Carl Eugen Hoyos | 1 Aug 01:05 2011
Picon

Re: [PATCH]Fix ffplay -s -pix_fmt

Stefano Sabatini <stefano.sabatini-lala <at> poste.it> writes:

> > Attached patch fixes Ticket 347, I will apply if nobody objects.

> This option has
> been deprecated in favour of private options, see for example
> the video_size option in the rawvideo format.

Pushed a hopefully acceptable variant.

Thank you, Carl Eugen
Carl Eugen Hoyos | 1 Aug 01:36 2011
Picon

Re: [PATCH]Remove non-existing fourcc " Y8"

Carl Eugen Hoyos <cehoyos <at> ag.or.at> writes:

> Hi!
> 
> Attached patch reverts a line from r1842 (!) that I believe is not correct.

Pushed.

Carl Eugen
compn | 1 Aug 01:43 2011
Picon

Re: License violation

On Mon, 1 Aug 2011 01:09:36 +0400, Max Lapshin wrote:
>Will anybody reply on sales <at> ffmpeg.org if I want to buy ffmpeg under
>non-gpl license?
>

no, there isnt any sales <at> ffmpeg email as far as i know.

to answer your other question, ffmpeg is made from many people's
contributions and it would be very difficult to license ffmpeg under
any non-gpl license.

in short: if you want to distribute ffmpeg, just follow the gpl/lgpl
rules. the easiest thing to do is have some link on your website with
the source you used to create your ffmpeg binary.

-compn
compn | 1 Aug 01:51 2011
Picon

Re: License violation

On Sun, 31 Jul 2011 13:51:49 -0700, Howard Chu wrote:
>http://www.tubedigger.com/index.html
>
>Their package includes an ffmpeg.exe

CC'd to ronald bultje (ffmpeg/libav sflc contact)
Max Lapshin | 1 Aug 06:24 2011
Picon

Re: License violation

On Mon, Aug 1, 2011 at 3:43 AM, compn <tempn <at> twmi.rr.com> wrote:
> On Mon, 1 Aug 2011 01:09:36 +0400, Max Lapshin wrote:
>>Will anybody reply on sales <at> ffmpeg.org if I want to buy ffmpeg under
>>non-gpl license?
>>
>
> no, there isnt any sales <at> ffmpeg email as far as i know.
>
> to answer your other question, ffmpeg is made from many people's
> contributions and it would be very difficult to license ffmpeg under
> any non-gpl license.
>
> in short: if you want to distribute ffmpeg, just follow the gpl/lgpl
> rules. the easiest thing to do is have some link on your website with
> the source you used to create your ffmpeg binary.
>

This is a problem. Maybe that guy wanted to share his profit with
programmers, but nobody left him any chance and that is
why he decided to violate license.

libx264 guys keep ownership of code in one hands, so it becomes
possible to obtain libx264 under non-gpl and compile
ffmpeg in lgpl mode, which makes it possible to use in commercial app.
Michael Niedermayer | 1 Aug 08:52 2011
Picon
Picon

Re: [PATCH] Extend -xerror in ffmpeg.c to react to codec errors

sorry for my late awnser

On Wed, Jul 06, 2011 at 06:46:43PM -0400, yeratrin wrote:
> On 7/2/2011 12:02 AM, yeratrin wrote:
> > This seems to be consistent with the already-extant documentation
> > to "exit on error" and merely covers a missing case. Currently it
> > is possible for the ffmpeg console to show lots of "error"s and
> > xerror to do nothing.
> 
> Is this patch (same as in referred-to email) at least potentially
> reasonable or is it missing something critical?

the idea is fine if this kind of thing is usefull to someone

[...]
>  
> +static void log_callback_errorfilter(void* ptr, int level, const char* fmt, va_list vl)
> +{
> +    av_log_default_callback(ptr, level, fmt, vl);
> +
> +    /* level == AV_LOG_FATAL already caught */
> +    if(exit_on_error && level == AV_LOG_ERROR)
> +        ffmpeg_exit(1);

there are more codes between ERROR AND FATAL
also it ignores parent_log_context_offset, which means it can exit
out on things that would be presented as low as DEBUG or INFO

also i think this should be enabled seperately from -xerror.
a new cmdline option would also allow passing the log level at which
(Continue reading)

Michael Niedermayer | 1 Aug 09:01 2011
Picon
Picon

Re: [PATCH] pixdesc: remove PIX_FMT_FLAG flag from pixel formats which are *not* paletted

On Thu, Jul 28, 2011 at 01:27:31PM +0200, Stefano Sabatini wrote:
> The flag was required for an hack in libswscale. Since now libswscale
> distinguishes between formats which store palette in the image, and
> formats which use it internally in libswscale, the flag is no longer
> required.

treating 8 bit per pixel formats like paletted formats is not only
usefull in swscale. But also can be usefull in other filters, like
contrast/brightness/gamma filters. For them for example its just needed
to apply the change to the palette and then switch the pix format to
PAL8

of course that can be done  without the flag too but the flag 
might lead to slightly simpler code somewhere, not a major thing though

[...]
--

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Old school: Use the lowest level language in which you can solve the problem
            conveniently.
New school: Use the highest level language in which the latest supercomputer
            can solve the problem without the user falling asleep waiting.
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel <at> ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
(Continue reading)

Michael Niedermayer | 1 Aug 11:29 2011
Picon
Picon

Re: [PATCH]Set bits_per_coded_sample when encoding MS ADPCM

On Mon, Jul 11, 2011 at 01:35:39AM +0200, Carl Eugen Hoyos wrote:
> Hi!
> 
> The value is read on decoding, so imo, it makes sense to set it on encoding.
> 
> Please comment, Carl Eugen

>  adpcm.c |    2 ++
>  1 file changed, 2 insertions(+)
> 99073da0c03c88daefa396cf0132a06b61a5bd9d  patchadpcmbitspersample.diff

LGTM

[...]

--

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Why not whip the teacher when the pupil misbehaves? -- Diogenes of Sinope
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel <at> ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Clément Bœsch | 1 Aug 11:32 2011
Picon

Re: Improve timecode support

On Sun, Jul 31, 2011 at 01:51:10AM +0200, Michael Niedermayer wrote:
[...]
> Does this support keeping timecodes when transcoding?

I'm afraid not, you indeed need to use the -timecode option for the
output.

> To elaborate a bit on the question
> 
> With private options holding the timecode, transcoding could become
> tricky if for example the values where to be interpreted differently
> for 2 codecs.
> Iam not asking for transcoding support to be added just if there are
> any expected obstacles if one wanted to add it and if changes to the
> design could be made to avoid that.
> 

You mean if timecode is muxed differently in the formats, it won't be easy
to use a common way of transmitting the timecode value between source
demuxer and destination muxer when transcoding?

I guess I completely ignored that part. If we need to transmit customized
crafted data between the demuxer and the muxer, what can are the different
ways to achieve that? Should we use the AVPacket side data?

Also, the mpeg encoder might be a specific case; we are mostly dealing
with formats with the timecode (right now mxf and dv, maybe gxf later),
but there is the exception of the mpeg encoder, see patch 2).

After a discussion on IRC, it seems there is no obvious way to do that,
(Continue reading)

Mina Nagy Zaki | 1 Aug 10:40 2011
Picon

Re: [ffmpeg-devel] [PATCH 2/4] lavfi: add asink_abuffer - audio sink buffer filter

On Wed, Jul 27, 2011 at 12:04:35PM +0200, Stefano Sabatini wrote:
> On date Wednesday 2011-07-27 01:05:18 +0300, Mina Nagy Zaki encoded:
[...]

Fixed all problems.

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel <at> ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Gmane