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

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_str
   7 from ..utils import (
   8     parse_duration,
   9     unified_strdate,
  10     str_to_int,
  11     int_or_none,
  12     float_or_none,
  13     ISO639Utils,
  14     determine_ext,
  15 )
  16
  17
  18 class AdobeTVBaseIE(InfoExtractor):
  19     _API_BASE_URL = 'http://tv.adobe.com/api/v4/'
  20
  21
  22 class AdobeTVIE(AdobeTVBaseIE):
  23     _VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?watch/(?P<show_urlname>[^/]+)/(?P<id>[^/]+)'
  24
  25     _TEST = {
  26         'url': 'http://tv.adobe.com/watch/the-complete-picture-with-julieanne-kost/quick-tip-how-to-draw-a-circle-around-an-object-in-photoshop/',
  27         'md5': '9bc5727bcdd55251f35ad311ca74fa1e',
  28         'info_dict': {
  29             'id': '10981',
  30             'ext': 'mp4',
  31             'title': 'Quick Tip - How to Draw a Circle Around an Object in Photoshop',
  32             'description': 'md5:99ec318dc909d7ba2a1f2b038f7d2311',
  33             'thumbnail': 're:https?://.*\.jpg$',
  34             'upload_date': '20110914',
  35             'duration': 60,
  36             'view_count': int,
  37         },
  38     }
  39
  40     def _real_extract(self, url):
  41         language, show_urlname, urlname = re.match(self._VALID_URL, url).groups()
  42         if not language:
  43             language = 'en'
  44
  45         video_data = self._download_json(
  46             self._API_BASE_URL + 'episode/get/?language=%s&show_urlname=%s&urlname=%s&disclosure=standard' % (language, show_urlname, urlname),
  47             urlname)['data'][0]
  48
  49         formats = [{
  50             'url': source['url'],
  51             'format_id': source.get('quality_level') or source['url'].split('-')[-1].split('.')[0] or None,
  52             'width': int_or_none(source.get('width')),
  53             'height': int_or_none(source.get('height')),
  54             'tbr': int_or_none(source.get('video_data_rate')),
  55         } for source in video_data['videos']]
  56         self._sort_formats(formats)
  57
  58         return {
  59             'id': compat_str(video_data['id']),
  60             'title': video_data['title'],
  61             'description': video_data.get('description'),
  62             'thumbnail': video_data.get('thumbnail'),
  63             'upload_date': unified_strdate(video_data.get('start_date')),
  64             'duration': parse_duration(video_data.get('duration')),
  65             'view_count': str_to_int(video_data.get('playcount')),
  66             'formats': formats,
  67         }
  68
  69
  70 class AdobeTVPlaylistBaseIE(AdobeTVBaseIE):
  71     def _parse_page_data(self, page_data):
  72         return [self.url_result(self._get_element_url(element_data)) for element_data in page_data]
  73
  74     def _extract_playlist_entries(self, url, display_id):
  75         page = self._download_json(url, display_id)
  76         entries = self._parse_page_data(page['data'])
  77         for page_num in range(2, page['paging']['pages'] + 1):
  78             entries.extend(self._parse_page_data(
  79                 self._download_json(url + '&page=%d' % page_num, display_id)['data']))
  80         return entries
  81
  82
  83 class AdobeTVShowIE(AdobeTVPlaylistBaseIE):
  84     _VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?show/(?P<id>[^/]+)'
  85
  86     _TEST = {
  87         'url': 'http://tv.adobe.com/show/the-complete-picture-with-julieanne-kost',
  88         'info_dict': {
  89             'id': '36',
  90             'title': 'The Complete Picture with Julieanne Kost',
  91             'description': 'md5:fa50867102dcd1aa0ddf2ab039311b27',
  92         },
  93         'playlist_mincount': 136,
  94     }
  95
  96     def _get_element_url(self, element_data):
  97         return element_data['urls'][0]
  98
  99     def _real_extract(self, url):
 100         language, show_urlname = re.match(self._VALID_URL, url).groups()
 101         if not language:
 102             language = 'en'
 103         query = 'language=%s&show_urlname=%s' % (language, show_urlname)
 104
 105         show_data = self._download_json(self._API_BASE_URL + 'show/get/?%s' % query, show_urlname)['data'][0]
 106
 107         return self.playlist_result(
 108             self._extract_playlist_entries(self._API_BASE_URL + 'episode/?%s' % query, show_urlname),
 109             compat_str(show_data['id']),
 110             show_data['show_name'],
 111             show_data['show_description'])
 112
 113
 114 class AdobeTVChannelIE(AdobeTVPlaylistBaseIE):
 115     _VALID_URL = r'https?://tv\.adobe\.com/(?:(?P<language>fr|de|es|jp)/)?channel/(?P<id>[^/]+)(?:/(?P<category_urlname>[^/]+))?'
 116
 117     _TEST = {
 118         'url': 'http://tv.adobe.com/channel/development',
 119         'info_dict': {
 120             'id': 'development',
 121         },
 122         'playlist_mincount': 96,
 123     }
 124
 125     def _get_element_url(self, element_data):
 126         return element_data['url']
 127
 128     def _real_extract(self, url):
 129         language, channel_urlname, category_urlname = re.match(self._VALID_URL, url).groups()
 130         if not language:
 131             language = 'en'
 132         query = 'language=%s&channel_urlname=%s' % (language, channel_urlname)
 133         if category_urlname:
 134             query += '&category_urlname=%s' % category_urlname
 135
 136         return self.playlist_result(
 137             self._extract_playlist_entries(self._API_BASE_URL + 'show/?%s' % query, channel_urlname),
 138             channel_urlname)
 139
 140
 141 class AdobeTVVideoIE(InfoExtractor):
 142     _VALID_URL = r'https?://video\.tv\.adobe\.com/v/(?P<id>\d+)'
 143
 144     _TEST = {
 145         # From https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners
 146         'url': 'https://video.tv.adobe.com/v/2456/',
 147         'md5': '43662b577c018ad707a63766462b1e87',
 148         'info_dict': {
 149             'id': '2456',
 150             'ext': 'mp4',
 151             'title': 'New experience with Acrobat DC',
 152             'description': 'New experience with Acrobat DC',
 153             'duration': 248.667,
 154         },
 155     }
 156
 157     def _real_extract(self, url):
 158         video_id = self._match_id(url)
 159         webpage = self._download_webpage(url, video_id)
 160
 161         video_data = self._parse_json(self._search_regex(
 162             r'var\s+bridge\s*=\s*([^;]+);', webpage, 'bridged data'), video_id)
 163
 164         formats = [{
 165             'format_id': '%s-%s' % (determine_ext(source['src']), source.get('height')),
 166             'url': source['src'],
 167             'width': int_or_none(source.get('width')),
 168             'height': int_or_none(source.get('height')),
 169             'tbr': int_or_none(source.get('bitrate')),
 170         } for source in video_data['sources']]
 171         self._sort_formats(formats)
 172
 173         # For both metadata and downloaded files the duration varies among
 174         # formats. I just pick the max one
 175         duration = max(filter(None, [
 176             float_or_none(source.get('duration'), scale=1000)
 177             for source in video_data['sources']]))
 178
 179         subtitles = {}
 180         for translation in video_data.get('translations', []):
 181             lang_id = translation.get('language_w3c') or ISO639Utils.long2short(translation['language_medium'])
 182             if lang_id not in subtitles:
 183                 subtitles[lang_id] = []
 184             subtitles[lang_id].append({
 185                 'url': translation['vttPath'],
 186                 'ext': 'vtt',
 187             })
 188
 189         return {
 190             'id': video_id,
 191             'formats': formats,
 192             'title': video_data['title'],
 193             'description': video_data.get('description'),
 194             'thumbnail': video_data['video'].get('poster'),
 195             'duration': duration,
 196             'subtitles': subtitles,
 197         }