fix(e2e): traces really are always captured now (completes DEC-073) - #51
Merged
Conversation
v2.1.4 flipped the trace default to "on" in playwright-base.config.ts and I
reported DEC-073 as delivered. It was not: two other settings still forced
retain-on-failure and both took precedence over the one I changed.
1. The reusable workflow hardcoded
BDD_TRACE: ${{ inputs.capture-all-traces && "on" || "retain-on-failure" }}
so every consumer run explicitly passed retain-on-failure into the container.
2. The BDD suite has its OWN config (e2e/bdd/playwright.config.ts) — the suite
consumers actually run via run-bdd.sh — whose default was also
retain-on-failure. Same duplicated-setting pattern as the collector: I fixed
one copy and missed the others.
Proven by share-chart sync PR #19 on v2.1.4: "copied 0 trace(s), 4 screenshot(s),
4 video(s)" — screenshots and videos flowed (the collector fix landed) while
traces did not, on a fully green run.
Replaces the boolean capture-all-traces flip with a trace-mode string input
(default "on", accepts any Playwright mode) so there is a real escape hatch for
disk-pressure cases without silently reverting the decision.
capture-all-traces is kept so existing callers keep working, but it can no
longer suppress traces.
Merged
omar-nahhas
added a commit
that referenced
this pull request
Aug 4, 2026
Cuts the release carrying the DEC-073 completion (#51). Consumers pin the devkit submodule to a release tag, and the BDD suite config — where the surviving retain-on-failure default lived — ships inside that submodule, so the trace fix only reaches the fleet once this tag exists.
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
1 participant
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.
v2.1.4 flipped the trace default to
oninplaywright-base.config.tsand I reported DEC-073 as delivered. It was not. Two other settings still forcedretain-on-failure, and both took precedence over the one I changed.BDD_TRACE: ${{ inputs.capture-all-traces && 'on' || 'retain-on-failure' }}— so every consumer run explicitly passed
retain-on-failureinto the container.e2e/bdd/playwright.config.ts— the suite consumers actually run viarun-bdd.sh— whose default was alsoretain-on-failure. Same duplicated-setting pattern as the collector: I fixed one copy and missed the others.Evidence this is real
All four consumers, now on v2.1.4, fully green:
0 trace(s), 12 screenshot(s), 12 video(s)0 trace(s), 11 screenshot(s), 11 video(s)0 trace(s), 9 screenshot(s), 9 video(s)0 trace(s), 4 screenshot(s), 4 video(s)Screenshots and videos flow (the v2.1.4 collector fix landed), traces are still zero on green runs — exactly the defect DEC-073 was written to eliminate.
Change
Replaces the boolean flip with a
trace-modestring input (defaulton, accepts any Playwright mode), so there is a real escape hatch for disk-pressure cases without silently reverting the decision.capture-all-tracesis kept so existing callers keep working, but it can no longer suppress traces.Merging this needs a v2.1.5 tag to reach consumers — the workflow change flows immediately via
@main, but the BDD config lives in the pinned submodule.