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