_ Git - youtube-dl/blob - youtube_dl/extractor/yahoo.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import itertools
   5 import json
   6 import re
   7
   8 from .common import InfoExtractor, SearchInfoExtractor
   9 from ..compat import (
  10     compat_urllib_parse,
  11     compat_urlparse,
  12 )
  13 from ..utils import (
  14     clean_html,
  15     unescapeHTML,
  16     ExtractorError,
  17     int_or_none,
  18 )
  19
  20
  21 class YahooIE(InfoExtractor):
  22     IE_DESC = 'Yahoo screen and movies'
  23     _VALID_URL = r'(?P<url>(?P<host>https?://(?:[a-zA-Z]{2}\.)?[\da-zA-Z_-]+\.yahoo\.com)/(?:[^/]+/)*(?P<display_id>.+?)-(?P<id>[0-9]+)(?:-[a-z]+)?\.html)'
  24     _TESTS = [
  25         {
  26             'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
  27             'md5': '4962b075c08be8690a922ee026d05e69',
  28             'info_dict': {
  29                 'id': '2d25e626-2378-391f-ada0-ddaf1417e588',
  30                 'ext': 'mp4',
  31                 'title': 'Julian Smith & Travis Legg Watch Julian Smith',
  32                 'description': 'Julian and Travis watch Julian Smith',
  33                 'duration': 6863,
  34             },
  35         },
  36         {
  37             'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html',
  38             'md5': 'd6e6fc6e1313c608f316ddad7b82b306',
  39             'info_dict': {
  40                 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9',
  41                 'ext': 'mp4',
  42                 'title': 'Codefellas - The Cougar Lies with Spanish Moss',
  43                 'description': 'md5:66b627ab0a282b26352136ca96ce73c1',
  44                 'duration': 151,
  45             },
  46         },
  47         {
  48             'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
  49             'md5': '60e8ac193d8fb71997caa8fce54c6460',
  50             'info_dict': {
  51                 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb',
  52                 'ext': 'mp4',
  53                 'title': "Yahoo Saves 'Community'",
  54                 'description': 'md5:4d4145af2fd3de00cbb6c1d664105053',
  55                 'duration': 170,
  56             }
  57         },
  58         {
  59             'url': 'https://tw.screen.yahoo.com/taipei-opinion-poll/選情站報-街頭民調-台北市篇-102823042.html',
  60             'md5': '92a7fdd8a08783c68a174d7aa067dde8',
  61             'info_dict': {
  62                 'id': '7a23b569-7bea-36cb-85b9-bd5301a0a1fb',
  63                 'ext': 'mp4',
  64                 'title': '選情站報 街頭民調 台北市篇',
  65                 'description': '選情站報 街頭民調 台北市篇',
  66                 'duration': 429,
  67             }
  68         },
  69         {
  70             'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html',
  71             'md5': '0b51660361f0e27c9789e7037ef76f4b',
  72             'info_dict': {
  73                 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58',
  74                 'ext': 'mp4',
  75                 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder',
  76                 'description': 'md5:f66c890e1490f4910a9953c941dee944',
  77                 'duration': 97,
  78             }
  79         },
  80         {
  81             'url': 'https://ca.sports.yahoo.com/video/program-makes-hockey-more-affordable-013127711.html',
  82             'md5': '57e06440778b1828a6079d2f744212c4',
  83             'info_dict': {
  84                 'id': 'c9fa2a36-0d4d-3937-b8f6-cc0fb1881e73',
  85                 'ext': 'mp4',
  86                 'title': 'Program that makes hockey more affordable not offered in Manitoba',
  87                 'description': 'md5:c54a609f4c078d92b74ffb9bf1f496f4',
  88                 'duration': 121,
  89             }
  90         }, {
  91             'url': 'https://ca.finance.yahoo.com/news/20-most-valuable-brands-world-112600775.html',
  92             'md5': '3e401e4eed6325aa29d9b96125fd5b4f',
  93             'info_dict': {
  94                 'id': 'c1b4c09c-8ed8-3b65-8b05-169c55358a83',
  95                 'ext': 'mp4',
  96                 'title': "Apple Is The World's Most Valuable Brand",
  97                 'description': 'md5:73eabc1a11c6f59752593b2ceefa1262',
  98                 'duration': 21,
  99             }
 100         }, {
 101             'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
 102             'md5': '67010fdf3a08d290e060a4dd96baa07b',
 103             'info_dict': {
 104                 'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521',
 105                 'ext': 'mp4',
 106                 'title': 'China Moses Is Crazy About the Blues',
 107                 'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0',
 108                 'duration': 128,
 109             }
 110         }, {
 111             'url': 'https://in.lifestyle.yahoo.com/video/connect-dots-dark-side-virgo-090247395.html',
 112             'md5': 'd9a083ccf1379127bf25699d67e4791b',
 113             'info_dict': {
 114                 'id': '52aeeaa3-b3d1-30d8-9ef8-5d0cf05efb7c',
 115                 'ext': 'mp4',
 116                 'title': 'Connect the Dots: Dark Side of Virgo',
 117                 'description': 'md5:1428185051cfd1949807ad4ff6d3686a',
 118                 'duration': 201,
 119             }
 120         }, {
 121             'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html',
 122             'md5': '989396ae73d20c6f057746fb226aa215',
 123             'info_dict': {
 124                 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
 125                 'ext': 'mp4',
 126                 'title': '\'True Story\' Trailer',
 127                 'description': 'True Story',
 128                 'duration': 150,
 129             },
 130         }, {
 131             'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html',
 132             'only_matching': True,
 133         }
 134     ]
 135
 136     def _real_extract(self, url):
 137         mobj = re.match(self._VALID_URL, url)
 138         display_id = mobj.group('display_id')
 139         page_id = mobj.group('id')
 140         url = mobj.group('url')
 141         host = mobj.group('host')
 142         webpage = self._download_webpage(url, display_id)
 143
 144         # Look for iframed media first
 145         iframe_m = re.search(r'<iframe[^>]+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage)
 146         if iframe_m:
 147             iframepage = self._download_webpage(
 148                 host + iframe_m.group(1), display_id, 'Downloading iframe webpage')
 149             items_json = self._search_regex(
 150                 r'mediaItems: (\[.+?\])$', iframepage, 'items', flags=re.MULTILINE, default=None)
 151             if items_json:
 152                 items = json.loads(items_json)
 153                 video_id = items[0]['id']
 154                 return self._get_info(video_id, display_id, webpage)
 155
 156         items_json = self._search_regex(
 157             r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE,
 158             default=None)
 159         if items_json is None:
 160             CONTENT_ID_REGEXES = [
 161                 r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"',
 162                 r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"',
 163                 r'"first_videoid"\s*:\s*"([^"]+)"',
 164                 r'%s[^}]*"ccm_id"\s*:\s*"([^"]+)"' % re.escape(page_id),
 165             ]
 166             video_id = self._search_regex(CONTENT_ID_REGEXES, webpage, 'content ID')
 167         else:
 168             items = json.loads(items_json)
 169             info = items['mediaItems']['query']['results']['mediaObj'][0]
 170             # The 'meta' field is not always in the video webpage, we request it
 171             # from another page
 172             video_id = info['id']
 173         return self._get_info(video_id, display_id, webpage)
 174
 175     def _get_info(self, video_id, display_id, webpage):
 176         region = self._search_regex(
 177             r'\\?"region\\?"\s*:\s*\\?"([^"]+?)\\?"',
 178             webpage, 'region', fatal=False, default='US')
 179         query = ('SELECT * FROM yahoo.media.video.streams WHERE id="%s"'
 180                  ' AND plrs="86Gj0vCaSzV_Iuf6hNylf2" AND region="%s"'
 181                  ' AND protocol="http"' % (video_id, region))
 182         data = compat_urllib_parse.urlencode({
 183             'q': query,
 184             'env': 'prod',
 185             'format': 'json',
 186         })
 187         query_result = self._download_json(
 188             'http://video.query.yahoo.com/v1/public/yql?' + data,
 189             display_id, 'Downloading video info')
 190
 191         info = query_result['query']['results']['mediaObj'][0]
 192         meta = info.get('meta')
 193
 194         if not meta:
 195             msg = info['status'].get('msg')
 196             if msg:
 197                 raise ExtractorError(
 198                     '%s returned error: %s' % (self.IE_NAME, msg), expected=True)
 199             raise ExtractorError('Unable to extract media object meta')
 200
 201         formats = []
 202         for s in info['streams']:
 203             format_info = {
 204                 'width': int_or_none(s.get('width')),
 205                 'height': int_or_none(s.get('height')),
 206                 'tbr': int_or_none(s.get('bitrate')),
 207             }
 208
 209             host = s['host']
 210             path = s['path']
 211             if host.startswith('rtmp'):
 212                 format_info.update({
 213                     'url': host,
 214                     'play_path': path,
 215                     'ext': 'flv',
 216                 })
 217             else:
 218                 format_url = compat_urlparse.urljoin(host, path)
 219                 format_info['url'] = format_url
 220             formats.append(format_info)
 221
 222         self._sort_formats(formats)
 223
 224         return {
 225             'id': video_id,
 226             'display_id': display_id,
 227             'title': unescapeHTML(meta['title']),
 228             'formats': formats,
 229             'description': clean_html(meta['description']),
 230             'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage),
 231             'duration': int_or_none(meta.get('duration')),
 232         }
 233
 234
 235 class YahooSearchIE(SearchInfoExtractor):
 236     IE_DESC = 'Yahoo screen search'
 237     _MAX_RESULTS = 1000
 238     IE_NAME = 'screen.yahoo:search'
 239     _SEARCH_KEY = 'yvsearch'
 240
 241     def _get_n_results(self, query, n):
 242         """Get a specified number of results for a query"""
 243         entries = []
 244         for pagenum in itertools.count(0):
 245             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)
 246             info = self._download_json(result_url, query,
 247                                        note='Downloading results page ' + str(pagenum + 1))
 248             m = info['m']
 249             results = info['results']
 250
 251             for (i, r) in enumerate(results):
 252                 if (pagenum * 30) + i >= n:
 253                     break
 254                 mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r)
 255                 e = self.url_result('http://' + mobj.group('url'), 'Yahoo')
 256                 entries.append(e)
 257             if (pagenum * 30 + i >= n) or (m['last'] >= (m['total'] - 1)):
 258                 break
 259
 260         return {
 261             '_type': 'playlist',
 262             'id': query,
 263             'entries': entries,
 264         }