video_info['title'] = playlist_title
return video_info
+class SearchInfoExtractor(InfoExtractor):
+ """
+ Base class for paged search queries extractors.
+ They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query}
+ Instances should define _SEARCH_KEY and _MAX_RESULTS.
+ """
+
+ @classmethod
+ def _make_valid_url(cls):
+ return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY
+
+ @classmethod
+ def suitable(cls, url):
+ return re.match(cls._make_valid_url(), url) is not None
+
+ def _real_extract(self, query):
+ mobj = re.match(self._make_valid_url(), query)
+ if mobj is None:
+ raise ExtractorError(u'Invalid search query "%s"' % query)
+
+ prefix = mobj.group('prefix')
+ query = mobj.group('query')
+ if prefix == '':
+ return self._get_n_results(query, 1)
+ elif prefix == 'all':
+ return self._get_n_results(query, self._MAX_RESULTS)
+ else:
+ n = int(prefix)
+ if n <= 0:
+ raise ExtractorError(u'invalid download number %s for query "%s"' % (n, query))
+ elif n > self._MAX_RESULTS:
+ self._downloader.report_warning(u'%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n))
+ n = self._MAX_RESULTS
+ return self._get_n_results(query, n)
+
+ def _get_n_results(self, query, n):
+ """Get a specified number of results for a query"""
+ raise NotImplementedError("This method must be implemented by sublclasses")
+
class YoutubeIE(InfoExtractor):
"""Information extractor for youtube.com."""
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
video_id = mobj.group('id')
-
- # TODO: Check which url parameters are required
- info_url = 'http://cosmos.bcst.yahoo.com/rest/v2/pops;lmsoverride=1;outputformat=mrss;cb=974419660;id=%s;rd=news.yahoo.com;datacontext=mdb;lg=KCa2IihxG3qE60vQ7HtyUy' % video_id
- webpage = self._download_webpage(info_url, video_id, "Downloading info webpage")
- info_re = r'''<title><!\[CDATA\[(?P<title>.*?)\]\]></title>.*
- <description><!\[CDATA\[(?P<description>.*?)\]\]></description>.*
- <media:pubStart><!\[CDATA\[(?P<date>.*?)\ .*\]\]></media:pubStart>.*
- <media:content\ medium="image"\ url="(?P<thumb>.*?)"\ name="LARGETHUMB"
- '''
- self.report_extraction(video_id)
- m_info = re.search(info_re, webpage, re.VERBOSE|re.DOTALL)
- if m_info is None:
- raise ExtractorError(u'Unable to extract video info')
- video_title = m_info.group('title')
- video_description = m_info.group('description')
- video_thumb = m_info.group('thumb')
- video_date = m_info.group('date')
- video_date = datetime.datetime.strptime(video_date, '%m/%d/%Y').strftime('%Y%m%d')
-
- # TODO: Find a way to get mp4 videos
- rest_url = 'http://cosmos.bcst.yahoo.com/rest/v2/pops;element=stream;outputformat=mrss;id=%s;lmsoverride=1;bw=375;dynamicstream=1;cb=83521105;tech=flv,mp4;rd=news.yahoo.com;datacontext=mdb;lg=KCa2IihxG3qE60vQ7HtyUy' % video_id
- webpage = self._download_webpage(rest_url, video_id, 'Downloading video url webpage')
- m_rest = re.search(r'<media:content url="(?P<url>.*?)" path="(?P<path>.*?)"', webpage)
- if m_rest is None:
- raise ExtractorError(u'Unable to extract video url')
+ webpage = self._download_webpage(url, video_id)
+ m_id = re.search(r'YUI\.namespace\("Media"\)\.CONTENT_ID = "(?P<new_id>.+?)";', webpage)
+
+ if m_id is None:
+ # TODO: Check which url parameters are required
+ info_url = 'http://cosmos.bcst.yahoo.com/rest/v2/pops;lmsoverride=1;outputformat=mrss;cb=974419660;id=%s;rd=news.yahoo.com;datacontext=mdb;lg=KCa2IihxG3qE60vQ7HtyUy' % video_id
+ webpage = self._download_webpage(info_url, video_id, u'Downloading info webpage')
+ info_re = r'''<title><!\[CDATA\[(?P<title>.*?)\]\]></title>.*
+ <description><!\[CDATA\[(?P<description>.*?)\]\]></description>.*
+ <media:pubStart><!\[CDATA\[(?P<date>.*?)\ .*\]\]></media:pubStart>.*
+ <media:content\ medium="image"\ url="(?P<thumb>.*?)"\ name="LARGETHUMB"
+ '''
+ self.report_extraction(video_id)
+ m_info = re.search(info_re, webpage, re.VERBOSE|re.DOTALL)
+ if m_info is None:
+ raise ExtractorError(u'Unable to extract video info')
+ video_title = m_info.group('title')
+ video_description = m_info.group('description')
+ video_thumb = m_info.group('thumb')
+ video_date = m_info.group('date')
+ video_date = datetime.datetime.strptime(video_date, '%m/%d/%Y').strftime('%Y%m%d')
+
+ # TODO: Find a way to get mp4 videos
+ rest_url = 'http://cosmos.bcst.yahoo.com/rest/v2/pops;element=stream;outputformat=mrss;id=%s;lmsoverride=1;bw=375;dynamicstream=1;cb=83521105;tech=flv,mp4;rd=news.yahoo.com;datacontext=mdb;lg=KCa2IihxG3qE60vQ7HtyUy' % video_id
+ webpage = self._download_webpage(rest_url, video_id, u'Downloading video url webpage')
+ m_rest = re.search(r'<media:content url="(?P<url>.*?)" path="(?P<path>.*?)"', webpage)
+ video_url = m_rest.group('url')
+ video_path = m_rest.group('path')
+ if m_rest is None:
+ raise ExtractorError(u'Unable to extract video url')
+
+ else: # We have to use a different method if another id is defined
+ long_id = m_id.group('new_id')
+ info_url = 'http://video.query.yahoo.com/v1/public/yql?q=SELECT%20*%20FROM%20yahoo.media.video.streams%20WHERE%20id%3D%22' + long_id + '%22%20AND%20format%3D%22mp4%2Cflv%22%20AND%20protocol%3D%22rtmp%2Chttp%22%20AND%20plrs%3D%2286Gj0vCaSzV_Iuf6hNylf2%22%20AND%20acctid%3D%22389%22%20AND%20plidl%3D%22%22%20AND%20pspid%3D%22792700001%22%20AND%20offnetwork%3D%22false%22%20AND%20site%3D%22ivy%22%20AND%20lang%3D%22en-US%22%20AND%20region%3D%22US%22%20AND%20override%3D%22none%22%3B&env=prod&format=json&callback=YUI.Env.JSONP.yui_3_8_1_1_1368368376830_335'
+ webpage = self._download_webpage(info_url, video_id, u'Downloading info json')
+ json_str = re.search(r'YUI.Env.JSONP.yui.*?\((.*?)\);', webpage).group(1)
+ info = json.loads(json_str)
+ res = info[u'query'][u'results'][u'mediaObj'][0]
+ stream = res[u'streams'][0]
+ video_path = stream[u'path']
+ video_url = stream[u'host']
+ meta = res[u'meta']
+ video_title = meta[u'title']
+ video_description = meta[u'description']
+ video_thumb = meta[u'thumbnail']
+ video_date = None # I can't find it
info_dict = {
'id': video_id,
- 'url':m_rest.group('url'),
- 'play_path': m_rest.group('path'),
+ 'url': video_url,
+ 'play_path': video_path,
'title':video_title,
'description': video_description,
'thumbnail': video_thumb,
opener.add_handler(handler())
response = opener.open(HeadRequest(url))
+ if response is None:
+ raise ExtractorError(u'Invalid URL protocol')
new_url = response.geturl()
if url == new_url:
}]
-class YoutubeSearchIE(InfoExtractor):
+class YoutubeSearchIE(SearchInfoExtractor):
"""Information Extractor for YouTube search queries."""
- _VALID_URL = r'ytsearch(\d+|all)?:[\s\S]+'
_API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
- _max_youtube_results = 1000
+ _MAX_RESULTS = 1000
IE_NAME = u'youtube:search'
+ _SEARCH_KEY = 'ytsearch'
def report_download_page(self, query, pagenum):
"""Report attempt to download search page with given number."""
query = query.decode(preferredencoding())
self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
- def _real_extract(self, query):
- mobj = re.match(self._VALID_URL, query)
- if mobj is None:
- raise ExtractorError(u'Invalid search query "%s"' % query)
-
- prefix, query = query.split(':')
- prefix = prefix[8:]
- query = query.encode('utf-8')
- if prefix == '':
- return self._get_n_results(query, 1)
- elif prefix == 'all':
- self._get_n_results(query, self._max_youtube_results)
- else:
- try:
- n = int(prefix)
- if n <= 0:
- raise ExtractorError(u'Invalid download number %s for query "%s"' % (n, query))
- elif n > self._max_youtube_results:
- self._downloader.report_warning(u'ytsearch returns max %i results (you requested %i)' % (self._max_youtube_results, n))
- n = self._max_youtube_results
- return self._get_n_results(query, n)
- except ValueError: # parsing prefix as integer fails
- return self._get_n_results(query, 1)
-
def _get_n_results(self, query, n):
"""Get a specified number of results for a query"""
if len(video_ids) > n:
video_ids = video_ids[:n]
videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids]
- return videos
+ return self.playlist_result(videos, query)
-class GoogleSearchIE(InfoExtractor):
+class GoogleSearchIE(SearchInfoExtractor):
"""Information Extractor for Google Video search queries."""
- _VALID_URL = r'gvsearch(?P<prefix>|\d+|all):(?P<query>[\s\S]+)'
- _MORE_PAGES_INDICATOR = r'class="pn" id="pnnext"'
- _max_google_results = 1000
+ _MORE_PAGES_INDICATOR = r'id="pnnext" class="pn"'
+ _MAX_RESULTS = 1000
IE_NAME = u'video.google:search'
-
- def _real_extract(self, query):
- mobj = re.match(self._VALID_URL, query)
-
- prefix = mobj.group('prefix')
- query = mobj.group('query')
- if prefix == '':
- return self._get_n_results(query, 1)
- elif prefix == 'all':
- return self._get_n_results(query, self._max_google_results)
- else:
- n = int(prefix)
- if n <= 0:
- raise ExtractorError(u'invalid download number %s for query "%s"' % (n, query))
- elif n > self._max_google_results:
- self._downloader.report_warning(u'gvsearch returns max %i results (you requested %i)' % (self._max_google_results, n))
- n = self._max_google_results
- return self._get_n_results(query, n)
+ _SEARCH_KEY = 'gvsearch'
def _get_n_results(self, query, n):
"""Get a specified number of results for a query"""
}
for pagenum in itertools.count(1):
- result_url = u'http://video.google.com/videosearch?q=%s&start=%s&hl=en' % (compat_urllib_parse.quote_plus(query), pagenum*10)
+ result_url = u'http://www.google.com/search?tbm=vid&q=%s&start=%s&hl=en' % (compat_urllib_parse.quote_plus(query), pagenum*10)
webpage = self._download_webpage(result_url, u'gvsearch:' + query,
note='Downloading result page ' + str(pagenum))
if (pagenum * 10 > n) or not re.search(self._MORE_PAGES_INDICATOR, webpage):
return res
-class YahooSearchIE(InfoExtractor):
+class YahooSearchIE(SearchInfoExtractor):
"""Information Extractor for Yahoo! Video search queries."""
- _WORKING = False
- _VALID_URL = r'yvsearch(\d+|all)?:[\s\S]+'
- _TEMPLATE_URL = 'http://video.yahoo.com/search/?p=%s&o=%s'
- _VIDEO_INDICATOR = r'href="http://video\.yahoo\.com/watch/([0-9]+/[0-9]+)"'
- _MORE_PAGES_INDICATOR = r'\s*Next'
- _max_yahoo_results = 1000
- IE_NAME = u'video.yahoo:search'
-
- def report_download_page(self, query, pagenum):
- """Report attempt to download playlist page with given number."""
- query = query.decode(preferredencoding())
- self.to_screen(u'query "%s": Downloading page %s' % (query, pagenum))
-
- def _real_extract(self, query):
- mobj = re.match(self._VALID_URL, query)
- if mobj is None:
- raise ExtractorError(u'Invalid search query "%s"' % query)
-
- prefix, query = query.split(':')
- prefix = prefix[8:]
- query = query.encode('utf-8')
- if prefix == '':
- self._download_n_results(query, 1)
- return
- elif prefix == 'all':
- self._download_n_results(query, self._max_yahoo_results)
- return
- else:
- try:
- n = int(prefix)
- if n <= 0:
- raise ExtractorError(u'Invalid download number %s for query "%s"' % (n, query))
- elif n > self._max_yahoo_results:
- self._downloader.report_warning(u'yvsearch returns max %i results (you requested %i)' % (self._max_yahoo_results, n))
- n = self._max_yahoo_results
- self._download_n_results(query, n)
- return
- except ValueError: # parsing prefix as integer fails
- self._download_n_results(query, 1)
- return
-
- def _download_n_results(self, query, n):
- """Downloads a specified number of results for a query"""
-
- video_ids = []
- already_seen = set()
- pagenum = 1
+ _MAX_RESULTS = 1000
+ IE_NAME = u'screen.yahoo:search'
+ _SEARCH_KEY = 'yvsearch'
- while True:
- self.report_download_page(query, pagenum)
- result_url = self._TEMPLATE_URL % (compat_urllib_parse.quote_plus(query), pagenum)
- request = compat_urllib_request.Request(result_url)
- try:
- page = compat_urllib_request.urlopen(request).read()
- except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
- raise ExtractorError(u'Unable to download webpage: %s' % compat_str(err))
+ def _get_n_results(self, query, n):
+ """Get a specified number of results for a query"""
- # Extract video identifiers
- for mobj in re.finditer(self._VIDEO_INDICATOR, page):
- video_id = mobj.group(1)
- if video_id not in already_seen:
- video_ids.append(video_id)
- already_seen.add(video_id)
- if len(video_ids) == n:
- # Specified n videos reached
- for id in video_ids:
- self._downloader.download(['http://video.yahoo.com/watch/%s' % id])
- return
-
- if re.search(self._MORE_PAGES_INDICATOR, page) is None:
- for id in video_ids:
- self._downloader.download(['http://video.yahoo.com/watch/%s' % id])
- return
+ res = {
+ '_type': 'playlist',
+ 'id': query,
+ 'entries': []
+ }
+ for pagenum in itertools.count(0):
+ result_url = u'http://video.search.yahoo.com/search/?p=%s&fr=screen&o=js&gs=0&b=%d' % (compat_urllib_parse.quote_plus(query), pagenum * 30)
+ webpage = self._download_webpage(result_url, query,
+ note='Downloading results page '+str(pagenum+1))
+ info = json.loads(webpage)
+ m = info[u'm']
+ results = info[u'results']
+
+ for (i, r) in enumerate(results):
+ if (pagenum * 30) +i >= n:
+ break
+ mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r)
+ e = self.url_result('http://' + mobj.group('url'), 'Yahoo')
+ res['entries'].append(e)
+ if (pagenum * 30 +i >= n) or (m[u'last'] >= (m[u'total'] -1 )):
+ break
- pagenum = pagenum + 1
+ return res
class YoutubePlaylistIE(InfoExtractor):
video_id = m.group('videoID')
video_url = u'http://tcdn.ustream.tv/video/%s' % video_id
webpage = self._download_webpage(url, video_id)
- m = re.search(r'data-title="(?P<title>.+)"',webpage)
- title = m.group('title')
- m = re.search(r'<a class="state" data-content-type="channel" data-content-id="(?P<uploader>\d+)"',webpage)
- uploader = m.group('uploader')
+ self.report_extraction(video_id)
+ try:
+ m = re.search(r'data-title="(?P<title>.+)"',webpage)
+ title = m.group('title')
+ m = re.search(r'data-content-type="channel".*?>(?P<uploader>.*?)</a>',
+ webpage, re.DOTALL)
+ uploader = unescapeHTML(m.group('uploader').strip())
+ m = re.search(r'<link rel="image_src" href="(?P<thumb>.*?)"', webpage)
+ thumb = m.group('thumb')
+ except AttributeError:
+ raise ExtractorError(u'Unable to extract info')
info = {
'id':video_id,
'url':video_url,
'ext': 'flv',
'title': title,
- 'uploader': uploader
+ 'uploader': uploader,
+ 'thumbnail': thumb,
}
- return [info]
+ return info
class WorldStarHipHopIE(InfoExtractor):
_VALID_URL = r'https?://(?:www|m)\.worldstar(?:candy|hiphop)\.com/videos/video\.php\?v=(?P<id>.*)'
class InaIE(InfoExtractor):
"""Information Extractor for Ina.fr"""
- _VALID_URL = r'(?:http://)?(?:www.)?ina\.fr/video/(?P<id>I[0-9]+)/.*'
+ _VALID_URL = r'(?:http://)?(?:www\.)?ina\.fr/video/(?P<id>I[0-9]+)/.*'
def _real_extract(self,url):
mobj = re.match(self._VALID_URL, url)
'title': video_title,
}]
+class HowcastIE(InfoExtractor):
+ """Information Extractor for Ina.fr"""
+ _VALID_URL = r'(?:https?://)?(?:www\.)?howcast\.com/videos/(?P<id>[\d]+)'
+
+ def _real_extract(self, url):
+ mobj = re.match(self._VALID_URL, url)
+
+ video_id = mobj.group('id')
+ webpage_url = 'http://www.howcast.com/videos/' + video_id
+ webpage = self._download_webpage(webpage_url, video_id)
+
+ mobj = re.search(r'\'file\': "(http://mobile-media\.howcast\.com/\d+\.mp4)"', webpage)
+ if mobj is None:
+ raise ExtractorError(u'Unable to extract video URL')
+ video_url = mobj.group(1)
+
+ mobj = re.search(r'<meta content=(?:"([^"]+)"|\'([^\']+)\') property=\'og:title\'', webpage)
+ if mobj is None:
+ raise ExtractorError(u'Unable to extract title')
+ video_title = mobj.group(1) or mobj.group(2)
+
+ mobj = re.search(r'<meta content=(?:"([^"]+)"|\'([^\']+)\') name=\'description\'', webpage)
+ if mobj is None:
+ self._downloader.report_warning(u'unable to extract description')
+ video_description = None
+ else:
+ video_description = mobj.group(1) or mobj.group(2)
+
+ return [{
+ 'id': video_id,
+ 'url': video_url,
+ 'ext': 'mp4',
+ 'title': video_title,
+ 'description': video_description,
+ }]
+
def gen_extractors():
""" Return a list of an instance of every supported extractor.
The order does matter; the first extractor matched is the one handling the URL.
BandcampIE(),
RedTubeIE(),
InaIE(),
+ HowcastIE(),
GenericIE()
]