_ Git - youtube-dl/blob - youtube_dl/extractor/adobetv.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     parse_duration,
   8     unified_strdate,
   9     str_to_int,
  10     int_or_none,
  11     float_or_none,
  12     ISO639Utils,
  13     determine_ext,
  14 )
  15
  16
  17 class AdobeTVBaseIE(InfoExtractor):
  18     _API_BASE_URL = 'http://tv.adobe.com/api/v4/'
  19
  20
  21 class AdobeTVIE(AdobeTVBaseIE):
  22     _VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?watch/(?P<show_urlname>[^/]+)/(?P<id>[^/]+)'
  23
  24     _TEST = {
  25         'url': 'http://tv.adobe.com/watch/the-complete-picture-with-julieanne-kost/quick-tip-how-to-draw-a-circle-around-an-object-in-photoshop/',
  26         'md5': '9bc5727bcdd55251f35ad311ca74fa1e',
  27         'info_dict': {
  28             'id': '10981',
  29             'ext': 'mp4',
  30             'title': 'Quick Tip - How to Draw a Circle Around an Object in Photoshop',
  31             'description': 'md5:99ec318dc909d7ba2a1f2b038f7d2311',
  32             'thumbnail': 're:https?://.*\.jpg$',
  33             'upload_date': '20110914',
  34             'duration': 60,
  35             'view_count': int,
  36         },
  37     }
  38
  39     def _real_extract(self, url):
  40         language, show_urlname, urlname = re.match(self._VALID_URL, url).groups()
  41         if not language:
  42             language = 'en'
  43
  44         video_data = self._download_json(
  45             self._API_BASE_URL + 'episode/get/?language=%s&show_urlname=%s&urlname=%s&disclosure=standard' % (language, show_urlname, urlname),
  46             urlname)['data'][0]
  47
  48         formats = [{
  49             'url': source['url'],
  50             'format_id': source.get('quality_level') or source['url'].split('-')[-1].split('.')[0] or None,
  51             'width': int_or_none(source.get('width')),
  52             'height': int_or_none(source.get('height')),
  53             'tbr': int_or_none(source.get('video_data_rate')),
  54         } for source in video_data['videos']]
  55         self._sort_formats(formats)
  56
  57         return {
  58             'id': str(video_data['id']),
  59             'title': video_data['title'],
  60             'description': video_data.get('description'),
  61             'thumbnail': video_data.get('thumbnail'),
  62             'upload_date': unified_strdate(video_data.get('start_date')),
  63             'duration': parse_duration(video_data.get('duration')),
  64             'view_count': str_to_int(video_data.get('playcount')),
  65             'formats': formats,
  66         }
  67
  68
  69 class AdobeTVPlaylistBaseIE(AdobeTVBaseIE):
  70     def _parse_page_data(self, page_data):
  71         return [self.url_result(self._get_element_url(element_data)) for element_data in page_data]
  72
  73     def _extract_playlist_entries(self, url, display_id):
  74         page = self._download_json(url, display_id)
  75         entries = self._parse_page_data(page['data'])
  76         for page_num in range(2, page['paging']['pages'] + 1):
  77             entries.extend(self._parse_page_data(
  78                 self._download_json(url + '&page=%d' % page_num, display_id)['data']))
  79         return entries
  80
  81
  82 class AdobeTVShowIE(AdobeTVPlaylistBaseIE):
  83     _VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?show/(?P<id>[^/]+)'
  84
  85     _TEST = {
  86         'url': 'http://tv.adobe.com/show/the-complete-picture-with-julieanne-kost',
  87         'info_dict': {
  88             'id': '36',
  89             'title': 'The Complete Picture with Julieanne Kost',
  90             'description': 'md5:fa50867102dcd1aa0ddf2ab039311b27',
  91         },
  92         'playlist_mincount': 136,
  93     }
  94
  95     def _get_element_url(self, element_data):
  96         return element_data['urls'][0]
  97
  98     def _real_extract(self, url):
  99         language, show_urlname = re.match(self._VALID_URL, url).groups()
 100         if not language:
 101             language = 'en'
 102         query = 'language=%s&show_urlname=%s' % (language, show_urlname)
 103
 104         show_data = self._download_json(self._API_BASE_URL + 'show/get/?%s' % query, show_urlname)['data'][0]
 105
 106         return self.playlist_result(
 107             self._extract_playlist_entries(self._API_BASE_URL + 'episode/?%s' % query, show_urlname),
 108             str(show_data['id']),
 109             show_data['show_name'],
 110             show_data['show_description'])
 111
 112
 113 class AdobeTVChannelIE(AdobeTVPlaylistBaseIE):
 114     _VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?channel/(?P<id>[^/]+)(?:/(?P<category_urlname>[^/]+))?'
 115
 116     _TEST = {
 117         'url': 'http://tv.adobe.com/channel/development',
 118         'info_dict': {
 119             'id': 'development',
 120         },
 121         'playlist_mincount': 96,
 122     }
 123
 124     def _get_element_url(self, element_data):
 125         return element_data['url']
 126
 127     def _real_extract(self, url):
 128         language, channel_urlname, category_urlname = re.match(self._VALID_URL, url).groups()
 129         if not language:
 130             language = 'en'
 131         query = 'language=%s&channel_urlname=%s' % (language, channel_urlname)
 132         if category_urlname:
 133             query += '&category_urlname=%s' % category_urlname
 134
 135         return self.playlist_result(
 136             self._extract_playlist_entries(self._API_BASE_URL + 'show/?%s' % query, channel_urlname),
 137             channel_urlname)
 138
 139
 140 class AdobeTVVideoIE(InfoExtractor):
 141     _VALID_URL = r'https?://video\.tv\.adobe\.com/v/(?P<id>\d+)'
 142
 143     _TEST = {
 144         # From https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners
 145         'url': 'https://video.tv.adobe.com/v/2456/',
 146         'md5': '43662b577c018ad707a63766462b1e87',
 147         'info_dict': {
 148             'id': '2456',
 149             'ext': 'mp4',
 150             'title': 'New experience with Acrobat DC',
 151             'description': 'New experience with Acrobat DC',
 152             'duration': 248.667,
 153         },
 154     }
 155
 156     def _real_extract(self, url):
 157         video_id = self._match_id(url)
 158         video_data = self._download_json(url + '?format=json', video_id)
 159
 160         formats = [{
 161             'format_id': '%s-%s' % (determine_ext(source['src']), source.get('height')),
 162             'url': source['src'],
 163             'width': int_or_none(source.get('width')),
 164             'height': int_or_none(source.get('height')),
 165             'tbr': int_or_none(source.get('bitrate')),
 166         } for source in video_data['sources']]
 167         self._sort_formats(formats)
 168
 169         # For both metadata and downloaded files the duration varies among
 170         # formats. I just pick the max one
 171         duration = max(filter(None, [
 172             float_or_none(source.get('duration'), scale=1000)
 173             for source in video_data['sources']]))
 174
 175         subtitles = {}
 176         for translation in video_data.get('translations', []):
 177             lang_id = translation.get('language_w3c') or ISO639Utils.long2short(translation['language_medium'])
 178             if lang_id not in subtitles:
 179                 subtitles[lang_id] = []
 180             subtitles[lang_id].append({
 181                 'url': translation['vttPath'],
 182                 'ext': 'vtt',
 183             })
 184
 185         return {
 186             'id': video_id,
 187             'formats': formats,
 188             'title': video_data['title'],
 189             'description': video_data.get('description'),
 190             'thumbnail': video_data['video'].get('poster'),
 191             'duration': duration,
 192             'subtitles': subtitles,
 193         }