Repository navigation
fix: dismiss the move to spark intro in outcome waits - #306
Conversation
tapAndExpect, tapStationaryAndExpect and typeAndExpect waited for their target without the Move to Spark intro recovery that tap() has, so the intro covering home failed send_2 and send_3.
Codex Review Summary: ✅ CompletedThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
talosmachina
left a comment
There was a problem hiding this comment.
No findings. Wraps the target waits in tapAndExpect, tapStationaryAndExpect and typeAndExpect in withMoveToSparkIntroRecovery, so the Move to Spark intro is dismissed with Later and the wait retried once, matching what tap() already does. Reviewed 7830527, light tier (static).
What I checked, and 4 candidates I ruled out
Read in full: test/helpers/outcome.ts; in test/helpers/actions.ts, tap, dismissMoveToSparkIntroIfShown, withMoveToSparkIntroRecovery
Build: skipped, the PR runner could not read the author's permission on this repo and fails closed, so tsc, eslint and the unit tests were not run here
CI: no checks reported on the branch
Ruled out
- Stale handle on retry in
tapAndExpect:currentis the lazyelementByIdchain, re-resolved bywaitForDisplayed, the same patterntap()uses with the same recovery. - Stale readiness state in
tapStationaryAndExpect: each poll resetspreviousandreadyand a covered target never sets them, so the retry starts clean and two matching samples are still required. - Doubled wall time: a recovered wait can take two timeouts, but only when the intro was actually dismissed; the outcome wait in
actAndExpectstarts after the action, so its own timeout is unchanged. - Intro appearing after the click, during the outcome wait: no recovery there, but nothing shows the intro surfacing over an open sheet, and the reported failures were in the pre-tap wait.
Merge confidence: 4/5, not built here and no unit test drives the recovery path in the outcome helpers.
The intro can surface on home a moment after Close, so send_2's total balance reads now run under the same recovery.
|
28b00ed: |
pwltr
left a comment
There was a problem hiding this comment.
Reviewed 28b00ed.
No consequential correctness finding was established in the readiness-recovery helpers and payment-balance assertion changes. Recovery retries waits and reads; the inspected clicks, input mutations and payment submissions remain outside the retry callbacks. This conclusion covers static source inspection.
Review details
General review
Model: GPT-6.1 Sol
Reasoning effort: high
|
Merging: approved by @pwltr at 28b00ed, and the repository runs no PR checks. synonymdev/bitkit-android#1774's E2E runs fall back to main once this lands. |
Description
This PR adds the Move to Spark intro recovery to the selector action/outcome waits.
tap()already dismisses the intro with Later when it covers its target (#297).tapAndExpect,tapStationaryAndExpectandtypeAndExpectfrom #296 wait for their target without that recovery. The send specs now tap Send throughtapAndExpect, so when the intro surfaces on home the tap times out.This failed
send_3on synonymdev/bitkit-android#1774: the intro showed as the Send tap started, and the wait ran out 30 s later with the intro still covering home.send_2reads the total balance right after each Close. The intro can surface on home a moment later and hide it, so those reads now run under the same recovery.Test
tsc --noEmit, eslint, prettier andtsx --test test/unit/outcome.test.tspass.