Routing reference test runs on today's routing loop; perf bench reads exported saves

The reference test from #106 now patches the loop as it looks after #96
(business-first passes) and also covers prefer, prepend and protect
customers. It catches an exit index that ignores RouteScript preferences,
which the shortcut test in speed.test.js cannot see.

npm run perf -- --save accepts files from the in-game Export save, so
scripts/benchmark-performance.js and its README section are not needed.
The performance skill documents --save, both equivalence tests and how to
handle noisy runs.
This commit is contained in:
Frederic Jung 2026-10-26 08:40:20 +01:00
parent 2366bb9a8f
commit e5deb90968
No known key found for this signature in database
6 changed files with 22 additions and 63 deletions

View file

@ -1,9 +1,9 @@
// Deliberately scan all exits for every flow, without byRank, local-city
// shortcuts or first-group selection. Stable sorting preserves E order on ties.
module.exports = `
for(const rank of RANKS){
for(const pass of PASSES||[null])for(const rank of RANKS){
for(f of flows){
if(f.rem<=1e-6||(rank===0&&!f.cache))continue;
if(f.rem<=1e-6||(rank===0&&!f.cache)||(pass&&!pass(f)))continue;
Dv=SDi[f.ci];Pv=SPi[f.ci];PP=walked[f.ci]||(walked[f.ci]=[]);vis++;
const candidates=E.filter(e=>e.used<e.cap&&e.ok(f.n)&&
(TR?eff(e,f.n):base(e))===rank&&Dv[CIDX[e.city]]<Infinity)