Files
openobserve/tests
Shrinath RaoandClaude Opus 4.8 8f81d41a80 test: heal logs Bug #4426 wrap-toggle assertion (switch to Table tab) (#12937)
## What

Heals the wrap-toggle assertion in the **Logs-Regression** shard — the
`text wrap toggle should not cause column misalignment on saved view
switch` test (Bug #4426) in `logs-bugs.spec.js`.

## Root cause

The test opens the log detail sidebar and asserts the wrap toggle is
visible. But the sidebar opens on the **JSON tab by default** (Bug
#9724), and the wrap toggle only renders in the **Table tab**. So the
assertion (`verifyWrapToggleVisibleInTableTab`) always waited out its
full `expect` timeout against the hidden toggle, then fell into the
`catch` and logged:

```
Wrap toggle not found in current log detail: Timed out 30000ms waiting for
expect(locator).toBeVisible()
Locator: locator('[data-test="log-detail-wrap-values-toggle-btn"]')
Received: hidden
```

This is the `Wrap toggle not found` warning visible in the scheduled
Logs-Regression run
[28373343106](https://github.com/openobserve/openobserve/actions/runs/28373343106).
The toggle assertion never actually verified anything and burned ~30s of
the shard's wall-clock as noise.

## Fix

Click the Table tab before asserting the toggle's visibility, so the
toggle actually renders and the check is meaningful (and fast).

```js
await pm.logsPage.clickLogDetailTableTab();
await pm.logsPage.verifyWrapToggleVisibleInTableTab();
```

## Verification

- Targeted run against a local OpenObserve instance: **passes in
10.9s**, now logging `✓ Wrap toggle is visible in Table tab` instead of
the timeout warning.
- Full Logs-Regression shard (all 4 spec files, 67 tests) run locally
beforehand: **67/67 passed** — no regressions.

```
npx playwright test playwright-tests/RegressionSet/Logs/logs-bugs.spec.js \
  -g "text wrap toggle should not cause column misalignment"
  ✓  ...logs-bugs.spec.js:1818:3 › ... @bug-4426 (10.9s)
  1 passed
```

## Note

The linked run's job-level failure was a runner **shutdown/cancel
signal** (infra reclaim) on the Logs-Regression shard plus a separate
failure in the `Merge Reports and Upload to TestDino` job — neither is a
test-code failure (all 67 logs tests pass locally). This PR addresses
the one genuine test-level defect surfaced in that run.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-29 16:31:34 +00:00
..
2024-09-22 17:57:47 +08:00