diff options
Diffstat (limited to 'youtube_dl/InfoExtractors.py')
| -rwxr-xr-x | youtube_dl/InfoExtractors.py | 972 |
1 files changed, 464 insertions, 508 deletions
diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py index e27e0cb7c..574d417be 100755 --- a/youtube_dl/InfoExtractors.py +++ b/youtube_dl/InfoExtractors.py @@ -191,6 +191,47 @@ 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. @@ -379,7 +420,7 @@ class YoutubeIE(InfoExtractor): 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_lang = self._downloader.params.get('subtitleslang') or 'en' 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) @@ -658,14 +699,14 @@ class YoutubeIE(InfoExtractor): pass else: # We report the original error - self._downloader.report_error(sub_error) + self._downloader.report_warning(sub_error) if self._downloader.params.get('allsubtitles', False): video_subtitles = self._extract_all_subtitles(video_id) for video_subtitle in video_subtitles: (sub_error, sub_lang, sub) = video_subtitle if sub_error: - self._downloader.report_error(sub_error) + self._downloader.report_warning(sub_error) if self._downloader.params.get('listsubtitles', False): sub_lang_list = self._list_available_subtitles(video_id) @@ -691,8 +732,11 @@ class YoutubeIE(InfoExtractor): for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','): url_data = compat_parse_qs(url_data_str) if 'itag' in url_data and 'url' in url_data: - url = url_data['url'][0] + '&signature=' + url_data['sig'][0] - if not 'ratebypass' in url: url += '&ratebypass=yes' + url = url_data['url'][0] + if 'sig' in url_data: + url += '&signature=' + url_data['sig'][0] + if 'ratebypass' not in url: + url += '&ratebypass=yes' url_map[url_data['itag'][0]] = url format_limit = self._downloader.params.get('format_limit', None) @@ -899,16 +943,10 @@ class DailymotionIE(InfoExtractor): video_title = unescapeHTML(mobj.group('title')) video_uploader = None - mobj = re.search(r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>', webpage) - if mobj is None: - # lookin for official user - mobj_official = re.search(r'<span rel="author"[^>]+?>([^<]+?)</span>', 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)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>', + # Looking for official user + r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'], + webpage, 'video uploader') video_upload_date = None mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage) @@ -964,18 +1002,13 @@ class PhotobucketIE(InfoExtractor): }] # We try looking in other parts of the webpage - mobj = re.search(r'<link rel="video_src" href=".*\?file=([^"]+)" />', 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'<link rel="video_src" href=".*\?file=([^"]+)" />', + webpage, u'video URL') mobj = re.search(r'<title>(.*) video by (.*) - Photobucket</title>', 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 [{ @@ -1400,6 +1433,13 @@ class GenericIE(InfoExtractor): # Try to find twitter cards info mobj = re.search(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage) if mobj is None: + # We look for Open Graph info: + # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am) + m_video_type = re.search(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage) + # We only look in og:video if the MIME type is a video, don't try if it's a Flash player: + if m_video_type is not None: + mobj = re.search(r'<meta.*?property="og:video".*?content="(.*?)"', webpage) + if mobj is None: raise ExtractorError(u'Invalid URL: %s' % url) # It's possible that one of the regexes @@ -1420,16 +1460,12 @@ class GenericIE(InfoExtractor): # Site Name | Video Title # Video Title - Tagline | Site Name # and so on and so forth; it's just not practical - mobj = re.search(r'<title>(.*)</title>', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = mobj.group(1) + video_title = self._html_search_regex(r'<title>(.*)</title>', + 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, @@ -1567,7 +1603,7 @@ 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' @@ -1603,9 +1639,10 @@ class YoutubePlaylistIE(InfoExtractor): # Number of videos is a multiple of self._MAX_RESULTS break - videos += [ (entry['yt$position']['$t'], entry['content']['src']) - for entry in response['feed']['entry'] - if 'content' in entry ] + for entry in response['feed']['entry']: + index = entry['yt$position']['$t'] + if 'media$group' in entry and 'media$player' in entry['media$group']: + videos.append((index, entry['media$group']['media$player']['url'])) if len(response['feed']['entry']) < self._MAX_RESULTS: break @@ -1828,10 +1865,7 @@ class DepositFilesIE(InfoExtractor): file_extension = os.path.splitext(file_url)[1][1:] # Search for file title - mobj = re.search(r'<b title="(.*?)">', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - file_title = mobj.group(1).decode('utf-8') + file_title = self._search_regex(r'<b title="(.*?)">', webpage, u'title') return [{ 'id': file_id.decode('utf-8'), @@ -1925,10 +1959,8 @@ class FacebookIE(InfoExtractor): video_duration = int(video_data['video_duration']) thumbnail = video_data['thumbnail_src'] - m = re.search('<h2 class="uiHeaderTitle">([^<]+)</h2>', webpage) - if not m: - raise ExtractorError(u'Cannot find title in webpage') - video_title = unescapeHTML(m.group(1)) + video_title = self._html_search_regex('<h2 class="uiHeaderTitle">([^<]+)</h2>', + webpage, u'title') info = { 'id': video_id, @@ -2090,15 +2122,10 @@ class MyVideoIE(InfoExtractor): self.report_extraction(video_id) video_url = mobj.group(1) + '.flv' - mobj = re.search('<title>([^<]+)</title>', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = mobj.group(1) + video_title = self._html_search_regex('<title>([^<]+)</title>', + webpage, u'title') - mobj = re.search('[.](.+?)$', video_url) - if mobj is None: - raise ExtractorError(u'Unable to extract extention') - video_ext = mobj.group(1) + video_ext = self._search_regex('[.](.+?)$', video_url, u'extension') return [{ 'id': video_id, @@ -2146,25 +2173,23 @@ class MyVideoIE(InfoExtractor): # extracting infos self.report_extraction(video_id) + video_url = None mobj = re.search('connectionurl=\'(.*?)\'', dec_data) - if mobj is None: - raise ExtractorError(u'unable to extract rtmpurl') - video_rtmpurl = compat_urllib_parse.unquote(mobj.group(1)) - if 'myvideo2flash' in video_rtmpurl: - self._downloader.report_warning(u'forcing RTMPT ...') - video_rtmpurl = video_rtmpurl.replace('rtmpe://', 'rtmpt://') - - # extract non rtmp videos - if (video_rtmpurl is None) or (video_rtmpurl == ''): + 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_rtmpurl = compat_urllib_parse.unquote(mobj.group(1)) + compat_urllib_parse.unquote(mobj.group(2)) + video_url = compat_urllib_parse.unquote(mobj.group(1)) + compat_urllib_parse.unquote(mobj.group(2)) - mobj = re.search('source=\'(.*?)\'', dec_data) - if mobj is None: - raise ExtractorError(u'unable to extract swfobj') - video_file = compat_urllib_parse.unquote(mobj.group(1)) + 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('.') @@ -2176,20 +2201,16 @@ class MyVideoIE(InfoExtractor): video_filepath + video_file ).replace('.f4m', '.m3u8') - mobj = re.search('swfobject.embedSWF\(\'(.+?)\'', webpage) - if mobj is None: - raise ExtractorError(u'unable to extract swfobj') - video_swfobj = compat_urllib_parse.unquote(mobj.group(1)) + video_swfobj = self._search_regex('swfobject.embedSWF\(\'(.+?)\'', webpage, u'swfobj') + video_swfobj = compat_urllib_parse.unquote(video_swfobj) - mobj = re.search("<h1(?: class='globalHd')?>(.*?)</h1>", webpage) - if mobj is None: - raise ExtractorError(u'unable to extract title') - video_title = mobj.group(1) + video_title = self._html_search_regex("<h1(?: class='globalHd')?>(.*?)</h1>", + webpage, u'title') return [{ 'id': video_id, - 'url': video_rtmpurl, - 'tc_url': video_rtmpurl, + 'url': video_url, + 'tc_url': video_url, 'uploader': None, 'upload_date': None, 'title': video_title, @@ -2200,6 +2221,7 @@ class MyVideoIE(InfoExtractor): 'player_url': video_swfobj, }] + class ComedyCentralIE(InfoExtractor): """Information extractor for The Daily Show and Colbert Report """ @@ -2381,19 +2403,25 @@ 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('<meta name="description" content="([^"]*)"', + webpage, u'description', fatal=False) + + imgUrl = self._html_search_regex('<meta property="og:image" content="([^"]*)"', + webpage, u'thumbnail', fatal=False) + + playerUrl = self._html_search_regex('<meta property="og:video" content="([^"]*)"', + webpage, u'player url') - descMatch = re.search('<meta name="description" content="([^"]*)"', webPage) - description = unescapeHTML(descMatch.group(1)) - imgMatch = re.search('<meta property="og:image" content="([^"]*)"', webPage) - imgUrl = unescapeHTML(imgMatch.group(1)) - playerUrlMatch = re.search('<meta property="og:video" content="([^"]*)"', webPage) - playerUrl = unescapeHTML(playerUrlMatch.group(1)) - configUrlMatch = re.search('config=(.*)$', playerUrl) - configUrl = compat_urllib_parse.unquote(configUrlMatch.group(1)) + title = self._html_search_regex('<meta name="title" content="([^"]*)"', + webpage, u'player url').split(' : ')[-1] - configJSON = self._download_webpage(configUrl, showName, + configUrl = self._search_regex('config=(.*)$', playerUrl, u'config url') + configUrl = compat_urllib_parse.unquote(configUrl) + + configJSON = self._download_webpage(configUrl, videoId, u'Downloading configuration', u'unable to download configuration') @@ -2413,10 +2441,10 @@ class EscapistIE(InfoExtractor): 'url': videoUrl, 'uploader': showName, 'upload_date': None, - 'title': showName, + 'title': title, 'ext': 'mp4', 'thumbnail': imgUrl, - 'description': description, + 'description': videoDesc, 'player_url': playerUrl, } @@ -2501,26 +2529,17 @@ class XVideosIE(InfoExtractor): self.report_extraction(video_id) - # Extract video URL - mobj = re.search(r'flv_url=(.+?)&', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract video url') - video_url = compat_urllib_parse.unquote(mobj.group(1)) - + video_url = compat_urllib_parse.unquote(self._search_regex(r'flv_url=(.+?)&', + webpage, u'video URL')) # Extract title - mobj = re.search(r'<title>(.*?)\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'<title>(.*?)\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, @@ -2677,16 +2696,12 @@ 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('.') @@ -2857,15 +2872,10 @@ 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'] = [ @@ -2926,25 +2936,17 @@ 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) @@ -3092,20 +3094,15 @@ 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, @@ -3125,26 +3122,6 @@ 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) @@ -3157,47 +3134,31 @@ 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""" @@ -3230,7 +3191,7 @@ 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): @@ -3239,28 +3200,27 @@ 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] @@ -3408,30 +3368,21 @@ 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>[^<]+?)</title>', webpage) - if not m: - raise ExtractorError(u'Cannot find video title') - title = clean_html(m.group('title')) + title = self._html_search_regex((r"<h1 class='player_page_h1'.*?>(?P<title>.*?)</h1>", + r'<title>(?P<title>[^<]+?)</title>'), webpage, 'title', flags=re.DOTALL) - 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, flags=re.DOTALL) info = { 'id': video_id, 'url': video_url, 'ext': 'mp4', 'title': title, - 'description': desc, + 'description': video_description, } return [info] @@ -3442,6 +3393,8 @@ class SteamIE(InfoExtractor): (?P<gameID>\d+)/? (?P<videoID>\d*)(?P<extra>\??) #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): @@ -3451,11 +3404,19 @@ 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'<h2 class="pageheader">(?P<game_title>.*?)</h2>', webpage).group('game_title') - + + if re.search('<h2>Please enter your birth date to continue:</h2>', 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'<h2 class="pageheader">(.*?)</h2>', + webpage, 'game title') + urlRE = r"'movie_(?P<videoID>\d+)': \{\s*FILENAME: \"(?P<videoURL>[\w:/\.\?=]+)\"(,\s*MOVIE_NAME: \"(?P<videoName>[\w:/\.\?=\+-]+)\")?\s*\}," mweb = re.finditer(urlRE, webpage) namesRE = r'<span class="title">(?P<videoName>.+?)</span>' @@ -3487,27 +3448,29 @@ 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) + 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') + + 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, + 'title': video_title, 'uploader': uploader, - 'thumbnail': thumb, - } + 'thumbnail': thumbnail, + } return info class WorldStarHipHopIE(InfoExtractor): @@ -3515,45 +3478,36 @@ class WorldStarHipHopIE(InfoExtractor): 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>(.*)</title>", webpage_src) - - if mobj is None: - raise ExtractorError(u'Cannot determine title') - title = mobj.group(1) + video_title = self._html_search_regex(r"<title>(.*)</title>", + webpage_src, u'title') - 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.*>(.*)</span>""" 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, }] @@ -3567,10 +3521,9 @@ class RBMARadioIE(InfoExtractor): video_id = m.group('videoID') webpage = self._download_webpage(url, video_id) - m = re.search(r'<script>window.gon = {.*?};gon\.show=(.+?);</script>', 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) @@ -3617,42 +3570,33 @@ 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'<h1.*?>(?P<title>.*)</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>[^"]+)">' @@ -3676,19 +3620,18 @@ 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): @@ -3729,17 +3672,13 @@ 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, @@ -3766,10 +3705,8 @@ class YouJizzIE(InfoExtractor): webpage = self._download_webpage(url, video_id) # Get the video title - result = re.search(r'<title>(?P<title>.*)</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'<title>(?P<title>.*)</title>', + webpage, u'title').strip() # Get the embed page result = re.search(r'https?://www.youjizz.com/videos/embed/(?P<videoid>[0-9]+)', webpage) @@ -3782,10 +3719,8 @@ class YouJizzIE(InfoExtractor): webpage = self._download_webpage(embed_page_url, video_id) # Get the video URL - result = re.search(r'so.addVariable\("file",encodeURIComponent\("(?P<source>[^"]+)"\)\);', 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\("(?P<source>[^"]+)"\)\);', + webpage, u'video URL') info = {'id': video_id, 'url': video_url, @@ -3808,10 +3743,7 @@ 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)) @@ -3847,18 +3779,22 @@ 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'<meta property="og:title" content="(?P<title>.*?)"', webpage) - title = unescapeHTML(m.group('title')) - m = re.search(r'<div class="user-name-and-bio">[\S\s]+?<h2>(?P<uploader>.+?)</h2>', webpage) - uploader = clean_html(m.group('uploader')) + + video_title = self._html_search_regex(r'<meta property="og:title" content="(?P<title>.*?)"', + webpage, u'title') + + uploader = self._html_search_regex(r'<div class="user-name-and-bio">[\S\s]+?<h2>(?P<uploader>.+?)</h2>', + webpage, u'uploader', fatal=False) + info = { 'id': video_id, 'url': video_url, 'ext': 'mp4', - 'title': title, + 'title': video_title, 'thumbnail': thumbnail, 'uploader': uploader } @@ -3890,10 +3826,6 @@ 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''' @@ -3906,9 +3838,8 @@ 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*?)<h1>(\s*?)<span>(?P<playlist_title>.*?)</span>' - 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*?<h1>\s*?<span>(.*?)</span>', + webpage, 'playlist title') playlist_entries = [] for m_video, m_name in zip(m_videos,m_names): @@ -3919,27 +3850,28 @@ 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'<span id="altHeadline" >(?P<title>.*)</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 @@ -4005,10 +3937,9 @@ class SpiegelIE(InfoExtractor): video_id = m.group('videoID') webpage = self._download_webpage(url, video_id) - m = re.search(r'<div class="module-title">(.*?)</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, @@ -4044,35 +3975,25 @@ 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] @@ -4188,23 +4109,23 @@ 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 found") - 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>' - title = unescapeHTML(re.search(re_title, webpage, re.DOTALL).group('title')) + video_title = self._html_search_regex(r'<title>(?P<title>.*?)</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 }] @@ -4218,7 +4139,7 @@ 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<id>\d*?)$', @@ -4246,10 +4167,10 @@ 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] @@ -4266,17 +4187,14 @@ 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'<source src="'+'(.+)'+'" type="video/mp4">',webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract media URL') + video_url = self._html_search_regex(r'<source src="(.+?)" type="video/mp4">', + webpage, u'video URL') - video_url = mobj.group(1) - mobj = re.search('<h1 class="videoTitle slidePanelMovable">(.+)</h1>',webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = mobj.group(1) + video_title = self._html_search_regex('<h1 class="videoTitle slidePanelMovable">(.+?)</h1>', + webpage, u'title') return [{ 'id': video_id, @@ -4297,15 +4215,13 @@ class InaIE(InfoExtractor): video_extension = 'mp4' webpage = self._download_webpage(mrss_url, video_id) - mobj = re.search(r'<media:player url="(?P<mp4url>http://mp4.ina.fr/[^"]+\.mp4)', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract media URL') - video_url = mobj.group(1) + self.report_extraction(video_id) - mobj = re.search(r'<title><!\[CDATA\[(?P<titre>.*?)]]></title>', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = mobj.group(1) + video_url = self._html_search_regex(r'<media:player url="(?P<mp4url>http://mp4.ina.fr/[^"]+\.mp4)', + webpage, u'video URL') + + video_title = self._search_regex(r'<title><!\[CDATA\[(?P<titre>.*?)]]></title>', + webpage, u'title') return [{ 'id': video_id, @@ -4327,27 +4243,17 @@ class HowcastIE(InfoExtractor): self.report_extraction(video_id) - mobj = re.search(r'\'?file\'?: "(http://mobile-media\.howcast\.com/[0-9]+\.mp4)"', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract video URL') - video_url = mobj.group(1) + video_url = self._search_regex(r'\'?file\'?: "(http://mobile-media\.howcast\.com/[0-9]+\.mp4)', + webpage, u'video URL') - 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) + video_title = self._html_search_regex(r'<meta content=(?:"([^"]+)"|\'([^\']+)\') property=\'og:title\'', + webpage, u'title') - 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) + video_description = self._html_search_regex(r'<meta content=(?:"([^"]+)"|\'([^\']+)\') name=\'description\'', + webpage, u'description', fatal=False) - mobj = re.search(r'<meta content=\'(.+?)\' property=\'og:image\'', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract thumbnail') - thumbnail = mobj.group(1) + thumbnail = self._html_search_regex(r'<meta content=\'(.+?)\' property=\'og:image\'', + webpage, u'thumbnail', fatal=False) return [{ 'id': video_id, @@ -4363,7 +4269,6 @@ class VineIE(InfoExtractor): _VALID_URL = r'(?:https?://)?(?:www\.)?vine\.co/v/(?P<id>\w+)' def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') @@ -4372,25 +4277,17 @@ class VineIE(InfoExtractor): self.report_extraction(video_id) - mobj = re.search(r'<meta property="twitter:player:stream" content="(.+?)"', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract video URL') - video_url = mobj.group(1) + video_url = self._html_search_regex(r'<meta property="twitter:player:stream" content="(.+?)"', + webpage, u'video URL') - mobj = re.search(r'<meta property="og:title" content="(.+?)"', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = mobj.group(1) + video_title = self._html_search_regex(r'<meta property="og:title" content="(.+?)"', + webpage, u'title') - mobj = re.search(r'<meta property="og:image" content="(.+?)(\?.*?)?"', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract thumbnail') - thumbnail = mobj.group(1) + thumbnail = self._html_search_regex(r'<meta property="og:image" content="(.+?)(\?.*?)?"', + webpage, u'thumbnail', fatal=False) - mobj = re.search(r'<div class="user">.*?<h2>(.+?)</h2>', webpage, re.DOTALL) - if mobj is None: - raise ExtractorError(u'Unable to extract uploader') - uploader = mobj.group(1) + uploader = self._html_search_regex(r'<div class="user">.*?<h2>(.+?)</h2>', + webpage, u'uploader', fatal=False, flags=re.DOTALL) return [{ 'id': video_id, @@ -4413,18 +4310,13 @@ class FlickrIE(InfoExtractor): webpage_url = 'http://www.flickr.com/photos/' + video_uploader_id + '/' + video_id webpage = self._download_webpage(webpage_url, video_id) - mobj = re.search(r"photo_secret: '(\w+)'", webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract video secret') - secret = mobj.group(1) + 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') - mobj = re.search(r'<Item id="id">(\d+-\d+)</Item>', first_xml) - if mobj is None: - raise ExtractorError(u'Unable to extract node_id') - node_id = mobj.group(1) + node_id = self._html_search_regex(r'<Item id="id">(\d+-\d+)</Item>', + 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') @@ -4436,22 +4328,14 @@ class FlickrIE(InfoExtractor): raise ExtractorError(u'Unable to extract video url') video_url = mobj.group(1) + unescapeHTML(mobj.group(2)) - mobj = re.search(r'<meta property="og:title" content=(?:"([^"]+)"|\'([^\']+)\')', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = mobj.group(1) or mobj.group(2) + video_title = self._html_search_regex(r'<meta property="og:title" content=(?:"([^"]+)"|\'([^\']+)\')', + webpage, u'video title') - mobj = re.search(r'<meta property="og:description" content=(?:"([^"]+)"|\'([^\']+)\')', 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) + video_description = self._html_search_regex(r'<meta property="og:description" content=(?:"([^"]+)"|\'([^\']+)\')', + webpage, u'description', fatal=False) - mobj = re.search(r'<meta property="og:image" content=(?:"([^"]+)"|\'([^\']+)\')', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract thumbnail') - thumbnail = mobj.group(1) or mobj.group(2) + thumbnail = self._html_search_regex(r'<meta property="og:image" content=(?:"([^"]+)"|\'([^\']+)\')', + webpage, u'thumbnail', fatal=False) return [{ 'id': video_id, @@ -4473,32 +4357,25 @@ class TeamcocoIE(InfoExtractor): url_title = mobj.group('url_title') webpage = self._download_webpage(url, url_title) - mobj = re.search(r'<article class="video" data-id="(\d+?)"', webpage) - video_id = mobj.group(1) + video_id = self._html_search_regex(r'<article class="video" data-id="(\d+?)"', + webpage, u'video id') self.report_extraction(video_id) - mobj = re.search(r'<meta property="og:title" content="(.+?)"', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = mobj.group(1) + video_title = self._html_search_regex(r'<meta property="og:title" content="(.+?)"', + webpage, u'title') - mobj = re.search(r'<meta property="og:image" content="(.+?)"', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract thumbnail') - thumbnail = mobj.group(1) + thumbnail = self._html_search_regex(r'<meta property="og:image" content="(.+?)"', + webpage, u'thumbnail', fatal=False) - mobj = re.search(r'<meta property="og:description" content="(.*?)"', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract description') - description = mobj.group(1) + video_description = self._html_search_regex(r'<meta property="og:description" content="(.*?)"', + webpage, u'description', fatal=False) data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id data = self._download_webpage(data_url, video_id, 'Downloading data webpage') - mobj = re.search(r'<file type="high".*?>(.*?)</file>', data) - if mobj is None: - raise ExtractorError(u'Unable to extract video url') - video_url = mobj.group(1) + + video_url = self._html_search_regex(r'<file type="high".*?>(.*?)</file>', + data, u'video URL') return [{ 'id': video_id, @@ -4506,9 +4383,9 @@ class TeamcocoIE(InfoExtractor): 'ext': 'mp4', 'title': video_title, 'thumbnail': thumbnail, - 'description': description, + 'description': video_description, }] - + class XHamsterIE(InfoExtractor): """Information Extractor for xHamster""" _VALID_URL = r'(?:http://)?(?:www.)?xhamster\.com/movies/(?P<id>[0-9]+)/.*\.html' @@ -4517,8 +4394,9 @@ class XHamsterIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') - mrss_url='http://xhamster.com/movies/%s/.html' % video_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<server>[^\']*)\',\s*\'file\': \'(?P<file>[^\']+)\',', webpage) if mobj is None: raise ExtractorError(u'Unable to extract media URL') @@ -4528,39 +4406,33 @@ class XHamsterIE(InfoExtractor): video_url = mobj.group('server')+'/key='+mobj.group('file') video_extension = video_url.split('.')[-1] - mobj = re.search(r'<title>(?P<title>.+?) - xHamster\.com</title>', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = unescapeHTML(mobj.group('title')) + video_title = self._html_search_regex(r'<title>(?P<title>.+?) - xHamster\.com</title>', + webpage, u'title') - mobj = re.search(r'<span>Description: </span>(?P<description>[^<]+)', webpage) - if mobj is None: - video_description = u'' - else: - video_description = unescapeHTML(mobj.group('description')) + # Can't see the description anywhere in the UI + # video_description = self._html_search_regex(r'<span>Description: </span>(?P<description>[^<]+)', + # webpage, u'description', fatal=False) + # if video_description: video_description = unescapeHTML(video_description) mobj = re.search(r'hint=\'(?P<upload_date_Y>[0-9]{4})-(?P<upload_date_m>[0-9]{2})-(?P<upload_date_d>[0-9]{2}) [0-9]{2}:[0-9]{2}:[0-9]{2} [A-Z]{3,4}\'', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract upload date') - video_upload_date = mobj.group('upload_date_Y')+mobj.group('upload_date_m')+mobj.group('upload_date_d') - - mobj = re.search(r'<a href=\'/user/[^>]+>(?P<uploader_id>[^>]+)', webpage) - if mobj is None: - video_uploader_id = u'anonymous' + if mobj: + video_upload_date = mobj.group('upload_date_Y')+mobj.group('upload_date_m')+mobj.group('upload_date_d') else: - video_uploader_id = mobj.group('uploader_id') + video_upload_date = None + self._downloader.report_warning(u'Unable to extract upload date') - mobj = re.search(r'\'image\':\'(?P<thumbnail>[^\']+)\'', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract thumbnail URL') - video_thumbnail = mobj.group('thumbnail') + video_uploader_id = self._html_search_regex(r'<a href=\'/user/[^>]+>(?P<uploader_id>[^<]+)', + webpage, u'uploader id', default=u'anonymous') + + video_thumbnail = self._search_regex(r'\'image\':\'(?P<thumbnail>[^\']+)\'', + webpage, u'thumbnail', fatal=False) return [{ 'id': video_id, 'url': video_url, 'ext': video_extension, 'title': video_title, - 'description': video_description, + # 'description': video_description, 'upload_date': video_upload_date, 'uploader_id': video_uploader_id, 'thumbnail': video_thumbnail @@ -4584,10 +4456,9 @@ class HypemIE(InfoExtractor): cookie = urlh.headers.get('Set-Cookie', '') self.report_extraction(track_id) - mobj = re.search(r'<script type="application/json" id="displayList-data">(.*?)</script>', response, flags=re.MULTILINE|re.DOTALL) - if mobj is None: - raise ExtractorError(u'Unable to extrack tracks') - html_tracks = mobj.group(1).strip() + + html_tracks = self._html_search_regex(r'<script type="application/json" id="displayList-data">(.*?)</script>', + response, u'tracks', flags=re.MULTILINE|re.DOTALL).strip() try: track_list = json.loads(html_tracks) track = track_list[u'tracks'][0] @@ -4628,11 +4499,12 @@ class Vbox7IE(InfoExtractor): video_id = mobj.group(1) redirect_page, urlh = self._download_webpage_handle(url, video_id) - redirect_url = urlh.geturl() + re.search(r'window\.location = \'(.*)\';', redirect_page).group(1) + 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 = re.search(r'<title>(.*)</title>', webpage) - title = (title.group(1)).split('/')[0].strip() + title = self._html_search_regex(r'<title>(.*)</title>', + webpage, u'title').split('/')[0].strip() ext = "flv" info_url = "http://vbox7.com/play/magare.do" @@ -4652,6 +4524,88 @@ class Vbox7IE(InfoExtractor): 'thumbnail': thumbnail_url, }] +class GametrailersIE(InfoExtractor): + _VALID_URL = r'http://www.gametrailers.com/(?P<type>videos|reviews|full-episodes)/(?P<id>.*?)/(?P<title>.*)' + + 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\[(?P<title>.*?)\]\]></title>.* + <description><!\[CDATA\[(?P<description>.*?)\]\]></description>.* + <image>.* + <url>(?P<thumb>.*?)</url>.* + </image>''' + + 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'<src>(?P<url>.*)</src>', 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, + } + +class StatigramIE(InfoExtractor): + _VALID_URL = r'(?:http://)?(?:www\.)?statigr\.am/p/([^/]+)' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + + video_id = mobj.group(1) + webpage = self._download_webpage(url, video_id) + video_url = self._html_search_regex( + r'<meta property="og:video:secure_url" content="(.+?)">', + webpage, u'video URL') + thumbnail_url = self._html_search_regex( + r'<meta property="og:image" content="(.+?)" />', + webpage, u'thumbnail URL', fatal=False) + html_title = self._html_search_regex( + r'<title>(.+?)</title>', + webpage, u'title') + title = html_title.rpartition(u' | Statigram')[0] + uploader = self._html_search_regex( + r'@(.+) \(Videos\)', title, u'uploader name', fatal=False) + ext = 'mp4' + + return [{ + 'id': video_id, + 'url': video_url, + 'ext': ext, + 'title': title, + 'thumbnail': thumbnail_url, + 'uploader' : uploader + }] + 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. @@ -4717,6 +4671,8 @@ def gen_extractors(): XHamsterIE(), HypemIE(), Vbox7IE(), + GametrailersIE(), + StatigramIE(), GenericIE() ] |