data/data_converter.py

import json

# Input and output file paths
input_filepath = 'commit_data_hpc.jsonl'
output_filepath = 'commit_data_hpc_modified.jsonl'

with open(input_filepath, 'r') as infile, open(output_filepath, 'w') as outfile:
    idx = 1  # Starting index
    for line in infile:
        # Parse the line as JSON
        data = json.loads(line.strip())
        
        # Rename 'commit_message' key to 'func'
        data['func'] = data.pop('commit_message')
        
        # Add 'idx' key
        data['idx'] = idx
        
        # Serialize back to JSON and write to the output file
        outfile.write(json.dumps(data) + '\n')
        
        # Increment the index
        idx += 1

print("File modification complete!")