[PATCH] Don't call memset() unless output buffer is non-NULL, otherwise warn if it is NULL.

This corrects roundup issue #2386.
---
 libavcodec/h264_mp4toannexb_bsf.c |    6 +++++-
 1 files changed, 5 insertions(+), 1 deletions(-)

diff --git a/libavcodec/h264_mp4toannexb_bsf.c b/libavcodec/h264_mp4toannexb_bsf.c
index d4a7f31..f8fd107 100644
--- a/libavcodec/h264_mp4toannexb_bsf.c
+++ b/libavcodec/h264_mp4toannexb_bsf.c
 <at>  <at>  -114,7 +114,11  <at>  <at>  static int h264_mp4toannexb_filter(AVBitStreamFilterContext *bsfc,
                 unit_nb = *extradata++; /* number of pps unit(s) */
         }

-        memset(out + total_size, 0, FF_INPUT_BUFFER_PADDING_SIZE);
+        if(out)
+            memset(out + total_size, 0, FF_INPUT_BUFFER_PADDING_SIZE);
+        else
+            av_log(avctx, AV_LOG_WARNING, "Warning: input buffer not allocated.\n");
+
         av_free(avctx->extradata);
         avctx->extradata      = out;
         avctx->extradata_size = total_size;
--

-- 
1.7.4

Re: Another possible bug in dvbsub.c

Ronen Mizrahi skrev 2011-02-28 18:31:
>>
>>
>>> -    for (region_id = 0; region_id<  h->num_rects; region_id++) {
>>> -        *q++ = region_id;
>>> -        *q++ = 0xff; /* reserved */
>>> -        bytestream_put_be16(&q, h->rects[region_id]->x); /* left pos */
>>> -        bytestream_put_be16(&q, h->rects[region_id]->y); /* top pos */
>>> +    if (!s->hide_state) {
>>> +        for (region_id = 0; region_id<  h->num_rects; region_id++) {
>>> +            *q++ = region_id;
>>> +            *q++ = 0xff; /* reserved */
>>> +            bytestream_put_be16(&q, h->rects[region_id]->x); /* left pos
>>> */
>>> +            bytestream_put_be16(&q, h->rects[region_id]->y); /* top pos
>>> */
>>> +        }
>>>      }
>>>
>>
>> Indent should be done in a separate patch. That way it's easier so see that
>> this one doesn't change the behavior of the loop.
>>
>
> There is no indentation for the sake of indentation, an if statement is now
> enclosing the loop and that is why it is indented.

Yes, but it makes it harder to figure out what actual functional changes 
were made to the code when browsing the git history or doing git blame. 
I'll simply point to what the website has to say:

Re: [rfc] merge the applehttp-urlprotocol into main

Tomas Härdin <tomas.hardin <at> codemill.se> writes:

> >
> > will be parsed as applehttp://http://xx.yy.zz.ww/abc/
> > while file urls can be specified explicitly, as in
> > applehttp://file://usr/files/myfile.m3u8
> >              ----------------------------
> >              inner url part
> 
> You can't nest URIs like that - please read RFC 3986 ( 
> http://www.ietf.org/rfc/rfc3986.txt ).
> 
> A better solution IMO would be to go the svn+ssh route - a plus sign. So 
> applehttp+file:/foo/bar/playlist.m3u8 for instance.
> 

You can the way we do it in xbmc. protocol://[url encoded inner 
protocol]/subparts. This is how we handle files in rars or files in rars in zips 
in rars for example.

You just have to make sure you don't url decode pointlessly.

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel <at> mplayerhq.hu
https://lists.mplayerhq.hu/mailman/listinfo/ffmpeg-devel

Re: [PATCH] Non-blocking HTTP

Le decadi 30 pluviôse, an CCXIX, Martin Storsjö a écrit :
> I guess you've done extensive tests with handling EAGAIN in inconvenient 
> places, e.g. halfway through chunk size and such. The code looks like it's 
> originally quite well-prepared for that situation at least.

I thought I did, but it seems it was not enough: this is broken for chunked
uploads. I can make it work, at the cost of some code complexity; but for
now, I withdraw "4/4 HTTP: handle non-blocking".

"3/4 avio: add url_set_nonblock" is suspended pending the great rename of
url_* functions (although I notice that some functions already have the av_
prefix, av_url_set_nonblock would not be out of place).

On the other hand, "1/4 URLContext: introduce a slave URLContext" and
"2/4 HTTP: use URLContext.slave" could still be applied soon I believe.

Regards,

--

-- 
  Nicolas George

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel <at> mplayerhq.hu
https://lists.mplayerhq.hu/mailman/listinfo/ffmpeg-devel

Jittery sound using ffplay

Hi,
    Iam using Qemu 0.13.0 and ffmpeg 0.6.1.to play audio.
whenever Iam using ffplay to play any wave or mp3 file...the audio is having
alot of glitches and jittery sound.but the same file is playing proper while
using aplay...donno whats going..any clue???
Plz let me know as soon as possible.

--Thanx
AK

Re: Jittery sound using ffplay

On 03/01/2011 06:44 AM, asim khan wrote:

> Hi,
>     Iam using Qemu 0.13.0 and ffmpeg 0.6.1.to play audio.
> whenever Iam using ffplay to play any wave or mp3 file...the audio is having
> alot of glitches and jittery sound.but the same file is playing proper while
> using aplay...donno whats going..any clue???
> Plz let me know as soon as possible.

SDL issue possibly?  I've had audio issues with SDL in the past.  Does
ffmpeg with "-f oss" or "-f alsa" work ok?

-Justin

Re: Jittery sound using ffplay

On 03/01/2011 12:44 PM, asim khan wrote:
> Hi,
>     Iam using Qemu 0.13.0 and ffmpeg 0.6.1.to play audio.
> whenever Iam using ffplay to play any wave or mp3 file...the audio is having
> alot of glitches and jittery sound.but the same file is playing proper while
> using aplay...donno whats going..any clue???
> Plz let me know as soon as possible.
>
>
> --Thanx
> AK

Here's a clue. Wrong mailing list for these kind of questions. Try the
users mailing list.

MvH
Benjamin Larsson

[PATCH] source snapshots back in action

$subject, I restored source snapshot generation on mphq.

Diego

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel <at> mplayerhq.hu
https://lists.mplayerhq.hu/mailman/listinfo/ffmpeg-devel

[PATCH] aac_latm_dec: use aac context and aac m4ac

When decoding latm config, use the corresponding aac context and its
m4ac instead of using NULL and a local variable. This fixes decoding of
audio in MPEG TS from SBTVD (the Brazillian Digital TV Sytem), when
there is no extradata. This is the case when using the decoder with
gst-ffmpeg and a GStreamer mpegts demuxer.

Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo <at> holoscopio.com>
---
 libavcodec/aacdec.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/libavcodec/aacdec.c b/libavcodec/aacdec.c
index 81a94f2..6317e42 100644
--- a/libavcodec/aacdec.c
+++ b/libavcodec/aacdec.c
 <at>  <at>  -2248,7 +2248,6  <at>  <at>  static int latm_decode_audio_specific_config(struct LATMContext *latmctx,
                                              GetBitContext *gb)
 {
     AVCodecContext *avctx = latmctx->aac_ctx.avctx;
-    MPEG4AudioConfig m4ac;
     int  config_start_bit = get_bits_count(gb);
     int     bits_consumed, esize;

 <at>  <at>  -2258,7 +2257,8  <at>  <at>  static int latm_decode_audio_specific_config(struct LATMContext *latmctx,
         return AVERROR_INVALIDDATA;
     } else {
         bits_consumed =
-            decode_audio_specific_config(NULL, avctx, &m4ac,
+            decode_audio_specific_config(&latmctx->aac_ctx, avctx,
+                                         &latmctx->aac_ctx.m4ac,

Re: [PATCH] source snapshots back in action

Il giorno mar, 01/03/2011 alle 13.27 +0100, Diego Biurrun ha scritto:
> 
> +<p>
> +Instead of using Git to clone our repository you can also download
> our nightly
> +<a href="releases/ffmpeg-snapshot-git.tar.bz2">git snapshot</a>
> tarball to
> +bootstrap your local repository.
> +</p>
> +
> +<p>
> +Alternatively if you have no need for git metadata and/or prefer the
> smaller
> +download, get the bare
> +<a href="releases/ffmpeg-snapshot-git.tar.bz2">source snapshot</a>. 

Both the same file?

--

-- 
Diego Elio Pettenò — Flameeyes
http://blog.flameeyes.eu/

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel <at> mplayerhq.hu
https://lists.mplayerhq.hu/mailman/listinfo/ffmpeg-devel