initial

scripts/check_and_fix_column.py

@@ -0,0 +1,54 @@
+#!/usr/bin/env python3
+import pymysql
+from dotenv import load_dotenv
+import os
+
+load_dotenv()
+
+config = {
+    'host': os.getenv('DB_HOST'),
+    'user': os.getenv('DB_USER'),
+    'password': os.getenv('DB_PASSWORD'),
+    'database': os.getenv('DB_NAME'),
+    'port': int(os.getenv('DB_PORT', 3306)),
+}
+
+conn = pymysql.connect(**config)
+cursor = conn.cursor()
+
+# Check column type
+cursor.execute("SHOW COLUMNS FROM arc_jsons WHERE Field = 'json'")
+result = cursor.fetchone()
+print(f"Current column type: {result[1]}")
+
+# Check if we need to alter it
+if 'text' in result[1].lower() and 'medium' not in result[1].lower() and 'long' not in result[1].lower():
+    print("\n⚠  Column is TEXT (max 65,535 bytes)")
+    print("   One file (4a21e3da.json) is ~69KB and failed to insert")
+    print("\nRecommended fix: ALTER TABLE arc_jsons MODIFY json MEDIUMTEXT;")
+    print("   MEDIUMTEXT supports up to 16MB")
+
+    response = input("\nApply fix now? (yes/no): ").strip().lower()
+    if response in ['yes', 'y']:
+        print("\nAltering column to MEDIUMTEXT...")
+        cursor.execute("ALTER TABLE arc_jsons MODIFY json MEDIUMTEXT")
+        conn.commit()
+        print("✓ Column altered successfully!")
+
+        # Now insert the failed record
+        print("\nRe-inserting failed record (4a21e3da)...")
+        with open('arc_data/evaluation/4a21e3da.json', 'r') as f:
+            json_content = f.read().strip()
+
+        cursor.execute("INSERT INTO arc_jsons (id, json) VALUES (%s, %s)", ('4a21e3da', json_content))
+        conn.commit()
+        print("✓ Record inserted successfully!")
+
+        # Final count
+        cursor.execute("SELECT COUNT(*) FROM arc_jsons")
+        count = cursor.fetchone()[0]
+        print(f"\n✓ Total records in database: {count}")
+else:
+    print(f"✓ Column type is sufficient: {result[1]}")
+
+conn.close()