test(flutter): Remove flaky frames measurement tests#3783
Merged
Conversation
The two 'measures frames' integration tests drive real frames through the binding's wall-clock Stopwatch, so on a loaded CI runner a 100ms slow frame can stretch past the 700ms frozen threshold and be reclassified. Earlier hardening (#3750) relaxed the assertions to combined counts and lower bounds, but the tests still flake. Their coverage is redundant: classification math is covered by sentry_delayed_frames_tracker_test, attribute/measurement application by span_frame_metrics_collector_test, and callback registration and cleanup by the remaining tests in this file. Remove the two cases along with their now-dead helpers and the unused material import. Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
Contributor
Semver Impact of This PR⚪ None (no version bump detected) 📋 Changelog PreviewThis is how your changes will appear in the changelog. Features
Enhancements
DependenciesDeps
Internal Changes
🤖 This preview updates automatically when you update the PR. |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #3783 +/- ##
==========================================
+ Coverage 87.07% 91.46% +4.39%
==========================================
Files 336 105 -231
Lines 12064 3763 -8301
==========================================
- Hits 10505 3442 -7063
+ Misses 1559 321 -1238
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Harness. |
Contributor
Android Performance metrics 🚀
|
| Revision | Plain | With Sentry | Diff |
|---|---|---|---|
| e2d675d | 457.92 ms | 529.17 ms | 71.25 ms |
| 5367726 | 380.04 ms | 361.33 ms | -18.71 ms |
| e200a70 | 433.06 ms | 437.00 ms | 3.94 ms |
| c97f488 | 502.43 ms | 492.47 ms | -9.97 ms |
| 5a95d04 | 378.92 ms | 364.33 ms | -14.59 ms |
| 8af916c | 367.83 ms | 379.60 ms | 11.77 ms |
| 73a3c38 | 478.18 ms | 526.62 ms | 48.44 ms |
| 40751bf | 421.09 ms | 424.82 ms | 3.73 ms |
| f872f8e | 402.38 ms | 416.20 ms | 13.82 ms |
| 1fff351 | 423.68 ms | 408.96 ms | -14.72 ms |
App size
| Revision | Plain | With Sentry | Diff |
|---|---|---|---|
| e2d675d | 6.54 MiB | 7.69 MiB | 1.15 MiB |
| 5367726 | 14.31 MiB | 15.49 MiB | 1.19 MiB |
| e200a70 | 13.93 MiB | 15.18 MiB | 1.25 MiB |
| c97f488 | 14.31 MiB | 15.49 MiB | 1.19 MiB |
| 5a95d04 | 13.93 MiB | 15.06 MiB | 1.13 MiB |
| 8af916c | 14.31 MiB | 15.56 MiB | 1.25 MiB |
| 73a3c38 | 6.54 MiB | 7.69 MiB | 1.15 MiB |
| 40751bf | 14.31 MiB | 15.49 MiB | 1.19 MiB |
| f872f8e | 13.93 MiB | 15.18 MiB | 1.25 MiB |
| 1fff351 | 14.31 MiB | 15.49 MiB | 1.19 MiB |
Contributor
iOS Performance metrics 🚀
|
| Revision | Plain | With Sentry | Diff |
|---|---|---|---|
| d0371bb | 1258.85 ms | 1259.44 ms | 0.60 ms |
| dc53d48 | 1252.06 ms | 1255.61 ms | 3.55 ms |
| c26ed0a | 1244.11 ms | 1263.85 ms | 19.75 ms |
| 81f83eb | 1259.53 ms | 1273.39 ms | 13.86 ms |
| 40c8f93 | 1234.27 ms | 1261.98 ms | 27.71 ms |
| dbd526b | 1244.78 ms | 1259.02 ms | 14.24 ms |
| 73dca78 | 1246.65 ms | 1265.42 ms | 18.76 ms |
| 737ef5d | 1255.65 ms | 1252.24 ms | -3.41 ms |
| 3801d52 | 1267.76 ms | 1266.10 ms | -1.65 ms |
| 5367726 | 1211.00 ms | 1216.73 ms | 5.73 ms |
App size
| Revision | Plain | With Sentry | Diff |
|---|---|---|---|
| d0371bb | 5.73 MiB | 6.18 MiB | 464.41 KiB |
| dc53d48 | 5.65 MiB | 6.09 MiB | 446.96 KiB |
| c26ed0a | 5.53 MiB | 5.97 MiB | 453.76 KiB |
| 81f83eb | 7.86 MiB | 9.44 MiB | 1.58 MiB |
| 40c8f93 | 5.53 MiB | 6.00 MiB | 479.94 KiB |
| dbd526b | 7.86 MiB | 9.44 MiB | 1.58 MiB |
| 73dca78 | 7.86 MiB | 9.44 MiB | 1.58 MiB |
| 737ef5d | 5.73 MiB | 6.18 MiB | 466.16 KiB |
| 3801d52 | 5.73 MiB | 6.17 MiB | 455.54 KiB |
| 5367726 | 5.73 MiB | 6.17 MiB | 455.87 KiB |
Contributor
There was a problem hiding this comment.
Pull request overview
This PR removes two flaky Flutter integration tests that measured delayed frames via real wall-clock timing, reducing nondeterministic CI failures while keeping deterministic coverage in unit tests and the remaining integration tests in this file.
Changes:
- Deleted the two
measures frameswidget/integration tests (stream + static lifecycle) and their helper utilities. - Removed now-unused Flutter Material import (
package:flutter/material.dart). - Kept the remaining frames-tracking integration tests that validate callback registration and cleanup behavior.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
denrase
approved these changes
Jun 24, 2026
6 tasks
This was referenced Jun 25, 2026
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.
📜 Description
Removes the two
measures framesintegration tests (with stream lifecycleandwith static lifecycle) fromframes_tracking_integration_test.dart, along with their now-dead helpers (simulateFrames,buildTestApp, thehub/slowFrame/frozenFramesetup) and the now-unusedpackage:flutter/material.dartimport.💡 Motivation and Context
These tests drive real frames through the binding's wall-clock
Stopwatch, so on a loaded CI runner a 100ms "slow" frame can stretch past the 700ms "frozen" threshold and be reclassified — producing nondeterministic slow/frozen counts. Earlier hardening (#3750) relaxed the assertions to combined counts and lower bounds, but the tests still flake (they failed again on a recent unrelated PR run).Their coverage is redundant with existing deterministic tests:
sentry_delayed_frames_tracker_test.dartspan_frame_metrics_collector_test.dartThe only coverage lost is the binding's real-time
Stopwatchmeasurement end to end — the ~6-line block that is the sole source of the flake.💚 How did you test it?
flutter analyzereports no issues and the remaining 11 tests in the file pass locally.📝 Checklist
sendDefaultPiiis enabled🔮 Next steps
Optionally add a focused, deterministic binding-measurement test (behind a small injectable-time seam) to recover that single coverage gap without reintroducing wall-clock flakiness.