From befd88b786dc41ff075693fd17bafbc7fa4c100e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?= Date: Fri, 29 Nov 2013 15:25:43 +0100 Subject: [yahoo] Add an extractor for yahoo news (closes #1849) --- youtube_dl/extractor/yahoo.py | 34 +++++++++++++++++++++++++++++++++- 1 file changed, 33 insertions(+), 1 deletion(-) (limited to 'youtube_dl/extractor/yahoo.py') diff --git a/youtube_dl/extractor/yahoo.py b/youtube_dl/extractor/yahoo.py index 617e3bb06..2d87e81b2 100644 --- a/youtube_dl/extractor/yahoo.py +++ b/youtube_dl/extractor/yahoo.py @@ -53,8 +53,11 @@ class YahooIE(InfoExtractor): # The 'meta' field is not always in the video webpage, we request it # from another page long_id = info['id'] + return self._get_info(info['id'], video_id) + + def _get_info(self, long_id, video_id): query = ('SELECT * FROM yahoo.media.video.streams WHERE id="%s"' - ' AND plrs="86Gj0vCaSzV_Iuf6hNylf2"' % long_id) + ' AND plrs="86Gj0vCaSzV_Iuf6hNylf2" AND region="US"' % long_id) data = compat_urllib_parse.urlencode({ 'q': query, 'env': 'prod', @@ -100,6 +103,35 @@ class YahooIE(InfoExtractor): } +class YahooNewsIE(YahooIE): + IE_NAME = 'yahoo:news' + _VALID_URL = r'http://news\.yahoo\.com/video/.*?-(?P\d*?)\.html' + + _TEST = { + u'url': u'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html', + u'info_dict': { + u'id': u'104538833', + u'ext': u'flv', + u'title': u'China Moses Is Crazy About the Blues', + u'description': u'md5:9900ab8cd5808175c7b3fe55b979bed0', + }, + u'params': { + # Requires rtmpdump + u'skip_download': True, + }, + } + + # Overwrite YahooIE properties we don't want + _TESTS = [] + + 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, u'long id') + return self._get_info(long_id, video_id) + + class YahooSearchIE(SearchInfoExtractor): IE_DESC = u'Yahoo screen search' _MAX_RESULTS = 1000 -- cgit v1.2.3 From c5171c454b4392f7276b7f9e94c25d7f1ad60375 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?= Date: Fri, 29 Nov 2013 22:06:17 +0100 Subject: [yahoo] Force use of the http protocol for downloading the videos. --- youtube_dl/extractor/yahoo.py | 24 ++++++++---------------- 1 file changed, 8 insertions(+), 16 deletions(-) (limited to 'youtube_dl/extractor/yahoo.py') diff --git a/youtube_dl/extractor/yahoo.py b/youtube_dl/extractor/yahoo.py index 2d87e81b2..e457c4707 100644 --- a/youtube_dl/extractor/yahoo.py +++ b/youtube_dl/extractor/yahoo.py @@ -17,27 +17,21 @@ class YahooIE(InfoExtractor): _TESTS = [ { u'url': u'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html', - u'file': u'214727115.flv', + u'file': u'214727115.mp4', + u'md5': u'4962b075c08be8690a922ee026d05e69', u'info_dict': { u'title': u'Julian Smith & Travis Legg Watch Julian Smith', u'description': u'Julian and Travis watch Julian Smith', }, - u'params': { - # Requires rtmpdump - u'skip_download': True, - }, }, { u'url': u'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html', - u'file': u'103000935.flv', + u'file': u'103000935.mp4', + u'md5': u'd6e6fc6e1313c608f316ddad7b82b306', u'info_dict': { u'title': u'Codefellas - The Cougar Lies with Spanish Moss', u'description': u'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?', }, - u'params': { - # Requires rtmpdump - u'skip_download': True, - }, }, ] @@ -57,7 +51,8 @@ class YahooIE(InfoExtractor): def _get_info(self, long_id, video_id): query = ('SELECT * FROM yahoo.media.video.streams WHERE id="%s"' - ' AND plrs="86Gj0vCaSzV_Iuf6hNylf2" AND region="US"' % long_id) + ' AND plrs="86Gj0vCaSzV_Iuf6hNylf2" AND region="US"' + ' AND protocol="http"' % long_id) data = compat_urllib_parse.urlencode({ 'q': query, 'env': 'prod', @@ -109,16 +104,13 @@ class YahooNewsIE(YahooIE): _TEST = { u'url': u'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html', + u'md5': u'67010fdf3a08d290e060a4dd96baa07b', u'info_dict': { u'id': u'104538833', - u'ext': u'flv', + u'ext': u'mp4', u'title': u'China Moses Is Crazy About the Blues', u'description': u'md5:9900ab8cd5808175c7b3fe55b979bed0', }, - u'params': { - # Requires rtmpdump - u'skip_download': True, - }, } # Overwrite YahooIE properties we don't want -- cgit v1.2.3