Repository navigation
Conversation
`at_each()` carries a `pause 0.5` before every capture that nothing in the repo explains; it arrived with the helper in #508. Measured on the fleet, the pause serves one server: wayvnc composites the pointer on the move and takes it back out again between 60ms and 80ms later. Near/far pairs over one connection, as `at_each()` takes them, carried a pointer 12/20 with the pause omitted and 20/20 at 0.05 and 0.06, then 0/20 at every pause from 0.08 to 0.5. Separate connections per capture gave the same edge. Every other fleet server was clean 10/10 with no pause at all. What the pause looks like it must be for -- the detection window for a server that starts painting again -- it is not. With x11vnc and libvncserver-example forced back into painting by `--cursor server`, the first full capture after the move already carried the pointer over 30 moves each, at a median 16ms and 61ms, which is the capture round trip itself (20ms and 59ms with nothing moving); the CLI comparison caught the painted pointer 15/15 at every pause tried, including none. So CURSOR_SETTLE is 0.2, a little over twice what wayvnc needs, with room for slower hardware. The cursor module drops from 23.5s to 16.1s locally, green over five consecutive runs. `CursorShapeOffered` and `CursorPositionIndependent` keep their 0.5s: both also run against the Windows servers of the os-servers workflow, which none of this measured. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #548 +/- ##
=======================================
Coverage 90.82% 90.82%
=======================================
Files 36 36
Lines 3509 3509
Branches 429 429
=======================================
Hits 3187 3187
Misses 239 239
Partials 83 83
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
This branch has not been deployed
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.
Cuts the cursor tests' unexplained
pause 0.5to a measuredCURSOR_SETTLE = 0.2, saving about 7s of the fleet job.The pause exists for wayvnc alone, which the spec table listed as never painting: it composites the pointer on a move and removes it 60–80ms later. Pairs were dirty 20/20 at 0.06s and clean 0/20 from 0.08s. Every other fleet server was clean with no pause at all. The pause is not a detection window for a server that paints: a painted pointer shows up within the capture's own round trip.
Measured locally against the Docker fleet, which CI doesn't do;
specs/cursor.mdhas the tables. The two 0.5s pauses inutils.pyalso run on the Windows os-servers, which weren't measured, so they're unchanged.🤖 Generated with Claude Code