The incorrect variable was used in a comparison, causing playback to
stop prematurely depending on whether the audio data or video data ended
first.
Closesobs-project/obs-studio#8422
Also modifies libobs & deps/media-playback.
AV_CODEC_CAP_TRUNCATED was removed in avcodec 60 [1].
We ifdef the code depending on it to allow compilation.
[1] avcodec: remove FF_API_FLAG_TRUNCATED
3ceffe7839
Signed-off-by: pkv <pkv@obsproject.com>
Stingers -- and especially track matte stingers -- are currently subject
to real time decoding, which can be detrimental in a production
environment where a stinger video may not be able to decode in a timely
fashion.
To remedy this, this change adds an option to fully decode stingers
immediately and cache the decoded video/audio in RAM for playback to
greatly improve stinger performance.
This fixes deprecation warnings since the channels member of
AVCodecContext is marked as deprecated [1], as well as the channels
member of AVFrame [2].
In all instances where a warning appear, a switch to the new API is
done.
[1] lavc: switch to the new channel layout API
FFmpeg/FFmpeg@548aeb9
[2] Bump minor versions after the channel layout changes
FFmpeg/FFmpeg@cdba98bb80
Signed-off-by: pkv <pkv@obsproject.com>
When loading a new Python script, the GIL might be released while
importing the module, allowing the tick to run and change and reset the
cur_python_script state variable. Use the busy_script variable to save
and restore the value if not null.
- Requires MbedTLS on Linux
- Enabled by default on macOS and Flatpak
- Enabled on linux via ENABLE_WHATSNEW_LINUX
- Enables compilation of blake2 on Linux/macOS
- Makes header name check also work with lowercase header
- Changes WahtsNew to be only enabled when browser panels are available
Calling `PyEval_InitThreads` has been deprecated in Python 3.7 and the
function itself will be removed in Python 3.11. The current check
guards this function behind a version check that only happens at compile
time.
This in turn leads to crashes when run on Python 3.6, as the necessary
initialization for `PyEval_ReleaseThread` did not take place.
This commit ensures the manual initialization takes place based on the
runtime version of Python and avoids loading the associated symbols
on Python 3.9 or later.
This changes the way obs-scripting looks for and loads an available
Python 3 library. It tries to find a best possible version (starting
with Python 3.10) down to and including Python 3.6 by existing file
naming conventions and loads the most recent variant it can find.
User specified search path is either a Python installation directory
(Windows), or a Framework directory containing `Python.framework`
(macOS). The dll or dylib names are composed automatically.
The Python home path is also composed automatically on macOS (where
it has to point inside the Framework directory).
Fixes issues introduced by https://github.com/swig/swig/pull/2238. As
obs-scripting is not compatible with the SWIGPYTHON_BUILTIN, obspython
needs to disable this feature.
Due to swig using a hardcoded swig library path in self-compiled
variants, we need to pass our custom swig library path with every
invocation of swig on macOS.
CMake's `add_custom_command` function does _not_ inherit the environment
variables used during configuration, hence why it needs to be passed
explicitly on invocation.
Usage of CMake's "-E env" command module should allow for cross-platform
compatible invocation.
The obspython.i file requires `ENABLE_UI` to be defined for the swig
compiler to pull in the actual obs-frontend-api header. If this is not
the case, swig will not discover the required methods and ENUMs and
in turn will not make those available in the scripting environment.
The obslua.i file requires `ENABLE_UI` to be defined for the swig
compiler to pull in the actual obs-frontend-api header. If this is not
the case, swig will not discover the required methods and ENUMs and
in turn will not make those available in the scripting environment.
Previous FFmpeg deprecation fix, and optimization attempt causes short
videos to loop badly. Keep deprecation fix, and fix the rest later.
Fixes d528f6cafa
This updates glad for EGL and adds in several extensions.
EGL_EXT_platform_base
EGL_EXT_platform_xcb
EGL_KHR_image_base
EGL_KHR_image_pixmap
The xcb platform extension is unused but good to include if we want to
try removing more Xlib dependencies. platform_base is included for good
measure as other platform extensions technically build on it. image_base
and image_pixmap are for upcoming xcomposite over EGL support.
It also drops the unused create_context_no_error extensions.
In Python 3.9+, `PyCFunction_New` and `PyCFunction_NewEx` are themselves
macros around `PyMethod_New`. To fix the warning, both macros need to
be untouched for those Python versions, instead the actual function
`PyMethod_New` needs to be imported and aliased.
This commit:
- Adds the import of the `PyCMethod_New` symbol
- Adds the function definition for `Import_PyCMethod_New`
- Adds the `PyCMethod_New` function alias
macOS uses runtime lookup for symbols in obs-scripting. While this guard
fixes a compiler warning, it also leads to the `_PyCMethod_New` symbol
being undefined.
In turn any library linking with obs-scripting will then fail to load
because of that undefined symbol. Removing this guard fixes that issue.
* obs-scripting: Fix unused 'varargs' warning
* obs-scripting: Fix PyEval deprecation warnings
Remove PyEval_InitThreads() and PyEval_ThreadsInitialized() call for
Python 3.7 and later as Python calls it automatically now.
This removes deprecation warnings when using Python 3.9 or later.
https://docs.python.org/3/c-api/init.html#c.PyEval_InitThreads
Rec. 2020 is really an SDR spec, but I think HDR10 made it okay to slap
PQ on it, call it an HDR spec. Rec. 2100 came along after and formally
allowed the use of PQ/HLG, so we should use 2100 instead.
Status output related to OBS configuration is prefixed with the string
"OBS" and added padding for enabled and disabled features. This padding
was not aligned between platforms.
By moving the padding and prefix decoration into its own function,
both elements are controlled in a single place. CMake scripts were
changed to use this new function `obs_status` instead of using CMake's
`message` function directly.
In the years since this code was added, ALPN is now widely supported,
and NPN is being removed entirely in the latest version of nginx. It's
time to start using ALPN!
Makes the callback variable "removed" atomic, and on script unload,
first sets all callbacks to removed before actually unloading the script
out of a safety precaution. (See note at the bottom for further details)
This minimizes the possibility of a race condition where the script
callback could be called while those callbacks were being removed.
Big note for this change, this change should eventually be replaced with
a reference counting ownership method where script callbacks can hold a
reference and share ownership of the script if it's still alive while
the script callback is being called. That way the script callbacks can
safely execute. May require a fair amount of reworking of the script
object.
Deprecates:
obs_source_addref()
obs_output_addref()
obs_encoder_addref()
obs_service_addref()
obs_scene_addref()
These functions should be considered unsafe and not used. Instead, use:
obs_source_get_ref()
obs_output_get_ref()
obs_encoder_get_ref()
obs_service_get_ref()
obs_scene_get_ref()
These functions return a pointer to the incremented object only if the
object is still valid, otherwise they will return null, indicating that
the object is no longer valid or is unsafe to use.
The reason why this is being done is because certain third party plugins
seem to be using addref, and are somehow managing to call addref on
sources that have already been fully released. For the sake of safety,
almost all usage of these functions within OBS have also been replaced
as well.
We currently pass a 'buffer_size' option to avformat in bytes.
Unfortunately, avformat/librist.c admits an option with the same name but in milliseconds.
The default values set in the UI (2 MB) go beyond the range admitted in ms (0-30 000)
so avformat issues errors.
This fix disables the option when rist protocol is detected.
Signed-off-by: pkv <pkv@obsproject.com>
AVFormatContext::oformat was made const on April 27, 2021 [1]. If we
respect the constness of AVOutputFormat and do not cast results from
FFmpeg functions to non-const, we cannot modify the results after the
fact. Our choices are either to cast them to non-const (and presumably
have them implicitly casted back to const on later function calls), or
only try to modify the results in versions of FFmpeg where these are not
expected to be const.
Instead of relying on casts, we can set the encoder values in the
ffmpeg_cfg struct, which are later passed to new_stream.
Also modifies deps/media-playback. Removes compiler warnings.
Some avformat functions return const AV(In/Out)putFormat per [1], so
ifdef as needed.
[1]: 56450a0ee4
lavf 59.0.100 avformat.h
avformat: Constify the API wrt AV(In|Out)putFormat
Also constify AVProbeData.
If there is a PTS discontinuity in the playback stream that causes the PTS to
reset, it is likely that new audio frames will come in while there are still
video frames waiting to be dequeued. This will cause the audio frames to be
played while the video frames wait for the PTS to get back up to where they
were before the discontinuity.