Raphaël G. Git Repositories - youtubedl/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 from .nbc import NBCSportsVPlayerIE
  21
  22
  23 class YahooIE(InfoExtractor):
  24     IE_DESC = 'Yahoo screen and movies'
  25     _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)'
  26     _TESTS = [
  27         {
  28             'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
  29             'info_dict': {
  30                 'id': '2d25e626-2378-391f-ada0-ddaf1417e588',
  31                 'ext': 'mp4',
  32                 'title': 'Julian Smith & Travis Legg Watch Julian Smith',
  33                 'description': 'Julian and Travis watch Julian Smith',
  34                 'duration': 6863,
  35             },
  36         },
  37         {
  38             'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html',
  39             'md5': 'd6e6fc6e1313c608f316ddad7b82b306',
  40             'info_dict': {
  41                 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9',
  42                 'ext': 'mp4',
  43                 'title': 'Codefellas - The Cougar Lies with Spanish Moss',
  44                 'description': 'md5:66b627ab0a282b26352136ca96ce73c1',
  45                 'duration': 151,
  46             },
  47         },
  48         {
  49             'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
  50             'md5': '60e8ac193d8fb71997caa8fce54c6460',
  51             'info_dict': {
  52                 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb',
  53                 'ext': 'mp4',
  54                 'title': "Yahoo Saves 'Community'",
  55                 'description': 'md5:4d4145af2fd3de00cbb6c1d664105053',
  56                 'duration': 170,
  57             }
  58         },
  59         {
  60             'url': 'https://tw.screen.yahoo.com/election-2014-askmayor/敢問市長-黃秀霜批賴清德-非常高傲-033009720.html',
  61             'md5': '3a09cf59349cfaddae1797acc3c087fc',
  62             'info_dict': {
  63                 'id': 'cac903b3-fcf4-3c14-b632-643ab541712f',
  64                 'ext': 'mp4',
  65                 'title': '敢問市長／黃秀霜批賴清德「非常高傲」',
  66                 'description': '直言台南沒捷運 交通居五都之末',
  67                 'duration': 396,
  68             }
  69         },
  70         {
  71             'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html',
  72             'md5': '0b51660361f0e27c9789e7037ef76f4b',
  73             'info_dict': {
  74                 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58',
  75                 'ext': 'mp4',
  76                 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder',
  77                 'description': 'md5:f66c890e1490f4910a9953c941dee944',
  78                 'duration': 97,
  79             }
  80         },
  81         {
  82             'url': 'https://ca.sports.yahoo.com/video/program-makes-hockey-more-affordable-013127711.html',
  83             'md5': '57e06440778b1828a6079d2f744212c4',
  84             'info_dict': {
  85                 'id': 'c9fa2a36-0d4d-3937-b8f6-cc0fb1881e73',
  86                 'ext': 'mp4',
  87                 'title': 'Program that makes hockey more affordable not offered in Manitoba',
  88                 'description': 'md5:c54a609f4c078d92b74ffb9bf1f496f4',
  89                 'duration': 121,
  90             }
  91         }, {
  92             'url': 'https://ca.finance.yahoo.com/news/hackers-sony-more-trouble-well-154609075.html',
  93             'md5': '226a895aae7e21b0129e2a2006fe9690',
  94             'info_dict': {
  95                 'id': 'e624c4bc-3389-34de-9dfc-025f74943409',
  96                 'ext': 'mp4',
  97                 'title': '\'The Interview\' TV Spot: War',
  98                 'description': 'The Interview',
  99                 'duration': 30,
 100             }
 101         }, {
 102             'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
 103             'md5': '67010fdf3a08d290e060a4dd96baa07b',
 104             'info_dict': {
 105                 'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521',
 106                 'ext': 'mp4',
 107                 'title': 'China Moses Is Crazy About the Blues',
 108                 'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0',
 109                 'duration': 128,
 110             }
 111         }, {
 112             'url': 'https://in.lifestyle.yahoo.com/video/connect-dots-dark-side-virgo-090247395.html',
 113             'md5': 'd9a083ccf1379127bf25699d67e4791b',
 114             'info_dict': {
 115                 'id': '52aeeaa3-b3d1-30d8-9ef8-5d0cf05efb7c',
 116                 'ext': 'mp4',
 117                 'title': 'Connect the Dots: Dark Side of Virgo',
 118                 'description': 'md5:1428185051cfd1949807ad4ff6d3686a',
 119                 'duration': 201,
 120             }
 121         }, {
 122             'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html',
 123             'md5': '989396ae73d20c6f057746fb226aa215',
 124             'info_dict': {
 125                 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
 126                 'ext': 'mp4',
 127                 'title': '\'True Story\' Trailer',
 128                 'description': 'True Story',
 129                 'duration': 150,
 130             },
 131         }, {
 132             'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html',
 133             'only_matching': True,
 134         }, {
 135             'note': 'NBC Sports embeds',
 136             'url': 'http://sports.yahoo.com/blogs/ncaab-the-dagger/tyler-kalinoski-s-buzzer-beater-caps-davidson-s-comeback-win-185609842.html?guid=nbc_cbk_davidsonbuzzerbeater_150313',
 137             'info_dict': {
 138                 'id': '9CsDKds0kvHI',
 139                 'ext': 'flv',
 140                 'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
 141                 'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
 142             }
 143         }, {
 144             'url': 'https://tw.news.yahoo.com/-100120367.html',
 145             'only_matching': True,
 146         }
 147     ]
 148
 149     def _real_extract(self, url):
 150         mobj = re.match(self._VALID_URL, url)
 151         display_id = mobj.group('display_id') or self._match_id(url)
 152         page_id = mobj.group('id')
 153         url = mobj.group('url')
 154         host = mobj.group('host')
 155         webpage = self._download_webpage(url, display_id)
 156
 157         # Look for iframed media first
 158         iframe_m = re.search(r'<iframe[^>]+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage)
 159         if iframe_m:
 160             iframepage = self._download_webpage(
 161                 host + iframe_m.group(1), display_id, 'Downloading iframe webpage')
 162             items_json = self._search_regex(
 163                 r'mediaItems: (\[.+?\])$', iframepage, 'items', flags=re.MULTILINE, default=None)
 164             if items_json:
 165                 items = json.loads(items_json)
 166                 video_id = items[0]['id']
 167                 return self._get_info(video_id, display_id, webpage)
 168         # Look for NBCSports iframes
 169         nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
 170         if nbc_sports_url:
 171             return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
 172
 173         items_json = self._search_regex(
 174             r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE,
 175             default=None)
 176         if items_json is None:
 177             CONTENT_ID_REGEXES = [
 178                 r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"',
 179                 r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"',
 180                 r'"first_videoid"\s*:\s*"([^"]+)"',
 181                 r'%s[^}]*"ccm_id"\s*:\s*"([^"]+)"' % re.escape(page_id),
 182             ]
 183             video_id = self._search_regex(CONTENT_ID_REGEXES, webpage, 'content ID')
 184         else:
 185             items = json.loads(items_json)
 186             info = items['mediaItems']['query']['results']['mediaObj'][0]
 187             # The 'meta' field is not always in the video webpage, we request it
 188             # from another page
 189             video_id = info['id']
 190         return self._get_info(video_id, display_id, webpage)
 191
 192     def _get_info(self, video_id, display_id, webpage):
 193         region = self._search_regex(
 194             r'\\?"region\\?"\s*:\s*\\?"([^"]+?)\\?"',
 195             webpage, 'region', fatal=False, default='US')
 196         data = compat_urllib_parse.urlencode({
 197             'protocol': 'http',
 198             'region': region,
 199         })
 200         query_url = (
 201             'https://video.media.yql.yahoo.com/v1/video/sapi/streams/'
 202             '{id}?{data}'.format(id=video_id, data=data))
 203         query_result = self._download_json(
 204             query_url, display_id, 'Downloading video info')
 205
 206         info = query_result['query']['results']['mediaObj'][0]
 207         meta = info.get('meta')
 208
 209         if not meta:
 210             msg = info['status'].get('msg')
 211             if msg:
 212                 raise ExtractorError(
 213                     '%s returned error: %s' % (self.IE_NAME, msg), expected=True)
 214             raise ExtractorError('Unable to extract media object meta')
 215
 216         formats = []
 217         for s in info['streams']:
 218             format_info = {
 219                 'width': int_or_none(s.get('width')),
 220                 'height': int_or_none(s.get('height')),
 221                 'tbr': int_or_none(s.get('bitrate')),
 222             }
 223
 224             host = s['host']
 225             path = s['path']
 226             if host.startswith('rtmp'):
 227                 format_info.update({
 228                     'url': host,
 229                     'play_path': path,
 230                     'ext': 'flv',
 231                 })
 232             else:
 233                 format_url = compat_urlparse.urljoin(host, path)
 234                 format_info['url'] = format_url
 235             formats.append(format_info)
 236
 237         self._sort_formats(formats)
 238
 239         return {
 240             'id': video_id,
 241             'display_id': display_id,
 242             'title': unescapeHTML(meta['title']),
 243             'formats': formats,
 244             'description': clean_html(meta['description']),
 245             'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage),
 246             'duration': int_or_none(meta.get('duration')),
 247         }
 248
 249
 250 class YahooSearchIE(SearchInfoExtractor):
 251     IE_DESC = 'Yahoo screen search'
 252     _MAX_RESULTS = 1000
 253     IE_NAME = 'screen.yahoo:search'
 254     _SEARCH_KEY = 'yvsearch'
 255
 256     def _get_n_results(self, query, n):
 257         """Get a specified number of results for a query"""
 258         entries = []
 259         for pagenum in itertools.count(0):
 260             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)
 261             info = self._download_json(result_url, query,
 262                                        note='Downloading results page ' + str(pagenum + 1))
 263             m = info['m']
 264             results = info['results']
 265
 266             for (i, r) in enumerate(results):
 267                 if (pagenum * 30) + i >= n:
 268                     break
 269                 mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r)
 270                 e = self.url_result('http://' + mobj.group('url'), 'Yahoo')
 271                 entries.append(e)
 272             if (pagenum * 30 + i >= n) or (m['last'] >= (m['total'] - 1)):
 273                 break
 274
 275         return {
 276             '_type': 'playlist',
 277             'id': query,
 278             'entries': entries,
 279         }