[FFmpeg-devel] [PATCH 3/3] Make opt_default(), return an error code rather than exit from the program.

Michael Niedermayer michaelni
Sun Nov 7 16:18:10 CET 2010


On Sun, Nov 07, 2010 at 10:42:12AM +0100, Stefano Sabatini wrote:
> On date Sunday 2010-11-07 01:04:49 +0100, Michael Niedermayer encoded:
> > On Sun, Nov 07, 2010 at 12:53:35AM +0100, Stefano Sabatini wrote:
> > > On date Sunday 2010-11-07 00:38:48 +0100, Michael Niedermayer encoded:
> > > > On Sat, Nov 06, 2010 at 03:02:57PM +0100, Stefano Sabatini wrote:
> > > > > This allows the program to nicely handle the error, for example
> > > > > providing detailed error explanation to the user or cleaning up the
> > > > > allocated resources of the program.
> > > > 
> > > > there is plenty of code that does not check its return value
> > > > and i dont think adding checks everywhere is a good idea
> > > 
> > > In general returning meaningful values *is* a good idea, but I'm
> > > not going to add them to all the FFmpeg codebase, but *this patch* is
> > > required by the previous patches in this series, to make ffmpeg prints
> > > something useful after opt_default() is called, rather than just
> > > abort.
> > 
> > grep for opt_default and you will see there are alot of cases where the
> > return value is not checked and not exiting anymore will lead to actual
> > bugs
> 
> All but one of those uses are in opt_target() where the value is fixed

until someone changes it ...
anyway, i dont strongly care about the return vs exit but i strongly object
to adding 20 checks in all uses of opt_default()

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Opposition brings concord. Out of discord comes the fairest harmony.
-- Heraclitus
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20101107/bd1cd7e0/attachment.pgp>



More information about the ffmpeg-devel mailing list