[FFmpeg-devel] [FFmpeg-commits] lavf: replace all uses of url_fskip with avio_seek
Peter Ross
pross
Wed Mar 2 16:08:19 CET 2011
On Tue, Mar 01, 2011 at 06:32:10PM +0100, Anton Khirnov wrote:
> Module: ffmpeg
> Branch: master
> Commit: e356fc57a2e9887370caec58d8aafeafd1f336dc
>
> Author: Anton Khirnov <anton at khirnov.net>
> Committer: Ronald S. Bultje <rsbultje at gmail.com>
> Date: Mon Feb 28 14:57:55 2011 +0100
>
> lavf: replace all uses of url_fskip with avio_seek
>
> Signed-off-by: Ronald S. Bultje <rsbultje at gmail.com>
>
> ---
[..]
> libavformat/vocdec.c | 8 +++---
> libavformat/vqf.c | 6 ++--
> libavformat/wav.c | 6 ++--
> libavformat/wtv.c | 58 +++++++++++++++++++++---------------------
Guys, the patch subject indicates what was done, but not why.
Can somebody explain WHY the fskip convention has been replaced
with seek. Im ignoring the original patch thread because that
has gone political.
This function has existed since I started using FFmpeg a decade ago.
In all my demuxers url_fskip has been used to skip a block of unknown
or irrelevant data. I don't 'seek' to another location in the file,
I 'skip' over the data. The intent of the code becomes less obvious
when fseek is used in both instances.
Fskip has never been a problem. What possible benefit does this
change give us? Objectively please!
By logical extension, the next patch in this series will be to
universally replace get_bits1(gb) with get_bits(gb, 1). Seriously,
why not?
I am not one normally to be critical of API and broader framework
changes, and dont have the time to read all mails, but this change,
and the get/put_tag renaming insanity really pinches a nerve. This
is a backward step on the master public API.
Its really late here -- expect an avio_skip() patch from me tomorrow.
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
-------------- 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/20110303/a040ce2f/attachment.pgp>
More information about the ffmpeg-devel
mailing list