_ 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 AdobeTVVideoIE(InfoExtractor):
  66     _VALID_URL = r'https?://video\.tv\.adobe\.com/v/(?P<id>\d+)'
  67
  68     _TEST = {
  69         # From https://helpx.adobe.com/acrobat/how-to/new-experience-acrobat-dc.html?set=acrobat--get-started--essential-beginners
  70         'url': 'https://video.tv.adobe.com/v/2456/',
  71         'md5': '43662b577c018ad707a63766462b1e87',
  72         'info_dict': {
  73             'id': '2456',
  74             'ext': 'mp4',
  75             'title': 'New experience with Acrobat DC',
  76             'description': 'New experience with Acrobat DC',
  77             'duration': 248.667,
  78         },
  79     }
  80
  81     def _real_extract(self, url):
  82         video_id = self._match_id(url)
  83         video_data = self._download_json(url + '?format=json', video_id)
  84
  85         formats = [{
  86             'format_id': '%s-%s' % (determine_ext(source['src']), source.get('height')),
  87             'url': source['src'],
  88             'width': int_or_none(source.get('width')),
  89             'height': int_or_none(source.get('height')),
  90             'tbr': int_or_none(source.get('bitrate')),
  91         } for source in video_data['sources']]
  92         self._sort_formats(formats)
  93
  94         # For both metadata and downloaded files the duration varies among
  95         # formats. I just pick the max one
  96         duration = max(filter(None, [
  97             float_or_none(source.get('duration'), scale=1000)
  98             for source in video_data['sources']]))
  99
 100         subtitles = {}
 101         for translation in video_data.get('translations', []):
 102             lang_id = translation.get('language_w3c') or ISO639Utils.long2short(translation['language_medium'])
 103             if lang_id not in subtitles:
 104                 subtitles[lang_id] = []
 105             subtitles[lang_id].append({
 106                 'url': translation['vttPath'],
 107                 'ext': 'vtt',
 108             })
 109
 110         return {
 111             'id': video_id,
 112             'formats': formats,
 113             'title': video_data['title'],
 114             'description': video_data.get('description'),
 115             'thumbnail': video_data['video'].get('poster'),
 116             'duration': duration,
 117             'subtitles': subtitles,
 118         }