import json

def check_jsonl_file(file_path):
    with open(file_path, "r", encoding="utf-8") as file:
        total = 0
        custom_id_set = set()
        for line in file:
            if line.strip() == "":
                continue
            try:
                line_dict = json.loads(line)
            except json.decoder.JSONDecodeError:
                raise Exception(f"Format error in the batch inference input file. The data on line {total + 1} is not in JSON format.")
            if not line_dict.get("custom_id"):
                raise Exception(f"Format error in the batch inference input file. The custom_id on line {total + 1} does not exist.")
            if line_dict.get("custom_id") in custom_id_set:
                raise Exception(
                    f"Format error in the batch inference input file. The custom_id={line_dict.get('custom_id', '')} is duplicated."
                )
            else:
                custom_id_set.add(line_dict.get("custom_id"))
            if not isinstance(line_dict.get("body", ""), dict):
                raise Exception(
                    f"Format error in the batch inference input file. The body of custom_id={line_dict.get('custom_id', '')} is not a JSON string."
                )
            total += 1
    return total


# Replace <YOUR_JSONL_FILE> with the path of your JSONL file.
file_path = "<YOUR_JSONL_FILE>"
total_lines = check_jsonl_file(file_path)
print(f"The number of lines with valid JSON data in the file is: {total_lines}")