Skip to content

fix(sgcr): straighten fork in-line child at the parity boundary#206

Merged
ivanmkc merged 1 commit into
masterfrom
fix/sgcr-fork-spine-straighten
Jun 22, 2026
Merged

fix(sgcr): straighten fork in-line child at the parity boundary#206
ivanmkc merged 1 commit into
masterfrom
fix/sgcr-fork-spine-straighten

Conversation

@ivanmkc

@ivanmkc ivanmkc commented Jun 22, 2026

Copy link
Copy Markdown
Owner

A fork parent's in-line child connector was left diagonal (~9px) when the target port fell a parity-hair outside the child's narrower face — the forced-edge straightening skipped instead of snapping. Now snaps with a 1px parity slack (clamped to face); genuine off-side edges still turn. Stress fuzz 10,060 layouts ZERO violations, vitest 26/26, arch spine now one straight line.

…p, don't skip)

A fork parent distributes its child ports around centre, so the in-line
child's connector was straight only when its target port fell strictly
inside the (possibly narrower / ½-grid-parity-shifted) child face. At the
boundary the forced-edge straightening SKIPPED, leaving a ~9px diagonal jog
even though parent and child were centre-aligned. Now snap with a 1px parity
slack, clamped to the face — straight when reachable, ≤1px at the boundary —
while still skipping genuinely-off-to-the-side edges (which must turn).

Verified: stress fuzz 10,060 layouts zero violations; vitest 26/26; the
arch demo's edge→api→svc→cache spine now renders as one straight line.
@ivanmkc ivanmkc merged commit eeaf49f into master Jun 22, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants