Skip to content

test: repro #670 — labeled rail nets get long bus traces instead of net labels#704

Closed
DPS0340 wants to merge 1 commit into
tscircuit:mainfrom
DPS0340:repro/670-long-trace
Closed

test: repro #670 — labeled rail nets get long bus traces instead of net labels#704
DPS0340 wants to merge 1 commit into
tscircuit:mainfrom
DPS0340:repro/670-long-trace

Conversation

@DPS0340

@DPS0340 DPS0340 commented Jul 24, 2026

Copy link
Copy Markdown

Reproduction PR for #670, following the repo's repro-first workflow (per the review guidance on #690).

Pins the current behavior on the #670 board (_board#1332 input): LongDistancePairSolver connects pins on the labeled V3_3/GND rails with long orthogonal bus traces even though every one of those pins already receives a net label (V3_3 declared y+, GND y-):

JP1.2-JP2.2  len=12.7   (8.0,2.3)->(8.0,2.5)->(5.5,2.5)->(5.5,-7.6)
D2.2-JP6.1   len=11.3
U1.2-JP6.2   len=9.8

The test asserts exactly these three traces exist (with a visual snapshot), so a fix flips the assertion to .toEqual([]).

Proposed fix in #702 (separate PR): cap long-distance neighbor candidates at maxMspPairDistance for labeled nets with >2 pins — the same threshold MspConnectionPairSolver already uses when it decides a pair should stay label-connected.

…tead of net labels

Pins the current behavior on the tscircuit#670 board: LongDistancePairSolver
connects pins on the labeled V3_3/GND rails (JP1.2-JP2.2, D2.2-JP6.1,
U1.2-JP6.2 — each 8-10 units apart, every pin already receiving a net
label) with 9-13 unit orthogonal bus traces.

A fix should flip the assertion to .toEqual([]).
@vercel

vercel Bot commented Jul 24, 2026

Copy link
Copy Markdown

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated (UTC)
schematic-trace-solver Ready Ready Preview, Comment Jul 24, 2026 11:10am

Request Review

@DPS0340

DPS0340 commented Jul 25, 2026

Copy link
Copy Markdown
Author

Superseded by #702, which contains this same regression test (tests/repros/repro-long-trace-instead-of-netlabel-670.test.ts) plus the actual fix. Closing this to keep the review queue focused on the PR that resolves #670.

@DPS0340 DPS0340 closed this Jul 25, 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.

1 participant