[FFmpeg-devel] [PATCH/RFC] Constify src in av_fifo_generic_write()

Michael Niedermayer michaelni
Tue Feb 23 15:55:12 CET 2010


On Tue, Feb 23, 2010 at 01:41:49PM +0100, Tomas H?rdin wrote:
> On Tue, 2010-02-23 at 13:14 +0100, Michael Niedermayer wrote:
> > On Tue, Feb 23, 2010 at 10:50:03AM +0100, Tomas H?rdin wrote:
> > > Good morning
> > > 
> > > This is a rather simple patch that merely makes av_fifo_generic_write()
> > > take const void *src instead of just void *src. It causes no warnings
> > > and regression tests pass.
> > > 
> > > The reason is simple: guarantee the user that the function does not
> > > modify src.
> > > 
> > > One small issue though: this necessitated changing func to int(*)(const
> > > void*,void*,int), which is OK for all code within FFmpeg. However, this
> > > seems to count as a minor API change since g++ (and gcc I presume) won't
> > > allow implicitly casting int(*)(void*,void*,int) to int(*)(const
> > > void*,void*,int). I presume that part requires a bit of discussion,
> > > which is why I marked this thread "RFC".
> > > 
> > > Patch attached.
> > 
> > ive added a comment explaining why src cannot be const as you are at
> > least the 2nd person trying to introduce this bug
> 
> Ah, ok. Might I suggest that you put a comment in fifo.h? That way users
> can see it too (especially if only the headers are installed), and there
> is less chance anyone else gets confused about this :o

EPATCHWELCOME

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

The real ebay dictionary, page 3
"Rare item" - "Common item with rare defect or maybe just a lie"
"Professional" - "'Toy' made in china, not functional except as doorstop"
"Experts will know" - "The seller hopes you are not an expert"
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20100223/d930ea69/attachment.pgp>



More information about the ffmpeg-devel mailing list