from:"Gilles"

Re: [FFmpeg-devel] Fixing rotation metadata override

2018-10-15 Thread Gilles

Oh my bad, I had 2 tickets on this subject and the one that is still opened
is:

https://trac.ffmpeg.org/ticket/6370

Anyway, the issue is still around

Le lun. 15 oct. 2018 à 13:27, Carl Eugen Hoyos  a
écrit :

> 2018-10-15 12:11 GMT+02:00, Gilles :
>
> > the one in the first post: https://trac.ffmpeg.org/ticket/4560
>
> Again:
> This ticket was fixed years ago and is not reproducible with
> current FFmpeg git head.
>
> Please stop top-posting here, Carl Eugen
> ___
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] Fixing rotation metadata override

2018-10-15 Thread Gilles

Hi,

the one in the first post: https://trac.ffmpeg.org/ticket/4560

Le lun. 15 oct. 2018 à 11:40, Carl Eugen Hoyos  a
écrit :

> 2018-10-15 10:29 GMT+02:00, Gilles :
>
> > it looks like to me that the issue is still present.
>
> Which issue? Please fix the commit message accordingly.
>
> Please do not top-post here, Carl Eugen
> ___
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] Fixing rotation metadata override

2018-10-15 Thread Gilles

0] 8x8 transform intra:53.6% inter:86.2%
[libx264 @ 0x3cbea80] coded y,uvDC,uvAC intra: 11.6% 21.3% 1.3% inter: 7.6%
11.0% 0.2%
[libx264 @ 0x3cbea80] i16 v,h,dc,p: 41% 24% 11% 23%
[libx264 @ 0x3cbea80] i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 24% 12% 56%  1%  1%
2%  1%  1%  1%
[libx264 @ 0x3cbea80] i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 38% 20% 18%  3%  4%
6%  4%  5%  3%
[libx264 @ 0x3cbea80] i8c dc,h,v,p: 64% 17% 18%  1%
[libx264 @ 0x3cbea80] Weighted P-Frames: Y:0.0% UV:0.0%
[libx264 @ 0x3cbea80] ref P L0: 69.8% 13.2% 13.8%  3.2%
[libx264 @ 0x3cbea80] ref B L0: 91.0%  9.0%
[libx264 @ 0x3cbea80] kb/s:5739.08

--

The resulting file still has a 90degres rotation:
mediainfo out.mp4 | grep -i rot
Rotation : 90?


Le sam. 13 oct. 2018 à 13:41, Carl Eugen Hoyos  a
écrit :

> 2018-10-13 8:31 GMT+02:00, Gilles :
>
> > here is a fix for issue: https://trac.ffmpeg.org/ticket/4560
>
> This ticket was fixed years ago and the issue is not
> reproducible with current FFmpeg git head.
>
> > Background: since FFmpeg 3.3, it is not possible to force
> > the rotation metadata to 0, when video is being converted.
> > It only work in -c:v copy mode.
>
> Sounds like ticket #6370.
>
> Carl Eugen
> ___
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

[FFmpeg-devel] Fixing rotation metadata override

2018-10-13 Thread Gilles

Hi,

here is a fix for issue: https://trac.ffmpeg.org/ticket/4560
Background: since FFmpeg 3.3, it is not possible to force the rotation
metadata to 0, when video is being converted. It only work in -c:v copy
mode.

Can you please have a look at it ?

Thanks


0001-Fix-rotation-metadata-override-when-transcoding.patch
Description: Binary data
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 3/3] avformat: add youtube-dl based demuxer

2015-04-10 Thread Gilles Chanteperdrix

On Thu, Apr 09, 2015 at 10:59:24PM +0200, Lukasz Marek wrote:
 On 09.04.2015 13:50, Kieran Kunhya wrote:
 On 9 April 2015 at 12:17, Rodger Combs rodger.co...@gmail.com wrote:
 Agreed, this belongs in a higher layer. I think it'd be reasonable for 
 FFmpeg to have a higher-layer library handling things like playlists and 
 this, but that's another conversation.
 
 +1
 
 So instead of complaining and spamming with +1, prepare a patch and submit
 it? Community will discuss it and do what's the best for the project (and
 yes, project is not single person who has their wishes). Personally I woulnt
 complain about that separation if done properly, but just don't see a reason
 for it. Most of the stuff you don't need/like you can disable at compilation
 level.
 We have separation libavformat/libavdevice and most of patches I submitted
 regarding libavdevice were complained by one person.
 
 Sometimes I have a feeling that ffmpeg is an open source project to feed
 wm4's project. Everything he doesn't need is retarded, rotten, dumb,
 etc, etc.
 
 I don't want to be drama queen here, but the way author of this patch was
 treated is bellow any level of dignity.

I assure you the author of this patch is not vexed or discouraged.
My main interest in using ffmpeg is an an RTSP client library (ok,
maybe some say this should not be a demuxer either ?), the
youtube-dl thing would just have been a nice addition if it had been
possible to have a quick solution, but this is nothing I care to
spend to much time on. My main effort for free-software go to
another project which has less maintainers, contributors and users
than ffmpeg.

Now for the high level API, if you mind the opinion of someone
which has a rather new view on this project and is implementing a
simplistic player, I believe what could be done relatively easy is
to split ffplay into say libffplay which would implement a
high-level API for players with calls such as open, play, pause,
seek, close, and ffplay which would use this API and implement the
human machine interface. This would make implementing a custom
player relatively easy by using libffplay being free to use GTK or
Qt, or whatever else for the GUI. Maybe the same goes for the ffmpeg
tool.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 3/3] avformat: add youtube-dl based demuxer

2015-04-09 Thread Gilles Chanteperdrix

On Wed, Apr 08, 2015 at 10:22:42PM +0200, Nicolas George wrote:
 Le nonidi 19 germinal, an CCXXIII, Gilles Chanteperdrix a écrit :
  So, when you just need to run exec, it is stupid 
 
 Exactly what I wrote: not your problem / premature optimization.
 
  I do not follow you here. youtube-dl -J has a uselessly verbose
  output, much harder to parse than youtube-dl -g which gives you just
  one URL. But you still need to run youtube-dl. So, I do not see the
  upside.
 
 Scratch uselessly and you will get it right: -J gives you everything you
 may need to download the video correctly. -g only gives you one piece of
 information, which is sometimes sufficient (but not not enough that you do
 not feel the need to call a second time for -e) but only for a few sites.
 
 So here is the upside of -J: it works.
 
 Second upside: it works with a single call.
 
 Third upside: since the information is completely self-contained, you can
 pass it around easily. People here whine about ffmpeg -f youtubedl, and
 they have valid arguments. This:
 
 youtube-dl -J $url | ffmpeg -f youtubedl -i -
 
 is just as easy, and nobody will whine.

Well, in that case, 
ffmpeg -i `youtube-dl -f bestvideo -g $url`

Is even simpler.

-- 
Gilles.


pgpbhFxbUoSjt.pgp
Description: PGP signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 3/3] avformat: add youtube-dl based demuxer

2015-04-09 Thread Gilles Chanteperdrix

On Thu, Apr 09, 2015 at 01:08:21PM +0200, wm4 wrote:
 On Thu, 9 Apr 2015 00:02:50 +0200
 Lukasz Marek lukasz.m.lu...@gmail.com wrote:
 
  W dniu środa, 8 kwietnia 2015 Gilles Chanteperdrix 
  gilles.chanteperd...@xenomai.org
  
  And forgot to add my apologies for these bad patches and the noise
   that followed.
  
  
  No need for that.
  
  Your demuxer seems to use public API only so you can try to move this
  implementation to your project and register this demuxer from your
  application.
 
 I'm pretty sure implementing a demuxer requires using private API.
 
 What's more, this stuff doesn't make sense as a demuxer.

I am not sure whether it makes sense, but implementing a fallback in
the application which spawns youtube-dl then retry when
avformat_open_input fails for an http/https URL is in fact simpler
than implementing a demuxer. And it avoids the noticeable added
latency which results from running youtube-dl (or libquvi for that
matter) for every http URL directly supported by ffmpeg.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 3/3] avformat: add youtube-dl based demuxer

2015-04-09 Thread Gilles Chanteperdrix

On Thu, Apr 09, 2015 at 03:37:30PM +0200, Nicolas George wrote:
 Le decadi 20 germinal, an CCXXIII, Gilles Chanteperdrix a écrit :
  Well, in that case, 
  ffmpeg -i `youtube-dl -f bestvideo -g $url`
 
 That could work too, but you still fail to acknowledge that -g is frequently
 not enough.

And -J is to much: it gives you all the available videos and forces
you to reimplement youtube-dl -f option to choose only one from the
list. In what case is -g not sufficient, for handling play lists ?

-- 
Gilles.


pgpzg4vkDQlzD.pgp
Description: PGP signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

[FFmpeg-devel] Segmentation fault in libquvi.c

2015-04-08 Thread Gilles Chanteperdrix

Hi, 

I just triend libquvi, and get a segmentation fault in the
libquvi_read_header function, because ff_copy_whitelists is called
before qc-fmtctx is allocated by avformat_open_input. I added a
call to avformat_alloc_context() before ff_copy_whitelists and the
libquvi demuxer works. 

However, I wonder how to fix this properly: the error handling
labels look backward, so that I am not sure where to free the
allocated context in case of error.

Regards.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] Segmentation fault in libquvi.c

2015-04-08 Thread Gilles Chanteperdrix

On Wed, Apr 08, 2015 at 11:02:47AM +0200, wm4 wrote:
 On Wed, 8 Apr 2015 10:55:45 +0200
 Gilles Chanteperdrix gilles.chanteperd...@xenomai.org wrote:
 
  Hi, 
  
  I just triend libquvi, and get a segmentation fault in the
  libquvi_read_header function, because ff_copy_whitelists is called
  before qc-fmtctx is allocated by avformat_open_input. I added a
  call to avformat_alloc_context() before ff_copy_whitelists and the
  libquvi demuxer works. 
  
  However, I wonder how to fix this properly: the error handling
  labels look backward, so that I am not sure where to free the
  allocated context in case of error.
  
  Regards.
  
 
 It's probably best to remove libquvi. It's dead.

Indeed, no activity since 2013. Is there any other ffmpeg demuxer
providing the same functionality ?

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] Segmentation fault in libquvi.c

2015-04-08 Thread Gilles Chanteperdrix

On Wed, Apr 08, 2015 at 05:29:13PM +0200, wm4 wrote:
 On Wed, 8 Apr 2015 17:22:58 +0200
 Gilles Chanteperdrix gilles.chanteperd...@xenomai.org wrote:
 
  On Wed, Apr 08, 2015 at 04:13:58PM +0200, wm4 wrote:
   On Wed, 8 Apr 2015 15:37:03 +0200
   Gilles Chanteperdrix gilles.chanteperd...@xenomai.org wrote:
   
On Wed, Apr 08, 2015 at 02:40:52PM +0200, Michael Niedermayer wrote:
 On Wed, Apr 08, 2015 at 10:55:45AM +0200, Gilles Chanteperdrix wrote:
  Hi, 
  
  I just triend libquvi, and get a segmentation fault in the
  libquvi_read_header function, because ff_copy_whitelists is called
  before qc-fmtctx is allocated by avformat_open_input. I added a
  call to avformat_alloc_context() before ff_copy_whitelists and the
  libquvi demuxer works.
 
  However, I wonder how to fix this properly: the error handling
  labels look backward, so that I am not sure where to free the
  allocated context in case of error.
 
 applied this, yes its correct

Ok, there are other details missing, the stream does not get a
duration, start_time and bitrate. This can easily be fixed, but as
wm4 said libquvi seems an abandoned project.

Would there be any interest in a solution based on youtube-dl? It
seems to be the standard, these days. I just ran a few tests:

youtube -qs 'url' returns 0 or 1 depending on whether the url can be
parsed by the tool

youtube -e 'url' prints the stream title

and youtube -f 'url' prints the video url.

To use this portably, we can use system() and redirect the
output to a temporary file, and read the title or URL from the file.
Or is popen available on all platforms where ffmpeg runs?

Is it a clean enough solution? If yes, I can submit the patch adding
this solution. From what I could see, all solutions to parse youtube
(including quvi) are based on scripts anyway.

   
   Calling external processes from within libavformat doesn't sound like
   the right thing to do. You can do it perfectly fine in a higher level,
   of course (I'm doing this too).
  
  The point is: the youtube-dl developers are doing a terrific job
  maintaining their software and supporting a lot of sites: 
  run youtube-dl --list-extractors to be convinced. So, I do not think
  it makes sense to duplicate this effort at a higher level, this is
  a waste of ressources. Maybe ask them kindly to provide us with some
  command line arguments that would arrange us would make sense,
  though.
  
 
 I'm not talking about duplicating this. I'm saying that you should put
 calling youtube-dl not into libavformat, but into the application which
 uses libavformat.

So, everyone should duplicate the job of using youtube-dl instead of
having it done once and for all in a common library ?

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] Segmentation fault in libquvi.c

2015-04-08 Thread Gilles Chanteperdrix

On Wed, Apr 08, 2015 at 02:40:52PM +0200, Michael Niedermayer wrote:
 On Wed, Apr 08, 2015 at 10:55:45AM +0200, Gilles Chanteperdrix wrote:
  Hi, 
  
  I just triend libquvi, and get a segmentation fault in the
  libquvi_read_header function, because ff_copy_whitelists is called
  before qc-fmtctx is allocated by avformat_open_input. I added a
  call to avformat_alloc_context() before ff_copy_whitelists and the
  libquvi demuxer works.
 
  However, I wonder how to fix this properly: the error handling
  labels look backward, so that I am not sure where to free the
  allocated context in case of error.
 
 applied this, yes its correct

Ok, there are other details missing, the stream does not get a
duration, start_time and bitrate. This can easily be fixed, but as
wm4 said libquvi seems an abandoned project.

Would there be any interest in a solution based on youtube-dl? It
seems to be the standard, these days. I just ran a few tests:

youtube -qs 'url' returns 0 or 1 depending on whether the url can be
parsed by the tool

youtube -e 'url' prints the stream title

and youtube -f 'url' prints the video url.

To use this portably, we can use system() and redirect the
output to a temporary file, and read the title or URL from the file.
Or is popen available on all platforms where ffmpeg runs?

Is it a clean enough solution? If yes, I can submit the patch adding
this solution. From what I could see, all solutions to parse youtube
(including quvi) are based on scripts anyway.

-- 
Gilles.


pgpuRmdT_1gda.pgp
Description: PGP signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] Segmentation fault in libquvi.c

2015-04-08 Thread Gilles Chanteperdrix

On Wed, Apr 08, 2015 at 03:51:36PM +0200, Lukasz Marek wrote:
 On 8 April 2015 at 15:37, Gilles Chanteperdrix 
 gilles.chanteperd...@xenomai.org wrote:
 
  On Wed, Apr 08, 2015 at 02:40:52PM +0200, Michael Niedermayer wrote:
   On Wed, Apr 08, 2015 at 10:55:45AM +0200, Gilles Chanteperdrix wrote:
Hi,
   
I just triend libquvi, and get a segmentation fault in the
libquvi_read_header function, because ff_copy_whitelists is called
before qc-fmtctx is allocated by avformat_open_input. I added a
call to avformat_alloc_context() before ff_copy_whitelists and the
libquvi demuxer works.
   
However, I wonder how to fix this properly: the error handling
labels look backward, so that I am not sure where to free the
allocated context in case of error.
  
   applied this, yes its correct
 
  Ok, there are other details missing, the stream does not get a
  duration, start_time and bitrate. This can easily be fixed, but as
  wm4 said libquvi seems an abandoned project.
 
  Would there be any interest in a solution based on youtube-dl? It
  seems to be the standard, these days. I just ran a few tests:
 
  youtube -qs 'url' returns 0 or 1 depending on whether the url can be
  parsed by the tool
 
  youtube -e 'url' prints the stream title
 
  and youtube -f 'url' prints the video url.
 
  To use this portably, we can use system() and redirect the
  output to a temporary file, and read the title or URL from the file.
  Or is popen available on all platforms where ffmpeg runs?
 
  Is it a clean enough solution? If yes, I can submit the patch adding
  this solution. From what I could see, all solutions to parse youtube
  (including quvi) are based on scripts anyway.
 
 
 I submitted a patch more that year ago to support youtube natively, as I
 remember this was based on the script you mention, but it was rejected
 because we have libquvi. [sic!]
 Project not actively developed is not the same as project not working.
 In corner case it is not developed because everything is working
 perfectly.

libquvi seems to work on youtube yes, but not on daily motion and
other streaming sites I tried where youtube-dl works, it simply does
not work. The sites change, so the tool to parse them must be up to
date to work correctly. So no, libquvi does not work perfectly.

And libquvi ffmpeg demuxer does not seem to be in a very good shape
either, it simply segfaults, and even when the segfault is resolved,
has other issues, the extern AVFormatcontext fields are simply not
filled.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] Segmentation fault in libquvi.c

2015-04-08 Thread Gilles Chanteperdrix

On Wed, Apr 08, 2015 at 04:13:58PM +0200, wm4 wrote:
 On Wed, 8 Apr 2015 15:37:03 +0200
 Gilles Chanteperdrix gilles.chanteperd...@xenomai.org wrote:
 
  On Wed, Apr 08, 2015 at 02:40:52PM +0200, Michael Niedermayer wrote:
   On Wed, Apr 08, 2015 at 10:55:45AM +0200, Gilles Chanteperdrix wrote:
Hi, 

I just triend libquvi, and get a segmentation fault in the
libquvi_read_header function, because ff_copy_whitelists is called
before qc-fmtctx is allocated by avformat_open_input. I added a
call to avformat_alloc_context() before ff_copy_whitelists and the
libquvi demuxer works.
   
However, I wonder how to fix this properly: the error handling
labels look backward, so that I am not sure where to free the
allocated context in case of error.
   
   applied this, yes its correct
  
  Ok, there are other details missing, the stream does not get a
  duration, start_time and bitrate. This can easily be fixed, but as
  wm4 said libquvi seems an abandoned project.
  
  Would there be any interest in a solution based on youtube-dl? It
  seems to be the standard, these days. I just ran a few tests:
  
  youtube -qs 'url' returns 0 or 1 depending on whether the url can be
  parsed by the tool
  
  youtube -e 'url' prints the stream title
  
  and youtube -f 'url' prints the video url.
  
  To use this portably, we can use system() and redirect the
  output to a temporary file, and read the title or URL from the file.
  Or is popen available on all platforms where ffmpeg runs?
  
  Is it a clean enough solution? If yes, I can submit the patch adding
  this solution. From what I could see, all solutions to parse youtube
  (including quvi) are based on scripts anyway.
  
 
 Calling external processes from within libavformat doesn't sound like
 the right thing to do. You can do it perfectly fine in a higher level,
 of course (I'm doing this too).

The point is: the youtube-dl developers are doing a terrific job
maintaining their software and supporting a lot of sites: 
run youtube-dl --list-extractors to be convinced. So, I do not think
it makes sense to duplicate this effort at a higher level, this is
a waste of ressources. Maybe ask them kindly to provide us with some
command line arguments that would arrange us would make sense,
though.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] Segmentation fault in libquvi.c

2015-04-08 Thread Gilles Chanteperdrix

On Wed, Apr 08, 2015 at 04:26:20PM +0200, Michael Niedermayer wrote:
 On Wed, Apr 08, 2015 at 04:01:38PM +0200, Gilles Chanteperdrix wrote:
  On Wed, Apr 08, 2015 at 03:51:36PM +0200, Lukasz Marek wrote:
   On 8 April 2015 at 15:37, Gilles Chanteperdrix 
   gilles.chanteperd...@xenomai.org wrote:
   
On Wed, Apr 08, 2015 at 02:40:52PM +0200, Michael Niedermayer wrote:
 On Wed, Apr 08, 2015 at 10:55:45AM +0200, Gilles Chanteperdrix wrote:
  Hi,
 
  I just triend libquvi, and get a segmentation fault in the
  libquvi_read_header function, because ff_copy_whitelists is called
  before qc-fmtctx is allocated by avformat_open_input. I added a
  call to avformat_alloc_context() before ff_copy_whitelists and the
  libquvi demuxer works.
 
  However, I wonder how to fix this properly: the error handling
  labels look backward, so that I am not sure where to free the
  allocated context in case of error.

 applied this, yes its correct
   
Ok, there are other details missing, the stream does not get a
duration, start_time and bitrate. This can easily be fixed, but as
wm4 said libquvi seems an abandoned project.
   
Would there be any interest in a solution based on youtube-dl? It
seems to be the standard, these days. I just ran a few tests:
   
youtube -qs 'url' returns 0 or 1 depending on whether the url can be
parsed by the tool
   
youtube -e 'url' prints the stream title
   
and youtube -f 'url' prints the video url.
   
To use this portably, we can use system() and redirect the
output to a temporary file, and read the title or URL from the file.
Or is popen available on all platforms where ffmpeg runs?
   
Is it a clean enough solution? If yes, I can submit the patch adding
this solution. From what I could see, all solutions to parse youtube
(including quvi) are based on scripts anyway.
   
   
   I submitted a patch more that year ago to support youtube natively, as I
   remember this was based on the script you mention, but it was rejected
   because we have libquvi. [sic!]
   Project not actively developed is not the same as project not working.
   In corner case it is not developed because everything is working
   perfectly.
  
  libquvi seems to work on youtube yes, but not on daily motion and
  other streaming sites I tried where youtube-dl works, it simply does
  not work. The sites change, so the tool to parse them must be up to
  date to work correctly. So no, libquvi does not work perfectly.
  
  And libquvi ffmpeg demuxer does not seem to be in a very good shape
  either, it simply segfaults, and even when the segfault is resolved,
  has other issues, the extern AVFormatcontext fields are simply not
  filled.
 
 the whitelist segfault is already fixed
 and it worked with git master (download from youtube) today
 
 maybe someone wants to take over libquvi maintaince ? i mean libquvi
 not our wraper. ive seen there are patches on teh libquvi developer
 mailing list which supposedly fix some of the not working services

I think the set of sites supported by libquvi is really small
compared to the ones supported by youtube-dl, and even smaller the
site that actually work, because they have changed and quvi did not
change.

 
 but really iam happy with any solution that works and that people
 like, libquvi, youtube-dl, anythng else
 
 but a plain ./ffplay http://youtube.com/...
 should work ideally
 or less ideally with a forced demuxer -f xxx http://youtube.com/...

Meanwhile, I get that solution working, however, there is a catch:
youtube-dl has a generic extractor which accepts plain http urls
to videos, so, when the input format opens the nested import
formats, the probe function succeeds again, and we have an infinite
recursion, I have not found a way to disable this extractor on the
command line, and have to admit that my ffmpeg knowledge is not
sufficient at this point, the only solution I have found to prevent
recursion is to return 0 in the probe function all the time, and
only be able to use the format with -f. Shall I send the patch, or
could you explain me how to avoid this ?

-- 
Gilles.


pgpg1OIwMtKqq.pgp
Description: PGP signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 3/3] avformat: add youtube-dl based demuxer

2015-04-08 Thread Gilles Chanteperdrix

On Wed, Apr 08, 2015 at 02:08:21PM -0400, compn wrote:
 On Wed, 8 Apr 2015 19:51:12 +0200
 Gilles Chanteperdrix gilles.chanteperd...@xenomai.org wrote:
 
  On Wed, Apr 08, 2015 at 07:44:13PM +0200, wm4 wrote:
   On Wed, 8 Apr 2015 19:39:00 +0200
   Gilles Chanteperdrix gilles.chanteperd...@xenomai.org wrote:
   
On Wed, Apr 08, 2015 at 07:24:27PM +0200, wm4 wrote:
  +snprintf(buffer, sizeof(buffer), youtube-dl -f %s -g
  '%s', 
  +   yc-format, s-filename);

Ok, missing single quotes here around the format.

   
   Doesn't help. You can't fix it. You need to use something other than
   system() if you want it to be secure.
  
  You can fix it, you can escape the quotes in the string or refuse a
  string that contains single quotes, but as I said, this starts being
  cumbersome.
  
 
 why not machine translate the python to c and import that into
 ffmpeg? :P
 
 https://github.com/pradyun/Py2C

Because then you will have to maintain that C, update it as soon as
a youtube-dl python file is updated because a site changed.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 3/3] avformat: add youtube-dl based demuxer

2015-04-08 Thread Gilles Chanteperdrix

On Wed, Apr 08, 2015 at 04:31:34PM -0400, compn wrote:
 On Wed, 8 Apr 2015 22:26:23 +0200
 Gilles Chanteperdrix gilles.chanteperd...@xenomai.org wrote:
 
  ready to forget all about these patches.
 
 thanks for the patches, even if some people do not like them.
 
 please dont be discouraged by strongly worded mails.
 
 its a long term bikeshed that you have stepped in. the developers are
 mostly shouting into the wind around here about what should be in each
 library and what should not.

I am not discouraged, it is better to see a lively mailing list than
one where people do not care, but I will pass on these patches, I am
not very much interested by the functionality, I agree that calling
the script from the demuxer was a bad idea, I was just looking for a
quick solution, and have other things to do.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 3/3] avformat: add youtube-dl based demuxer

2015-04-08 Thread Gilles Chanteperdrix

On Wed, Apr 08, 2015 at 09:59:17PM +0200, Nicolas George wrote:
 Le nonidi 19 germinal, an CCXXIII, Gilles Chanteperdrix a écrit :
  As I explained, I did it this way for portability. system is ANSI
  not POSIX, redirections work with windows shell too, and av_tempfile
  is implemented by ffmpeg, so I expect it to be portable.
 
 Implementing pipe reading for other systems looks like a better time
 investment.
 
  Well, actually, it looks not so hard to me.
 
 Every person who got it wrong thought that too.
 
  One argument is an URL so, we could url encode it before passing it to
  youtube-dl
 
 No, you can not, because it is probably already URL-encoded.
 
  Well, actually, vfork() + exec() is better than fork() + exec(),
  both performance wise and because it works on machines without an
  MMU, and the implementation of posix_spawn() in glibc is too
  conservative and uses fork() + exec() even in situations where
  vfork() could be used, see:
  https://sourceware.org/bugzilla/show_bug.cgi?id=10354
  
  So, better use vfork() + exec() (that is what system does for
  instance). I did not use it simply because I thought ffmpeg could
  have to be compiled on machines without the POSIX interface.
 
 That was arguments for 20 years ago. Nowadays, vfork() was left to bitrot
 for compatibility reasons while fork() was insanely optimized. And in the
 meantime, threads were invented, with their own set of problems when it
 comes to spawning a child process.

That is what the doc says, but the doc is wrong. Measurements show
that fork() has a detrimental effect, not as much at the moment of
the fork, this is true, this instant is short, but later: all
private writable mappings in the parent process are made read-only
and need to take a fault and go to kernel-space to re-change the MMU
configuration. You can make a simple experiment, map a large private
anonymous area (say 512 MB), and measure the time it takes to write
one byte on each page before and after a fork. At least the
difference was huge last time I tested.

And again, on a platform without MMU, you have vfork(), not fork().
So, when you just need to run exec, it is stupid 

 
 So the short of it: use posix_spawn().
 
 The way it is implemented is at best not your concern, at worst premature
 optimization.
 
  I did not use it simply because I thought ffmpeg could have to be compiled
  on machines without the POSIX interface.
 
 FFmpeg requires POSIX. posix_spawn() is a POSIX option, yet, so it requires
 a configure check. But apart from that, let someone who cares implement for
 other systems.
 
 Anyway:
 
   Then I suggest you just endeavour to write a pseudo-demuxer for the output
   of youtube-dl -J. No need of a shell or an external command.
 
 Just do that, no need for any fancy system call for that, and it would serve
 as a base for a more fully automatic version anyway later.

I do not follow you here. youtube-dl -J has a uselessly verbose
output, much harder to parse than youtube-dl -g which gives you just
one URL. But you still need to run youtube-dl. So, I do not see the
upside.

-- 
Gilles.


pgp4T9wkD6zIe.pgp
Description: PGP signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 3/3] avformat: add youtube-dl based demuxer

2015-04-08 Thread Gilles Chanteperdrix

On Wed, Apr 08, 2015 at 10:37:13PM +0200, Gilles Chanteperdrix wrote:
 On Wed, Apr 08, 2015 at 04:31:34PM -0400, compn wrote:
  On Wed, 8 Apr 2015 22:26:23 +0200
  Gilles Chanteperdrix gilles.chanteperd...@xenomai.org wrote:
  
   ready to forget all about these patches.
  
  thanks for the patches, even if some people do not like them.
  
  please dont be discouraged by strongly worded mails.
  
  its a long term bikeshed that you have stepped in. the developers are
  mostly shouting into the wind around here about what should be in each
  library and what should not.
 
 I am not discouraged, it is better to see a lively mailing list than
 one where people do not care, but I will pass on these patches, I am
 not very much interested by the functionality, I agree that calling
 the script from the demuxer was a bad idea, I was just looking for a
 quick solution, and have other things to do.

And forgot to add my apologies for these bad patches and the noise
that followed.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 3/3] avformat: add youtube-dl based demuxer

2015-04-08 Thread Gilles Chanteperdrix

On Wed, Apr 08, 2015 at 08:52:12PM +0100, Derek Buitenhuis wrote:
 On 4/8/2015 6:16 PM, Gilles Chanteperdrix wrote:
  Signed-off-by: Gilles Chanteperdrix gilles.chanteperd...@xenomai.org
  ---
   libavformat/Makefile |   1 +
   libavformat/allformats.c |   1 +
   libavformat/youtubedl.c  | 221 
  +++
   3 files changed, 223 insertions(+)
   create mode 100644 libavformat/youtubedl.c
 
 I must say this is incredibly dumb and has no place at all in libav*format*. 
 This
 should be done in the layer above whatever avformat (and not even to mention 
 the
 hilarity of running a cli script from a library).
 
 I am registering my dissent here, in writing, so I can point to it in the 
 future
 when asked (and I always get asked).
 
 NOTE: I am sure people will respond and explain how avformat is totally the 
 right
 place, just like it was the right place for a dir listing api, and an http 
 server,
 and probably soon systemd. I am not going to respond, or otherwise engage in
 mailing list flamewars. This mail is read-only.

For my defense: I asked if it would not be to ugly before starting
and (mis)writing it.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 3/3] avformat: add youtube-dl based demuxer

2015-04-08 Thread Gilles Chanteperdrix

On Wed, Apr 08, 2015 at 10:08:25PM +0200, wm4 wrote:
 On Wed, 08 Apr 2015 20:52:12 +0100
 Derek Buitenhuis derek.buitenh...@gmail.com wrote:
 
  On 4/8/2015 6:16 PM, Gilles Chanteperdrix wrote:
   Signed-off-by: Gilles Chanteperdrix gilles.chanteperd...@xenomai.org
   ---
libavformat/Makefile |   1 +
libavformat/allformats.c |   1 +
libavformat/youtubedl.c  | 221 
   +++
3 files changed, 223 insertions(+)
create mode 100644 libavformat/youtubedl.c
  
  I must say this is incredibly dumb and has no place at all in 
  libav*format*. This
  should be done in the layer above whatever avformat (and not even to 
  mention the
  hilarity of running a cli script from a library).
  
  I am registering my dissent here, in writing, so I can point to it in the 
  future
  when asked (and I always get asked).
  
  NOTE: I am sure people will respond and explain how avformat is totally the 
  right
  place, just like it was the right place for a dir listing api, and an http 
  server,
  and probably soon systemd. I am not going to respond, or otherwise engage in
  mailing list flamewars. This mail is read-only.
 
 +1
 
 As usual, people are trying to force things that don't belong into a
 low level API, just because the low level API is the only API in ffmpeg.

For my defense again: the libquvi demuxer caused a segmentation
fault, I posted here, you answered that the project was dead, so I looked
for a replacement. I am not trying to force anything. I am ready to
forget all about these patches.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH] avformat/srtenc: avoid segmentation fault when transcoding from SSA to SRT

2015-03-09 Thread Gilles Chanteperdrix

On Sun, Mar 08, 2015 at 10:57:54AM +0100, Nicolas George wrote:
 L'octidi 18 ventôse, an CCXXIII, Clement Boesch a écrit :
  Try adding tags with no text maybe. You may also try ASS drawing mode, but
  FFmpeg probably doesn't do cray stuff about it.
 
 I tried various combinations, including empty styles, drawing mode, Picture,
 Comment, empty lines, missing comma, nothing produces a crash.

Ok, I will remove the patch on my side, and chime in if I get the
segfault again. Thanks for investigating, sorry for the noise.

Regards.

-- 
Gilles.


pgpfq8h5xrkbH.pgp
Description: PGP signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH] avformat/srtenc: avoid segmentation fault when transcoding from SSA to SRT

2015-03-07 Thread Gilles Chanteperdrix

On Sat, Mar 07, 2015 at 11:46:44PM +0100, Nicolas George wrote:
 Le septidi 17 ventôse, an CCXXIII, Gilles Chanteperdrix a écrit :
  You need an ASS file that produces a sub-title frame without text. I
  have been carrying this patch since 2013, I am afraid I no longer
  have the problematic file.
 
 I tried an ASS file with an empty text part, and it did not crash
 either.

Is not there some way in the ssa format to declare some graphic
elements which are not text?

 It
 is entirely possible the problem has disappeared due to some of the reworks
 of the text subtitles code since you experienced it.

In my modest experience, bugs do not just disappear if you do not
fix them.

 
 I am sorry to insist, but I am always suspicious of changes that go if the
 value is not what we expect, silently ignore it, because they frequently
 should return an error or hide a bug earlier in the code; 

Maybe it is not a perfect patch, but if subtitles can contain
graphic elements mixed with text, it seems better to translate the
text only when the target codec only supports text, than to fail
with a segmentation fault.

 for these reasons,
 I would like either a more detailed analysis or a chance to see for myself
 before this patch is applied.

Just to make things clear: you will not get such an analysis for me.
I am perfectly fine with the patch applied in my local copy of
ffmpeg.

-- 
Gilles.


pgpgGnMK76Xzl.pgp
Description: PGP signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

[FFmpeg-devel] [PATCH] avformat/srtenc: avoid segmentation fault when transcoding from SSA to SRT

2015-03-07 Thread Gilles Chanteperdrix

When transcoding to SRT, do not assume that each sub-title frame contains
text.

Signed-off-by: Gilles Chanteperdrix gilles.chanteperd...@xenomai.org
---
 libavcodec/srtenc.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/libavcodec/srtenc.c b/libavcodec/srtenc.c
index 3287970..ab790f8 100644
--- a/libavcodec/srtenc.c
+++ b/libavcodec/srtenc.c
@@ -241,6 +241,9 @@ static int srt_encode_frame(AVCodecContext *avctx,
 for (; dialog  num--; dialog++) {
 s-alignment_applied = 0;
 srt_style_apply(s, dialog-style);
+if (!dialog-text)
+continue;
+
 ff_ass_split_override_codes(srt_callbacks, s, dialog-text);
 }
 }
-- 
1.8.4

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH] avformat/srtenc: avoid segmentation fault when transcoding from SSA to SRT

2015-03-07 Thread Gilles Chanteperdrix

On Sat, Mar 07, 2015 at 11:46:44PM +0100, Nicolas George wrote:
 Le septidi 17 ventôse, an CCXXIII, Gilles Chanteperdrix a écrit :
  You need an ASS file that produces a sub-title frame without text. I
  have been carrying this patch since 2013, I am afraid I no longer
  have the problematic file.
 
 I tried an ASS file with an empty text part, and it did not crash either. It
 is entirely possible the problem has disappeared due to some of the reworks
 of the text subtitles code since you experienced it.
 
 I am sorry to insist, but I am always suspicious of changes that go if the
 value is not what we expect, silently ignore it, because they frequently
 should return an error or hide a bug earlier in the code; for these reasons,
 I would like either a more detailed analysis or a chance to see for myself
 before this patch is applied.

If I read ffmpeg current documentation here:

http://www.ffmpeg.org/doxygen/2.5/group__lavc__core.html#ga828218dcb8874ab2c5af8d486c365421

It seems that when a subtitle rect type is SUBTITLE_ASS (which is
the case in the code which my patch modified), there is no guarantee
that the text field contains something.

-- 
Gilles.


pgpoVl06GLm91.pgp
Description: PGP signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH] avformat/srtenc: avoid segmentation fault when transcoding from SSA to SRT

2015-03-07 Thread Gilles Chanteperdrix

On Sat, Mar 07, 2015 at 09:18:41PM +0100, Nicolas George wrote:
 Le septidi 17 ventôse, an CCXXIII, Gilles Chanteperdrix a écrit :
  When transcoding to SRT, do not assume that each sub-title frame contains
  text.
 
 Can you explain how to reproduce the segfault? I tried transcoding an ASS
 file and it worked.

You need an ASS file that produces a sub-title frame without text. I
have been carrying this patch since 2013, I am afraid I no longer
have the problematic file.

-- 
Gilles.


pgprvrq1x6I6o.pgp
Description: PGP signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

[FFmpeg-devel] [PATCH 1/2] avformat/rtpdec_mpa_robust: fix commit 96084251e57d1738fde02a2b0d37ca609d9efd71

2015-03-02 Thread Gilles Chanteperdrix

Commit 96084251e57d1738fde02a2b0d37ca609d9efd71 introduced a change in
the parser implementation which broke it. Restore the original
implementation.

Signed-off-by: Gilles Chanteperdrix gilles.chanteperd...@xenomai.org
---
 libavformat/rtpdec_mpa_robust.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/libavformat/rtpdec_mpa_robust.c b/libavformat/rtpdec_mpa_robust.c
index 776e245..2439f9e 100644
--- a/libavformat/rtpdec_mpa_robust.c
+++ b/libavformat/rtpdec_mpa_robust.c
@@ -98,7 +98,7 @@ static int mpa_robust_parse_packet(AVFormatContext *ctx, 
PayloadContext *data,
 pkt-stream_index = st-index;
 memcpy(pkt-data, buf, adu_size);
 
-data-split_pos += adu_size;
+data-split_pos = (buf - data-split_buf) + adu_size;
 
 if (data-split_pos == data-split_buf_size) {
 av_freep(data-split_buf);
-- 
1.8.4

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

[FFmpeg-devel] [PATCH 2/2] MAINTAINERS: add myself as rtpdec_mpa_robust maintainer

2015-03-02 Thread Gilles Chanteperdrix

Signed-off-by: Gilles Chanteperdrix gilles.chanteperd...@xenomai.org
---
 MAINTAINERS | 1 +
 1 file changed, 1 insertion(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 0cd8c26..007540c 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -471,6 +471,7 @@ Muxers/Demuxers:
   rtpdec_dv.*   Thomas Volkert
   rtpdec_h261.*, rtpenc_h261.*  Thomas Volkert
   rtpdec_hevc.*, rtpenc_hevc.*  Thomas Volkert
+  rtpdec_mpa_robust.*   Gilles Chanteperdrix
   rtpdec_asf.*  Ronald S. Bultje
   rtpdec_vp9.c  Thomas Volkert
   rtpenc_mpv.*, rtpenc_aac.*Martin Storsjo
-- 
1.8.4

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

[FFmpeg-devel] [PATCH] avformat/rtsp: punch holes again after pause

2015-02-21 Thread Gilles Chanteperdrix

When a client behind a NAT issues a pause command, and stay paused for a
long time, the router may stop the RTP/RTCP port redirection. Resend the
hole punching packets after each PLAY command to cause the router to
restart the port redirection in that case.

Signed-off-by: Gilles Chanteperdrix gilles.chanteperd...@xenomai.org
---
 libavformat/rtspdec.c | 4 
 1 file changed, 4 insertions(+)

diff --git a/libavformat/rtspdec.c b/libavformat/rtspdec.c
index fa7383c..bfbe52b 100644
--- a/libavformat/rtspdec.c
+++ b/libavformat/rtspdec.c
@@ -542,6 +542,10 @@ static int rtsp_read_play(AVFormatContext *s)
 AVStream *st = NULL;
 if (!rtpctx || rtsp_st-stream_index  0)
 continue;
+if (CONFIG_RTPDEC 
+!(rt-server_type == RTSP_SERVER_WMS  i  1)  
s-iformat)
+ff_rtp_send_punch_packets(rtsp_st-rtp_handle);
+
 st = s-streams[rtsp_st-stream_index];
 rtpctx-range_start_offset =
 av_rescale_q(reply-range_start, AV_TIME_BASE_Q,
-- 
1.8.4

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 1/3] avformat/rtpdec_mpeg4: reassemble fragmented AAC frames

2015-02-21 Thread Gilles Chanteperdrix

On Sat, Feb 21, 2015 at 04:54:12PM +0100, Thomas Volkert wrote:
 +if (len  data-au_headers[0].size) {
 +av_log(ctx, AV_LOG_ERROR, First AU larger than packet size\n);
 
 Can this really happen?
 In the lines above you already check for the case that data is missing while
 you have already received the last packet of the frame.

The first test tests for 
data-nb_au_headers == 1  len  data-au_headers[0].size

We could still have:
data-nb_au_headers = 2  len  data-au_headers[0].size

Which is illegal per the RFC (a packet contains either a piece of a
fragmented frame, or one or several entire frames, but never both),
but some malformed packets could happen.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 1/3] avformat/rtpdec_mpeg4: reassemble fragmented AAC frames

2015-02-16 Thread Gilles Chanteperdrix

On Fri, Feb 13, 2015 at 10:51:32PM +0100, Gilles Chanteperdrix wrote:
 Signed-off-by: Gilles Chanteperdrix gilles.chanteperd...@xenomai.org
 ---
  libavformat/rtpdec_mpeg4.c | 84 
 +-
  1 file changed, 76 insertions(+), 8 deletions(-)

Hi, what about this patch?

Regards.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 2/3] avformat/rtpdec_mpeg12: add robust MPEG audio depacketization (RFC 5219)

2015-02-14 Thread Gilles Chanteperdrix

On Sat, Feb 14, 2015 at 05:56:49PM +0100, Thomas Volkert wrote:
 On 02/13/2015 10:51 PM, Gilles Chanteperdrix wrote:
   #define LIBAVFORMAT_VERSION_MAJOR 56
 -#define LIBAVFORMAT_VERSION_MINOR  19
 -#define LIBAVFORMAT_VERSION_MICRO 101
 +#define LIBAVFORMAT_VERSION_MINOR  20
 +#define LIBAVFORMAT_VERSION_MICRO 102
 In this case it should be:
 
 +#define LIBAVFORMAT_VERSION_MINOR  20
 +#define LIBAVFORMAT_VERSION_MICRO 100
 
 My comment was maybe a bit misleading. But I think this can be fixed when the 
 patch gets committed.

Sorry, I was completely ignorant of how major/minor/micro version
worked. I can post new versions of the patches if needed.

Thanks for your patient review.
Regards.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 0/7] using ffmpeg as an RTSP client

2015-02-13 Thread Gilles Chanteperdrix

On Sun, Feb 08, 2015 at 10:20:20PM +0100, Gilles Chanteperdrix wrote:
 Hi,
 
 I am using ffmpeg 2.5.3 as an RTSP client and found a few missing
 pieces, which the following series of patches should address.

Hi,

ping ?

I sent these patches almost a week ago and received no answer. Is it
because the patches are not conform to the submission rules, or
simply because people are to busy for reviewing them?

Thanks in advance.
Regards.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 0/7] using ffmpeg as an RTSP client

2015-02-13 Thread Gilles Chanteperdrix

On Fri, Feb 13, 2015 at 07:45:27PM +0100, Gilles Chanteperdrix wrote:
 On Sun, Feb 08, 2015 at 10:20:20PM +0100, Gilles Chanteperdrix wrote:
  Hi,
  
  I am using ffmpeg 2.5.3 as an RTSP client and found a few missing
  pieces, which the following series of patches should address.
 
 Hi,
 
 ping ?
 
 I sent these patches almost a week ago and received no answer. Is it
 because the patches are not conform to the submission rules, or
 simply because people are to busy for reviewing them?

Never mind. Sorry for the noise.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Re: [FFmpeg-devel] [PATCH 1/7] avformat/rtpdec_ac3: add AC3 RTP depacketization

2015-02-13 Thread Gilles Chanteperdrix

On Thu, Feb 12, 2015 at 10:07:39PM +0100, Thomas Volkert wrote:
 +if (nr_frames  1) {
 +av_log(ctx, AV_LOG_ERROR,
 +   Unimplemented multiple AC3 frames per packet\n);
 
 You could use avpriv_report_missing_feature() here.

I am not sure that the case where there are multiple frames in one
packet is not already working: frames are concatenated without not
headers, so the full AC3 parser should be able to split the packet
in multiple frames right ? Unfortunately, I do not have an example
using this feature.


 +if (nr_frames  1) {
 +av_log(ctx, AV_LOG_ERROR,
 +   Unimplemented multiple AC3 frames per packet\n);
 
 You could use avpriv_report_missing_feature() here.

I am not sure that the case where there are multiple frames in one
packet is not already working: frames are concatenated without not
headers, so the full AC3 parser should be able to split the packet
in multiple frames right ? Unfortunately, I do not have an example
using this feature.

 
 Do you see any use for the optional SDP parameter?
 If yes, you could add an AC3 specific parser for SDP line and link it as
 .parse_sdp_a_line line here.

Which optional SDP parameter? channels,ptime,maxptime ?

 What do you think about a packetizer as counterpart?

Currently, I am using live555 as an RTSP server. I did not check
ffmpeg documentation to see how to run an RTSP server. I will check.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

[FFmpeg-devel] [PATCH 2/3] avformat/rtpdec_mpeg12: add robust MPEG audio depacketization (RFC 5219)

2015-02-13 Thread Gilles Chanteperdrix

Signed-off-by: Gilles Chanteperdrix gilles.chanteperd...@xenomai.org
---
 Changelog   |   1 +
 libavcodec/mpegaudio_parser.c   |  14 ++-
 libavcodec/mpegaudiodecheader.c |   3 +-
 libavformat/rtpdec.c|   1 +
 libavformat/rtpdec_formats.h|   1 +
 libavformat/rtpdec_mpeg12.c | 204 
 libavformat/version.h   |   4 +-
 7 files changed, 223 insertions(+), 5 deletions(-)

diff --git a/Changelog b/Changelog
index c663d5e..775fc79 100644
--- a/Changelog
+++ b/Changelog
@@ -23,6 +23,7 @@ version next:
 - Changed default DNxHD colour range in QuickTime .mov derivatives to mpeg 
range
 - ported softpulldown filter from libmpcodecs as repeatfields filter
 - dcshift filter
+- RTP parser for loss tolerant payload format for MP3 audio (RFC 5219)
 
 
 version 2.5:
diff --git a/libavcodec/mpegaudio_parser.c b/libavcodec/mpegaudio_parser.c
index 79dbf63..42f4706 100644
--- a/libavcodec/mpegaudio_parser.c
+++ b/libavcodec/mpegaudio_parser.c
@@ -64,7 +64,7 @@ static int mpegaudio_parse(AVCodecParserContext *s1,
 }else{
 while(ibuf_size){
 int ret, sr, channels, bit_rate, frame_size;
-enum AVCodecID codec_id;
+enum AVCodecID codec_id = avctx-codec_id;
 
 state= (state8) + buf[i++];
 
@@ -90,6 +90,16 @@ static int mpegaudio_parse(AVCodecParserContext *s1,
 avctx-bit_rate += (bit_rate - avctx-bit_rate) / 
(s-header_count - header_threshold);
 }
 }
+
+if (s1-flags  PARSER_FLAG_COMPLETE_FRAMES) {
+s-frame_size = 0;
+next = buf_size;
+} else if (codec_id == AV_CODEC_ID_MP3ADU) {
+av_log(avctx, AV_LOG_ERROR,
+MP3ADU full parser not implemented);
+return AVERROR_PATCHWELCOME;
+}
+
 break;
 }
 }
@@ -110,7 +120,7 @@ static int mpegaudio_parse(AVCodecParserContext *s1,
 
 
 AVCodecParser ff_mpegaudio_parser = {
-.codec_ids  = { AV_CODEC_ID_MP1, AV_CODEC_ID_MP2, AV_CODEC_ID_MP3 },
+.codec_ids  = { AV_CODEC_ID_MP1, AV_CODEC_ID_MP2, AV_CODEC_ID_MP3, 
AV_CODEC_ID_MP3ADU },
 .priv_data_size = sizeof(MpegAudioParseContext),
 .parser_parse   = mpegaudio_parse,
 .parser_close   = ff_parse_close,
diff --git a/libavcodec/mpegaudiodecheader.c b/libavcodec/mpegaudiodecheader.c
index 5db1957..6af6e4b 100644
--- a/libavcodec/mpegaudiodecheader.c
+++ b/libavcodec/mpegaudiodecheader.c
@@ -134,7 +134,8 @@ int avpriv_mpa_decode_header2(uint32_t head, int 
*sample_rate, int *channels, in
 break;
 default:
 case 3:
-*codec_id = AV_CODEC_ID_MP3;
+if (*codec_id != AV_CODEC_ID_MP3ADU)
+*codec_id = AV_CODEC_ID_MP3;
 if (s-lsf)
 *frame_size = 576;
 else
diff --git a/libavformat/rtpdec.c b/libavformat/rtpdec.c
index 4ff209c..c632340 100644
--- a/libavformat/rtpdec.c
+++ b/libavformat/rtpdec.c
@@ -89,6 +89,7 @@ void ff_register_rtp_dynamic_payload_handlers(void)
 ff_register_dynamic_payload_handler(ff_mp4a_latm_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_mp4v_es_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_mpeg_audio_dynamic_handler);
+ff_register_dynamic_payload_handler(ff_mpeg_audio_robust_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_mpeg_video_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_mpeg4_generic_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_mpegts_dynamic_handler);
diff --git a/libavformat/rtpdec_formats.h b/libavformat/rtpdec_formats.h
index 87e316f..da22d70 100644
--- a/libavformat/rtpdec_formats.h
+++ b/libavformat/rtpdec_formats.h
@@ -52,6 +52,7 @@ extern RTPDynamicProtocolHandler ff_jpeg_dynamic_handler;
 extern RTPDynamicProtocolHandler ff_mp4a_latm_dynamic_handler;
 extern RTPDynamicProtocolHandler ff_mp4v_es_dynamic_handler;
 extern RTPDynamicProtocolHandler ff_mpeg_audio_dynamic_handler;
+extern RTPDynamicProtocolHandler ff_mpeg_audio_robust_dynamic_handler;
 extern RTPDynamicProtocolHandler ff_mpeg_video_dynamic_handler;
 extern RTPDynamicProtocolHandler ff_mpeg4_generic_dynamic_handler;
 extern RTPDynamicProtocolHandler ff_mpegts_dynamic_handler;
diff --git a/libavformat/rtpdec_mpeg12.c b/libavformat/rtpdec_mpeg12.c
index d73ff1e..d99c507 100644
--- a/libavformat/rtpdec_mpeg12.c
+++ b/libavformat/rtpdec_mpeg12.c
@@ -2,6 +2,9 @@
  * Common code for the RTP depacketization of MPEG-1/2 formats.
  * Copyright (c) 2002 Fabrice Bellard
  *
+ * RTP parser for loss tolerant payload format for MP3 audio (RFC 5219)
+ * Copyright (c) 2015 Gilles Chanteperdrix g...@xenomai.org
+ *
  * This file is part of FFmpeg.
  *
  * FFmpeg is free software; you can redistribute it and/or
@@ -23,6 +26,8

[FFmpeg-devel] [PATCH 3/3] avformat/rtpdec_ac3: add AC3 RTP depacketization (RFC 4184)

2015-02-13 Thread Gilles Chanteperdrix

Signed-off-by: Gilles Chanteperdrix gilles.chanteperd...@xenomai.org
---
 Changelog|   1 +
 MAINTAINERS  |   1 +
 libavformat/Makefile |   1 +
 libavformat/rtpdec.c |   1 +
 libavformat/rtpdec_ac3.c | 157 +++
 libavformat/rtpdec_formats.h |   1 +
 libavformat/version.h|   4 +-
 7 files changed, 164 insertions(+), 2 deletions(-)
 create mode 100644 libavformat/rtpdec_ac3.c

diff --git a/Changelog b/Changelog
index 775fc79..085fd17 100644
--- a/Changelog
+++ b/Changelog
@@ -24,6 +24,7 @@ version next:
 - ported softpulldown filter from libmpcodecs as repeatfields filter
 - dcshift filter
 - RTP parser for loss tolerant payload format for MP3 audio (RFC 5219)
+- RTP parser for AC3 payload format (RFC 4184)
 
 
 version 2.5:
diff --git a/MAINTAINERS b/MAINTAINERS
index 13b211e..8430222 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -465,6 +465,7 @@ Muxers/Demuxers:
   rmdec.c, rmenc.c  Ronald S. Bultje, Kostya Shishkov
   rtmp* Kostya Shishkov
   rtp.c, rtpenc.c   Martin Storsjo
+  rtpdec_ac3.*  Gilles Chanteperdrix
   rtpdec_h261.*, rtpenc_h261.*  Thomas Volkert
   rtpdec_hevc.*, rtpenc_hevc.*  Thomas Volkert
   rtpdec_asf.*  Ronald S. Bultje
diff --git a/libavformat/Makefile b/libavformat/Makefile
index ec312ba..073a42d 100644
--- a/libavformat/Makefile
+++ b/libavformat/Makefile
@@ -29,6 +29,7 @@ OBJS-$(CONFIG_RIFFENC)   += riffenc.o
 OBJS-$(CONFIG_RTPDEC)+= rdt.o   \
 rtp.o   \
 rtpdec.o\
+rtpdec_ac3.o\
 rtpdec_amr.o\
 rtpdec_asf.o\
 rtpdec_g726.o   \
diff --git a/libavformat/rtpdec.c b/libavformat/rtpdec.c
index c632340..550da4a 100644
--- a/libavformat/rtpdec.c
+++ b/libavformat/rtpdec.c
@@ -72,6 +72,7 @@ void 
ff_register_dynamic_payload_handler(RTPDynamicProtocolHandler *handler)
 
 void ff_register_rtp_dynamic_payload_handlers(void)
 {
+ff_register_dynamic_payload_handler(ff_ac3_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_amr_nb_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_amr_wb_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_g726_16_dynamic_handler);
diff --git a/libavformat/rtpdec_ac3.c b/libavformat/rtpdec_ac3.c
new file mode 100644
index 000..728f421
--- /dev/null
+++ b/libavformat/rtpdec_ac3.c
@@ -0,0 +1,157 @@
+/*
+ * RTP parser for AC3 payload format (RFC 4184)
+ * Copyright (c) 2015 Gilles Chanteperdrix g...@xenomai.org
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * FFmpeg is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with FFmpeg; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include avformat.h
+#include rtpdec_formats.h
+
+#define RTP_AC3_PAYLOAD_HEADER_SIZE 2
+
+struct PayloadContext {
+unsigned nr_frames;
+unsigned last_frame;
+uint32_t timestamp;
+AVIOContext *fragment;
+};
+
+static av_cold int ac3_init(AVFormatContext *s, int st_index, 
+   PayloadContext *data)
+{
+if (st_index  0)
+return 0;
+s-streams[st_index]-need_parsing = AVSTREAM_PARSE_FULL;
+return 0;
+}
+
+static PayloadContext *ac3_new_context(void)
+{
+return av_mallocz(sizeof(PayloadContext));
+}
+
+static inline void free_fragment_if_needed(PayloadContext *data)
+{
+if (data-fragment) {
+uint8_t *p;
+avio_close_dyn_buf(data-fragment, p);
+av_free(p);
+data-fragment = NULL;
+}
+}
+
+static void ac3_free_context(PayloadContext *data)
+{
+free_fragment_if_needed(data);
+av_free(data);
+}
+
+static int ac3_handle_packet(AVFormatContext *ctx, PayloadContext *data,
+  AVStream *st, AVPacket *pkt, uint32_t *timestamp,
+  const uint8_t *buf, int len, uint16_t seq,
+  int flags)
+{
+unsigned frame_type;
+unsigned

[FFmpeg-devel] [PATCH 1/3] avformat/rtpdec_mpeg4: reassemble fragmented AAC frames

2015-02-13 Thread Gilles Chanteperdrix

Signed-off-by: Gilles Chanteperdrix gilles.chanteperd...@xenomai.org
---
 libavformat/rtpdec_mpeg4.c | 84 +-
 1 file changed, 76 insertions(+), 8 deletions(-)

diff --git a/libavformat/rtpdec_mpeg4.c b/libavformat/rtpdec_mpeg4.c
index 9655d30..f2814b5 100644
--- a/libavformat/rtpdec_mpeg4.c
+++ b/libavformat/rtpdec_mpeg4.c
@@ -33,6 +33,8 @@
 #include libavutil/avstring.h
 #include libavcodec/get_bits.h
 
+#define MAX_AAC_HBR_FRAME_SIZE 8191
+
 /** Structure listing useful vars to parse RTP packet payload */
 struct PayloadContext {
 int sizelength;
@@ -59,8 +61,9 @@ struct PayloadContext {
 int au_headers_length_bytes;
 int cur_au_index;
 
-uint8_t buf[RTP_MAX_PACKET_LENGTH];
+uint8_t buf[FFMAX(RTP_MAX_PACKET_LENGTH, MAX_AAC_HBR_FRAME_SIZE)];
 int buf_pos, buf_size;
+uint32_t timestamp;
 };
 
 typedef struct {
@@ -168,30 +171,95 @@ static int aac_parse_packet(AVFormatContext *ctx, 
PayloadContext *data,
 {
 int ret;
 
+
 if (!buf) {
-if (data-cur_au_index  data-nb_au_headers)
+if (data-cur_au_index  data-nb_au_headers) {
+av_log(ctx, AV_LOG_ERROR, Invalid parser state\n);
 return AVERROR_INVALIDDATA;
-if (data-buf_size - data-buf_pos  
data-au_headers[data-cur_au_index].size)
+}
+if (data-buf_size - data-buf_pos  
data-au_headers[data-cur_au_index].size) {
+av_log(ctx, AV_LOG_ERROR, Invalid AU size\n);
 return AVERROR_INVALIDDATA;
-if ((ret = av_new_packet(pkt, 
data-au_headers[data-cur_au_index].size))  0)
+}
+if ((ret = av_new_packet(pkt, 
data-au_headers[data-cur_au_index].size))  0) {
+av_log(ctx, AV_LOG_ERROR, Out of memory\n);
 return ret;
+}
 memcpy(pkt-data, data-buf[data-buf_pos], 
data-au_headers[data-cur_au_index].size);
 data-buf_pos += data-au_headers[data-cur_au_index].size;
 pkt-stream_index = st-index;
 data-cur_au_index++;
-return data-cur_au_index  data-nb_au_headers;
+
+if (data-cur_au_index == data-nb_au_headers) {
+data-buf_pos = 0;
+return 0;
+}
+
+return 1;
 }
 
-if (rtp_parse_mp4_au(data, buf, len))
+if (rtp_parse_mp4_au(data, buf, len)) {
+av_log(ctx, AV_LOG_ERROR, Error parsing AU headers\n);
 return -1;
+}
 
 buf += data-au_headers_length_bytes + 2;
 len -= data-au_headers_length_bytes + 2;
+if (data-nb_au_headers == 1  len  data-au_headers[0].size) {
+/* Packet is fragmented */
+
+if (!data-buf_pos) {
+if (data-au_headers[0].size  MAX_AAC_HBR_FRAME_SIZE) {
+av_log(ctx, AV_LOG_ERROR, Invalid AU size\n);
+return AVERROR_INVALIDDATA;
+}
+
+data-buf_size = data-au_headers[0].size;
+data-timestamp = *timestamp;
+}
+
+if (data-timestamp != *timestamp
+|| data-au_headers[0].size != data-buf_size
+|| data-buf_pos + len  MAX_AAC_HBR_FRAME_SIZE) {
+data-buf_pos = 0;
+data-buf_size = 0;
+av_log(ctx, AV_LOG_ERROR, Invalid packet received\n);
+return AVERROR_INVALIDDATA;
+}
 
-if (len  data-au_headers[0].size)
+memcpy(data-buf[data-buf_pos], buf, len);
+data-buf_pos += len;
+
+if (!(flags  RTP_FLAG_MARKER))
+return AVERROR(EAGAIN);
+
+if (data-buf_pos != data-buf_size) {
+data-buf_pos = 0;
+av_log(ctx, AV_LOG_ERROR, Missed some packets, discarding 
frame\n);
+return AVERROR_INVALIDDATA;
+}
+
+data-buf_pos = 0;
+ret = av_new_packet(pkt, data-buf_size);
+if (ret  0) {
+av_log(ctx, AV_LOG_ERROR, Out of memory\n);
+return ret;
+}
+pkt-stream_index = st-index;
+
+memcpy(pkt-data, data-buf, data-buf_size);
+
+return 0;
+}
+
+if (len  data-au_headers[0].size) {
+av_log(ctx, AV_LOG_ERROR, First AU larger than packet size\n);
 return AVERROR_INVALIDDATA;
-if ((ret = av_new_packet(pkt, data-au_headers[0].size))  0)
+}
+if ((ret = av_new_packet(pkt, data-au_headers[0].size))  0) {
+av_log(ctx, AV_LOG_ERROR, Out of memory\n);
 return ret;
+}
 memcpy(pkt-data, buf, data-au_headers[0].size);
 len -= data-au_headers[0].size;
 buf += data-au_headers[0].size;
-- 
1.8.4

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

[FFmpeg-devel] RTP patches v2

2015-02-13 Thread Gilles Chanteperdrix

Hi,

please find a second version of some patches for various RTP audio
payload format. This second version addresses Thomas Volkert
comments.

Regards.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

[FFmpeg-devel] [PATCH 0/7] using ffmpeg as an RTSP client

2015-02-08 Thread Gilles Chanteperdrix

Hi,

I am using ffmpeg 2.5.3 as an RTSP client and found a few missing
pieces, which the following series of patches should address. I used
some test files streamed with an RTSP server on the xenomai.org
server, the files are available for download by replacing
rtsp://xenomai.org:8554 with https://xenomai.org in the URL

- support for depacketization of AC3 RTP frames is missing, patch 1
adds it, following https://tools.ietf.org/html/draft-ietf-avt-rtp-ac3-07
it can be tested with:
rtsp://xenomai.org:8554/video-tests/big_buck_bunny/BigBuckBunny_DivX_HD720p_ASP.divx
(using audio track 2, as audio track 1 uses mp3adu)

- support for receiving sub-titles via RTP is missing. live555 uses
the T140 RTP packetization for that (RFC 4103), and vlc decodes
subrip subtitles sent that way. Patch 2 and 3 implement a compatible
solution, meaning that the AV_CODEC_ID_SUBRIP is used for T140
tracks, of course another codec id could be used. These patches can
be tested with:
rtsp://xenomai.org:8554/video-tests/sintel/sintel-1280-surround.mp4

- support for receiving MP3ADU audio via RTP, following RFC 3119 is
missing. Patch 4 modifies the MPEG audio parser to decode MP3ADU
headers in MP3ADU frames only when the PARSER_FLAG_COMPLETE_FRAMES
is set, because for some reason the parser creates incorrect frames
when trying to parse MP3ADU frames fully with the test file I used.
Patch 5 implements the depacketizer. These patches can be tested
with:
rtsp://xenomai.org:8554/video-tests/big_buck_bunny/BigBuckBunny_DivX_HD720p_ASP.divx
(this time using audio track 1)

- support for reassembly of fragmented AAC frames via RTP is
missing. Patch 6 modifies libavformat/rtpdec_mpeg4.c to add that
support. This patch can be tested with:
rtsp://xenomai.org:8554/video-tests/sintel/sintel-1280-surround.mp4
Without this patch, fragmented AAC frames are silently dropped,
which causes underruns in the audio driver when playing the test
file.

- support for parsing the lang attribute in RTSP tracks is missing.
Patch 7 adds it. This patch can be tested with:
rtsp://xenomai.org:8554/video-tests/sintel/sintel-1280-surround.mp4
Launching ffprobe with this URL should display the languages of the
audio track and subtitle tracks.

Thanks in advance for your review.
Regards.

-- 
Gilles.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

[FFmpeg-devel] [PATCH 1/7] avformat/rtpdec_ac3: add AC3 RTP depacketization

2015-02-08 Thread Gilles Chanteperdrix

---
 libavformat/Makefile |   1 +
 libavformat/rtpdec.c |   1 +
 libavformat/rtpdec_ac3.c | 166 +++
 libavformat/rtpdec_formats.h |   1 +
 4 files changed, 169 insertions(+)
 create mode 100644 libavformat/rtpdec_ac3.c

diff --git a/libavformat/Makefile b/libavformat/Makefile
index 6bf0761..5acb292 100644
--- a/libavformat/Makefile
+++ b/libavformat/Makefile
@@ -30,6 +30,7 @@ OBJS-$(CONFIG_RIFFENC)   += riffenc.o
 OBJS-$(CONFIG_RTPDEC)+= rdt.o   \
 rtp.o   \
 rtpdec.o\
+rtpdec_ac3.o\
 rtpdec_amr.o\
 rtpdec_asf.o\
 rtpdec_g726.o   \
diff --git a/libavformat/rtpdec.c b/libavformat/rtpdec.c
index 33205aa..f5557d8 100644
--- a/libavformat/rtpdec.c
+++ b/libavformat/rtpdec.c
@@ -66,6 +66,7 @@ void 
ff_register_dynamic_payload_handler(RTPDynamicProtocolHandler *handler)
 
 void ff_register_rtp_dynamic_payload_handlers(void)
 {
+ff_register_dynamic_payload_handler(ff_ac3_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_amr_nb_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_amr_wb_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_g726_16_dynamic_handler);
diff --git a/libavformat/rtpdec_ac3.c b/libavformat/rtpdec_ac3.c
new file mode 100644
index 000..2a47b75
--- /dev/null
+++ b/libavformat/rtpdec_ac3.c
@@ -0,0 +1,166 @@
+/*
+ * RTP parser for AC3 payload format
+ * Copyright (c) 2015 Gilles Chanteperdrix g...@xenomai.org
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * FFmpeg is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with FFmpeg; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include avformat.h
+#include rtpdec_formats.h
+#include libavcodec/get_bits.h
+
+/*
+ * From:
+ * https://tools.ietf.org/html/draft-ietf-avt-rtp-ac3-07
+ */
+
+struct PayloadContext {
+unsigned nr_frames;
+unsigned last_frame;
+uint32_t timestamp;
+AVIOContext *fragment;
+};
+
+static av_cold int
+ac3_init(AVFormatContext *s, int st_index, PayloadContext *data)
+{
+if (st_index  0)
+return 0;
+s-streams[st_index]-need_parsing = AVSTREAM_PARSE_FULL;
+return 0;
+}
+
+static PayloadContext *ac3_new_context(void)
+{
+return av_mallocz(sizeof(PayloadContext));
+}
+
+static inline void free_fragment_if_needed(PayloadContext *data)
+{
+if (data-fragment) {
+uint8_t *p;
+avio_close_dyn_buf(data-fragment, p);
+av_free(p);
+data-fragment = NULL;
+}
+}
+
+static void ac3_free_context(PayloadContext *data)
+{
+free_fragment_if_needed(data);
+av_free(data);
+}
+
+static int ac3_handle_packet(AVFormatContext *ctx, PayloadContext *data,
+  AVStream *st, AVPacket *pkt, uint32_t *timestamp,
+  const uint8_t *buf, int len, uint16_t seq,
+  int flags)
+{
+unsigned frame_type;
+unsigned nr_frames;
+int err;
+
+if (len  2) {
+av_log(ctx, AV_LOG_ERROR, Invalid %d bytes packet\n, len);
+return AVERROR_INVALIDDATA;
+}
+
+frame_type = buf[0]  0x3;
+nr_frames = buf[1];
+buf += 2;
+len -= 2;
+
+switch (frame_type) {
+case 0: /* One or more complete frames */
+if (nr_frames  1) {
+av_log(ctx, AV_LOG_ERROR,
+   Unimplemented multiple AC3 frames per packet\n);
+return AVERROR_PATCHWELCOME;
+}
+if (nr_frames == 0) {
+av_log(ctx, AV_LOG_ERROR, Invalid AC3 packet data\n);
+return AVERROR_INVALIDDATA;
+}
+if (av_new_packet(pkt, len)) {
+av_log(ctx, AV_LOG_ERROR, Out of memory.\n);
+return AVERROR(ENOMEM);
+}
+
+pkt-stream_index = st-index;
+memcpy(pkt-data, buf, len);
+return 0;
+
+case 1:
+case 2: /* First fragment */
+free_fragment_if_needed(data);
+
+data-last_frame = 1;
+data-nr_frames = nr_frames;
+err

[FFmpeg-devel] [PATCH 2/7] avformat/rtsp: allow receiving subtitles via RTP

2015-02-08 Thread Gilles Chanteperdrix

---
 libavformat/rtsp.c | 9 ++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/libavformat/rtsp.c b/libavformat/rtsp.c
index ae62252..77407dd 100644
--- a/libavformat/rtsp.c
+++ b/libavformat/rtsp.c
@@ -68,10 +68,11 @@
 { filter_src, only receive packets from the negotiated peer IP, 0, 
AV_OPT_TYPE_CONST, {.i64 = RTSP_FLAG_FILTER_SRC}, 0, 0, DEC, rtsp_flags }
 
 #define RTSP_MEDIATYPE_OPTS(name, longname) \
-{ name, longname, OFFSET(media_type_mask), AV_OPT_TYPE_FLAGS, { .i64 = (1 
 (AVMEDIA_TYPE_DATA+1)) - 1 }, INT_MIN, INT_MAX, DEC, allowed_media_types 
}, \
+{ name, longname, OFFSET(media_type_mask), AV_OPT_TYPE_FLAGS, { .i64 = (1 
 (AVMEDIA_TYPE_SUBTITLE+1)) - 1 }, INT_MIN, INT_MAX, DEC, 
allowed_media_types }, \
 { video, Video, 0, AV_OPT_TYPE_CONST, {.i64 = 1  
AVMEDIA_TYPE_VIDEO}, 0, 0, DEC, allowed_media_types }, \
 { audio, Audio, 0, AV_OPT_TYPE_CONST, {.i64 = 1  
AVMEDIA_TYPE_AUDIO}, 0, 0, DEC, allowed_media_types }, \
-{ data, Data, 0, AV_OPT_TYPE_CONST, {.i64 = 1  AVMEDIA_TYPE_DATA}, 
0, 0, DEC, allowed_media_types }
+{ data, Data, 0, AV_OPT_TYPE_CONST, {.i64 = 1  AVMEDIA_TYPE_DATA}, 
0, 0, DEC, allowed_media_types }, \
+{ subtitle, Subtitle, 0, AV_OPT_TYPE_CONST, {.i64 = 1  
AVMEDIA_TYPE_SUBTITLE}, 0, 0, DEC, allowed_media_types }
 
 #define RTSP_REORDERING_OPTS() \
 { reorder_queue_size, set number of packets to buffer for handling of 
reordered packets, OFFSET(reordering_queue_size), AV_OPT_TYPE_INT, { .i64 = -1 
}, -1, INT_MAX, DEC }
@@ -399,6 +400,8 @@ static void sdp_parse_line(AVFormatContext *s, 
SDPParseState *s1,
 codec_type = AVMEDIA_TYPE_VIDEO;
 } else if (!strcmp(st_type, application)) {
 codec_type = AVMEDIA_TYPE_DATA;
+} else if (!strcmp(st_type, text)) {
+codec_type = AVMEDIA_TYPE_SUBTITLE;
 }
 if (codec_type == AVMEDIA_TYPE_UNKNOWN || !(rt-media_type_mask  (1 
 codec_type))) {
 s1-skip_media = 1;
@@ -2368,7 +2371,7 @@ static int rtp_read_header(AVFormatContext *s)
 /* sdp_read_header initializes this again */
 ff_network_close();
 
-rt-media_type_mask = (1  (AVMEDIA_TYPE_DATA+1)) - 1;
+rt-media_type_mask = (1  (AVMEDIA_TYPE_SUBTITLE+1)) - 1;
 
 ret = sdp_read_header(s);
 s-pb = NULL;
-- 
1.8.4

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

[FFmpeg-devel] [PATCH 7/7] avformat/rtsp: parse lang attribute in SDP

2015-02-08 Thread Gilles Chanteperdrix

---
 libavformat/rtsp.c | 12 
 libavformat/rtsp.h |  2 ++
 2 files changed, 14 insertions(+)

diff --git a/libavformat/rtsp.c b/libavformat/rtsp.c
index 77407dd..3a68020 100644
--- a/libavformat/rtsp.c
+++ b/libavformat/rtsp.c
@@ -477,6 +477,8 @@ static void sdp_parse_line(AVFormatContext *s, 
SDPParseState *s1,
 handler-init(s, st-index,
   rtsp_st-dynamic_protocol_context);
 }
+if (rt-default_lang[0])
+av_dict_set(st-metadata, language, rt-default_lang, 0);
 }
 /* put a default control url */
 av_strlcpy(rtsp_st-control_url, rt-control_uri,
@@ -540,6 +542,16 @@ static void sdp_parse_line(AVFormatContext *s, 
SDPParseState *s1,
 /* AV_NOPTS_VALUE means live broadcast (and can't seek) */
 s-duration   = (end == AV_NOPTS_VALUE) ?
 AV_NOPTS_VALUE : end - start;
+} else if (av_strstart(p, lang:, p)) {
+if (s-nb_streams  0) {
+get_word(buf1, sizeof(buf1), p);
+rtsp_st = rt-rtsp_streams[rt-nb_rtsp_streams - 1];
+if (rtsp_st-stream_index = 0) {
+st = s-streams[rtsp_st-stream_index];
+av_dict_set(st-metadata, language, buf1, 0);
+}
+} else
+get_word(rt-default_lang, sizeof(rt-default_lang), p);
 } else if (av_strstart(p, IsRealDataType:integer;,p)) {
 if (atoi(p) == 1)
 rt-transport = RTSP_TRANSPORT_RDT;
diff --git a/libavformat/rtsp.h b/libavformat/rtsp.h
index a9f2f1e..3cfcfb6 100644
--- a/libavformat/rtsp.h
+++ b/libavformat/rtsp.h
@@ -405,6 +405,8 @@ typedef struct RTSPState {
  * User-Agent string
  */
 char *user_agent;
+
+char default_lang[4];
 } RTSPState;
 
 #define RTSP_FLAG_FILTER_SRC  0x1/** Filter incoming UDP packets -
-- 
1.8.4

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

[FFmpeg-devel] [PATCH 4/7] avcocdec/mpegaudio_parser: add MP3 ADU headers parser

2015-02-08 Thread Gilles Chanteperdrix

---
 libavcodec/mpegaudio_parser.c   | 14 --
 libavcodec/mpegaudiodecheader.c |  3 ++-
 2 files changed, 14 insertions(+), 3 deletions(-)

diff --git a/libavcodec/mpegaudio_parser.c b/libavcodec/mpegaudio_parser.c
index 79dbf63..42f4706 100644
--- a/libavcodec/mpegaudio_parser.c
+++ b/libavcodec/mpegaudio_parser.c
@@ -64,7 +64,7 @@ static int mpegaudio_parse(AVCodecParserContext *s1,
 }else{
 while(ibuf_size){
 int ret, sr, channels, bit_rate, frame_size;
-enum AVCodecID codec_id;
+enum AVCodecID codec_id = avctx-codec_id;
 
 state= (state8) + buf[i++];
 
@@ -90,6 +90,16 @@ static int mpegaudio_parse(AVCodecParserContext *s1,
 avctx-bit_rate += (bit_rate - avctx-bit_rate) / 
(s-header_count - header_threshold);
 }
 }
+
+if (s1-flags  PARSER_FLAG_COMPLETE_FRAMES) {
+s-frame_size = 0;
+next = buf_size;
+} else if (codec_id == AV_CODEC_ID_MP3ADU) {
+av_log(avctx, AV_LOG_ERROR,
+MP3ADU full parser not implemented);
+return AVERROR_PATCHWELCOME;
+}
+
 break;
 }
 }
@@ -110,7 +120,7 @@ static int mpegaudio_parse(AVCodecParserContext *s1,
 
 
 AVCodecParser ff_mpegaudio_parser = {
-.codec_ids  = { AV_CODEC_ID_MP1, AV_CODEC_ID_MP2, AV_CODEC_ID_MP3 },
+.codec_ids  = { AV_CODEC_ID_MP1, AV_CODEC_ID_MP2, AV_CODEC_ID_MP3, 
AV_CODEC_ID_MP3ADU },
 .priv_data_size = sizeof(MpegAudioParseContext),
 .parser_parse   = mpegaudio_parse,
 .parser_close   = ff_parse_close,
diff --git a/libavcodec/mpegaudiodecheader.c b/libavcodec/mpegaudiodecheader.c
index 5db1957..6af6e4b 100644
--- a/libavcodec/mpegaudiodecheader.c
+++ b/libavcodec/mpegaudiodecheader.c
@@ -134,7 +134,8 @@ int avpriv_mpa_decode_header2(uint32_t head, int 
*sample_rate, int *channels, in
 break;
 default:
 case 3:
-*codec_id = AV_CODEC_ID_MP3;
+if (*codec_id != AV_CODEC_ID_MP3ADU)
+*codec_id = AV_CODEC_ID_MP3;
 if (s-lsf)
 *frame_size = 576;
 else
-- 
1.8.4

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

[FFmpeg-devel] [PATCH 3/7] avformat/rtpdec: add T.140 RTP depacketization (RFC 4103)

2015-02-08 Thread Gilles Chanteperdrix

---
 libavformat/rtpdec.c | 7 +++
 1 file changed, 7 insertions(+)

diff --git a/libavformat/rtpdec.c b/libavformat/rtpdec.c
index f5557d8..4091bdc 100644
--- a/libavformat/rtpdec.c
+++ b/libavformat/rtpdec.c
@@ -56,6 +56,12 @@ static RTPDynamicProtocolHandler opus_dynamic_handler = {
 .codec_id   = AV_CODEC_ID_OPUS,
 };
 
+static RTPDynamicProtocolHandler ff_t140_dynamic_handler = {
+.enc_name   = t140,
+.codec_type = AVMEDIA_TYPE_SUBTITLE,
+.codec_id   = AV_CODEC_ID_SUBRIP,
+};
+
 static RTPDynamicProtocolHandler *rtp_first_dynamic_payload_handler = NULL;
 
 void ff_register_dynamic_payload_handler(RTPDynamicProtocolHandler *handler)
@@ -96,6 +102,7 @@ void ff_register_rtp_dynamic_payload_handlers(void)
 ff_register_dynamic_payload_handler(ff_quicktime_rtp_aud_handler);
 ff_register_dynamic_payload_handler(ff_quicktime_rtp_vid_handler);
 ff_register_dynamic_payload_handler(ff_svq3_dynamic_handler);
+ff_register_dynamic_payload_handler(ff_t140_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_theora_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_vorbis_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_vp8_dynamic_handler);
-- 
1.8.4

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

[FFmpeg-devel] [PATCH 6/7] avformat/rtpdec_mpeg4: reassemble fragmented AAC frames

2015-02-08 Thread Gilles Chanteperdrix

---
 libavformat/rtpdec_mpeg4.c | 84 +-
 1 file changed, 76 insertions(+), 8 deletions(-)

diff --git a/libavformat/rtpdec_mpeg4.c b/libavformat/rtpdec_mpeg4.c
index 53921f2..f791a5c 100644
--- a/libavformat/rtpdec_mpeg4.c
+++ b/libavformat/rtpdec_mpeg4.c
@@ -33,6 +33,8 @@
 #include libavutil/avstring.h
 #include libavcodec/get_bits.h
 
+#define MAX_AAC_HBR_FRAME_SIZE 8191
+
 /** Structure listing useful vars to parse RTP packet payload */
 struct PayloadContext {
 int sizelength;
@@ -59,8 +61,9 @@ struct PayloadContext {
 int au_headers_length_bytes;
 int cur_au_index;
 
-uint8_t buf[RTP_MAX_PACKET_LENGTH];
+uint8_t buf[FFMAX(RTP_MAX_PACKET_LENGTH, MAX_AAC_HBR_FRAME_SIZE)];
 int buf_pos, buf_size;
+uint32_t timestamp;
 };
 
 typedef struct {
@@ -168,30 +171,95 @@ static int aac_parse_packet(AVFormatContext *ctx, 
PayloadContext *data,
 {
 int ret;
 
+
 if (!buf) {
-if (data-cur_au_index  data-nb_au_headers)
+if (data-cur_au_index  data-nb_au_headers) {
+av_log(ctx, AV_LOG_ERROR, Invalid parser state\n);
 return AVERROR_INVALIDDATA;
-if (data-buf_size - data-buf_pos  
data-au_headers[data-cur_au_index].size)
+}
+if (data-buf_size - data-buf_pos  
data-au_headers[data-cur_au_index].size) {
+av_log(ctx, AV_LOG_ERROR, Invalid AU size\n);
 return AVERROR_INVALIDDATA;
-if ((ret = av_new_packet(pkt, 
data-au_headers[data-cur_au_index].size))  0)
+}
+if ((ret = av_new_packet(pkt, 
data-au_headers[data-cur_au_index].size))  0) {
+av_log(ctx, AV_LOG_ERROR, Out of memory\n);
 return ret;
+}
 memcpy(pkt-data, data-buf[data-buf_pos], 
data-au_headers[data-cur_au_index].size);
 data-buf_pos += data-au_headers[data-cur_au_index].size;
 pkt-stream_index = st-index;
 data-cur_au_index++;
-return data-cur_au_index  data-nb_au_headers;
+
+if (data-cur_au_index == data-nb_au_headers) {
+data-buf_pos = 0;
+return 0;
+}
+
+return 1;
 }
 
-if (rtp_parse_mp4_au(data, buf, len))
+if (rtp_parse_mp4_au(data, buf, len)) {
+av_log(ctx, AV_LOG_ERROR, Error parsing AU headers\n);
 return -1;
+}
 
 buf += data-au_headers_length_bytes + 2;
 len -= data-au_headers_length_bytes + 2;
+if (data-nb_au_headers == 1  len  data-au_headers[0].size) {
+/* Packet is fragmented */
+
+if (data-buf_pos == 0) {
+if (data-au_headers[0].size  MAX_AAC_HBR_FRAME_SIZE) {
+av_log(ctx, AV_LOG_ERROR, Invalid AU size\n);
+return AVERROR_INVALIDDATA;
+}
+
+data-buf_size = data-au_headers[0].size;
+data-timestamp = *timestamp;
+}
+
+if (data-timestamp != *timestamp
+|| data-au_headers[0].size != data-buf_size
+|| data-buf_pos + len  MAX_AAC_HBR_FRAME_SIZE) {
+data-buf_pos = 0;
+data-buf_size = 0;
+av_log(ctx, AV_LOG_ERROR, Invalid packet received\n);
+return AVERROR_INVALIDDATA;
+}
 
-if (len  data-au_headers[0].size)
+memcpy(data-buf[data-buf_pos], buf, len);
+data-buf_pos += len;
+
+if (!(flags  RTP_FLAG_MARKER))
+return AVERROR(EAGAIN);
+
+if (data-buf_pos != data-buf_size) {
+data-buf_pos = 0;
+av_log(ctx, AV_LOG_ERROR, Missed some packets, discarding 
frame\n);
+return AVERROR_INVALIDDATA;
+}
+
+data-buf_pos = 0;
+ret = av_new_packet(pkt, data-buf_size);
+if (ret  0) {
+av_log(ctx, AV_LOG_ERROR, Out of memory\n);
+return ret;
+}
+pkt-stream_index = st-index;
+
+memcpy(pkt-data, data-buf, data-buf_size);
+
+return 0;
+}
+
+if (len  data-au_headers[0].size) {
+av_log(ctx, AV_LOG_ERROR, First AU larger than packet size\n);
 return AVERROR_INVALIDDATA;
-if ((ret = av_new_packet(pkt, data-au_headers[0].size))  0)
+}
+if ((ret = av_new_packet(pkt, data-au_headers[0].size))  0) {
+av_log(ctx, AV_LOG_ERROR, Out of memory\n);
 return ret;
+}
 memcpy(pkt-data, buf, data-au_headers[0].size);
 len -= data-au_headers[0].size;
 buf += data-au_headers[0].size;
-- 
1.8.4

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

[FFmpeg-devel] [PATCH 5/7] avformat/rtpdec_mpeg12: add robust MPEG audio depacketization (RFC 3119)

2015-02-08 Thread Gilles Chanteperdrix

---
 libavformat/rtpdec.c |   1 +
 libavformat/rtpdec_formats.h |   1 +
 libavformat/rtpdec_mpeg12.c  | 201 +++
 3 files changed, 203 insertions(+)

diff --git a/libavformat/rtpdec.c b/libavformat/rtpdec.c
index 4091bdc..711a94d 100644
--- a/libavformat/rtpdec.c
+++ b/libavformat/rtpdec.c
@@ -90,6 +90,7 @@ void ff_register_rtp_dynamic_payload_handlers(void)
 ff_register_dynamic_payload_handler(ff_mp4a_latm_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_mp4v_es_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_mpeg_audio_dynamic_handler);
+ff_register_dynamic_payload_handler(ff_mpeg_audio_robust_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_mpeg_video_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_mpeg4_generic_dynamic_handler);
 ff_register_dynamic_payload_handler(ff_mpegts_dynamic_handler);
diff --git a/libavformat/rtpdec_formats.h b/libavformat/rtpdec_formats.h
index e5aff59..3cbd0aa 100644
--- a/libavformat/rtpdec_formats.h
+++ b/libavformat/rtpdec_formats.h
@@ -57,6 +57,7 @@ extern RTPDynamicProtocolHandler ff_jpeg_dynamic_handler;
 extern RTPDynamicProtocolHandler ff_mp4a_latm_dynamic_handler;
 extern RTPDynamicProtocolHandler ff_mp4v_es_dynamic_handler;
 extern RTPDynamicProtocolHandler ff_mpeg_audio_dynamic_handler;
+extern RTPDynamicProtocolHandler ff_mpeg_audio_robust_dynamic_handler;
 extern RTPDynamicProtocolHandler ff_mpeg_video_dynamic_handler;
 extern RTPDynamicProtocolHandler ff_mpeg4_generic_dynamic_handler;
 extern RTPDynamicProtocolHandler ff_mpegts_dynamic_handler;
diff --git a/libavformat/rtpdec_mpeg12.c b/libavformat/rtpdec_mpeg12.c
index d73ff1e..7d01ef8 100644
--- a/libavformat/rtpdec_mpeg12.c
+++ b/libavformat/rtpdec_mpeg12.c
@@ -71,3 +71,204 @@ RTPDynamicProtocolHandler ff_mpeg_video_dynamic_handler = {
 .parse_packet  = mpeg_parse_packet,
 .static_payload_id = 32,
 };
+
+/* MPA-ROBUST, RFC 3119 */
+struct PayloadContext {
+unsigned adu_size;
+unsigned cur_size;
+uint32_t timestamp;
+uint8_t *split_buf;
+int split_pos, split_buf_size, split_pkts;
+AVIOContext *fragment;
+};
+
+static av_cold int
+mpa_robust_init(AVFormatContext *ctx, int st_index, PayloadContext *data)
+{
+if (st_index  0)
+return 0;
+ctx-streams[st_index]-need_parsing = AVSTREAM_PARSE_HEADERS;
+return 0;
+}
+
+static PayloadContext *mpa_robust_new_context(void)
+{
+return av_mallocz(sizeof(PayloadContext));
+}
+
+static inline void free_fragment_if_needed(PayloadContext *data)
+{
+if (data-fragment) {
+uint8_t *p;
+avio_close_dyn_buf(data-fragment, p);
+av_free(p);
+data-fragment = NULL;
+}
+}
+
+static void mpa_robust_free_context(PayloadContext *data)
+{
+free_fragment_if_needed(data);
+av_free(data);
+}
+
+static int mpa_robust_parse_rtp_header(AVFormatContext *ctx,
+const uint8_t *buf, int len,
+unsigned *adu_size, unsigned *cont)
+{
+unsigned header_size;
+
+if (len  1) {
+av_log(ctx, AV_LOG_ERROR, Invalid %d bytes packet\n, len);
+return AVERROR_INVALIDDATA;
+}
+
+*cont = !!(buf[0]  0x80);
+if ((buf[0]  0x40) == 0) {
+header_size = 1;
+*adu_size = buf[0]  ~0xc0;
+} else {
+if (len == 1) {
+av_log(ctx, AV_LOG_ERROR, Invalid %d bytes packet\n, len);
+return AVERROR_INVALIDDATA;
+}
+header_size = 2;
+*adu_size = AV_RB16(buf)  ~0xc000;
+}
+
+return header_size;
+}
+
+static int
+mpa_robust_parse_packet(AVFormatContext *ctx, PayloadContext *data,
+AVStream *st, AVPacket *pkt, uint32_t *timestamp,
+const uint8_t *buf, int len, uint16_t seq,
+int flags)
+{
+unsigned adu_size, continuation;
+int err;
+
+if (!buf) {
+buf = data-split_buf[data-split_pos];
+len = data-split_buf_size - data-split_pos;
+
+err = mpa_robust_parse_rtp_header(ctx, buf, len, adu_size, 
continuation);
+if (err  0) {
+av_freep(data-split_buf);
+return err;
+}
+buf += err;
+len -= err;
+
+if (continuation != 0 || adu_size  len) {
+av_freep(data-split_buf);
+av_log(ctx, AV_LOG_ERROR, Invalid frame\n);
+return AVERROR_INVALIDDATA;
+}
+
+if (av_new_packet(pkt, adu_size)) {
+av_log(ctx, AV_LOG_ERROR, Out of memory.\n);
+return AVERROR(ENOMEM);
+}
+
+pkt-stream_index = st-index;
+memcpy(pkt-data, buf, adu_size);
+
+data-split_pos = (buf - data-split_buf) + adu_size;
+
+if (data-split_pos == data-split_buf_size) {
+av_freep(data-split_buf);
+return 0;
+}
+
+return 1;
+}
+
+
+err =

47 matches

Mail list logo