2 from __future__ import unicode_literals
8 from .common import InfoExtractor, SearchInfoExtractor
17 class YahooIE(InfoExtractor):
18 IE_DESC = 'Yahoo screen and movies'
19 _VALID_URL = r'(?P<url>https?://(?:.+?\.)?(?:screen|movies)\.yahoo\.com/.*?-(?P<id>[0-9]+)(?:-[a-z]+)?\.html)'
22 'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
23 'md5': '4962b075c08be8690a922ee026d05e69',
25 'id': '2d25e626-2378-391f-ada0-ddaf1417e588',
27 'title': 'Julian Smith & Travis Legg Watch Julian Smith',
28 'description': 'Julian and Travis watch Julian Smith',
33 'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html',
34 'md5': 'd6e6fc6e1313c608f316ddad7b82b306',
36 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9',
38 'title': 'Codefellas - The Cougar Lies with Spanish Moss',
39 'description': 'md5:66b627ab0a282b26352136ca96ce73c1',
44 'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
45 'md5': '60e8ac193d8fb71997caa8fce54c6460',
47 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb',
49 'title': "Yahoo Saves 'Community'",
50 'description': 'md5:4d4145af2fd3de00cbb6c1d664105053',
55 'url': 'https://tw.screen.yahoo.com/taipei-opinion-poll/選情站報-街頭民調-台北市篇-102823042.html',
56 'md5': '92a7fdd8a08783c68a174d7aa067dde8',
58 'id': '7a23b569-7bea-36cb-85b9-bd5301a0a1fb',
60 'title': '選情站報 街頭民調 台北市篇',
61 'description': '選情站報 街頭民調 台北市篇',
66 'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html ',
67 'md5': '0b51660361f0e27c9789e7037ef76f4b',
69 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58',
71 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder',
72 'description': 'md5:f66c890e1490f4910a9953c941dee944',
78 def _real_extract(self, url):
79 mobj = re.match(self._VALID_URL, url)
80 video_id = mobj.group('id')
81 url = mobj.group('url')
82 webpage = self._download_webpage(url, video_id)
84 items_json = self._search_regex(
85 r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE,
87 if items_json is None:
88 CONTENT_ID_REGEXES = [
89 r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"',
90 r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"',
91 r'"first_videoid"\s*:\s*"([^"]+)"',
93 long_id = self._search_regex(CONTENT_ID_REGEXES, webpage, 'content ID')
96 items = json.loads(items_json)
97 info = items['mediaItems']['query']['results']['mediaObj'][0]
98 # The 'meta' field is not always in the video webpage, we request it
101 return self._get_info(long_id, video_id, webpage)
103 def _get_info(self, long_id, video_id, webpage):
104 region = self._search_regex(
105 r'"region"\s*:\s*"([^"]+)"', webpage, 'region', fatal=False, default='US')
106 query = ('SELECT * FROM yahoo.media.video.streams WHERE id="%s"'
107 ' AND plrs="86Gj0vCaSzV_Iuf6hNylf2" AND region="%s"'
108 ' AND protocol="http"' % (long_id, region))
109 data = compat_urllib_parse.urlencode({
114 query_result = self._download_json(
115 'http://video.query.yahoo.com/v1/public/yql?' + data,
116 video_id, 'Downloading video info')
117 info = query_result['query']['results']['mediaObj'][0]
121 for s in info['streams']:
123 'width': int_or_none(s.get('width')),
124 'height': int_or_none(s.get('height')),
125 'tbr': int_or_none(s.get('bitrate')),
130 if host.startswith('rtmp'):
137 format_url = compat_urlparse.urljoin(host, path)
138 format_info['url'] = format_url
139 formats.append(format_info)
141 self._sort_formats(formats)
145 'title': meta['title'],
147 'description': clean_html(meta['description']),
148 'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage),
149 'duration': int_or_none(meta.get('duration')),
153 class YahooNewsIE(YahooIE):
154 IE_NAME = 'yahoo:news'
155 _VALID_URL = r'http://news\.yahoo\.com/video/.*?-(?P<id>\d*?)\.html'
158 'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
159 'md5': '67010fdf3a08d290e060a4dd96baa07b',
163 'title': 'China Moses Is Crazy About the Blues',
164 'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0',
168 def _real_extract(self, url):
169 mobj = re.match(self._VALID_URL, url)
170 video_id = mobj.group('id')
171 webpage = self._download_webpage(url, video_id)
172 long_id = self._search_regex(r'contentId: \'(.+?)\',', webpage, 'long id')
173 return self._get_info(long_id, video_id, webpage)
176 class YahooSearchIE(SearchInfoExtractor):
177 IE_DESC = 'Yahoo screen search'
179 IE_NAME = 'screen.yahoo:search'
180 _SEARCH_KEY = 'yvsearch'
182 def _get_n_results(self, query, n):
183 """Get a specified number of results for a query"""
185 for pagenum in itertools.count(0):
186 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)
187 info = self._download_json(result_url, query,
188 note='Downloading results page '+str(pagenum+1))
190 results = info['results']
192 for (i, r) in enumerate(results):
193 if (pagenum * 30) + i >= n:
195 mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r)
196 e = self.url_result('http://' + mobj.group('url'), 'Yahoo')
198 if (pagenum * 30 + i >= n) or (m['last'] >= (m['total'] - 1)):