| Message ID | 20240213212456.167386-4-leo.izen@gmail.com |
|---|---|
| State | New |
| Headers | show |
| Series | EXIF overhaul | expand |
| Context | Check | Description |
|---|---|---|
| yinshiyou/make_loongarch64 | success | Make finished |
| yinshiyou/make_fate_loongarch64 | fail | Make fate failed |
| andriy/make_x86 | success | Make finished |
| andriy/make_fate_x86 | fail | Make fate failed |
On Tue, Feb 13, 2024 at 04:24:56PM -0500, Leo Izen wrote: > Write EXIF metadata exposed AV_FRAME_DATA_EXIF as an eXIf chunk > to PNG files, if present. > > Signed-off-by: Leo Izen <leo.izen@gmail.com> > --- > libavcodec/pngenc.c | 4 ++++ > 1 file changed, 4 insertions(+) This seems to break: --- ./tests/ref/fate/cover-art-mp3-id3v2-remux 2024-02-14 19:33:50.756012174 +0100 +++ tests/data/fate/cover-art-mp3-id3v2-remux 2024-02-14 19:36:59.150052682 +0100 @@ -1,5 +1,5 @@ -94946f0efd5f9bb0061ac1fbff7d731f *tests/data/fate/cover-art-mp3-id3v2-remux.mp3 -399346 tests/data/fate/cover-art-mp3-id3v2-remux.mp3 +0643e306430fbadeb1c89da6ab36a303 *tests/data/fate/cover-art-mp3-id3v2-remux.mp3 +399414 tests/data/fate/cover-art-mp3-id3v2-remux.mp3 #tb 0: 1/14112000 #media_type 0: audio #codec_id 0: mp3 @@ -23,7 +23,7 @@ 0, -353590, -353590, 368640, 417, 0x15848290, S=1, 10 1, 0, 0, 0, 208350, 0x291b44d1 2, 0, 0, 0, 15760, 0x71d5c418 -3, 0, 0, 0, 165671, 0x7c1c8070 +3, 0, 0, 0, 165739, 0x2e108b69 0, 15050, 15050, 368640, 418, 0x46f684a4 0, 383690, 383690, 368640, 418, 0x46f684a4 0, 752330, 752330, 368640, 418, 0x46f684a4 Test cover-art-mp3-id3v2-remux failed. Look at tests/data/fate/cover-art-mp3-id3v2-remux.err for details. tests/Makefile:318: recipe for target 'fate-cover-art-mp3-id3v2-remux' failed make: *** [fate-cover-art-mp3-id3v2-remux] Error 1 [...]
Leo Izen: > Write EXIF metadata exposed AV_FRAME_DATA_EXIF as an eXIf chunk > to PNG files, if present. > > Signed-off-by: Leo Izen <leo.izen@gmail.com> > --- > libavcodec/pngenc.c | 4 ++++ > 1 file changed, 4 insertions(+) > > diff --git a/libavcodec/pngenc.c b/libavcodec/pngenc.c > index 50689cb50c..a302c879da 100644 > --- a/libavcodec/pngenc.c > +++ b/libavcodec/pngenc.c > @@ -413,6 +413,10 @@ static int encode_headers(AVCodecContext *avctx, const AVFrame *pict) > } > } > > + side_data = av_frame_get_side_data(pict, AV_FRAME_DATA_EXIF); > + if (side_data) > + png_write_chunk(&s->bytestream, MKTAG('e', 'X', 'I', 'f'), side_data->data, FFMIN(side_data->size, INT_MAX)); > + > side_data = av_frame_get_side_data(pict, AV_FRAME_DATA_ICC_PROFILE); > if ((ret = png_write_iccp(s, side_data))) > return ret; If I see this correctly, then these patches can lead to a situation where an input packet has rotation metadata in exif which gets exported twice -- as displaymatrix and as exif metadata side data. If the user changes the displaymatrix (e.g. applies the transformation to the image data and removes the displaymatrix side data before reencoding), the exif data (that the user would probably not be aware of) would still be there and get propagated into the output, corrupting it. - Andreas
On 2/14/24 13:53, Andreas Rheinhardt wrote: > Leo Izen: >> Write EXIF metadata exposed AV_FRAME_DATA_EXIF as an eXIf chunk >> to PNG files, if present. >> >> Signed-off-by: Leo Izen <leo.izen@gmail.com> >> --- >> libavcodec/pngenc.c | 4 ++++ >> 1 file changed, 4 insertions(+) >> >> diff --git a/libavcodec/pngenc.c b/libavcodec/pngenc.c >> index 50689cb50c..a302c879da 100644 >> --- a/libavcodec/pngenc.c >> +++ b/libavcodec/pngenc.c >> @@ -413,6 +413,10 @@ static int encode_headers(AVCodecContext *avctx, const AVFrame *pict) >> } >> } >> >> + side_data = av_frame_get_side_data(pict, AV_FRAME_DATA_EXIF); >> + if (side_data) >> + png_write_chunk(&s->bytestream, MKTAG('e', 'X', 'I', 'f'), side_data->data, FFMIN(side_data->size, INT_MAX)); >> + >> side_data = av_frame_get_side_data(pict, AV_FRAME_DATA_ICC_PROFILE); >> if ((ret = png_write_iccp(s, side_data))) >> return ret; > > If I see this correctly, then these patches can lead to a situation > where an input packet has rotation metadata in exif which gets exported > twice -- as displaymatrix and as exif metadata side data. If the user > changes the displaymatrix (e.g. applies the transformation to the image > data and removes the displaymatrix side data before reencoding), the > exif data (that the user would probably not be aware of) would still be > there and get propagated into the output, corrupting it. > Hm. This is true, but it's also currently how the mjpeg decoder functions (i.e. it attaches displaymatrix side data). There are no encoders that currently save EXIF metadata, so there's no precedent. How do you suggest this be reconciled? - Leo Izen (Traneptora)
diff --git a/libavcodec/pngenc.c b/libavcodec/pngenc.c index 50689cb50c..a302c879da 100644 --- a/libavcodec/pngenc.c +++ b/libavcodec/pngenc.c @@ -413,6 +413,10 @@ static int encode_headers(AVCodecContext *avctx, const AVFrame *pict) } } + side_data = av_frame_get_side_data(pict, AV_FRAME_DATA_EXIF); + if (side_data) + png_write_chunk(&s->bytestream, MKTAG('e', 'X', 'I', 'f'), side_data->data, FFMIN(side_data->size, INT_MAX)); + side_data = av_frame_get_side_data(pict, AV_FRAME_DATA_ICC_PROFILE); if ((ret = png_write_iccp(s, side_data))) return ret;
Write EXIF metadata exposed AV_FRAME_DATA_EXIF as an eXIf chunk to PNG files, if present. Signed-off-by: Leo Izen <leo.izen@gmail.com> --- libavcodec/pngenc.c | 4 ++++ 1 file changed, 4 insertions(+)