get-lyrics.py

import re
from glob import glob
from os import path, environ
from pprint import pprint
from mutagen.mp4 import MP4
from googleapiclient.discovery import build
import requests
from bs4 import BeautifulSoup

MUSIC_ROOT = r"C:\Users\tomokazu\Music\iTunes\iTunes Media\Music"

cse = build(serviceName="customsearch", version="v1", developerKey=environ["GOOGLE_API_KEY"])

for file in glob(path.join(MUSIC_ROOT, "**", "*.m4a"), recursive=True):
    audio = MP4(file)
    title = audio.tags.get('\xa9nam', [None])[0]
    singer = audio.tags.get('\xa9ART', [None])[0]
    if audio.tags.get('\xa9lyr', [None])[0] is not None:
        audio.tags['\xa9lyr'] = audio.tags.get('\xa9lyr', [None])[0].replace("&", "&")
        print(re.findall(r"&[A-z].*;", audio.tags.get('\xa9lyr', [None])[0]))
        print(audio.tags.get('\xa9lyr', [None]))
        print("already added lyric...skip...", title, singer)
        continue
    if "nstrumental" in title:
        print("Instrumental,skip....", title, singer)
        continue
    print(file, title, singer)
    # title = re.sub(r"\(.*?\)", "", title)
    query = f"{title} {singer} 歌詞 uta-net.com/song"
    print(query)
    search_res = cse.cse().list(q=query, cx=environ["CSE_ID"], num=3).execute()
    try:
        search_res["items"]
    except:
        print("search result not found...skip...")
        continue
    for search_res_item in search_res["items"]:
        print(search_res_item["link"])
        if "uta-net.com/song/" in search_res_item["link"]:
            uta_net_html = requests.get(search_res_item["link"]).text
            soup = BeautifulSoup(uta_net_html, "lxml")
            song = soup.select_one("#kashi_area").decode_contents()
            print(
                " ".join([text.strip() for text in soup.select_one("div.song-infoboard").findAll(string=True)]).strip())
            print(song)
            if "n" in input():
                break
            audio.tags["\xa9lyr"] = song.replace("<br/>", "\n")
            audio.save()
            break