,
with automatic consecutive values, and use tables whenever the values
for 23001-8_2013 are needed.
The fact that the values currently align with 23001-8_2013 is of course
not a problem, but keeping them that way is problematic.
Regards,
--
Nicolas George
signature.asc
Description: D
way?
I can see the benefit of organizing code in separate directories, of
course, but separating the libraries only bring headaches to users and
developers. What am I missing?
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
__
t sees all the dependencies, it does
not rely on recursive make, and thus does the almost minimal amount of
work every time and works in parallel flawlessly. With split projects,
we lose that.
Regards,
--
Nicolas George
___
libav-devel mailing list
el, and place the actual library and cli source files inside a
> src/ folder.
And that is awfully annoying, always one more step to access the
interesting files, for no good reason.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
ror during scaling: %d\n", err);
Does the numeric value of the error code have any meaning for users? If
not, I think the code needs to be translated into a human-readable
string.
> ret = AVERROR_UNKNOWN;
> goto fail;
> }
Regards,
--
Nicolas George
signatu
always going to require reading the source code.
I think an error message or any other kind of information that is never
of use to simple users does not belong in error messages above the
"verbose" level.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
_
the error should clearly state that it is a BUG, and suggest
the user to report it, along with diagnostics relevant for the
developers obtained at loglevel debug.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-deve
it is even worse.
Try to put yourself in the shoes of a completely clueless user: does
this error message tell you anything useful at all?
To get more relevant advice, I believe you need to explain, here and
now, what causes the error: user settings, system/hardware, or bug.
Regards,
--
Nico
al error code, if useful, does not belong in a message
at AV_LOG_ERROR message but at AV_LOG_DEBUG.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
L'octidi 8 ventôse, an CCXXV, Mark Thompson a écrit :
> I have no idea
Then the user have even less idea: the error code belongs at
AV_LOG_DEBUG, not AV_LOG_ERROR.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
_
(and lavu has several levels of asserts for
that).
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
our application to CRASH RIGHT NOW, not allow
remote code execution.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
ue."
What part of this would you want to refute?
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
lanned tonight. If you are still not convinced, I can self-caricature the
anti-constructive rivality: "Please, replace all asserts by untested dead
code; please make avconv more bloated and less secure. We will just fix it
back in FFmpeg."
Regards,
Le decadi 10 germinal, an CCXXIII, Vittorio Giovara a écrit :
> also OSS is named oss_audio, should this be renamed alsa_audio_dec for
> consistency?
I concur. It could also be just "alsa", since the name is completely
distinctive ("oss" is quite ambiguous).
Rega
replace every one of them by a
pedestrian version?
And pointers! Pointers are very confusing for people not familiar with them.
Let us eliminate them from the code base completely.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
_
e to express and force
on others ideas on readability and validity about a language that I do not
actually know.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-devel mailing list
libav-devel@libav.org
https://lists.
a "fair, productive
environment for developers and contributors" does not apply to me, so please
disregard my advice.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-devel mailing list
libav-devel@libav.or
r proprietary extension. I advise you not to use proprietary
extensions.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
Hi. I suspect that this bug affects you too, so in the spirit of cooperation
here is the mail I sent about to ffmpeg-devel.
- Forwarded message from Nicolas George -
Reply-To: FFmpeg development discussions and patches
Date: Wed, 6 May 2015 11:55:25 +0200
From: Nicolas George
To
I
more fragile.
The convention chosen in FFmpeg is: if 0/0, that means the filter did not
bother to set it, then assume the simplest case and let the framework set
the value as the same as the input.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
_
parate project, the practice has
a number of drawbacks.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
r a context is opened or
closed, then there is something deeply wrong in its design.
Sprinkling FFmpeg's code to work with random tests to work around broken
application designs is not good for maintainability.
Regards,
--
Nicolas George
signature.asc
Description: Digital s
Le primidi 11 frimaire, an CCXXIII, Nicolas George a écrit :
> Sprinkling FFmpeg's code
Sorry, I was a bit careless when redacting the code; I was not trying to
troll.
I stand with the core of my statement though.
Regards,
--
Nicolas George
signature.asc
Description: Digital s
rce_id:%u minor_code:%u major_code:%u.\n",
> + e->response_type, e->error_code,
> + e->sequence, e->resource_id, e->minor_code, e->major_code);
> +return AVERROR(EACCES);
EACCES is "Permission denied", this is proba
ubtitle frame at pts=5, duplicate it to
pts=5.99, it should allow processing the video frame at pts=5.
Another solution for CFR would be to add some kind of "min_frame_duration"
option to the framesync utility.
Hope this helps.
Regards,
-
of them, preferably "x11grab-xcb" if it is
available. I believe that is what I would have done.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
> Subject: Re: [libav-devel] [PATCH 2/3] [RFC] xcbgrab: Add "xcbgrab" as input
> device name alias for "x11grab_xcb"
The name of the device is not "x11grab_xcb".
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
_
->params.i_bitdepth
tells us what x264 proposes; they are not necessarily the same thing.
You may want to check the similar change you made on FFmpeg.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
li
f it is just a matter of the
name "pipe" suggesting it is not seekable, I do not think it warrants a
completely new protocol.
Also, the doc is missing.
Regards,
--
Nicolas George
signature.asc
Description: PGP signature
___
libav-devel mai
sted in subtitles, and they usually reject patches
that are "not the good long-term solution but immediately useful to users in
the meantime"; or ask for the good long-term solution, which has the same
result.
Regards,
--
Nicolas George
signatu
Le tridi 23 thermidor, an CCXXI, Nicolas George a écrit :
> Please forgive me for replying in private, but I am not very well liked on
> this mailing-list.
This obviously should not have arrived here. I will make public excuses if
anyone has been offended by parts of this message. I am so
Le quartidi 14 messidor, an CCXX, Kostya Shishkov a écrit :
> also maybe put it into loop so it can crypt more than 8 bytes at once?
You are not suggesting to design an API that would encourage people to use
ECB mode, are you?
Regards,
--
Nicolas George
signature.asc
Description: Digi
lly handle various block ciphers.
Providing a public API that implements ECB will encourage people with no
cryptography background to use it and produce flawed protocols, while it
will not help anyone serious since ECB is mostly useless.
Regards,
--
Nicolas George
signature.asc
Descript
Le quintidi 15 messidor, an CCXX, Kostya Shishkov a écrit :
> We have one compromised protocol (RTMPE) right here to support!
Well, if RTMPE really uses ECB, then the loop can go in the RTMPE
implementation rather than the public API of good block ciphers.
Regards,
--
Nicolas Geo
eg.git;a=commit;h=a9e7e98021ca9f2eba260c6a6f22a9a476332b24
And it is not just a warning, it is a real bug when the error code is
immediately cast to int64_t, like in the seek functions for example.
Feel free to merge the commit, of course.
Regards,
--
Nicolas George
signature.asc
Descr
Le septidi 27 thermidor, an CCXX, Måns Rullgård a écrit :
> That is not a fix.
Yes it is, unlike the thing you posted at 11:18:48 +0100. I could take the
time to explain why your answer is completely wrong, but considering the way
you addressed me, I do not feel like it.
Regards,
--
Nico
asts just to silence
compiler warnings." Nothing offensive.
And it was not the first time.
My purpose is not to pollute your mailing list, I will not reply again in
this thread.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
Le quartidi 24 frimaire, an CCXX, Luca Barbato a écrit :
> "Bug detected, please report the issue" ?
Then AVERROR_BUG is maybe a better name, don't you think?
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
_
Purely formal review, I do not know about stereoscopic metadata.
Le tridi 23 ventôse, an CCXXI, Vittorio Giovara a écrit :
> ---
> libavcodec/avcodec.h|3 +++
> libavcodec/options.c|2 ++
> libavutil/Makefile |2 ++
> libavutil/stereoscopy.c | 11 +++
> libavuti
idea how the microsoft
implementations behave though.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
val = avfilter_poll_frame(link->src->inputs[0]);
> +return poll_frame(link);
Are you sure this will not recurse too deep?
> }
> assert(yadif->next || !val);
Also, is it possible to tell Gmail to use text/plain for patches?
Regards,
--
Nicolas George
s
.start_frame = start_frame,
> +.end_frame = end_frame, },
> + { .name = NULL}},
> +.outputs = (AVFilterPad[]) {{ .name= "default",
> +.type= AVMEDIA_TYPE_VIDEO,
> +.request_frame = request_frame,
> +.config_props= out_config_props},
> + { .name = NULL}},
No permissions mask? I believe we need REUSE at least.
> +};
> +
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
compensation) if the user needs it.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
and resubmit them.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
This patch also introduces CODEC_ID_CELT.
Signed-off-by: Nicolas George
---
libavcodec/avcodec.h |1 +
libavformat/Makefile |1 +
libavformat/oggdec.c |1 +
libavformat/oggdec.h |1 +
libavformat/oggparsecelt.c | 99
there was an unresolved discussion about the API of framed
codecs, so I did not write the encoder side. I did not yet had time to come
back to it.
As for Opus, I fear the specs of the Ogg mapping and the RTP packetization
may be still too unstable yet.
Regards,
--
Nicolas George
signatu
This patch also introduces CODEC_ID_CELT.
Signed-off-by: Nicolas George
---
libavcodec/avcodec.h |1 +
libavcodec/version.h |2 +-
libavformat/Makefile |1 +
libavformat/oggdec.c |1 +
libavformat/oggdec.h |1 +
libavformat/oggparsecelt.c | 99
This patch also introduces CODEC_ID_CELT.
Signed-off-by: Nicolas George
---
libavcodec/avcodec.h |1 +
libavcodec/version.h |2 +-
libavformat/Makefile |1 +
libavformat/oggdec.c |1 +
libavformat/oggdec.h |1 +
libavformat/oggparsecelt.c | 99
ut it around 2011-03-06 for another
reason, but I do not think anything came of it yet.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
This patch includes some fixes by Carl Eugen Hoyos.
Signed-off-by: Nicolas George
---
configure|5 ++
doc/general.texi |2 +
libavcodec/Makefile |1 +
libavcodec/allcodecs.c |1 +
libavcodec/libcelt_dec.c | 137
codec, and that is where you will find (or not) news and roadmaps.
Thus, a native decoder would not be really useful, and the same goes for
encoding with libcelt. On the other hand some files may already float
around, and basic decoding may still be useful.
Regards,
--
Nicolas
f (
Done.
> Align the last '=' as well.
It would go beyond 80 columns, and I did not find any combination that does
not make it look worse than the current version.
New patch incoming.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
_
This patch includes some fixes by Carl Eugen Hoyos.
Signed-off-by: Nicolas George
---
Changelog|1 +
configure|5 ++
doc/general.texi |2 +
libavcodec/Makefile |1 +
libavcodec/allcodecs.c |1 +
libavcodec/libcelt_dec.c | 138
It would be rather dumb not to support it.
> Do you mean OS X or not? If yes, which version?
Someone else has answered to that point.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
___
libav-devel mailing list
libav-devel@l
This patch includes some fixes by Carl Eugen Hoyos.
Signed-off-by: Nicolas George
---
Changelog|1 +
configure|5 ++
doc/general.texi |2 +
libavcodec/Makefile |1 +
libavcodec/allcodecs.c |1 +
libavcodec/libcelt_dec.c | 138
56 matches
Mail list logo
