This uses the dockcross manylinux docker image to enable crosscompiling aarch64 python wheels. The wheels built for aarch64 linux are excluded from the release for now, pending testing (pending, in PR#8392).
The setuptools package was added to allow definition of namespaces using
the now outdated (and discouraged from use) pkg_resources-style.
The code here, for a long while now, uses a try/except (`ImportError`)
protection around the setuptools code, and falls back to the more
encoraged pkgutil-style.
Removing this library won't affect any current workflow, and in the case
of setuptools not found, it'll actually use a more modern (and
encouraged) flow.
This change updates the README.md to describe the order of resolution
between `options['generate_py_protobufs']`, `--protoc`, and the `PROTOC`
env var.
Since the `setup.py` script is not the main way to find docs, this change
deletes the parts of the docstring that are redundant with `README.md`.
Currently, the logic in `generate_py_protobufs` cannot find a custom `protoc` (even though it's supposed to be possible). Fixes:
1. Mark the `--protoc` flag as accepting an argument.
2. If the `--protoc` flag was not passed, try finding `PROTOC` in the environment.
3. (Existing behavior) Otherwise, fall back to `spawn.find_executable`.
Hat tip to @bobhancock for uncovering the problem(s).
Using non-versioned scripts for `build_file` means every Python version is
tested (under Tox), and this is multiplied by each Python version running Tox.
For example, when the `python38` config is used to run Tox, the
Tox layer runs on Python 3.8. This then tests all of the Python versions
individually, including, for example, python27 tests.
This change fixes the `build_file` paths to point to the same-pathed build
script. For example, `kokoro/linux/python27/presubmit.cfg` now uses
`kokoro/linux/python27/build.sh`.
Some additional fixes:
* Use `python -m tox` in tests.sh instead of just `tox`. This helps non-site
installations of tox, where the `tox` script may not be on `$PATH`.
* Ensure tox (and other Python build-related packages) are available in
Python testing images. (New images have been pushed.)
* Disable `--warnings_as_errors` due to deprecated function.
* Remove apt lists per [Docker best practices][1].
[1]: https://docs.docker.com/develop/develop-images/dockerfile_best-practices/
In #713 and #1296, the `google` package in protobuf sources was found
to cause conflicts with other Google projects, because it was not
properly configured as a namespace package [1]. The initial fix in
786f80f addressed part of the issue, and #1298 fixed the rest.
However, 786f80f (the initial fix) also made `google.protobuf` and
`google.protobuf.pyext` into namespace packages. This was not correct:
they are both regular, non-namespace, sub-subpackages.
However (still), the follow-up #1298 did not nominate them as
namespace packages, so the namespace registration behavior has
remained, but without benefit.
This change removes the unnecessary namespace registration, which has
substantial overhead, thus reducing startup time substantially when
using protobufs.
Because this change affects the import internals, quantifying the
overhead requires a full tear-down/start-up of the Python interpreter.
So, to capture the full cost for every run, I measured the time to
launching a _fresh_ Python instance in a subprocess, varying the
imports and code under test. In other words, I used `timeit` to
measure the time to launch a _fresh_ Python subprocess which actually
performs the imports.
* Reference: normal Python startup (i.e., don't import protobuf at all).
```
% python3 -m timeit -s 'import subprocess' -r 3 -n 10 'subprocess.call(["python3", "-c", "pass"])'
10 loops, best of 3: 27.1 msec per loop
```
* Baseline: cost to import `google.protobuf.descriptor`, with
extraneous namespace packages.
```
% python3 -m timeit -s 'import subprocess' -r 3 -n 10 'subprocess.call(["python3", "-c", "import google.protobuf.descriptor"])'
10 loops, best of 3: 133 msec per loop
```
* This change: cost to import `google.protobuf.descriptor`, without
extraneous namespace packages.
```
% python3 -m timeit -s 'import subprocess' -r 3 -n 10 'subprocess.call(["python3", "-c", "import google.protobuf.descriptor"])'
10 loops, best of 3: 43.1 msec per loop
```
[1]: https://packaging.python.org/guides/packaging-namespace-packages/