Server-scripts/audiobook_metadata/metadata_edit_database.py

#!/usr/bin/python3
import json
import re
import sqlite3
import argparse
from pathlib import Path

def load_genre(conn, unique=False):
    """
    Return all genres from the books table.

    Args:
        conn: sqlite3 connection
        unique: if True, return unique genres only

    Returns:
        list[str]
    """
    genres = []

    for (genres_json,) in conn.execute("SELECT genres FROM books"):
        if not genres_json:
            continue

        try:
            genres.extend(json.loads(genres_json))
        except Exception:
            pass

    if unique:
        return sorted(set(genres))

    return genres

def main():
    parser = argparse.ArgumentParser()
    parser.add_argument(
        "-p", "--print-only",
        action="store_true",
        help="Print data only",
    )
    parser.add_argument(
        "database",
        type=Path,
        help="SQL Database Path",
    )

    args = parser.parse_args()

    database = args.database.resolve()

    if not database.is_file():
        parser.error(f"Database is not a file: {database}")

    db = sqlite3.connect(database)
    cur = db.cursor()

    list_genres = load_genre(cur,True)

    #for genre in list_genres:
    #    print(genre)
    rows = cur.execute(
        "SELECT id, genres FROM books"
    ).fetchall()

    for book_id, genres_json in rows:
        if not genres_json:
            continue
        genres = json.loads(genres_json)
        new_genres = []
        print("=================================")
        print(f"Before sed: {genres}")
        for genre in genres:
            genre = re.sub(r'(?i)^audio\s*books?$', '', genre)
            genre = re.sub(r'(?i)^audiobooks?$', '', genre)
            genre = re.sub(r'(?i)\bhörb(uch|ücher)\b', '', genre)
            genre = re.sub(r'(?i)^(hörspiele?|hoerspiele?)$', '', genre)
            genre = re.sub(r'(?i)^spoken\s*word$', '', genre)
            genre = re.sub(r'(?i)^speech$', '', genre)

            genre = re.sub(r'(?i)^(sci[- ]?fi|sf)$', 'Science Fiction', genre)
            genre = re.sub(r'(?i)^nonfiction$', 'NonFiction', genre)

            genre = re.sub(r'^-\s*|\s*-$', '', genre)
            genre = genre.strip()

            new_genres.append(genre)

        new_genres = sorted(set(new_genres))
        
        print("Write to Database")

        print(f"After sed: {new_genres}")
        print("=================================")

        list_difference = False

        for old,new in zip(genres,new_genres):
            if old != new:
                list_difference = True
                break

        if not list_difference:
            print("No difference: Dont Commit to Database")
            continue

        if args.print_only:
            print("Print only: Dont Commit to database")
            continue

        print("Write to Database")
        cur.execute(
            "UPDATE books SET genres=? WHERE id=?",
            (json.dumps(new_genres), book_id),
        )
        print("Database Updated")

    db.commit()

if __name__ == "__main__":
    main()