This fixes all occurences of `cmake --build` as it already was in one
case with commit HEAD~7.
If omitted, CMake leaves the number of processes used on underlying
implementation, which turns out to `make -j`, thus unlimited number of
cores, which is undesirable.
Using `cmake --build . --parallel` actually doesn't limit the number
of processes because it just passes the number (here none) to make,
resulting in `make -j`.
This is a possible explanations of macOS runner freezes, as eg. here:
<https://github.com/MartinPulec/UltraGrid/actions/runs/7641831068>
(last relevant message in log is `[100%] Built target joysticks`).
`hdiutil umount` occasionally fails with `hdiutil: create failed -
Resource busy`. Perhaps some macOS process is touching the content
of the DMG (there are some reports in different GH about
`XProtectBehaviorService` or something).
refer to GH-367
VulkanLoader has been updated, with the CUDA _EXT prefix, which is on
contrary incompatible with Vulkan-Headers v1.3.270.
This reverts commit 62ae77117d.
After previous commit, UG configure finds the .pc file because
the file was copied incorrectly before - it is copied to
/usr/local/lib/pkgconfig, which was perhaps not a directory, creating
file _named_ pkgconfig instead the file inside that dir. Interestingly
this didn't look like it influenced other libraries, that might also
want to copy its files to something, that was then file.
But the contents is not usable without adjustnments:
1. prefix is C:/Program Files/CineFormSDK
2. lib is set to -lCFHDCodecShared in the PC file, but generated
library has name just CFHDCodec (something similar as already solved
by 0001-CMakeList.txt-remove-output-lib-name-force-UNIX.patch)
While the pkg-config file was not used previously, it is perhaps easiest
to leave it as it used to be instead of trying to fix the .pc file.
This saves some 2.5 minute from the Windows runner run time.
There is also commented-out code to download the Release Asset directly
from CineForm GH. This unfortunatelly doesn't work right now (commented
in the file).
When GITHUB_TOKEN was not provided, fetch_json didn't work, because it
uses positional arguments ("$@"), which were set only in this case but
not unset otherwise.
closes GH-352
It looks like the pre-build library BUILD/Binaries/x64/SpoutLibrary.dll
is no longer compatible with the headers in Git.
Pre-built library version is 2.007.010 while the latest version on GitHub
is Spout 2.007.012. When building Spout from sources, the referenced
issue no longer occurs. Also the libraries are not expected to be included
in the Git in future (leadedge/Spout2#103).
Homebrew builds now dependencies from sources on unsupported macOSes
(which is 11 now), which makes it slow. Relevant snippet from brew output:
```
Warning: You are using macOS 11.
We (and Apple) do not provide support for this old version.
It is expected behaviour that some formulae will fail to build in this old version.
It is expected behaviour that Homebrew will be buggy and slow.
```
This makes the script reentrant, allowing to run it also interactively
repeatedly, provided that the `apple_key_p12_b64` and
`notarytool_credentials` environment variables are correctly set.
Allow 32-bit builds on Buster. This has some advantages
regarding supported HW accelerations in UltraGrid, eg. h264_omx is not
supported in higher versions of the system.
It is not enabled by default but can be enabled easily by switching the
value in YAML file.
refers to GH-344
Deleted 2 unneeded lines:
1. OLDPWD - not used
2. comment - no longer valid - the cached environment key contains hash
of .github/scripts/Linux/arm/bootstrap.sh, which means that if the file
changes, the old env is not used, anyways
MMAL headers not present in 64-bit RaspiOS 11 libraspberrypi-dev (but
mmal.pc is still there!)
The MMAL API doesn't seem to be longer supported, see necro-posting
in GH-129.
