Script to add lyrics tag from text file in directories

Lyrics Utility

# coding: utf-8

# --------------------------------------------------------------------------------
# 音楽ファイルのタグ情報を元にテキスト形式の歌詞ファイルを検索し歌詞を書き込む
# --------------------------------------------------------------------------------
# targets               対象の音楽ファイルorフォルダ（可変長）
# -d --dir              歌詞フォルダ（def:カレント以下）
# -l --matchlevel       文字列一致判定の類似度のレベル(1..3=def)
# -a --ischeckartist   アーティストとフォルダ名の一致判定を行うか（def:False）
# -n --isnormalize      歌手表示を規格化するか（def:False）
# -w --isoverwrite      上書きするか（def:False）
# -f --format           歌手表示のフォーマット
# -t --istest           実際の書き込みは行わない（def:False）
# --------------------------------------------------------------------------------

import os, sys, codecs
import glob
import difflib
import argparse
import traceback
import mutagen
from mutagen.id3 import USLT
from mutagen.mp3 import MP3
from mutagen.oggvorbis import OggVorbis
from mutagen.easyid3 import EasyID3

MP3_EXTENTION = ".mp3"
OGG_EXTENTION = ".ogg"
TXT_EXTENTION = ".txt"
DEFAULT_FORMAT = ""
IsTestMode = False
IsVerbose = False

# ----------------------------------------
# コマンドラインの解析
# ----------------------------------------
def parseArgument():

  parser = argparse.ArgumentParser(description='')
  parser.add_argument('--version', action='version', version='%(prog)s 1.0')
  parser.add_argument('targets', nargs='*')
  parser.add_argument('-d', '--dir', default='./', help='')
  parser.add_argument('-l', '--matchlevel', type=int, default=3, choices=xrange(1, 4), help='')
  parser.add_argument('-a', '--ischeckartist', default=False, action='store_true', help='')
  parser.add_argument('-n', '--isnormalize', default=False, action='store_true', help='')
  parser.add_argument('-w', '--isoverwrite', default=False, action='store_true', help='')
  parser.add_argument('-f', '--format', type=str, default=DEFAULT_FORMAT, help='')
  parser.add_argument('-t', '--istest', default=False, action='store_true', help='')
  parser.add_argument('--verbose', default=False, action='store_true', help='')
  args = parser.parse_args()

  return args

# ----------------------------------------
# 音楽ファイルの読み込み
# ----------------------------------------
def loadAudio (fname):
  if not os.path.exists(fname):
    return None

  ext = os.path.splitext(fname)[1]
  if ext == MP3_EXTENTION:
    return MP3(fname)
  elif ext == OGG_EXTENTION:
    return OggVorbis(fname)

  return None

# ----------------------------------------
# 一致強度の取得
# ----------------------------------------
def getMatchRatio (level):
  ratio = 1.0
  if level >= 3:
    ratio = 1.0
  elif level >= 2:
    ratio = 0.8
  elif level >= 1:
    ratio = 0.6

  return ratio

# ----------------------------------------
# 曲名とファイル名との判定
# ----------------------------------------
def checkLyricsTitle (level, fname, title):
  name = os.path.splitext(fname)[0]
  match = difflib.SequenceMatcher(None, name, title)
  ratio = match.ratio()
  if ratio > 0.4:
    printLog(u"  Match ratio: "+name+u" - "+title+u" > "+str(ratio))
  return (ratio >= getMatchRatio(level))

# ----------------------------------------
# アーティスト名とディレクトリ名との判定
# ----------------------------------------
def checkLyricsArtist (level, dname, artist):
  name = os.path.splitext(dname)[0]
  match = difflib.SequenceMatcher(None, name, artist)
  ratio = match.ratio()
  if ratio > 0.4:
    printLog(u"  Match ratio: "+name+u" - "+artist+u" > "+str(ratio))
  return (ratio >= getMatchRatio(level))

# ----------------------------------------
# 歌詞の取得
# ----------------------------------------
def getLyrics (rpath, title, artist, album, isCheckArtist, stringMatchLevel):
  # 走査
  for dpath, dnames, fnames in os.walk(rpath):
    for fname in fnames:
      ext = os.path.splitext(fname)[1]
      if ext == TXT_EXTENTION:
        # 曲名の判定
        if checkLyricsTitle(stringMatchLevel, fname.decode("mbcs"), title):
          # アーティスト名の判定
          if isCheckArtist:
            absPath = os.path.abspath(os.path.join(dpath, fname))
            dname = os.path.basename(os.path.dirname(absPath))
            if not checkLyricsArtist(stringMatchLevel, dname.decode("mbcs"), artist):
              continue
          # 取得
          f = codecs.open(os.path.join(dpath, fname), "r", "shift_jis")
          buf = f.read()
          f.close()
          printLog(u" Get lyrics: "+fname.decode("mbcs"))
          return buf

  return None

# ----------------------------------------
# 歌詞を正規化する
# ----------------------------------------
def getNormalizedLyrics (lyrics, format):
  # TODO
  return lyrics

# ----------------------------------------
# 歌詞をタグに書き込む
# ----------------------------------------
def addLyrics (isOverWrite, audio, lyrics):
  if 'audio/mp3' in audio.mime:
    if not isOverWrite:
      if len(audio.tags.getall('USLT')) > 0:
        printLog(u" Already exists..")
        return False
    audio.tags.delall('USLT')
    audio.tags.add(USLT(encoding=3, text=lyrics))
    if not IsTestMode:
      audio.save()
    printLog(u" Saved..")
    return True

  elif 'audio/vorbis' in audio.mime:
    if not isOverWrite:
      if "UNSYNCEDLYRICS" in audio.tags:
        printLog(u" Already exists..")
        return False
    audio.tags["UNSYNCEDLYRICS"] = lyrics
    if not IsTestMode:
      audio.save()
    printLog(u" Saved..")
    return True

# ----------------------------------------
# メインプロセス
# ----------------------------------------
def mainProcess (fname, dpath, isOverWrite, isNormalize, isCheckArtist, stringMatchLevel, format):
  printLog("Start: "+fname)

  # 読み込み
  audio = loadAudio(fname)
  if not audio == None:
    # 必要な情報の取得
    title = ""
    artist = ""
    album = ""
    if 'audio/mp3' in audio.mime:
      if "TIT2" in audio:
        title = audio["TIT2"].text[0]
      if "TPE1" in audio:
        artist = audio["TPE1"].text[0]
      if "TALB" in audio:
        album = audio["TALB"].text[0]
    if 'audio/vorbis' in audio.mime:
      if "title" in audio:
        title = audio["title"][0]
      if "artist" in audio:
        artist = audio["artist"][0]
      if "album" in audio:
        album = audio["album"][0]

    printLog(u" Load audio: "+fname.decode("mbcs")+u" - "+title+u", "+artist+u", "+album)

    # 歌詞ファイルの取得
    lyrics = getLyrics(dpath, title, artist, album, isCheckArtist, stringMatchLevel)

    if not lyrics == None:
      # 歌詞のフォーマット
      if isNormalize:
        lyrics = getNormalizedLyrics(lyrics, format)
      # 保存
      isSaved = addLyrics(isOverWrite, audio, lyrics)
      if isSaved:
        print u"Done: saved lyrics.. "+fname.decode("mbcs")
      else:
        print u"Warn: already exists.. "+fname.decode("mbcs")
    else:
      print u"Warn: lyrics not found.. " + fname.decode("mbcs")

# ----------------------------------------
# メイン
# ----------------------------------------
def main():
  args = parseArgument()

  targets = args.targets
  dpath = args.dir
  stringMatchLevel = args.matchlevel
  isCheckArtist = args.ischeckartist
  isOverWrite = args.isoverwrite
  isNormalize = args.isnormalize
  format = args.format

  global IsTestMode
  IsTestMode = args.istest
  global IsVerbose
  IsVerbose = args.verbose

  printLog(args)

  # 対象の走査
  for target in targets:
    # フォルダ名の時
    fnames = glob.glob(target)
    if len(fnames) > 0:
      for fname in fnames:
        mainProcess(fname, dpath, isOverWrite, isNormalize, isCheckArtist, stringMatchLevel, format)
    # ファイルの時
    else:
      mainProcess(target, dpath, isOverWrite, isNormalize, isCheckArtist, stringMatchLevel, format)

# ----------------------------------------
# ログ出力
# ----------------------------------------
def printLog(mes):
  if IsVerbose: print(mes)

# ----------------------------------------
# スクリプト実行
# ----------------------------------------
if __name__ == "__main__":
  try:
    main()
  except Exception as err:
    print str(type(err))
    print err.message
    print traceback.print_exc()