Merge tag 'upstream/2015.05.15'

[youtubedl] / youtube_dl / extractor / yahoo.py

diff --git a/youtube_dl/extractor/yahoo.py b/youtube_dl/extractor/yahoo.py
index d84be25620eecb944845b74299510067772c583f..bf4e659ac6981c77f7e5f3c77578c4808634d766 100644 (file)
--- a/youtube_dl/extractor/yahoo.py
+++ b/youtube_dl/extractor/yahoo.py
@@ -1,3 +1,4 @@
+# coding: utf-8

 from __future__ import unicode_literals
 
 import itertools
 from __future__ import unicode_literals
 
 import itertools


@@ -5,26 +6,32 @@ import json
 import re
 
 from .common import InfoExtractor, SearchInfoExtractor
 import re
 
 from .common import InfoExtractor, SearchInfoExtractor

-from ..utils import (
+from ..compat import (

     compat_urllib_parse,
     compat_urlparse,
     compat_urllib_parse,
     compat_urlparse,

+)
+from ..utils import (

     clean_html,
     clean_html,

+    unescapeHTML,
+    ExtractorError,

     int_or_none,
 )
 
     int_or_none,
 )
 

+from .nbc import NBCSportsVPlayerIE
+

 
 class YahooIE(InfoExtractor):
     IE_DESC = 'Yahoo screen and movies'
 
 class YahooIE(InfoExtractor):
     IE_DESC = 'Yahoo screen and movies'

-    _VALID_URL = r'https?://(?:screen|movies)\.yahoo\.com/.*?-(?P<id>[0-9]+)(?:-[a-z]+)?\.html'
+    _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)'

     _TESTS = [
         {
             'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
     _TESTS = [
         {
             'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',

-            'md5': '4962b075c08be8690a922ee026d05e69',

             'info_dict': {
                 'id': '2d25e626-2378-391f-ada0-ddaf1417e588',
                 'ext': 'mp4',
                 'title': 'Julian Smith & Travis Legg Watch Julian Smith',
                 'description': 'Julian and Travis watch Julian Smith',
             'info_dict': {
                 'id': '2d25e626-2378-391f-ada0-ddaf1417e588',
                 'ext': 'mp4',
                 'title': 'Julian Smith & Travis Legg Watch Julian Smith',
                 'description': 'Julian and Travis watch Julian Smith',

+                'duration': 6863,

             },
         },
         {
             },
         },
         {


@@ -34,25 +41,134 @@ class YahooIE(InfoExtractor):
                 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9',
                 'ext': 'mp4',
                 'title': 'Codefellas - The Cougar Lies with Spanish Moss',
                 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9',
                 'ext': 'mp4',
                 'title': 'Codefellas - The Cougar Lies with Spanish Moss',

-                '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?',
+                'description': 'md5:66b627ab0a282b26352136ca96ce73c1',
+                'duration': 151,

             },
         },
         {
             },
         },
         {

-            'url': 'https://movies.yahoo.com/video/world-loves-spider-man-190819223.html',
-            'md5': '410b7104aa9893b765bc22787a22f3d9',
+            'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
+            'md5': '60e8ac193d8fb71997caa8fce54c6460',
+            'info_dict': {
+                'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb',
+                'ext': 'mp4',
+                'title': "Yahoo Saves 'Community'",
+                'description': 'md5:4d4145af2fd3de00cbb6c1d664105053',
+                'duration': 170,
+            }
+        },
+        {
+            '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',
+            'md5': '3a09cf59349cfaddae1797acc3c087fc',
+            'info_dict': {
+                'id': 'cac903b3-fcf4-3c14-b632-643ab541712f',
+                'ext': 'mp4',
+                'title': '敢問市長／黃秀霜批賴清德「非常高傲」',
+                'description': '直言台南沒捷運 交通居五都之末',
+                'duration': 396,
+            }
+        },
+        {
+            'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html',
+            'md5': '0b51660361f0e27c9789e7037ef76f4b',
+            'info_dict': {
+                'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58',
+                'ext': 'mp4',
+                'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder',
+                'description': 'md5:f66c890e1490f4910a9953c941dee944',
+                'duration': 97,
+            }
+        },
+        {
+            'url': 'https://ca.sports.yahoo.com/video/program-makes-hockey-more-affordable-013127711.html',
+            'md5': '57e06440778b1828a6079d2f744212c4',
+            'info_dict': {
+                'id': 'c9fa2a36-0d4d-3937-b8f6-cc0fb1881e73',
+                'ext': 'mp4',
+                'title': 'Program that makes hockey more affordable not offered in Manitoba',
+                'description': 'md5:c54a609f4c078d92b74ffb9bf1f496f4',
+                'duration': 121,
+            }
+        }, {
+            'url': 'https://ca.finance.yahoo.com/news/hackers-sony-more-trouble-well-154609075.html',
+            'md5': '226a895aae7e21b0129e2a2006fe9690',

             'info_dict': {
             'info_dict': {

-                'id': '516ed8e2-2c4f-339f-a211-7a8b49d30845',
+                'id': 'e624c4bc-3389-34de-9dfc-025f74943409',

                 'ext': 'mp4',
                 'ext': 'mp4',

-                'title': 'The World Loves Spider-Man',
-                '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.''',
+                'title': '\'The Interview\' TV Spot: War',
+                'description': 'The Interview',
+                'duration': 30,

             }
             }

+        }, {
+            'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
+            'md5': '67010fdf3a08d290e060a4dd96baa07b',
+            'info_dict': {
+                'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521',
+                'ext': 'mp4',
+                'title': 'China Moses Is Crazy About the Blues',
+                'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0',
+                'duration': 128,
+            }
+        }, {
+            'url': 'https://in.lifestyle.yahoo.com/video/connect-dots-dark-side-virgo-090247395.html',
+            'md5': 'd9a083ccf1379127bf25699d67e4791b',
+            'info_dict': {
+                'id': '52aeeaa3-b3d1-30d8-9ef8-5d0cf05efb7c',
+                'ext': 'mp4',
+                'title': 'Connect the Dots: Dark Side of Virgo',
+                'description': 'md5:1428185051cfd1949807ad4ff6d3686a',
+                'duration': 201,
+            }
+        }, {
+            'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html',
+            'md5': '989396ae73d20c6f057746fb226aa215',
+            'info_dict': {
+                'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
+                'ext': 'mp4',
+                'title': '\'True Story\' Trailer',
+                'description': 'True Story',
+                'duration': 150,
+            },
+        }, {
+            'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html',
+            'only_matching': True,
+        }, {
+            'note': 'NBC Sports embeds',
+            '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',
+            'info_dict': {
+                'id': '9CsDKds0kvHI',
+                'ext': 'flv',
+                'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
+                'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
+            }
+        }, {
+            'url': 'https://tw.news.yahoo.com/-100120367.html',
+            'only_matching': True,

         }
     ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         }
     ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

-        video_id = mobj.group('id')
-        webpage = self._download_webpage(url, video_id)
+        display_id = mobj.group('display_id') or self._match_id(url)
+        page_id = mobj.group('id')
+        url = mobj.group('url')
+        host = mobj.group('host')
+        webpage = self._download_webpage(url, display_id)
+
+        # Look for iframed media first
+        iframe_m = re.search(r'<iframe[^>]+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage)
+        if iframe_m:
+            iframepage = self._download_webpage(
+                host + iframe_m.group(1), display_id, 'Downloading iframe webpage')
+            items_json = self._search_regex(
+                r'mediaItems: (\[.+?\])$', iframepage, 'items', flags=re.MULTILINE, default=None)
+            if items_json:
+                items = json.loads(items_json)
+                video_id = items[0]['id']
+                return self._get_info(video_id, display_id, webpage)
+        # Look for NBCSports iframes
+        nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
+        if nbc_sports_url:
+            return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')

 
         items_json = self._search_regex(
             r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE,
 
         items_json = self._search_regex(
             r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE,


@@ -60,32 +176,42 @@ class YahooIE(InfoExtractor):
         if items_json is None:
             CONTENT_ID_REGEXES = [
                 r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"',
         if items_json is None:
             CONTENT_ID_REGEXES = [
                 r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"',

-                r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"'
+                r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"',
+                r'"first_videoid"\s*:\s*"([^"]+)"',
+                r'%s[^}]*"ccm_id"\s*:\s*"([^"]+)"' % re.escape(page_id),

             ]
             ]

-            long_id = self._search_regex(CONTENT_ID_REGEXES, webpage, 'content ID')
-            video_id = long_id
+            video_id = self._search_regex(CONTENT_ID_REGEXES, webpage, 'content ID')

         else:
             items = json.loads(items_json)
             info = items['mediaItems']['query']['results']['mediaObj'][0]
             # The 'meta' field is not always in the video webpage, we request it
             # from another page
         else:
             items = json.loads(items_json)
             info = items['mediaItems']['query']['results']['mediaObj'][0]
             # The 'meta' field is not always in the video webpage, we request it
             # from another page

-            long_id = info['id']
-        return self._get_info(long_id, video_id, webpage)
+            video_id = info['id']
+        return self._get_info(video_id, display_id, webpage)

 
 

-    def _get_info(self, long_id, video_id, webpage):
-        query = ('SELECT * FROM yahoo.media.video.streams WHERE id="%s"'
-                 ' AND plrs="86Gj0vCaSzV_Iuf6hNylf2" AND region="US"'
-                 ' AND protocol="http"' % long_id)
+    def _get_info(self, video_id, display_id, webpage):
+        region = self._search_regex(
+            r'\\?"region\\?"\s*:\s*\\?"([^"]+?)\\?"',
+            webpage, 'region', fatal=False, default='US')

         data = compat_urllib_parse.urlencode({
         data = compat_urllib_parse.urlencode({

-            'q': query,
-            'env': 'prod',
-            'format': 'json',
+            'protocol': 'http',
+            'region': region,

         })
         })

+        query_url = (
+            'https://video.media.yql.yahoo.com/v1/video/sapi/streams/'
+            '{id}?{data}'.format(id=video_id, data=data))

         query_result = self._download_json(
         query_result = self._download_json(

-            'http://video.query.yahoo.com/v1/public/yql?' + data,
-            video_id, 'Downloading video info')
+            query_url, display_id, 'Downloading video info')
+

         info = query_result['query']['results']['mediaObj'][0]
         info = query_result['query']['results']['mediaObj'][0]

-        meta = info['meta']
+        meta = info.get('meta')
+
+        if not meta:
+            msg = info['status'].get('msg')
+            if msg:
+                raise ExtractorError(
+                    '%s returned error: %s' % (self.IE_NAME, msg), expected=True)
+            raise ExtractorError('Unable to extract media object meta')

 
         formats = []
         for s in info['streams']:
 
         formats = []
         for s in info['streams']:


@@ -112,36 +238,15 @@ class YahooIE(InfoExtractor):
 
         return {
             'id': video_id,
 
         return {
             'id': video_id,

-            'title': meta['title'],
+            'display_id': display_id,
+            'title': unescapeHTML(meta['title']),

             'formats': formats,
             'description': clean_html(meta['description']),
             'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage),
             'formats': formats,
             'description': clean_html(meta['description']),
             'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage),

+            'duration': int_or_none(meta.get('duration')),

         }
 
 
         }
 
 

-class YahooNewsIE(YahooIE):
-    IE_NAME = 'yahoo:news'
-    _VALID_URL = r'http://news\.yahoo\.com/video/.*?-(?P<id>\d*?)\.html'
-
-    _TESTS = [{
-        'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
-        'md5': '67010fdf3a08d290e060a4dd96baa07b',
-        'info_dict': {
-            'id': '104538833',
-            'ext': 'mp4',
-            'title': 'China Moses Is Crazy About the Blues',
-            'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0',
-        },
-    }]
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        webpage = self._download_webpage(url, video_id)
-        long_id = self._search_regex(r'contentId: \'(.+?)\',', webpage, 'long id')
-        return self._get_info(long_id, video_id, webpage)
-
-

 class YahooSearchIE(SearchInfoExtractor):
     IE_DESC = 'Yahoo screen search'
     _MAX_RESULTS = 1000
 class YahooSearchIE(SearchInfoExtractor):
     IE_DESC = 'Yahoo screen search'
     _MAX_RESULTS = 1000


@@ -154,7 +259,7 @@ class YahooSearchIE(SearchInfoExtractor):
         for pagenum in itertools.count(0):
             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)
             info = self._download_json(result_url, query,
         for pagenum in itertools.count(0):
             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)
             info = self._download_json(result_url, query,

-                note='Downloading results page '+str(pagenum+1))
+                                       note='Downloading results page ' + str(pagenum + 1))

             m = info['m']
             results = info['results']
 
             m = info['m']
             results = info['results']