ix.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
ngineered shit we realize it is useless.
> Setting From: to be @ffmpeg.org seems to be the only practical
> solution. Sorry for the noise.
If we both arrive to the conclusion…
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- ff
oubtful.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
single logic to work for every kind of sinks.
Really, sinks have never worked reliably, they need to express their own
logic.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
Not cool.
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
atch.
> for https://lists.ffmpeg.org/mailman3/lists/ffmpeg-devel.ffmpeg.org/
> iam not sure where to change that
Alas, that would be the place where it is most useful.
Regards,
--
Nicolas George
>From 607f9bc14f46fbbb6f73b83f20d0688e5535a3b6 Mon Sep 17 00:00:00 2001
From: Nicolas Geor
ff --git a/libavformat/oggenc.c b/libavformat/oggenc.c
I would have hoped that whoever pushed this would have fixed the commit
message to include “lavf/oggenc:” at the beginning. Please be more
careful in the future.
Regards,
--
Nicolas George
___
ffmpeg-
d neg_step, num_conv = 0, num_mergers = 0;
>
> if (!link)
> continue;
> @@ -578,6 +580,8 @@ retry:
> conv_opts[num_conv] = FF_FIELD_AT(char *,
> m->conversion_opts_offset, *graph);
> num_conv
.
Too bad, DMARC killed that loophole.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
it.
There is a back-and-forth between coding and design.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
around hardware filters” is mostly
design.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
;
…
const double sample = fabs(samples_ch[nb_channels * i]);
Anyway, the fix looks good to me.
Niklas, it has become mostly your code, what do you make of it?
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- ffmpeg-devel@
ourselves to the decision means pushing the
patches. But for the design, we do not have such a mechanism, and that
is an issue.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to ffm
n.org” in the mails sent from our
mailing-lists: it will only work if their-domain.org has SPF and DMARC
policies lax enough, and these will vanish. The new scheme where the
header from is replaced with the address of the mailing-list seems
unavoidable.
Regards,
--
Nicolas George
___
nd it is much more convenient for
users.
> And not how we do it for other filters
We should.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
rs additional
> flexibility.
Then we need a patch series that really allows to do that, including
examples.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
s FFmpeg into a company in all but
legal status. That would be detrimental to our freedom to experiment,
and eventually to the quality of the code. Not counting being halfway
between two status, i.e. getting the problems of both.
Regards,
--
Nicolas George
_
uess
if the feature is useful or if there are better ways to achieve what you
want it to do.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
something to this extent.
Using the word “corporate” unironically about FFmpeg is a sure sign
somebody misunderstands something fundamental about the nature of this
project.
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To
t you aren't part of CC going forward.
This kind of victim-blaming is unacceptable. However bad you think
somebody's attitude is, expressing approval for a social media mob makes
you the bad guy.
--
Nicolas George
___
ffmpeg-devel mailing list
o similar entitlement expressed on users mailing-lists or bug
trackers, “patch welcome” and so on.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
ns, by implementing them in
libavutil. We can even gain a little more performance by making them
more specifically suited to our needs and style.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to
not know that. But if you limit “users” to people who use the
library directly, then obviously the majority will be from the languages
where it is possible to use the library directly. So what?
Regards,
--
Nicolas George
___
ffmpeg-devel mailing
base, although we choose to minimize the use of
> C++ inside FFmpeg itself.
Do you have evidence for that surprising statement? Intuitively, I would
say that the ultra-majority of our users do not any programming
language, or maybe some python on high school.
Regards,
--
n, while review happens in its time and while
requesting user comments and accepting incompatible changes in the
syntax.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
y it is useful or necessary?
Preferably in the commit message.
By the way, the first line of the commit message must start with
something like: “lavc/libcelt_dec:”.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- ffmpeg-devel@ffmpeg.
hing uses. If this API is
useful, then it probably can be made use of by one of the fftools, most
likely ffmpeg itself. Please submit a patch series that also adds this
use.
Also, please explain what this makes possible that was not already
possible with existing functions, in particular sim
we considered accepting, it should not be as an extra filter
with its own set of options. It should be as a single filter with an
option to choose the implementation but common options for the shared
features.
Regards,
--
Nicolas George
___
ffmpeg-devel
the patch. What is the benefit of this library over
librubberband. That last one has the merit of being included in major
distributions.
Also, please explain the deal with “Bungee Pro”: if it is a proprietary
variant, I am not sure we want it, and it will require --enable-nonfree
at least.
Regards,
ichael.
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
Nicolas George via ffmpeg-devel (HE12025-10-13):
> Thanks. It sucks a little less than the first impression. But it still
> sucks. The fact that it defaults to a non-monospace font, for starters,
> and tries to “pretty print” the mail is terrible for our use.
Also: threading. For exam
Nicolas George via ffmpeg-devel (HE12025-10-17):
> > URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20715
My answer did not reach there. The promised ability to reply by mail is
still missing.
Regards,
--
Nicolas George
___
ffmpeg-devel m
age of the
list?
> https://www.mail-archive.com/[email protected]/
This is good too, but it is not ours.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
es.
Focus on making good code state of the art and let people who want
stability work for it.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
t", "volume", NULL, 0, 0);
>
> Signed-off-by: YiboFang
Hi. The file option seems to be documented but not implemented, it is
still using a hard-coded default directory, and a problematic one. And
you have not answered Andreas's question about using asplit.
Regards,
--
themselves and their sponsor rather than
to the project.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
less than the first impression. But it still
sucks. The fact that it defaults to a non-monospace font, for starters,
and tries to “pretty print” the mail is terrible for our use.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- ffm
icolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
cceptable spam threshold.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
t; it from a different computer or just see if CI passes or something.
If this web monster does not let you push code in your private fork
without asking to include it upstream, it is one more reason not to use
that thing.
PS: please do not bypass
egards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
a perfectly normal way of managing the
project.
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
> There is one, but OK, I will make the existing comment a bit more verbose.
Thanks.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list -- [email protected]
To unsubscribe send an email to [email protected]
the users. If a filter is wrongly assumed to not work, the system will
just insert conversion filters and waste a few cycles.
One is unacceptable, the other is negligible. The default should be to
choose the issue with negligible drawbacks.
Regards,
--
Nicolas George
rgii Zagoruiko
On Sat, Aug 30, 2025, 10:55 PM Martin Storsjö wrote:
> On Sat, 30 Aug 2025, George Zaguri wrote:
>
> > You mentioned submition of patch as a PR, you mean open merge request I
> > guess, how it works with mailing list? Do I need to update the patch in
> > ma
, you mean open merge request I
guess, how it works with mailing list? Do I need to update the patch in
mailing list as well? Should I reflect opened PR somewhere?
All the best,
Georgii Zagoruiko
On Sun, Aug 3, 2025 at 6:54 PM George Zaguri
wrote:
> Thank you, Martin! I will have a look at
-142,6 +119,14 @@ void av_vbprintf(AVBPrint *buf, const char *fmt, va_list
> vl_arg)
> av_bprint_grow(buf, extra_len);
> }
>
> +void av_bprintf(AVBPrint *buf, const char *fmt, ...)
> +{
> +va_list vl;
> + va_start(vl, fmt);
> +av_vbprintf(buf, fmt, vl);
> +va_end(
es not tell us how much room it would need: let us
> retry with twice as much until the buffer is large enough */
> -room = !room ? strlen(fmt) + 1 :
> +room = !room ? fmt_len + 1 :
> room <= INT_MAX / 2 ? room * 2 : INT_MAX;
> if (av_bpri
there is no doubt about, and
much better than the quick and dirty version I sent, and it is a good
thing you did it. But it was not on my demand and it does not change the
fact that support must be opt-in.)
Regards,
--
Nicolas George
___
ffmpeg-devel ma
I do not know the code and my assembly is very rusty.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@f
t values to the range [0,
> 1].",
> + OFFSET(clamplutoutput), AV_OPT_TYPE_BOOL, { .i64 = 0},
> + 0, 1, FLAGS },
> +
> { "iall", "Set all input color properties together",
>OFFSET(user_iall), AV_OPT_TYPE_INT, { .i64 = CS_UNSPECIFIE
kind of “bugs” too.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
[email protected] with subject "unsubscribe".
dded support for unknown channel layouts while keeping
API compatibility with the forks: filters that had not been explicitly
vetted for support were assumed to not support them. Now like before, it
is the only reasonable way to do things.
--
Nicolas George
knows the code only
tangentially and could have missed subtle issues (⇒ basis for waiting a
little less before pushing, but not immediate) from an approval by a
newbie who overestimate their own competence (we had a few of those in
the past).
--
Nicolas
ave, or for somebody who exchanged mails with me
recently on the very subject of libavfilter.
> Having a clearly defined process is a human solution to a human problem.
A bot is not a human solution.
--
Nicolas George
___
ffmpeg-devel maili
ay, and will
therefore forgive the ad-hominem attacks at the end of this mail.
I will not apologize for the miscommunication caused by the awful tool
that is being pushed on us.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmp
it is not the main objective, and chasing two hares is a
recipe for catching zero.
I have quite a lot to answer to the technical points you made below and
that I will now snip, but before that I think it is important we check
we are on the same page about the goals.
Regards,
--
Nicolas George
___
t;
fi
git pull --no-rebase --log --stat --commit --no-edit "$1" "$branch"
Also, better not forget quoting shell variables, even if you trust the
caller to avoid dangerous characters.
> }
>
> unset succeeded failed version
Not judging on the merits of havi
r all.
With this system, we would implement whitelists by adding only trusted
components to the AVLibrary instance. And people can add custom
components because the lists are mutable.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmp
ion 1.
The difference between options 2-5 is just window dressing: the code is
the same, the only difference is whether the plugins are listed on the
static website, the wiki or nowhere under our control.
Regards,
--
Nicolas George
___
ffmpeg-dev
Jacob Lifshay (HE12025-08-12):
> framework is misspelled here
Thanks, locally fixed.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link ab
I have been maintaining the framework and orphaned filters
for ages but apparently it was never recorded in this file.
Signed-off-by: Nicolas George
---
MAINTAINERS | 4
1 file changed, 4 insertions(+)
… but it was common knowledge.
diff --git a/MAINTAINERS b/MAINTAINERS
index
will not solve human problems.
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
[email protected] with subject "unsubscribe".
t. The only approval worth something comes
from somebody who knows the code.
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg
or no forge, and a spit in my face for
the time I invested into starting to look at the code, thankfully
limited.
Never do that again.
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmp
Nicolas George (HE12025-08-11):
> I do not like it either, especially because it weakens the intent to
> bring libpostproc back properly, but Michael made the decision and the
> time to object it was before
> “12b853530a Add tools/merge-all-source-plugins”.
I see it has been pushed
ks or features.
I do not like it either, especially because it weakens the intent to
bring libpostproc back properly, but Michael made the decision and the
time to object it was before
“12b853530a Add tools/merge-all-source-plugins”.
Regards,
--
Nicolas George
_
ogus software.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
[email protected] with subject "unsubscribe".
the
format, then we should not either.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
[email protected] w
e to specify “-f mp3”.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
[email protected] with subject "unsubscribe".
users from creating
invalid output.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
[email protected] with subject "unsubscribe".
explain better what the commit does.
What other software manages to recognize these files as valid?
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, vi
atches because that's not what email was intended for. Git didn't
> even exist then.
Any skill need to be updated.
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-d
Jorge Estrada (HE12025-08-08):
> As recommended by Nicholas George. Uses refcounting for planar formats
> similar to this change for alphamerge
I'd prefer the first sentence to be left out. And if not, then please
fix the typo in my name ;)
> https://patchwork.ffmpeg.org/projec
;
> @@ -212,4 +222,4 @@ const FFFilter ff_vf_alphamerge = {
> FILTER_QUERY_FUNC2(query_formats),
> .uninit = uninit,
> .activate = activate,
> -};
> +};
> \ No newline at end of file
You really need to fix you editor to not have it remove the final
n
icolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
[email protected] with subject "unsubscribe".
onal inbox a mail from a member of one of the top
computer science research teams in the world complaining that GitLab
errors with an abstruse message about corrupted commits. Mature indeed.
--
Nicolas George
___
ffmpeg-devel mailing list
ffmpeg-deve
re broken since about four days, and suddenly we got a
new notification, and it was your patch. So I assumed somebody was
working at fixing the issue, and I answered to give thanks but report it
was still not perfect.
Regards,
--
Nicolas George
___
ffm
Wu Jianhua (HE12025-08-08):
> Yes I should separte them to different subjects. I will pay attention next
> time.
I do not understand: did you send this mail yourself manually?
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
ffmpeg
That is better, thanks, but the fact that it combines multiple
completely unrelated commits in one mail with the subject of only one of
the commit is still not good.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https:/
,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
[email protected] with subject "unsubscribe".
Nicolas George (HE12025-08-08):
> Sorry, I thought it was Paul's code and you were waiting for review. In
> the process of applying.
Pushed.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.
view. In
the process of applying.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
[email protected] with subject "unsubscribe".
.
So no objection to the patch, just take half a second considering my
suggestion.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link abov
at patch is
what introduced the issue. It is quite obvious in that case with a
cosmetic commit.
Regards,
--
Nicolas George
1: Strange construction, but apparently idiomatic.
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman
t;
> rtp_ctx->pb = avio_alloc_context(buffer, buffer_size, 1, s, NULL,
> on_rtp_write_packet, NULL);
> if (!rtp_ctx->pb) {
> +av_freep(&buffer);
> ret = AVERROR(ENOMEM);
> goto end;
That belong to after end, otherwise yo
Jiasheng Jiang (HE12025-08-07):
> Add check for the return value of av_mallocz() to avoid potential NULL
> pointer dereference.
LGTM, thanks.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmp
s still in test.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
[email protected] with subject "unsubscribe".
never spend five minutes learning how to use mail?
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
[email protected] with s
; it to fail to compile on certain platforms with --assert-level=2. We
> fix this by #including stdbool.h to declare false as a literal.
Better replace the av_assert2() with av_unreachable().
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
this one.
These are rather low-hanging fruits.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg
e
Still that idiotic idea that attracting more newbies will solve the
issue that we do not have enough senior developers.
> to improve productivity.
The current bullshit decreases the productivity of the most productive
member of the project.
--
Nicolas George
__
blic headers and creates a
binary-incompatible version of the library.
And IIRC you cannot use CONFIG_SOMETHING in public headers anyway.
Better leave the function as stubs in the libraries.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing l
Nicolas George (HE12025-08-07):
> > URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20154
> > Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20154.patch
> I do not like that it bloats the public headers and creates a
> binary-incompatible version of the library.
>
hing, saving him
>time
>
> This reverts commit cc6ad703b41e318bd1e4cb6196defaf505dbfea3.
LGTM.
We certainly are not at the point where policies affecting the whole
project can be enacted without discussion on the mailing-list.
Regards,
--
Nicolas George
d, 5 insertions(+), 1 deletion(-)
Still looks good to me, of course, thanks.
I do not maintain the docs.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscr
This one looks good to me, thanks.
> Fixes: f5df897c4b ("examples/avcodec: split audio decoding into a separate
> example")
???
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mail
exit(1); /* or proper cleanup and returning */
And in any case, the commit message should inaccurately pretend to fix
possible leaks.
Regards,
--
Nicolas George
___
ffmpeg-devel mailing list
[email protected]
https://ffmpeg.org/mailman/li
1 - 100 of 1470 matches
Mail list logo
