Server-scripts/media handling/compare_playlists.py

#!/usr/bin/env python3

import re
import sys
from pathlib import Path
from difflib import SequenceMatcher

def normalize(name):
    name = Path(name).stem.lower()

    # remove brackets
    name = re.sub(r"\(.*?\)|\[.*?\]", "", name)

    # replace punctuation with spaces
    name = re.sub(r"[^a-z0-9]", " ", name)

    # collapse whitespace
    name = " ".join(name.split())

    return name


def similarity(a, b):
    words_a = set(normalize(a).split())
    words_b = set(normalize(b).split())

    if not words_a or not words_b:
        return 0.0

    # Jaccard similarity
    intersection = words_a & words_b
    union = words_a | words_b

    return len(intersection) / len(union)


def load_playlist(path):
    entries = []

    with open(path, encoding="utf-8", errors="ignore") as f:
        for line in f:
            line = line.strip()

            if not line or line.startswith("#"):
                continue

            entries.append(Path(line).name)

    return entries


def compare_playlists(orig_file, new_file, threshold=0.85):
    orig_files = load_playlist(orig_file)
    new_files = load_playlist(new_file)

    print(f"\n=== {orig_file.parent} ===")
    print(f"Comparing:")
    print(f"  {orig_file.name}")
    print(f"  {new_file.name}")

    missing = []

    for orig in orig_files:
        best_match = None
        best_score = 0

        for candidate in new_files:
            score = similarity(orig, candidate)

            if score > best_score:
                best_score = score
                best_match = candidate

        if best_score >= threshold:
            pass
           # print(
           #     f"MATCH   {best_score:.1%}  "
           #     f"{orig} -> {best_match}"
           # )
        else:
            missing.append((orig, best_score, best_match))

    if missing:
        print("\nMISSING:")
        for orig, score, best_match in sorted(
            missing,
            key=lambda x: x[1],
            reverse=True,
        ):
            print(
                f"  {orig}"
                f" (best={score:.1%}"
                f"{', candidate=' + best_match if best_match else ''})"
            )

    return len(missing)


def compare_directory(root, threshold=0.85):
    total_missing = 0
    playlist_pairs = 0

    for orig_file in root.rglob("*_orig.m3u"):
        new_file = orig_file.with_name(
            orig_file.name.replace("_orig.m3u", ".m3u")
        )

        if not new_file.exists():
            print(
                f"\nWARNING: Missing playlist for "
                f"{orig_file}"
            )
            continue

        playlist_pairs += 1
        total_missing += compare_playlists(
            orig_file,
            new_file,
            threshold,
        )

    print("\n" + "=" * 80)
    print(f"Playlists compared : {playlist_pairs}")
    print(f"Total missing      : {total_missing}")


def main():
    args = sys.argv[1:]

    threshold = 0.85

    if len(args) == 3 and args[0] == "--threshold":
        threshold = float(args[1])
        args = [args[2]]

    elif len(args) == 4 and args[0] == "--threshold":
        threshold = float(args[1])
        args = [args[2], args[3]]

    if len(args) == 1:
        path = Path(args[0])

        if not path.is_dir():
            print("Single argument must be a directory")
            sys.exit(1)

        compare_directory(path, threshold)

    elif len(args) == 2:
        compare_playlists(
            Path(args[0]),
            Path(args[1]),
            threshold,
        )

    else:
        print(
            f"Usage:\n"
            f"  {Path(sys.argv[0]).name} "
            f"<playlist_orig.m3u> <playlist.m3u>\n\n"
            f"  {Path(sys.argv[0]).name} "
            f"<directory>\n\n"
            f"Optional:\n"
            f"  --threshold 0.90"
        )
        sys.exit(1)


if __name__ == "__main__":
    main()