X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/28d5370e4a3ed87a622c07b81844b14420e5bcba..478c6cb48296d1059eebc41f68d770254539f9a4:/youtube_dl/InfoExtractors.py diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py index 4bcd8b9..17e0f83 100755 --- a/youtube_dl/InfoExtractors.py +++ b/youtube_dl/InfoExtractors.py @@ -16,6 +16,9 @@ import xml.etree.ElementTree import random import math import operator +import hashlib +import binascii +import urllib from .utils import * @@ -188,6 +191,86 @@ class InfoExtractor(object): video_info['title'] = playlist_title return video_info + def _search_regex(self, pattern, string, name, default=None, fatal=True, flags=0): + """ + Perform a regex search on the given string, using a single or a list of + patterns returning the first matching group. + In case of failure return a default value or raise a WARNING or a + ExtractorError, depending on fatal, specifying the field name. + """ + if isinstance(pattern, (str, compat_str, compiled_regex_type)): + mobj = re.search(pattern, string, flags) + else: + for p in pattern: + mobj = re.search(p, string, flags) + if mobj: break + + if sys.stderr.isatty() and os.name != 'nt': + _name = u'\033[0;34m%s\033[0m' % name + else: + _name = name + + if mobj: + # return the first matching group + return next(g for g in mobj.groups() if g is not None) + elif default is not None: + return default + elif fatal: + raise ExtractorError(u'Unable to extract %s' % _name) + else: + self._downloader.report_warning(u'unable to extract %s; ' + u'please report this issue on GitHub.' % _name) + return None + + def _html_search_regex(self, pattern, string, name, default=None, fatal=True, flags=0): + """ + Like _search_regex, but strips HTML tags and unescapes entities. + """ + res = self._search_regex(pattern, string, name, default, fatal, flags) + if res: + return clean_html(res).strip() + else: + return res + +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|[1-9][0-9]*|all):(?P[\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.""" @@ -334,6 +417,34 @@ class YoutubeIE(InfoExtractor): return (u'Did not fetch video subtitles', None, None) return (None, sub_lang, sub) + def _request_automatic_caption(self, video_id, webpage): + """We need the webpage for getting the captions url, pass it as an + argument to speed up the process.""" + sub_lang = self._downloader.params.get('subtitleslang') + sub_format = self._downloader.params.get('subtitlesformat') + self.to_screen(u'%s: Looking for automatic captions' % video_id) + mobj = re.search(r';ytplayer.config = ({.*?});', webpage) + err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang + if mobj is None: + return [(err_msg, None, None)] + player_config = json.loads(mobj.group(1)) + try: + args = player_config[u'args'] + caption_url = args[u'ttsurl'] + timestamp = args[u'timestamp'] + params = compat_urllib_parse.urlencode({ + 'lang': 'en', + 'tlang': sub_lang, + 'fmt': sub_format, + 'ts': timestamp, + 'kind': 'asr', + }) + subtitles_url = caption_url + '&' + params + sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions') + return [(None, sub_lang, sub)] + except KeyError: + return [(err_msg, None, None)] + def _extract_subtitle(self, video_id): """ Return a list with a tuple: @@ -581,7 +692,14 @@ class YoutubeIE(InfoExtractor): if video_subtitles: (sub_error, sub_lang, sub) = video_subtitles[0] if sub_error: - self._downloader.report_error(sub_error) + # We try with the automatic captions + video_subtitles = self._request_automatic_caption(video_id, video_webpage) + (sub_error_auto, sub_lang, sub) = video_subtitles[0] + if sub is not None: + pass + else: + # We report the original error + self._downloader.report_error(sub_error) if self._downloader.params.get('allsubtitles', False): video_subtitles = self._extract_all_subtitles(video_id) @@ -822,16 +940,10 @@ class DailymotionIE(InfoExtractor): video_title = unescapeHTML(mobj.group('title')) video_uploader = None - mobj = re.search(r'(?im)[^<]+?]+?>([^<]+?)', webpage) - if mobj is None: - # lookin for official user - mobj_official = re.search(r'', webpage) - if mobj_official is None: - self._downloader.report_warning(u'unable to extract uploader nickname') - else: - video_uploader = mobj_official.group(1) - else: - video_uploader = mobj.group(1) + video_uploader = self._search_regex([r'(?im)[^<]+?]+?>([^<]+?)', + # Looking for official user + r'<(?:span|a) .*?rel="author".*?>([^<]+?)([0-9]{2})-([0-9]{2})-([0-9]{4})', webpage) @@ -887,18 +999,13 @@ class PhotobucketIE(InfoExtractor): }] # We try looking in other parts of the webpage - mobj = re.search(r'', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract media URL') - mediaURL = compat_urllib_parse.unquote(mobj.group(1)) - - video_url = mediaURL + video_url = self._search_regex(r'', + webpage, u'video URL') mobj = re.search(r'(.*) video by (.*) - Photobucket', webpage) if mobj is None: raise ExtractorError(u'Unable to extract title') video_title = mobj.group(1).decode('utf-8') - video_uploader = mobj.group(2).decode('utf-8') return [{ @@ -983,7 +1090,7 @@ class VimeoIE(InfoExtractor): """Information extractor for vimeo.com.""" # _VALID_URL matches Vimeo URLs - _VALID_URL = r'(?Phttps?://)?(?:(?:www|player)\.)?vimeo\.com/(?:(?:groups|album)/[^/]+/)?(?Pplay_redirect_hls\?clip_id=)?(?:videos?/)?(?P[0-9]+)' + _VALID_URL = r'(?Phttps?://)?(?:(?:www|player)\.)?vimeo(?Ppro)?\.com/(?:(?:(?:groups|album)/[^/]+)|(?:.*?)/)?(?Pplay_redirect_hls\?clip_id=)?(?:videos?/)?(?P[0-9]+)' IE_NAME = u'vimeo' def _real_extract(self, url, new_video=True): @@ -995,7 +1102,7 @@ class VimeoIE(InfoExtractor): video_id = mobj.group('id') if not mobj.group('proto'): url = 'https://' + url - if mobj.group('direct_link'): + if mobj.group('direct_link') or mobj.group('pro'): url = 'https://vimeo.com/' + video_id # Retrieve video webpage to extract further information @@ -1022,7 +1129,7 @@ class VimeoIE(InfoExtractor): # Extract uploader and uploader_id video_uploader = config["video"]["owner"]["name"] - video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] + video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None # Extract video thumbnail video_thumbnail = config["video"]["thumbnail"] @@ -1265,6 +1372,8 @@ class GenericIE(InfoExtractor): 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: @@ -1294,6 +1403,9 @@ class GenericIE(InfoExtractor): if mobj is None: # Broaden the search a little bit: JWPlayer JS loader mobj = re.search(r'[^A-Za-z0-9]?file:\s*["\'](http[^\'"&]*)', webpage) + if mobj is None: + # Try to find twitter cards info + mobj = re.search(r'(.*)', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = mobj.group(1) + video_title = self._html_search_regex(r'(.*)', + webpage, u'video title') # video uploader is domain name - mobj = re.match(r'(?:https?://)?([^/]*)/.*', url) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_uploader = mobj.group(1) + video_uploader = self._search_regex(r'(?:https?://)?([^/]*)/.*', + url, u'video uploader') return [{ 'id': video_id, @@ -1336,42 +1444,17 @@ class GenericIE(InfoExtractor): }] -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""" @@ -1404,30 +1487,12 @@ class YoutubeSearchIE(InfoExtractor): return self.playlist_result(videos, query) -class GoogleSearchIE(InfoExtractor): +class GoogleSearchIE(SearchInfoExtractor): """Information Extractor for Google Video search queries.""" - _VALID_URL = r'gvsearch(?P|\d+|all):(?P[\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""" @@ -1453,37 +1518,12 @@ class GoogleSearchIE(InfoExtractor): 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""" @@ -1530,7 +1570,7 @@ class YoutubePlaylistIE(InfoExtractor): | ((?:PL|EC|UU)[0-9A-Za-z-_]{10,}) )""" - _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json' + _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none' _MAX_RESULTS = 50 IE_NAME = u'youtube:playlist' @@ -1791,10 +1831,7 @@ class DepositFilesIE(InfoExtractor): file_extension = os.path.splitext(file_url)[1][1:] # Search for file title - mobj = re.search(r'', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - file_title = mobj.group(1).decode('utf-8') + file_title = self._search_regex(r'', webpage, u'title') return [{ 'id': file_id.decode('utf-8'), @@ -1888,10 +1925,8 @@ class FacebookIE(InfoExtractor): video_duration = int(video_data['video_duration']) thumbnail = video_data['thumbnail_src'] - m = re.search('

([^<]+)

', webpage) - if not m: - raise ExtractorError(u'Cannot find title in webpage') - video_title = unescapeHTML(m.group(1)) + video_title = self._html_search_regex('

([^<]+)

', + webpage, u'title') info = { 'id': video_id, @@ -1907,7 +1942,7 @@ class FacebookIE(InfoExtractor): class BlipTVIE(InfoExtractor): """Information extractor for blip.tv""" - _VALID_URL = r'^(?:https?://)?(?:\w+\.)?blip\.tv(/.+)$' + _VALID_URL = r'^(?:https?://)?(?:\w+\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(.+)$' _URL_EXT = r'^.*\.([a-z0-9]+)$' IE_NAME = u'blip.tv' @@ -1920,6 +1955,10 @@ class BlipTVIE(InfoExtractor): if mobj is None: raise ExtractorError(u'Invalid URL: %s' % url) + # See https://github.com/rg3/youtube-dl/issues/857 + api_mobj = re.match(r'http://a\.blip\.tv/api\.swf#(?P[\d\w]+)', url) + if api_mobj is not None: + url = 'http://blip.tv/play/g_%s' % api_mobj.group('video_id') urlp = compat_urllib_parse_urlparse(url) if urlp.path.startswith('/play/'): request = compat_urllib_request.Request(url) @@ -2005,39 +2044,150 @@ class MyVideoIE(InfoExtractor): _VALID_URL = r'(?:http://)?(?:www\.)?myvideo\.de/watch/([0-9]+)/([^?/]+).*' IE_NAME = u'myvideo' + # Original Code from: https://github.com/dersphere/plugin.video.myvideo_de.git + # Released into the Public Domain by Tristan Fischer on 2013-05-19 + # https://github.com/rg3/youtube-dl/pull/842 + def __rc4crypt(self,data, key): + x = 0 + box = list(range(256)) + for i in list(range(256)): + x = (x + box[i] + compat_ord(key[i % len(key)])) % 256 + box[i], box[x] = box[x], box[i] + x = 0 + y = 0 + out = '' + for char in data: + x = (x + 1) % 256 + y = (y + box[x]) % 256 + box[x], box[y] = box[y], box[x] + out += chr(compat_ord(char) ^ box[(box[x] + box[y]) % 256]) + return out + + def __md5(self,s): + return hashlib.md5(s).hexdigest().encode() + def _real_extract(self,url): mobj = re.match(self._VALID_URL, url) if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) + raise ExtractorError(u'invalid URL: %s' % url) video_id = mobj.group(1) + GK = ( + b'WXpnME1EZGhNRGhpTTJNM01XVmhOREU0WldNNVpHTTJOakpt' + b'TW1FMU5tVTBNR05pWkRaa05XRXhNVFJoWVRVd1ptSXhaVEV3' + b'TnpsbA0KTVRkbU1tSTRNdz09' + ) + # Get video webpage webpage_url = 'http://www.myvideo.de/watch/%s' % video_id webpage = self._download_webpage(webpage_url, video_id) - self.report_extraction(video_id) - mobj = re.search(r'([^<]+)', + webpage, u'title') + + video_ext = self._search_regex('[.](.+?)$', video_url, u'extension') + + return [{ + 'id': video_id, + 'url': video_url, + 'uploader': None, + 'upload_date': None, + 'title': video_title, + 'ext': u'flv', + }] - mobj = re.search('([^<]+)', webpage) + # try encxml + mobj = re.search('var flashvars={(.+?)}', webpage) if mobj is None: - raise ExtractorError(u'Unable to extract title') + raise ExtractorError(u'Unable to extract video') + + params = {} + encxml = '' + sec = mobj.group(1) + for (a, b) in re.findall('(.+?):\'(.+?)\',?', sec): + if not a == '_encxml': + params[a] = b + else: + encxml = compat_urllib_parse.unquote(b) + if not params.get('domain'): + params['domain'] = 'www.myvideo.de' + xmldata_url = '%s?%s' % (encxml, compat_urllib_parse.urlencode(params)) + if 'flash_playertype=MTV' in xmldata_url: + self._downloader.report_warning(u'avoiding MTV player') + xmldata_url = ( + 'http://www.myvideo.de/dynamic/get_player_video_xml.php' + '?flash_playertype=D&ID=%s&_countlimit=4&autorun=yes' + ) % video_id + + # get enc data + enc_data = self._download_webpage(xmldata_url, video_id).split('=')[1] + enc_data_b = binascii.unhexlify(enc_data) + sk = self.__md5( + base64.b64decode(base64.b64decode(GK)) + + self.__md5( + str(video_id).encode('utf-8') + ) + ) + dec_data = self.__rc4crypt(enc_data_b, sk) + + # extracting infos + self.report_extraction(video_id) - video_title = mobj.group(1) + video_url = None + mobj = re.search('connectionurl=\'(.*?)\'', dec_data) + if mobj: + video_url = compat_urllib_parse.unquote(mobj.group(1)) + if 'myvideo2flash' in video_url: + self._downloader.report_warning(u'forcing RTMPT ...') + video_url = video_url.replace('rtmpe://', 'rtmpt://') + + if not video_url: + # extract non rtmp videos + mobj = re.search('path=\'(http.*?)\' source=\'(.*?)\'', dec_data) + if mobj is None: + raise ExtractorError(u'unable to extract url') + video_url = compat_urllib_parse.unquote(mobj.group(1)) + compat_urllib_parse.unquote(mobj.group(2)) + + video_file = self._search_regex('source=\'(.*?)\'', dec_data, u'video file') + video_file = compat_urllib_parse.unquote(video_file) + + if not video_file.endswith('f4m'): + ppath, prefix = video_file.split('.') + video_playpath = '%s:%s' % (prefix, ppath) + video_hls_playlist = '' + else: + video_playpath = '' + video_hls_playlist = ( + video_filepath + video_file + ).replace('.f4m', '.m3u8') + + video_swfobj = self._search_regex('swfobject.embedSWF\(\'(.+?)\'', webpage, u'swfobj') + video_swfobj = compat_urllib_parse.unquote(video_swfobj) + + video_title = self._html_search_regex("(.*?)", + webpage, u'title') return [{ - 'id': video_id, - 'url': video_url, - 'uploader': None, - 'upload_date': None, - 'title': video_title, - 'ext': u'flv', + 'id': video_id, + 'url': video_url, + 'tc_url': video_url, + 'uploader': None, + 'upload_date': None, + 'title': video_title, + 'ext': u'flv', + 'play_path': video_playpath, + 'video_file': video_file, + 'video_hls_playlist': video_hls_playlist, + 'player_url': video_swfobj, }] + class ComedyCentralIE(InfoExtractor): """Information extractor for The Daily Show and Colbert Report """ @@ -2219,19 +2369,25 @@ class EscapistIE(InfoExtractor): showName = mobj.group('showname') videoId = mobj.group('episode') - self.report_extraction(showName) - webPage = self._download_webpage(url, showName) + self.report_extraction(videoId) + webpage = self._download_webpage(url, videoId) + + videoDesc = self._html_search_regex('(.*?)\s+-\s+XVID', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract video title') - video_title = mobj.group(1) - + video_title = self._html_search_regex(r'(.*?)\s+-\s+XVID', + webpage, u'title') # Extract video thumbnail - mobj = re.search(r'http://(?:img.*?\.)xvideos.com/videos/thumbs/[a-fA-F0-9]+/[a-fA-F0-9]+/[a-fA-F0-9]+/[a-fA-F0-9]+/([a-fA-F0-9.]+jpg)', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract video thumbnail') - video_thumbnail = mobj.group(0) + video_thumbnail = self._search_regex(r'http://(?:img.*?\.)xvideos.com/videos/thumbs/[a-fA-F0-9]+/[a-fA-F0-9]+/[a-fA-F0-9]+/[a-fA-F0-9]+/([a-fA-F0-9.]+jpg)', + webpage, u'thumbnail', fatal=False) info = { 'id': video_id, @@ -2515,16 +2662,12 @@ class InfoQIE(InfoExtractor): video_url = 'rtmpe://video.infoq.com/cfx/st/' + real_id # Extract title - mobj = re.search(r'contentTitle = "(.*?)";', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract video title') - video_title = mobj.group(1) + video_title = self._search_regex(r'contentTitle = "(.*?)";', + webpage, u'title') # Extract description - video_description = u'No description available.' - mobj = re.search(r'<meta name="description" content="(.*)"(?:\s*/)?>', webpage) - if mobj is not None: - video_description = mobj.group(1) + video_description = self._html_search_regex(r'<meta name="description" content="(.*)"(?:\s*/)?>', + webpage, u'description', fatal=False) video_filename = video_url.split('/')[-1] video_id, extension = video_filename.split('.') @@ -2695,15 +2838,10 @@ class StanfordOpenClassroomIE(InfoExtractor): note='Downloading course info page', errnote='Unable to download course info page') - m = re.search('<h1>([^<]+)</h1>', coursepage) - if m: - info['title'] = unescapeHTML(m.group(1)) - else: - info['title'] = info['id'] + info['title'] = self._html_search_regex('<h1>([^<]+)</h1>', coursepage, 'title', default=info['id']) - m = re.search('<description>([^<]+)</description>', coursepage) - if m: - info['description'] = unescapeHTML(m.group(1)) + info['description'] = self._html_search_regex('<description>([^<]+)</description>', + coursepage, u'description', fatal=False) links = orderedSet(re.findall('<a href="(VideoPage.php\?[^"]+)">', coursepage)) info['list'] = [ @@ -2764,25 +2902,17 @@ class MTVIE(InfoExtractor): webpage = self._download_webpage(url, video_id) - mobj = re.search(r'<meta name="mtv_vt" content="([^"]+)"/>', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract song name') - song_name = unescapeHTML(mobj.group(1).decode('iso-8859-1')) - mobj = re.search(r'<meta name="mtv_an" content="([^"]+)"/>', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract performer') - performer = unescapeHTML(mobj.group(1).decode('iso-8859-1')) - video_title = performer + ' - ' + song_name + song_name = self._html_search_regex(r'<meta name="mtv_vt" content="([^"]+)"/>', + webpage, u'song name', fatal=False) - mobj = re.search(r'<meta name="mtvn_uri" content="([^"]+)"/>', webpage) - if mobj is None: - raise ExtractorError(u'Unable to mtvn_uri') - mtvn_uri = mobj.group(1) + video_title = self._html_search_regex(r'<meta name="mtv_an" content="([^"]+)"/>', + webpage, u'title') - mobj = re.search(r'MTVN.Player.defaultPlaylistId = ([0-9]+);', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract content id') - content_id = mobj.group(1) + mtvn_uri = self._html_search_regex(r'<meta name="mtvn_uri" content="([^"]+)"/>', + webpage, u'mtvn_uri', fatal=False) + + content_id = self._search_regex(r'MTVN.Player.defaultPlaylistId = ([0-9]+);', + webpage, u'content id', fatal=False) videogen_url = 'http://www.mtv.com/player/includes/mediaGen.jhtml?uri=' + mtvn_uri + '&id=' + content_id + '&vid=' + video_id + '&ref=www.mtvn.com&viewUri=' + mtvn_uri self.report_extraction(video_id) @@ -2930,20 +3060,15 @@ class XNXXIE(InfoExtractor): # Get webpage content webpage = self._download_webpage(url, video_id) - result = re.search(self.VIDEO_URL_RE, webpage) - if result is None: - raise ExtractorError(u'Unable to extract video url') - video_url = compat_urllib_parse.unquote(result.group(1)) + video_url = self._search_regex(self.VIDEO_URL_RE, + webpage, u'video URL') + video_url = compat_urllib_parse.unquote(video_url) - result = re.search(self.VIDEO_TITLE_RE, webpage) - if result is None: - raise ExtractorError(u'Unable to extract video title') - video_title = result.group(1) + video_title = self._html_search_regex(self.VIDEO_TITLE_RE, + webpage, u'title') - result = re.search(self.VIDEO_THUMB_RE, webpage) - if result is None: - raise ExtractorError(u'Unable to extract video thumbnail') - video_thumbnail = result.group(1) + video_thumbnail = self._search_regex(self.VIDEO_THUMB_RE, + webpage, u'thumbnail', fatal=False) return [{ 'id': video_id, @@ -2963,26 +3088,6 @@ class GooglePlusIE(InfoExtractor): _VALID_URL = r'(?:https://)?plus\.google\.com/(?:[^/]+/)*?posts/(\w+)' IE_NAME = u'plus.google' - def report_extract_entry(self, url): - """Report downloading extry""" - self.to_screen(u'Downloading entry: %s' % url) - - def report_date(self, upload_date): - """Report downloading extry""" - self.to_screen(u'Entry date: %s' % upload_date) - - def report_uploader(self, uploader): - """Report downloading extry""" - self.to_screen(u'Uploader: %s' % uploader) - - def report_title(self, video_title): - """Report downloading extry""" - self.to_screen(u'Title: %s' % video_title) - - def report_extract_vid_page(self, video_page): - """Report information extraction.""" - self.to_screen(u'Extracting video page: %s' % video_page) - def _real_extract(self, url): # Extract id from URL mobj = re.match(self._VALID_URL, url) @@ -2995,47 +3100,31 @@ class GooglePlusIE(InfoExtractor): video_extension = 'flv' # Step 1, Retrieve post webpage to extract further information - self.report_extract_entry(post_url) webpage = self._download_webpage(post_url, video_id, u'Downloading entry webpage') + self.report_extraction(video_id) + # Extract update date - upload_date = None - pattern = 'title="Timestamp">(.*?)</a>' - mobj = re.search(pattern, webpage) - if mobj: - upload_date = mobj.group(1) + upload_date = self._html_search_regex('title="Timestamp">(.*?)</a>', + webpage, u'upload date', fatal=False) + if upload_date: # Convert timestring to a format suitable for filename upload_date = datetime.datetime.strptime(upload_date, "%Y-%m-%d") upload_date = upload_date.strftime('%Y%m%d') - self.report_date(upload_date) # Extract uploader - uploader = None - pattern = r'rel\="author".*?>(.*?)</a>' - mobj = re.search(pattern, webpage) - if mobj: - uploader = mobj.group(1) - self.report_uploader(uploader) + uploader = self._html_search_regex(r'rel\="author".*?>(.*?)</a>', + webpage, u'uploader', fatal=False) # Extract title # Get the first line for title - video_title = u'NA' - pattern = r'<meta name\=\"Description\" content\=\"(.*?)[\n<"]' - mobj = re.search(pattern, webpage) - if mobj: - video_title = mobj.group(1) - self.report_title(video_title) + video_title = self._html_search_regex(r'<meta name\=\"Description\" content\=\"(.*?)[\n<"]', + webpage, 'title', default=u'NA') # Step 2, Stimulate clicking the image box to launch video - pattern = '"(https\://plus\.google\.com/photos/.*?)",,"image/jpeg","video"\]' - mobj = re.search(pattern, webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract video page URL') - - video_page = mobj.group(1) + video_page = self._search_regex('"(https\://plus\.google\.com/photos/.*?)",,"image/jpeg","video"\]', + webpage, u'video page URL') webpage = self._download_webpage(video_page, video_id, u'Downloading video page') - self.report_extract_vid_page(video_page) - # Extract video links on video page """Extract video links of all sizes""" @@ -3068,7 +3157,7 @@ class GooglePlusIE(InfoExtractor): }] class NBAIE(InfoExtractor): - _VALID_URL = r'^(?:https?://)?(?:watch\.|www\.)?nba\.com/(?:nba/)?video(/[^?]*)(\?.*)?$' + _VALID_URL = r'^(?:https?://)?(?:watch\.|www\.)?nba\.com/(?:nba/)?video(/[^?]*?)(?:/index\.html)?(?:\?.*)?$' IE_NAME = u'nba' def _real_extract(self, url): @@ -3077,28 +3166,27 @@ class NBAIE(InfoExtractor): raise ExtractorError(u'Invalid URL: %s' % url) video_id = mobj.group(1) - if video_id.endswith('/index.html'): - video_id = video_id[:-len('/index.html')] webpage = self._download_webpage(url, video_id) video_url = u'http://ht-mobile.cdn.turner.com/nba/big' + video_id + '_nba_1280x720.mp4' - def _findProp(rexp, default=None): - m = re.search(rexp, webpage) - if m: - return unescapeHTML(m.group(1)) - else: - return default shortened_video_id = video_id.rpartition('/')[2] - title = _findProp(r'<meta property="og:title" content="(.*?)"', shortened_video_id).replace('NBA.com: ', '') + title = self._html_search_regex(r'<meta property="og:title" content="(.*?)"', + webpage, 'title', default=shortened_video_id).replace('NBA.com: ', '') + + # It isn't there in the HTML it returns to us + # uploader_date = self._html_search_regex(r'<b>Date:</b> (.*?)</div>', webpage, 'upload_date', fatal=False) + + description = self._html_search_regex(r'<meta name="description" (?:content|value)="(.*?)" />', webpage, 'description', fatal=False) + info = { 'id': shortened_video_id, 'url': video_url, 'ext': 'mp4', 'title': title, - 'uploader_date': _findProp(r'<b>Date:</b> (.*?)</div>'), - 'description': _findProp(r'<div class="description">(.*?)</h1>'), + # 'uploader_date': uploader_date, + 'description': description, } return [info] @@ -3246,30 +3334,21 @@ class FunnyOrDieIE(InfoExtractor): video_id = mobj.group('id') webpage = self._download_webpage(url, video_id) - m = re.search(r'<video[^>]*>\s*<source[^>]*>\s*<source src="(?P<url>[^"]+)"', webpage, re.DOTALL) - if not m: - raise ExtractorError(u'Unable to find video information') - video_url = unescapeHTML(m.group('url')) + video_url = self._html_search_regex(r'<video[^>]*>\s*<source[^>]*>\s*<source src="(?P<url>[^"]+)"', + webpage, u'video URL', flags=re.DOTALL) - m = re.search(r"<h1 class='player_page_h1'.*?>(?P<title>.*?)</h1>", webpage, flags=re.DOTALL) - if not m: - m = re.search(r'<title>(?P<title>[^<]+?)', webpage) - if not m: - raise ExtractorError(u'Cannot find video title') - title = clean_html(m.group('title')) + title = self._html_search_regex((r"

(?P.*?)</h1>", + r'<title>(?P<title>[^<]+?)'), webpage, 'title', flags=re.DOTALL) - m = re.search(r'\d+)/? (?P\d*)(?P\??) #For urltype == video we sometimes get the videoID """ + _VIDEO_PAGE_TEMPLATE = 'http://store.steampowered.com/video/%s/' + _AGECHECK_TEMPLATE = 'http://store.steampowered.com/agecheck/video/%s/?snr=1_agecheck_agecheck__age-gate&ageDay=1&ageMonth=January&ageYear=1970' @classmethod def suitable(cls, url): @@ -3289,11 +3370,19 @@ class SteamIE(InfoExtractor): def _real_extract(self, url): m = re.match(self._VALID_URL, url, re.VERBOSE) gameID = m.group('gameID') - videourl = 'http://store.steampowered.com/agecheck/video/%s/?snr=1_agecheck_agecheck__age-gate&ageDay=1&ageMonth=January&ageYear=1970' % gameID - self.report_age_confirmation() + + videourl = self._VIDEO_PAGE_TEMPLATE % gameID webpage = self._download_webpage(videourl, gameID) - game_title = re.search(r'

(?P.*?)

', webpage).group('game_title') - + + if re.search('

Please enter your birth date to continue:

', webpage) is not None: + videourl = self._AGECHECK_TEMPLATE % gameID + self.report_age_confirmation() + webpage = self._download_webpage(videourl, gameID) + + self.report_extraction(gameID) + game_title = self._html_search_regex(r'

(.*?)

', + webpage, 'game title') + urlRE = r"'movie_(?P\d+)': \{\s*FILENAME: \"(?P[\w:/\.\?=]+)\"(,\s*MOVIE_NAME: \"(?P[\w:/\.\?=\+-]+)\")?\s*\}," mweb = re.finditer(urlRE, webpage) namesRE = r'(?P.+?)' @@ -3325,65 +3414,66 @@ class UstreamIE(InfoExtractor): def _real_extract(self, url): m = re.match(self._VALID_URL, url) 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.+)"',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) + + video_title = self._html_search_regex(r'data-title="(?P<title>.+)"', + webpage, u'title') + + uploader = self._html_search_regex(r'data-content-type="channel".*?>(?P<uploader>.*?)</a>', + webpage, u'uploader', fatal=False, flags=re.DOTALL) + + thumbnail = self._html_search_regex(r'<link rel="image_src" href="(?P<thumb>.*?)"', + webpage, u'thumbnail', fatal=False) + info = { - 'id':video_id, - 'url':video_url, + 'id': video_id, + 'url': video_url, 'ext': 'flv', - 'title': title, - 'uploader': uploader - } - return [info] + 'title': video_title, + 'uploader': uploader, + 'thumbnail': thumbnail, + } + return info class WorldStarHipHopIE(InfoExtractor): _VALID_URL = r'https?://(?:www|m)\.worldstar(?:candy|hiphop)\.com/videos/video\.php\?v=(?P<id>.*)' IE_NAME = u'WorldStarHipHop' def _real_extract(self, url): - _src_url = r'so\.addVariable\("file","(.*?)"\)' - m = re.match(self._VALID_URL, url) video_id = m.group('id') - webpage_src = self._download_webpage(url, video_id) + webpage_src = self._download_webpage(url, video_id) - mobj = re.search(_src_url, webpage_src) + video_url = self._search_regex(r'so\.addVariable\("file","(.*?)"\)', + webpage_src, u'video URL') - if mobj is not None: - video_url = mobj.group(1) - if 'mp4' in video_url: - ext = 'mp4' - else: - ext = 'flv' + if 'mp4' in video_url: + ext = 'mp4' else: - raise ExtractorError(u'Cannot find video url for %s' % video_id) + ext = 'flv' - mobj = re.search(r"<title>(.*)", webpage_src) + video_title = self._html_search_regex(r"(.*)", + webpage_src, u'title') - if mobj is None: - raise ExtractorError(u'Cannot determine title') - title = mobj.group(1) - - mobj = re.search(r'rel="image_src" href="(.*)" />', webpage_src) # Getting thumbnail and if not thumbnail sets correct title for WSHH candy video. - if mobj is not None: - thumbnail = mobj.group(1) - else: + thumbnail = self._html_search_regex(r'rel="image_src" href="(.*)" />', + webpage_src, u'thumbnail', fatal=False) + + if not thumbnail: _title = r"""candytitles.*>(.*)""" mobj = re.search(_title, webpage_src) if mobj is not None: - title = mobj.group(1) - thumbnail = None + video_title = mobj.group(1) results = [{ 'id': video_id, 'url' : video_url, - 'title' : title, + 'title' : video_title, 'thumbnail' : thumbnail, 'ext' : ext, }] @@ -3397,10 +3487,9 @@ class RBMARadioIE(InfoExtractor): video_id = m.group('videoID') webpage = self._download_webpage(url, video_id) - m = re.search(r'', webpage) - if not m: - raise ExtractorError(u'Cannot find metadata') - json_data = m.group(1) + + json_data = self._search_regex(r'window\.gon.*?gon\.show=(.+?);$', + webpage, u'json data', flags=re.MULTILINE) try: data = json.loads(json_data) @@ -3447,42 +3536,33 @@ class YouPornIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) if mobj is None: raise ExtractorError(u'Invalid URL: %s' % url) - video_id = mobj.group('videoid') req = compat_urllib_request.Request(url) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, video_id) - # Get the video title - result = re.search(r'(?P.*)</h1>', webpage) - if result is None: - raise ExtractorError(u'Unable to extract video title') - video_title = result.group('title').strip() - - # Get the video date - result = re.search(r'Date:</label>(?P<date>.*) </li>', webpage) - if result is None: - self._downloader.report_warning(u'unable to extract video date') - upload_date = None - else: - upload_date = unified_strdate(result.group('date').strip()) + # Get JSON parameters + json_params = self._search_regex(r'var currentVideo = new Video\((.*)\);', webpage, u'JSON parameters') + try: + params = json.loads(json_params) + except: + raise ExtractorError(u'Invalid JSON') - # Get the video uploader - result = re.search(r'Submitted:</label>(?P<uploader>.*)</li>', webpage) - if result is None: - self._downloader.report_warning(u'unable to extract uploader') - video_uploader = None - else: - video_uploader = result.group('uploader').strip() - video_uploader = clean_html( video_uploader ) + self.report_extraction(video_id) + try: + video_title = params['title'] + upload_date = unified_strdate(params['release_date_f']) + video_description = params['description'] + video_uploader = params['submitted_by'] + thumbnail = params['thumbnails'][0]['image'] + except KeyError: + raise ExtractorError('Missing JSON parameter: ' + sys.exc_info()[1]) # Get all of the formats available DOWNLOAD_LIST_RE = r'(?s)<ul class="downloadList">(?P<download_list>.*?)</ul>' - result = re.search(DOWNLOAD_LIST_RE, webpage) - if result is None: - raise ExtractorError(u'Unable to extract download list') - download_list_html = result.group('download_list').strip() + download_list_html = self._search_regex(DOWNLOAD_LIST_RE, + webpage, u'download list').strip() # Get all of the links from the page LINK_RE = r'(?s)<a href="(?P<url>[^"]+)">' @@ -3506,19 +3586,18 @@ class YouPornIE(InfoExtractor): size = format[0] bitrate = format[1] format = "-".join( format ) - title = u'%s-%s-%s' % (video_title, size, bitrate) + # title = u'%s-%s-%s' % (video_title, size, bitrate) formats.append({ 'id': video_id, 'url': video_url, 'uploader': video_uploader, 'upload_date': upload_date, - 'title': title, + 'title': video_title, 'ext': extension, 'format': format, - 'thumbnail': None, - 'description': None, - 'player_url': None + 'thumbnail': thumbnail, + 'description': video_description }) if self._downloader.params.get('listformats', None): @@ -3559,17 +3638,13 @@ class PornotubeIE(InfoExtractor): # Get the video URL VIDEO_URL_RE = r'url: "(?P<url>http://video[0-9].pornotube.com/.+\.flv)",' - result = re.search(VIDEO_URL_RE, webpage) - if result is None: - raise ExtractorError(u'Unable to extract video url') - video_url = compat_urllib_parse.unquote(result.group('url')) + video_url = self._search_regex(VIDEO_URL_RE, webpage, u'video url') + video_url = compat_urllib_parse.unquote(video_url) #Get the uploaded date VIDEO_UPLOADED_RE = r'<div class="video_added_by">Added (?P<date>[0-9\/]+) by' - result = re.search(VIDEO_UPLOADED_RE, webpage) - if result is None: - raise ExtractorError(u'Unable to extract video title') - upload_date = unified_strdate(result.group('date')) + upload_date = self._html_search_regex(VIDEO_UPLOADED_RE, webpage, u'upload date', fatal=False) + if upload_date: upload_date = unified_strdate(upload_date) info = {'id': video_id, 'url': video_url, @@ -3596,10 +3671,8 @@ class YouJizzIE(InfoExtractor): webpage = self._download_webpage(url, video_id) # Get the video title - result = re.search(r'<title>(?P<title>.*)', webpage) - if result is None: - raise ExtractorError(u'ERROR: unable to extract video title') - video_title = result.group('title').strip() + video_title = self._html_search_regex(r'(?P<title>.*)', + webpage, u'title').strip() # Get the embed page result = re.search(r'https?://www.youjizz.com/videos/embed/(?P[0-9]+)', webpage) @@ -3612,10 +3685,8 @@ class YouJizzIE(InfoExtractor): webpage = self._download_webpage(embed_page_url, video_id) # Get the video URL - result = re.search(r'so.addVariable\("file",encodeURIComponent\("(?P[^"]+)"\)\);', webpage) - if result is None: - raise ExtractorError(u'ERROR: unable to extract video url') - video_url = result.group('source') + video_url = self._search_regex(r'so.addVariable\("file",encodeURIComponent\("(?P[^"]+)"\)\);', + webpage, u'video URL') info = {'id': video_id, 'url': video_url, @@ -3638,10 +3709,7 @@ class EightTracksIE(InfoExtractor): webpage = self._download_webpage(url, playlist_id) - m = re.search(r"PAGE.mix = (.*?);\n", webpage, flags=re.DOTALL) - if not m: - raise ExtractorError(u'Cannot find trax information') - json_like = m.group(1) + json_like = self._search_regex(r"PAGE.mix = (.*?);\n", webpage, u'trax information', flags=re.DOTALL) data = json.loads(json_like) session = str(random.randint(0, 1000000000)) @@ -3677,18 +3745,22 @@ class KeekIE(InfoExtractor): def _real_extract(self, url): m = re.match(self._VALID_URL, url) video_id = m.group('videoID') + video_url = u'http://cdn.keek.com/keek/video/%s' % video_id thumbnail = u'http://cdn.keek.com/keek/thumbnail/%s/w100/h75' % video_id webpage = self._download_webpage(url, video_id) - m = re.search(r'[\S\s]+?

(?P.+?)

', webpage) - uploader = clean_html(m.group('uploader')) + + video_title = self._html_search_regex(r'[\S\s]+?

(?P.+?)

', + webpage, u'uploader', fatal=False) + info = { 'id': video_id, 'url': video_url, 'ext': 'mp4', - 'title': title, + 'title': video_title, 'thumbnail': thumbnail, 'uploader': uploader } @@ -3720,10 +3792,6 @@ class TEDIE(InfoExtractor): self.to_screen(u'Getting info of playlist %s: "%s"' % (playlist_id,name)) return [self._playlist_videos_info(url,name,playlist_id)] - def _talk_video_link(self,mediaSlug): - '''Returns the video link for that mediaSlug''' - return 'http://download.ted.com/talks/%s.mp4' % mediaSlug - def _playlist_videos_info(self,url,name,playlist_id=0): '''Returns the videos of the playlist''' video_RE=r''' @@ -3736,9 +3804,8 @@ class TEDIE(InfoExtractor): m_videos=re.finditer(video_RE,webpage,re.VERBOSE) m_names=re.finditer(video_name_RE,webpage) - playlist_RE = r'div class="headline">(\s*?)

(\s*?)(?P.*?)' - m_playlist = re.search(playlist_RE, webpage) - playlist_title = m_playlist.group('playlist_title') + playlist_title = self._html_search_regex(r'div class="headline">\s*?

\s*?(.*?)', + webpage, 'playlist title') playlist_entries = [] for m_video, m_name in zip(m_videos,m_names): @@ -3749,27 +3816,28 @@ class TEDIE(InfoExtractor): def _talk_info(self, url, video_id=0): """Return the video for the talk in the url""" - m=re.match(self._VALID_URL, url,re.VERBOSE) - videoName=m.group('name') - webpage=self._download_webpage(url, video_id, 'Downloading \"%s\" page' % videoName) + m = re.match(self._VALID_URL, url,re.VERBOSE) + video_name = m.group('name') + webpage = self._download_webpage(url, video_id, 'Downloading \"%s\" page' % video_name) + self.report_extraction(video_name) # If the url includes the language we get the title translated - title_RE=r'(?P.*)</span>' - title=re.search(title_RE, webpage).group('title') - info_RE=r'''<script\ type="text/javascript">var\ talkDetails\ =(.*?) - "id":(?P<videoID>[\d]+).*? - "mediaSlug":"(?P<mediaSlug>[\w\d]+?)"''' - thumb_RE=r'</span>[\s.]*</div>[\s.]*<img src="(?P<thumbnail>.*?)"' - thumb_match=re.search(thumb_RE,webpage) - info_match=re.search(info_RE,webpage,re.VERBOSE) - video_id=info_match.group('videoID') - mediaSlug=info_match.group('mediaSlug') - video_url=self._talk_video_link(mediaSlug) + title = self._html_search_regex(r'<span id="altHeadline" >(?P<title>.*)</span>', + webpage, 'title') + json_data = self._search_regex(r'<script.*?>var talkDetails = ({.*?})</script>', + webpage, 'json data') + info = json.loads(json_data) + desc = self._html_search_regex(r'<div class="talk-intro">.*?<p.*?>(.*?)</p>', + webpage, 'description', flags = re.DOTALL) + + thumbnail = self._search_regex(r'</span>[\s.]*</div>[\s.]*<img src="(.*?)"', + webpage, 'thumbnail') info = { - 'id': video_id, - 'url': video_url, + 'id': info['id'], + 'url': info['htmlStreams'][-1]['file'], 'ext': 'mp4', 'title': title, - 'thumbnail': thumb_match.group('thumbnail') + 'thumbnail': thumbnail, + 'description': desc, } return info @@ -3835,10 +3903,9 @@ class SpiegelIE(InfoExtractor): video_id = m.group('videoID') webpage = self._download_webpage(url, video_id) - m = re.search(r'<div class="spVideoTitle">(.*?)</div>', webpage) - if not m: - raise ExtractorError(u'Cannot find title') - video_title = unescapeHTML(m.group(1)) + + video_title = self._html_search_regex(r'<div class="module-title">(.*?)</div>', + webpage, u'title') xml_url = u'http://video2.spiegel.de/flash/' + video_id + u'.xml' xml_code = self._download_webpage(xml_url, video_id, @@ -3874,35 +3941,25 @@ class LiveLeakIE(InfoExtractor): webpage = self._download_webpage(url, video_id) - m = re.search(r'file: "(.*?)",', webpage) - if not m: - raise ExtractorError(u'Unable to find video url') - video_url = m.group(1) + video_url = self._search_regex(r'file: "(.*?)",', + webpage, u'video URL') - m = re.search(r'<meta property="og:title" content="(?P<title>.*?)"', webpage) - if not m: - raise ExtractorError(u'Cannot find video title') - title = unescapeHTML(m.group('title')).replace('LiveLeak.com -', '').strip() + video_title = self._html_search_regex(r'<meta property="og:title" content="(?P<title>.*?)"', + webpage, u'title').replace('LiveLeak.com -', '').strip() - m = re.search(r'<meta property="og:description" content="(?P<desc>.*?)"', webpage) - if m: - desc = unescapeHTML(m.group('desc')) - else: - desc = None + video_description = self._html_search_regex(r'<meta property="og:description" content="(?P<desc>.*?)"', + webpage, u'description', fatal=False) - m = re.search(r'By:.*?(\w+)</a>', webpage) - if m: - uploader = clean_html(m.group(1)) - else: - uploader = None + video_uploader = self._html_search_regex(r'By:.*?(\w+)</a>', + webpage, u'uploader', fatal=False) info = { 'id': video_id, 'url': video_url, 'ext': 'mp4', - 'title': title, - 'description': desc, - 'uploader': uploader + 'title': video_title, + 'description': video_description, + 'uploader': video_uploader } return [info] @@ -3946,6 +4003,64 @@ class ARDIE(InfoExtractor): info["url"] = stream["video_url"] return [info] +class ZDFIE(InfoExtractor): + _VALID_URL = r'^http://www\.zdf\.de\/ZDFmediathek\/(.*beitrag\/video\/)(?P<video_id>[^/\?]+)(?:\?.*)?' + _TITLE = r'<h1(?: class="beitragHeadline")?>(?P<title>.*)</h1>' + _MEDIA_STREAM = r'<a href="(?P<video_url>.+(?P<media_type>.streaming).+/zdf/(?P<quality>[^\/]+)/[^"]*)".+class="play".+>' + _MMS_STREAM = r'href="(?P<video_url>mms://[^"]*)"' + _RTSP_STREAM = r'(?P<video_url>rtsp://[^"]*.mp4)' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + if mobj is None: + raise ExtractorError(u'Invalid URL: %s' % url) + video_id = mobj.group('video_id') + + html = self._download_webpage(url, video_id) + streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)] + if streams is None: + raise ExtractorError(u'No media url found.') + + # s['media_type'] == 'wstreaming' -> use 'Windows Media Player' and mms url + # s['media_type'] == 'hstreaming' -> use 'Quicktime' and rtsp url + # choose first/default media type and highest quality for now + for s in streams: #find 300 - dsl1000mbit + if s['quality'] == '300' and s['media_type'] == 'wstreaming': + stream_=s + break + for s in streams: #find veryhigh - dsl2000mbit + if s['quality'] == 'veryhigh' and s['media_type'] == 'wstreaming': # 'hstreaming' - rtsp is not working + stream_=s + break + if stream_ is None: + raise ExtractorError(u'No stream found.') + + media_link = self._download_webpage(stream_['video_url'], video_id,'Get stream URL') + + self.report_extraction(video_id) + mobj = re.search(self._TITLE, html) + if mobj is None: + raise ExtractorError(u'Cannot extract title') + title = unescapeHTML(mobj.group('title')) + + mobj = re.search(self._MMS_STREAM, media_link) + if mobj is None: + mobj = re.search(self._RTSP_STREAM, media_link) + if mobj is None: + raise ExtractorError(u'Cannot extract mms:// or rtsp:// URL') + mms_url = mobj.group('video_url') + + mobj = re.search('(.*)[.](?P<ext>[^.]+)', mms_url) + if mobj is None: + raise ExtractorError(u'Cannot extract extention') + ext = mobj.group('ext') + + return [{'id': video_id, + 'url': mms_url, + 'title': title, + 'ext': ext + }] + class TumblrIE(InfoExtractor): _VALID_URL = r'http://(?P<blog_name>.*?)\.tumblr\.com/((post)|(video))/(?P<id>\d*)/(.*?)' @@ -3960,23 +4075,23 @@ class TumblrIE(InfoExtractor): re_video = r'src=\\x22(?P<video_url>http://%s\.tumblr\.com/video_file/%s/(.*?))\\x22 type=\\x22video/(?P<ext>.*?)\\x22' % (blog, video_id) video = re.search(re_video, webpage) if video is None: - self.to_screen("No video founded") - return [] + raise ExtractorError(u'Unable to extract video') video_url = video.group('video_url') ext = video.group('ext') - re_thumb = r'posters(.*?)\[\\x22(?P<thumb>.*?)\\x22' # We pick the first poster - thumb = re.search(re_thumb, webpage).group('thumb').replace('\\', '') + video_thumbnail = self._search_regex(r'posters(.*?)\[\\x22(?P<thumb>.*?)\\x22', + webpage, u'thumbnail', fatal=False) # We pick the first poster + if video_thumbnail: video_thumbnail = video_thumbnail.replace('\\', '') # The only place where you can get a title, it's not complete, # but searching in other places doesn't work for all videos - re_title = r'<title>(?P<title>.*?)' - title = unescapeHTML(re.search(re_title, webpage, re.DOTALL).group('title')) + video_title = self._html_search_regex(r'(?P<title>.*?)', + webpage, u'title', flags=re.DOTALL) return [{'id': video_id, 'url': video_url, - 'title': title, - 'thumbnail': thumb, + 'title': video_title, + 'thumbnail': video_thumbnail, 'ext': ext }] @@ -3990,7 +4105,7 @@ class BandcampIE(InfoExtractor): # We get the link to the free download page m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage) if m_download is None: - raise ExtractorError(u'No free songs founded') + raise ExtractorError(u'No free songs found') download_link = m_download.group(1) id = re.search(r'var TralbumData = {(.*?)id: (?P\d*?)$', @@ -4018,10 +4133,10 @@ class BandcampIE(InfoExtractor): track_info = {'id':id, 'title' : info[u'title'], - 'ext' : 'mp3', - 'url' : final_url, + 'ext' : 'mp3', + 'url' : final_url, 'thumbnail' : info[u'thumb_url'], - 'uploader' : info[u'artist'] + 'uploader' : info[u'artist'] } return [track_info] @@ -4038,17 +4153,14 @@ class RedTubeIE(InfoExtractor): video_id = mobj.group('id') video_extension = 'mp4' webpage = self._download_webpage(url, video_id) + self.report_extraction(video_id) - mobj = re.search(r'',webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract media URL') + video_url = self._html_search_regex(r'', + webpage, u'video URL') - video_url = mobj.group(1) - mobj = re.search('

(.+)

',webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = mobj.group(1) + video_title = self._html_search_regex('

(.+?)

', + webpage, u'title') return [{ 'id': video_id, @@ -4059,7 +4171,7 @@ class RedTubeIE(InfoExtractor): class InaIE(InfoExtractor): """Information Extractor for Ina.fr""" - _VALID_URL = r'(?:http://)?(?:www.)?ina\.fr/video/(?PI[0-9]+)/.*' + _VALID_URL = r'(?:http://)?(?:www\.)?ina\.fr/video/(?PI[0-9]+)/.*' def _real_extract(self,url): mobj = re.match(self._VALID_URL, url) @@ -4069,23 +4181,366 @@ class InaIE(InfoExtractor): video_extension = 'mp4' webpage = self._download_webpage(mrss_url, video_id) - mobj = re.search(r'.*?)]]>', + webpage, u'title') + + return [{ + 'id': video_id, + 'url': video_url, + 'ext': video_extension, + 'title': video_title, + }] + +class HowcastIE(InfoExtractor): + """Information Extractor for Howcast.com""" + _VALID_URL = r'(?:https?://)?(?:www\.)?howcast\.com/videos/(?P\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) + + self.report_extraction(video_id) + + video_url = self._search_regex(r'\'?file\'?: "(http://mobile-media\.howcast\.com/[0-9]+\.mp4)', + webpage, u'video URL') + + video_title = self._html_search_regex(r'\w+)' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + + video_id = mobj.group('id') + webpage_url = 'https://vine.co/v/' + video_id + webpage = self._download_webpage(webpage_url, video_id) + + self.report_extraction(video_id) + + video_url = self._html_search_regex(r'.*?

(.+?)

', + webpage, u'uploader', fatal=False, flags=re.DOTALL) + + return [{ + 'id': video_id, + 'url': video_url, + 'ext': 'mp4', + 'title': video_title, + 'thumbnail': thumbnail, + 'uploader': uploader, + }] + +class FlickrIE(InfoExtractor): + """Information Extractor for Flickr videos""" + _VALID_URL = r'(?:https?://)?(?:www\.)?flickr\.com/photos/(?P[\w\-_@]+)/(?P\d+).*' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + + video_id = mobj.group('id') + video_uploader_id = mobj.group('uploader_id') + webpage_url = 'http://www.flickr.com/photos/' + video_uploader_id + '/' + video_id + webpage = self._download_webpage(webpage_url, video_id) + + secret = self._search_regex(r"photo_secret: '(\w+)'", webpage, u'secret') + + first_url = 'https://secure.flickr.com/apps/video/video_mtl_xml.gne?v=x&photo_id=' + video_id + '&secret=' + secret + '&bitrate=700&target=_self' + first_xml = self._download_webpage(first_url, video_id, 'Downloading first data webpage') + + node_id = self._html_search_regex(r'(\d+-\d+)', + first_xml, u'node_id') + + second_url = 'https://secure.flickr.com/video_playlist.gne?node_id=' + node_id + '&tech=flash&mode=playlist&bitrate=700&secret=' + secret + '&rd=video.yahoo.com&noad=1' + second_xml = self._download_webpage(second_url, video_id, 'Downloading second data webpage') + + self.report_extraction(video_id) + + mobj = re.search(r'.*?)]]>', webpage) + thumbnail = self._html_search_regex(r'.*)' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = mobj.group(1) + raise ExtractorError(u'Invalid URL: %s' % url) + url_title = mobj.group('url_title') + webpage = self._download_webpage(url, url_title) + + video_id = self._html_search_regex(r'
(.*?)', + data, u'video URL') + + return [{ + 'id': video_id, + 'url': video_url, + 'ext': 'mp4', + 'title': video_title, + 'thumbnail': thumbnail, + 'description': video_description, + }] + +class XHamsterIE(InfoExtractor): + """Information Extractor for xHamster""" + _VALID_URL = r'(?:http://)?(?:www.)?xhamster\.com/movies/(?P[0-9]+)/.*\.html' + + def _real_extract(self,url): + mobj = re.match(self._VALID_URL, url) + + video_id = mobj.group('id') + mrss_url = 'http://xhamster.com/movies/%s/.html' % video_id + webpage = self._download_webpage(mrss_url, video_id) + + mobj = re.search(r'\'srv\': \'(?P[^\']*)\',\s*\'file\': \'(?P[^\']+)\',', webpage) + if mobj is None: + raise ExtractorError(u'Unable to extract media URL') + if len(mobj.group('server')) == 0: + video_url = compat_urllib_parse.unquote(mobj.group('file')) + else: + video_url = mobj.group('server')+'/key='+mobj.group('file') + video_extension = video_url.split('.')[-1] + + video_title = self._html_search_regex(r'(?P<title>.+?) - xHamster\.com', + webpage, u'title') + + # Can't see the description anywhere in the UI + # video_description = self._html_search_regex(r'Description: (?P[^<]+)', + # webpage, u'description', fatal=False) + # if video_description: video_description = unescapeHTML(video_description) + + mobj = re.search(r'hint=\'(?P[0-9]{4})-(?P[0-9]{2})-(?P[0-9]{2}) [0-9]{2}:[0-9]{2}:[0-9]{2} [A-Z]{3,4}\'', webpage) + if mobj: + video_upload_date = mobj.group('upload_date_Y')+mobj.group('upload_date_m')+mobj.group('upload_date_d') + else: + video_upload_date = None + self._downloader.report_warning(u'Unable to extract upload date') + + video_uploader_id = self._html_search_regex(r']+>(?P[^<]+)', + webpage, u'uploader id', default=u'anonymous') + + video_thumbnail = self._search_regex(r'\'image\':\'(?P[^\']+)\'', + webpage, u'thumbnail', fatal=False) return [{ 'id': video_id, 'url': video_url, 'ext': video_extension, 'title': video_title, + # 'description': video_description, + 'upload_date': video_upload_date, + 'uploader_id': video_uploader_id, + 'thumbnail': video_thumbnail + }] + +class HypemIE(InfoExtractor): + """Information Extractor for hypem""" + _VALID_URL = r'(?:http://)?(?:www\.)?hypem\.com/track/([^/]+)/([^/]+)' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + if mobj is None: + raise ExtractorError(u'Invalid URL: %s' % url) + track_id = mobj.group(1) + + data = { 'ax': 1, 'ts': time.time() } + data_encoded = compat_urllib_parse.urlencode(data) + complete_url = url + "?" + data_encoded + request = compat_urllib_request.Request(complete_url) + response, urlh = self._download_webpage_handle(request, track_id, u'Downloading webpage with the url') + cookie = urlh.headers.get('Set-Cookie', '') + + self.report_extraction(track_id) + + html_tracks = self._html_search_regex(r'', + response, u'tracks', flags=re.MULTILINE|re.DOTALL).strip() + try: + track_list = json.loads(html_tracks) + track = track_list[u'tracks'][0] + except ValueError: + raise ExtractorError(u'Hypemachine contained invalid JSON.') + + key = track[u"key"] + track_id = track[u"id"] + artist = track[u"artist"] + title = track[u"song"] + + serve_url = "http://hypem.com/serve/source/%s/%s" % (compat_str(track_id), compat_str(key)) + request = compat_urllib_request.Request(serve_url, "" , {'Content-Type': 'application/json'}) + request.add_header('cookie', cookie) + song_data_json = self._download_webpage(request, track_id, u'Downloading metadata') + try: + song_data = json.loads(song_data_json) + except ValueError: + raise ExtractorError(u'Hypemachine contained invalid JSON.') + final_url = song_data[u"url"] + + return [{ + 'id': track_id, + 'url': final_url, + 'ext': "mp3", + 'title': title, + 'artist': artist, }] +class Vbox7IE(InfoExtractor): + """Information Extractor for Vbox7""" + _VALID_URL = r'(?:http://)?(?:www\.)?vbox7\.com/play:([^/]+)' + + def _real_extract(self,url): + mobj = re.match(self._VALID_URL, url) + if mobj is None: + raise ExtractorError(u'Invalid URL: %s' % url) + video_id = mobj.group(1) + + redirect_page, urlh = self._download_webpage_handle(url, video_id) + new_location = self._search_regex(r'window\.location = \'(.*)\';', redirect_page, u'redirect location') + redirect_url = urlh.geturl() + new_location + webpage = self._download_webpage(redirect_url, video_id, u'Downloading redirect page') + + title = self._html_search_regex(r'(.*)', + webpage, u'title').split('/')[0].strip() + + ext = "flv" + info_url = "http://vbox7.com/play/magare.do" + data = compat_urllib_parse.urlencode({'as3':'1','vid':video_id}) + info_request = compat_urllib_request.Request(info_url, data) + info_request.add_header('Content-Type', 'application/x-www-form-urlencoded') + info_response = self._download_webpage(info_request, video_id, u'Downloading info webpage') + if info_response is None: + raise ExtractorError(u'Unable to extract the media url') + (final_url, thumbnail_url) = map(lambda x: x.split('=')[1], info_response.split('&')) + + return [{ + 'id': video_id, + 'url': final_url, + 'ext': ext, + 'title': title, + 'thumbnail': thumbnail_url, + }] + +class GametrailersIE(InfoExtractor): + _VALID_URL = r'http://www.gametrailers.com/(?Pvideos|reviews|full-episodes)/(?P.*?)/(?P.*)' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + if mobj is None: + raise ExtractorError(u'Invalid URL: %s' % url) + video_id = mobj.group('id') + video_type = mobj.group('type') + webpage = self._download_webpage(url, video_id) + if video_type == 'full-episodes': + mgid_re = r'data-video="(?P<mgid>mgid:.*?)"' + else: + mgid_re = r'data-contentId=\'(?P<mgid>mgid:.*?)\'' + mgid = self._search_regex(mgid_re, webpage, u'mgid') + data = compat_urllib_parse.urlencode({'uri': mgid, 'acceptMethods': 'fms'}) + + info_page = self._download_webpage('http://www.gametrailers.com/feeds/mrss?' + data, + video_id, u'Downloading video info') + links_webpage = self._download_webpage('http://www.gametrailers.com/feeds/mediagen/?' + data, + video_id, u'Downloading video urls info') + + self.report_extraction(video_id) + info_re = r'''<title><!\[CDATA\[(?P<title>.*?)\]\]>.* + .*?)\]\]>.* + .* + (?P.*?).* + ''' + + m_info = re.search(info_re, info_page, 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') + + m_urls = list(re.finditer(r'(?P.*)', links_webpage)) + if m_urls is None or len(m_urls) == 0: + raise ExtractError(u'Unable to extrat video url') + # They are sorted from worst to best quality + video_url = m_urls[-1].group('url') + + return {'url': video_url, + 'id': video_id, + 'title': video_title, + # Videos are actually flv not mp4 + 'ext': 'flv', + 'thumbnail': video_thumb, + '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. @@ -4104,8 +4559,8 @@ def gen_extractors(): YahooSearchIE(), DepositFilesIE(), FacebookIE(), - BlipTVUserIE(), BlipTVIE(), + BlipTVUserIE(), VimeoIE(), MyVideoIE(), ComedyCentralIE(), @@ -4139,10 +4594,19 @@ def gen_extractors(): SpiegelIE(), LiveLeakIE(), ARDIE(), + ZDFIE(), TumblrIE(), BandcampIE(), RedTubeIE(), InaIE(), + HowcastIE(), + VineIE(), + FlickrIE(), + TeamcocoIE(), + XHamsterIE(), + HypemIE(), + Vbox7IE(), + GametrailersIE(), GenericIE() ]