X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2FInfoExtractors.py;h=8b2442bacc21e86db7ffdc920116b51e79fd3820;hb=5e34d2ebbf9906bded4201d7bd8bb82e9353de9f;hp=3450f0d17e19d95d67645a50c79f0680b05379cc;hpb=3820df0106d6065f50cc1eb90823906410dc9543;p=youtube-dl diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py index 3450f0d17..8b2442bac 100755 --- a/youtube_dl/InfoExtractors.py +++ b/youtube_dl/InfoExtractors.py @@ -562,12 +562,7 @@ class YoutubeIE(InfoExtractor): mobj = re.search(r'id="eow-date.*?>(.*?)', video_webpage, re.DOTALL) if mobj is not None: upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) - format_expressions = ['%d %B %Y', '%B %d %Y', '%b %d %Y'] - for expression in format_expressions: - try: - upload_date = datetime.datetime.strptime(upload_date, expression).strftime('%Y%m%d') - except: - pass + upload_date = unified_strdate(upload_date) # description video_description = get_element_by_id("eow-description", video_webpage) @@ -784,13 +779,11 @@ class MetacafeIE(InfoExtractor): 'ext': video_extension.decode('utf-8'), }] - class DailymotionIE(InfoExtractor): """Information Extractor for Dailymotion""" _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/video/([^/]+)' IE_NAME = u'dailymotion' - _WORKING = False def _real_extract(self, url): # Extract id and simplified title from URL @@ -1075,13 +1068,7 @@ class VimeoIE(InfoExtractor): # Retrieve video webpage to extract further information request = compat_urllib_request.Request(url, None, std_headers) - try: - self.report_download_webpage(video_id) - webpage_bytes = compat_urllib_request.urlopen(request).read() - webpage = webpage_bytes.decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_error(u'Unable to retrieve video webpage: %s' % compat_str(err)) - return + webpage = self._download_webpage(request, video_id) # Now we begin extracting as much information as we can from what we # retrieved. First we extract the information common to all extractors, @@ -1688,10 +1675,6 @@ class YoutubePlaylistIE(InfoExtractor): """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, re.VERBOSE) @@ -1705,14 +1688,8 @@ class YoutubePlaylistIE(InfoExtractor): videos = [] while True: - 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(url).read().decode('utf8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_error(u'unable to download webpage: %s' % compat_str(err)) - return + page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) try: response = json.loads(page) @@ -1723,12 +1700,11 @@ class YoutubePlaylistIE(InfoExtractor): if 'feed' not in response: self._downloader.report_error(u'Got a malformed response from YouTube API') return + playlist_title = response['feed']['title']['$t'] if 'entry' not in response['feed']: # Number of videos is a multiple of self._MAX_RESULTS break - playlist_title = response['feed']['title']['$t'] - videos += [ (entry['yt$position']['$t'], entry['content']['src']) for entry in response['feed']['entry'] if 'content' in entry ] @@ -1752,10 +1728,6 @@ class YoutubeChannelIE(InfoExtractor): _MORE_PAGES_URL = 'http://www.youtube.com/channel_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' IE_NAME = u'youtube:channel' - def report_download_page(self, channel_id, pagenum): - """Report attempt to download channel page with given number.""" - self._downloader.to_screen(u'[youtube] Channel %s: Downloading page #%s' % (channel_id, pagenum)) - def extract_videos_from_page(self, page): ids_in_page = [] for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): @@ -1775,14 +1747,9 @@ class YoutubeChannelIE(InfoExtractor): video_ids = [] pagenum = 1 - self.report_download_page(channel_id, pagenum) url = self._TEMPLATE_URL % (channel_id, pagenum) - request = compat_urllib_request.Request(url) - try: - page = compat_urllib_request.urlopen(request).read().decode('utf8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_error(u'unable to download webpage: %s' % compat_str(err)) - return + page = self._download_webpage(url, channel_id, + u'Downloading page #%s' % pagenum) # Extract video identifiers ids_in_page = self.extract_videos_from_page(page) @@ -1793,14 +1760,9 @@ class YoutubeChannelIE(InfoExtractor): while True: pagenum = pagenum + 1 - self.report_download_page(channel_id, pagenum) url = self._MORE_PAGES_URL % (pagenum, channel_id) - request = compat_urllib_request.Request(url) - try: - page = compat_urllib_request.urlopen(request).read().decode('utf8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_error(u'unable to download webpage: %s' % compat_str(err)) - return + page = self._download_webpage(url, channel_id, + u'Downloading page #%s' % pagenum) page = json.loads(page) @@ -1827,11 +1789,6 @@ class YoutubeUserIE(InfoExtractor): _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]' IE_NAME = u'youtube:user' - def report_download_page(self, username, start_index): - """Report attempt to download user page.""" - self._downloader.to_screen(u'[youtube] user %s: Downloading video ids from %d to %d' % - (username, start_index, start_index + self._GDATA_PAGE_SIZE)) - def _real_extract(self, url): # Extract username mobj = re.match(self._VALID_URL, url) @@ -1851,15 +1808,10 @@ class YoutubeUserIE(InfoExtractor): while True: start_index = pagenum * self._GDATA_PAGE_SIZE + 1 - self.report_download_page(username, start_index) - request = compat_urllib_request.Request(self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)) - - try: - 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.report_error(u'unable to download webpage: %s' % compat_str(err)) - return + gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) + page = self._download_webpage(gdata_url, username, + u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE)) # Extract video identifiers ids_in_page = [] @@ -1893,11 +1845,6 @@ class BlipTVUserIE(InfoExtractor): _PAGE_SIZE = 12 IE_NAME = u'blip.tv:user' - def report_download_page(self, username, pagenum): - """Report attempt to download user page.""" - self.to_screen(u'user %s: Downloading video ids from page %d' % - (username, pagenum)) - def _real_extract(self, url): # Extract username mobj = re.match(self._VALID_URL, url) @@ -1909,15 +1856,9 @@ class BlipTVUserIE(InfoExtractor): page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1' - request = compat_urllib_request.Request(url) - - try: - page = compat_urllib_request.urlopen(request).read().decode('utf-8') - mobj = re.search(r'data-users-id="([^"]+)"', page) - page_base = page_base % mobj.group(1) - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_error(u'unable to download webpage: %s' % compat_str(err)) - return + page = self._download_webpage(url, username, u'Downloading user page') + mobj = re.search(r'data-users-id="([^"]+)"', page) + page_base = page_base % mobj.group(1) # Download video ids using BlipTV Ajax calls. Result size per @@ -1929,14 +1870,9 @@ class BlipTVUserIE(InfoExtractor): pagenum = 1 while True: - self.report_download_page(username, 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: - self._downloader.report_error(u'unable to download webpage: %s' % str(err)) - return + page = self._download_webpage(url, username, + u'Downloading video ids from page %d' % pagenum) # Extract video identifiers ids_in_page = [] @@ -2295,12 +2231,6 @@ class ComedyCentralIE(InfoExtractor): """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_config_download(self, episode_id, media_id): - self.to_screen(u'%s: Downloading configuration for %s' % (episode_id, media_id)) - - def report_index_download(self, episode_id): - self.to_screen(u'%s: Downloading show index' % episode_id) - def _print_formats(self, formats): print('Available formats:') for x in formats: @@ -2334,15 +2264,8 @@ class ComedyCentralIE(InfoExtractor): else: epTitle = mobj.group('episode') - req = compat_urllib_request.Request(url) self.report_extraction(epTitle) - try: - htmlHandle = compat_urllib_request.urlopen(req) - html = htmlHandle.read() - webpage = html.decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_error(u'unable to download webpage: %s' % compat_str(err)) - return + webpage = self._download_webpage(url, epTitle) if dlNewest: url = htmlHandle.geturl() mobj = re.match(self._VALID_URL, url, re.VERBOSE) @@ -2370,12 +2293,9 @@ class ComedyCentralIE(InfoExtractor): uri = mMovieParams[0][1] indexUrl = 'http://shadow.comedycentral.com/feeds/video_player/mrss/?' + compat_urllib_parse.urlencode({'uri': uri}) - self.report_index_download(epTitle) - try: - indexXml = compat_urllib_request.urlopen(indexUrl).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_error(u'unable to download episode index: ' + compat_str(err)) - return + indexXml = self._download_webpage(indexUrl, epTitle, + u'Downloading show index', + u'unable to download episode index') results = [] @@ -2386,17 +2306,12 @@ class ComedyCentralIE(InfoExtractor): shortMediaId = mediaId.split(':')[-1] showId = mediaId.split(':')[-2].replace('.com', '') officialTitle = itemEl.findall('./title')[0].text - officialDate = itemEl.findall('./pubDate')[0].text + officialDate = unified_strdate(itemEl.findall('./pubDate')[0].text) configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' + compat_urllib_parse.urlencode({'uri': mediaId})) - configReq = compat_urllib_request.Request(configUrl) - self.report_config_download(epTitle, shortMediaId) - try: - configXml = compat_urllib_request.urlopen(configReq).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_error(u'unable to download webpage: %s' % compat_str(err)) - return + configXml = self._download_webpage(configUrl, epTitle, + u'Downloading configuration for %s' % shortMediaId) cdoc = xml.etree.ElementTree.fromstring(configXml) turls = [] @@ -2453,9 +2368,6 @@ class EscapistIE(InfoExtractor): _VALID_URL = r'^(https?://)?(www\.)?escapistmagazine\.com/videos/view/(?P[^/]+)/(?P[^/?]+)[/?]?.*$' IE_NAME = u'escapist' - def report_config_download(self, showName): - self.to_screen(u'%s: Downloading configuration' % showName) - def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) if mobj is None: @@ -2465,14 +2377,7 @@ class EscapistIE(InfoExtractor): videoId = mobj.group('episode') self.report_extraction(showName) - try: - webPage = compat_urllib_request.urlopen(url) - webPageBytes = webPage.read() - m = re.match(r'text/html; charset="?([^"]+)"?', webPage.headers['Content-Type']) - webPage = webPageBytes.decode(m.group(1) if m else 'utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_error(u'unable to download webpage: ' + compat_str(err)) - return + webPage = self._download_webpage(url, showName) descMatch = re.search('video|app)/ #If the page is only for videos or for a game (?P\d+)/? (?P\d*)(?P\??) #For urltype == video we sometimes get the videoID @@ -3630,16 +3485,15 @@ class WorldStarHipHopIE(InfoExtractor): IE_NAME = u'WorldStarHipHop' def _real_extract(self, url): - _src_url = r"""(http://hw-videos.*(?:mp4|flv))""" - - webpage_src = compat_urllib_request.urlopen(url).read() - webpage_src = webpage_src.decode('utf-8') - - mobj = re.search(_src_url, webpage_src) + _src_url = r"""(http://(hw-videos|hw-post1).*(?:mp4|flv))""" m = re.match(self._VALID_URL, url) video_id = m.group('id') + webpage_src = self._download_webpage(url, video_id) + + mobj = re.search(_src_url, webpage_src) + if mobj is not None: video_url = mobj.group() if 'mp4' in video_url: @@ -3759,7 +3613,7 @@ class YouPornIE(InfoExtractor): self._downloader.report_warning(u'unable to extract video date') upload_date = None else: - upload_date = result.group('date').strip() + upload_date = unified_strdate(result.group('date').strip()) # Get the video uploader result = re.search(r'Submitted:(?P.*)', webpage) @@ -3866,7 +3720,7 @@ class PornotubeIE(InfoExtractor): if result is None: self._downloader.report_error(u'unable to extract video title') return - upload_date = result.group('date') + upload_date = unified_strdate(result.group('date')) info = {'id': video_id, 'url': video_url, @@ -3993,12 +3847,13 @@ class KeekIE(InfoExtractor): return [info] class TEDIE(InfoExtractor): - _VALID_URL=r'''http://www.ted.com/ + _VALID_URL=r'''http://www\.ted\.com/ ( ((?Pplaylists)/(?P\d+)) # We have a playlist | ((?Ptalks)) # We have a simple talk ) + (/lang/(.*?))? # The url may contain the language /(?P\w+) # Here goes the name and then ".html" ''' @@ -4249,7 +4104,7 @@ class ARDIE(InfoExtractor): return [info] class TumblrIE(InfoExtractor): - _VALID_URL = r'http://(?P.*?).tumblr.com/((post)|(video))/(?P\d*)/(.*?)' + _VALID_URL = r'http://(?P.*?)\.tumblr\.com/((post)|(video))/(?P\d*)/(.*?)' def _real_extract(self, url): m_url = re.match(self._VALID_URL, url) @@ -4259,7 +4114,7 @@ class TumblrIE(InfoExtractor): url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id) webpage = self._download_webpage(url, video_id) - re_video = r'src=\\x22(?Phttp://%s.tumblr.com/video_file/%s/(.*?))\\x22 type=\\x22video/(?P.*?)\\x22' % (blog, video_id) + re_video = r'src=\\x22(?Phttp://%s\.tumblr\.com/video_file/%s/(.*?))\\x22 type=\\x22video/(?P.*?)\\x22' % (blog, video_id) video = re.search(re_video, webpage) if video is None: self.to_screen("No video founded") @@ -4272,8 +4127,8 @@ class TumblrIE(InfoExtractor): # 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'(.*?) - (?P<title>.*?)' - title = unescapeHTML(re.search(re_title, webpage).group('title')) + re_title = r'(?P<title>.*?)' + title = unescapeHTML(re.search(re_title, webpage, re.DOTALL).group('title')) return [{'id': video_id, 'url': video_url, @@ -4282,6 +4137,85 @@ class TumblrIE(InfoExtractor): 'ext': ext }] +class BandcampIE(InfoExtractor): + _VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P.*)' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + title = mobj.group('title') + webpage = self._download_webpage(url, title) + # We get the link to the free download page + m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage) + if m_download is None: + self._downloader.report_error('No free songs founded') + return + download_link = m_download.group(1) + id = re.search(r'var TralbumData = {(.*?)id: (?P<id>\d*?)$', + webpage, re.MULTILINE|re.DOTALL).group('id') + + download_webpage = self._download_webpage(download_link, id, + 'Downloading free downloads page') + # We get the dictionary of the track from some javascrip code + info = re.search(r'items: (.*?),$', + download_webpage, re.MULTILINE).group(1) + info = json.loads(info)[0] + # We pick mp3-320 for now, until format selection can be easily implemented. + mp3_info = info[u'downloads'][u'mp3-320'] + # If we try to use this url it says the link has expired + initial_url = mp3_info[u'url'] + re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$' + m_url = re.match(re_url, initial_url) + #We build the url we will use to get the final track url + # This url is build in Bandcamp in the script download_bunde_*.js + request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), id, m_url.group('ts')) + final_url_webpage = self._download_webpage(request_url, id, 'Requesting download url') + # If we could correctly generate the .rand field the url would be + #in the "download_url" key + final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1) + + track_info = {'id':id, + 'title' : info[u'title'], + 'ext' : 'mp3', + 'url' : final_url, + 'thumbnail' : info[u'thumb_url'], + 'uploader' : info[u'artist'] + } + + return [track_info] + +class RedtubeIE(InfoExtractor): + """Information Extractor for redtube""" + _VALID_URL = r'(?:http://)?(?:www\.)?redtube\.com/(?P<id>[0-9]+)' + IE_NAME = u'redtube' + + def _real_extract(self,url): + mobj = re.match(self._VALID_URL, url) + if mobj is None: + self._downloader.report_error(u'invalid URL: %s' % url) + return + 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 not None: + video_url = mobj.group(1) + else: + self._downloader.report_error(u'unable to extract media URL') + return + mobj = re.search('<h1 class="videoTitle slidePanelMovable">'+r'(.+)'+r'</h1>',webpage) + if mobj is not None: + video_title = mobj.group(1) + else: + video_title = 'Redtube - %s' % time.ctime() + + return [{ + 'id': video_id, + 'url': video_url, + 'ext': video_extension, + 'title': video_title, + }] + def gen_extractors(): """ Return a list of an instance of every supported extractor. @@ -4337,6 +4271,8 @@ def gen_extractors(): LiveLeakIE(), ARDIE(), TumblrIE(), + BandcampIE(), + RedtubeIE(), GenericIE() ]