Age | Commit message (Collapse) | Author | Files | Lines |
|
From https://docs.python.org/3/library/tempfile.html#tempfile.TemporaryDirectory.name:
> The directory name can be retrieved from the name attribute of the
> returned object. When the returned object is used as a context
> manager, the name will be assigned to the target of the as clause in
> the with statement, if there is one.
Bug added in https://gitlab.freedesktop.org/gstreamer/cerbero/-/commit/5a8b7625730d8aad90c29996a0918c3a99d87abd
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1638>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1623>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1619>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1620>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1618>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1600>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1600>
|
|
These steps can also fail in a flaky way on Android or MSVC.
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1606>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1605>
|
|
https://gitlab.freedesktop.org/gstreamer/cerbero/-/jobs/65264611
https://gitlab.freedesktop.org/slomo/cerbero/-/jobs/65261197
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1596>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1594>
|
|
Don't use our mirror as primary.
https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1582#note_2604060
(cherry picked from commit 4aa90ed8fe2a0fea159e5f9f867a07201de6ed7c)
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1585>
|
|
Fixes https://gitlab.freedesktop.org/gstreamer/cerbero/-/issues/495
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1585>
|
|
(cherry picked from commit df3144cf9f58a9d934f292c34ac314f542289d2b)
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1591>
|
|
Fedora 40 switched from wget to wget2, and doesn't backport critical
bugfixes such as returning a non-zero status code on error, which can
happen due to any number of reasons, such as certificate
incompatibilities, network issues, domain not existing, server issues,
etc.
In these cases we get a success code and we don't try the fallback URL
and we fail in extract() with a confusing error message saying that
the downloaded tarball was not found.
Switch to curl, which doesn't have such issues.
(cherry picked from commit 412c7642a88ff0c5398c38660e202bd29edca665)
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1591>
|
|
Python 3.13 adds a predefined instance variable,
`__static_attributes__`, that matches the expected syntax for build
steps' identification.
See https://docs.python.org/3.13/whatsnew/3.13.html#summary-release-highlights
Co-Authored-by: Nirbheek Chauhan <nirbheek@centricular.com>
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1586>
|
|
Import the patch from the 1.82 version, this is required to build
it inside Fedora 40.
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1580>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1575>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1571>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1570>
|
|
Fixes gstreamer/gstreamer#3765
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1561>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1557>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1554>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1553>
|
|
We only support binutils >=2.36 on Linux, which means Ubuntu 22.04 or
Debian Bookworm, or thereabouts.
Fixes https://gitlab.freedesktop.org/gstreamer/cerbero/-/issues/490
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1546>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1543>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1541>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1534>
|
|
platform
Fixes #487
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1535>
|
|
(cherry picked from commit ddd686b57011e2f32af6701adea4e69387c071c9)
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1535>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1533>
|
|
See https://github.com/mesonbuild/meson/pull/13398
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1528>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1524>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1526>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1525>
|
|
We want both dynamic and static libraries on macOS, and only static
libs on iOS, Android, Windows.
We used to have a static library on all platforms because libvpx's
custom build system was too finicky regarding shared builds. With the
port to Meson, for some reason the libvpx recipe was specifying *only*
dylibs on macOS, which is not intended.
We continue to require static libs on Windows because libvpx only
supports those.
Co-Authored-by: L. E. Segovia <amy@centricular.com>
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1517>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1509>
|
|
With GCC
../glib/gspawn-win32.c:548:33: error:
passing argument 2 of '_spawnve' from incompatible pointer type
And MSVC
../glib/gspawn-win32.c(548): warning C4133: 'function':
incompatible types - from 'wchar_t *' to 'const char *'
Applying upstream MR https://gitlab.gnome.org/GNOME/glib/-/merge_requests/3237
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1509>
|
|
This is unexpected fallout from !1478 and !1485. It was reported on the
GStreamer Matrix room that GStreamer.framework would not either load or,
once its RPATH is manually inserted, it would refuse to forward any of
the reexported dylibs.
This is because of two reasons:
- the framework is generated as `<cerbero dist folder>/lib/GStreamer`,
then repacked (post having been relocated and merged as a normal lib)
- the framework entrypoint is already relocatable, so once OSXRelocator
touches it, its relocatable, working dylib ID is replaced with a
"correct" (for a lib) but broken one
This commit works around it by validating the dylib ID and skipping it
if it's already relocatable, and adding the relevant RPATHs.
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1508>
|
|
This applies and propagates the workaround from the dep common to all
affected recipes.
See <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1485>
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1499>
|
|
BaseUniversalRecipe's __getattr__ yields over to the proxy recipe for
any nonexisting methods. This is no problem for the merge step, as
it's defined in the universal recipe directly, but steps such as
codesign and relocation are instead executed only in whatever
recipe was designated as the proxy.
Due to how recipes are loaded (I think it follows filesystem's sort
order), this recipe will usually be arm64 -- x86_64 libraries and
executables being left out of the relocation, and thus resulting
in a half-baked universal binary.
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1499>
|
|
error: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/install_name_tool:
changing install names or rpaths can't be redone for:
cargo-cinstall (for architecture arm64) because larger updated load commands do not fit
(the program must be relinked, and you may need to use -headerpad or -headerpad_max_install_names)
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1498>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1489>
|
|
With the removal of the `lib` subfolder in the dylib IDs, this needs to be accounted for in the RPATH nesting calculations.
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1491>
|
|
This works around an undocumented semantic in Meson as regard dylibs'
IDs. The fix performed for dyld in 2c058e57282577cbb77a3cd99bd16fee3b6cddaa
meant that anyone attempting to use relocated libraries within Meson
would have their libraries or executables crash at launch, since Meson
does not insert any RPATH entries for dependencies, only for build
targets in the current project.
See:
https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1478
https://github.com/mesonbuild/meson/pull/3691
(NOTE: for a more comprehensive fix, implement the post-install step on
osxuniversalgenerator.py:do_merge inside the copy-pc action.)
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1491>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1488>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1480>
|
|
When researching the construction of the monolithic GStreamer
library/framework (see !1466), I found that Qt applications
were totally unable to load GStreamer once deployed through
macdeployqt. In my case, I was consuming the libraries in raw form,
through a tarball I packaged myself, but @thewildtree also ran into
the same issue when testing an app that consumes the official release.
Upon looking at the libraries, I quickly realised that all libraries
had what looked like wrongly nested load commands, of the form
`@rpath/lib/libyadda.dylib`. Although the RPATH entries looked
reasonable at first glance, this is quickly not the case once the
libraries are deployed, because the @rpath of such an app will point to
the root of the Frameworks folder, and macdeployqt deploys the libraries
in raw form there.
However, that's not all the story. @thewildtree's case revealed a much
subtler and deadlier problem: the load commands themselves do not
respect Apple's convention, leading dyld(1) to kill the application on
sight. This is because, although OSXUniversalGenerator tries making the
fat libraries relocatable (correctly) by changing their ID, there's no
equivalent change made to any consumer. All load commands must equal the
ID of the dylib being loaded [1].
This is easily fixed at a given recipe's post-install time by adjusting
the library ID there, and fixing the rpaths so that they always point to
the root of the library path.
[1]: https://developer.apple.com/forums/thread/736728
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1481>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1481>
|
|
Part-of: <https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/1481>
|