<feed xmlns='http://www.w3.org/2005/Atom'>
<title>dcpomatic/src/lib/ffmpeg_decoder.cc, branch v2.1.4</title>
<subtitle>DCP-o-matic DCP tools</subtitle>
<id>https://git.carlh.net/cgit/dcpomatic/atom?h=v2.1.4</id>
<link rel='self' href='https://git.carlh.net/cgit/dcpomatic/atom?h=v2.1.4'/>
<link rel='alternate' type='text/html' href='https://git.carlh.net/cgit/dcpomatic/'/>
<updated>2015-06-14T21:52:42Z</updated>
<entry>
<title>Don't seek before 0 in FFmpegDecoder; fixes errors on seek close to 0.</title>
<updated>2015-06-14T21:52:42Z</updated>
<author>
<name>Carl Hetherington</name>
<email>cth@carlh.net</email>
</author>
<published>2015-06-14T21:52:42Z</published>
<link rel='alternate' type='text/html' href='https://git.carlh.net/cgit/dcpomatic/commit/?id=b6aa4eaa6e4335c3b5a7144c0ab126eb8dbe94ec'/>
<id>urn:sha1:b6aa4eaa6e4335c3b5a7144c0ab126eb8dbe94ec</id>
<content type='text'>
</content>
</entry>
<entry>
<title>Fix failure on 1-frame-back seek (#604).</title>
<updated>2015-06-14T21:05:45Z</updated>
<author>
<name>Carl Hetherington</name>
<email>cth@carlh.net</email>
</author>
<published>2015-06-14T21:05:45Z</published>
<link rel='alternate' type='text/html' href='https://git.carlh.net/cgit/dcpomatic/commit/?id=2651ef8c2fd486332457630d459c1e620e626c1b'/>
<id>urn:sha1:2651ef8c2fd486332457630d459c1e620e626c1b</id>
<content type='text'>
I don't know why this works, but it's what 1.x has been using;
I can't find any documentation on what AVSEEK_FLAG_BACKWARD is supposed
to mean.
</content>
</entry>
<entry>
<title>Remove PassReason stuff.</title>
<updated>2015-06-12T12:47:35Z</updated>
<author>
<name>Carl Hetherington</name>
<email>cth@carlh.net</email>
</author>
<published>2015-06-12T12:47:35Z</published>
<link rel='alternate' type='text/html' href='https://git.carlh.net/cgit/dcpomatic/commit/?id=bdbfd6b08856a445446bfd845f1c43109d184250'/>
<id>urn:sha1:bdbfd6b08856a445446bfd845f1c43109d184250</id>
<content type='text'>
This feels wrong: it means that it is possible for FFmpegDecoder
to discard packets.  I can't see how this is ok in all cases:
maybe we were lucky that it worked at all.
</content>
</entry>
<entry>
<title>Small optimisation.</title>
<updated>2015-06-12T08:40:31Z</updated>
<author>
<name>Carl Hetherington</name>
<email>cth@carlh.net</email>
</author>
<published>2015-06-12T08:40:31Z</published>
<link rel='alternate' type='text/html' href='https://git.carlh.net/cgit/dcpomatic/commit/?id=977c874929e28792af0d8709c32dd0695f6c7f5e'/>
<id>urn:sha1:977c874929e28792af0d8709c32dd0695f6c7f5e</id>
<content type='text'>
</content>
</entry>
<entry>
<title>Handle multiple audio streams in a single piece of content</title>
<updated>2015-06-02T12:38:21Z</updated>
<author>
<name>Carl Hetherington</name>
<email>cth@carlh.net</email>
</author>
<published>2015-05-27T19:55:51Z</published>
<link rel='alternate' type='text/html' href='https://git.carlh.net/cgit/dcpomatic/commit/?id=0a93237cb5e4642d3b698ff9b7d0cfae5401478c'/>
<id>urn:sha1:0a93237cb5e4642d3b698ff9b7d0cfae5401478c</id>
<content type='text'>
in a similar way to the V1 patch.
</content>
</entry>
<entry>
<title>Token effort toward non-bitmap subtitles from FFmpegDecoder.</title>
<updated>2015-05-13T22:38:49Z</updated>
<author>
<name>Carl Hetherington</name>
<email>cth@carlh.net</email>
</author>
<published>2015-05-13T22:38:49Z</published>
<link rel='alternate' type='text/html' href='https://git.carlh.net/cgit/dcpomatic/commit/?id=419c025709132e21ce948c4a6051a323f06ee61d'/>
<id>urn:sha1:419c025709132e21ce948c4a6051a323f06ee61d</id>
<content type='text'>
</content>
</entry>
<entry>
<title>Don't bother decoding video frames when we're seeking around trying to find subtitles.</title>
<updated>2015-05-13T09:15:26Z</updated>
<author>
<name>Carl Hetherington</name>
<email>cth@carlh.net</email>
</author>
<published>2015-05-13T09:15:26Z</published>
<link rel='alternate' type='text/html' href='https://git.carlh.net/cgit/dcpomatic/commit/?id=9922c1f2eaea674ba2ff6cce5f5853655fd8ad7a'/>
<id>urn:sha1:9922c1f2eaea674ba2ff6cce5f5853655fd8ad7a</id>
<content type='text'>
</content>
</entry>
<entry>
<title>Look up unknown subtitle end times from the data prepared by the examiner.</title>
<updated>2015-05-12T23:41:50Z</updated>
<author>
<name>Carl Hetherington</name>
<email>cth@carlh.net</email>
</author>
<published>2015-05-12T23:41:50Z</published>
<link rel='alternate' type='text/html' href='https://git.carlh.net/cgit/dcpomatic/commit/?id=83e5567530d0be24490abdda46d196e4279c5030'/>
<id>urn:sha1:83e5567530d0be24490abdda46d196e4279c5030</id>
<content type='text'>
</content>
</entry>
<entry>
<title>Assorted image subtitle fixes.</title>
<updated>2015-05-12T15:13:48Z</updated>
<author>
<name>Carl Hetherington</name>
<email>cth@carlh.net</email>
</author>
<published>2015-05-12T15:13:48Z</published>
<link rel='alternate' type='text/html' href='https://git.carlh.net/cgit/dcpomatic/commit/?id=2bfd531137f1a4874493186015046e33c5a07c1e'/>
<id>urn:sha1:2bfd531137f1a4874493186015046e33c5a07c1e</id>
<content type='text'>
</content>
</entry>
<entry>
<title>Remove the code, on seeking, to seek to the requested timestamp</title>
<updated>2015-01-30T22:08:08Z</updated>
<author>
<name>Carl Hetherington</name>
<email>cth@carlh.net</email>
</author>
<published>2015-01-30T22:08:08Z</published>
<link rel='alternate' type='text/html' href='https://git.carlh.net/cgit/dcpomatic/commit/?id=e520faf63aa7c4446e6e6477553824645d2b1f47'/>
<id>urn:sha1:e520faf63aa7c4446e6e6477553824645d2b1f47</id>
<content type='text'>
in the audio stream if it is lower than that for the video stream.
The FFmpeg docs don't support it, AFAICS, as they suggest that
the the seek should be in the timebase of the specified stream.

It also causes problems with the "preview" test when seeking
inaccurately in the viewer.
</content>
</entry>
</feed>
