Repository navigation
ci: test files in parallel; one Chrome job per compiler - #61
Merged
Merged
Conversation
- `dart_test.yaml`: no `concurrency: 1` (test files ran one at a time); the browser suites run each in its own frame and pass in parallel. - CI: `--concurrency=4` (the runner's vCPUs), and `test_chrome` as a matrix (dart2wasm, dart2js) running in parallel instead of one after the other; the coverage from the dart2js job, as before. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #61 +/- ##
==========================================
+ Coverage 93.02% 93.05% +0.03%
==========================================
Files 34 34
Lines 9537 9537
==========================================
+ Hits 8872 8875 +3
+ Misses 665 662 -3
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
- `test_chrome`: a matrix of compiler (dart2wasm, dart2js) x 3 shards of the test files (picked round-robin by name at run time: new test files are included), all in parallel; most of the time is compiling each test file. - dart2wasm shards run 4 test files at a time. dart2js shards, collecting the coverage, run one at a time: run in parallel, the coverage missed 32 lines run asynchronously after a test file's tests (async contents, loadings, calendar events...), the -0.09% codecov/project reported. - Each dart2js shard uploads its coverage; `.codecov.yml` makes Codecov wait for the 3 uploads (`after_n_builds`) before reporting. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
In parallel on the runner's 4 vCPUs, timing tests failed in the dart2wasm shards (timers late: UIComponentAsync auto refresh, a removed UIMultiSelection's cancelled selection). The parallelism is the 6 jobs. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
3 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Why
CI took ~10–11 min, almost all of it
test_chrome: the Wasm run (~4.2 min) and then the JS run with coverage (~5.6 min), one after the other in one job, each running the test files one at a time (concurrency: 1indart_test.yaml, from 2020, no reason recorded).build(~30s) andtest_vm(~1 min) were never the bottleneck.Changes
dart_test.yaml: dropconcurrency: 1.dart testdefaults to half the processors; each browser suite runs in its own frame.dart.yml:test_chromeis a matrix (dart2wasm,dart2js),fail-fast: false, so both compilers run in parallel and one failure doesn't cancel the other. Coverage is still collected from the dart2js run and uploaded only there.--concurrency=4(GitHub's Linux runners have 4 vCPUs) for the Chrome and VM runs.test_chrometimeout 30 → 20 min.master, so the renamed jobs (test_chrome (dart2wasm),test_chrome (dart2js)) block nothing.Measured locally (all 561 browser tests passing each run)
-j 4, 3 runs: 0:35–0:38)-j 4)-j 8)-j)This PR's own CI run gives the real numbers on the runners.
No version bump: CI and test configuration only, nothing in the published package's behavior changes.
🤖 Generated with Claude Code