Method getSegmentMatches

src/index.ts:1321–1397

* Graph-derived prompt matching for the front-load hook's MEDIUM tier: * which indexed symbols do these prose words name? "state machine des * commandes" → `OrderStateMachine`, in any human language whose technical * nouns are Latin script — no keyword list involved. * * Precision com

(words: string[], limit: number = 6)

   * returned symbol is guaranteed to exist right now.
   */
  getSegmentMatches(words: string[], limit: number = 6): SegmentMatch[] {
    if (words.length === 0) return [];
    // Variant → original word (plural folding), for coverage accounting.
    const variantToWord = new Map<string, string>();
    for (const word of words) {
      for (const variant of segmentLookupVariants(word)) {
        if (!variantToWord.has(variant)) variantToWord.set(variant, word);
      }
    }
    const variants = [...variantToWord.keys()];

    // Tier A: co-occurrence. The SQL folds variants back to their original
    // word (#1146), so minWords=2 means two distinct PROMPT WORDS — a name
    // matching both `service` and `services` can't tie with (or crowd past
    // the LIMIT) a genuine two-word match. The JS re-check below recomputes
    // the fold from live segments as the honesty layer.
    const variantPairs = [...variantToWord.entries()].map(([segment, word]) => ({ segment, word }));
    const candidates: Array<{ name: string; matchedWords: Set<string> }> = [];
    for (const hit of this.queries.getSegmentCoOccurrence(variantPairs, 2, 24)) {
      const matched = this.wordsMatchingName(hit.name, variantToWord);
      if (matched.size >= 2) candidates.push({ name: hit.name, matchedWords: matched });
    }

    // Tier B: single rare word. Only when co-occurrence found nothing — a
    // co-occurring name is categorically stronger evidence — and under
    // stricter rules, because one word is thin: the word must be ≥5 chars
    // (measured FPs: "this", "typo"); the segment must appear in AT LEAST TWO
    // names (a concept the codebase is about clusters across names —
    // CheckoutService/CheckoutController — while a prose coincidence is a
    // singleton: measured FP "deploy to PRODUCTION" → the one name
    // matchesNonProductionDir); and the candidate name must have ≥2 segments
    // (a bare common verb matching a bare function name — "write" → `write` —
    // is prose coincidence, not the user naming a symbol).
    if (candidates.length === 0) {
      const singleWordVariants = variants.filter((v) => variantToWord.get(v)!.length >= 5);
      const counts = this.queries.getSegmentNameCounts(singleWordVariants);
      const rare = [...counts.entries()]
        .filter(([, n]) => n >= 2 && n <= CodeGraph.SEGMENT_RARITY_CEILING)
        .sort((a, b) => a[1] - b[1])
        .slice(0, 2);
      for (const [variant] of rare) {
        const word = variantToWord.get(variant)!;
        for (const name of this.queries.getNamesForSegment(variant, 12)) {
          if (splitIdentifierSegments(name).length < 2) continue;
          candidates.push({ name, matchedWords: new Set([word]) });
        }
      }
    }

    // Verify against nodes (the honesty gate) and pick a representative
    // definition per name. A name whose only nodes are file/import kind has
    // no real definition to point at — surfacing the import statement instead
    // reads as a matched symbol but isn't one (#1144) — so it's skipped, the
    // same way an orphaned vocab row is. (Import names no longer enter the
    // vocab at write time, but rows written before that exclusion persist
    // until the next full index.)
    const out: SegmentMatch[] = [];
    const seen = new Set<string>();