Commit Graph

1034 Commits

Author SHA1 Message Date
Jonathan Thomas
0bec048acc Copy max_audio_samples with Frame DeepCopy 2020-01-02 17:15:49 -06:00
Jonathan Thomas
3dbf372b11 Merge pull request #393 from ferdnyc/colorshift-classname
ColorShift effect: Correct class_name
2020-01-02 14:41:03 -06:00
Jonathan Thomas
0fc831f152 Merge pull request #394 from ferdnyc/effect-shortname
EffectBase: Remove unused short_name
2020-01-02 14:40:11 -06:00
Jonathan Thomas
4e91a2acf3 Merge pull request #387 from ferdnyc/switch-enums
Ensure switch statements on enums cover all values
2020-01-02 14:38:08 -06:00
Jonathan Thomas
8c91db610e Merge pull request #389 from ferdnyc/sign-compare
Don't compare differently-signed types
2020-01-02 14:37:09 -06:00
Jonathan Thomas
64d93404de Merge pull request #398 from ckirmse/fix_buffer_read
correctly calculate remaining_frame_samples
2020-01-02 14:29:51 -06:00
Jonathan Thomas
b8d3514ef6 Merge pull request #391 from ckirmse/fix_audio_crash
fix incorrect buffer size being passed to avcodec_fill_audio_frame
2020-01-02 14:28:35 -06:00
FeRD (Frank Dana)
e49f62247e Use C++11 range-based for loops where we can 2019-12-28 09:50:09 -05:00
Chris Kirmse
d60678af0d correctly calculate remaining_frame_samples
- sometimes the output from the SWR_CONVERT is not the exact amount we asked for
- without this fix, the old code would sometimes read past the end of a buffer
- valgrind complained about the old code
- read https://stackoverflow.com/questions/39587839/libswresample-swr-convert-not-producing-enough-samples
2019-12-27 09:54:57 -08:00
FeRD (Frank Dana)
cbf1f17059 Merge branch 'develop' into sign-compare 2019-12-25 16:35:04 -05:00
FeRD (Frank Dana)
c11b4ac266 EffectBase: Remove unused short_name 2019-12-19 07:59:30 -05:00
FeRD (Frank Dana)
07b6e2c403 ColorShift effect: Correct class_name 2019-12-19 07:57:26 -05:00
Chris Kirmse
4806f2ff75 fix incorrect buffer size being passed to avcodec_fill_audio_frame
- most of the time the fixed number being sent in was large enough,
  but not always
- also added logging if avcodec_fill_audio_frame fails
2019-12-18 18:27:04 -08:00
FeRD (Frank Dana)
e502f97d8a Don't compare differently-signed types 2019-12-15 14:22:59 -05:00
FeRD (Frank Dana)
bd4d2bfdc4 Cover all values in switch(enum_type) 2019-12-15 12:40:32 -05:00
FeRD (Frank Dana)
2c71ae49fc FFmpegWriter: Fixes for building with libav
- libav defines AV_CODEC_ID_HEVC but not AV_CODEC_ID_H265
- AVCodecContext->framerate was actually added in FFmpeg 2.2
  which corresponds to libavcodec 56.26
2019-12-11 05:42:41 -05:00
Frank Dana
99565bb342 Move feature summary to root CMakeLists (#383) 2019-12-07 14:53:54 -05:00
Jonathan Thomas
06e6de5c99 Merge pull request #381 from ferdnyc/swig-python
Add repr to openshot.Version
2019-12-06 17:14:56 -06:00
Jonathan Thomas
0347ad2073 Merge pull request #374 from musteresel/new-keyframe-implementation
New Keyframe implementation
2019-12-06 17:10:11 -06:00
Daniel Jour
c940c1f42b Keyframe: Cleanup duplicate binary search code
GetRepeatFraction uses two binary searches; reuse that code!
2019-12-06 01:21:25 +01:00
Daniel Jour
1fbdc521ca Keyframe::GetRepeatFraction(): Binary search, skipping when constant
The old implementation did a linear scan over the values.  This was
slow with slowly changing keyframes.  This new implementation skips
over constant (when rounded) segments and performs binary search
in (possibly long) interpolated segments to find the X coordinates
where a change occurs quickly.
2019-12-06 01:04:47 +01:00
Daniel Jour
f00edbad7e Keyframe interpolation: In own function; only for Y coordinate 2019-12-06 01:03:56 +01:00
FeRD (Frank Dana)
4a5eb20202 Add __repr__ to openshot.Version 2019-12-03 21:30:22 -05:00
Daniel Jour
ed0b081803 Keyframe::IsIncreasing(): Search over points, not values
Searching over the keyframe points is considerably faster than
calculating interpolating values and searching over them.
2019-12-03 17:27:28 +01:00
Daniel Jour
b40fa6922e Keyframe::GetMaxPoint() simplify loop 2019-12-03 16:58:53 +01:00