Function prepare_input_files

scripts/data/processing/text_to_jsonl.py:194–223

Prepare list of input files from various input types. Args: input_path: Single file path, directory path, or list of file paths Returns: List of transcript file paths Raises: ValueError: If input is invalid or no files are found

(input_path: Union[str, List[str]])

def prepare_input_files(input_path: Union[str, List[str]]) -> FileList:
    """
    Prepare list of input files from various input types.

    Args:
        input_path: Single file path, directory path, or list of file paths

    Returns:
        List of transcript file paths

    Raises:
        ValueError: If input is invalid or no files are found
    """
    if isinstance(input_path, str):
        if os.path.isdir(input_path):
            # Directory: discover transcript files
            return discover_transcript_files(input_path)
        elif os.path.isfile(input_path):
            # Single file
            return [input_path]
        else:
            raise ValueError(f"Input path does not exist: {input_path}")
    elif isinstance(input_path, list):
        # List of files: validate each exists
        for file_path in input_path:
            if not os.path.isfile(file_path):
                raise ValueError(f"File does not exist: {file_path}")
        return input_path
    else:
        raise ValueError(f"Invalid input_path type: {type(input_path)}")


def write_jsonl_output(