Update src/api/agent.py

src/api/agent.py

@@ -1400,57 +1400,64 @@ async def _semantic_scholar_search(topic: str, limit: int = 8) -> List[dict]:
 async def _openalex_search(topic: str, limit: int = 8) -> List[dict]:
     """
     Search OpenAlex using title.search + abstract.search filters for precision.
-    This targets only the title and abstract fields — far more relevant than the
-    general `search` parameter which also hits author names, journals, institutions.
+    Targets only title and abstract fields — more relevant than general search.
     Free, no API key, typically < 2s.
     """
     try:
-        from urllib.parse import urlencode
+        from urllib.parse import quote
 
-        # Build a clean search term — strip boolean operators, keep the core phrase
+        # Strip boolean operators — OpenAlex filter search uses its own relevance engine
         clean = re.sub(r'\b(AND|OR|NOT)\b|[()"\']', ' ', topic, flags=re.IGNORECASE)
         clean = re.sub(r'\s+', ' ', clean).strip()[:200]
+        encoded = quote(clean, safe='')
 
-        # Use title.search OR abstract.search — papers that match in either field
-        params = {
-            "filter": (
-                f"title.search:{clean}|abstract.search:{clean},"
-                f"has_abstract:true,type:article,is_paratext:false"
-            ),
-            "sort": "relevance_score:desc",
-            "per_page": str(limit),
-            "select": ",".join([
-                "id", "title", "authorships", "publication_year",
-                "primary_location", "abstract_inverted_index",
-                "doi", "cited_by_count", "open_access",
-            ]),
-        }
-        url = "https://api.openalex.org/works?" + urlencode(params)
+        # Build URL manually — urlencode encodes | and : which breaks OpenAlex filter syntax
+        filter_str = (
+            f"title.search:{encoded}|abstract.search:{encoded},"
+            f"has_abstract:true,type:article,is_paratext:false"
+        )
+        select = "id,title,authorships,publication_year,primary_location,abstract_inverted_index,doi,cited_by_count,open_access"
+        url = (
+            f"https://api.openalex.org/works"
+            f"?filter={filter_str}"
+            f"&sort=relevance_score:desc"
+            f"&per_page={limit}"
+            f"&select={select}"
+        )
         headers = {"User-Agent": "LibBee/3.2 (mailto:libse@ku.ac.ae)"}
 
         async with httpx.AsyncClient(timeout=10) as client:
             r = await client.get(url, headers=headers)
             if r.status_code != 200:
-                logger.warning(f"OpenAlex returned {r.status_code} for topic '{clean[:40]}'")
-                return []
+                logger.warning(f"OpenAlex title+abstract search {r.status_code} — retrying with general search")
+                # Fallback: general search
+                url_fb = (
+                    f"https://api.openalex.org/works"
+                    f"?search={encoded}"
+                    f"&filter=has_abstract:true,type:article,is_paratext:false"
+                    f"&sort=relevance_score:desc"
+                    f"&per_page={limit}"
+                    f"&select={select}"
+                )
+                r = await client.get(url_fb, headers=headers)
+                if r.status_code != 200:
+                    logger.warning(f"OpenAlex fallback also failed: {r.status_code}")
+                    return []
 
-            data = r.json()
-            works = data.get("results", [])
+            works = r.json().get("results", [])
 
-            # If title+abstract search returns nothing, fall back to general search
+            # If title+abstract returns nothing, retry with general search
             if not works:
-                logger.info(f"OpenAlex title+abstract search returned 0 — retrying with general search")
-                params_fallback = {
-                    "search": clean,
-                    "filter": "has_abstract:true,type:article,is_paratext:false",
-                    "sort": "relevance_score:desc",
-                    "per_page": str(limit),
-                    "select": params["select"],
-                }
-                r2 = await client.get(
-                    "https://api.openalex.org/works?" + urlencode(params_fallback),
-                    headers=headers,
+                logger.info("OpenAlex title+abstract returned 0 — retrying with general search")
+                url_fb = (
+                    f"https://api.openalex.org/works"
+                    f"?search={encoded}"
+                    f"&filter=has_abstract:true,type:article,is_paratext:false"
+                    f"&sort=relevance_score:desc"
+                    f"&per_page={limit}"
+                    f"&select={select}"
                 )
+                r2 = await client.get(url_fb, headers=headers)
                 if r2.status_code == 200:
                     works = r2.json().get("results", [])
 
@@ -1472,7 +1479,6 @@ async def _openalex_search(topic: str, limit: int = 8) -> List[dict]:
                 if not abstract:
                     continue
 
-                # Authors
                 authorships = work.get("authorships") or []
                 author_names = [
                     a.get("author", {}).get("display_name", "")
@@ -1493,9 +1499,9 @@ async def _openalex_search(topic: str, limit: int = 8) -> List[dict]:
                     or work.get("id", "https://openalex.org")
                 )
                 venue = (
-                    (work.get("primary_location") or {})
-                    .get("source", {}) or {}
-                ).get("display_name") or ""
+                    ((work.get("primary_location") or {}).get("source") or {})
+                    .get("display_name") or ""
+                )
 
                 results.append({
                     "title": title,