chore: implement shard weights feedback (#38902)

Author: Skn0tt

.github/workflows/tests_primary.yml

@@ -139,14 +139,15 @@ jobs:
     - uses: ./.github/actions/run-test
       with:
         node-version: ${{matrix.node-version}}
-        command: npm run ttest -- --shard ${{ matrix.shardIndex }}/${{ matrix.shardTotal }} --shard-weights=${{ matrix.shardWeights }}
+        command: npm run ttest -- --shard ${{ matrix.shardIndex }}/${{ matrix.shardTotal }}
         bot-name: "${{ matrix.os }}-node${{ matrix.node-version }}"
         shard-index: ${{ matrix.shardIndex }}
         flakiness-client-id: ${{ secrets.AZURE_FLAKINESS_DASHBOARD_CLIENT_ID }}
         flakiness-tenant-id: ${{ secrets.AZURE_FLAKINESS_DASHBOARD_TENANT_ID }}
         flakiness-subscription-id: ${{ secrets.AZURE_FLAKINESS_DASHBOARD_SUBSCRIPTION_ID }}
       env:
         PWTEST_CHANNEL: firefox-beta
+        PWTEST_SHARD_WEIGHTS: ${{ matrix.shardWeights }}
 
   test_vscode_extension:
     name: VSCode Extension

docs/src/images/speedboard-shards.png

@@ -132,11 +132,6 @@ Reporter is allowed to override the status and hence affect the exit code of the
   - `status` <[FullStatus]<"passed"|"failed"|"timedout"|"interrupted">> Test run status.
   - `startTime` <[Date]> Test run start wall time.
   - `duration` <[int]> Test run duration in milliseconds.
-  - `shards` <[Array]<[Object]>> Only present on merged reports
-    - `shardIndex` ?<[int]> The index of the shard, one-based.
-    - `tag` ?<[Array]<[string]>> Global [`property: TestConfig.tag`] that differentiates CI environments
-    - `startTime` <[Date]> Start wall time of shard.
-    - `duration` <[int]> Shard run duration in milliseconds.
 
 Result of the full test run, `status` can be one of:
 * `'passed'` - Everything went as expected.

docs/src/test-reporter-api/class-reporter.md

@@ -100,9 +100,3 @@ Learn more about [parallelism and sharding](../test-parallel.md) with Playwright
 - type: <[int]>
 
 The index of the worker between `0` and `workers - 1`. It is guaranteed that workers running at the same time have a different `parallelIndex`.
-
-## property: TestResult.shardIndex
-* since: v1.58
-- type: ?<[int]>
-
-The index of the shard between `1` and [`shards`](../test-sharding.md).

docs/src/test-reporter-api/class-testresult.md

@@ -45,21 +45,6 @@ Without the fullyParallel setting, Playwright Test defaults to file-level granul
 - **Without** `fullyParallel`: Tests are split at the file level, so to balance the shards, it's important to keep your test files small and evenly sized.
 - To ensure the most effective use of sharding, especially in CI environments, it is recommended to use `fullyParallel: true` when aiming for balanced distribution across shards. Otherwise, you may need to manually organize your test files to avoid imbalances.
 
-### Rebalancing Shards
-
-If tests in one of your shards take longer, you can manually assign less work to it using the `--shard-weights` option:
-
-```bash
-npx playwright test --shard=x/4 --shard-weights=3:2:3:3
-```
-
-Make sure you pass the same `--shard-weights` value to all shards.
-In this example, Shard 2 gets assigned less tests because they take longer, evening out the full duration.
-
-In the Speedboard section of the combined HTML report, you can see a visualisation of your individual shard durations, and a recommendation for how to set your shard weights.
-
-![Speedboard Shard Diagram](./images/speedboard-shards.png)
-
 ## Merging reports from multiple shards
 
 In the previous example, each test shard has its own test report. If you want to have a combined report showing all the test results from all the shards, you can merge them.

docs/src/test-sharding-js.md

@@ -14,14 +14,14 @@
   limitations under the License.
 */
 
-.barchart-bar {
+.gantt-bar {
   transition: opacity 0.2s;
   cursor: pointer;
   outline: none;
 }
 
-.barchart-bar:hover,
-.barchart-bar:focus {
+.gantt-bar:hover,
+.gantt-bar:focus {
   opacity: 0.8;
   stroke: var(--color-fg-default);
   stroke-width: 2;