• Home
  • Features
  • Pricing
  • Docs
  • Announcements
  • Sign In

solidjs / solid / 41676814354
79%
main: 89%

Build:
Build:
LAST BUILD BRANCH: next
DEFAULT BRANCH: main
Ran 03 Nov 2026 04:49PM UTC
Jobs 1
Files 12
Run time 1min
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst

03 Nov 2026 04:41PM UTC coverage: 78.249%. Remained the same
41676814354

push

github

web-flow
fix(signals): a latest() read of a flight does not hold a Loading on switch (#4661) (#4663)

* fix(signals): a latest() read of a flight does not hold a Loading on switch (#4661)

A memo whose last run read `latest(() => list())` is still the hold's verdict
reader on its next run (D-25). Its read of the switch's own staging (`f`) is
the screen's, and that stale-reader mark (REACTIVE_FRAME_READ) made
`blockedBy` count it as a frame reader of `list`'s new flight too — a flight
it read only inside `latest()`, ahead of it (F-6). The switch was held and
the `on` fallback never showed.

At the seam, a verdict reader that read a flight of the frame only ahead of
it this pass (and no flight plainly: `observeFlight`, `frameRead`) drops the
mark and its rerun on the frame's landing: it holds nothing, keeps its
committed value (maintainer, 2026-11-03: "intentional tearing"), and the
flight's own landing re-derives it. Engine only (lanes.ts, verdict.ts).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0117BoWNSSgtemsSYsaX2hXE

* perf(signals): track #4661's read-ahead readers with one list, no WeakMap

The seam drops the stale-reader mark of a verdict reader whose pass read a
flight inside latest() before any plain flight read routed it (F-6). A plain
read of a flight first routes the pass (REACTIVE_LANE_READ), so the read-ahead
push is skipped and the reader still holds (A10, A15); later plain reads in a
routed pass are the lane's and mark nothing. That replaces the pass-generation
tuples, the plain-read WeakMap and the laneServed hook. A new variant pins
the plain-read-first hold.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0117BoWNSSgtemsSYsaX2hXE

* fix(signals): a memo's plain read of a flight after its latest() window still holds (#4661)

A window read routes the pass into the verdict lane, and `read` then served a
later plain read of a flight... (continued)

807 of 1108 branches covered (72.83%)

Branch coverage included in aggregate %.

1132 of 1370 relevant lines covered (82.63%)

30.26 hits per line

Jobs
ID Job ID Ran Files Coverage
1 41676814354.1 03 Nov 2026 04:49PM UTC 12
78.25
GitHub Action Run
Source Files on build 41676814354
  • Tree
  • List 12
  • Changed 0
  • Source Changed 0
  • Coverage Changed 0
Coverage ∆ File Lines Relevant Covered Missed Hits/Line Branch Hits Branch Misses
  • Back to Repo
  • Github Actions Build #41676814354
  • 061b013c on github
  • Prev Build on next (#41673859564)
  • Next Build on next (#41680892396)
  • Delete
STATUS · Troubleshooting · Open an Issue · Sales · Support · CAREERS · ENTERPRISE · START FREE TRIAL · SCHEDULE DEMO
ANNOUNCEMENTS · TWITTER · TOS & SLA · Supported CI Services · What's a CI service? · Automated Testing

© 2026 Coveralls, Inc