FIX: first_seen re-stamp bug — persist true source publish date, idempotent upsert, backfill DB

pipeline.py

@@ -25,6 +25,7 @@ from datetime import datetime, timezone
 sys.path.insert(0, os.path.dirname(__file__))
 
 from adapters import SourceAdapter
+from adapters._store import upsert_entries
 
 # Adapter registry — add new adapters here (one line each)
 ADAPTERS = {
@@ -51,27 +52,14 @@ def init_db(db_path: str, schema_path: str) -> sqlite3.Connection:
 
 
 def store_entries(conn: sqlite3.Connection, entries: list[dict]) -> int:
-    """Store entries using INSERT OR REPLACE (dedup by source+source_id)."""
-    cur = conn.cursor()
-    stored = 0
-    for entry in entries:
-        try:
-            cur.execute("""
-                INSERT OR REPLACE INTO entries
-                (source, source_id, url, title, extracted_text, summary,
-                 category_tags, signal_score, raw_metadata, first_seen, last_updated)
-                VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
-            """, (
-                entry["source"], entry["source_id"], entry["url"], entry["title"],
-                entry["extracted_text"] or "", entry["summary"],  # None → NULL in DB
-                entry["category_tags"], entry["signal_score"],
-                entry["raw_metadata"], entry["first_seen"], entry["last_updated"],
-            ))
-            stored += 1
-        except Exception as e:
-            print(f"  ⚠ DB error on {entry.get('source', '?')}/{entry.get('source_id', '?')}: {e}")
-    conn.commit()
-    return stored
+    """Store entries idempotently (dedup by url).
+
+    FIX (2026-07-13, Tony): was INSERT OR REPLACE which OVERWROTE first_seen
+    with the harvest time on every re-harvest, turning stale stories into
+    "today". Now uses an UPSERT that preserves the ORIGINAL first_seen and only
+    bumps last_updated. Entries must already carry first_seen = true publish date.
+    """
+    return upsert_entries(conn, entries)
 
 
 def verify_entries(conn: sqlite3.Connection, source: str, sample_size: int = 3):