refactor: total change, no longer ripping gigabytes but going directly to datasets, must find way to automate download.
This commit is contained in:
Leonard Excoffier
139
write_to_db.py
139
write_to_db.py
@@ -1,129 +1,18 @@
|
||||
import os
|
||||
import mariadb
|
||||
import json
|
||||
from dotenv import load_dotenv
|
||||
import pandas as pd
|
||||
|
||||
# Load environment variables from .env file
|
||||
load_dotenv()
|
||||
# Read the data into a Pandas DataFrame
|
||||
file_path = 'sec_data/2024q1/tag.txt'
|
||||
df = pd.read_csv(file_path, sep='\t')
|
||||
|
||||
def connect_to_db():
|
||||
try:
|
||||
# Read the connection parameters from the environment
|
||||
conn = mariadb.connect(
|
||||
user=os.getenv("DB_USER"),
|
||||
password=os.getenv("DB_PASSWORD"),
|
||||
host=os.getenv("DB_HOST"),
|
||||
port=int(os.getenv("DB_PORT")),
|
||||
database=os.getenv("DB_NAME")
|
||||
)
|
||||
return conn
|
||||
except mariadb.Error as e:
|
||||
print(f"Error connecting to MariaDB: {e}")
|
||||
return None
|
||||
# Inspect the DataFrame
|
||||
print("First rows of the DataFrame:")
|
||||
print(df.head(10))
|
||||
|
||||
def insert_entity(cursor, cik, entity_name):
|
||||
cursor.execute(
|
||||
"INSERT IGNORE INTO entities (cik, name) VALUES (?, ?)", (cik, entity_name))
|
||||
# Get the DataFrame Information
|
||||
print("\nSummary Information:")
|
||||
print(df.info())
|
||||
|
||||
def insert_fact(cursor, taxonomy, fact_id, label, description, unit):
|
||||
cursor.execute(
|
||||
"INSERT IGNORE INTO facts (id, taxonomy, label, description, unit) VALUES (?, ?, ?, ?, ?)",
|
||||
(fact_id, taxonomy, label, description, unit)
|
||||
)
|
||||
|
||||
def insert_data(cursor, cik, fact_id, start, end, val, accn, fy, fp, form, filed, frame):
|
||||
cursor.execute(
|
||||
"""INSERT IGNORE INTO data (cik, fact_id, end, start, val, accn, fy, fp, form, filed, frame)
|
||||
VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)""",
|
||||
(cik, fact_id, end, start, val, accn, fy, fp, form, filed, frame)
|
||||
)
|
||||
|
||||
def cik_exists(cursor, cik):
|
||||
cursor.execute("SELECT 1 FROM entities WHERE cik = ?", (cik,))
|
||||
return cursor.fetchone() is not None
|
||||
|
||||
def parse_json_and_insert_data(file_path):
|
||||
with open(file_path, 'r') as file:
|
||||
data = json.load(file)
|
||||
|
||||
cik = data.get('cik')
|
||||
|
||||
# Start a new connection for each file
|
||||
conn = connect_to_db()
|
||||
if conn is None:
|
||||
return False
|
||||
|
||||
try:
|
||||
cursor = conn.cursor()
|
||||
|
||||
# Optional: Check if cik already exists in the database.
|
||||
# You can comment this block out if you do not want this check.
|
||||
if cik_exists(cursor, cik):
|
||||
print(f"CIK {cik} already exists in the database. Skipping file {file_path}.")
|
||||
return False
|
||||
|
||||
# Insert the entity
|
||||
entity_name = data.get('entityName')
|
||||
insert_entity(cursor, cik, entity_name)
|
||||
|
||||
# Iterate over facts
|
||||
for taxonomy, fact_details in data['facts'].items():
|
||||
for fact_id, fact in fact_details.items():
|
||||
# Get fact details
|
||||
label = fact.get('label')
|
||||
description = fact.get('description')
|
||||
|
||||
for unit, unit_vals in fact.get('units', {}).items():
|
||||
# Insert fact
|
||||
insert_fact(cursor, taxonomy, fact_id, label, description, unit)
|
||||
|
||||
# Insert each data point
|
||||
for entry in unit_vals:
|
||||
start = entry.get('start', None)
|
||||
end = entry['end']
|
||||
val = entry['val']
|
||||
accn = entry['accn']
|
||||
fy = entry['fy']
|
||||
fp = entry['fp']
|
||||
form = entry['form']
|
||||
filed = entry['filed']
|
||||
frame = entry.get('frame', None)
|
||||
|
||||
insert_data(cursor, cik, fact_id, start, end, val, accn, fy, fp, form, filed, frame)
|
||||
|
||||
# Commit transaction
|
||||
conn.commit()
|
||||
return True
|
||||
|
||||
except Exception as e:
|
||||
print(f"Error occurred while processing {file_path}: {e}")
|
||||
conn.rollback()
|
||||
return False
|
||||
finally:
|
||||
cursor.close()
|
||||
conn.close()
|
||||
|
||||
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()
|
||||
# Check if there are any missing values in the DataFrame
|
||||
missing_values = df.isnull().sum()
|
||||
print("\nMissing Values:")
|
||||
print(missing_values)
|
||||
|
||||
Reference in New Issue
Block a user