Lyrics embedding update

Fixed lyrics embedding, they will now work even if the lyrics were left out of the track data as we now insert it ourselves by grabbing it from the tracks individual page.

This of course slows things down when using the option so be aware.

README.rst

@@ -49,18 +49,10 @@ Details
 ::
 
     Usage:
-     bandcamp-dl [url]
-     bandcamp-dl [--template=<template>] [--base-dir=<dir>]
-                 [--full-album]
-                 (<url> | --artist=<artist> --album=<album>)
-                 [--overwrite]
-                 [--no-art]
-                 [--embed-lyrics]
-                 [--group]
-                 [--embed-art]
-                 [--no-slugify]
-     bandcamp-dl (-h | --help)
-     bandcamp-dl (--version)
+        bandcamp-dl [options] [URL]
+
+    Arguments:
+        URL         Bandcamp album/track URL
 
 Options
 -------
@@ -68,20 +60,21 @@ Options
 ::
 
     Options:
-     -h --help                   Show this screen.
-     -v --version                Show version.
-     -a --artist=<artist>        The artist's slug (from the URL)
-     -b --album=<album>          The album's slug (from the URL)
-     -t --template=<template>    Output filename template.
-                                 [default: %{artist}/%{album}/%{track} - %{title}]
-     -d --base-dir=<dir>         Base location of which all files are downloaded.
-     -f --full-album             Download only if all tracks are available.
-     -o --overwrite              Overwrite tracks that already exist. Default is False.
-     -n --no-art                 Skip grabbing album art
-     -e --embed-lyrics           Embed track lyrics (If available)
-     -g --group                  Use album/track Label as iTunes grouping
-     -r --embed-art              Embed album art (If available)
-     -n --no-slugify             Disable slugification of track, album, and artist names.
+        -h --help               Show this screen.
+        -v --version            Show version.
+        --artist=<artist>       The artist's slug (from the URL)
+        --track=<track>         The track's slug (from the URL)
+        --album=<album>         The album's slug (from the URL)
+        --template=<template>   Output filename template.
+                                [default: %{artist}/%{album}/%{track} - %{title}]
+        --base-dir=<dir>        Base location of which all files are downloaded.
+        -f --full-album         Download only if all tracks are available.
+        -o --overwrite          Overwrite tracks that already exist. Default is False.
+        -n --no-art             Skip grabbing album art
+        -e --embed-lyrics       Embed track lyrics (If available)
+        -g --group              Use album/track Label as iTunes grouping
+        -r --embed-art          Embed album art (If available)
+        -y --no-slugify         Disable slugification of track, album, and artist names.
 
 Filename Template
 -----------------

bandcamp_dl/__main__.py

@@ -79,9 +79,9 @@ def main():
         url = arguments['URL']
 
     if arguments['--no-art']:
-        album = bandcamp.parse(url, False)
+        album = bandcamp.parse(url, False, arguments['--embed-lyrics'])
     else:
-        album = bandcamp.parse(url)
+        album = bandcamp.parse(url, True, arguments['--embed-lyrics'])
 
     if arguments['--full-album'] and not album['full']:
         print("Full album not available. Skipping...")

bandcamp_dl/bandcamp.py

@@ -13,11 +13,12 @@ class Bandcamp:
     def __init__(self):
         self.headers = {'User-Agent': 'bandcamp-dl/{} (https://github.com/iheanyi/bandcamp-dl)'.format(__version__)}
 
-    def parse(self, url: str, art: bool=True) -> dict or None:
+    def parse(self, url: str, art: bool=True, lyrics: bool=False) -> dict or None:
         """Requests the page, cherry picks album info
 
         :param url: album/track url
         :param art: if True download album art
+        :param lyrics: if True fetch track lyrics
         :return: album metadata
         """
         try:
@@ -61,7 +62,10 @@ class Bandcamp:
             "date": str(dt.strptime(album_release, "%d %b %Y %H:%M:%S GMT").year)
         }
 
+        artist_url = album_json['url'].rpartition('/album/')[0]
         for track in self.tracks:
+            if lyrics:
+                track['lyrics'] = self.get_track_lyrics("{}{}#lyrics".format(artist_url, track['title_link']))
             if track['file'] is not None:
                 track = self.get_track_metadata(track)
                 album['tracks'].append(track)
@@ -72,6 +76,18 @@ class Bandcamp:
 
         return album
 
+    def get_track_lyrics(self, track_url):
+        track_page = requests.get(track_url, headers=self.headers)
+        try:
+            track_soup = BeautifulSoup(track_page.text, "lxml")
+        except FeatureNotFound:
+            track_soup = BeautifulSoup(track_page.text, "html.parser")
+        track_lyrics = track_soup.find("div", {"class": "lyricsText"})
+        if track_lyrics:
+            return track_lyrics.text
+        else:
+            return "lyrics unavailable"
+
     def all_tracks_available(self) -> bool:
         """Verify that all tracks have a url
 

bandcamp_dl/bandcampdownloader.py

@@ -108,12 +108,11 @@ class BandcampDownloader:
                 "album": album['title'],
                 "title": track['title'],
                 "track": track['track'],
+                # TODO: Find out why the 'lyrics' key seems to vanish.
+                "lyrics": track.get('lyrics', "lyrics unavailable"),
                 "date": album['date']
             }
 
-            if 'lyrics' in track and self.embed_lyrics:
-                track_meta['lyrics'] = track['lyrics']
-
             self.num_tracks = len(album['tracks'])
             self.track_num = track_index + 1
 
@@ -218,7 +217,7 @@ class BandcampDownloader:
         if self.grouping and 'label' in meta:
             audio["TIT1"] = TIT1(encoding=3, text=meta["label"])
 
-        if self.embed_lyrics and 'lyrics' in meta:
+        if self.embed_lyrics:
             audio["USLT"] = USLT(encoding=3, lang='eng', desc='', text=meta['lyrics'])
 
         if self.embed_art:

bandcamp_dl/bandcampjson.py

@@ -9,8 +9,6 @@ class BandcampJSON:
         self.targets = ['TralbumData', 'EmbedData', 'pagedata']
         self.json_data = []
 
-    # Lyrics can be found by appending #lyrics to the /track/ url
-
     def generate(self) -> list:
         """Iterate through targets grabbing needed data"""
         for target in self.targets: