If the user provides a `--link-objects` (or `-O`) flag that looks like it points to libprotobuf, use that for static linking instead of using the hard-coded path `../src/.libs/libprotobuf.a`. (Also, if we see such a flag, assume they want to link it statically, even if they don't provide `--compile_static_extension`.)
Similarly, if they provide a `--library-dirs=` (or `-L`), assume that is the location to search, and don't add the hard-coded path `../src/.libs`.
After some debugging I was able to determine that the Win32 DLL was failing to load libgcc_s_sjlj-1.dll. If we statically link libgcc instead, there should be no runtime dependency on libgcc.
PiperOrigin-RevId: 448338965
When I tried the previous `arm64` wheel on macOS, I discovered that `pip` on macOS only supports the `arm64` platform tag in a limited set of circumstances. pip seems to prefer `universal2` wheels.
To build a `universal2` wheel, we must run the `llvm-lipo` tool to bundle multiple `cc_binary()` outputs into a single output wheel. We use a transition to depend on multiple architectures for the extension, if we see that we are building for a multiarch CPU.
PiperOrigin-RevId: 447486256
While I've made some progress debugging b/231485326, I'm not going to be able to get this fully working in time for -rc1.
If I can debug this before the final release, we can re-enable the win32 wheels.
PiperOrigin-RevId: 446849335
This should get us the full set of binary wheels we need for release. Also added acceptance tests for all wheels except aarch64 (on Linux and macos): we cannot test these without emulation, which will require a somewhat more complicated setup.
PiperOrigin-RevId: 446574958
* Rollback pull request #9852, which will break clients who don't IWYU and depend on us for these STL headers
* Rolling forward https://github.com/protocolbuffers/protobuf/pull/9852 after confirming it doesn't break google internals
1. For some reason the version script was not working, it was failing to export the main symbol for the Python extension. I fixed this by using the `visibility` attribute instead to export the `PyInit__message` function.
2. We were not properly stripping the `python/dist/` prefix for the C module, which was making the module exported under the name `dist.google._upb` instead of `google._upb`.
3. The `py_library()` rule was failing to actually propagate the module file into the wheel, so I just removed it.
PiperOrigin-RevId: 445446611
* WIP.
* Restore deleted __init__.py files.
* Updated version of upb.
* Moved upb from a submodule to a Bazel dep.
* Updated to new upb which can tolerate missing system_python.
* Updated to a new upb.
* Fix NPE during encoding and add regression test for issue 9507.
(cherry picked from commit 58e320a732)
* Implement `respond_to?` in RubyMessage (#9677)
All synthetic methods implemented by `method_missing` are now supported by `respond_to?`.
Fixes issue #9202.
* Fix null pointer exceptions exposed by new regression tests.
* Fix clear_ on oneofs so that it is safe to call repeatedly and so that respond_to? does not depend on whether the oneof is currently cleared.
* Code cleanup: reenable more tests on JRuby.
* Align JRuby behavior with CRuby by throwing a RuntimeError when attempting to assign to a oneof.
(cherry picked from commit 8e7f936696)
* Update protobuf version
* Merge pull request #9727 from mlocati/build-packaged-php-extension
Fix building packaged PHP extension
(cherry picked from commit 7f9901c5f6)
* Update protobuf version
* Update changelogs for 3.20.1-rc1
Co-authored-by: Jason Lunn <jason.lunn@gmail.com>
Co-authored-by: Jorg Brown <jorg.brown@gmail.com>
Joshua Haberman
David L. Jones
Deanna Garcia
Protobuf Team
mkruskal-google
Joshua Haberman
David Benjamin
Protobuf Team
noahdietz
Mike Kruskal
Darly Paredes
theodorerose
Adam Cozzette