On Wed, Dec 14, 2016 at 08:46:28AM +0100, Paul B Mahol wrote:
> On 12/14/16, Michael Niedermayer wrote:
> > On Tue, Dec 13, 2016 at 10:23:48PM +0100, Martin Vignali wrote:
> >> 2016-11-24 21:35 GMT+01:00 Martin Vignali :
> >>
> >> > Hello
> >> >
> >> > New patchs in attach
> >> >
> >> > I changed
On 12/14/16, Michael Niedermayer wrote:
> On Tue, Dec 13, 2016 at 10:23:48PM +0100, Martin Vignali wrote:
>> 2016-11-24 21:35 GMT+01:00 Martin Vignali :
>>
>> > Hello
>> >
>> > New patchs in attach
>> >
>> > I changed the dimensions check, the check is now :
>> > --
On Tue, Dec 13, 2016 at 10:23:48PM +0100, Martin Vignali wrote:
> 2016-11-24 21:35 GMT+01:00 Martin Vignali :
>
> > Hello
> >
> > New patchs in attach
> >
> > I changed the dimensions check, the check is now :
> >
> > if ((s->height > 3) && (s->avctx->s
2016-11-24 21:35 GMT+01:00 Martin Vignali :
> Hello
>
> New patchs in attach
>
> I changed the dimensions check, the check is now :
>
> if ((s->height > 3) && (s->avctx->strict_std_compliance >
> FF_COMPLIANCE_EXPERIMENTAL)) {
> av_log(s->avctx,
Hello
>
> If you want to maintain this code, then please add yourself to the
> MAINTAINER file too
>
I'm not very familiar with git. So it's probably not a good idea to be a
maintainer.
But i can take a look for patch on psd (and exr) if need.
Martin
_
On Thu, Nov 24, 2016 at 09:35:42PM +0100, Martin Vignali wrote:
> Hello
>
> New patchs in attach
>
> I changed the dimensions check, the check is now :
>
> if ((s->height > 3) && (s->avctx->strict_std_compliance >
> FF_COMPLIANCE_EXPERIMENTAL)) {
>
Hello
New patchs in attach
I changed the dimensions check, the check is now :
if ((s->height > 3) && (s->avctx->strict_std_compliance >
FF_COMPLIANCE_EXPERIMENTAL)) {
av_log(s->avctx, AV_LOG_ERROR,
"Height > 3 is experimental
On Tue, Nov 22, 2016 at 12:15:16AM +0100, Martin Vignali wrote:
> Hello,
>
> 2016-11-21 23:16 GMT+01:00 Moritz Barsnick :
>
> > On Mon, Nov 21, 2016 at 21:44:55 +0100, Martin Vignali wrote:
> > > +avpriv_request_sample(s->avctx, "ZIP without predictor
> > compression");
> > > +ret
On Tue, 22 Nov 2016 01:21:08 +0100, Carl Eugen Hoyos wrote:
> If an application writes psd files with width or height > 3 (or,
> horribile dictu, Adobe changes the limits), we want to decode
> such images (unless strict strict was set).
We shouldn't be encouraging theoretical random broken cr
2016-11-22 0:57 GMT+01:00 Andreas Cadhalpun :
> On 20.11.2016 17:57, Martin Vignali wrote:
+s->height = bytestream2_get_be32(&s->gb);
+
+if ((s->height < 1) || (s->height > 3)) {
>>>
>>> Why 3?
>>> ff_set_dimensions already checks for sane dimensions.
>>>
>>
>> Follo
On 20.11.2016 17:57, Martin Vignali wrote:
>>> Support uncompress and rle compression in Image Data Section.
>> ^
>> decompression
>>
>
> Not sure i understand, do you mean :
> Support uncompress and rle decompression in Image Data Section.
> ?
> Because currently, the psd reader, can ma
Hello,
2016-11-21 23:16 GMT+01:00 Moritz Barsnick :
> On Mon, Nov 21, 2016 at 21:44:55 +0100, Martin Vignali wrote:
> > +avpriv_request_sample(s->avctx, "ZIP without predictor
> compression");
> > +return AVERROR_PATCHWELCOME;
> > +break;
> > +case 3:
> > +avpr
On Mon, Nov 21, 2016 at 21:44:55 +0100, Martin Vignali wrote:
> +avpriv_request_sample(s->avctx, "ZIP without predictor compression");
> +return AVERROR_PATCHWELCOME;
> +break;
> +case 3:
> +avpriv_request_sample(s->avctx, "ZIP with predictor compression");
> +
On 21 November 2016 at 21:47, Ronald S. Bultje wrote:
> Hi,
>
> On Mon, Nov 21, 2016 at 4:40 PM, Rostislav Pehlivanov >
> wrote:
>
> > On 21 November 2016 at 20:44, Martin Vignali
> > wrote:
> >
> > > Hello,
> > >
> > > New patchs in attach.
> > > Correction have been made followings comments f
Hi,
On Mon, Nov 21, 2016 at 4:40 PM, Rostislav Pehlivanov
wrote:
> On 21 November 2016 at 20:44, Martin Vignali
> wrote:
>
> > Hello,
> >
> > New patchs in attach.
> > Correction have been made followings comments from Andreas and Carl.
> >
> > @Rotislav : thanks for your answer.
> >
> > Martin
On 21 November 2016 at 20:44, Martin Vignali
wrote:
> Hello,
>
> New patchs in attach.
> Correction have been made followings comments from Andreas and Carl.
>
> @Rotislav : thanks for your answer.
>
> Martin
>
> ___
> ffmpeg-devel mailing list
> ffmpeg
Hello,
New patchs in attach.
Correction have been made followings comments from Andreas and Carl.
@Rotislav : thanks for your answer.
Martin
From d923d4e8031fae4fa87cab39b431d055d2744ffe Mon Sep 17 00:00:00 2001
From: Martin Vignali
Date: Mon, 21 Nov 2016 21:39:07 +0100
Subject: [PATCH 1/2] lib
Le primidi 1er frimaire, an CCXXV, wm4 a écrit :
> I do not understand this. Please explain why requesting features is
> against policy or not useful.
Please stop your passive-aggressive flame bait.
signature.asc
Description: Digital signature
___
ffmp
On Mon, 21 Nov 2016 13:08:37 +0100
Carl Eugen Hoyos wrote:
> 2016-11-21 13:06 GMT+01:00 wm4 :
> > On Mon, 21 Nov 2016 12:48:31 +0100
> > Carl Eugen Hoyos wrote:
> >
> >> 2016-11-21 12:04 GMT+01:00 wm4 :
> >> > On Mon, 21 Nov 2016 07:36:44 +
> >>
> >> > Can we get support for .doc and
2016-11-21 13:06 GMT+01:00 wm4 :
> On Mon, 21 Nov 2016 12:48:31 +0100
> Carl Eugen Hoyos wrote:
>
>> 2016-11-21 12:04 GMT+01:00 wm4 :
>> > On Mon, 21 Nov 2016 07:36:44 +
>>
>> > Can we get support for .doc and .html too? I think these would be great
>> > features.
>>
>> Maybe such comments are
On Mon, 21 Nov 2016 12:48:31 +0100
Carl Eugen Hoyos wrote:
> 2016-11-21 12:04 GMT+01:00 wm4 :
> > On Mon, 21 Nov 2016 07:36:44 +
>
> > Can we get support for .doc and .html too? I think these would be great
> > features.
>
> Maybe such comments are not against our policy (that you reque
2016-11-21 12:04 GMT+01:00 wm4 :
> On Mon, 21 Nov 2016 07:36:44 +
> Can we get support for .doc and .html too? I think these would be great
> features.
Maybe such comments are not against our policy (that you requested so
much), in any case they are not useful.
Carl Eugen
___
On Mon, 21 Nov 2016 07:36:44 +
Rostislav Pehlivanov wrote:
> On 20 November 2016 at 17:05, Kieran Kunhya wrote:
>
> > I object to this patch for the reasons Rostislav outlined..
> >
> > Kieran
> > ___
> > ffmpeg-devel mailing list
> > ffmpeg-devel
On 20 November 2016 at 17:05, Kieran Kunhya wrote:
> I object to this patch for the reasons Rostislav outlined..
>
> Kieran
> ___
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
I actually don't
2016-11-20 18:05 GMT+01:00 Kieran Kunhya :
> I object to this patch for the reasons Rostislav outlined..
Since this version of the decoder does not only support
uncompressed images, I don't think the "reasons" are
still valid.
Iirc, FFmpeg does not support all features of practically
every standa
2016-11-20 18:04 GMT+01:00 Martin Vignali :
>>
>> > I doesn't find a planar, for YA, so there is now two way to store data
>> > one for YA, and one for Gray and GBR.
>>
>> Sorry, I don't understand this sentence...
>
> Not sure i understand what is your question.
> I use planar for RGB and Gray, bu
> However, all the rle samples seem to not decode entirely correct, instead
> containing blue/white rectangular areas.
>
>
> Sorry, forget to answer to that.
The white/blue rect have been add by me to the sample file, in order to
"enable" the rle compression of photoshop.
So i think, the conversion
Hello,
2016-11-20 18:05 GMT+01:00 Kieran Kunhya :
> I object to this patch for the reasons Rostislav outlined..
>
Thanks for saying this now.
I answer to Rotislav in July.
No answer from him from this moment,
and the answers from other people, doesn't seems to go in the same way.
Martin
>
> > I doesn't find a planar, for YA, so there is now two way to store data
> > one for YA, and one for Gray and GBR.
>
> Sorry, I don't understand this sentence...
>
Not sure i understand what is your question.
I use planar for RGB and Gray, bu for YA, i doesn't find a planar something
like YAP
I object to this patch for the reasons Rostislav outlined..
Kieran
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> > Subject: [PATCH 1/2] libavcodec : add decoder for Photoshop PSD image
> file.
>
> ^
> image file*s* and no dot at the end
>
> > Decode the Image Data Section (who contain merge picture).
> ^
> (which contains merged pictures)
>
> > Support RGB/A and Grayscale/
2016-11-18 18:27 GMT+01:00 Martin Vignali :
>> Did you see AV_PIX_FMT_GBRP, AV_PIX_FMT_GBRAP, AV_PIX_FMT_GBRP16
>> and AV_PIX_FMT_GBRAP16?
>>
>
> New patch in attach use AV_PIX_FMT_GBRP, AV_PIX_FMT_GBRAP,
> AV_PIX_FMT_GBRP16BE
> and AV_PIX_FMT_GBRAP16BE for RGB mode
> I doesn't find a planar, for
On 18.11.2016 18:27, Martin Vignali wrote:
> From c41e6c79ed42478a1658c8922d19c832d3a89424 Mon Sep 17 00:00:00 2001
> From: Martin Vignali
> Date: Fri, 18 Nov 2016 18:18:20 +0100
> Subject: [PATCH 1/2] libavcodec : add decoder for Photoshop PSD image file.
>
> Can you share a few small samples?
> I'd like to fuzz test this decoder a bit.
>
> You can find sample here :
https://we.tl/BHKIQCDZZm
Martin
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
On 18.11.2016 18:27, Martin Vignali wrote:
> From c41e6c79ed42478a1658c8922d19c832d3a89424 Mon Sep 17 00:00:00 2001
> From: Martin Vignali
> Date: Fri, 18 Nov 2016 18:18:20 +0100
> Subject: [PATCH 1/2] libavcodec : add decoder for Photoshop PSD image file.
>
> Decode the Image Data Section (who c
Hello,
Did you see AV_PIX_FMT_GBRP, AV_PIX_FMT_GBRAP, AV_PIX_FMT_GBRP16
> and AV_PIX_FMT_GBRAP16?
>
New patch in attach use AV_PIX_FMT_GBRP, AV_PIX_FMT_GBRAP,
AV_PIX_FMT_GBRP16BE
and AV_PIX_FMT_GBRAP16BE for RGB mode
I doesn't find a planar, for YA, so there is now two way to store data
one for
2016-10-24 22:21 GMT+02:00 Martin Vignali :
> Hello,
>
> In attach new patchs.
> /* reorganize uncompress data. Each channel is stored one after the other */
(Sorry if I misread the code.)
Did you see AV_PIX_FMT_GBRP, AV_PIX_FMT_GBRAP, AV_PIX_FMT_GBRP16
and AV_PIX_FMT_GBRAP16?
And the more I thin
Hello,
In attach new patchs.
I tested with zzuf using
zzuf -c -s0:1 -r0.01 ./ffmpeg -i inFile -f null -
on one sample
and with
zzuf -c -s0:5000 -r0.01 ./ffmpeg -i inFile -f null
on 10 others samples
Martin
From 37a7ee0fb3b32336c5376f85f1caeafb03f9c0e5 Mon Sep 17 00:00:00 2001
From: Mart
On Fri, Aug 05, 2016 at 01:05:38PM +0200, Martin Vignali wrote:
[...]
> >
> > did you test this code with a fuzzer ?
> >
>
> No i didn't know how to do that.
there are many fuzzers, an older one is:
zzuf -c -s0: -r0.01 ./ffmpeg -i someinputfile -f null -
see the zzuf manual page for mor inform
>
>
> > +static int decode_header(PSDContext * s)
> > +{
> > +int signature, version, color_mode, len_section, compression;
> > +int ret = 0;
> > +
> > +if (bytestream2_get_bytes_left(&s->gb) < 30) {/* File header
> section + color map data section length */
> > +av_log(s->avctx
On Thu, Aug 04, 2016 at 05:00:27PM +0200, Martin Vignali wrote:
> >
> > I think you should fail probe with less than 26 bytes
> > or are there smaller files that can be valid ?
> >
> >
> > Fix
>
> >
> > this is not based on git master
> >
>
> New patch in attach
>
> Martin
[...]
> +static int
>
> I think you should fail probe with less than 26 bytes
> or are there smaller files that can be valid ?
>
>
> Fix
>
> this is not based on git master
>
New patch in attach
Martin
From 7af7660813e2abe7174d12fef80f9198ee612192 Mon Sep 17 00:00:00 2001
From: Martin Vignali
Date: Thu, 4 Aug 2016
On Tue, Jul 26, 2016 at 03:07:37PM +0200, Martin Vignali wrote:
> Sorry, forgot one out of array issue in the previous patch
>
> Correct patch in attach
>
> Martin
> Makefile |1 +
> allformats.c |1 +
> img2dec.c| 33 +
> 3 files changed, 35 in
Sorry, forgot one out of array issue in the previous patch
Correct patch in attach
Martin
From 99c60ace42f29b681fc8e5729f4b0081a204cfe5 Mon Sep 17 00:00:00 2001
From: Martin Vignali
Date: Tue, 26 Jul 2016 15:05:57 +0200
Subject: [PATCH 1/2] libavformat : add Photoshop PSD file.
---
libavformat
2016-07-25 2:46 GMT+02:00 Michael Niedermayer :
> On Sun, Jul 24, 2016 at 03:21:00PM +0200, Martin Vignali wrote:
> > > breaks fate
> > >
> > > make: *** [fate-filter-metadata-ebur128] Error 1
> > > --- - 2016-07-23 18:50:11.633752058 +0200
> > > +++ tests/data/fate/probe-format-roundup14142
On Sun, Jul 24, 2016 at 03:21:00PM +0200, Martin Vignali wrote:
> > breaks fate
> >
> > make: *** [fate-filter-metadata-ebur128] Error 1
> > --- - 2016-07-23 18:50:11.633752058 +0200
> > +++ tests/data/fate/probe-format-roundup14142016-07-23
> > 18:50:10.913635588 +0200
> > @@ -1 +1 @@
> > -m
2016-07-24 16:36 GMT+02:00 Moritz Barsnick :
> On Sun, Jul 24, 2016 at 15:21:00 +0200, Martin Vignali wrote:
> > Subject: [PATCH] libavcodec : add decoder for .psd image file.
> [...]
> > +- Psd Decoder
> [...]
> > +@item PSD @tab @tab X
> > +@tab Photoshop
> [...]
> > +.lon
On Sun, Jul 24, 2016 at 15:21:00 +0200, Martin Vignali wrote:
> Subject: [PATCH] libavcodec : add decoder for .psd image file.
[...]
> +- Psd Decoder
[...]
> +@item PSD @tab @tab X
> +@tab Photoshop
[...]
> +.long_name = NULL_IF_CONFIG_SMALL("Photoshop file"),
".psd", "psd",
> breaks fate
>
> make: *** [fate-filter-metadata-ebur128] Error 1
> --- - 2016-07-23 18:50:11.633752058 +0200
> +++ tests/data/fate/probe-format-roundup14142016-07-23
> 18:50:10.913635588 +0200
> @@ -1 +1 @@
> -mpeg
> +psd_pipe
> Test probe-format-roundup1414 failed. Look at
> tests/data/fat
On Sat, Jul 23, 2016 at 04:37:19PM +0200, Martin Vignali wrote:
> Hello,
>
> New patch in attach.
>
> - add support for RLE compression (who can be use by Photoshop and After
> Effects)
>
> RLE samples can be found here :
> https://we.tl/LCQLqwuvZ4
>
> - fix trouble with not even width.
> - mod
Hello,
New patch in attach.
- add support for RLE compression (who can be use by Photoshop and After
Effects)
RLE samples can be found here :
https://we.tl/LCQLqwuvZ4
- fix trouble with not even width.
- modification of the probe function, but i'm not sure of mine for this
- cosmetics fix.
Co
On 09.07.2016 19:37, Martin Vignali wrote:
You're right, this is a very complex file format, with lot of features.
indeed..
But i disagree, with the idea, that this decoder have no interest. The
"useful question", highly depend of the point of view.
I write this decoder, because i sometime nee
Thanks for your comments
@James Almer : MKTAG, is maybe more comprehensive, than the 32i value.
Modify locally
@Carl Eugen Hoyos :
> > +{ AV_CODEC_ID_EXR, MKTAG('p', 's', 'd', ' ') }, /* Psd */
>
> What produces such files?
>
I think i do a mistake here. Remove locally
> psd_probe(
On 7/9/16, Rostislav Pehlivanov wrote:
> On 9 July 2016 at 16:29, Martin Vignali wrote:
>
>> Hello,
>>
>> In attach patch to add decoding of .psd file (Photoshop file).
>>
>> PSD file can contain merge layers data at the end of the file (Image Data
>> Section)
>>
>> This patch add support for thi
On 9 July 2016 at 16:29, Martin Vignali wrote:
> Hello,
>
> In attach patch to add decoding of .psd file (Photoshop file).
>
> PSD file can contain merge layers data at the end of the file (Image Data
> Section)
>
> This patch add support for this kind of files, using uncompress data Image
> Data
On 7/9/16, Martin Vignali wrote:
> Hello,
>
> In attach patch to add decoding of .psd file (Photoshop file).
>
> PSD file can contain merge layers data at the end of the file (Image Data
> Section)
>
> This patch add support for this kind of files, using uncompress data Image
> Data Section
>
> Su
On 7/9/2016 1:09 PM, James Almer wrote:
> On 7/9/2016 12:29 PM, Martin Vignali wrote:
>> +static int psd_probe(AVProbeData *p)
>> +{
>> +const uint8_t *b = p->buf;
>> +
>> +if (AV_RL32(b) == 1397768760)
>
> if (AV_RL32(b) == MKTAG('S','P','B','8'))
>
MKTAG('8','B','P','S'), sorry.
Also,
Martin Vignali gmail.com> writes:
> psd_probe()
Please also test for version (and color mode) and increase
the probe score / if easily possible, don't detect things
that are immediately rejected by the decoder.
And don't remove the empty line from Changelog...
> +{ AV_CODEC_ID_EXR,
On 7/9/2016 12:29 PM, Martin Vignali wrote:
> Hello,
>
> In attach patch to add decoding of .psd file (Photoshop file).
>
> PSD file can contain merge layers data at the end of the file (Image Data
> Section)
>
> This patch add support for this kind of files, using uncompress data Image
> Data S
59 matches
Mail list logo
