metadata.py

   1 #!/usr/bin/env python
   2
   3 import os
   4 import subprocess
   5 from datetime import datetime
   6 from xml.dom import minidom
   7
   8 import mutagen
   9 from lrucache import LRUCache
  10
  11 import config
  12
  13 # Something to strip
  14 TRIBUNE_CR = ' Copyright Tribune Media Services, Inc.'
  15
  16 tivo_cache = LRUCache(50)
  17 mp4_cache = LRUCache(50)
  18
  19 def tag_data(element, tag):
  20     for name in tag.split('/'):
  21         new_element = element.getElementsByTagName(name)
  22         if not new_element:
  23             return ''
  24         element = new_element[0]
  25     if not element.firstChild:
  26         return ''
  27     return element.firstChild.data
  28
  29 def _vtag_data(element, tag):
  30     for name in tag.split('/'):
  31         new_element = element.getElementsByTagName(name)
  32         if not new_element:
  33             return []
  34         element = new_element[0]
  35     elements = element.getElementsByTagName('element')
  36     return [x.firstChild.data for x in elements if x.firstChild]
  37
  38 def _tag_value(element, tag):
  39     item = element.getElementsByTagName(tag)
  40     if item:
  41         value = item[0].attributes['value'].value
  42         name = item[0].firstChild.data
  43         return name[0] + value[0]
  44
  45 def from_moov(full_path):
  46     if full_path in mp4_cache:
  47         return mp4_cache[full_path]
  48
  49     metadata = {}
  50     len_desc = 0
  51
  52     try:
  53         mp4meta = mutagen.File(full_path)
  54         assert(mp4meta)
  55     except:
  56         mp4_cache[full_path] = {}
  57         return {}
  58
  59     for key, value in mp4meta.items():
  60         # The following 1-to-1 correspondence of atoms to pyTivo
  61         # variables is TV-biased
  62         keys = {'tvnn': 'callsign', 'tven': 'episodeNumber',
  63                 'tvsh': 'seriesTitle'}
  64         if type(value) == list:
  65             value = value[0]
  66         if key == 'stik':
  67             metadata['isEpisode'] = ['false', 'true'][value == 'TV Show']
  68         elif key in keys:
  69             metadata[keys[key]] = value
  70         # These keys begin with the copyright symbol \xA9
  71         elif key == '\xa9day':
  72             if len(value) == 4:
  73                 value += '-01-01T16:00:00Z'
  74             metadata['originalAirDate'] = value
  75             #metadata['time'] = value
  76         elif key in ['\xa9gen', 'gnre']:
  77             for k in ('vProgramGenre', 'vSeriesGenre'):
  78                 if k in metadata:
  79                     metadata[k].append(value)
  80                 else:
  81                     metadata[k] = [value]
  82         elif key == '\xa9nam':
  83             if 'tvsh' in mp4meta:
  84                 metadata['episodeTitle'] = value
  85             else:
  86                 metadata['title'] = value
  87
  88         # Description in desc, cmt, and/or ldes tags. Keep the longest.
  89         elif key in ['desc', '\xa9cmt', 'ldes'] and len(value) > len_desc:
  90             metadata['description'] = value
  91             len_desc = len(value)
  92
  93         # A common custom "reverse DNS format" tag
  94         # Possible TV values: TV-Y7 TV-Y TV-G TV-PG TV-14 TV-MA Unrated
  95         # Possible MPAA values: G PG PG-13 R NC-17 Unrated
  96         elif (key == '----:com.apple.iTunes:iTunEXTC' and
  97               ('us-tv' in value or 'mpaa' in value)):
  98             ratings = {'TV-Y7': 'x1', 'TV-Y': 'x2', 'TV-G': 'x3',
  99                        'TV-PG': 'x4', 'TV-14': 'x5', 'TV-MA': 'x6',
 100                        'Unrated': 'x7', 'G': 'G1', 'PG': 'P2',
 101                        'PG-13': 'P3', 'R': 'R4', 'NC-17': 'N6'}
 102             rating = value.split("|")[1]
 103             if rating in ratings:
 104                 if 'us-tv' in value:
 105                     metadata['tvRating'] = ratings[rating]
 106                 elif 'mpaa' in value:
 107                     metadata['mpaaRating'] = ratings[rating]
 108
 109         # Actors, directors, producers, AND screenwriters may be in a long
 110         # embedded XML plist, with key '----' and rDNS 'iTunMOVI'. Ughh!
 111
 112     mp4_cache[full_path] = metadata
 113     return metadata
 114
 115 def from_text(full_path):
 116     metadata = {}
 117     path, name = os.path.split(full_path)
 118     for metafile in [os.path.join(path, 'default.txt'), full_path + '.txt',
 119                      os.path.join(path, '.meta', name) + '.txt']:
 120         if os.path.exists(metafile):
 121             for line in file(metafile):
 122                 if line.strip().startswith('#') or not ':' in line:
 123                     continue
 124                 key, value = [x.strip() for x in line.split(':', 1)]
 125                 if key.startswith('v'):
 126                     if key in metadata:
 127                         metadata[key].append(value)
 128                     else:
 129                         metadata[key] = [value]
 130                 else:
 131                     metadata[key] = value
 132     return metadata
 133
 134 def basic(full_path):
 135     base_path, name = os.path.split(full_path)
 136     title, ext = os.path.splitext(name)
 137     mtime = os.stat(full_path).st_mtime
 138     if (mtime < 0):
 139         mtime = 0
 140     originalAirDate = datetime.fromtimestamp(mtime)
 141
 142     metadata = {'title': title,
 143                 'originalAirDate': originalAirDate.isoformat()}
 144     if ext.lower() in ['.mp4', '.m4v', '.mov']:
 145         metadata.update(from_moov(full_path))
 146     metadata.update(from_text(full_path))
 147
 148     return metadata
 149
 150 def from_container(xmldoc):
 151     metadata = {}
 152
 153     keys = {'title': 'Title', 'episodeTitle': 'EpisodeTitle',
 154             'description': 'Description', 'seriesId': 'SeriesId',
 155             'episodeNumber': 'EpisodeNumber', 'tvRating': 'TvRating',
 156             'displayMajorNumber': 'SourceChannel', 'callsign': 'SourceStation'}
 157
 158     details = xmldoc.getElementsByTagName('Details')[0]
 159
 160     for key in keys:
 161         data = tag_data(details, keys[key])
 162         if data:
 163             if key == 'description':
 164                 data = data.replace(TRIBUNE_CR, '')
 165             elif key == 'tvRating':
 166                 data = 'x' + data
 167             elif key == 'displayMajorNumber':
 168                 if '-' in data:
 169                     data, metadata['displayMinorNumber'] = data.split('-')
 170             metadata[key] = data
 171
 172     return metadata
 173
 174 def from_details(xmldoc):
 175     metadata = {}
 176
 177     showing = xmldoc.getElementsByTagName('showing')[0]
 178     program = showing.getElementsByTagName('program')[0]
 179
 180     items = {'description': 'program/description',
 181              'title': 'program/title',
 182              'episodeTitle': 'program/episodeTitle',
 183              'episodeNumber': 'program/episodeNumber',
 184              'seriesId': 'program/series/uniqueId',
 185              'seriesTitle': 'program/series/seriesTitle',
 186              'originalAirDate': 'program/originalAirDate',
 187              'isEpisode': 'program/isEpisode',
 188              'movieYear': 'program/movieYear',
 189              'partCount': 'partCount',
 190              'partIndex': 'partIndex'}
 191
 192     for item in items:
 193         data = tag_data(showing, items[item])
 194         if data:
 195             if item == 'description':
 196                 data = data.replace(TRIBUNE_CR, '')
 197             metadata[item] = data
 198
 199     vItems = ['vActor', 'vChoreographer', 'vDirector',
 200               'vExecProducer', 'vProgramGenre', 'vGuestStar',
 201               'vHost', 'vProducer', 'vWriter']
 202
 203     for item in vItems:
 204         data = _vtag_data(program, item)
 205         if data:
 206             metadata[item] = data
 207
 208     sb = showing.getElementsByTagName('showingBits')
 209     if sb:
 210         metadata['showingBits'] = sb[0].attributes['value'].value
 211
 212     for tag in ['starRating', 'mpaaRating', 'colorCode']:
 213         value = _tag_value(program, tag)
 214         if value:
 215             metadata[tag] = value
 216
 217     rating = _tag_value(showing, 'tvRating')
 218     if rating:
 219         metadata['tvRating'] = 'x' + rating[1]
 220
 221     return metadata
 222
 223 def from_tivo(full_path):
 224     if full_path in tivo_cache:
 225         return tivo_cache[full_path]
 226
 227     tdcat_path = config.get_bin('tdcat')
 228     tivo_mak = config.get_server('tivo_mak')
 229     if tdcat_path and tivo_mak:
 230         tcmd = [tdcat_path, '-m', tivo_mak, '-2', full_path]
 231         tdcat = subprocess.Popen(tcmd, stdout=subprocess.PIPE)
 232         xmldoc = minidom.parse(tdcat.stdout)
 233         metadata = from_details(xmldoc)
 234         tivo_cache[full_path] = metadata
 235     else:
 236         metadata = {}
 237
 238     return metadata
 239
 240 if __name__ == '__main__':
 241     import sys
 242     if len(sys.argv) > 1:
 243         metadata = {}
 244         ext = os.path.splitext(sys.argv[1])[1].lower()
 245         if ext == '.tivo':
 246             config.init([])
 247             metadata.update(from_tivo(sys.argv[1]))
 248         elif ext in ['.mp4', '.m4v', '.mov']:
 249             metadata.update(from_moov(sys.argv[1]))
 250         for key in metadata:
 251             value = metadata[key]
 252             if type(value) == list:
 253                 for item in value:
 254                     print '%s: %s' % (key, item.encode('utf8'))
 255             else:
 256                 print '%s: %s' % (key, value.encode('utf8'))