[Geany] [patch]win32_spawn, GError, status code timeout

[Pierre Joye]

Hi Enrico,

On Wed, Feb 27, 2008 at 5:06 PM, Enrico Tröger <enrico.troeger at uvena.de> wrote:
> On Tue, 26 Feb 2008 20:10:24 +0100, "Pierre Joye"
>  <pierre.php at gmail.com> wrote:
>
>  Hi.
>
>  > Here is another patch to improve a little bit the win32_spawn
>  Thanks a lot, Pierre.
>
>
>  > functions. It adds:
>  >
>  > - GError support for the win32 functions (thanks to Jeff for the link!)
>  I changed the usage a little bit to avoid memory leaks.
>  g_win32_error_message() returns a newly-allocated buffer which needs to
>  be freed. But not a big deal at all.

Thanks for this fix!

We need valgrind on windows or a VC build (there is some tools to work
with VC binaries :).

>  > The timeout is set to 30 secondbut it is not really required right
>  > now, the default behavior being to wait until that the process exists.
>  > The upcoming async support will use it more intensively (still some
>  > work, don't hold your breath :).
>  Hehe, no need to hurry. It's done when it's done.
>
>  Attached is a patch which changes the used function for spawning
>  commands in build.c for View DVI/PDF file and for the general Run
>  command (e.g. execute a compiled C file or run a python script).
>  But it shows that there must be a little bug whether in your code or in
>  the code in build.c. I get these messages on the console:
>  CreateChildProcess: CreateProcess failed
>  win32_spawn: Create process failed
>  utils_spawn_async() failed: Das System kann die angegebene Datei nicht
>  finden.
>
>  (the German message at the end means: the system could not find the
>  specified file)

thanks :)

btw, I can read/speak German (write too but it is painful/slower ;).

>  The created "cmdline" string looked good:
>  C:\WINNT\System32\cmd.exe /Q /C G:\geany_run_script.bat
>
>  (geany_run_script.bat actually existed and the above command line was
>  executed as expected in a usually opened cmd.exe)
>
>  I didn't look depper into this problem. Maybe you'll find the problem.

I will take a quick look asap. If it is easily fixable, fine, if not I
rather prefer to focus on the real async  + io_add_watch
implementation.

Off topic, where are you located in Germany? I'm wondering if we can
have a drink, spring is coming early this year ;-)

Cheers,
-- 
Pierre
http://blog.thepimp.net | http://www.libgd.org