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."""
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'id="pnnext" class="pn"'
- _max_google_results = 1000
+ _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://www.google.com/search?tbm=vid&q=%s&start=%s&hl=en' % (compat_urllib_parse.quote_plus(query), pagenum*10)
- print(result_url)
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."""
- _VALID_URL = r'yvsearch(\d+|all)?:[\s\S]+'
-
- _max_yahoo_results = 1000
+ _MAX_RESULTS = 1000
IE_NAME = u'screen.yahoo:search'
-
- 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':
- return self._get_n_results(query, self._max_yahoo_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_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
- return self._get_n_results(query, n)
- except ValueError: # parsing prefix as integer fails
- return self._get_n_results(query, 1)
+ _SEARCH_KEY = 'yvsearch'
def _get_n_results(self, query, n):
"""Get a specified number of results for a query"""
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()
]
projects / youtube-dl / blobdiff