addons/metadata.generic.albums/lib/allmusic.py

   1 # -*- coding: utf-8 -*-
   2
   3 import datetime
   4 import difflib
   5 import time
   6 import re
   7
   8 def allmusic_albumfind(data, artist, album):
   9     data = data.decode('utf-8')
  10     albums = []
  11     albumlist = re.findall('class="album">\s*(.*?)\s*</li', data, re.S)
  12     for item in albumlist:
  13         albumdata = {}
  14         albumartist = re.search('class="artist">.*?>(.*?)</a', item, re.S)
  15         if albumartist:
  16             albumdata['artist'] = albumartist.group(1)
  17         else: # classical album
  18             continue
  19         albumname = re.search('class="title">.*?>(.*?)</a', item, re.S)
  20         if albumname:
  21             albumdata['album'] = albumname.group(1)
  22         else: # not likely to happen, but just in case
  23             continue
  24         # filter inaccurate results
  25         artistmatch = difflib.SequenceMatcher(None, artist.lower(), albumdata['artist'].lower()).ratio()
  26         albummatch = difflib.SequenceMatcher(None, album.lower(), albumdata['album'].lower()).ratio()
  27         if artistmatch > 0.90 and albummatch > 0.90:
  28             albumurl = re.search('class="title">\s*<a href="(.*?)"', item)
  29             if albumurl:
  30                 albumdata['url'] = albumurl.group(1)
  31             else: # not likely to happen, but just in case
  32                 continue
  33             albums.append(albumdata)
  34             # we are only interested in the top result
  35             break
  36     return albums
  37
  38 def allmusic_albumdetails(data):
  39     data = data.decode('utf-8')
  40     albumdata = {}
  41     releasedata = re.search('class="release-date">.*?<span>(.*?)<', data, re.S)
  42     if releasedata:
  43         dateformat = releasedata.group(1)
  44         if len(dateformat) > 4:
  45             try:
  46                 # month day, year
  47                 albumdata['releasedate'] = datetime.datetime(*(time.strptime(dateformat, '%B %d, %Y')[0:3])).strftime('%Y-%m-%d')
  48             except:
  49                 # month, year
  50                 albumdata['releasedate'] = datetime.datetime(*(time.strptime(dateformat, '%B, %Y')[0:3])).strftime('%Y-%m')
  51         else:
  52             # year
  53             albumdata['releasedate'] = dateformat
  54     yeardata = re.search('class="year".*?>\s*(.*?)\s*<', data)
  55     if yeardata:
  56         albumdata['year'] = yeardata.group(1)
  57     genredata = re.search('class="genre">.*?">(.*?)<', data, re.S)
  58     if genredata:
  59         albumdata['genre'] = genredata.group(1)
  60     styledata = re.search('class="styles">.*?div>\s*(.*?)\s*</div', data, re.S)
  61     if styledata:
  62         stylelist = re.findall('">(.*?)<', styledata.group(1))
  63         if stylelist:
  64             albumdata['styles'] =  ' / '.join(stylelist)
  65     mooddata = re.search('class="moods">.*?div>\s*(.*?)\s*</div', data, re.S)
  66     if mooddata:
  67         moodlist = re.findall('">(.*?)<', mooddata.group(1))
  68         if moodlist:
  69             albumdata['moods'] =  ' / '.join(moodlist)
  70     themedata = re.search('class="themes">.*?div>\s*(.*?)\s*</div', data, re.S)
  71     if themedata:
  72         themelist = re.findall('">(.*?)<', themedata.group(1))
  73         if themelist:
  74             albumdata['themes'] =  ' / '.join(themelist)
  75     ratingdata = re.search('itemprop="ratingValue">\s*(.*?)\s*</div', data)
  76     if ratingdata:
  77         albumdata['rating'] = ratingdata.group(1)
  78     albumdata['votes'] = ''
  79     titledata = re.search('class="album-title".*?>\s*(.*?)\s*<', data, re.S)
  80     if titledata:
  81         albumdata['album'] = titledata.group(1)
  82     labeldata = re.search('class="label-catalog".*?<.*?>(.*?)<', data, re.S)
  83     if labeldata:
  84         albumdata['label'] = labeldata.group(1)
  85     artistdata = re.search('class="album-artist".*?<span.*?>\s*(.*?)\s*</span', data, re.S)
  86     if artistdata:
  87         artistlist = re.findall('">(.*?)<', artistdata.group(1))
  88         artists = []
  89         for item in artistlist:
  90             artistinfo = {}
  91             artistinfo['artist'] = item
  92             artists.append(artistinfo)
  93         if artists:
  94             albumdata['artist'] = artists
  95             albumdata['artist_description'] = ' / '.join(artistlist)
  96     thumbsdata = re.search('class="album-contain".*?src="(.*?)"', data, re.S)
  97     if thumbsdata:
  98         thumbs = []
  99         thumbdata = {}
 100         thumb = thumbsdata.group(1).rstrip('?partner=allrovi.com')
 101         # ignore internal blank thumb
 102         if thumb.startswith('http'):
 103             # 0=largest / 1=75 / 2=150 / 3=250 / 4=400 / 5=500 / 6=1080
 104             if thumb.endswith('f=5'):
 105                 thumbdata['image'] = thumb.replace('f=5', 'f=0')
 106                 thumbdata['preview'] = thumb.replace('f=5', 'f=2')
 107             else:
 108                 thumbdata['image'] = thumb
 109                 thumbdata['preview'] = thumb
 110             thumbdata['aspect'] = 'thumb'
 111             thumbs.append(thumbdata)
 112             albumdata['thumb'] = thumbs
 113     return albumdata