youtube_dl/extractor/zype.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_HTTPError
   8 from ..utils import (
   9     dict_get,
  10     ExtractorError,
  11     int_or_none,
  12     js_to_json,
  13     parse_iso8601,
  14 )
  15
  16
  17 class ZypeIE(InfoExtractor):
  18     _ID_RE = r'[\da-fA-F]+'
  19     _COMMON_RE = r'//player\.zype\.com/embed/%s\.(?:js|json|html)\?.*?(?:access_token|(?:ap[ip]|player)_key)='
  20     _VALID_URL = r'https?:%s[^&]+' % (_COMMON_RE % ('(?P<id>%s)' % _ID_RE))
  21     _TEST = {
  22         'url': 'https://player.zype.com/embed/5b400b834b32992a310622b9.js?api_key=jZ9GUhRmxcPvX7M3SlfejB6Hle9jyHTdk2jVxG7wOHPLODgncEKVdPYBhuz9iWXQ&autoplay=false&controls=true&da=false',
  23         'md5': 'eaee31d474c76a955bdaba02a505c595',
  24         'info_dict': {
  25             'id': '5b400b834b32992a310622b9',
  26             'ext': 'mp4',
  27             'title': 'Smoky Barbecue Favorites',
  28             'thumbnail': r're:^https?://.*\.jpe?g',
  29             'description': 'md5:5ff01e76316bd8d46508af26dc86023b',
  30             'timestamp': 1504915200,
  31             'upload_date': '20170909',
  32         },
  33     }
  34
  35     @staticmethod
  36     def _extract_urls(webpage):
  37         return [
  38             mobj.group('url')
  39             for mobj in re.finditer(
  40                 r'<script[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?%s.+?)\1' % (ZypeIE._COMMON_RE % ZypeIE._ID_RE),
  41                 webpage)]
  42
  43     def _real_extract(self, url):
  44         video_id = self._match_id(url)
  45
  46         try:
  47             response = self._download_json(re.sub(
  48                 r'\.(?:js|html)\?', '.json?', url), video_id)['response']
  49         except ExtractorError as e:
  50             if isinstance(e.cause, compat_HTTPError) and e.cause.code in (400, 401, 403):
  51                 raise ExtractorError(self._parse_json(
  52                     e.cause.read().decode(), video_id)['message'], expected=True)
  53             raise
  54
  55         body = response['body']
  56         video = response['video']
  57         title = video['title']
  58
  59         if isinstance(body, dict):
  60             formats = []
  61             for output in body.get('outputs', []):
  62                 output_url = output.get('url')
  63                 if not output_url:
  64                     continue
  65                 name = output.get('name')
  66                 if name == 'm3u8':
  67                     formats = self._extract_m3u8_formats(
  68                         output_url, video_id, 'mp4',
  69                         'm3u8_native', m3u8_id='hls', fatal=False)
  70                 else:
  71                     f = {
  72                         'format_id': name,
  73                         'tbr': int_or_none(output.get('bitrate')),
  74                         'url': output_url,
  75                     }
  76                     if name in ('m4a', 'mp3'):
  77                         f['vcodec'] = 'none'
  78                     else:
  79                         f.update({
  80                             'height': int_or_none(output.get('height')),
  81                             'width': int_or_none(output.get('width')),
  82                         })
  83                     formats.append(f)
  84             text_tracks = body.get('subtitles') or []
  85         else:
  86             m3u8_url = self._search_regex(
  87                 r'(["\'])(?P<url>(?:(?!\1).)+\.m3u8(?:(?!\1).)*)\1',
  88                 body, 'm3u8 url', group='url', default=None)
  89             if not m3u8_url:
  90                 source = self._search_regex(
  91                     r'(?s)sources\s*:\s*\[\s*({.+?})\s*\]', body, 'source')
  92
  93                 def get_attr(key):
  94                     return self._search_regex(
  95                         r'\b%s\s*:\s*([\'"])(?P<val>(?:(?!\1).)+)\1' % key,
  96                         source, key, group='val')
  97
  98                 if get_attr('integration') == 'verizon-media':
  99                     m3u8_url = 'https://content.uplynk.com/%s.m3u8' % get_attr('id')
 100             formats = self._extract_m3u8_formats(
 101                 m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls')
 102             text_tracks = self._search_regex(
 103                 r'textTracks\s*:\s*(\[[^]]+\])',
 104                 body, 'text tracks', default=None)
 105             if text_tracks:
 106                 text_tracks = self._parse_json(
 107                     text_tracks, video_id, js_to_json, False)
 108         self._sort_formats(formats)
 109
 110         subtitles = {}
 111         if text_tracks:
 112             for text_track in text_tracks:
 113                 tt_url = dict_get(text_track, ('file', 'src'))
 114                 if not tt_url:
 115                     continue
 116                 subtitles.setdefault(text_track.get('label') or 'English', []).append({
 117                     'url': tt_url,
 118                 })
 119
 120         thumbnails = []
 121         for thumbnail in video.get('thumbnails', []):
 122             thumbnail_url = thumbnail.get('url')
 123             if not thumbnail_url:
 124                 continue
 125             thumbnails.append({
 126                 'url': thumbnail_url,
 127                 'width': int_or_none(thumbnail.get('width')),
 128                 'height': int_or_none(thumbnail.get('height')),
 129             })
 130
 131         return {
 132             'id': video_id,
 133             'display_id': video.get('friendly_title'),
 134             'title': title,
 135             'thumbnails': thumbnails,
 136             'description': dict_get(video, ('description', 'ott_description', 'short_description')),
 137             'timestamp': parse_iso8601(video.get('published_at')),
 138             'duration': int_or_none(video.get('duration')),
 139             'view_count': int_or_none(video.get('request_count')),
 140             'average_rating': int_or_none(video.get('rating')),
 141             'season_number': int_or_none(video.get('season')),
 142             'episode_number': int_or_none(video.get('episode')),
 143             'formats': formats,
 144             'subtitles': subtitles,
 145         }