Upload app.py

app.py

@@ -935,6 +935,129 @@ async def chat(req: ChatRequest, request: Request):
     )
 
 
+# ── Journal meta lookup — ISSN + publisher auto-fill ─────────────────────
+@app.get("/api/lookup-journal-meta")
+async def lookup_journal_meta(request: Request, name: str):
+    """
+    Two-layer journal lookup:
+    Layer 1: SCImago (exact) → OpenAlex (title search) → returns ISSN + publisher
+    Layer 2: If not found, LLM corrects spelling only — never returns ISSN
+    """
+    ip = get_client_ip(request)
+    if not check_rate_limit(ip, "global"):
+        raise HTTPException(429, "Too many requests.")
+
+    name = name.strip()
+    if not name or len(name) < 3:
+        return {"found": False}
+
+    # ── Layer 1a: SCImago fuzzy title match ───────────────────────────────
+    from scimago_index import load_subject_index
+    rows = load_subject_index()
+    name_lower = name.lower()
+
+    # Exact match first
+    exact = next((r for r in rows if r["title"].lower() == name_lower), None)
+    if exact:
+        return {
+            "found":      True,
+            "issn":       exact.get("issn", ""),
+            "publisher":  exact.get("publisher", ""),
+            "title":      exact["title"],
+            "source":     "SCImago",
+            "confidence": "high",
+        }
+
+    # Partial match — title contains all words
+    words = [w for w in name_lower.split() if len(w) > 2]
+    partial = next((r for r in rows if all(w in r["title"].lower() for w in words)), None)
+    if partial:
+        return {
+            "found":      True,
+            "issn":       partial.get("issn", ""),
+            "publisher":  partial.get("publisher", ""),
+            "title":      partial["title"],
+            "source":     "SCImago",
+            "confidence": "medium",
+        }
+
+    # ── Layer 1b: OpenAlex title search ───────────────────────────────────
+    try:
+        oa_client = openalex._get_client()
+        r = await oa_client.get(
+            "https://api.openalex.org/sources",
+            params={"search": name, "per_page": 1, "filter": "type:journal"},
+            timeout=5.0,
+        )
+        if r.is_success:
+            results = r.json().get("results", [])
+            if results:
+                s = results[0]
+                title      = s.get("display_name", "")
+                issns      = s.get("issn", [])
+                issn_l     = s.get("issn_l", "")
+                publisher  = s.get("host_organization_name", "")
+                best_issn  = issn_l or (issns[0] if issns else "")
+                ni         = norm_issn(best_issn)
+                if title:
+                    return {
+                        "found":      True,
+                        "issn":       ni or best_issn,
+                        "publisher":  publisher,
+                        "title":      title,
+                        "source":     "OpenAlex",
+                        "confidence": "high" if title.lower() == name_lower else "medium",
+                    }
+    except Exception as e:
+        logger.warning(f"[LookupJournal] OpenAlex error: {e}")
+
+    # ── Layer 2: LLM spelling correction only — NO ISSN ──────────────────
+    try:
+        client = get_client()
+        resp = await client.chat.completions.create(
+            model=MODEL,
+            response_format={"type": "json_object"},
+            messages=[
+                {
+                    "role": "system",
+                    "content": (
+                        "You are an academic journal name expert. "
+                        "The user has typed a journal name that may be misspelled or abbreviated. "
+                        "Return ONLY valid JSON: "
+                        "{corrected_name: string, confidence: high|medium|low, known: bool}. "
+                        "corrected_name: the most likely correct full journal name. "
+                        "If you do not recognise it, return the input unchanged. "
+                        "confidence: high if certain, medium if likely, low if guessing. "
+                        "known: true only if you are confident this journal exists. "
+                        "Do NOT return ISSN or publisher. Return JSON only."
+                    ),
+                },
+                {"role": "user", "content": f"Journal name input: {name}"},
+            ],
+            max_tokens=100,
+            temperature=0.1,
+        )
+        raw = resp.choices[0].message.content or "{}"
+        parsed = json.loads(raw)
+        corrected  = parsed.get("corrected_name", name).strip()
+        confidence = parsed.get("confidence", "low")
+        known      = parsed.get("known", False)
+
+        # Only suggest if different from input and model is confident it exists
+        if known and corrected.lower() != name_lower and confidence in ("high", "medium"):
+            return {
+                "found":          False,
+                "suggestion":     corrected,
+                "suggestion_confidence": confidence,
+                "source":         "LLM",
+                "message":        f"Did you mean: {corrected}?",
+            }
+    except Exception as e:
+        logger.warning(f"[LookupJournal] LLM error: {e}")
+
+    return {"found": False}
+
+
 # ── Static frontend ────────────────────────────────────────────────────────
 @app.get("/")
 async def root():