mirror of
https://github.com/samsonjs/spirit-tracker.git
synced 2026-03-25 09:25:51 +00:00
link sku
This commit is contained in:
parent
37ebe4862a
commit
76ed7362b8
1 changed files with 85 additions and 36 deletions
|
|
@ -582,33 +582,52 @@ function recommendSimilar(allAgg, pinned, limit, otherPinnedSku, mappedSkus, isI
|
|||
const pinRawToks = tokenizeQuery(pinNorm);
|
||||
const pinToks = filterSimTokens(pinRawToks);
|
||||
|
||||
// "brand" = first meaningful token (usually distillery)
|
||||
const pinBrand = pinToks[0] || "";
|
||||
const pinAge = extractAgeFromText(pinNorm);
|
||||
|
||||
const pinnedSmws = smwsKeyFromName(pinned.name || "");
|
||||
const scored = [];
|
||||
|
||||
// Precompute set for cheap overlap checks
|
||||
const pinTokSet = new Set(pinToks);
|
||||
|
||||
// ---- Tuning knobs ----
|
||||
const MAX_SCAN = 4500; // cap scan cost if your catalog gets huge
|
||||
const MAX_CHEAP_KEEP = 220; // keep only top cheap candidates
|
||||
const MAX_FINE = 40; // run expensive similarityScore on only top N
|
||||
const CHEAP_MIN = 0.35; // drop obviously bad cheap matches
|
||||
const REQUIRE_SHARED_IF_BRAND_DIFF = 2; // if first token differs, require at least this many shared tokens
|
||||
// ----------------------
|
||||
|
||||
// Fast insert into top-K list (descending)
|
||||
function pushTopK(arr, item, k) {
|
||||
arr.push(item);
|
||||
// tiny lists; sort is fine
|
||||
if (arr.length > k) {
|
||||
arr.sort((a, b) => b.s - a.s);
|
||||
arr.length = k;
|
||||
}
|
||||
}
|
||||
|
||||
const cheap = [];
|
||||
let scanned = 0;
|
||||
|
||||
for (const it of allAgg) {
|
||||
if (!it) continue;
|
||||
if (scanned++ > MAX_SCAN) break;
|
||||
|
||||
const itSku = String(it.sku || "");
|
||||
if (!itSku) continue;
|
||||
|
||||
if (itSku === pinnedSku) continue;
|
||||
if (otherSku && itSku === otherSku) continue;
|
||||
if (storesOverlap(pinned, it)) continue;
|
||||
|
||||
if (typeof isIgnoredPairFn === "function" && isIgnoredPairFn(pinnedSku, itSku)) continue;
|
||||
|
||||
// SMWS exact NUM.NUM match => force to top (requires SMWS + code match)
|
||||
// SMWS exact code match: keep (still cheap)
|
||||
if (pinnedSmws) {
|
||||
const k = smwsKeyFromName(it.name || "");
|
||||
if (k && k === pinnedSmws) {
|
||||
const stores = it.stores ? it.stores.size : 0;
|
||||
const hasPrice = it.cheapestPriceNum != null ? 1 : 0;
|
||||
const s = 1e9 + stores * 10 + hasPrice; // tie-break within exact matches
|
||||
scored.push({ it, s });
|
||||
pushTopK(cheap, { it, s: 1e9 + stores * 10 + hasPrice }, MAX_CHEAP_KEEP);
|
||||
continue;
|
||||
}
|
||||
}
|
||||
|
|
@ -618,55 +637,85 @@ function recommendSimilar(allAgg, pinned, limit, otherPinnedSku, mappedSkus, isI
|
|||
|
||||
const itRawToks = tokenizeQuery(itNorm);
|
||||
const itToks = filterSimTokens(itRawToks);
|
||||
const itBrand = itToks[0] || "";
|
||||
if (!itToks.length) continue;
|
||||
|
||||
const itBrand = itToks[0] || "";
|
||||
const firstMatch = pinBrand && itBrand && pinBrand === itBrand;
|
||||
|
||||
// Cheap “shared tokens” count
|
||||
let shared = 0;
|
||||
for (const t of itToks) if (pinTokSet.has(t)) shared++;
|
||||
|
||||
// If brands differ, don’t block completely — but avoid total junk
|
||||
if (!firstMatch && shared < REQUIRE_SHARED_IF_BRAND_DIFF) continue;
|
||||
|
||||
// Cheap score first (no Levenshtein)
|
||||
let s0 = fastSimilarityScore(pinRawToks, itRawToks, pinNorm, itNorm);
|
||||
if (s0 <= CHEAP_MIN) continue;
|
||||
|
||||
// Soft first-token mismatch penalty based on containment (fastSimilarityScore already uses contain)
|
||||
if (!firstMatch) {
|
||||
const contain = tokenContainmentScore(pinRawToks, itRawToks); // 0..1
|
||||
const smallN = Math.min(pinToks.length || 0, itToks.length || 0);
|
||||
|
||||
let mult = 0.10 + 0.95 * contain;
|
||||
if (smallN <= 3 && contain < 0.78) mult *= 0.18;
|
||||
|
||||
s0 *= Math.min(1.0, mult);
|
||||
if (s0 <= CHEAP_MIN) continue;
|
||||
}
|
||||
|
||||
// Apply size penalty early (cheap stage) so mismatched sizes don’t waste fine scoring
|
||||
if (typeof sizePenaltyFn === "function") {
|
||||
s0 *= sizePenaltyFn(pinnedSku, itSku);
|
||||
if (s0 <= CHEAP_MIN) continue;
|
||||
}
|
||||
|
||||
// Apply age penalty/boost early too (cheap)
|
||||
const itAge = extractAgeFromText(itNorm);
|
||||
if (pinAge && itAge) {
|
||||
if (pinAge === itAge) s0 *= 1.6;
|
||||
else s0 *= 0.22;
|
||||
if (s0 <= CHEAP_MIN) continue;
|
||||
}
|
||||
|
||||
pushTopK(cheap, { it, s: s0, itNorm, itRawToks }, MAX_CHEAP_KEEP);
|
||||
}
|
||||
|
||||
// Fine stage: expensive scoring only on top candidates
|
||||
cheap.sort((a, b) => b.s - a.s);
|
||||
const fine = [];
|
||||
for (const x of cheap.slice(0, MAX_FINE)) {
|
||||
const it = x.it;
|
||||
const itSku = String(it.sku || "");
|
||||
|
||||
// score first
|
||||
let s = similarityScore(base, it.name || "");
|
||||
if (s <= 0) continue;
|
||||
|
||||
// soft first-token mismatch penalty (never blocks)
|
||||
const contain = tokenContainmentScore(pinRawToks, itRawToks); // 0..1
|
||||
const firstMatch = pinBrand && itBrand && pinBrand === itBrand;
|
||||
|
||||
if (!firstMatch) {
|
||||
const smallN = Math.min(pinToks.length || 0, itToks.length || 0);
|
||||
|
||||
// 0.10..~1.0 based on containment
|
||||
let mult = 0.10 + 0.95 * contain;
|
||||
|
||||
// Short names: first token matters more unless containment is *very* high
|
||||
if (smallN <= 3 && contain < 0.78) mult *= 0.18;
|
||||
|
||||
s *= Math.min(1.0, mult);
|
||||
if (s <= 0) continue;
|
||||
}
|
||||
|
||||
// size penalty (your existing hook)
|
||||
// Keep these here too for correct ordering vs other candidates
|
||||
if (typeof sizePenaltyFn === "function") {
|
||||
s *= sizePenaltyFn(pinnedSku, itSku);
|
||||
if (s <= 0) continue;
|
||||
}
|
||||
|
||||
// age boost/penalty (existing)
|
||||
const itAge = extractAgeFromText(itNorm);
|
||||
const itAge = extractAgeFromText(x.itNorm || normSearchText(it.name || ""));
|
||||
if (pinAge && itAge) {
|
||||
if (pinAge === itAge) s *= 2.0;
|
||||
else s *= 0.15;
|
||||
}
|
||||
|
||||
// unknown boost (existing)
|
||||
const aUnknown = pinnedSku.startsWith("u:");
|
||||
const bUnknown = itSku.startsWith("u:");
|
||||
if (aUnknown || bUnknown) s *= 1.12;
|
||||
|
||||
if (s > 0) scored.push({ it, s });
|
||||
|
||||
fine.push({ it, s });
|
||||
}
|
||||
|
||||
scored.sort((a, b) => b.s - a.s);
|
||||
return scored.slice(0, limit).map((x) => x.it);
|
||||
fine.sort((a, b) => b.s - a.s);
|
||||
return fine.slice(0, limit).map((x) => x.it);
|
||||
}
|
||||
|
||||
|
||||
function computeInitialPairsFast(allAgg, mappedSkus, limitPairs, isIgnoredPairFn) {
|
||||
const itemsAll = allAgg.filter((it) => !!it);
|
||||
|
||||
|
|
@ -1191,7 +1240,7 @@ export async function renderSkuLinker($app) {
|
|||
return recommendSimilar(
|
||||
allAgg,
|
||||
otherPinned,
|
||||
60,
|
||||
30,
|
||||
otherSku,
|
||||
mappedSkus,
|
||||
isIgnoredPair,
|
||||
|
|
|
|||
Loading…
Reference in a new issue