect that registration is required only to create or edit pages, but not to read them. That way if someone does click a (grey?) link, it will be more obvious that the documentation isn't there to read yet. -- Wes Sheldahl wes.sheldahl at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061221/2e465555/attachment.html From vikingtux at gmail.com Thu Dec 21 10:35:51 2006 From: vikingtux at gmail.com (Alexandre Gravem) Date: Thu, 21 Dec 2006 13:35:51 -0200 Subject: [Nitro] Repo update In-Reply-To: References: Message-ID: <40b05ebe0612210735u63e19bfcw6a6ae007a8b7d430@mail.gmail.com> This is an amazing change. A totally flexible transformation pipeline!! On 12/21/06, George Moschovitis wrote: > > Dear devs, > > the latest repo update contains a change that may require a minor > change to make your apps run. I have made the Template.transformation > stage to be just a stage in the transformation pipeline instead of > beeing applied by default on every action. This way nitro now can > support pluggable template engines. This change was suggested by > manveru. > > regards, > George. > > -- > http://blog.gmosx.com > http://cull.gr > http://www.joy.gr > http://nitroproject.org > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061221/bb8bb715/attachment.html From george.moschovitis at gmail.com Thu Dec 21 10:47:21 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 21 Dec 2006 17:47:21 +0200 Subject: [Nitro] Repo update In-Reply-To: <40b05ebe0612210735u63e19bfcw6a6ae007a8b7d430@mail.gmail.com> References: <40b05ebe0612210735u63e19bfcw6a6ae007a8b7d430@mail.gmail.com> Message-ID: On 12/21/06, Alexandre Gravem wrote: > This is an amazing change. A totally flexible transformation pipeline!! I am glad you like this (but, thank manv for bugging me to do this), more cool features are in the pipeline. -g. > > > On 12/21/06, George Moschovitis < george.moschovitis at gmail.com> wrote: > > > > Dear devs, > > > > the latest repo update contains a change that may require a minor > > change to make your apps run. I have made the Template.transformation > > stage to be just a stage in the transformation pipeline instead of > > beeing applied by default on every action. This way nitro now can > > support pluggable template engines. This change was suggested by > > manveru. > > > > regards, > > George. > > > > -- > > http://blog.gmosx.com > > http://cull.gr > > http://www.joy.gr > > http://nitroproject.org > > _______________________________________________ > > Nitro-general mailing list > > Nitro-general at rubyforge.org > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Thu Dec 21 10:55:35 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 21 Dec 2006 17:55:35 +0200 Subject: [Nitro] Nitro dir structure In-Reply-To: References: <40b05ebe0612180715n2173f47j132fe06197562435@mail.gmail.com> <45896F6F.30705@tastapod.com> Message-ID: Will do something like that... -g. On 12/21/06, Jonathan Buch wrote: > Hi, > > > One tiny suggestion: > > > > Rather than --gen, why not have the second param be a command, like > > subversion or apachectl? > > > > So you'd have nitro start --opts myapp, nitro stop --opts myapp, nitro > > gen --opts myapp, etc. > > > > The "gen" isn't an option to nitro as such - it's an operating mode. > > Oh, yes, I like that even better. :) > > Jo > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From m.fellinger at gmail.com Thu Dec 21 14:56:50 2006 From: m.fellinger at gmail.com (Michael Fellinger) Date: Fri, 22 Dec 2006 04:56:50 +0900 Subject: [Nitro] Repo update In-Reply-To: <40b05ebe0612210735u63e19bfcw6a6ae007a8b7d430@mail.gmail.com> References: <40b05ebe0612210735u63e19bfcw6a6ae007a8b7d430@mail.gmail.com> Message-ID: <9c00d3e00612211156qd1a3213l66c244c70ca7c2f3@mail.gmail.com> On 12/22/06, Alexandre Gravem wrote: > This is an amazing change. A totally flexible transformation pipeline!! Now, here comes your sugar-topping: http://pastie.caboo.se/29038 after tinkering around, i found this is the way it can be used most efficient... but G. could you explain why we need to give something eval-ready? especially in the case of erubis, this is just slowing it down... ^manveru > > On 12/21/06, George Moschovitis < george.moschovitis at gmail.com> wrote: > > Dear devs, > > > > the latest repo update contains a change that may require a minor > > change to make your apps run. I have made the Template.transformation > > stage to be just a stage in the transformation pipeline instead of > > beeing applied by default on every action. This way nitro now can > > support pluggable template engines. This change was suggested by > > manveru. > > > > regards, > > George. > > > > -- > > http://blog.gmosx.com > > http://cull.gr > > http://www.joy.gr > > http://nitroproject.org > > _______________________________________________ > > Nitro-general mailing list > > Nitro-general at rubyforge.org > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > From george.moschovitis at gmail.com Thu Dec 21 18:34:52 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Fri, 22 Dec 2006 01:34:52 +0200 Subject: [Nitro] Repo update In-Reply-To: <9c00d3e00612211156qd1a3213l66c244c70ca7c2f3@mail.gmail.com> References: <40b05ebe0612210735u63e19bfcw6a6ae007a8b7d430@mail.gmail.com> <9c00d3e00612211156qd1a3213l66c244c70ca7c2f3@mail.gmail.com> Message-ID: > efficient... but G. could you explain why we need to give something > eval-ready? > especially in the case of erubis, this is just slowing it down... what do you mean? -g. > > ^manveru > > > > > On 12/21/06, George Moschovitis < george.moschovitis at gmail.com> wrote: > > > Dear devs, > > > > > > the latest repo update contains a change that may require a minor > > > change to make your apps run. I have made the Template.transformation > > > stage to be just a stage in the transformation pipeline instead of > > > beeing applied by default on every action. This way nitro now can > > > support pluggable template engines. This change was suggested by > > > manveru. > > > > > > regards, > > > George. > > > > > > -- > > > http://blog.gmosx.com > > > http://cull.gr > > > http://www.joy.gr > > > http://nitroproject.org > > > _______________________________________________ > > > Nitro-general mailing list > > > Nitro-general at rubyforge.org > > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > > > > > _______________________________________________ > > Nitro-general mailing list > > Nitro-general at rubyforge.org > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From noe.rubinstein at gmail.com Sat Dec 23 09:06:50 2006 From: noe.rubinstein at gmail.com (=?UTF-8?Q?No=C3=A9_Rubinstein?=) Date: Sat, 23 Dec 2006 15:06:50 +0100 Subject: [Nitro] [BUG] Can't create new member on nitroproject.org In-Reply-To: References: Message-ID: mmh, It's a bug with "-" and "_" in page adresses. Links are given like this : http://www.nitroproject.org/wiki/pages/read/Documentation__Nitro_Guide wich redirect there : http://www.nitroproject.org/wiki/pages/create/Documentation__Nitro_Guide but the page is here : http://www.nitroproject.org/wiki/pages/read/Documentation__Nitro-Guide And even when you create the page http://www.nitroproject.org/wiki/pages/read/Documentation__Nitro_Guide , the page created is http://www.nitroproject.org/wiki/pages/read/Documentation__Nitro-Guide 2006/12/21, Wes Sheldahl : > On 12/21/06, Jonathan Buch wrote: > > Hi, > > > > > Let me try asking a different way. > > > > > > I am trying to get a handle on what Nitro can do, but much of the > > > documentation (like the Nitro Guide) is in the wiki which denies me > access > > > to most pages until I register. My attempts to register are met with the > > > error below. > > > > > > How do I get access to the documentation in the wiki? > > > > there's a good and a bad message now. The good one is, that you're not > being > > left out from anything. The pages which require registering are actually > > empty and lead to a form where you can fill the page. That's why it asks > > for registration. > > Bad message is now, that there is even less documentation than you > expected. > > > > George, can you make it visible (make the links gray) when the pages > behind > > are not available/empty? > > > > Jo > > Changing the link color would be good. Also, if it's not too much work, it > might be helpful if the registration page said something to the effect that > registration is required only to create or edit pages, but not to read them. > That way if someone does click a (grey?) link, it will be more obvious > that the documentation isn't there to read yet. > > > > -- > Wes Sheldahl > wes.sheldahl at gmail.com > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > From george.moschovitis at gmail.com Sun Dec 24 07:31:22 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sun, 24 Dec 2006 14:31:22 +0200 Subject: [Nitro] Request for help Message-ID: Dear devs, is someone on this list experienced enough to convert this html filtering code from php to Ruby? Nitro's current sanitize method sucks. I have update my repo to use Rick Olson's WhiteList Rails plugin code but this version too doesn't handle unballanced tags. I really hope someone with php experience will take the time to conver this and make Nitro so much better. -g. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Sun Dec 24 07:32:44 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sun, 24 Dec 2006 14:32:44 +0200 Subject: [Nitro] Request for help In-Reply-To: References: Message-ID: Btw, I am referring to this piece of code: http://code.iamcal.com/php/lib_filter/ Also if someone knows of a similar library available for Ruby let me know. -g. On 12/24/06, George Moschovitis wrote: > Dear devs, > > is someone on this list experienced enough to convert this html > filtering code from php to Ruby? Nitro's current sanitize method > sucks. I have update my repo to use Rick Olson's WhiteList Rails > plugin code but this version too doesn't handle unballanced tags. > > I really hope someone with php experience will take the time to conver > this and make Nitro so much better. > > -g. > > -- > http://blog.gmosx.com > http://cull.gr > http://www.joy.gr > http://nitroproject.org > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Sun Dec 24 07:45:15 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sun, 24 Dec 2006 14:45:15 +0200 Subject: [Nitro] Request for help In-Reply-To: References: Message-ID: Another thing, I will add automatic html filtering to all string serializable attributes in the attributeutils populate_object method used in request.fill etc. -g. On 12/24/06, George Moschovitis wrote: > Btw, > > I am referring to this piece of code: > > http://code.iamcal.com/php/lib_filter/ > > Also if someone knows of a similar library available for Ruby let me know. > > -g. > > On 12/24/06, George Moschovitis wrote: > > Dear devs, > > > > is someone on this list experienced enough to convert this html > > filtering code from php to Ruby? Nitro's current sanitize method > > sucks. I have update my repo to use Rick Olson's WhiteList Rails > > plugin code but this version too doesn't handle unballanced tags. > > > > I really hope someone with php experience will take the time to conver > > this and make Nitro so much better. > > > > -g. > > > > -- > > http://blog.gmosx.com > > http://cull.gr > > http://www.joy.gr > > http://nitroproject.org > > > > > -- > http://blog.gmosx.com > http://cull.gr > http://www.joy.gr > http://nitroproject.org > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From transfire at gmail.com Sun Dec 24 11:24:43 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Sun, 24 Dec 2006 16:24:43 -0000 Subject: [Nitro] Ratchets Message-ID: <1166977483.673491.75250@48g2000cwx.googlegroups.com> As many of you are aware I've been working on a new build tool to replace Reap, called Ratchets. I have had nearly all the core functionality in place for a while now, but I have been struggling with the "engine" that drives the system. I've set my bar fairly high, as I wanted to provide something more capable, and even more elegant than rake if possible (NOTE of course you can easily use Rathcets' via Rake if you prefer). It took quite a bit of deep consideration (and frustration), but I think that I have finally arrived at the nice solution. I would like to share it with you to get your take and to make sure I haven't overlooked anything. ... and I figure some of you could use a break from the Holiday pageantry ;-) Let's say you have a project: myproject/ README ProjectInfo lib/ bin/ script/ As with Reap, Ratchets uses a ProjectInfo file, but rather than define tasks in this file, they are defined as stand-alone scripts in a separate folder -- I haven't yet settled on a conventional name for this folder, one can set the name in the ProjectInfo file, but for now I'm calling it script/. Ratchets build scripts are ruby scripts, of course. But what's elegant about them is how little they differ from any ruby script --there is very little special notation. Here's a stupid example: # script/tryme.rb def tryme puts "It worked!" end Then on the command line: $ project tryme It worked! I know. This seems too simple. How will we deal with prerequisites? How do we call other targets if each one is in a separate file? Well, that's were the elegance come into play. Every _public_ method is dealt with as a target (eg. Ratchets' term for task). So public methods are never triggered more than once per run session. Have a look at this: # script/tryme.rb def tryme sayit sayit sayit end def sayit puts "It worked again!" end Now: $ project tryme It worked again! Notice "It worked again!" only printed once. But if 'sayit' were made a private method instead, then it would print three times. To access targets in the other files you can simply require them as normal. Say we have a target script called more.rb: # script/more.rb def more puts "more..." end Then in tryme.rb: require 'more' def tryme sayit more end At this point, you may be wondering why bother with the separate scripts. There are two reasons. Firstly they act as namespaces. Instead of 'require' we can use 'utilize'. Let say more.rb has three target methods more1, more2 and more3. Then: utilize 'more' def tryme more.more1 more.more2 more.more3 end The effect of this is that we can simply drop build scripts into our script directly, and not worry about any name clashes between target method names by using #utilize, instead or #require. Everything is kept in well separated modules. Now Ratchets has a bunch of built-in build scripts already, and these scripts do not differ in construction from personal build scripts --another elegance which make it very easy to donate build scripts to Rathcets. I could go one and layout a few other details and features, like targets ability to take parameters and access to projectinfo data, but I've covered the core of the build system, and this is getting long, so I'll leave it there. Happy Holidays, T. From transfire at gmail.com Sun Dec 24 11:34:45 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Sun, 24 Dec 2006 16:34:45 -0000 Subject: [Nitro] Request for help In-Reply-To: References: Message-ID: <1166978085.809422.22010@73g2000cwn.googlegroups.com> George Moschovitis wrote: > Btw, > > I am referring to this piece of code: > > http://code.iamcal.com/php/lib_filter/ > > Also if someone knows of a similar library available for Ruby let me know. What does it do? Is it like http://rubyforge.org/projects/tidy ? It's not long, I could probably translate it for you in short order (assuming there's nothing out of the ordinary in there). T. From transfire at gmail.com Sun Dec 24 12:18:15 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Sun, 24 Dec 2006 17:18:15 -0000 Subject: [Nitro] Request for help In-Reply-To: <1166978085.809422.22010@73g2000cwn.googlegroups.com> References: <1166978085.809422.22010@73g2000cwn.googlegroups.com> Message-ID: <1166980695.673202.261850@a3g2000cwd.googlegroups.com> transfire at gmail.com wrote: > George Moschovitis wrote: > > Btw, > > > > I am referring to this piece of code: > > > > http://code.iamcal.com/php/lib_filter/ > > > > Also if someone knows of a similar library available for Ruby let me know. > > What does it do? Is it like http://rubyforge.org/projects/tidy ? > > It's not long, I could probably translate it for you in short order > (assuming there's nothing out of the ordinary in there). Okay. Let me know if you find anything else/better. But if not, I've already translated about half of it. I just need to look up a couple things to finish it. T. From james.britt at gmail.com Sun Dec 24 13:59:05 2006 From: james.britt at gmail.com (James Britt) Date: Sun, 24 Dec 2006 11:59:05 -0700 Subject: [Nitro] Request for help In-Reply-To: <1166980695.673202.261850@a3g2000cwd.googlegroups.com> References: <1166978085.809422.22010@73g2000cwn.googlegroups.com> <1166980695.673202.261850@a3g2000cwd.googlegroups.com> Message-ID: <458ECDF9.90800@gmail.com> transfire at gmail.com wrote: > transfire at gmail.com wrote: >> George Moschovitis wrote: >>> Btw, >>> >>> I am referring to this piece of code: >>> >>> http://code.iamcal.com/php/lib_filter/ >>> >>> Also if someone knows of a similar library available for Ruby let me know. >> What does it do? Is it like http://rubyforge.org/projects/tidy ? >> >> It's not long, I could probably translate it for you in short order >> (assuming there's nothing out of the ordinary in there). > > Okay. Let me know if you find anything else/better. But if not, I've > already translated about half of it. I just need to look up a couple > things to finish it. http://www.neurogami.com/code/filter.rb.txt :) Needs actual unit tests to see that it does something good. -- James Britt "I never dispute another person's delusions, just their facts." - Len Bullard From james.britt at gmail.com Sun Dec 24 14:02:15 2006 From: james.britt at gmail.com (James Britt) Date: Sun, 24 Dec 2006 12:02:15 -0700 Subject: [Nitro] Request for help In-Reply-To: <1166978085.809422.22010@73g2000cwn.googlegroups.com> References: <1166978085.809422.22010@73g2000cwn.googlegroups.com> Message-ID: <458ECEB7.3000009@gmail.com> transfire at gmail.com wrote: > George Moschovitis wrote: >> Btw, >> >> I am referring to this piece of code: >> >> http://code.iamcal.com/php/lib_filter/ >> >> Also if someone knows of a similar library available for Ruby let me know. > > What does it do? Is it like http://rubyforge.org/projects/tidy ? > > It's not long, I could probably translate it for you in short order > (assuming there's nothing out of the ordinary in there). I posted my version, but I forgot to mention that I did not re-implement certain PHP methods. The Ruby version passes though the interpreter, but there are no practical unit tests. But it's broken because, for example, there are numerous calls to preg_replace (or whatever it's called) that will bomb. But it's a start. -- James Britt "I never dispute another person's delusions, just their facts." - Len Bullard From george.moschovitis at gmail.com Sun Dec 24 14:07:39 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sun, 24 Dec 2006 21:07:39 +0200 Subject: [Nitro] Ratchets In-Reply-To: <1166977483.673491.75250@48g2000cwx.googlegroups.com> References: <1166977483.673491.75250@48g2000cwx.googlegroups.com> Message-ID: This looks very intuitive to use but I would like to see some source code or try this with a real world scenario to give more constructive comments. -g. On 12/24/06, transfire at gmail.com wrote: > As many of you are aware I've been working on a new build tool to > replace Reap, called Ratchets. I have had nearly all the core > functionality in place for a while now, but I have been struggling with > the "engine" that drives the system. I've set my bar fairly high, as I > wanted to provide something more capable, and even more elegant than > rake if possible (NOTE of course you can easily use Rathcets' via Rake > if you prefer). It took quite a bit of deep consideration (and > frustration), but I think that I have finally arrived at the nice > solution. I would like to share it with you to get your take and to > make sure I haven't overlooked anything. ... and I figure some of you > could use a break from the Holiday pageantry ;-) > > Let's say you have a project: > > myproject/ > README > ProjectInfo > lib/ > bin/ > script/ > > As with Reap, Ratchets uses a ProjectInfo file, but rather than define > tasks in this file, they are defined as stand-alone scripts in a > separate folder -- I haven't yet settled on a conventional name for > this folder, one can set the name in the ProjectInfo file, but for now > I'm calling it script/. > > Ratchets build scripts are ruby scripts, of course. But what's elegant > about them is how little they differ from any ruby script --there is > very little special notation. Here's a stupid example: > > # script/tryme.rb > > def tryme > puts "It worked!" > end > > Then on the command line: > > $ project tryme > It worked! > > I know. This seems too simple. How will we deal with prerequisites? How > do we call other targets if each one is in a separate file? Well, > that's were the elegance come into play. Every _public_ method is dealt > with as a target (eg. Ratchets' term for task). So public methods are > never triggered more than once per run session. Have a look at this: > > # script/tryme.rb > > def tryme > sayit > sayit > sayit > end > > def sayit > puts "It worked again!" > end > > Now: > > $ project tryme > It worked again! > > Notice "It worked again!" only printed once. But if 'sayit' were made a > private method instead, then it would print three times. > > To access targets in the other files you can simply require them as > normal. Say we have a target script called more.rb: > > # script/more.rb > > def more > puts "more..." > end > > Then in tryme.rb: > > require 'more' > > def tryme > sayit > more > end > > At this point, you may be wondering why bother with the separate > scripts. There are two reasons. Firstly they act as namespaces. Instead > of 'require' we can use 'utilize'. Let say more.rb has three target > methods more1, more2 and more3. Then: > > utilize 'more' > > def tryme > more.more1 > more.more2 > more.more3 > end > > The effect of this is that we can simply drop build scripts into our > script directly, and not worry about any name clashes between target > method names by using #utilize, instead or #require. Everything is kept > in well separated modules. > > Now Ratchets has a bunch of built-in build scripts already, and these > scripts do not differ in construction from personal build scripts > --another elegance which make it very easy to donate build scripts to > Rathcets. > > I could go one and layout a few other details and features, like > targets ability to take parameters and access to projectinfo data, but > I've covered the core of the build system, and this is getting long, so > I'll leave it there. > > Happy Holidays, > T. > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Sun Dec 24 14:09:40 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sun, 24 Dec 2006 21:09:40 +0200 Subject: [Nitro] Request for help In-Reply-To: <1166980695.673202.261850@a3g2000cwd.googlegroups.com> References: <1166978085.809422.22010@73g2000cwn.googlegroups.com> <1166980695.673202.261850@a3g2000cwd.googlegroups.com> Message-ID: > Okay. Let me know if you find anything else/better. But if not, I've > already translated about half of it. I just need to look up a couple > things to finish it. I haven;t found something better. And I am pretty sure this php implementation is great. So if you finish the porting let me know ;-) ;-) thanks, George. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Sun Dec 24 14:11:42 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sun, 24 Dec 2006 21:11:42 +0200 Subject: [Nitro] Request for help In-Reply-To: <458ECEB7.3000009@gmail.com> References: <1166978085.809422.22010@73g2000cwn.googlegroups.com> <458ECEB7.3000009@gmail.com> Message-ID: > I posted my version where ? -g. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From james.britt at gmail.com Sun Dec 24 14:50:44 2006 From: james.britt at gmail.com (James Britt) Date: Sun, 24 Dec 2006 12:50:44 -0700 Subject: [Nitro] Request for help In-Reply-To: References: <1166978085.809422.22010@73g2000cwn.googlegroups.com> <458ECEB7.3000009@gmail.com> Message-ID: <458EDA14.1080303@gmail.com> George Moschovitis wrote: >> I posted my version > > where ? > In a separate post, in response to Trans saying he had gotten halfway through. http://www.neurogami.com/code/filter.rb.txt -- James Britt http://beginningruby.com - Beginning Ruby: The Online Book http://www.artima.com/rubycs/ - The Journal By & For Rubyists http://www.rubystuff.com - The Ruby Store for Ruby Stuff From transfire at gmail.com Sun Dec 24 15:46:05 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Sun, 24 Dec 2006 20:46:05 -0000 Subject: [Nitro] Request for help In-Reply-To: <458ECEB7.3000009@gmail.com> References: <1166978085.809422.22010@73g2000cwn.googlegroups.com> <458ECEB7.3000009@gmail.com> Message-ID: <1166993165.085895.132210@42g2000cwt.googlegroups.com> James Britt wrote: > I posted my version, but I forgot to mention that I did not re-implement > certain PHP methods. The Ruby version passes though the interpreter, > but there are no practical unit tests. But it's broken because, for > example, there are numerous calls to preg_replace (or whatever it's > called) that will bomb. Where did you post it? I have pretty much finished it except for two things 1) HtmlSpecialChars needs implementation (see http://us3.php.net/manual/en/function.htmlspecialchars.php) I seem to recall Ruby has something for that but I can't seem to find it. And 2) It doesn't work ;-) Something's not quite right. I'm not sure if my #scan calls in #process_tag are right or what, and I'm also not sure what #strip_single is doing. But it's pretty close. I suppose the next thing is to write some test for each function to figure out what's a miss. Want to tag team this? I can send you my code and you can work on it some? T. From george.moschovitis at gmail.com Sun Dec 24 16:23:47 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sun, 24 Dec 2006 23:23:47 +0200 Subject: [Nitro] Request for help In-Reply-To: <458ECDF9.90800@gmail.com> References: <1166978085.809422.22010@73g2000cwn.googlegroups.com> <1166980695.673202.261850@a3g2000cwd.googlegroups.com> <458ECDF9.90800@gmail.com> Message-ID: > Needs actual unit tests to see that it does something good. It also needs some cleanup to make it look more like nitro and/or facets/ruby code ;-) I wonder if Tom could post his version as well. thanks, -g. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From transfire at gmail.com Sun Dec 24 17:59:01 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Sun, 24 Dec 2006 22:59:01 -0000 Subject: [Nitro] Request for help In-Reply-To: References: <1166978085.809422.22010@73g2000cwn.googlegroups.com> <1166980695.673202.261850@a3g2000cwd.googlegroups.com> <458ECDF9.90800@gmail.com> Message-ID: <1167001141.682836.180060@42g2000cwt.googlegroups.com> George Moschovitis wrote: > > Needs actual unit tests to see that it does something good. > > It also needs some cleanup to make it look more like nitro and/or > facets/ruby code ;-) > I wonder if Tom could post his version as well. Okay, I basically finsihed it and emailed you and Britt a copy. Do you have any preferences on the interface? T. From james.britt at gmail.com Sun Dec 24 18:07:25 2006 From: james.britt at gmail.com (James Britt) Date: Sun, 24 Dec 2006 16:07:25 -0700 Subject: [Nitro] Request for help In-Reply-To: References: <1166978085.809422.22010@73g2000cwn.googlegroups.com> <1166980695.673202.261850@a3g2000cwd.googlegroups.com> <458ECDF9.90800@gmail.com> Message-ID: <458F082D.2040804@gmail.com> George Moschovitis wrote: >> Needs actual unit tests to see that it does something good. > > It also needs some cleanup to make it look more like nitro and/or > facets/ruby code ;-) I was first aiming to make it valid Ruby code. Then I got distracted with Christmas stuff. :) -- James Britt "I was born not knowing and have had only a little time to change that here and there." - Richard P. Feynman From james.britt at gmail.com Mon Dec 25 00:24:17 2006 From: james.britt at gmail.com (James Britt) Date: Sun, 24 Dec 2006 22:24:17 -0700 Subject: [Nitro] Request for help In-Reply-To: <1167001141.682836.180060@42g2000cwt.googlegroups.com> References: <1166978085.809422.22010@73g2000cwn.googlegroups.com> <1166980695.673202.261850@a3g2000cwd.googlegroups.com> <458ECDF9.90800@gmail.com> <1167001141.682836.180060@42g2000cwt.googlegroups.com> Message-ID: <458F6081.6020607@gmail.com> transfire at gmail.com wrote: > George Moschovitis wrote: >>> Needs actual unit tests to see that it does something good. >> It also needs some cleanup to make it look more like nitro and/or >> facets/ruby code ;-) >> I wonder if Tom could post his version as well. > > Okay, I basically finsihed it and emailed you and Britt a copy. I've not seen this arrive, though. -- James Britt "Trying to port the desktop metaphor to the Web is like working on how to fuel your car with hay because that is what horses eat." - Dare Obasanjo From george.moschovitis at gmail.com Mon Dec 25 02:48:41 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Mon, 25 Dec 2006 09:48:41 +0200 Subject: [Nitro] Merry Xmas Message-ID: Dear devs, I wish you merry xmas! regards, George. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From transfire at gmail.com Tue Dec 26 10:11:05 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Tue, 26 Dec 2006 15:11:05 -0000 Subject: [Nitro] Ratchets In-Reply-To: References: <1166977483.673491.75250@48g2000cwx.googlegroups.com> Message-ID: <1167145865.157685.214890@48g2000cwx.googlegroups.com> George Moschovitis wrote: > This looks very intuitive to use but I would like to see some source > code or try this with a real world scenario to give more constructive > comments. Actually I could use some help here. I have two implementations of the core functionality I describe, but I'm not sure which is better. The first is less concise but more traditional, while the second is cleaner but uses the fancy #prepend method which is defined: class Module # Prepend an +aspect+ module to a module. # # module X # def x; "x"; end # end # # module U # def x; '{' + super + '}'; end # end # # X.prepend U # # X.x # => "{x}" def prepend( aspect ) aspect.send(:include, self) extend aspect end end The only reason I hesitate to use the second defintion is b/c I have yet to fully wrap my head around prepend. Also note these use a slightly modified version of script.rb by Joel VanderWerf (see http://redshift.sourceforge.net/script/doc/index.html). # Target Script # # NOTE The access methods created by utilize are being # cached. They don't need to be, and if possible should # not be. But I think it doesn't hurt anything unless one # tries to reuse the same name with a different utilize # call, which is farily unlikely. class TargetScript < Module # integrate $NOHARM into session ? include($NOHARM ? FileUtils::DryRun : FileUtils) def initialize(file, session) session = session project = session.project stack = session.stack cache = session.cache # # kernel # kernel = Module.new do define_method(:project) do project end define_method(:session) do session end define_method(:utilize) do |*args| opts = (Hash === args.last ? args.pop : {}) args.each{ |a| opts[a] = nil } opts.each do |k,v| target = session.targets.lookup(k).first m = target.to_module( session ) define_method(v||target.name){m} end end end # # script # script = Script.new(file) do include kernel end names = script.public_instance_methods - ['project', 'utilize', 'session'] include script # # cache # cache_module = Module.new do names.each do |name| send(:define_method, name) do |*args| key = [file, name, *args] unless cache.key?(key) stack << file cache[key] = nil cache[key] = super(*args) stack.pop end return cache[key] end end end include cache_module # # self # extend self end end Prepend version: # Target Script # # NOTE This seems to work well. The use of +prepend+ is # new and hence a bit expiremental however. One issue is # that the access methods created by utilize are cached. # They don't need to be, and if possible should not be. # But I think it doesn't hurt anything unless one tries to # reuse the same name with a different utilize call, which # is farily unlikely. class TargetScript < Script # integrate $NOHARM into session ? include($NOHARM ? FileUtils::DryRun : FileUtils) attr_reader :session, :project def initialize(file, session) @session = session @project = session.project stack = session.stack cache = session.cache super(file) names = public_instance_methods cache_module = Module.new do names.each do |name| send(:define_method, name) do |*args| key = [file, name, *args] unless cache.key?(key) stack << file cache[key] = nil # stymie infinite recursion cache[key] = super(*args) stack.pop end return cache[key] end end end prepend cache_module end # # utilize # def utilize(*args) opts = (Hash === args.last ? args.pop : {}) args.each{ |a| opts[a] = nil } opts.each do |k,v| target = session.targets.lookup(k).first mod = target.to_module( session ) define_method(v||target.name){mod} end end end T. From coldredlemur at gmail.com Tue Dec 26 10:49:59 2006 From: coldredlemur at gmail.com (Ry Dahl) Date: Tue, 26 Dec 2006 12:49:59 -0300 Subject: [Nitro] Og in Rails Message-ID: <21ee31950612260749m5d41bb8cy1d62b4e6dcf1d86c@mail.gmail.com> Is it possible to use Og as a replacement for AR in Rails? If not, what would have to be done to make that happen? Ryan From james.britt at gmail.com Tue Dec 26 11:15:38 2006 From: james.britt at gmail.com (James Britt) Date: Tue, 26 Dec 2006 09:15:38 -0700 Subject: [Nitro] Og in Rails In-Reply-To: <21ee31950612260749m5d41bb8cy1d62b4e6dcf1d86c@mail.gmail.com> References: <21ee31950612260749m5d41bb8cy1d62b4e6dcf1d86c@mail.gmail.com> Message-ID: <45914AAA.3020905@gmail.com> Ry Dahl wrote: > Is it possible to use Og as a replacement for AR in Rails? If not, > what would have to be done to make that happen? You would have to figure out what API calls Rails expected of AR, but are not implemented in Og. I.e., most. Then add message handlers to Og to provide expected behavior. Given how much code is generated at runtime, this may be a non-trivial task. There are, however, examples on the Web of using AR without a database. I did some work where I needed an XML DOM to stand in for AR; the various calls for reading or assigning model values were mapped to XPath calls and such. Conceivably, you could do this, with the difference that the "no database" model object would delegate to an Og enchanted object. So you have a Rails model that does not extend AR (so Rails thinks there's no DB involved), while the model goes about its persistence behavior using Og. (This all presumes that calling 'require "og"' doesn't load code that conflicts with Rails.) Anyway, some pre-coffee ruminations. -- James Britt "The use of anthropomorphic terminology when dealing with computing systems is a symptom of professional immaturity." - Edsger W. Dijkstra From george.moschovitis at gmail.com Tue Dec 26 15:31:25 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Tue, 26 Dec 2006 22:31:25 +0200 Subject: [Nitro] Ratchets In-Reply-To: <1167145865.157685.214890@48g2000cwx.googlegroups.com> References: <1166977483.673491.75250@48g2000cwx.googlegroups.com> <1167145865.157685.214890@48g2000cwx.googlegroups.com> Message-ID: Wow, prepend and Script (the original version) are very interesting ideas ;-) -g. On 12/26/06, transfire at gmail.com wrote: > > George Moschovitis wrote: > > This looks very intuitive to use but I would like to see some source > > code or try this with a real world scenario to give more constructive > > comments. > > Actually I could use some help here. I have two implementations of the > core functionality I describe, but I'm not sure which is better. The > first is less concise but more traditional, while the second is cleaner > but uses the fancy #prepend method which is defined: > > class Module > > # Prepend an +aspect+ module to a module. > # > # module X > # def x; "x"; end > # end > # > # module U > # def x; '{' + super + '}'; end > # end > # > # X.prepend U > # > # X.x # => "{x}" > > def prepend( aspect ) > aspect.send(:include, self) > extend aspect > end > > end > > The only reason I hesitate to use the second defintion is b/c I have > yet to fully wrap my head around prepend. > > Also note these use a slightly modified version of script.rb by Joel > VanderWerf (see > http://redshift.sourceforge.net/script/doc/index.html). > > # Target Script > # > # NOTE The access methods created by utilize are being > # cached. They don't need to be, and if possible should > # not be. But I think it doesn't hurt anything unless one > # tries to reuse the same name with a different utilize > # call, which is farily unlikely. > > class TargetScript < Module > > # integrate $NOHARM into session ? > include($NOHARM ? FileUtils::DryRun : FileUtils) > > def initialize(file, session) > session = session > project = session.project > > stack = session.stack > cache = session.cache > > # > # kernel > # > kernel = Module.new do > define_method(:project) do project end > define_method(:session) do session end > define_method(:utilize) do |*args| > opts = (Hash === args.last ? args.pop : {}) > args.each{ |a| opts[a] = nil } > opts.each do |k,v| > target = session.targets.lookup(k).first > m = target.to_module( session ) > define_method(v||target.name){m} > end > end > end > > # > # script > # > script = Script.new(file) do > include kernel > end > > names = script.public_instance_methods - ['project', 'utilize', > 'session'] > > include script > > # > # cache > # > cache_module = Module.new do > names.each do |name| > send(:define_method, name) do |*args| > key = [file, name, *args] > unless cache.key?(key) > stack << file > cache[key] = nil > cache[key] = super(*args) > stack.pop > end > return cache[key] > end > end > end > > include cache_module > > # > # self > # > extend self > end > > end > > Prepend version: > > # Target Script > # > # NOTE This seems to work well. The use of +prepend+ is > # new and hence a bit expiremental however. One issue is > # that the access methods created by utilize are cached. > # They don't need to be, and if possible should not be. > # But I think it doesn't hurt anything unless one tries to > # reuse the same name with a different utilize call, which > # is farily unlikely. > > class TargetScript < Script > > # integrate $NOHARM into session ? > include($NOHARM ? FileUtils::DryRun : FileUtils) > > attr_reader :session, :project > > def initialize(file, session) > @session = session > @project = session.project > > stack = session.stack > cache = session.cache > > super(file) > > names = public_instance_methods > > cache_module = Module.new do > names.each do |name| > send(:define_method, name) do |*args| > key = [file, name, *args] > unless cache.key?(key) > stack << file > cache[key] = nil # stymie infinite recursion > cache[key] = super(*args) > stack.pop > end > return cache[key] > end > end > end > > prepend cache_module > end > > # > # utilize > # > > def utilize(*args) > opts = (Hash === args.last ? args.pop : {}) > args.each{ |a| opts[a] = nil } > opts.each do |k,v| > target = session.targets.lookup(k).first > mod = target.to_module( session ) > define_method(v||target.name){mod} > end > end > > end > > > T. > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From vikingtux at gmail.com Wed Dec 27 06:40:35 2006 From: vikingtux at gmail.com (Alexandre Gravem) Date: Wed, 27 Dec 2006 09:40:35 -0200 Subject: [Nitro] Nitro Real World example In-Reply-To: References: Message-ID: <40b05ebe0612270340q16a93560w1f8f70993f3a359a@mail.gmail.com> Very useful for me indeed. :P On 12/21/06, George Moschovitis wrote: > > Dear devs, > > as promised yesterday on #nitro, I am releasing a real world example > of using Nitro. You can find the code in my latest blog entry: > > http://blog.gmosx.com/2006/12/21/real-world-nitro-example > > I hope this will be useful for you. > > regards, > George. > > -- > http://blog.gmosx.com > http://cull.gr > http://nitroproject.org > http://www.joy.gr > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061227/116e8062/attachment.html From john at oxyliquit.de Wed Dec 27 07:11:37 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 27 Dec 2006 13:11:37 +0100 Subject: [Nitro] Merry Xmas In-Reply-To: References: Message-ID: > I wish you merry xmas! > George. Me too! Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From john at oxyliquit.de Wed Dec 27 07:11:40 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 27 Dec 2006 13:11:40 +0100 Subject: [Nitro] Nitro/Og RDOC Message-ID: Hi, George, could you update the rdoc for Nitro/Og/Glue on your page? http://static.nitroproject.org/rdoc/nitro/index.html Thanks! Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From john at oxyliquit.de Wed Dec 27 07:11:40 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 27 Dec 2006 13:11:40 +0100 Subject: [Nitro] Og in Rails In-Reply-To: <45914AAA.3020905@gmail.com> References: <21ee31950612260749m5d41bb8cy1d62b4e6dcf1d86c@mail.gmail.com> <45914AAA.3020905@gmail.com> Message-ID: Hi, >> Is it possible to use Og as a replacement for AR in Rails? If not, >> what would have to be done to make that happen? > (This all presumes that calling 'require "og"' doesn't load code that > conflicts with Rails.) Keep in mind, that Og uses Facets, and quite extensively so. I'm not sure if it clashes with ActiveSupport or whatever the counterpart for small general support functionality is in the Rails world. I think this attempt was made before, I'm not sure but I think it failed. But don't let that scare you, please try and report regardless of the outcome. :) Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From transfire at gmail.com Wed Dec 27 08:32:58 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Wed, 27 Dec 2006 13:32:58 -0000 Subject: [Nitro] Og in Rails In-Reply-To: References: <21ee31950612260749m5d41bb8cy1d62b4e6dcf1d86c@mail.gmail.com> <45914AAA.3020905@gmail.com> Message-ID: <1167226378.930970.176540@a3g2000cwd.googlegroups.com> Jonathan Buch wrote: > Hi, > > >> Is it possible to use Og as a replacement for AR in Rails? If not, > >> what would have to be done to make that happen? > > > (This all presumes that calling 'require "og"' doesn't load code that > > conflicts with Rails.) > > Keep in mind, that Og uses Facets, and quite extensively so. I'm not > sure if it clashes with ActiveSupport or whatever the counterpart for > small general support functionality is in the Rails world. That's an interesting point. Thre is a good bit of overlap between ActiveSupport and Facets --Facets largely being a superset of ActiveSupport, but not everything is included, nor always 100% compatiable. The first big question however is how/where does ActiveSupport include it's extensions. ActiveSupport uses modules. In the case of most classes, like Array, String, Hash, etc. any clash will fall to Facets b/c Facets injects method directly instead. (I've long considered using modules too, but I haven't seen a pressing reason to do so). > I think this attempt was made before, I'm not sure but I think it > failed. But don't let that scare you, please try and report regardless > of the outcome. :) I would be interesting if we could automatically show where ActiveSupport and Facets overlap. T. > Jo > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general From billk at cts.com Wed Dec 27 09:21:26 2006 From: billk at cts.com (Bill Kelly) Date: Wed, 27 Dec 2006 06:21:26 -0800 Subject: [Nitro] og-0.41.0: postgresql needs table creation WITHOUT OIDS? Message-ID: <062401c729c2$4b91f780$6442a8c0@musicbox> Hi, I just upgraded from og-0.30.0 to og-0.41.0, and it looks like postgres support may be broken: Previously, og-0.30.0/lib/og/store/psql.rb would specify "WITHOUT OIDS" when creating tables. This is apparently important, because unless "WITHOUT OIDS" is specified, postgres creates its own system column called 'oid': http://www.postgresql.org/docs/7.4/static/ddl-system-columns.html It seems the "WITHOUT OIDS" is missing in og's create_table() now, which appears to have been factored out of og-0.41.0/lib/og/adapter/postgresql.rb and now resides in og-0.41.0/lib/og/store/sql.rb I can kludge store/sql.rb for now, but I'm not sure how to fix it properly. :-( Regards, Bill From john at oxyliquit.de Wed Dec 27 09:46:59 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 27 Dec 2006 15:46:59 +0100 Subject: [Nitro] og-0.41.0: postgresql needs table creation WITHOUT OIDS? In-Reply-To: <062401c729c2$4b91f780$6442a8c0@musicbox> References: <062401c729c2$4b91f780$6442a8c0@musicbox> Message-ID: Hi, > I just upgraded from og-0.30.0 to og-0.41.0, and it looks like > postgres support may be broken: > > Previously, og-0.30.0/lib/og/store/psql.rb would specify > "WITHOUT OIDS" when creating tables. > > This is apparently important, because unless "WITHOUT OIDS" is > specified, postgres creates its own system column called 'oid': > http://www.postgresql.org/docs/7.4/static/ddl-system-columns.html > > It seems the "WITHOUT OIDS" is missing in og's create_table() > now, which appears to have been factored out of > og-0.41.0/lib/og/adapter/postgresql.rb > and now resides in > og-0.41.0/lib/og/store/sql.rb > > > I can kludge store/sql.rb for now, but I'm not sure how to fix > it properly. :-( thank you for reporting this, I will see to it that this gets fixed for the next release. Which postgresql version are you using? 7.4 as the docs show? I'm using 8.1, I think the OID stuff was changed on 8.0 so I didn't catch this. Again, thanks for reporting! Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From john at oxyliquit.de Wed Dec 27 12:12:43 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Wed, 27 Dec 2006 18:12:43 +0100 Subject: [Nitro] [PATCH] some fixes + enhancements Message-ID: Hi, This also fixes the bug reported by Bill Kelly. There are a few more fixes for Oracle as well, so anyone who wants further Oracle support, please test these patches and report back. Fri Dec 8 19:17:06 CET 2006 Jonathan Buch * sql.rb split create_table, oracle fixes Fri Dec 8 20:48:11 CET 2006 Jonathan Buch * oracle fixes, resolve_limit_options works Sat Dec 9 13:33:35 CET 2006 Jonathan Buch * split method_missing in entity.rb move functionality to 2 extra methods, find_by_() and find_or_create_by_() Wed Dec 13 11:58:40 CET 2006 Jonathan Buch * Fix for sql.rb create_field map to make it even more general also some minor enhancements in tcs Wed Dec 13 12:02:15 CET 2006 Jonathan Buch * sqlite enhancements Wed Dec 13 12:03:12 CET 2006 Jonathan Buch * fix nasty sti bug it wanted the 'ogtype' field always as the first field. Fix to not rely on that. Wed Dec 13 12:07:10 CET 2006 Jonathan Buch * set force_boolean for populate_object (.assign) to true as default Wed Dec 27 15:56:33 CET 2006 Jonathan Buch * WITHOUT OIDS for psql, psql < 8.0 add oid column without that Wed Dec 27 17:11:04 CET 2006 Jonathan Buch * some sti fixes, minor other stuff -- Feel the love http://pinkjuice.com/pics/ruby.png -------------- next part -------------- A non-text attachment was scrubbed... Name: pbndl.patch.tar.bz2 Type: application/bzip2 Size: 7135 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061227/674d7610/attachment-0001.bin From george.moschovitis at gmail.com Wed Dec 27 12:15:43 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Wed, 27 Dec 2006 19:15:43 +0200 Subject: [Nitro] Nitro Real World example In-Reply-To: <40b05ebe0612270340q16a93560w1f8f70993f3a359a@mail.gmail.com> References: <40b05ebe0612270340q16a93560w1f8f70993f3a359a@mail.gmail.com> Message-ID: nice :) -g. On 12/27/06, Alexandre Gravem wrote: > Very useful for me indeed. :P > > > On 12/21/06, George Moschovitis wrote: > > > > Dear devs, > > > > as promised yesterday on #nitro, I am releasing a real world example > > of using Nitro. You can find the code in my latest blog entry: > > > > http://blog.gmosx.com/2006/12/21/real-world-nitro-example > > > > I hope this will be useful for you. > > > > regards, > > George. > > > > -- > > http://blog.gmosx.com > > http://cull.gr > > http://nitroproject.org > > http://www.joy.gr > > _______________________________________________ > > Nitro-general mailing list > > Nitro-general at rubyforge.org > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Wed Dec 27 12:16:08 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Wed, 27 Dec 2006 19:16:08 +0200 Subject: [Nitro] Nitro/Og RDOC In-Reply-To: References: Message-ID: Ok ;-) On 12/27/06, Jonathan Buch wrote: > Hi, > > George, could you update the rdoc for Nitro/Og/Glue on your > page? http://static.nitroproject.org/rdoc/nitro/index.html > > Thanks! > > Jo > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Wed Dec 27 12:17:42 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Wed, 27 Dec 2006 19:17:42 +0200 Subject: [Nitro] [PATCH] some fixes + enhancements In-Reply-To: References: Message-ID: Many thanks for this patch :) -g. On 12/27/06, Jonathan Buch wrote: > Hi, > > This also fixes the bug reported by Bill Kelly. > > There are a few more fixes for Oracle as well, so anyone who wants > further Oracle support, please test these patches and report back. > > > Fri Dec 8 19:17:06 CET 2006 Jonathan Buch > * sql.rb split create_table, oracle fixes > > Fri Dec 8 20:48:11 CET 2006 Jonathan Buch > * oracle fixes, resolve_limit_options works > > Sat Dec 9 13:33:35 CET 2006 Jonathan Buch > * split method_missing in entity.rb > move functionality to 2 extra methods, find_by_() and find_or_create_by_() > > Wed Dec 13 11:58:40 CET 2006 Jonathan Buch > * Fix for sql.rb create_field map to make it even more general > also some minor enhancements in tcs > > Wed Dec 13 12:02:15 CET 2006 Jonathan Buch > * sqlite enhancements > > Wed Dec 13 12:03:12 CET 2006 Jonathan Buch > * fix nasty sti bug > it wanted the 'ogtype' field always as the first field. Fix to not rely on > that. > > Wed Dec 13 12:07:10 CET 2006 Jonathan Buch > * set force_boolean for populate_object (.assign) to true as default > > Wed Dec 27 15:56:33 CET 2006 Jonathan Buch > * WITHOUT OIDS for psql, psql < 8.0 add oid column without that > > Wed Dec 27 17:11:04 CET 2006 Jonathan Buch > * some sti fixes, minor other stuff > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From billk at cts.com Wed Dec 27 14:02:24 2006 From: billk at cts.com (Bill Kelly) Date: Wed, 27 Dec 2006 11:02:24 -0800 Subject: [Nitro] og-0.41.0: postgresql needs table creation WITHOUT OIDS? References: <062401c729c2$4b91f780$6442a8c0@musicbox> Message-ID: <069c01c729e9$8b890500$6442a8c0@musicbox> From: "Jonathan Buch" > > Which postgresql version are you using? 7.4 as the docs show? > I'm using 8.1, I think the OID stuff was changed on 8.0 so I didn't > catch this. Yeah, 7.4 seems to be still the cutting edge according to debian stable... ;) > Again, thanks for reporting! Thank you all for this awesome software :) Regards, Bill From george.moschovitis at gmail.com Wed Dec 27 17:29:47 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 28 Dec 2006 00:29:47 +0200 Subject: [Nitro] Consistency Message-ID: Dear devs, I think, we should make Nitro/Og more consistent with names. Let me present a list with alternatives for the same concept and the synonym I propose to keep: template, view -> template (already converted) adapter, store -> adapter (already converted) model, entity, managed_class -> entity scaffold, enchant -> enchant If you can find more synonyms, please let me know. If you think that we should keep a different synonym, let me know as well. Please note that I would like to keep the same synonyms throughout Nitro and Og (for example not scaffold in Nitro, enchant in Og) Using consistent names will make Nitro more accessible to new comers. regards, George. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From john at oxyliquit.de Wed Dec 27 19:01:11 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Thu, 28 Dec 2006 01:01:11 +0100 Subject: [Nitro] Consistency In-Reply-To: References: Message-ID: Hi, > I think, we should make Nitro/Og more consistent with names. Let me > present a list with alternatives for the same concept and the synonym > I propose to keep: > > template, view -> template (already converted) Sidenote: This brings me to a thought I had since I was last in a course in university here: They tought us the MVC concept and it was a little different from what we do with Nitro. The *View* in that concept is the calculating and presenting data to the outside. This is what our templating mechanism and (surprise) _the Nitro::Controller_ does. The *Model* is the complete application logic together with the data. This is what currently the Og models are, and also what we currently put often in the controller. The *Controller* finally is the part accepting requests from the outside. This picture was shown: +------------------------+ <- | View | | +------------| Model | -> | Controller | | +------------------------+ I'm coming to that, since I just read the documentation for the Python framework Django. In http://www.djangoproject.com/documentation/faq/ it explains quite nicely on the question: "Django appears to be a MVC framework, but you call the Controller the "view", and the View the "template". How come you don't use the standard names?". Note, that I would never ever propose to rename anything in Nitro based on this information. As the Django doc make clear, it's just how you get the job done. :P (This was just a little 'excurse' to look a little beyond one's own nose.) > adapter, store -> adapter (already converted) I'm not sure that they are synonyms here... you explained quite nicely a long time ago why they are different. The distinction here is quite important (unimportant for someone just using Og) for the developer to see where he is in the 'general layout'. > model, entity, managed_class -> entity This makes sense, yes. Although 'entity' is less descriptive than managed_class or model. Less descriptive doesn't mean bad, but a little 'redundancy' in the 'language' helps to understand the general intent. Og::Manager.entities Og::Manager.managed_classes Here entities sounds a little ambiguous. Since 'entity' only means 'thing/unit' in the english language, it could also mean 'all manager instances'. > scaffold, enchant -> enchant I had a look inside the nitro scaffolding thing, I didn't even know about it, and it does not seem like I would ever use it... Yes, the scaffold -> enchant is probably the right thing to do. > If you can find more synonyms, please let me know. If you think that > we should keep a different synonym, let me know as well. Please note > that I would like to keep the same synonyms throughout Nitro and Og > (for example not scaffold in Nitro, enchant in Og) > > Using consistent names will make Nitro more accessible to new comers. Make that 'newcomers browing the source'. :P Sorry for the post with probably more question than answers, I felt a little like making a post about general 'philosophy'. :) I feel a little like William now (in a good sense). :P Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From james.britt at gmail.com Wed Dec 27 20:14:26 2006 From: james.britt at gmail.com (James Britt) Date: Wed, 27 Dec 2006 18:14:26 -0700 Subject: [Nitro] Consistency In-Reply-To: References: Message-ID: <45931A72.5060008@gmail.com> Jonathan Buch wrote: > > Sidenote: > > This brings me to a thought I had since I was last in a course in > university here: They tought us the MVC concept and it was a little > different from what we do with Nitro. MVC came about in Smalltalk, and since then assorted variations have evolved. http://st-www.cs.uiuc.edu/users/smarch/st-docs/mvc.html http://heim.ifi.uio.no/~trygver/themes/mvc/mvc-index.html There seems to be a much tighter coupling among view and controller than is appropriate for Web apps. Some interesting thoughts here: http://wardley.org/computers/web/mvc.html (Essay question: How does MVC reconcile with REST? ) Also, some Django-heads I know have told me that Django is MVP: Model, view, presenter. http://www.martinfowler.com/eaaDev/ModelViewPresenter.html http://www.martinfowler.com/eaaDev/uiArchs.html -- James Britt From john at oxyliquit.de Wed Dec 27 20:45:24 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Thu, 28 Dec 2006 02:45:24 +0100 Subject: [Nitro] Consistency In-Reply-To: <45931A72.5060008@gmail.com> References: <45931A72.5060008@gmail.com> Message-ID: Hi, > MVC came about in Smalltalk, and since then assorted variations have > evolved. > [many interesting things snipped here] thanks for those links! Food for thought. :D Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From transfire at gmail.com Wed Dec 27 22:53:54 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Thu, 28 Dec 2006 03:53:54 -0000 Subject: [Nitro] Consistency In-Reply-To: <45931A72.5060008@gmail.com> References: <45931A72.5060008@gmail.com> Message-ID: <1167278034.396327.276900@48g2000cwx.googlegroups.com> George wrote: > template, view -> template (already converted) > adapter, store -> adapter (already converted) > model, entity, managed_class -> entity > scaffold, enchant -> enchant Sigh. I hate to be the contrarian here, but I have to offer a couple of points. 1) It's better to utilize standard terminology when you can --even if it's not perfect 1:1 match. It still makes things easier to talk about and more importantly it's good for MARKETING. When you can say Nito/Og is "MVC" then that means something to people. Using those terms in the code then furthers that connection. 2) Some of the terms are simply too general. If I were to come across one out of the blue/on it's own, I would have no idea what it was. The terms 'entity' especially just doesn't have much semantic value, might I suggest "domain" (see MVC on wikipedia). I would also argue that "adapter" is much more ambiguous than "store". I think "enchant" is good though. Having said that I want to say a couple of things about MVC in light of what others are saying. I think the general notaion of MVC is a little stilted. I have seen numerous discussions about what it actually means, and I think that b/c of what James was saying, in that MVC came from GUI apps and was sort of roughly fitted to Webapps. With the way web apps work however the Controller is largely the web server platform itself, ie. managing the HTTP request (this is where REST comes in). And the View is the response. Whats left to Nitro then is just the model to view part of control. But what goes unmentioned in the system is the very back-end --the Data. When you add that to the picture, a broader picture of "MVC" comes to light where there are "controllers" all along teh chain: Like this: User <-> Webserver <-> View/REST <-> Model <-> Adapter <-> Database <-> Data We can think of the database as a DataController and so an adapter is really a "ModelController". The Model is the managed/enchanted class (DomainController), then the remaining is the usual Controller of MVC. The first is a "Rest/ViewController", that includes the template system in the View part, then the "WebController" ie. the Webserver. We could also add a UserController just before the user, which is the web browser, but well let that go since the line is big enough ;-) I know that seems a little confusing, but my point is simply that these are really different "controllers" all along the flow chain from data to user and back. T. From william.full.moon at gmail.com Wed Dec 27 23:42:51 2006 From: william.full.moon at gmail.com (* William) Date: Thu, 28 Dec 2006 15:42:51 +1100 Subject: [Nitro] Consistency In-Reply-To: Message-ID: <00a701c72a3a$aaacf580$0201a8c0@ghostgum> In support, YES! g>> If you can find more synonyms, please let me know. If you think g>> that we should keep a different synonym, let me know as well. g>> Please note that I would like to keep the same synonyms g>> throughout Nitro and Og (for example not scaffold in Nitro, g>> enchant in Og) g>> g>> Using consistent names will make Nitro more accessible to g>> new comers. My view on this is that a common terminology will only come about wish a high-level architectural description or better yet picture (similar to the ones supplied by Jo. But of course with more separation: jo>> This picture was shown: +---------+ +------------+ +----------+ <--> | View 1 | <--> | | <--> | Model | +---------| | Controller | +----------+ <--> | View 2 | <--> | | A +---------+ +------------+ | | V +------------+ | +-----------+ | interfaces | <--> O | store | +------------+ | interface | +-----------+ This picture might be more involved than is required. I wanted to show that applications especially the one we will need in the 21st century will not only rely on a GUI or is it now to become a "graphical web interface" (GWI)? Google, SAP, Microsoft, IBM, etc offer various API-s or WebServices to access interface (and reuse) the same application. I really recommend formulating a useful architectural block structure. And developing some nice documentation stuff that identifies where various bits belong in the "big picture" in the header comments. As to design and theory of m-v-c the wikipedia entry is absolutely fantastic. * http://en.wikipedia.org/wiki/MVC Lovely stuff, ... William MBA, B.Math, STI, NLP MP _My_Bright_Ideas__________________________________________abn:77425560110_ "You must BE the change you want." ~ Gandhi Success Coaching http://adroit-process.blogspot.com/ ph/sms: 041-476-1217 -----Original Message----- From: nitro-general-bounces at rubyforge.org [mailto:nitro-general-bounces at rubyforge.org] On Behalf Of George Moschovitis Sent: Thursday, 28 December 2006 09:30 To: General discussion about Nitro Subject: [Nitro] Consistency Importance: Low Dear devs, -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.432 / Virus Database: 268.15.28/604 - Release Date: 26-Dec-2006 12:23 From william.full.moon at gmail.com Wed Dec 27 23:49:21 2006 From: william.full.moon at gmail.com (* William) Date: Thu, 28 Dec 2006 15:49:21 +1100 Subject: [Nitro] (oops) Model-view-controller In-Reply-To: Message-ID: <00a801c72a3b$9220ab00$0201a8c0@ghostgum> I gave the disambiguation link in my earlier mail. It should have been: http://en.wikipedia.org/wiki/Model-view-controller I'm sure this is redundant, most of you will figure it out ;-) /w. -----Original Message----- From: nitro-general-bounces at rubyforge.org [mailto:nitro-general-bounces at rubyforge.org] On Behalf Of George Moschovitis Sent: Thursday, 28 December 2006 09:30 To: General discussion about Nitro Subject: [Nitro] Consistency Importance: Low Dear devs, -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.432 / Virus Database: 268.15.28/604 - Release Date: 26-Dec-2006 12:23 From james.britt at gmail.com Thu Dec 28 00:39:04 2006 From: james.britt at gmail.com (James Britt) Date: Wed, 27 Dec 2006 22:39:04 -0700 Subject: [Nitro] Consistency In-Reply-To: <1167278034.396327.276900@48g2000cwx.googlegroups.com> References: <45931A72.5060008@gmail.com> <1167278034.396327.276900@48g2000cwx.googlegroups.com> Message-ID: <45935878.8000600@gmail.com> transfire at gmail.com wrote: > > Sigh. I hate to be the contrarian here, but I have to offer a couple of > points. > > 1) It's better to utilize standard terminology when you can --even if > it's not perfect 1:1 match. It still makes things easier to talk about > and more importantly it's good for MARKETING. When you can say Nito/Og > is "MVC" then that means something to people. Using those terms in the > code then furthers that connection. > > 2) Some of the terms are simply too general. If I were to come across > one out of the blue/on it's own, I would have no idea what it was. The > terms 'entity' especially just doesn't have much semantic value, might > I suggest "domain" (see MVC on wikipedia). I would also argue that > "adapter" is much more ambiguous than "store". I think "enchant" is > good though. > These are good points. If a familiar term can be used then it should be considered. (A caveat is that the term should be more or less accurate. For example if something is called a 'view', but a full explanation ends up having to describe all the ways it differs from what people usually expect of a view, then that's just the wrong term to use.) > Having said that I want to say a couple of things about MVC in light of > what others are saying. I think the general notaion of MVC is a little > stilted. I have seen numerous discussions about what it actually means, > and I think that b/c of what James was saying, in that MVC came from > GUI apps and was sort of roughly fitted to Webapps. > > With the way web apps work however the Controller is largely the web > server platform itself, ie. managing the HTTP request (this is where > REST comes in). And the View is the response. Whats left to Nitro then > is just the model to view part of control. But what goes unmentioned > in the system is the very back-end --the Data. When you add that to the > picture, a broader picture of "MVC" comes to light where there are > "controllers" all along teh chain: Like this: > > User <-> Webserver <-> View/REST <-> Model <-> Adapter <-> Database > <-> Data > More food for thought along these lines: http://c2.com/cgi/wiki?HexagonalArchitecture http://alistair.cockburn.us/index.php/Hexagonal_architecture I like this very much; it's simpler. The symmetry is very appealing. Some time ago I was hacking on the Nitro compiler pipeline (some here may recall my seemingly endless questions on transformations) because I wanted to manipulate a page *after* Nitro had decided it was finished with it. I didn't find a way to just drop one more transformer into the mix because, as I recall, the very last bit of transformation was coupled to the server request/response code. I was thinking that Nitro was "just" a text or request transformation tool, and the fact that it would generate Web pages was (or could be ) incidental. Stuff goes in, stuff comes out; "out" just happens to be a Web server most of the time. But needn't always be; it could be another application. The stuff that comes out would usually be a view of some sort, but it could also just be data requested from some other process. (Thought: When executing a SQL select, you get back one or more rows; aren't these just 'views', representations of stuff suitable for manipulation by some client, with the DB engine being a controller, and whatever is really storing the bits being the models? ) So you have a chain of local MVC environments, where one section's model is another section's view: ... v] <--> [c] <--> [m|v] <--> [c] <--> [m ... Or something roughly like that. And it need not be 2-D (the c2.org page has one comment referring to a honeycomb). -- James Britt http://www.rubyaz.org - Hacking in the Desert http://www.jamesbritt.com - Playing with Better Toys From george.moschovitis at gmail.com Thu Dec 28 03:28:45 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 28 Dec 2006 10:28:45 +0200 Subject: [Nitro] Consistency In-Reply-To: References: Message-ID: > Note, that I would never ever propose to rename anything in Nitro based > on this information. As the Django doc make clear, it's just how you > get the job done. :P hmm, I don't feel compeled to follow the smalltalk definitions. > > (This was just a little 'excurse' to look a little beyond one's own nose.) > > > adapter, store -> adapter (already converted) > > I'm not sure that they are synonyms here... you explained quite nicely > a long time ago why they are different. The distinction here is quite > important (unimportant for someone just using Og) for the developer to > see where he is in the 'general layout'. yeah, they are not synonyms my last explanation still stands. I just splitted the original store of og to adapter, store to make more consistent with nitro. > > model, entity, managed_class -> entity > > This makes sense, yes. Although 'entity' is less descriptive than > managed_class or model. Less descriptive doesn't mean bad, but a > little 'redundancy' in the 'language' helps to understand the general > intent. entity comes from Entity-Relation a DB term. I think it is much more descriptive than model. > Og::Manager.entities > Og::Manager.managed_classes managed_classes is too long, entities is more succint > > scaffold, enchant -> enchant > > I had a look inside the nitro scaffolding thing, I didn't even know > about it, and it does not seem like I would ever use it... > > Yes, the scaffold -> enchant is probably the right thing to do. > > > If you can find more synonyms, please let me know. If you think that > > we should keep a different synonym, let me know as well. Please note > > that I would like to keep the same synonyms throughout Nitro and Og > > (for example not scaffold in Nitro, enchant in Og) > > > > Using consistent names will make Nitro more accessible to new comers. > > Make that 'newcomers browing the source'. :P > > Sorry for the post with probably more question than answers, I felt a > little like making a post about general 'philosophy'. :) > I feel a little like William now (in a good sense). :P > > Jo > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Thu Dec 28 03:33:04 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 28 Dec 2006 10:33:04 +0200 Subject: [Nitro] Consistency In-Reply-To: <45935878.8000600@gmail.com> References: <45931A72.5060008@gmail.com> <1167278034.396327.276900@48g2000cwx.googlegroups.com> <45935878.8000600@gmail.com> Message-ID: Interesting stuff, I will check the links... On 12/28/06, James Britt wrote: > transfire at gmail.com wrote: > > > > > Sigh. I hate to be the contrarian here, but I have to offer a couple of > > points. > > > ... > > -- > James Britt > > http://www.rubyaz.org - Hacking in the Desert > http://www.jamesbritt.com - Playing with Better Toys > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From william.full.moon at gmail.com Thu Dec 28 03:39:24 2006 From: william.full.moon at gmail.com (* William) Date: Thu, 28 Dec 2006 19:39:24 +1100 Subject: [Nitro] Consistency In-Reply-To: <45935878.8000600@gmail.com> Message-ID: <00bd01c72a5b$b25e1810$0201a8c0@ghostgum> Happy New 007 to come all. That last point James falls back into the UNIX idea of daisy chaining transformations or pipes. j>> So you have a chain of local MVC environments, where one j>> section's model is another section's view: j>> j>> ... v] <--> [c] <--> [m|v] <--> [c] <--> [m ... j>> Where we'd do things like: cp packeddata | tar | prog < inputs | make-metric | make-graphs | | report.awk | ghostscript | tee email | cp archive I agree that definitely fits the hexagonal architecture, or a LEGO plug-n-play. Each step is a transformation. You can see this at work with tools such as XSLT. I am not certain that type structure helps people develop an application. Too many transformations on a non-parallel machine "makes jack a dull boy". I like the multiple-adapter architecture, then again, I always have. There's a lot to be said for a "set of transformations" like the pipeline chain inside the framework (There was code like this inside Nirto -- it wasn't working last time I looked -- I think it has been orphaned). I definitely think the hexagonal architecture makes a lot of sense as a high-level model. It is a much prettier picture than the lame diagram I sent earlier. Aloha, Will. ___________________________________________________________________________ "You must BE the change you want." ~ Gandhi Success Coaching http://adroit-process.blogspot.com/ ph/sms: 041-476-1217 -----Original Message----- From: nitro-general-bounces at rubyforge.org [mailto:nitro-general-bounces at rubyforge.org] On Behalf Of James Britt Sent: Thursday, 28 December 2006 16:39 To: General discussion about Nitro Subject: Re: [Nitro] Consistency Importance: Low transfire at gmail.com wrote: (Thought: When executing a SQL select, you get back one or more rows; aren't these just 'views', representations of stuff suitable for manipulation by some client, with the DB engine being a controller, and whatever is really storing the bits being the models? ) So you have a chain of local MVC environments, where one section's model is another section's view: ... v] <--> [c] <--> [m|v] <--> [c] <--> [m ... Or something roughly like that. And it need not be 2-D (the c2.org page has one comment referring to a honeycomb). -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.432 / Virus Database: 268.15.28/604 - Release Date: 26-Dec-2006 12:23 From john at oxyliquit.de Thu Dec 28 08:07:11 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Thu, 28 Dec 2006 14:07:11 +0100 Subject: [Nitro] Consistency In-Reply-To: <45935878.8000600@gmail.com> References: <45931A72.5060008@gmail.com> <1167278034.396327.276900@48g2000cwx.googlegroups.com> <45935878.8000600@gmail.com> Message-ID: Hi, > Some time ago I was hacking on the Nitro compiler pipeline (some here > may recall my seemingly endless questions on transformations) because I > wanted to manipulate a page *after* Nitro had decided it was finished > with it. I didn't find a way to just drop one more transformer into the > mix because, as I recall, the very last bit of transformation was > coupled to the server request/response code. just as a sidenote, this 'last step' has now fully moved into the compiler pipeline (current glycerin). Jo -- Feel the love http://pinkjuice.com/pics/ruby.png From transfire at gmail.com Thu Dec 28 10:45:04 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Thu, 28 Dec 2006 15:45:04 -0000 Subject: [Nitro] Consistency In-Reply-To: <45935878.8000600@gmail.com> References: <45931A72.5060008@gmail.com> <1167278034.396327.276900@48g2000cwx.googlegroups.com> <45935878.8000600@gmail.com> Message-ID: <1167320704.647114.109370@79g2000cws.googlegroups.com> James Britt wrote: > transfire at gmail.com wrote: > These are good points. If a familiar term can be used then it should be > considered. (A caveat is that the term should be more or less > accurate. For example if something is called a 'view', but a full > explanation ends up having to describe all the ways it differs from > what people usually expect of a view, then that's just the wrong term to > use.) That's a good point too. > More food for thought along these lines: > > http://c2.com/cgi/wiki?HexagonalArchitecture > > http://alistair.cockburn.us/index.php/Hexagonal_architecture > > I like this very much; it's simpler. The symmetry is very appealing. Nice links. > Some time ago I was hacking on the Nitro compiler pipeline (some here > may recall my seemingly endless questions on transformations) because I > wanted to manipulate a page *after* Nitro had decided it was finished > with it. I didn't find a way to just drop one more transformer into the > mix because, as I recall, the very last bit of transformation was > coupled to the server request/response code. > > I was thinking that Nitro was "just" a text or request transformation > tool, and the fact that it would generate Web pages was (or could be ) > incidental. Stuff goes in, stuff comes out; "out" just happens to be a > Web server most of the time. But needn't always be; it could be another > application. > > The stuff that comes out would usually be a view of some sort, but it > could also just be data requested from some other process. > > (Thought: When executing a SQL select, you get back one or more rows; > aren't these just 'views', representations of stuff suitable for > manipulation by some client, with the DB engine being a controller, and > whatever is really storing the bits being the models? ) > > So you have a chain of local MVC environments, where one section's model > is another section's view: > > ... v] <--> [c] <--> [m|v] <--> [c] <--> [m ... Oooo... Pretty. An "MVC pipeline". T. From transfire at gmail.com Thu Dec 28 10:54:55 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Thu, 28 Dec 2006 15:54:55 -0000 Subject: [Nitro] Consistency In-Reply-To: <1167278034.396327.276900@48g2000cwx.googlegroups.com> References: <45931A72.5060008@gmail.com> <1167278034.396327.276900@48g2000cwx.googlegroups.com> Message-ID: <1167321295.306422.315200@n51g2000cwc.googlegroups.com> transfire at gmail.com wrote: > 1) It's better to utilize standard terminology when you can --even if > it's not perfect 1:1 match. It still makes things easier to talk about > and more importantly it's good for MARKETING. When you can say Nito/Og > is "MVC" then that means something to people. Using those terms in the > code then furthers that connection. > > 2) Some of the terms are simply too general. If I were to come across > one out of the blue/on it's own, I would have no idea what it was. The > terms 'entity' especially just doesn't have much semantic value, might > I suggest "domain" (see MVC on wikipedia). I would also argue that > "adapter" is much more ambiguous than "store". I think "enchant" is > good though. I feel like I "over-talked" this. I'll just give the up shot: I think "adapter" is week and "entity" is really poor, becuase the normal usage of those words is too general. T. From george.moschovitis at gmail.com Thu Dec 28 11:19:56 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 28 Dec 2006 18:19:56 +0200 Subject: [Nitro] Consistency In-Reply-To: References: <45931A72.5060008@gmail.com> <1167278034.396327.276900@48g2000cwx.googlegroups.com> <45935878.8000600@gmail.com> Message-ID: > just as a sidenote, this 'last step' has now fully moved into the compiler > pipeline (current glycerin). correct, the pipeline is now fully customizable. -g. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Thu Dec 28 12:50:08 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 28 Dec 2006 19:50:08 +0200 Subject: [Nitro] create_on_insert Message-ID: Dear devs, I added a new option to has_one relations. It's called create_on_insert. If it is set to true, the target object is automatically created when the parent object is created (insert). class User has_one :profile, :create_on_insert => true end class Profile belongs_to :user end then User.new.save automatically creates and links a default profile to the user. The reason I am writing this email is because I am looking for a better name. Any suggestions? -g. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org From george.moschovitis at gmail.com Thu Dec 28 15:06:50 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Thu, 28 Dec 2006 22:06:50 +0200 Subject: [Nitro] Resending earlier post about attributes... Message-ID: Hello Tom, I posted this some time ago but did not get any reply, perhaps you didn't see this, so I am resending this: Hello Tom, I would like to make Og work like this: class Article attr_accessor :title attr_accessor :hits end class Article ann :title, String ann :hits, Fixnum end be equivalent to: class Article attr_accessor :title, String attr_accessor :hits, Fixnum end I would appreciate it if you had a look at: glue/attribute.rb glue/attributeutils.rb and suggest ideas how this could be elegantly done. thanks in advance, George -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061228/1070c92e/attachment.html From james.britt at gmail.com Thu Dec 28 18:27:50 2006 From: james.britt at gmail.com (James Britt) Date: Thu, 28 Dec 2006 16:27:50 -0700 Subject: [Nitro] Consistency In-Reply-To: References: <45931A72.5060008@gmail.com> <1167278034.396327.276900@48g2000cwx.googlegroups.com> <45935878.8000600@gmail.com> Message-ID: <459452F6.6060803@gmail.com> Jonathan Buch wrote: > Hi, > >> Some time ago I was hacking on the Nitro compiler pipeline (some here >> may recall my seemingly endless questions on transformations) because I >> wanted to manipulate a page *after* Nitro had decided it was finished >> with it. I didn't find a way to just drop one more transformer into the >> mix because, as I recall, the very last bit of transformation was >> coupled to the server request/response code. > > just as a sidenote, this 'last step' has now fully moved into the compiler > pipeline (current glycerin). Cool. I haven't looked into the code lately; I had something working, then re-thought the whole page-generation process, and ended up with an external app to generate static pages when needed. But I know I'll want to so something like that for dynamic content again. -- James Britt http://www.rubyaz.org - Hacking in the Desert http://www.jamesbritt.com - Playing with Better Toys From dan at tastapod.com Fri Dec 29 16:48:49 2006 From: dan at tastapod.com (Dan North) Date: Fri, 29 Dec 2006 21:48:49 +0000 Subject: [Nitro] create_on_insert In-Reply-To: References: Message-ID: <45958D41.5070103@tastapod.com> :cascade_create ? George Moschovitis wrote: > Dear devs, > > I added a new option to has_one relations. It's called > create_on_insert. If it is set to true, the target object is > automatically created when the parent object is created (insert). > > class User > has_one :profile, :create_on_insert => true > end > > class Profile > belongs_to :user > end > > then > > User.new.save > > automatically creates and links a default profile to the user. > > The reason I am writing this email is because I am looking for a > better name. Any suggestions? > > -g. > > > From john at oxyliquit.de Fri Dec 29 18:38:09 2006 From: john at oxyliquit.de (Jonathan Buch) Date: Sat, 30 Dec 2006 00:38:09 +0100 Subject: [Nitro] [PATCH] psql sequence patch Message-ID: Hi, thanks go to berkus making me work on this: Sat Dec 30 00:25:29 CET 2006 Jonathan Buch * don't override sequence in psql adapter Usage: class Foo attr_accessor :oid, Fixnum, :sequence => 'some_custom_seq', :sql => 'primary key' end class Bar < Foo end This will help in the case where you don't want STI but still uniform oids. Jo -- Feel the love http://pinkjuice.com/pics/ruby.png -------------- next part -------------- A non-text attachment was scrubbed... Name: psqlseq.patch.tar.bz2 Type: application/bzip2 Size: 3430 bytes Desc: not available Url : http://rubyforge.org/pipermail/nitro-general/attachments/20061230/41948ee3/attachment.bin From transfire at gmail.com Fri Dec 29 19:48:05 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Sat, 30 Dec 2006 00:48:05 -0000 Subject: [Nitro] Resending earlier post about attributes... In-Reply-To: References: Message-ID: <1167439685.276191.120190@s34g2000cwa.googlegroups.com> George Moschovitis wrote: > Hello Tom, > > I posted this some time ago but did not get any reply, perhaps you didn't > see this, so I am resending this: > > Hello Tom, > > I would like to make Og work like this: > > class Article > attr_accessor :title > attr_accessor :hits > end > > class Article > ann :title, String > ann :hits, Fixnum > end > > be equivalent to: > > class Article > attr_accessor :title, String > attr_accessor :hits, Fixnum > end > > I would appreciate it if you had a look at: > > glue/attribute.rb > glue/attributeutils.rb > > and suggest ideas how this could be elegantly done. i'll look at it again soon. t. From george.moschovitis at gmail.com Sat Dec 30 04:04:20 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sat, 30 Dec 2006 11:04:20 +0200 Subject: [Nitro] Resending earlier post about attributes... In-Reply-To: <1167439685.276191.120190@s34g2000cwa.googlegroups.com> References: <1167439685.276191.120190@s34g2000cwa.googlegroups.com> Message-ID: > > > i'll look at it again soon. > thanks ;-) -g. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061230/7ef3dc50/attachment.html From george.moschovitis at gmail.com Sat Dec 30 04:04:44 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sat, 30 Dec 2006 11:04:44 +0200 Subject: [Nitro] [PATCH] psql sequence patch In-Reply-To: References: Message-ID: thanks! -g. On 12/30/06, Jonathan Buch wrote: > > Hi, > > thanks go to berkus making me work on this: > > Sat Dec 30 00:25:29 CET 2006 Jonathan Buch > * don't override sequence in psql adapter > > Usage: > > class Foo > attr_accessor :oid, Fixnum, :sequence => 'some_custom_seq', :sql => > 'primary key' > end > > class Bar < Foo > end > > This will help in the case where you don't want STI but still uniform > oids. > > Jo > > -- > Feel the love > http://pinkjuice.com/pics/ruby.png > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > > > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061230/253d03b6/attachment.html From transfire at gmail.com Sat Dec 30 10:08:54 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Sat, 30 Dec 2006 15:08:54 -0000 Subject: [Nitro] css-dryer Message-ID: <1167491334.707201.188740@s34g2000cwa.googlegroups.com> http://blog.airbladesoftware.com/2006/12/11/cssdryer-dry-up-your-css t. From george.moschovitis at gmail.com Sat Dec 30 12:06:24 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sat, 30 Dec 2006 19:06:24 +0200 Subject: [Nitro] css-dryer In-Reply-To: <1167491334.707201.188740@s34g2000cwa.googlegroups.com> References: <1167491334.707201.188740@s34g2000cwa.googlegroups.com> Message-ID: interesting... btw, these days I have found a better way to do such things... when I fully formulate my idea I will post to this list, then implement in Nitro. -g. On 12/30/06, transfire at gmail.com wrote: > > http://blog.airbladesoftware.com/2006/12/11/cssdryer-dry-up-your-css > > t. > > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061230/a8c12945/attachment.html From george.moschovitis at gmail.com Sat Dec 30 12:09:20 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sat, 30 Dec 2006 19:09:20 +0200 Subject: [Nitro] css-dryer In-Reply-To: References: <1167491334.707201.188740@s34g2000cwa.googlegroups.com> Message-ID: Btw, nitro already supports dynamice css (ie rubu code interpolation in css files). The current implementation can be improved though... -g. On 12/30/06, George Moschovitis wrote: > > interesting... > > btw, these days I have found a better way to do such things... when I > fully formulate my idea I will post to this list, then implement in Nitro. > > -g. > > On 12/30/06, transfire at gmail.com wrote: > > > > http://blog.airbladesoftware.com/2006/12/11/cssdryer-dry-up-your-css > > > > t. > > > > _______________________________________________ > > Nitro-general mailing list > > Nitro-general at rubyforge.org > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > > -- > http://blog.gmosx.com > http://cull.gr > http://www.joy.gr > http://nitroproject.org -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061230/94010358/attachment.html From james.britt at gmail.com Sat Dec 30 13:19:13 2006 From: james.britt at gmail.com (James Britt) Date: Sat, 30 Dec 2006 11:19:13 -0700 Subject: [Nitro] css-dryer In-Reply-To: <1167491334.707201.188740@s34g2000cwa.googlegroups.com> References: <1167491334.707201.188740@s34g2000cwa.googlegroups.com> Message-ID: <4596ADA1.4030606@gmail.com> transfire at gmail.com wrote: > http://blog.airbladesoftware.com/2006/12/11/cssdryer-dry-up-your-css Interesting. It pains me, though, to see this: <% sleek_grey = '#fefefe' %> #sidebar { border: 1px solid <%= sleek_grey %> } #footer { footer: 1px dashed <%= sleek_grey %> } The <% ... %> stuff is fugly and tedious. How hard is it to allow this? sleek_grey = '#fefefe' #sidebar { border: 1px solid sleek_grey } #footer { footer: 1px dashed sleek_grey } Answer: not very. -- James Britt "Design depends largely on constraints." - Charles Eames From transfire at gmail.com Sat Dec 30 17:00:14 2006 From: transfire at gmail.com (transfire at gmail.com) Date: Sat, 30 Dec 2006 22:00:14 -0000 Subject: [Nitro] css-dryer In-Reply-To: <4596ADA1.4030606@gmail.com> References: <1167491334.707201.188740@s34g2000cwa.googlegroups.com> <4596ADA1.4030606@gmail.com> Message-ID: <1167516014.481501.324290@h40g2000cwb.googlegroups.com> James Britt wrote: > transfire at gmail.com wrote: > > http://blog.airbladesoftware.com/2006/12/11/cssdryer-dry-up-your-css > > Interesting. > > It pains me, though, to see this: > > <% sleek_grey = '#fefefe' %> > #sidebar { border: 1px solid <%= sleek_grey %> } > #footer { footer: 1px dashed <%= sleek_grey %> } > > > The <% ... %> stuff is fugly and tedious. > > > How hard is it to allow this? > > sleek_grey = '#fefefe' > #sidebar { border: 1px solid sleek_grey } > #footer { footer: 1px dashed sleek_grey } > > > Answer: not very. +1 and Spot on! t. From george.moschovitis at gmail.com Sun Dec 31 06:02:20 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sun, 31 Dec 2006 13:02:20 +0200 Subject: [Nitro] css-dryer In-Reply-To: <1167516014.481501.324290@h40g2000cwb.googlegroups.com> References: <1167491334.707201.188740@s34g2000cwa.googlegroups.com> <4596ADA1.4030606@gmail.com> <1167516014.481501.324290@h40g2000cwb.googlegroups.com> Message-ID: > > > > How hard is it to allow this? > > > > sleek_grey = '#fefefe' > > #sidebar { border: 1px solid sleek_grey } > > #footer { footer: 1px dashed sleek_grey } > > > > > > Answer: not very. > > +1 and Spot on! +1 Nice!!! -g. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061231/12218479/attachment.html From noe.rubinstein at gmail.com Sun Dec 31 08:41:19 2006 From: noe.rubinstein at gmail.com (=?UTF-8?Q?No=C3=A9_Rubinstein?=) Date: Sun, 31 Dec 2006 14:41:19 +0100 Subject: [Nitro] Repo update In-Reply-To: References: <40b05ebe0612210735u63e19bfcw6a6ae007a8b7d430@mail.gmail.com> <9c00d3e00612211156qd1a3213l66c244c70ca7c2f3@mail.gmail.com> Message-ID: It seems the template method has to return ruby code, for nitro playing with it afterwards. Most of the template engines seem to directly return the result, not the Ruby code generated by the template. Thus, returning Ruby code containing the result is slowing down the transformation, and it would be simpler to return the result. (manv code explains that better ;) ) 2006/12/22, George Moschovitis : > > efficient... but G. could you explain why we need to give something > > eval-ready? > > especially in the case of erubis, this is just slowing it down... > > what do you mean? > > -g. > > > > > ^manveru > > > > > > > > On 12/21/06, George Moschovitis < george.moschovitis at gmail.com> wrote: > > > > Dear devs, > > > > > > > > the latest repo update contains a change that may require a minor > > > > change to make your apps run. I have made the Template.transformation > > > > stage to be just a stage in the transformation pipeline instead of > > > > beeing applied by default on every action. This way nitro now can > > > > support pluggable template engines. This change was suggested by > > > > manveru. > > > > > > > > regards, > > > > George. > > > > > > > > -- > > > > http://blog.gmosx.com > > > > http://cull.gr > > > > http://www.joy.gr > > > > http://nitroproject.org > > > > _______________________________________________ > > > > Nitro-general mailing list > > > > Nitro-general at rubyforge.org > > > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > > > > > > > > > _______________________________________________ > > > Nitro-general mailing list > > > Nitro-general at rubyforge.org > > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > > > > _______________________________________________ > > Nitro-general mailing list > > Nitro-general at rubyforge.org > > http://rubyforge.org/mailman/listinfo/nitro-general > > > > > -- > http://blog.gmosx.com > http://cull.gr > http://www.joy.gr > http://nitroproject.org > _______________________________________________ > Nitro-general mailing list > Nitro-general at rubyforge.org > http://rubyforge.org/mailman/listinfo/nitro-general > From george.moschovitis at gmail.com Sun Dec 31 09:36:26 2006 From: george.moschovitis at gmail.com (George Moschovitis) Date: Sun, 31 Dec 2006 16:36:26 +0200 Subject: [Nitro] create_on_insert In-Reply-To: <45958D41.5070103@tastapod.com> References: <45958D41.5070103@tastapod.com> Message-ID: nice idea thanks. -g. -- http://blog.gmosx.com http://cull.gr http://www.joy.gr http://nitroproject.org -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/nitro-general/attachments/20061231/ce2bc0e3/attachment.html