feat: now loops over all the facts jsons

write_to_db.py

@@ -1,14 +1,14 @@
+import os
 import mariadb
 import json
 from dotenv import load_dotenv
-import os
 
 # Load environment variables from .env file
 load_dotenv()
 
 def connect_to_db():
     try:
-        # Read the needed variables from the environment
+        # Read the connection parameters from the environment
         conn = mariadb.connect(
             user=os.getenv("DB_USER"),
             password=os.getenv("DB_PASSWORD"),
@@ -38,13 +38,21 @@ def insert_data(cursor, cik, fact_id, start, end, val, accn, fy, fp, form, filed
         (cik, fact_id, start, end, val, accn, fy, fp, form, filed, frame)
     )
 
-def parse_json_and_insert_data(file_path, cursor):
+def parse_json_and_insert_data(file_path):
     with open(file_path, 'r') as file:
         data = json.load(file)
 
+        # Get the connection and cursor
+        conn = connect_to_db()
+        if conn is None:
+            return False
+
+        try:
+            cursor = conn.cursor()
+
             # Insert the entity
-        cik = data['cik']
-        entity_name = data['entityName']
+            cik = data.get('cik')
+            entity_name = data.get('entityName')
             insert_entity(cursor, cik, entity_name)
 
             # Iterate over facts
@@ -72,29 +80,39 @@ def parse_json_and_insert_data(file_path, cursor):
 
                             insert_data(cursor, cik, fact_id, start, end, val, accn, fy, fp, form, filed, frame)
 
-def main():
-    # Connect to the database
-    conn = connect_to_db()
-    if conn is None:
-        return
-
-    try:
-        cursor = conn.cursor()
-
-        # Load JSON and insert data
-        json_file_path = 'CIK0000320193.json'
-        parse_json_and_insert_data(json_file_path, cursor)
-
-        # Commit the transaction
+            # Commit the transaction for the whole file
             conn.commit()
+            return True
 
         except Exception as e:
-        print(f"Error occurred: {e}")
+            print(f"Error occurred while processing {file_path}: {e}")
             conn.rollback()
+            return False
         finally:
             cursor.close()
             conn.close()
-        print("Connection closed")
+
+def process_all_files_in_directory(directory_path):
+    files = [f for f in os.listdir(directory_path) if f.endswith('.json')]
+    total_files = len(files)
+    processed_files = 0
+
+    for idx, file_name in enumerate(files, start=1):
+        file_path = os.path.join(directory_path, file_name)
+        print(f"Processing file {idx} of {total_files}: {file_name}")
+
+        if parse_json_and_insert_data(file_path):
+            processed_files += 1
+            print(f"Successfully processed {file_name}")
+        else:
+            print(f"Failed to process {file_name}")
+
+    print(f"Finished processing {processed_files} out of {total_files} files.")
+
+def main():
+    # Process all JSON files in the directory
+    directory_path = './sec_data/companyfacts/'
+    process_all_files_in_directory(directory_path)
 
 if __name__ == "__main__":
     main()