A small test harness bundle was recently added that is breaking the docker image build. It could be added to the docker image, but that would introduce a bunch of extraneous test file dependencies. So this tweaks the build to simply skip the test bundle if its primary source file is not found.
Also added some other test fixes along the way:
* make a custom widget test more reliable
* update a localization test now that `pt` exists
* store more log info in artifact on error
Summary:
This uses a newer version of mocha in grist-core so that tests can be run in parallel. That allows more tests to be moved without slowing things down overall. Tests moved are venerable browser tests; only the ones that "just work" or worked without too much trouble to are moved, in order to keep the diff from growing too large. Will wrestle with more in follow up.
Parallelism is at the file level, rather than the individual test.
The newer version of mocha isn't needed for grist-saas repo; tests are parallelized in our internal CI by other means. I've chosen to allocate files to workers in a cruder way than our internal CI, based on initial characters rather than an automated process. The automated process would need some reworking to be compatible with mocha running in parallel mode.
Test Plan: this diff was tested first on grist-core, then ported to grist-saas so saas repo history will correctly track history of moved files.
Reviewers: jarek
Reviewed By: jarek
Subscribers: jarek
Differential Revision: https://phab.getgrist.com/D3927
This sets up a framework for running tests in parallel.
It increases the total time taken (since some steps are
repeated) but reduces the turn-around time significantly
overall.
The main objective is to make it possible to release more
test batches to grist-core without bringing CI to a crawl.
The clever little test/split-test.js script is from the
Grist Labs mono-repo and is Dmitry's work.
I considered doing the build in one job, and copying
it to test jobs, since it feels wasteful to repeat it.
That may be worth trying, especially if we start getting
jobs backing up (total concurrent Linux jobs on free plan
is quoted at 20).
It might also be worth looking at doing some tests in
parallel on the same worker, perhaps using the relatively
new MOCHA_WORKER_ID feature, since the tests are often not
actually CPU or I/O bound.
Use python3 for running tests. Grist has supported python3 for a while,
and is slowly slowly starting to inch away from python2 support. Grist
now needs python3 when doing imports, although python2 is still supported
for regular document operation.
* remove stray redis dependency in test
* tweak handling of database connection between tests
* upgrade node versions in tests, type guessing in node 10 has problems
Summary:
This activates the new tests that can run on a grist-core image,
and will not push a new version to `latest` tag unless those tests
pass.
This also runs a smoke test in the regular CI tests, to confirm the
code continues to work without GRIST_TEST_LOGIN enabled.
Test Plan: tested on a fork of grist-core
Reviewers: georgegevoian
Reviewed By: georgegevoian
Differential Revision: https://phab.getgrist.com/D3186
Summary:
* Keeps `grist-core:latest` docker image up to date with grist-core
source, with a daily cron job.
* Update yarn.lock file.
* In passing, tag current version for a versioned release.
When a push to `main` branch passes tests, the result is placed in `latest_candidate` branch. The cron job will update docker up from `latest_candidate`, placing code used for image in `latest` branch.
Test Plan: tested in a grist-core fork
Reviewers: georgegevoian
Reviewed By: georgegevoian
Differential Revision: https://phab.getgrist.com/D3156
Summary:
Activate CI for grist-core using github actions. Can be improved
with caching but starting simple.
Test Plan: tested in a fork of grist-core
Reviewers: dsagal
Reviewed By: dsagal
Differential Revision: https://phab.getgrist.com/D2771
Paul Fitzpatrick
Jarosław Sadziński
George Gevoian