import os
import pandas as pd
def split_csv(input_file, output_dir):
# Create the output directory if it doesn't exist
os.makedirs(output_dir, exist_ok=True)
# Read the CSV file into chunks
chunksize = 10000
reader = pd.read_csv(input_file, chunksize=chunksize)
# Save each chunk to a separate file
for i, chunk in enumerate(reader):
output_file = os.path.join(output_dir, f"{os.path.splitext(os.path.basename(input_file))[0]}-small-{i+1}.csv")
chunk.to_csv(output_file, index=False)
if __name__ == "__main__":
import sys
if len(sys.argv) != 2:
print("Usage: python splitcsv.py dir/filename.csv")
sys.exit(1)
input_file = sys.argv[1]
output_dir = os.path.dirname(input_file)
split_csv(input_file, output_dir)