Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/tapely.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_urllib_request,
   9 )
  10 from ..utils import (
  11     clean_html,
  12     ExtractorError,
  13     float_or_none,
  14     parse_iso8601,
  15 )
  16
  17
  18 class TapelyIE(InfoExtractor):
  19     _VALID_URL = r'https?://(?:www\.)?tape\.ly/(?P<id>[A-Za-z0-9\-_]+)(?:/(?P<songnr>\d+))?'
  20     _API_URL = 'http://tape.ly/showtape?id={0:}'
  21     _S3_SONG_URL = 'http://mytape.s3.amazonaws.com/{0:}'
  22     _SOUNDCLOUD_SONG_URL = 'http://api.soundcloud.com{0:}'
  23     _TESTS = [
  24         {
  25             'url': 'http://tape.ly/my-grief-as-told-by-water',
  26             'info_dict': {
  27                 'id': 23952,
  28                 'title': 'my grief as told by water',
  29                 'thumbnail': 're:^https?://.*\.png$',
  30                 'uploader_id': 16484,
  31                 'timestamp': 1411848286,
  32                 'description': 'For Robin and Ponkers, whom the tides of life have taken out to sea.',
  33             },
  34             'playlist_count': 13,
  35         },
  36         {
  37             'url': 'http://tape.ly/my-grief-as-told-by-water/1',
  38             'md5': '79031f459fdec6530663b854cbc5715c',
  39             'info_dict': {
  40                 'id': 258464,
  41                 'title': 'Dreaming Awake  (My Brightest Diamond)',
  42                 'ext': 'm4a',
  43             },
  44         },
  45     ]
  46
  47     def _real_extract(self, url):
  48         mobj = re.match(self._VALID_URL, url)
  49         display_id = mobj.group('id')
  50
  51         playlist_url = self._API_URL.format(display_id)
  52         request = compat_urllib_request.Request(playlist_url)
  53         request.add_header('X-Requested-With', 'XMLHttpRequest')
  54         request.add_header('Accept', 'application/json')
  55         request.add_header('Referer', url)
  56
  57         playlist = self._download_json(request, display_id)
  58
  59         tape = playlist['tape']
  60
  61         entries = []
  62         for s in tape['songs']:
  63             song = s['song']
  64             entry = {
  65                 'id': song['id'],
  66                 'duration': float_or_none(song.get('songduration'), 1000),
  67                 'title': song['title'],
  68             }
  69             if song['source'] == 'S3':
  70                 entry.update({
  71                     'url': self._S3_SONG_URL.format(song['filename']),
  72                 })
  73                 entries.append(entry)
  74             elif song['source'] == 'YT':
  75                 self.to_screen('YouTube video detected')
  76                 yt_id = song['filename'].replace('/youtube/', '')
  77                 entry.update(self.url_result(yt_id, 'Youtube', video_id=yt_id))
  78                 entries.append(entry)
  79             elif song['source'] == 'SC':
  80                 self.to_screen('SoundCloud song detected')
  81                 sc_url = self._SOUNDCLOUD_SONG_URL.format(song['filename'])
  82                 entry.update(self.url_result(sc_url, 'Soundcloud'))
  83                 entries.append(entry)
  84             else:
  85                 self.report_warning('Unknown song source: %s' % song['source'])
  86
  87         if mobj.group('songnr'):
  88             songnr = int(mobj.group('songnr')) - 1
  89             try:
  90                 return entries[songnr]
  91             except IndexError:
  92                 raise ExtractorError(
  93                     'No song with index: %s' % mobj.group('songnr'),
  94                     expected=True)
  95
  96         return {
  97             '_type': 'playlist',
  98             'id': tape['id'],
  99             'display_id': display_id,
 100             'title': tape['name'],
 101             'entries': entries,
 102             'thumbnail': tape.get('image_url'),
 103             'description': clean_html(tape.get('subtext')),
 104             'like_count': tape.get('likescount'),
 105             'uploader_id': tape.get('user_id'),
 106             'timestamp': parse_iso8601(tape.get('published_at')),
 107         }