feat(docs): 관련 문서(유사도 KNN) 엔드포인트+패널 + 법령/지침 splitter

scripts/paper_openalex_test.py

@@ -0,0 +1,30 @@
+#!/usr/bin/env python3
+"""OpenAlex 보강 타당성 테스트 — 소수 논문 제목으로 매칭/메타 확인 (외부 API)."""
+import asyncio, os, re
+
+async def main():
+    import asyncpg, httpx
+    conn = await asyncpg.connect(os.environ['DATABASE_URL'].replace('+asyncpg', ''))
+    rows = await conn.fetch("SELECT id, title FROM documents WHERE material_type='paper' "
+                            "AND doc_kind='standard' AND deleted_at IS NULL AND title IS NOT NULL "
+                            "AND length(title) > 15 ORDER BY id LIMIT 6")
+    async with httpx.AsyncClient(timeout=20) as client:
+        for r in rows:
+            title = re.sub(r'\s+', ' ', r['title']).strip()
+            try:
+                resp = await client.get("https://api.openalex.org/works",
+                    params={"search": title[:200], "per_page": 1, "mailto": "hyun49196@gmail.com"})
+                js = resp.json()
+                res = (js.get("results") or [])
+                if not res:
+                    print(f"[{r['id']}] NO MATCH | {title[:50]}"); continue
+                w = res[0]
+                oid = (w.get("id") or "").split("/")[-1]
+                print(f"[{r['id']}] {title[:46]}")
+                print(f"   → OA {oid} | {(w.get('title') or '')[:46]} | {w.get('publication_year')} | "
+                      f"cited_by={w.get('cited_by_count')} | refs={len(w.get('referenced_works') or [])} | doi={w.get('doi')}")
+            except Exception as e:
+                print(f"[{r['id']}] ERROR {type(e).__name__}: {e}")
+    await conn.close()
+
+asyncio.run(main())