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