From djberg96 at gmail.com  Tue Jun 12 19:05:20 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Tue, 12 Jun 2012 13:05:20 -0600
Subject: [Win32utils-devel] Problem with Process.spawn + Process.kill
Message-ID: <CAGSpibkkMDX1bU_DU=+TY+PPSN8oTWxQPAQ3JsWpmUeADoYYPQ@mail.gmail.com>

Hi,

This little bit of code is causing me trouble. Ruby 1.9.3-p194 on
Windows 7, btw.

# kill_test.rb
cmd = "ruby -e 'sleep 5; at_exit{ puts \"cmd finished\" }"

pid = Process.spawn(cmd)

Process.kill('INT', pid)

puts "Done"

This results in popup dialogue that says, "The application was unable
to start correctly. Click OK to close the application."

Any ideas?

Dan

From phasis at gmail.com  Wed Jun 13 01:04:35 2012
From: phasis at gmail.com (Heesob Park)
Date: Wed, 13 Jun 2012 10:04:35 +0900
Subject: [Win32utils-devel] Problem with Process.spawn + Process.kill
In-Reply-To: <CAGSpibkkMDX1bU_DU=+TY+PPSN8oTWxQPAQ3JsWpmUeADoYYPQ@mail.gmail.com>
References: <CAGSpibkkMDX1bU_DU=+TY+PPSN8oTWxQPAQ3JsWpmUeADoYYPQ@mail.gmail.com>
Message-ID: <CAHw=Fdup30BzBSZ3irf-vGHMZFAcxDaadN384ScerybfNAieHA@mail.gmail.com>

Hi,

2012/6/13 Daniel Berger <djberg96 at gmail.com>

> Hi,
>
> This little bit of code is causing me trouble. Ruby 1.9.3-p194 on
> Windows 7, btw.
>
> # kill_test.rb
> cmd = "ruby -e 'sleep 5; at_exit{ puts \"cmd finished\" }"
>
> pid = Process.spawn(cmd)
>
> Process.kill('INT', pid)
>
> puts "Done"
>
> This results in popup dialogue that says, "The application was unable
> to start correctly. Click OK to close the application."
>
> Any ideas?
>
> Did you see the popup dialogue after Process.kill?
I guess this issue is similer to http://bugs.ruby-lang.org/issues/6535 .
If so, it is fixed on the trunk.

Regards,
Park Heesob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/win32utils-devel/attachments/20120613/319f3573/attachment.html>

From djberg96 at gmail.com  Wed Jun 13 13:24:20 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Wed, 13 Jun 2012 07:24:20 -0600
Subject: [Win32utils-devel] Problem with Process.spawn + Process.kill
In-Reply-To: <CAHw=Fdup30BzBSZ3irf-vGHMZFAcxDaadN384ScerybfNAieHA@mail.gmail.com>
References: <CAGSpibkkMDX1bU_DU=+TY+PPSN8oTWxQPAQ3JsWpmUeADoYYPQ@mail.gmail.com>
	<CAHw=Fdup30BzBSZ3irf-vGHMZFAcxDaadN384ScerybfNAieHA@mail.gmail.com>
Message-ID: <CAGSpib=NP3fR+V0+znOzaYJuUHhnnG4E-+_2mSEjjxuqdnDk_Q@mail.gmail.com>

On Tue, Jun 12, 2012 at 7:04 PM, Heesob Park <phasis at gmail.com> wrote:
> Hi,
>
> 2012/6/13 Daniel Berger <djberg96 at gmail.com>
>>
>> Hi,
>>
>> This little bit of code is causing me trouble. Ruby 1.9.3-p194 on
>> Windows 7, btw.
>>
>> # kill_test.rb
>> cmd = "ruby -e 'sleep 5; at_exit{ puts \"cmd finished\" }"
>>
>> pid = Process.spawn(cmd)
>>
>> Process.kill('INT', pid)
>>
>> puts "Done"
>>
>> This results in popup dialogue that says, "The application was unable
>> to start correctly. Click OK to close the application."
>>
>> Any ideas?
>>
> Did you see the popup dialogue after Process.kill?
> I guess this issue is similer to?http://bugs.ruby-lang.org/issues/6535?.
> If so, it is fixed on the trunk.

Yes, that's it. Thanks!

Dan

From djberg96 at gmail.com  Wed Jun 13 13:20:08 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Wed, 13 Jun 2012 07:20:08 -0600
Subject: [Win32utils-devel] Process.kill and at_exit
Message-ID: <CAGSpib=22B-pOU9kVdgjShgf3FybFKMG=NRGDGYPrburJEtGvQ@mail.gmail.com>

I could have sworn our custom Process.kill implementation for signals
1 and 4-8 (the one that uses CreateRemoteThread + ExitProcess) allowed
processes to call exit handlers. But it doesn't seem to be the case.

# kill.rb
require 'win32/process'

cmd = "ruby -e 'sleep 3; at_exit{ puts \"done\" }"
pid = Process.spawn(cmd)

Process.kill(1, pid)

But it doesn't seem that the at_exit block fires. Why not?

Regards,

Dan

From phasis at gmail.com  Fri Jun 15 04:10:25 2012
From: phasis at gmail.com (Heesob Park)
Date: Fri, 15 Jun 2012 13:10:25 +0900
Subject: [Win32utils-devel] Process.kill and at_exit
In-Reply-To: <CAGSpib=22B-pOU9kVdgjShgf3FybFKMG=NRGDGYPrburJEtGvQ@mail.gmail.com>
References: <CAGSpib=22B-pOU9kVdgjShgf3FybFKMG=NRGDGYPrburJEtGvQ@mail.gmail.com>
Message-ID: <CAHw=FdtJY3J8NMvjJcwtZY2N9THh+dT+WDq9Rw3HW4bfsyThSQ@mail.gmail.com>

Hi,

2012/6/13 Daniel Berger <djberg96 at gmail.com>

> I could have sworn our custom Process.kill implementation for signals
> 1 and 4-8 (the one that uses CreateRemoteThread + ExitProcess) allowed
> processes to call exit handlers. But it doesn't seem to be the case.
>
> # kill.rb
> require 'win32/process'
>
> cmd = "ruby -e 'sleep 3; at_exit{ puts \"done\" }"
> pid = Process.spawn(cmd)
>
> Process.kill(1, pid)
>
> But it doesn't seem that the at_exit block fires. Why not?
>
>
>
The at_exit block is called by ruby_finalize_0, ruby_finalize,
ruby_cleanup, rb_f_exit, rb_f_abort or rb_exit function.
Call ExitProcess with ruby process means skipping any of these functions.

After modifying CreateRemoteThread function in the process.rb like
following, you can see at_exit block fires.

thread = CreateRemoteThread(
                  handle,
                  0,
                  0,
                  GetProcAddress(GetModuleHandle('msvcrt-ruby191'),
'rb_f_exit'),
                  0,
                  0,
                  thread_id
                )

But you can also see the annoying error dialog.

Regards,
Park Heesob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/win32utils-devel/attachments/20120615/47181bd9/attachment.html>

From djberg96 at gmail.com  Sat Jun 16 01:35:34 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Fri, 15 Jun 2012 19:35:34 -0600
Subject: [Win32utils-devel] Process.kill and at_exit
In-Reply-To: <CAHw=FdtJY3J8NMvjJcwtZY2N9THh+dT+WDq9Rw3HW4bfsyThSQ@mail.gmail.com>
References: <CAGSpib=22B-pOU9kVdgjShgf3FybFKMG=NRGDGYPrburJEtGvQ@mail.gmail.com>
	<CAHw=FdtJY3J8NMvjJcwtZY2N9THh+dT+WDq9Rw3HW4bfsyThSQ@mail.gmail.com>
Message-ID: <CAGSpibkBjFUgOCr5Mh37-EGpm6Lqtytk-oOwtOyo-OszPBS8cg@mail.gmail.com>

On Thu, Jun 14, 2012 at 10:10 PM, Heesob Park <phasis at gmail.com> wrote:
> Hi,
>
> 2012/6/13 Daniel Berger <djberg96 at gmail.com>
>>
>> I could have sworn our custom Process.kill implementation for signals
>> 1 and 4-8 (the one that uses CreateRemoteThread + ExitProcess) allowed
>> processes to call exit handlers. But it doesn't seem to be the case.
>>
>> # kill.rb
>> require 'win32/process'
>>
>> cmd = "ruby -e 'sleep 3; at_exit{ puts \"done\" }"
>> pid = Process.spawn(cmd)
>>
>> Process.kill(1, pid)
>>
>> But it doesn't seem that the at_exit block fires. Why not?
>>
>>
>
> The at_exit block is called by?ruby_finalize_0, ruby_finalize, ruby_cleanup,
> rb_f_exit, rb_f_abort or rb_exit function.
> Call ExitProcess with ruby process means skipping any of these functions.
>
> After modifying?CreateRemoteThread?function in the process.rb like
> following, you can see at_exit block fires.
>
> thread = CreateRemoteThread(
> ? ? ? ? ? ? ? ? ? handle,
> ? ? ? ? ? ? ? ? ? 0,
> ? ? ? ? ? ? ? ? ? 0,
> ? ? ? ? ? ? ? ? ? GetProcAddress(GetModuleHandle('msvcrt-ruby191'),
> 'rb_f_exit'),
> ? ? ? ? ? ? ? ? ? 0,
> ? ? ? ? ? ? ? ? ? 0,
> ? ? ? ? ? ? ? ? ? thread_id
> ? ? ? ? ? ? ? ? )
>
> But you can also see the annoying error dialog.

Thanks for this! I'm rewriting Process.kill using FFI, and now I'm
thinking I might want to alter the interface so that people can pass
in more types of arguments.

Regards,

Dan

From djberg96 at gmail.com  Sun Jun 17 17:29:38 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Sun, 17 Jun 2012 11:29:38 -0600
Subject: [Win32utils-devel] Process.kill and at_exit
In-Reply-To: <CAHw=FdtJY3J8NMvjJcwtZY2N9THh+dT+WDq9Rw3HW4bfsyThSQ@mail.gmail.com>
References: <CAGSpib=22B-pOU9kVdgjShgf3FybFKMG=NRGDGYPrburJEtGvQ@mail.gmail.com>
	<CAHw=FdtJY3J8NMvjJcwtZY2N9THh+dT+WDq9Rw3HW4bfsyThSQ@mail.gmail.com>
Message-ID: <CAGSpibm5S-CtPWjRrm_6ikSQ5dHRVzJ_XSJ3cZWESc4GreDAJw@mail.gmail.com>

On Thu, Jun 14, 2012 at 10:10 PM, Heesob Park <phasis at gmail.com> wrote:
> Hi,
>
> 2012/6/13 Daniel Berger <djberg96 at gmail.com>
>>
>> I could have sworn our custom Process.kill implementation for signals
>> 1 and 4-8 (the one that uses CreateRemoteThread + ExitProcess) allowed
>> processes to call exit handlers. But it doesn't seem to be the case.
>>
>> # kill.rb
>> require 'win32/process'
>>
>> cmd = "ruby -e 'sleep 3; at_exit{ puts \"done\" }"
>> pid = Process.spawn(cmd)
>>
>> Process.kill(1, pid)
>>
>> But it doesn't seem that the at_exit block fires. Why not?
>>
>>
>
> The at_exit block is called by?ruby_finalize_0, ruby_finalize, ruby_cleanup,
> rb_f_exit, rb_f_abort or rb_exit function.
> Call ExitProcess with ruby process means skipping any of these functions.
>
> After modifying?CreateRemoteThread?function in the process.rb like
> following, you can see at_exit block fires.
>
> thread = CreateRemoteThread(
> ? ? ? ? ? ? ? ? ? handle,
> ? ? ? ? ? ? ? ? ? 0,
> ? ? ? ? ? ? ? ? ? 0,
> ? ? ? ? ? ? ? ? ? GetProcAddress(GetModuleHandle('msvcrt-ruby191'),
> 'rb_f_exit'),
> ? ? ? ? ? ? ? ? ? 0,
> ? ? ? ? ? ? ? ? ? 0,
> ? ? ? ? ? ? ? ? ? thread_id
> ? ? ? ? ? ? ? ? )
>
> But you can also see the annoying error dialog.

I tried this with:  ruby 1.9.3p194 (2012-04-20 revision 35410)
[i386-mswin32_100]

I did not have much luck, however.

When I ran this using rb_f_exit, rb_exit or rb_f_abort I did not see
the at_exit function call get executed, though the process was killed
successfully.

I got mixed results with ruby_cleanup, sometimes it worked, others it
seemed to cause a segfault. With ruby_finalize I get a
SystemStackError. It doesn't seem ruby_finalize_0 is actually
exported.

If you checkout the "ffi" branch for win32-process you can see that
I've added some hash options, so you can now do this;

Process.kill(1, pid, :exit_proc => "rb_f_exit", :dll_module =>
"msvcr100-ruby191", :wait_time => 25)

There's also a shortcut for Ruby processes that just sets :exit_proc
and :dll_module to default values:

Process.kill(1, pid, :ruby_proc => true)

Can you please take a look and see if this is working for you?

Regards,

Dan

From phasis at gmail.com  Mon Jun 18 01:55:56 2012
From: phasis at gmail.com (Heesob Park)
Date: Mon, 18 Jun 2012 10:55:56 +0900
Subject: [Win32utils-devel] Process.kill and at_exit
In-Reply-To: <CAGSpibm5S-CtPWjRrm_6ikSQ5dHRVzJ_XSJ3cZWESc4GreDAJw@mail.gmail.com>
References: <CAGSpib=22B-pOU9kVdgjShgf3FybFKMG=NRGDGYPrburJEtGvQ@mail.gmail.com>
	<CAHw=FdtJY3J8NMvjJcwtZY2N9THh+dT+WDq9Rw3HW4bfsyThSQ@mail.gmail.com>
	<CAGSpibm5S-CtPWjRrm_6ikSQ5dHRVzJ_XSJ3cZWESc4GreDAJw@mail.gmail.com>
Message-ID: <CAHw=Fds97my0+ssBJPsDnGnV417=e7FCxHO-_0t+PaT_B1+FdQ@mail.gmail.com>

Hi,

2012/6/18 Daniel Berger <djberg96 at gmail.com>

> On Thu, Jun 14, 2012 at 10:10 PM, Heesob Park <phasis at gmail.com> wrote:
> > Hi,
> >
> > 2012/6/13 Daniel Berger <djberg96 at gmail.com>
> >>
> >> I could have sworn our custom Process.kill implementation for signals
> >> 1 and 4-8 (the one that uses CreateRemoteThread + ExitProcess) allowed
> >> processes to call exit handlers. But it doesn't seem to be the case.
> >>
> >> # kill.rb
> >> require 'win32/process'
> >>
> >> cmd = "ruby -e 'sleep 3; at_exit{ puts \"done\" }"
> >> pid = Process.spawn(cmd)
> >>
> >> Process.kill(1, pid)
> >>
> >> But it doesn't seem that the at_exit block fires. Why not?
> >>
> >>
> >
> > The at_exit block is called by ruby_finalize_0, ruby_finalize,
> ruby_cleanup,
> > rb_f_exit, rb_f_abort or rb_exit function.
> > Call ExitProcess with ruby process means skipping any of these functions.
> >
> > After modifying CreateRemoteThread function in the process.rb like
> > following, you can see at_exit block fires.
> >
> > thread = CreateRemoteThread(
> >                   handle,
> >                   0,
> >                   0,
> >                   GetProcAddress(GetModuleHandle('msvcrt-ruby191'),
> > 'rb_f_exit'),
> >                   0,
> >                   0,
> >                   thread_id
> >                 )
> >
> > But you can also see the annoying error dialog.
>
> I tried this with:  ruby 1.9.3p194 (2012-04-20 revision 35410)
> [i386-mswin32_100]
>
> I did not have much luck, however.
>
> When I ran this using rb_f_exit, rb_exit or rb_f_abort I did not see
> the at_exit function call get executed, though the process was killed
> successfully.
>
> I got mixed results with ruby_cleanup, sometimes it worked, others it
> seemed to cause a segfault. With ruby_finalize I get a
> SystemStackError. It doesn't seem ruby_finalize_0 is actually
> exported.
>
> If you checkout the "ffi" branch for win32-process you can see that
> I've added some hash options, so you can now do this;
>
> Process.kill(1, pid, :exit_proc => "rb_f_exit", :dll_module =>
> "msvcr100-ruby191", :wait_time => 25)
>
> There's also a shortcut for Ruby processes that just sets :exit_proc
> and :dll_module to default values:
>
> Process.kill(1, pid, :ruby_proc => true)
>
> Can you please take a look and see if this is working for you?
>
>
If you comment out SetErrorMode process.rb like following, you can see the
at_exit function call get executed.

# SetErrorMode(SEM_FAILCRITICALERRORS | SEM_NOGPFAULTERRORBOX)


Regards,
Park Heesob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/win32utils-devel/attachments/20120618/398808e3/attachment.html>

From phasis at gmail.com  Mon Jun 18 02:51:13 2012
From: phasis at gmail.com (Heesob Park)
Date: Mon, 18 Jun 2012 11:51:13 +0900
Subject: [Win32utils-devel] Process.kill and at_exit
In-Reply-To: <CAHw=Fds97my0+ssBJPsDnGnV417=e7FCxHO-_0t+PaT_B1+FdQ@mail.gmail.com>
References: <CAGSpib=22B-pOU9kVdgjShgf3FybFKMG=NRGDGYPrburJEtGvQ@mail.gmail.com>
	<CAHw=FdtJY3J8NMvjJcwtZY2N9THh+dT+WDq9Rw3HW4bfsyThSQ@mail.gmail.com>
	<CAGSpibm5S-CtPWjRrm_6ikSQ5dHRVzJ_XSJ3cZWESc4GreDAJw@mail.gmail.com>
	<CAHw=Fds97my0+ssBJPsDnGnV417=e7FCxHO-_0t+PaT_B1+FdQ@mail.gmail.com>
Message-ID: <CAHw=FdtoMJNUWwatLs8HW9-K5ouctoasbFs0UYHJUTQPrX1vuQ@mail.gmail.com>

Hi,

2012/6/18 Heesob Park <phasis at gmail.com>

> Hi,
>
> 2012/6/18 Daniel Berger <djberg96 at gmail.com>
>
>> On Thu, Jun 14, 2012 at 10:10 PM, Heesob Park <phasis at gmail.com> wrote:
>> > Hi,
>> >
>> > 2012/6/13 Daniel Berger <djberg96 at gmail.com>
>> >>
>> >> I could have sworn our custom Process.kill implementation for signals
>> >> 1 and 4-8 (the one that uses CreateRemoteThread + ExitProcess) allowed
>> >> processes to call exit handlers. But it doesn't seem to be the case.
>> >>
>> >> # kill.rb
>> >> require 'win32/process'
>> >>
>> >> cmd = "ruby -e 'sleep 3; at_exit{ puts \"done\" }"
>> >> pid = Process.spawn(cmd)
>> >>
>> >> Process.kill(1, pid)
>> >>
>> >> But it doesn't seem that the at_exit block fires. Why not?
>> >>
>> >>
>> >
>> > The at_exit block is called by ruby_finalize_0, ruby_finalize,
>> ruby_cleanup,
>> > rb_f_exit, rb_f_abort or rb_exit function.
>> > Call ExitProcess with ruby process means skipping any of these
>> functions.
>> >
>> > After modifying CreateRemoteThread function in the process.rb like
>> > following, you can see at_exit block fires.
>> >
>> > thread = CreateRemoteThread(
>> >                   handle,
>> >                   0,
>> >                   0,
>> >                   GetProcAddress(GetModuleHandle('msvcrt-ruby191'),
>> > 'rb_f_exit'),
>> >                   0,
>> >                   0,
>> >                   thread_id
>> >                 )
>> >
>> > But you can also see the annoying error dialog.
>>
>> I tried this with:  ruby 1.9.3p194 (2012-04-20 revision 35410)
>> [i386-mswin32_100]
>>
>> I did not have much luck, however.
>>
>> When I ran this using rb_f_exit, rb_exit or rb_f_abort I did not see
>> the at_exit function call get executed, though the process was killed
>> successfully.
>>
>> I got mixed results with ruby_cleanup, sometimes it worked, others it
>> seemed to cause a segfault. With ruby_finalize I get a
>> SystemStackError. It doesn't seem ruby_finalize_0 is actually
>> exported.
>>
>> If you checkout the "ffi" branch for win32-process you can see that
>> I've added some hash options, so you can now do this;
>>
>> Process.kill(1, pid, :exit_proc => "rb_f_exit", :dll_module =>
>> "msvcr100-ruby191", :wait_time => 25)
>>
>> There's also a shortcut for Ruby processes that just sets :exit_proc
>> and :dll_module to default values:
>>
>> Process.kill(1, pid, :ruby_proc => true)
>>
>> Can you please take a look and see if this is working for you?
>>
>>
> If you comment out SetErrorMode process.rb like following, you can see the
> at_exit function call get executed.
>
> # SetErrorMode(SEM_FAILCRITICALERRORS | SEM_NOGPFAULTERRORBOX)
>
>
> Sorry, I realized that the above statement is wrong.
I guess it is not safe to call a ruby exit function in the different thread
which is not created in the ruby process.

Regards,
Park Heesob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/win32utils-devel/attachments/20120618/5d9482ce/attachment-0001.html>

From djberg96 at gmail.com  Mon Jun 18 02:53:48 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Sun, 17 Jun 2012 20:53:48 -0600
Subject: [Win32utils-devel] Process.kill and at_exit
In-Reply-To: <CAHw=Fds97my0+ssBJPsDnGnV417=e7FCxHO-_0t+PaT_B1+FdQ@mail.gmail.com>
References: <CAGSpib=22B-pOU9kVdgjShgf3FybFKMG=NRGDGYPrburJEtGvQ@mail.gmail.com>
	<CAHw=FdtJY3J8NMvjJcwtZY2N9THh+dT+WDq9Rw3HW4bfsyThSQ@mail.gmail.com>
	<CAGSpibm5S-CtPWjRrm_6ikSQ5dHRVzJ_XSJ3cZWESc4GreDAJw@mail.gmail.com>
	<CAHw=Fds97my0+ssBJPsDnGnV417=e7FCxHO-_0t+PaT_B1+FdQ@mail.gmail.com>
Message-ID: <CAGSpibn=JVxWn6zkHfVWvvZa06cOatkdNUTtN2ojwYeuJj=cKg@mail.gmail.com>

On Sun, Jun 17, 2012 at 7:55 PM, Heesob Park <phasis at gmail.com> wrote:
> Hi,
>
> 2012/6/18 Daniel Berger <djberg96 at gmail.com>
>>
>> On Thu, Jun 14, 2012 at 10:10 PM, Heesob Park <phasis at gmail.com> wrote:
>> > Hi,
>> >
>> > 2012/6/13 Daniel Berger <djberg96 at gmail.com>
>> >>
>> >> I could have sworn our custom Process.kill implementation for signals
>> >> 1 and 4-8 (the one that uses CreateRemoteThread + ExitProcess) allowed
>> >> processes to call exit handlers. But it doesn't seem to be the case.
>> >>
>> >> # kill.rb
>> >> require 'win32/process'
>> >>
>> >> cmd = "ruby -e 'sleep 3; at_exit{ puts \"done\" }"
>> >> pid = Process.spawn(cmd)
>> >>
>> >> Process.kill(1, pid)
>> >>
>> >> But it doesn't seem that the at_exit block fires. Why not?
>> >>
>> >>
>> >
>> > The at_exit block is called by?ruby_finalize_0, ruby_finalize,
>> > ruby_cleanup,
>> > rb_f_exit, rb_f_abort or rb_exit function.
>> > Call ExitProcess with ruby process means skipping any of these
>> > functions.
>> >
>> > After modifying?CreateRemoteThread?function in the process.rb like
>> > following, you can see at_exit block fires.
>> >
>> > thread = CreateRemoteThread(
>> > ? ? ? ? ? ? ? ? ? handle,
>> > ? ? ? ? ? ? ? ? ? 0,
>> > ? ? ? ? ? ? ? ? ? 0,
>> > ? ? ? ? ? ? ? ? ? GetProcAddress(GetModuleHandle('msvcrt-ruby191'),
>> > 'rb_f_exit'),
>> > ? ? ? ? ? ? ? ? ? 0,
>> > ? ? ? ? ? ? ? ? ? 0,
>> > ? ? ? ? ? ? ? ? ? thread_id
>> > ? ? ? ? ? ? ? ? )
>> >
>> > But you can also see the annoying error dialog.
>>
>> I tried this with: ?ruby 1.9.3p194 (2012-04-20 revision 35410)
>> [i386-mswin32_100]
>>
>> I did not have much luck, however.
>>
>> When I ran this using rb_f_exit, rb_exit or rb_f_abort I did not see
>> the at_exit function call get executed, though the process was killed
>> successfully.
>>
>> I got mixed results with ruby_cleanup, sometimes it worked, others it
>> seemed to cause a segfault. With ruby_finalize I get a
>> SystemStackError. It doesn't seem ruby_finalize_0 is actually
>> exported.
>>
>> If you checkout the "ffi" branch for win32-process you can see that
>> I've added some hash options, so you can now do this;
>>
>> Process.kill(1, pid, :exit_proc => "rb_f_exit", :dll_module =>
>> "msvcr100-ruby191", :wait_time => 25)
>>
>> There's also a shortcut for Ruby processes that just sets :exit_proc
>> and :dll_module to default values:
>>
>> Process.kill(1, pid, :ruby_proc => true)
>>
>> Can you please take a look and see if this is working for you?
>>
>
> If you comment out SetErrorMode process.rb like following, you can see?the
> at_exit function call get executed.
>
> # SetErrorMode(SEM_FAILCRITICALERRORS | SEM_NOGPFAULTERRORBOX)

I still don't see it. I see the dialogue come up, but I don't see the
at_exit function get executed. And is there any way to avoid this and
avoid the annoying dialogue? Or would people have to setup their own
SetConsoleCtrlHandler function?

Regards,

Dan

From phasis at gmail.com  Mon Jun 18 03:07:02 2012
From: phasis at gmail.com (Heesob Park)
Date: Mon, 18 Jun 2012 12:07:02 +0900
Subject: [Win32utils-devel] Process.kill and at_exit
In-Reply-To: <CAGSpibn=JVxWn6zkHfVWvvZa06cOatkdNUTtN2ojwYeuJj=cKg@mail.gmail.com>
References: <CAGSpib=22B-pOU9kVdgjShgf3FybFKMG=NRGDGYPrburJEtGvQ@mail.gmail.com>
	<CAHw=FdtJY3J8NMvjJcwtZY2N9THh+dT+WDq9Rw3HW4bfsyThSQ@mail.gmail.com>
	<CAGSpibm5S-CtPWjRrm_6ikSQ5dHRVzJ_XSJ3cZWESc4GreDAJw@mail.gmail.com>
	<CAHw=Fds97my0+ssBJPsDnGnV417=e7FCxHO-_0t+PaT_B1+FdQ@mail.gmail.com>
	<CAGSpibn=JVxWn6zkHfVWvvZa06cOatkdNUTtN2ojwYeuJj=cKg@mail.gmail.com>
Message-ID: <CAHw=FdtZYxnJeLB0wG2SUsxgVHBPEug8aQGiCWP4DgKQ7+je2g@mail.gmail.com>

Hi,

2012/6/18 Daniel Berger <djberg96 at gmail.com>

> On Sun, Jun 17, 2012 at 7:55 PM, Heesob Park <phasis at gmail.com> wrote:
> > Hi,
> >
> > 2012/6/18 Daniel Berger <djberg96 at gmail.com>
> >>
> >> On Thu, Jun 14, 2012 at 10:10 PM, Heesob Park <phasis at gmail.com> wrote:
> >> > Hi,
> >> >
> >> > 2012/6/13 Daniel Berger <djberg96 at gmail.com>
> >> >>
> >> >> I could have sworn our custom Process.kill implementation for signals
> >> >> 1 and 4-8 (the one that uses CreateRemoteThread + ExitProcess)
> allowed
> >> >> processes to call exit handlers. But it doesn't seem to be the case.
> >> >>
> >> >> # kill.rb
> >> >> require 'win32/process'
> >> >>
> >> >> cmd = "ruby -e 'sleep 3; at_exit{ puts \"done\" }"
> >> >> pid = Process.spawn(cmd)
> >> >>
> >> >> Process.kill(1, pid)
> >> >>
> >> >> But it doesn't seem that the at_exit block fires. Why not?
> >> >>
> >> >>
> >> >
> >> > The at_exit block is called by ruby_finalize_0, ruby_finalize,
> >> > ruby_cleanup,
> >> > rb_f_exit, rb_f_abort or rb_exit function.
> >> > Call ExitProcess with ruby process means skipping any of these
> >> > functions.
> >> >
> >> > After modifying CreateRemoteThread function in the process.rb like
> >> > following, you can see at_exit block fires.
> >> >
> >> > thread = CreateRemoteThread(
> >> >                   handle,
> >> >                   0,
> >> >                   0,
> >> >                   GetProcAddress(GetModuleHandle('msvcrt-ruby191'),
> >> > 'rb_f_exit'),
> >> >                   0,
> >> >                   0,
> >> >                   thread_id
> >> >                 )
> >> >
> >> > But you can also see the annoying error dialog.
> >>
> >> I tried this with:  ruby 1.9.3p194 (2012-04-20 revision 35410)
> >> [i386-mswin32_100]
> >>
> >> I did not have much luck, however.
> >>
> >> When I ran this using rb_f_exit, rb_exit or rb_f_abort I did not see
> >> the at_exit function call get executed, though the process was killed
> >> successfully.
> >>
> >> I got mixed results with ruby_cleanup, sometimes it worked, others it
> >> seemed to cause a segfault. With ruby_finalize I get a
> >> SystemStackError. It doesn't seem ruby_finalize_0 is actually
> >> exported.
> >>
> >> If you checkout the "ffi" branch for win32-process you can see that
> >> I've added some hash options, so you can now do this;
> >>
> >> Process.kill(1, pid, :exit_proc => "rb_f_exit", :dll_module =>
> >> "msvcr100-ruby191", :wait_time => 25)
> >>
> >> There's also a shortcut for Ruby processes that just sets :exit_proc
> >> and :dll_module to default values:
> >>
> >> Process.kill(1, pid, :ruby_proc => true)
> >>
> >> Can you please take a look and see if this is working for you?
> >>
> >
> > If you comment out SetErrorMode process.rb like following, you can
> see the
> > at_exit function call get executed.
> >
> > # SetErrorMode(SEM_FAILCRITICALERRORS | SEM_NOGPFAULTERRORBOX)
>
> I still don't see it. I see the dialogue come up, but I don't see the
> at_exit function get executed. And is there any way to avoid this and
> avoid the annoying dialogue? Or would people have to setup their own
> SetConsoleCtrlHandler function?
>
> You missed my last message.
Commenting out SetErrorMode is not the solution.
And calling a ruby function with a thread which is not created in the ruby
process is not safe and undesirable.

I think using inter-process communication like signal or event handling is
reliable.

Regards,
Park Heesob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/win32utils-devel/attachments/20120618/ae18c08c/attachment.html>

From djberg96 at gmail.com  Mon Jun 18 05:38:49 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Sun, 17 Jun 2012 23:38:49 -0600
Subject: [Win32utils-devel] Process.kill and at_exit
In-Reply-To: <CAHw=FdtZYxnJeLB0wG2SUsxgVHBPEug8aQGiCWP4DgKQ7+je2g@mail.gmail.com>
References: <CAGSpib=22B-pOU9kVdgjShgf3FybFKMG=NRGDGYPrburJEtGvQ@mail.gmail.com>
	<CAHw=FdtJY3J8NMvjJcwtZY2N9THh+dT+WDq9Rw3HW4bfsyThSQ@mail.gmail.com>
	<CAGSpibm5S-CtPWjRrm_6ikSQ5dHRVzJ_XSJ3cZWESc4GreDAJw@mail.gmail.com>
	<CAHw=Fds97my0+ssBJPsDnGnV417=e7FCxHO-_0t+PaT_B1+FdQ@mail.gmail.com>
	<CAGSpibn=JVxWn6zkHfVWvvZa06cOatkdNUTtN2ojwYeuJj=cKg@mail.gmail.com>
	<CAHw=FdtZYxnJeLB0wG2SUsxgVHBPEug8aQGiCWP4DgKQ7+je2g@mail.gmail.com>
Message-ID: <CAGSpibnjCn6iY--nTHr9Shjbi6zG7ymvgyF5c9B_wDjeaNwN8w@mail.gmail.com>

On Sun, Jun 17, 2012 at 9:07 PM, Heesob Park <phasis at gmail.com> wrote:
> Hi,
>
> 2012/6/18 Daniel Berger <djberg96 at gmail.com>
>>
>> On Sun, Jun 17, 2012 at 7:55 PM, Heesob Park <phasis at gmail.com> wrote:
>> > Hi,
>> >
>> > 2012/6/18 Daniel Berger <djberg96 at gmail.com>
>> >>
>> >> On Thu, Jun 14, 2012 at 10:10 PM, Heesob Park <phasis at gmail.com> wrote:
>> >> > Hi,
>> >> >
>> >> > 2012/6/13 Daniel Berger <djberg96 at gmail.com>
>> >> >>
>> >> >> I could have sworn our custom Process.kill implementation for
>> >> >> signals
>> >> >> 1 and 4-8 (the one that uses CreateRemoteThread + ExitProcess)
>> >> >> allowed
>> >> >> processes to call exit handlers. But it doesn't seem to be the case.
>> >> >>
>> >> >> # kill.rb
>> >> >> require 'win32/process'
>> >> >>
>> >> >> cmd = "ruby -e 'sleep 3; at_exit{ puts \"done\" }"
>> >> >> pid = Process.spawn(cmd)
>> >> >>
>> >> >> Process.kill(1, pid)
>> >> >>
>> >> >> But it doesn't seem that the at_exit block fires. Why not?
>> >> >>
>> >> >>
>> >> >
>> >> > The at_exit block is called by?ruby_finalize_0, ruby_finalize,
>> >> > ruby_cleanup,
>> >> > rb_f_exit, rb_f_abort or rb_exit function.
>> >> > Call ExitProcess with ruby process means skipping any of these
>> >> > functions.
>> >> >
>> >> > After modifying?CreateRemoteThread?function in the process.rb like
>> >> > following, you can see at_exit block fires.
>> >> >
>> >> > thread = CreateRemoteThread(
>> >> > ? ? ? ? ? ? ? ? ? handle,
>> >> > ? ? ? ? ? ? ? ? ? 0,
>> >> > ? ? ? ? ? ? ? ? ? 0,
>> >> > ? ? ? ? ? ? ? ? ? GetProcAddress(GetModuleHandle('msvcrt-ruby191'),
>> >> > 'rb_f_exit'),
>> >> > ? ? ? ? ? ? ? ? ? 0,
>> >> > ? ? ? ? ? ? ? ? ? 0,
>> >> > ? ? ? ? ? ? ? ? ? thread_id
>> >> > ? ? ? ? ? ? ? ? )
>> >> >
>> >> > But you can also see the annoying error dialog.
>> >>
>> >> I tried this with: ?ruby 1.9.3p194 (2012-04-20 revision 35410)
>> >> [i386-mswin32_100]
>> >>
>> >> I did not have much luck, however.
>> >>
>> >> When I ran this using rb_f_exit, rb_exit or rb_f_abort I did not see
>> >> the at_exit function call get executed, though the process was killed
>> >> successfully.
>> >>
>> >> I got mixed results with ruby_cleanup, sometimes it worked, others it
>> >> seemed to cause a segfault. With ruby_finalize I get a
>> >> SystemStackError. It doesn't seem ruby_finalize_0 is actually
>> >> exported.
>> >>
>> >> If you checkout the "ffi" branch for win32-process you can see that
>> >> I've added some hash options, so you can now do this;
>> >>
>> >> Process.kill(1, pid, :exit_proc => "rb_f_exit", :dll_module =>
>> >> "msvcr100-ruby191", :wait_time => 25)
>> >>
>> >> There's also a shortcut for Ruby processes that just sets :exit_proc
>> >> and :dll_module to default values:
>> >>
>> >> Process.kill(1, pid, :ruby_proc => true)
>> >>
>> >> Can you please take a look and see if this is working for you?
>> >>
>> >
>> > If you comment out SetErrorMode process.rb like following, you can
>> > see?the
>> > at_exit function call get executed.
>> >
>> > # SetErrorMode(SEM_FAILCRITICALERRORS | SEM_NOGPFAULTERRORBOX)
>>
>> I still don't see it. I see the dialogue come up, but I don't see the
>> at_exit function get executed. And is there any way to avoid this and
>> avoid the annoying dialogue? Or would people have to setup their own
>> SetConsoleCtrlHandler function?
>>
> You missed my last message.
> Commenting out SetErrorMode is not the solution.
> And calling a ruby function with a thread which is not created in the ruby
> process is not safe and undesirable.
>
> I think using inter-process communication like signal or event handling is
> reliable.

Darn, I was afraid of that. I guess I'll remove the :ruby_proc option then.

I'm pondering changes to core Ruby so that this -would- be a safer option.

Regards,

Dan

From djberg96 at gmail.com  Fri Jun 22 01:01:16 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Thu, 21 Jun 2012 19:01:16 -0600
Subject: [Win32utils-devel] Trying to get RubyInline to work with VS
Message-ID: <CAGSpib=VHw4kNBhAeG0nR=hnsHgiXpNv4LZ0BndnyYHzcM=4Pg@mail.gmail.com>

Hi,

I noticed that RubyInline did not work with Visual Studio (or mingw
for that matter). I forked RubyInline and made some patches that look
like they ought to work. Mostly it was just adjusting the build
command string so that linker options came after the -link option,
though I also changed the default extension to .dll and I make a
Dir.chdir call before the build happens so that it doesn't build in
the current directory.

https://github.com/djberg96/rubyinline

However, even after my patches I'm still having trouble. It builds,
but it won't load. I tried it with this simple bit of inline code:

# factorial.rb
require "inline"
class MyTest
  inline do |builder|
    builder.c "
      long factorial(int max) {
        int i=max, result=1;
        while (i >= 2) { result *= i--; }
        return result;
      }"
  end
end

t = MyTest.new()
p t.factorial(5)

I can see that it generated the .c code and built the .dll file in
c:\Users\djberge\.ruby_inline\ruby-1.9.1

Inline_MyTest_cb89593d1f9fe3ecdb2178215eee80ae.c
Inline_MyTest_cb89593d1f9fe3ecdb2178215eee80ae.dll

However, it fails trying to load the file:

c:/usr/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb:36:in
`require': cannot load such file --
   C:/Users/djberge/.ruby_inline/ruby-1.9.1/Inline_MyTest_cb89593d1f9fe3ecdb2178215eee80ae.dll
(LoadError)
        from c:/usr/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb:36:in
`require'
        from c:/usr/lib/ruby/gems/1.9.1/gems/RubyInline-3.11.2/lib/inline.rb:522:in
`load'
        from c:/usr/lib/ruby/gems/1.9.1/gems/RubyInline-3.11.2/lib/inline.rb:859:in
`inline'
        from factorial.rb:3:in `<class:MyTest>'
        from factorial.rb:2:in `<main>'

What have I done wrong?

Regards,

Dan

From phasis at gmail.com  Fri Jun 22 02:27:01 2012
From: phasis at gmail.com (Heesob Park)
Date: Fri, 22 Jun 2012 11:27:01 +0900
Subject: [Win32utils-devel] Trying to get RubyInline to work with VS
In-Reply-To: <CAGSpib=VHw4kNBhAeG0nR=hnsHgiXpNv4LZ0BndnyYHzcM=4Pg@mail.gmail.com>
References: <CAGSpib=VHw4kNBhAeG0nR=hnsHgiXpNv4LZ0BndnyYHzcM=4Pg@mail.gmail.com>
Message-ID: <CAHw=FdsWaSXaKG=JwbUnAfSAJzKYe2biUWQT9Yis2CczbtYkrg@mail.gmail.com>

Hi,

2012/6/22 Daniel Berger <djberg96 at gmail.com>

> Hi,
>
> I noticed that RubyInline did not work with Visual Studio (or mingw
> for that matter). I forked RubyInline and made some patches that look
> like they ought to work. Mostly it was just adjusting the build
> command string so that linker options came after the -link option,
> though I also changed the default extension to .dll and I make a
> Dir.chdir call before the build happens so that it doesn't build in
> the current directory.
>
> https://github.com/djberg96/rubyinline
>
> However, even after my patches I'm still having trouble. It builds,
> but it won't load. I tried it with this simple bit of inline code:
>
> # factorial.rb
> require "inline"
> class MyTest
>  inline do |builder|
>    builder.c "
>      long factorial(int max) {
>        int i=max, result=1;
>        while (i >= 2) { result *= i--; }
>        return result;
>      }"
>  end
> end
>
> t = MyTest.new()
> p t.factorial(5)
>
> I can see that it generated the .c code and built the .dll file in
> c:\Users\djberge\.ruby_inline\ruby-1.9.1
>
> Inline_MyTest_cb89593d1f9fe3ecdb2178215eee80ae.c
> Inline_MyTest_cb89593d1f9fe3ecdb2178215eee80ae.dll
>
> However, it fails trying to load the file:
>
> c:/usr/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb:36:in
> `require': cannot load such file --
>
> C:/Users/djberge/.ruby_inline/ruby-1.9.1/Inline_MyTest_cb89593d1f9fe3ecdb2178215eee80ae.dll
> (LoadError)
>        from
> c:/usr/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb:36:in
> `require'
>        from
> c:/usr/lib/ruby/gems/1.9.1/gems/RubyInline-3.11.2/lib/inline.rb:522:in
> `load'
>        from
> c:/usr/lib/ruby/gems/1.9.1/gems/RubyInline-3.11.2/lib/inline.rb:859:in
> `inline'
>        from factorial.rb:3:in `<class:MyTest>'
>        from factorial.rb:2:in `<main>'
>
> What have I done wrong?
>
>
Ruby wants not .dll but .so for the extension name.

Here is a patch for your forked Rubyinline.

diff --git a/inline.rb b/inline.rb.new
index 3ac2f41..de45ee1 100644
--- a/inline.rb
+++ b/inline.rb.new
@@ -377,11 +377,7 @@ module Inline

     def so_name
       unless defined? @so_name then
-        if WINDOZE
-          @so_name = "#{Inline.directory}/#{module_name}.dll"
-        else
           @so_name =
"#{Inline.directory}/#{module_name}.#{RbConfig::CONFIG["DLEXT"]}"
-        end
       end
       @so_name
     end
@@ -571,7 +567,7 @@ VALUE #{method}_equals(VALUE value) {
                             nil
                           end

-          if WINDOZE
+          if WINDOZE && RUBY_PLATFORM =~ /mswin/
             cmd = [ RbConfig::CONFIG['LDSHARED'],
                     flags,
                     RbConfig::CONFIG['CFLAGS'],
@@ -657,7 +653,7 @@ VALUE #{method}_equals(VALUE value) {
       # gawd windoze land sucks
       case RUBY_PLATFORM
       when /mswin32/ then
-        " -link /LIBPATH:\"#{RbConfig::CONFIG['libdir']}\"
/DEFAULTLIB:\"#{RbConfig::CONFIG['LIBRUBY']}\" /INCREMENTAL:no
/EXPORT:Init_#{module_name}"
+        " -link /OUT:\"#{self.so_name}\"
/LIBPATH:\"#{RbConfig::CONFIG['libdir']}\"
/DEFAULTLIB:\"#{RbConfig::CONFIG['LIBRUBY']}\" /INCREMENTAL:no
/EXPORT:Init_#{module_name}"
       when /mingw32/ then
         c = RbConfig::CONFIG
         " -Wl,--enable-auto-import -L#{c['libdir']}
-l#{c['RUBY_SO_NAME']}"


Regards,
Park Heesob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/win32utils-devel/attachments/20120622/1853edb4/attachment-0001.html>

From djberg96 at gmail.com  Fri Jun 22 04:23:14 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Thu, 21 Jun 2012 22:23:14 -0600
Subject: [Win32utils-devel] Trying to get RubyInline to work with VS
In-Reply-To: <CAHw=FdsWaSXaKG=JwbUnAfSAJzKYe2biUWQT9Yis2CczbtYkrg@mail.gmail.com>
References: <CAGSpib=VHw4kNBhAeG0nR=hnsHgiXpNv4LZ0BndnyYHzcM=4Pg@mail.gmail.com>
	<CAHw=FdsWaSXaKG=JwbUnAfSAJzKYe2biUWQT9Yis2CczbtYkrg@mail.gmail.com>
Message-ID: <CAGSpibmqhdjV0g81xWs6Gmg2=+-Zz5f-P80fv+0hkH9tkAxhgw@mail.gmail.com>

On Thu, Jun 21, 2012 at 8:27 PM, Heesob Park <phasis at gmail.com> wrote:
> Hi,
>
> 2012/6/22 Daniel Berger <djberg96 at gmail.com>
>>
>> Hi,
>>
>> I noticed that RubyInline did not work with Visual Studio (or mingw
>> for that matter). I forked RubyInline and made some patches that look
>> like they ought to work. Mostly it was just adjusting the build
>> command string so that linker options came after the -link option,
>> though I also changed the default extension to .dll and I make a
>> Dir.chdir call before the build happens so that it doesn't build in
>> the current directory.
>>
>> https://github.com/djberg96/rubyinline
>>
>> However, even after my patches I'm still having trouble. It builds,
>> but it won't load. I tried it with this simple bit of inline code:
>>
>> # factorial.rb
>> require "inline"
>> class MyTest
>> ?inline do |builder|
>> ? ?builder.c "
>> ? ? ?long factorial(int max) {
>> ? ? ? ?int i=max, result=1;
>> ? ? ? ?while (i >= 2) { result *= i--; }
>> ? ? ? ?return result;
>> ? ? ?}"
>> ?end
>> end
>>
>> t = MyTest.new()
>> p t.factorial(5)
>>
>> I can see that it generated the .c code and built the .dll file in
>> c:\Users\djberge\.ruby_inline\ruby-1.9.1
>>
>> Inline_MyTest_cb89593d1f9fe3ecdb2178215eee80ae.c
>> Inline_MyTest_cb89593d1f9fe3ecdb2178215eee80ae.dll
>>
>> However, it fails trying to load the file:
>>
>> c:/usr/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb:36:in
>> `require': cannot load such file --
>>
>> C:/Users/djberge/.ruby_inline/ruby-1.9.1/Inline_MyTest_cb89593d1f9fe3ecdb2178215eee80ae.dll
>> (LoadError)
>> ? ? ? ?from
>> c:/usr/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb:36:in
>> `require'
>> ? ? ? ?from
>> c:/usr/lib/ruby/gems/1.9.1/gems/RubyInline-3.11.2/lib/inline.rb:522:in
>> `load'
>> ? ? ? ?from
>> c:/usr/lib/ruby/gems/1.9.1/gems/RubyInline-3.11.2/lib/inline.rb:859:in
>> `inline'
>> ? ? ? ?from factorial.rb:3:in `<class:MyTest>'
>> ? ? ? ?from factorial.rb:2:in `<main>'
>>
>> What have I done wrong?
>>
>
> Ruby wants not .dll but .so for the extension name.
>
> Here is a patch for your forked Rubyinline.
>
> diff --git a/inline.rb b/inline.rb.new
> index 3ac2f41..de45ee1 100644
> --- a/inline.rb
> +++ b/inline.rb.new
> @@ -377,11 +377,7 @@ module Inline
>
> ? ? ?def so_name
> ? ? ? ?unless defined? @so_name then
> - ? ? ? ?if WINDOZE
> - ? ? ? ? ?@so_name = "#{Inline.directory}/#{module_name}.dll"
> - ? ? ? ?else
> ? ? ? ? ? ?@so_name =
> "#{Inline.directory}/#{module_name}.#{RbConfig::CONFIG["DLEXT"]}"
> - ? ? ? ?end
> ? ? ? ?end
> ? ? ? ?@so_name
> ? ? ?end
> @@ -571,7 +567,7 @@ VALUE #{method}_equals(VALUE value) {
> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?nil
> ? ? ? ? ? ? ? ? ? ? ? ? ? ?end
>
> - ? ? ? ? ?if WINDOZE
> + ? ? ? ? ?if WINDOZE && RUBY_PLATFORM =~ /mswin/
> ? ? ? ? ? ? ?cmd = [ RbConfig::CONFIG['LDSHARED'],
> ? ? ? ? ? ? ? ? ? ? ?flags,
> ? ? ? ? ? ? ? ? ? ? ?RbConfig::CONFIG['CFLAGS'],
> @@ -657,7 +653,7 @@ VALUE #{method}_equals(VALUE value) {
> ? ? ? ?# gawd windoze land sucks
> ? ? ? ?case RUBY_PLATFORM
> ? ? ? ?when /mswin32/ then
> - ? ? ? ?" -link /LIBPATH:\"#{RbConfig::CONFIG['libdir']}\"
> /DEFAULTLIB:\"#{RbConfig::CONFIG['LIBRUBY']}\" /INCREMENTAL:no
> /EXPORT:Init_#{module_name}"
> + ? ? ? ?" -link /OUT:\"#{self.so_name}\"
> /LIBPATH:\"#{RbConfig::CONFIG['libdir']}\"
> /DEFAULTLIB:\"#{RbConfig::CONFIG['LIBRUBY']}\" /INCREMENTAL:no
> /EXPORT:Init_#{module_name}"
> ? ? ? ?when /mingw32/ then
> ? ? ? ? ?c = RbConfig::CONFIG
> ? ? ? ? ?" -Wl,--enable-auto-import -L#{c['libdir']}
> -l#{c['RUBY_SO_NAME']}"
>
>
> Regards,
> Park Heesob

Thanks! The docs Ruby will try to load a .dll but it doesn't seem to
actually be the case.

Anyway, I have all the tests passing with VS but mingw has 6 failures,
though there are many more in the core version, so it's an
improvement!

Regards,

Dan

From djberg96 at gmail.com  Fri Jun 22 04:42:32 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Thu, 21 Jun 2012 22:42:32 -0600
Subject: [Win32utils-devel] Trying to get RubyInline to work with VS
In-Reply-To: <CAGSpibmqhdjV0g81xWs6Gmg2=+-Zz5f-P80fv+0hkH9tkAxhgw@mail.gmail.com>
References: <CAGSpib=VHw4kNBhAeG0nR=hnsHgiXpNv4LZ0BndnyYHzcM=4Pg@mail.gmail.com>
	<CAHw=FdsWaSXaKG=JwbUnAfSAJzKYe2biUWQT9Yis2CczbtYkrg@mail.gmail.com>
	<CAGSpibmqhdjV0g81xWs6Gmg2=+-Zz5f-P80fv+0hkH9tkAxhgw@mail.gmail.com>
Message-ID: <CAGSpibnW5MjrY1Jo0wtyBTZoUFtmZsc1x4=x9U9UsrpDZxmhyw@mail.gmail.com>

Hi,

<snip>

> Anyway, I have all the tests passing with VS but mingw has 6 failures,
> though there are many more in the core version, so it's an
> improvement!

It now passes with mingw, though it requires Devkit in your PATH. I'm
not sure how to detect the Devkit location from within a Ruby program,
however. Anyway, good enough for now.

Thanks again,

Dan

From djberg96 at gmail.com  Sat Jun 23 00:24:33 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Fri, 22 Jun 2012 18:24:33 -0600
Subject: [Win32utils-devel] Trying to get RubyInline to work with VS
In-Reply-To: <CAHw=FdsWaSXaKG=JwbUnAfSAJzKYe2biUWQT9Yis2CczbtYkrg@mail.gmail.com>
References: <CAGSpib=VHw4kNBhAeG0nR=hnsHgiXpNv4LZ0BndnyYHzcM=4Pg@mail.gmail.com>
	<CAHw=FdsWaSXaKG=JwbUnAfSAJzKYe2biUWQT9Yis2CczbtYkrg@mail.gmail.com>
Message-ID: <CAGSpib=tOHVjszckY8GAuZBZ-d8f+zHhg3k5m0=qYiakw=czZw@mail.gmail.com>

On Thu, Jun 21, 2012 at 8:27 PM, Heesob Park <phasis at gmail.com> wrote:
> Hi,
>
> 2012/6/22 Daniel Berger <djberg96 at gmail.com>
>>
>> Hi,
>>
>> I noticed that RubyInline did not work with Visual Studio (or mingw
>> for that matter). I forked RubyInline and made some patches that look
>> like they ought to work. Mostly it was just adjusting the build
>> command string so that linker options came after the -link option,
>> though I also changed the default extension to .dll and I make a
>> Dir.chdir call before the build happens so that it doesn't build in
>> the current directory.
>>
>> https://github.com/djberg96/rubyinline
>>
>> However, even after my patches I'm still having trouble. It builds,
>> but it won't load. I tried it with this simple bit of inline code:
>>
>> # factorial.rb
>> require "inline"
>> class MyTest
>> ?inline do |builder|
>> ? ?builder.c "
>> ? ? ?long factorial(int max) {
>> ? ? ? ?int i=max, result=1;
>> ? ? ? ?while (i >= 2) { result *= i--; }
>> ? ? ? ?return result;
>> ? ? ?}"
>> ?end
>> end
>>
>> t = MyTest.new()
>> p t.factorial(5)
>>
>> I can see that it generated the .c code and built the .dll file in
>> c:\Users\djberge\.ruby_inline\ruby-1.9.1
>>
>> Inline_MyTest_cb89593d1f9fe3ecdb2178215eee80ae.c
>> Inline_MyTest_cb89593d1f9fe3ecdb2178215eee80ae.dll
>>
>> However, it fails trying to load the file:
>>
>> c:/usr/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb:36:in
>> `require': cannot load such file --
>>
>> C:/Users/djberge/.ruby_inline/ruby-1.9.1/Inline_MyTest_cb89593d1f9fe3ecdb2178215eee80ae.dll
>> (LoadError)
>> ? ? ? ?from
>> c:/usr/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb:36:in
>> `require'
>> ? ? ? ?from
>> c:/usr/lib/ruby/gems/1.9.1/gems/RubyInline-3.11.2/lib/inline.rb:522:in
>> `load'
>> ? ? ? ?from
>> c:/usr/lib/ruby/gems/1.9.1/gems/RubyInline-3.11.2/lib/inline.rb:859:in
>> `inline'
>> ? ? ? ?from factorial.rb:3:in `<class:MyTest>'
>> ? ? ? ?from factorial.rb:2:in `<main>'
>>
>> What have I done wrong?
>>
>
> Ruby wants not .dll but .so for the extension name.
>
> Here is a patch for your forked Rubyinline.
>
> diff --git a/inline.rb b/inline.rb.new
> index 3ac2f41..de45ee1 100644
> --- a/inline.rb
> +++ b/inline.rb.new
> @@ -377,11 +377,7 @@ module Inline
>
> ? ? ?def so_name
> ? ? ? ?unless defined? @so_name then
> - ? ? ? ?if WINDOZE
> - ? ? ? ? ?@so_name = "#{Inline.directory}/#{module_name}.dll"
> - ? ? ? ?else
> ? ? ? ? ? ?@so_name =
> "#{Inline.directory}/#{module_name}.#{RbConfig::CONFIG["DLEXT"]}"
> - ? ? ? ?end
> ? ? ? ?end
> ? ? ? ?@so_name
> ? ? ?end
> @@ -571,7 +567,7 @@ VALUE #{method}_equals(VALUE value) {
> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?nil
> ? ? ? ? ? ? ? ? ? ? ? ? ? ?end
>
> - ? ? ? ? ?if WINDOZE
> + ? ? ? ? ?if WINDOZE && RUBY_PLATFORM =~ /mswin/
> ? ? ? ? ? ? ?cmd = [ RbConfig::CONFIG['LDSHARED'],
> ? ? ? ? ? ? ? ? ? ? ?flags,
> ? ? ? ? ? ? ? ? ? ? ?RbConfig::CONFIG['CFLAGS'],
> @@ -657,7 +653,7 @@ VALUE #{method}_equals(VALUE value) {
> ? ? ? ?# gawd windoze land sucks
> ? ? ? ?case RUBY_PLATFORM
> ? ? ? ?when /mswin32/ then
> - ? ? ? ?" -link /LIBPATH:\"#{RbConfig::CONFIG['libdir']}\"
> /DEFAULTLIB:\"#{RbConfig::CONFIG['LIBRUBY']}\" /INCREMENTAL:no
> /EXPORT:Init_#{module_name}"
> + ? ? ? ?" -link /OUT:\"#{self.so_name}\"
> /LIBPATH:\"#{RbConfig::CONFIG['libdir']}\"
> /DEFAULTLIB:\"#{RbConfig::CONFIG['LIBRUBY']}\" /INCREMENTAL:no
> /EXPORT:Init_#{module_name}"
> ? ? ? ?when /mingw32/ then
> ? ? ? ? ?c = RbConfig::CONFIG
> ? ? ? ? ?" -Wl,--enable-auto-import -L#{c['libdir']}
> -l#{c['RUBY_SO_NAME']}"

I don't suppose there's an easy way to have all of the compilation
happen in a different directory is there? I see options for individual
output files, but not one to say "build everything in directory X". Am
I missing it?

I only ask because Ryan doesn't like having to add a Dir.chdir call.

Regards,

Dan

From phasis at gmail.com  Sat Jun 23 03:59:19 2012
From: phasis at gmail.com (Heesob Park)
Date: Sat, 23 Jun 2012 12:59:19 +0900
Subject: [Win32utils-devel] Trying to get RubyInline to work with VS
In-Reply-To: <CAGSpib=tOHVjszckY8GAuZBZ-d8f+zHhg3k5m0=qYiakw=czZw@mail.gmail.com>
References: <CAGSpib=VHw4kNBhAeG0nR=hnsHgiXpNv4LZ0BndnyYHzcM=4Pg@mail.gmail.com>
	<CAHw=FdsWaSXaKG=JwbUnAfSAJzKYe2biUWQT9Yis2CczbtYkrg@mail.gmail.com>
	<CAGSpib=tOHVjszckY8GAuZBZ-d8f+zHhg3k5m0=qYiakw=czZw@mail.gmail.com>
Message-ID: <CAHw=FdtkUJb7v5=qsNJpU_iV8ha2c9O557_kHdj5gs674YxAqQ@mail.gmail.com>

Hi,

2012/6/23 Daniel Berger <djberg96 at gmail.com>

> On Thu, Jun 21, 2012 at 8:27 PM, Heesob Park <phasis at gmail.com> wrote:
> > Hi,
> >
> > 2012/6/22 Daniel Berger <djberg96 at gmail.com>
> >>
> >> Hi,
> >>
> >> I noticed that RubyInline did not work with Visual Studio (or mingw
> >> for that matter). I forked RubyInline and made some patches that look
> >> like they ought to work. Mostly it was just adjusting the build
> >> command string so that linker options came after the -link option,
> >> though I also changed the default extension to .dll and I make a
> >> Dir.chdir call before the build happens so that it doesn't build in
> >> the current directory.
> >>
> >> https://github.com/djberg96/rubyinline
> >>
> >> However, even after my patches I'm still having trouble. It builds,
> >> but it won't load. I tried it with this simple bit of inline code:
> >>
> >> # factorial.rb
> >> require "inline"
> >> class MyTest
> >>  inline do |builder|
> >>    builder.c "
> >>      long factorial(int max) {
> >>        int i=max, result=1;
> >>        while (i >= 2) { result *= i--; }
> >>        return result;
> >>      }"
> >>  end
> >> end
> >>
> >> t = MyTest.new()
> >> p t.factorial(5)
> >>
> >> I can see that it generated the .c code and built the .dll file in
> >> c:\Users\djberge\.ruby_inline\ruby-1.9.1
> >>
> >> Inline_MyTest_cb89593d1f9fe3ecdb2178215eee80ae.c
> >> Inline_MyTest_cb89593d1f9fe3ecdb2178215eee80ae.dll
> >>
> >> However, it fails trying to load the file:
> >>
> >> c:/usr/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb:36:in
> >> `require': cannot load such file --
> >>
> >>
> C:/Users/djberge/.ruby_inline/ruby-1.9.1/Inline_MyTest_cb89593d1f9fe3ecdb2178215eee80ae.dll
> >> (LoadError)
> >>        from
> >> c:/usr/lib/ruby/site_ruby/1.9.1/rubygems/custom_require.rb:36:in
> >> `require'
> >>        from
> >> c:/usr/lib/ruby/gems/1.9.1/gems/RubyInline-3.11.2/lib/inline.rb:522:in
> >> `load'
> >>        from
> >> c:/usr/lib/ruby/gems/1.9.1/gems/RubyInline-3.11.2/lib/inline.rb:859:in
> >> `inline'
> >>        from factorial.rb:3:in `<class:MyTest>'
> >>        from factorial.rb:2:in `<main>'
> >>
> >> What have I done wrong?
> >>
> >
> > Ruby wants not .dll but .so for the extension name.
> >
> > Here is a patch for your forked Rubyinline.
> >
> > diff --git a/inline.rb b/inline.rb.new
> > index 3ac2f41..de45ee1 100644
> > --- a/inline.rb
> > +++ b/inline.rb.new
> > @@ -377,11 +377,7 @@ module Inline
> >
> >      def so_name
> >        unless defined? @so_name then
> > -        if WINDOZE
> > -          @so_name = "#{Inline.directory}/#{module_name}.dll"
> > -        else
> >            @so_name =
> > "#{Inline.directory}/#{module_name}.#{RbConfig::CONFIG["DLEXT"]}"
> > -        end
> >        end
> >        @so_name
> >      end
> > @@ -571,7 +567,7 @@ VALUE #{method}_equals(VALUE value) {
> >                              nil
> >                            end
> >
> > -          if WINDOZE
> > +          if WINDOZE && RUBY_PLATFORM =~ /mswin/
> >              cmd = [ RbConfig::CONFIG['LDSHARED'],
> >                      flags,
> >                      RbConfig::CONFIG['CFLAGS'],
> > @@ -657,7 +653,7 @@ VALUE #{method}_equals(VALUE value) {
> >        # gawd windoze land sucks
> >        case RUBY_PLATFORM
> >        when /mswin32/ then
> > -        " -link /LIBPATH:\"#{RbConfig::CONFIG['libdir']}\"
> > /DEFAULTLIB:\"#{RbConfig::CONFIG['LIBRUBY']}\" /INCREMENTAL:no
> > /EXPORT:Init_#{module_name}"
> > +        " -link /OUT:\"#{self.so_name}\"
> > /LIBPATH:\"#{RbConfig::CONFIG['libdir']}\"
> > /DEFAULTLIB:\"#{RbConfig::CONFIG['LIBRUBY']}\" /INCREMENTAL:no
> > /EXPORT:Init_#{module_name}"
> >        when /mingw32/ then
> >          c = RbConfig::CONFIG
> >          " -Wl,--enable-auto-import -L#{c['libdir']}
> > -l#{c['RUBY_SO_NAME']}"
>
> I don't suppose there's an easy way to have all of the compilation
> happen in a different directory is there? I see options for individual
> output files, but not one to say "build everything in directory X". Am
> I missing it?
>
>  As far as I know, there is no shortcut to set whole building directory
once.

I think the following is the second best way.

" /Fe\"#{Inline.directory}/\" /Fd\"#{Inline.directory}/\"
/Fo\"#{Inline.directory}/\" -link  /OUT:\"#{self.so_name}\"
/LIBPATH:\"#{RbConfig::CONFIG['libdir']}\"
/DEFAULTLIB:\"#{RbConfig::CONFIG['LIBRUBY']}\" /INCREMENTAL:no
/EXPORT:Init_#{module_name}"


Regards,
Park Heesob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/win32utils-devel/attachments/20120623/94310be5/attachment-0001.html>