The Custom FFmpeg Output allows the user to configure it to output to a
URL instead of a file. In OBSBasic::StartRecording(), we unconditionally
call LowDiskSpace() to check for low disk space because we assume that a
"recording" will always be to disk. When os_get_free_disk_space() is
called on a non-existent path, it returns 0, which causes OBS to emit a
low disk space warning. Instead, only call DiskSpaceMessage() if not
"recording" to a URL.
The Auto-Configuration Wizard cannot progress when selecting a custom
service that does not require a stream key. Since we have no way to
validate if a custom service requires a stream key, we should just not
require a stream key is custom service is selected. To fix this, check
if a custom service has not been selected when checking if the stream
key is empty.
Co-authored-by: Ryan Foster <ryan@obsproject.com>
Add a COMPAT version of NV_ENC_INITIALIZE_PARAMS_VER, which changed
between SDK 12.0 and 12.1.
If using drivers older than the configured SDK/API requires, fall back
to the compatibility version of the API (11.1).
An example scenario would be encoding in H.264 on driver version 512.15.
SDK 12.1 requires driver version 531.61 on Windows. Currently, in this
scenario, OBS will fall back to the FFmpeg encoder path, which will do
its own driver check, which will fail on this driver version.
With this change, this scenario will instead leverage the existing
compatibility hack to fall back to the compatibility version of the API
without falling back to FFmpeg.
It should be noted that while neat, the compatibility hack that enables
Kepler GPUs to continue to be able to use NVENC in this way is expected
to be removed at some point.
Because 4a8e8bb1bd changed the
NVENCAPI_STRUCT_VERSION define to rely on enc->needs_compat_ver, we must
do the conditional check and define needs_compat_ver before any structs
are initialized because they will check the version when doing so by
calling NVENCAPI_STRUCT_VERSION internally.
In this case, make sure that we define enc->needs_compat_ver before
NV_ENCODE_API_FUNCTION_LIST_VER is called for
NV_ENCODE_API_FUNCTION_LIST.
The upload step in the job uses the output of two prior steps as its
condition. One of the jobs itself is a conditional job, so if the
primary condition is not met, the secondary output is not set and the
fromJSON call will fail.
Using a string comparison allows the check to still work as expected
(as the output will be an empty string otherwise)
Unlike filesystem addresses, which are paths,
abstract addresses are blobs.
This means that the address wasn't "\0/com/obsproject 2413747 version",
but rather "\0/com/obsproject 2402613 version\0\0\0\0\0\0\0\0\0\0\0\0\0
\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0
\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0";
this is reflected in both /proc/net/unix:
000000001bf56057: 00000002 00000000 00000000 0002 01 56719817
@/com/obsproject 2402613 version@@@@@@@@@@@
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
@@@@@@@@@@@@@@@@@@@@@@
000000003e897b41: 00000002 00000000 00000000 0002 01 56730132
@/com/obsproject 2413747 version
and ss -l, extending the whole address column to the point of
unusability.
This size calculation follows trivially from the manual,
and naturally behaves correctly, as above.
Fixes issue #10369.
Since the update to FFmpeg 6.1, streams to a Media Source are broken
if hardware decoding is enabled (both RTMP or SRT have been reported).
The video is black although the audio is decoded fine.
The manual copy of metadata introduced in commit [1] does not work any
more for some unfathomable reasons.
As a fix we call instead the av_frame_copy_props function used in FFmpeg
app in a similar context (hardware decoding) [2].
The metadata are copied without issues.
There is no need to guard the use of that function since it was
introduced 9 years ago in avutil/frame.c.
[1] 22fde5cdcd
[2] https://github.com/FFmpeg/FFmpeg/master/fftools/ffmpeg_dec.c
Signed-off-by: pkv <pkv@obsproject.com>
The MSVC_RUNTIME_LIBRARY property is not propagated to targets which
link against a target which has this property set. Thus the property
needs to be set on the actual virtualcam targets and not the interface
library.
- Rename from .appdata.xml to the modern .metainfo.xml
- Add a primary branding color (Yami blurple)
- Add "internet" to recommends and supports
- Update developer_id tag to developer name
- Shorten summary
- Remove obsoleted "kudos" tags
- Add a URL to the release tag
When OBS is closed and there is a floating dock, a Windows system
sound is emitted. This fixes that by not executing the close dock
dialog when OBS is closing.
w32-pthreads is not correctly added to the generated CMake project
if the UI subdirectory is not also included, which leads to builds
configured without UI to fail.
Before the CI checks for source code and CMake files would always
check the entire source code even if just a single files has been
changed.
With this update, the formatting script is enhanced to accept a list
of files (which is generated as a condition to run the script in the
first place) which ensures that a PR will only fail validation over
files it changed itself.
Adding to the previous commit, let's also use obs_output_set_last_error
to provide localized error messages to the user if we run into these
failure cases.
The Elgato Stream Deck plugin older than 5.5.2.0 invokes undefined
behavior that results in application crashes. Let's block older versions
to prevent the crashes.
If the "Pause timer when recording is paused" option in the Output Timer
settings was enabled, even if an Output Timer was not being used, a
recording may stop when attempting to unpause it. This was due to the
check in the UnpauseRecordingTimer function being too loose and only
checking for if the recording timer was not active. Let's initialize the
recordingTimeLeft variable to -1 and check that it's greater than 0
before attempting to restart a recording timer.
BGRA or other formats that do not use color primaries will not yield
a valid color primary value. Initializing the CFComparisonResult to a
default value and replacing it only if a non-NULL color primary value
was retrieved avoids a possible crash.
