import json
import re


def clean_gender_from_tags(tags):
    """
    Clean gender/sex related terms from tag strings.

    Args:
        tags (str): String containing tags

    Returns:
        str: Cleaned tags with gender terms removed
    """
    if not tags:
        return tags

    # Define patterns for gender/sex related terms to remove
    gender_patterns = [
        r"\bmale vocals?\b",
        r"\bfemale vocals?\b",
        r"\bmale voice\b",
        r"\bfemale voice\b",
        r"\bmale singer\b",
        r"\bfemale singer\b",
        r"\bwoman\b",
        r"\bwomen\b",
        r"\bman\b",
        r"\bmen\b",
        r"\bgirl\b",
        r"\bgirls\b",
        r"\bboy\b",
        r"\bboys\b",
        r"\bsouthern male\b",
    ]

    # Replace each gender pattern with an empty string
    cleaned_tags = tags
    for pattern in gender_patterns:
        cleaned_tags = re.sub(pattern, "", cleaned_tags, flags=re.IGNORECASE)

    # Clean up by removing extra commas and spaces
    cleaned_tags = re.sub(r",\s*,", ",", cleaned_tags)  # Remove double commas
    cleaned_tags = re.sub(r",\s*$", "", cleaned_tags)  # Remove trailing comma
    cleaned_tags = re.sub(r"^\s*,", "", cleaned_tags)  # Remove leading comma
    cleaned_tags = re.sub(
        r"\s+", " ", cleaned_tags
    )  # Replace multiple spaces with a single space
    cleaned_tags = cleaned_tags.strip()  # Trim white space

    return cleaned_tags


def process_json_file(input_filename, output_filename):
    """
    Process a JSON file to remove gender information from tags.

    Args:
        input_filename (str): Path to input JSON file
        output_filename (str): Path to save the cleaned JSON file
    """
    # Read the input JSON file
    with open(input_filename, "r", encoding="utf-8") as file:
        json_data = json.load(file)

    # Create a clean copy of the data
    cleaned_data = {}

    # Process each genre and its songs
    for genre, songs in json_data.items():
        cleaned_data[genre] = []
        for song in songs:
            cleaned_song = song.copy()
            cleaned_song["tags"] = clean_gender_from_tags(song.get("tags", ""))
            cleaned_data[genre].append(cleaned_song)

    # Write the cleaned data to the output file
    with open(output_filename, "w", encoding="utf-8") as file:
        json.dump(cleaned_data, file, indent=4, ensure_ascii=False)

    print(f"Processed JSON saved to {output_filename}")


if __name__ == "__main__":
    # Usage example
    input_file = (
        "/home/sara/glockenspiel/suno_utils/task_eval/labelbox/labelbox_sources.json"
    )
    output_file = "labelbox_sources_cleaned.json"
    process_json_file(input_file, output_file)