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

   1 from __future__ import unicode_literals
   2
   3 import itertools
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor, SearchInfoExtractor
   8 from ..utils import (
   9     compat_urllib_parse,
  10     compat_urlparse,
  11     clean_html,
  12     int_or_none,
  13 )
  14
  15
  16 class YahooIE(InfoExtractor):
  17     IE_DESC = 'Yahoo screen and movies'
  18     _VALID_URL = r'https?://(?:screen|movies)\.yahoo\.com/.*?-(?P<id>[0-9]+)(?:-[a-z]+)?\.html'
  19     _TESTS = [
  20         {
  21             'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
  22             'md5': '4962b075c08be8690a922ee026d05e69',
  23             'info_dict': {
  24                 'id': '2d25e626-2378-391f-ada0-ddaf1417e588',
  25                 'ext': 'mp4',
  26                 'title': 'Julian Smith & Travis Legg Watch Julian Smith',
  27                 'description': 'Julian and Travis watch Julian Smith',
  28             },
  29         },
  30         {
  31             'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html',
  32             'md5': 'd6e6fc6e1313c608f316ddad7b82b306',
  33             'info_dict': {
  34                 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9',
  35                 'ext': 'mp4',
  36                 'title': 'Codefellas - The Cougar Lies with Spanish Moss',
  37                 'description': 'Agent Topple\'s mustache does its dirty work, and Nicole brokers a deal for peace. But why is the NSA collecting millions of Instagram brunch photos? And if your waffles have nothing to hide, what are they so worried about?',
  38             },
  39         },
  40         {
  41             'url': 'https://movies.yahoo.com/video/world-loves-spider-man-190819223.html',
  42             'md5': '410b7104aa9893b765bc22787a22f3d9',
  43             'info_dict': {
  44                 'id': '516ed8e2-2c4f-339f-a211-7a8b49d30845',
  45                 'ext': 'mp4',
  46                 'title': 'The World Loves Spider-Man',
  47                 'description': '''People all over the world are celebrating the release of \"The Amazing Spider-Man 2.\" We're taking a look at the enthusiastic response Spider-Man has received from viewers all over the world.''',
  48             }
  49         }
  50     ]
  51
  52     def _real_extract(self, url):
  53         mobj = re.match(self._VALID_URL, url)
  54         video_id = mobj.group('id')
  55         webpage = self._download_webpage(url, video_id)
  56
  57         items_json = self._search_regex(
  58             r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE,
  59             default=None)
  60         if items_json is None:
  61             CONTENT_ID_REGEXES = [
  62                 r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"',
  63                 r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"'
  64             ]
  65             long_id = self._search_regex(CONTENT_ID_REGEXES, webpage, 'content ID')
  66             video_id = long_id
  67         else:
  68             items = json.loads(items_json)
  69             info = items['mediaItems']['query']['results']['mediaObj'][0]
  70             # The 'meta' field is not always in the video webpage, we request it
  71             # from another page
  72             long_id = info['id']
  73         return self._get_info(long_id, video_id, webpage)
  74
  75     def _get_info(self, long_id, video_id, webpage):
  76         query = ('SELECT * FROM yahoo.media.video.streams WHERE id="%s"'
  77                  ' AND plrs="86Gj0vCaSzV_Iuf6hNylf2" AND region="US"'
  78                  ' AND protocol="http"' % long_id)
  79         data = compat_urllib_parse.urlencode({
  80             'q': query,
  81             'env': 'prod',
  82             'format': 'json',
  83         })
  84         query_result = self._download_json(
  85             'http://video.query.yahoo.com/v1/public/yql?' + data,
  86             video_id, 'Downloading video info')
  87         info = query_result['query']['results']['mediaObj'][0]
  88         meta = info['meta']
  89
  90         formats = []
  91         for s in info['streams']:
  92             format_info = {
  93                 'width': int_or_none(s.get('width')),
  94                 'height': int_or_none(s.get('height')),
  95                 'tbr': int_or_none(s.get('bitrate')),
  96             }
  97
  98             host = s['host']
  99             path = s['path']
 100             if host.startswith('rtmp'):
 101                 format_info.update({
 102                     'url': host,
 103                     'play_path': path,
 104                     'ext': 'flv',
 105                 })
 106             else:
 107                 format_url = compat_urlparse.urljoin(host, path)
 108                 format_info['url'] = format_url
 109             formats.append(format_info)
 110
 111         self._sort_formats(formats)
 112
 113         return {
 114             'id': video_id,
 115             'title': meta['title'],
 116             'formats': formats,
 117             'description': clean_html(meta['description']),
 118             'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage),
 119         }
 120
 121
 122 class YahooNewsIE(YahooIE):
 123     IE_NAME = 'yahoo:news'
 124     _VALID_URL = r'http://news\.yahoo\.com/video/.*?-(?P<id>\d*?)\.html'
 125
 126     _TESTS = [{
 127         'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
 128         'md5': '67010fdf3a08d290e060a4dd96baa07b',
 129         'info_dict': {
 130             'id': '104538833',
 131             'ext': 'mp4',
 132             'title': 'China Moses Is Crazy About the Blues',
 133             'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0',
 134         },
 135     }]
 136
 137     def _real_extract(self, url):
 138         mobj = re.match(self._VALID_URL, url)
 139         video_id = mobj.group('id')
 140         webpage = self._download_webpage(url, video_id)
 141         long_id = self._search_regex(r'contentId: \'(.+?)\',', webpage, 'long id')
 142         return self._get_info(long_id, video_id, webpage)
 143
 144
 145 class YahooSearchIE(SearchInfoExtractor):
 146     IE_DESC = 'Yahoo screen search'
 147     _MAX_RESULTS = 1000
 148     IE_NAME = 'screen.yahoo:search'
 149     _SEARCH_KEY = 'yvsearch'
 150
 151     def _get_n_results(self, query, n):
 152         """Get a specified number of results for a query"""
 153         entries = []
 154         for pagenum in itertools.count(0):
 155             result_url = 'http://video.search.yahoo.com/search/?p=%s&fr=screen&o=js&gs=0&b=%d' % (compat_urllib_parse.quote_plus(query), pagenum * 30)
 156             info = self._download_json(result_url, query,
 157                 note='Downloading results page '+str(pagenum+1))
 158             m = info['m']
 159             results = info['results']
 160
 161             for (i, r) in enumerate(results):
 162                 if (pagenum * 30) + i >= n:
 163                     break
 164                 mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r)
 165                 e = self.url_result('http://' + mobj.group('url'), 'Yahoo')
 166                 entries.append(e)
 167             if (pagenum * 30 + i >= n) or (m['last'] >= (m['total'] - 1)):
 168                 break
 169
 170         return {
 171             '_type': 'playlist',
 172             'id': query,
 173             'entries': entries,
 174         }