Create ingest/cia_reading_room.py

ingest/cia_reading_room.py

@@ -1,6 +1,55 @@
+import asyncio
+import requests
+from typing import List, Dict
 from ingest.generic_public_foia import GenericFOIAAdapter
 
 
 class CIAAdapter(GenericFOIAAdapter):
-    source_name = "CIA Reading Room"
-    base_url = "https://www.cia.gov/readingroom/"
+    """
+    LIVE adapter for the CIA FOIA Electronic Reading Room
+    https://www.cia.gov/readingroom/
+    """
+
+    source_name = "CIA FOIA Reading Room"
+    base_url = "https://www.cia.gov/readingroom/search/site"
+
+    live = True
+    extended = False
+
+    async def search(self, query: str) -> List[Dict]:
+        await self._rate_limit()
+
+        params = {
+            "search_api_fulltext": query
+        }
+
+        try:
+            resp = requests.get(
+                self.base_url,
+                params=params,
+                timeout=10,
+                headers={
+                    "User-Agent": "FOIA-Research-Bot/1.0"
+                }
+            )
+            resp.raise_for_status()
+        except Exception:
+            return []
+
+        # CIA search pages are HTML — keep it SAFE + SIMPLE
+        results = []
+
+        # Minimal heuristic parse (intentionally conservative)
+        for line in resp.text.splitlines():
+            if "/readingroom/document/" in line:
+                url = line.split('"')[1]
+                results.append({
+                    "source": self.source_name,
+                    "title": "CIA FOIA Document",
+                    "url": f"https://www.cia.gov{url}",
+                    "snippet": "Publicly released CIA FOIA document.",
+                    "live": True,
+                    "extended": False,
+                })
+
+        return results[:10]  # HARD CAP (HF-safe)