Some of the continuous aarch64 builds started failing yesterday because
the autotools -> CMake change caused the protoc binary to end up in an
unexpected location. This change fixes the problem by putting a symlink
in the src/ directory.
* Revert "Added cmake abseil include guard"
This reverts commit b6ee841d7c.
* Revert "Update CMake configuration to add a dependency on Abseil (#9793)"
This reverts commit e9246cd789.
This is somewhat of a rough cut, since it doesn't split apart the lite and full targets, or unit tests. However, it does split sources under src/google/protobuf into a separate package, which is a fairly impactful change.
This change creates packages under src/google/protobuf/{io,stubs,testing} and moves build definitions there. Future changes will handle .../util and .../compiler, and finally src/google/protobuf.
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`.
The cmake tests are expected to fail for now due to Abseil sources missing from the distribution artifact. The tests are structured as expected failures.
These are all "toolchain-y" things, like copts, link_opts, and config_settings. These are very different from what is in //toolchain, though, so I chose the somewhat common name build_defs for the package. For now, I am only using this package for purely internal things. (Most public "defs"-type things should come from rules_proto/rules_cc, anyhow.)
Start reducing our reliance on autotools
This commit updates a few of our CI scripts to build protoc using Bazel
instead of autotools. I also updated the setuptools and Rake build
configs to look for protoc under bazel-bin/ so that they will
preferentially use a Bazel-built protoc binary. The goal is to start
chipping away at our use of autotools in preparation for eventually
dropping autotools support.
* Proof of concept for CMake Abseil dependency
* Hooked up Abseil linking
* Adding test binaries
* Reverting absl::string view use added for testing. This will still be broken for automake and bazel
* Adding new cmake config to dist list
* Whitespace fixes and an attempt at fix for Windows Python Release (CMake version < 3.13)
* Hook up LTS to windows cmake
* Fix 'git cd' typo
* Adding clarifying comment for fix
* Adding abseil dependency to protobuf-lite too
* Update submodules instead of cloning abseil-cpp
* Adding explicit pointer to abseil root directory
Co-authored-by: Jason Lunn <jason.lunn@gmail.com>
Co-authored-by: Jorg Brown <jorg.brown@gmail.com>
This enables googletest XML output on the macos-next builders, and adds logic to collect the results.
The log collection logic is slightly complex, but it should be reusable in other contexts. The idea is to capture stdout/stderr for build steps along with googletest XML reports into a temporary directory, then stage those into paths expected for artifacts.
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.
* Allow pre-compiled binaries for ruby 3.1.0 (#9566)
* 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
* Updated CHANGES.txt for Ruby changes.
* Fixed Ruby 3.1 tests by marking intersect? as unimplemented. (#9645)
* Fixed Ruby 3.1 tests by marking intersect? as unimplemented.
* Updated compatibility tests.
Co-authored-by: Marco Concetto Rudilosso <marcoconcettorudilosso@gmail.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
Our benchmark job starting failing when we stopped supporting Python 3.5
and 3.6, but this change fixes it by running it in our java_stretch
Docker container, which has a more recent Python version. I also deleted
all configs related to old Python versions that we no longer support.
* Drop Python versions <3.7.
* Updated README to clarify that Python 3.7 is the minimum.
* Removed more Python 3.5-specific code.
Also changed tests to skip missing interpreters.
* Invoke tox directly instead of through Python.
Hopefully this will pick up python3.
* Updated java_stretch image to bullseye to get Python >= 3.7.
* Use jdk11 instead of jdk8.
* Installed python2 for gtest.
* Use "python3 -m venv" instead of "virtualenv."
* Install python3-venv.
zhangskz
Adam Cozzette
mkruskal-google
David L. Jones
Deanna Garcia
Thomas Van Lenten
Elliotte Rusty Harold
Joshua Haberman
noahdietz
Mike Kruskal
Jason Lunn
Marco Concetto Rudilosso
Brett McBride