[Ffmpeg-devel] Re: [PATCH] Fix compilation when using --disable-opts

[Måns Rullgård]

Panagiotis Issaris said:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi Ramiro,
>
> Ramiro Polla schreef:
>> Panagiotis Issaris escreveu:
>>> Ramiro Polla schreef:
>>>> M?ns Rullg?rd escreveu:
>>>>> Panagiotis Issaris said:
>>>>>> M?ns Rullg?rd schreef:
>>>>>>> Diego Biurrun said:
>>>>>>>> On Thu, Mar 15, 2007 at 05:58:35PM +0100, Panagiotis Issaris wrote:
>>>>>>>>
>>>>>>>>> The attached patch fixes compilation when using the --disable-opts
>>>>>>>>> parameter. This to help those interested in using a debugger to
>>>>>>>>> debug
>>>>>>>>> FFmpeg.
> [...]
>>>> On Windows XP, crashes issue a dialog box (and so configure has to wait). Of
>>>> course
>>>> you won't decide if it's ok to crash a program on configure just because
>>>> of Windows users, but it would be nice to just put a warning before
>>>> running the tests: "This configure script might crash and generate a
>>>> core dump" (or something similar).
>>>>
>>> I have to confess that I have no idea on how to solve this on Windows
>>> :-( But, note that without the patch --disable-opts would just fail to
>>> compile, but now although apparantly while showing a dialog box, it will
>>> at least compile...
>>>
>>
>> Don't worry. I'm sure making the configure script not crash on Windows
>> is not a high priority for FFmpeg =).
> The configure script itself wont crash, right? The generated and
> executed testapp will crash but the configure script will keep on
> running after displaying some kind of warning? Or does it really end
> with the crash of the testapp?

The script should keep running.

>> A conditional warning is enough.
> During the execution of the configure script or on "configure --help"? A
> generic warning for all operating systems, or one specific for Windows?

On a Unix system I'd just use ulimit to disable core dumps and not worry
further about it.  The problem with windows is that it displays a monolog
box whenever an app crashes.  This might be perceived by a user as the
configure script crashing.  Hence printing message stating otherwise
prior to running the test on windows might be in order.

-- 
M?ns Rullg?rd
mans at mansr.com