If a plugin removes a property when processing 'modified'
property callback, OBS WILL CRASH.
Example of offending plugin: obs-source-record
This commit moves callback processing to after iterating over
properties, thus removing the chance to crash OBS.
This pattern uses fewer instructions and also avoids using max, which
does not work on infinity.
Also remove unreferenced techniques from scale filters.
One notable difference from the AVC/HEVC code is that it also inserts
the METADATA and SEQUENCE_HEADER OBUs into new_packet, otherwise the
resulting video file wouldn't play.
By default duplicate non-static symbols loaded by dynamic libraries are
de-duplicated by the dynamic library loader. This can lead to issues
with statically linked libraries inside obs plugins if the symbols
share their signature: Whichever plugin is loaded first gets to "set"
the symbol (which can become problematic especially for C++ template
functions).
Using RTLD_LOCAL ensures that all symbols are hidden and can only be
explicitly loaded using dlsym() which avoids this issue.
Unfortunately due to the way scripting works in obs-studio, Python
still needs to be loaded with RTLD_GLOBAL, hence the branch in the fix.
The API `obs_output_video` and `obs_output_audio` returned valid
pointers until the commit fb57eff21 and 645e31fa1.
The API `obs_output_video` is used by some plugins such as obs-websocket
and obs-midi to calculate the duration of streaming and recording.
To have a similar behavior, return the media (video and audio,
respectively) from the encoder.
With the lowest deployment target being macOS 11.0, there is no need
to check for the availability of the selectors on the NSProcessInfo
class anymore.
In c8d95005c1, it was incorrect assumed that just because the fix
worked on OpenGL, that the fix would also work on D3D11. Automated tests
would be ideal here, although if there were automated tests, ironically,
it wouldn't have picked this up, because we (read: I) made D3D11 fall
back to OpenGL of D3D11 failed. Basically there is no one to blame but
myself again.
This time, we've removed the OpenGL fallback in a06d6893b9 (thank you
Rodney), and additionally I properly tested on both rendering backends.
(Lain note: I'm just going to go sit in the corner for a bit and
ruminate after this)
Due to my jank transpiler, default.effect worked with Direct3D 11 but
not with OpengGL. Fixing this particular bug with the transpiler is a
bit non-trivial so instead just modify the shader to work with both.
When `obs_output_t` is an encoded output, `obs_output_set_media` will
ignore the video and audio so that `obs_output_t` will keep holding the
`video_t` and `audio_t` pointer when the output was created.
By this commit, `video` and `audio` member variables in `obs_output_t`
will never set if it is an encoded output.
In the auto-configuration wizard, `video_t` is released and created to
have a different video size while `obs_output_t` is not released. This
resulted in accessing the released `video_t` pointer.
Define PRAGMA_WARN_DEPRECATION for MSVC to enable using that define for
cross-platform code, which allows us to write slightly less code when we
need to suppress this warning.
To avoid a mismatch between the state of pressed keys as held by the
hotkey thread and keys pressed while the application is in focus, local
key events need to be monitored as well (even though they are already
handled by Cocoa's main event loop).
If there is filter_remove, it is reasonable to expect that there is also
filter_add. filter_add also enables filters to attach signal handlers
on the parent, and disconnect them in filter_remove.
During encoder shutdown `maybe_clear_encoder_core_video_mix` is
called to clear created mixes that are no longer needed; at that
time `obs_encoder_set_video` rejects changes to `media` since
the encoder is still active, so we bypass those checks
This is an issue e.g. when a rtmp stream disconnects (and thus
all encoders are cleared) and subsequently reconnects
This fixes flipping behaviour when positional alignement inside the
bounding box is other than centered, bringing it in line with the
behaviour when centered, i.e. flip the item in-place in the bounding
box.
This changes the return type for getting the filter index from a
size_t to an int. This makes it easier for developers to use,
as an invalid index just returns a -1.
