diff options
Diffstat (limited to 'youtube_dl/InfoExtractors.py')
| -rwxr-xr-x | youtube_dl/InfoExtractors.py | 326 |
1 files changed, 176 insertions, 150 deletions
diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py index ab8bd2104..14fd644a2 100755 --- a/youtube_dl/InfoExtractors.py +++ b/youtube_dl/InfoExtractors.py @@ -15,6 +15,7 @@ import email.utils import xml.etree.ElementTree import random import math +import operator from .utils import * @@ -73,13 +74,15 @@ class InfoExtractor(object): self._ready = False self.set_downloader(downloader) - def suitable(self, url): + @classmethod + def suitable(cls, url): """Receives a URL and returns True if suitable for this IE.""" - return re.match(self._VALID_URL, url) is not None + return re.match(cls._VALID_URL, url) is not None - def working(self): + @classmethod + def working(cls): """Getter method for _WORKING.""" - return self._WORKING + return cls._WORKING def initialize(self): """Initializes an instance (authentication, etc).""" @@ -123,8 +126,14 @@ class InfoExtractor(object): def _download_webpage(self, url_or_request, video_id, note=None, errnote=None): """ Returns the data of the page as a string """ urlh = self._request_webpage(url_or_request, video_id, note, errnote) + content_type = urlh.headers.get('Content-Type', '') + m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) + if m: + encoding = m.group(1) + else: + encoding = 'utf-8' webpage_bytes = urlh.read() - return webpage_bytes.decode('utf-8', 'replace') + return webpage_bytes.decode(encoding, 'replace') class YoutubeIE(InfoExtractor): @@ -136,7 +145,6 @@ class YoutubeIE(InfoExtractor): (?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/| tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains (?:.*?\#/)? # handle anchor (#/) redirect urls - (?!view_play_list|my_playlists|artist|playlist) # ignore playlist URLs (?: # the various things that can precede the ID: (?:(?:v|embed|e)/) # v/ or embed/ or e/ |(?: # or the v= param in all its forms @@ -188,9 +196,11 @@ class YoutubeIE(InfoExtractor): } IE_NAME = u'youtube' - def suitable(self, url): + @classmethod + def suitable(cls, url): """Receives a URL and returns True if suitable for this IE.""" - return re.match(self._VALID_URL, url, re.VERBOSE) is not None + if YoutubePlaylistIE.suitable(url): return False + return re.match(cls._VALID_URL, url, re.VERBOSE) is not None def report_lang(self): """Report attempt to set language.""" @@ -321,7 +331,7 @@ class YoutubeIE(InfoExtractor): else: raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) except (IOError, netrc.NetrcParseError) as err: - self._downloader.to_stderr(u'WARNING: parsing .netrc: %s' % compat_str(err)) + self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err)) return # Set language @@ -330,7 +340,7 @@ class YoutubeIE(InfoExtractor): self.report_lang() compat_urllib_request.urlopen(request).read() except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.to_stderr(u'WARNING: unable to set language: %s' % compat_str(err)) + self._downloader.report_warning(u'unable to set language: %s' % compat_str(err)) return # No authentication to be performed @@ -341,7 +351,7 @@ class YoutubeIE(InfoExtractor): try: login_page = compat_urllib_request.urlopen(request).read().decode('utf-8') except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.to_stderr(u'WARNING: unable to fetch login page: %s' % compat_str(err)) + self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err)) return galx = None @@ -385,10 +395,10 @@ class YoutubeIE(InfoExtractor): self.report_login() login_results = compat_urllib_request.urlopen(request).read().decode('utf-8') if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: - self._downloader.to_stderr(u'WARNING: unable to log in: bad username or password') + self._downloader.report_warning(u'unable to log in: bad username or password') return except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.to_stderr(u'WARNING: unable to log in: %s' % compat_str(err)) + self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) return # Confirm age @@ -1308,7 +1318,8 @@ class GenericIE(InfoExtractor): def report_download_webpage(self, video_id): """Report webpage download.""" - self._downloader.to_screen(u'WARNING: Falling back on generic information extractor.') + if not self._downloader.params.get('test', False): + self._downloader.to_screen(u'WARNING: Falling back on generic information extractor.') self._downloader.to_screen(u'[generic] %s: Downloading webpage' % video_id) def report_extraction(self, video_id): @@ -1378,13 +1389,8 @@ class GenericIE(InfoExtractor): if self._test_redirect(url): return video_id = url.split('/')[-1] - request = compat_urllib_request.Request(url) try: - self.report_download_webpage(video_id) - webpage = compat_urllib_request.urlopen(request).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % compat_str(err)) - return + webpage = self._download_webpage(url, video_id) except ValueError as err: # since this is the last-resort InfoExtractor, if # this error is thrown, it'll be thrown here @@ -1483,7 +1489,7 @@ class YoutubeSearchIE(InfoExtractor): self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query)) return elif n > self._max_youtube_results: - self._downloader.to_stderr(u'WARNING: ytsearch returns max %i results (you requested %i)' % (self._max_youtube_results, n)) + self._downloader.report_warning(u'ytsearch returns max %i results (you requested %i)' % (self._max_youtube_results, n)) n = self._max_youtube_results self._download_n_results(query, n) return @@ -1503,12 +1509,16 @@ class YoutubeSearchIE(InfoExtractor): result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) request = compat_urllib_request.Request(result_url) try: - data = compat_urllib_request.urlopen(request).read() + data = compat_urllib_request.urlopen(request).read().decode('utf-8') except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: self._downloader.trouble(u'ERROR: unable to download API page: %s' % compat_str(err)) return api_response = json.loads(data)['data'] + if not 'items' in api_response: + self._downloader.trouble(u'[youtube] No video results') + return + new_ids = list(video['id'] for video in api_response['items']) video_ids += new_ids @@ -1561,7 +1571,7 @@ class GoogleSearchIE(InfoExtractor): self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query)) return elif n > self._max_google_results: - self._downloader.to_stderr(u'WARNING: gvsearch returns max %i results (you requested %i)' % (self._max_google_results, n)) + self._downloader.report_warning(u'gvsearch returns max %i results (you requested %i)' % (self._max_google_results, n)) n = self._max_google_results self._download_n_results(query, n) return @@ -1645,7 +1655,7 @@ class YahooSearchIE(InfoExtractor): self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query)) return elif n > self._max_yahoo_results: - self._downloader.to_stderr(u'WARNING: yvsearch returns max %i results (you requested %i)' % (self._max_yahoo_results, n)) + self._downloader.report_warning(u'yvsearch returns max %i results (you requested %i)' % (self._max_yahoo_results, n)) n = self._max_yahoo_results self._download_n_results(query, n) return @@ -1693,80 +1703,94 @@ class YahooSearchIE(InfoExtractor): class YoutubePlaylistIE(InfoExtractor): """Information Extractor for YouTube playlists.""" - _VALID_URL = r'(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:(?:course|view_play_list|my_playlists|artist|playlist)\?.*?(p|a|list)=|user/.*?/user/|p/|user/.*?#[pg]/c/)(?:PL|EC)?|PL|EC)([0-9A-Za-z-_]{10,})(?:/.*?/([0-9A-Za-z_-]+))?.*' - _TEMPLATE_URL = 'http://www.youtube.com/%s?%s=%s&page=%s&gl=US&hl=en' - _VIDEO_INDICATOR_TEMPLATE = r'/watch\?v=(.+?)&([^&"]+&)*list=.*?%s' - _MORE_PAGES_INDICATOR = u"Next \N{RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK}" + _VALID_URL = r"""(?: + (?:https?://)? + (?:\w+\.)? + youtube\.com/ + (?: + (?:course|view_play_list|my_playlists|artist|playlist|watch) + \? (?:.*?&)*? (?:p|a|list)= + | user/.*?/user/ + | p/ + | user/.*?#[pg]/c/ + ) + ((?:PL|EC|UU)?[0-9A-Za-z-_]{10,}) + .* + | + ((?: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' + _MAX_RESULTS = 50 IE_NAME = u'youtube:playlist' def __init__(self, downloader=None): InfoExtractor.__init__(self, downloader) + @classmethod + def suitable(cls, url): + """Receives a URL and returns True if suitable for this IE.""" + return re.match(cls._VALID_URL, url, re.VERBOSE) is not None + def report_download_page(self, playlist_id, pagenum): """Report attempt to download playlist page with given number.""" self._downloader.to_screen(u'[youtube] PL %s: Downloading page #%s' % (playlist_id, pagenum)) def _real_extract(self, url): # Extract playlist id - mobj = re.match(self._VALID_URL, url) + mobj = re.match(self._VALID_URL, url, re.VERBOSE) if mobj is None: self._downloader.trouble(u'ERROR: invalid url: %s' % url) return - # Single video case - if mobj.group(3) is not None: - self._downloader.download([mobj.group(3)]) - return - - # Download playlist pages - # prefix is 'p' as default for playlists but there are other types that need extra care - playlist_prefix = mobj.group(1) - if playlist_prefix == 'a': - playlist_access = 'artist' - else: - playlist_prefix = 'p' - playlist_access = 'view_play_list' - playlist_id = mobj.group(2) - video_ids = [] - pagenum = 1 + # Download playlist videos from API + playlist_id = mobj.group(1) or mobj.group(2) + page_num = 1 + videos = [] while True: - self.report_download_page(playlist_id, pagenum) - url = self._TEMPLATE_URL % (playlist_access, playlist_prefix, playlist_id, pagenum) - request = compat_urllib_request.Request(url) + self.report_download_page(playlist_id, page_num) + + url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1) try: - page = compat_urllib_request.urlopen(request).read().decode('utf-8') + page = compat_urllib_request.urlopen(url).read().decode('utf8') except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: self._downloader.trouble(u'ERROR: unable to download webpage: %s' % compat_str(err)) return - # Extract video identifiers - ids_in_page = [] - for mobj in re.finditer(self._VIDEO_INDICATOR_TEMPLATE % playlist_id, page): - if mobj.group(1) not in ids_in_page: - ids_in_page.append(mobj.group(1)) - video_ids.extend(ids_in_page) + try: + response = json.loads(page) + except ValueError as err: + self._downloader.trouble(u'ERROR: Invalid JSON in API response: ' + compat_str(err)) + return - if self._MORE_PAGES_INDICATOR not in page: + if not 'feed' in response or not 'entry' in response['feed']: + self._downloader.trouble(u'ERROR: Got a malformed response from YouTube API') + return + videos += [ (entry['yt$position']['$t'], entry['content']['src']) + for entry in response['feed']['entry'] + if 'content' in entry ] + + if len(response['feed']['entry']) < self._MAX_RESULTS: break - pagenum = pagenum + 1 + page_num += 1 - total = len(video_ids) + videos = [v[1] for v in sorted(videos)] + total = len(videos) playliststart = self._downloader.params.get('playliststart', 1) - 1 playlistend = self._downloader.params.get('playlistend', -1) if playlistend == -1: - video_ids = video_ids[playliststart:] + videos = videos[playliststart:] else: - video_ids = video_ids[playliststart:playlistend] + videos = videos[playliststart:playlistend] - if len(video_ids) == total: + if len(videos) == total: self._downloader.to_screen(u'[youtube] PL %s: Found %i videos' % (playlist_id, total)) else: - self._downloader.to_screen(u'[youtube] PL %s: Found %i videos, downloading %i' % (playlist_id, total, len(video_ids))) + self._downloader.to_screen(u'[youtube] PL %s: Found %i videos, downloading %i' % (playlist_id, total, len(videos))) - for id in video_ids: - self._downloader.download(['http://www.youtube.com/watch?v=%s' % id]) + for video in videos: + self._downloader.download([video]) return @@ -1952,9 +1976,8 @@ class BlipTVUserIE(InfoExtractor): while True: self.report_download_page(username, pagenum) - - request = compat_urllib_request.Request( page_base + "&page=" + str(pagenum) ) - + url = page_base + "&page=" + str(pagenum) + request = compat_urllib_request.Request( url ) try: page = compat_urllib_request.urlopen(request).read().decode('utf-8') except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: @@ -2090,7 +2113,7 @@ class FacebookIE(InfoExtractor): else: raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) except (IOError, netrc.NetrcParseError) as err: - self._downloader.to_stderr(u'WARNING: parsing .netrc: %s' % compat_str(err)) + self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err)) return if useremail is None: @@ -2107,10 +2130,10 @@ class FacebookIE(InfoExtractor): self.report_login() login_results = compat_urllib_request.urlopen(request).read() if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None: - self._downloader.to_stderr(u'WARNING: unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.') + self._downloader.report_warning(u'unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.') return except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.to_stderr(u'WARNING: unable to log in: %s' % compat_str(err)) + self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) return def _real_extract(self, url): @@ -2175,6 +2198,17 @@ class BlipTVIE(InfoExtractor): self._downloader.trouble(u'ERROR: invalid URL: %s' % url) return + urlp = compat_urllib_parse_urlparse(url) + if urlp.path.startswith('/play/'): + request = compat_urllib_request.Request(url) + response = compat_urllib_request.urlopen(request) + redirecturl = response.geturl() + rurlp = compat_urllib_parse_urlparse(redirecturl) + file_id = compat_parse_qs(rurlp.fragment)['file'][0].rpartition('/')[2] + url = 'http://blip.tv/a/a-' + file_id + return self._real_extract(url) + + if '?' in url: cchar = '&' else: @@ -2330,9 +2364,10 @@ class ComedyCentralIE(InfoExtractor): '400': '384x216', } - def suitable(self, url): + @classmethod + def suitable(cls, url): """Receives a URL and returns True if suitable for this IE.""" - return re.match(self._VALID_URL, url, re.VERBOSE) is not None + return re.match(cls._VALID_URL, url, re.VERBOSE) is not None def report_extraction(self, episode_id): self._downloader.to_screen(u'[comedycentral] %s: Extracting information' % episode_id) @@ -2555,7 +2590,7 @@ class EscapistIE(InfoExtractor): 'uploader': showName, 'upload_date': None, 'title': showName, - 'ext': 'flv', + 'ext': 'mp4', 'thumbnail': imgUrl, 'description': description, 'player_url': playerUrl, @@ -3589,65 +3624,17 @@ class FunnyOrDieIE(InfoExtractor): } return [info] -class TweetReelIE(InfoExtractor): - _VALID_URL = r'^(?:https?://)?(?:www\.)?tweetreel\.com/[?](?P<id>[0-9a-z]+)$' - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - if mobj is None: - self._downloader.trouble(u'ERROR: invalid URL: %s' % url) - return - - video_id = mobj.group('id') - webpage = self._download_webpage(url, video_id) - - m = re.search(r'<div id="left" status_id="([0-9]+)">', webpage) - if not m: - self._downloader.trouble(u'ERROR: Cannot find status ID') - status_id = m.group(1) - - m = re.search(r'<div class="tweet_text">(.*?)</div>', webpage, flags=re.DOTALL) - if not m: - self._downloader.trouble(u'WARNING: Cannot find description') - desc = unescapeHTML(re.sub('<a.*?</a>', '', m.group(1))).strip() - - m = re.search(r'<div class="tweet_info">.*?from <a target="_blank" href="https?://twitter.com/(?P<uploader_id>.+?)">(?P<uploader>.+?)</a>', webpage, flags=re.DOTALL) - if not m: - self._downloader.trouble(u'ERROR: Cannot find uploader') - uploader = unescapeHTML(m.group('uploader')) - uploader_id = unescapeHTML(m.group('uploader_id')) - - m = re.search(r'<span unixtime="([0-9]+)"', webpage) - if not m: - self._downloader.trouble(u'ERROR: Cannot find upload date') - upload_date = datetime.datetime.fromtimestamp(int(m.group(1))).strftime('%Y%m%d') - - title = desc - video_url = 'http://files.tweetreel.com/video/' + status_id + '.mov' - - info = { - 'id': video_id, - 'url': video_url, - 'ext': 'mov', - 'title': title, - 'description': desc, - 'uploader': uploader, - 'uploader_id': uploader_id, - 'internal_id': status_id, - 'upload_date': upload_date - } - return [info] - class SteamIE(InfoExtractor): - _VALID_URL = r"""http://store.steampowered.com/ + _VALID_URL = r"""http://store.steampowered.com/ (?P<urltype>video|app)/ #If the page is only for videos or for a game (?P<gameID>\d+)/? (?P<videoID>\d*)(?P<extra>\??) #For urltype == video we sometimes get the videoID """ - def suitable(self, url): + @classmethod + def suitable(cls, url): """Receives a URL and returns True if suitable for this IE.""" - return re.match(self._VALID_URL, url, re.VERBOSE) is not None + return re.match(cls._VALID_URL, url, re.VERBOSE) is not None def _real_extract(self, url): m = re.match(self._VALID_URL, url, re.VERBOSE) @@ -3658,18 +3645,22 @@ class SteamIE(InfoExtractor): mweb = re.finditer(urlRE, webpage) namesRE = r'<span class="title">(?P<videoName>.+?)</span>' titles = re.finditer(namesRE, webpage) + thumbsRE = r'<img class="movie_thumb" src="(?P<thumbnail>.+?)">' + thumbs = re.finditer(thumbsRE, webpage) videos = [] - for vid,vtitle in zip(mweb,titles): + for vid,vtitle,thumb in zip(mweb,titles,thumbs): video_id = vid.group('videoID') title = vtitle.group('videoName') video_url = vid.group('videoURL') + video_thumb = thumb.group('thumbnail') if not video_url: self._downloader.trouble(u'ERROR: Cannot find video url for %s' % video_id) info = { 'id':video_id, 'url':video_url, 'ext': 'flv', - 'title': unescapeHTML(title) + 'title': unescapeHTML(title), + 'thumbnail': video_thumb } videos.append(info) return videos @@ -3735,7 +3726,7 @@ class RBMARadioIE(InfoExtractor): class YouPornIE(InfoExtractor): """Information extractor for youporn.com.""" _VALID_URL = r'^(?:https?://)?(?:\w+\.)?youporn\.com/watch/(?P<videoid>[0-9]+)/(?P<title>[^/]+)' - + def _print_formats(self, formats): """Print all available formats""" print(u'Available formats:') @@ -3771,7 +3762,7 @@ class YouPornIE(InfoExtractor): # Get the video date result = re.search(r'Date:</label>(?P<date>.*) </li>', webpage) if result is None: - self._downloader.to_stderr(u'WARNING: unable to extract video date') + self._downloader.report_warning(u'unable to extract video date') upload_date = None else: upload_date = result.group('date').strip() @@ -3779,7 +3770,7 @@ class YouPornIE(InfoExtractor): # Get the video uploader result = re.search(r'Submitted:</label>(?P<uploader>.*)</li>', webpage) if result is None: - self._downloader.to_stderr(u'WARNING: unable to extract uploader') + self._downloader.report_warning(u'unable to extract uploader') video_uploader = None else: video_uploader = result.group('uploader').strip() @@ -3797,8 +3788,8 @@ class YouPornIE(InfoExtractor): links = re.findall(LINK_RE, download_list_html) if(len(links) == 0): raise ExtractorError(u'ERROR: no known formats available for video') - - self._downloader.to_screen(u'[youporn] Links found: %d' % len(links)) + + self._downloader.to_screen(u'[youporn] Links found: %d' % len(links)) formats = [] for link in links: @@ -3849,7 +3840,7 @@ class YouPornIE(InfoExtractor): return return [format] - + class PornotubeIE(InfoExtractor): """Information extractor for pornotube.com.""" @@ -3921,7 +3912,7 @@ class YouJizzIE(InfoExtractor): embed_page_url = result.group(0).strip() video_id = result.group('videoid') - + webpage = self._download_webpage(embed_page_url, video_id) # Get the video URL @@ -4017,9 +4008,10 @@ class TEDIE(InfoExtractor): /(?P<name>\w+) # Here goes the name and then ".html" ''' - def suitable(self, url): + @classmethod + def suitable(cls, url): """Receives a URL and returns True if suitable for this IE.""" - return re.match(self._VALID_URL, url, re.VERBOSE) is not None + return re.match(cls._VALID_URL, url, re.VERBOSE) is not None def _real_extract(self, url): m=re.match(self._VALID_URL, url, re.VERBOSE) @@ -4042,31 +4034,30 @@ class TEDIE(InfoExtractor): ([.\s]*?)data-playlist_item_id="(\d+)" ([.\s]*?)data-mediaslug="(?P<mediaSlug>.+?)" ''' - video_name_RE=r'<p\ class="talk-title"><a href="/talks/(.+).html">(?P<fullname>.+?)</a></p>' + video_name_RE=r'<p\ class="talk-title"><a href="(?P<talk_url>/talks/(.+).html)">(?P<fullname>.+?)</a></p>' webpage=self._download_webpage(url, playlist_id, 'Downloading playlist webpage') m_videos=re.finditer(video_RE,webpage,re.VERBOSE) m_names=re.finditer(video_name_RE,webpage) info=[] for m_video, m_name in zip(m_videos,m_names): - video_dic={ - 'id': m_video.group('video_id'), - 'url': self._talk_video_link(m_video.group('mediaSlug')), - 'ext': 'mp4', - 'title': m_name.group('fullname') - } - info.append(video_dic) + video_id=m_video.group('video_id') + talk_url='http://www.ted.com%s' % m_name.group('talk_url') + info.append(self._talk_info(talk_url,video_id)) return info + 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) # If the url includes the language we get the title translated - title_RE=r'<h1><span id="altHeadline" >(?P<title>[\s\w:/\.\?=\+-\\\']*)</span></h1>' + title_RE=r'<h1><span id="altHeadline" >(?P<title>.*)</span></h1>' 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') @@ -4075,13 +4066,14 @@ class TEDIE(InfoExtractor): 'id': video_id, 'url': video_url, 'ext': 'mp4', - 'title': title + 'title': title, + 'thumbnail': thumb_match.group('thumbnail') } return info class MySpassIE(InfoExtractor): _VALID_URL = r'http://www.myspass.de/.*' - + def _real_extract(self, url): META_DATA_URL_TEMPLATE = 'http://www.myspass.de/myspass/includes/apps/video/getvideometadataxml.php?id=%s' @@ -4091,12 +4083,12 @@ class MySpassIE(InfoExtractor): url_parent_path, video_id = os.path.split(url_path) if not video_id: _, video_id = os.path.split(url_parent_path) - + # get metadata metadata_url = META_DATA_URL_TEMPLATE % video_id metadata_text = self._download_webpage(metadata_url, video_id) metadata = xml.etree.ElementTree.fromstring(metadata_text.encode('utf-8')) - + # extract values from metadata url_flv_el = metadata.find('url_flv') if url_flv_el is None: @@ -4135,6 +4127,40 @@ class MySpassIE(InfoExtractor): } return [info] +class SpiegelIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<videoID>[0-9]+)(?:\.html)?$' + + def _real_extract(self, url): + m = re.match(self._VALID_URL, url) + 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)) + + xml_url = u'http://video2.spiegel.de/flash/' + video_id + u'.xml' + xml_code = self._download_webpage(xml_url, video_id, + note=u'Downloading XML', errnote=u'Failed to download XML') + + idoc = xml.etree.ElementTree.fromstring(xml_code) + last_type = idoc[-1] + filename = last_type.findall('./filename')[0].text + duration = float(last_type.findall('./duration')[0].text) + + video_url = 'http://video2.spiegel.de/flash/' + filename + video_ext = filename.rpartition('.')[2] + info = { + 'id': video_id, + 'url': video_url, + 'ext': video_ext, + 'title': video_title, + 'duration': duration, + } + return [info] + + 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. @@ -4176,7 +4202,6 @@ def gen_extractors(): NBAIE(), JustinTVIE(), FunnyOrDieIE(), - TweetReelIE(), SteamIE(), UstreamIE(), RBMARadioIE(), @@ -4184,6 +4209,7 @@ def gen_extractors(): KeekIE(), TEDIE(), MySpassIE(), + SpiegelIE(), GenericIE() ] |