The new option will allow us to generate test XML output.
CTest can discover tests from googletest, but it generates one file
per case. We have several thousand cases, so one file each would be
far from optimal. The approach in this change will generate one file
per test (executable).
This enables build logic on the "macos-next" flavor of Kokoro tests.
The current runners use Big Sur with Xcode 13.2.1. This build uses cmake to configure and drive the build through Xcode. Tests are run under ctest.
This doesn't test *all* possibilities (e.g. the field being a
wrapper, or a message field, etc) - but I'm fairly confident that I
found all the places referring to the case.
When following the instructions directly (for developing on Windows) we should end up with a generator that's easy to run for csharp/generate_protos.sh, and we shouldn't end up trying to stage the build output.
This change adds separate libraries for common test utilities and test proto definitions, which are shared by different test targets.
Without this change, cmake fails when generating Xcode files with the error:
```
CMake Error in CMakeLists.txt:
The custom command generating
.../protobuf/src/google/protobuf/unittest_lite.pb.cc
is attached to multiple targets:
tests
lite-test
lite-arena-test
but none of these is a common dependency of the other(s). This is not
allowed by the Xcode "new build system".
```
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.
* Implement Utf8.decodeUtf8 by using String constructor and a search for the replacement string "\uFFFD". This greatly simplifies the implementation, speeds it up for ascii and saves in memory allocations for non ascii strings.
* Remove irrelevant comment about indexOf.
* Code style changes following review.
* Remove TODO + remove final per google style.
* Delete decodeUtf8 from UnsafeProcessor as it inherits the intended implementation from its parent.
* Move decodeUtf8 implementation from Utf8::Processor to Utf8 since it has only a single implementation which is independent of whether the processor is safe or unsafe.
* Change only the logic of UnsafeProcessor to use String constructor
This is done since some Android versions will see a performance regression if this change is applied.
So we are making this change only for UnsafeProcessor which is not used on Android.
* Remove duplicated Javadoc
Co-authored-by: ahadadi <ahadadi@outbrain.com>
* Allow pre-compiled binaries for ruby 3.1.1
* add comment
* fix build and use ruby 3.1.0
* add ruby31 to build CI for tests and release
* trying to fix ci
* install ruby 3.1.0 in ruby_build_environment.sh
* use head for rvm to install 3.1.0
* just install master version of rvm in prepare_build_macos_rc
* force install of master rvm in ruby_build_environment.sh
* Use coroutine=universal when compiling ruby31
* use ucontext
* fix filename
* fix coroutine name
* use git head for rake-compiler-dock
* use newest rake-compiler-dock version
