solidjs / solid / 37378208965

76%

main: 89%

push

github

fix(signals): the lane-membership ruling — F8 and adopted staging (F6 stopped on A18 (c)) (#3806)

* docs(signals): rule the lane-membership amendment (F8, F6, adopted staging)

The 2026-10-05 paragraph under the L2 section, ruled by the maintainer
(reading A for all three): a render effect off a lane that reads a lane
node directly is a stale reader of the lane whether or not the lane is
held (F8); a leaf reading through a boundary showing content is the
lane's frame reader through it (F6, A33); an on-screen reader whose
removal an adopted write stages follows the lane's work (adopted
staging). Derivations between the lane node and the leaf stay born held
under A29 (#3761). RULES-INDEX regenerated (line shifts only).

Co-authored-by: Claude via Cursor <noreply@cursor.com>
Co-authored-by: Cursor <cursoragent@cursor.com>

* fix(signals): a leaf reading a lane member in flight is a stale reader of the lane, held or not (fuzzer F8)

A mainline mount over a lane derivation whose re-ask nobody had observed
(the seam left the lane unheld) published its mount control while the
child entered the lane and waited on the flight: control and content
disagreed (fuzzer S2 "mount control"). `laneRead`'s stale-reader arm was
gated on `l._held`; it now also takes a render effect off the lane
reading a member whose own flight is up. The child shows the screen (the
lane's revealed derivation) beside its control, and re-derives at the
reveal.

Rule: A15 lanes corollary (#3460) and the lane-membership ruling
(2026-10-05, reading A): "a stale reader of the lane whether or not the
lane is held at that moment". Through a memo the mount stays born held
(A29, #3761) — pinned beside it.

Removing the gate outright livelocks the verdict lanes (F7b, F12: an
unheld, unblocked verdict lane re-queues its stale readers at every
reveal); the arm is widened only to a member with a flight up.

Co-authored-by: Claude via Cursor <noreply@cursor.com>
Co-authored-by: Cursor <cursoragent@cur... (continued)

649 of 925 branches covered (70.16%)

Branch coverage included in aggregate %.

962 of 1195 relevant lines covered (80.5%)

27.69 hits per line