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