diff options
Diffstat (limited to 'youtube_dl/InfoExtractors.py')
| -rwxr-xr-x | youtube_dl/InfoExtractors.py | 496 |
1 files changed, 241 insertions, 255 deletions
diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py index 4d145dfa1..36343882b 100755 --- a/youtube_dl/InfoExtractors.py +++ b/youtube_dl/InfoExtractors.py @@ -124,8 +124,8 @@ class InfoExtractor(object): errnote = u'Unable to download webpage' raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2]) - def _download_webpage(self, url_or_request, video_id, note=None, errnote=None): - """ Returns the data of the page as a string """ + def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None): + """ Returns a tuple (page content as string, URL handle) """ 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) @@ -142,7 +142,12 @@ class InfoExtractor(object): self.to_screen(u'Dumping request to ' + url) dump = base64.b64encode(webpage_bytes).decode('ascii') self._downloader.to_screen(dump) - return webpage_bytes.decode(encoding, 'replace') + content = webpage_bytes.decode(encoding, 'replace') + return (content, urlh) + + def _download_webpage(self, url_or_request, video_id, note=None, errnote=None): + """ Returns the data of the page as a string """ + return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0] def to_screen(self, msg): """Print msg to screen, prefixing it with '[ie_name]'""" @@ -206,7 +211,7 @@ class YoutubeIE(InfoExtractor): ([0-9A-Za-z_-]+) # here is it! the YouTube video ID (?(1).+)? # if we found the ID, everything can follow $""" - _LANG_URL = r'http://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' + _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' @@ -422,7 +427,7 @@ class YoutubeIE(InfoExtractor): # Log in login_form_strs = { - u'continue': u'http://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', + u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', u'Email': username, u'GALX': galx, u'Passwd': password, @@ -482,12 +487,12 @@ class YoutubeIE(InfoExtractor): # Extract original video URL from URL with redirection, like age verification, using next_url parameter mobj = re.search(self._NEXT_URL_RE, url) if mobj: - url = 'http://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') + url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') video_id = self._extract_id(url) # Get video webpage self.report_video_webpage_download(video_id) - url = 'http://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id + url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id request = compat_urllib_request.Request(url) try: video_webpage_bytes = compat_urllib_request.urlopen(request).read() @@ -562,12 +567,7 @@ class YoutubeIE(InfoExtractor): mobj = re.search(r'id="eow-date.*?>(.*?)</span>', 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 +784,11 @@ 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 +1073,7 @@ 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, @@ -1093,7 +1085,10 @@ class VimeoIE(InfoExtractor): config = webpage.split(' = {config:')[1].split(',assets:')[0] config = json.loads(config) except: - self._downloader.report_error(u'unable to extract info section') + if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage): + self._downloader.report_error(u'The author has restricted the access to this video, try with the "--referer" option') + else: + self._downloader.report_error(u'unable to extract info section') return # Extract title @@ -1685,10 +1680,6 @@ 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) @@ -1702,14 +1693,8 @@ 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) @@ -1720,12 +1705,11 @@ 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 ] @@ -1749,10 +1733,6 @@ 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): @@ -1772,14 +1752,9 @@ 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) @@ -1790,14 +1765,9 @@ 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) @@ -1824,11 +1794,6 @@ 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) @@ -1848,15 +1813,10 @@ 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 = [] @@ -1890,11 +1850,6 @@ 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) @@ -1906,15 +1861,9 @@ 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 @@ -1926,14 +1875,9 @@ 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 = [] @@ -2292,12 +2236,6 @@ 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: @@ -2331,24 +2269,15 @@ 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,htmlHandle = self._download_webpage_handle(url, epTitle) if dlNewest: url = htmlHandle.geturl() mobj = re.match(self._VALID_URL, url, re.VERBOSE) if mobj is None: - self._downloader.report_error(u'Invalid redirected URL: ' + url) - return + raise ExtractorError(u'Invalid redirected URL: ' + url) if mobj.group('episode') == '': - self._downloader.report_error(u'Redirected URL is still not specific: ' + url) - return + raise ExtractorError(u'Redirected URL is still not specific: ' + url) epTitle = mobj.group('episode') mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage) @@ -2360,19 +2289,15 @@ class ComedyCentralIE(InfoExtractor): altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video).*?:.*?)"', webpage) if len(altMovieParams) == 0: - self._downloader.report_error(u'unable to find Flash URL in webpage ' + url) - return + raise ExtractorError(u'unable to find Flash URL in webpage ' + url) else: mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])] 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 = [] @@ -2383,17 +2308,12 @@ 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 = [] @@ -2450,9 +2370,6 @@ class EscapistIE(InfoExtractor): _VALID_URL = r'^(https?://)?(www\.)?escapistmagazine\.com/videos/view/(?P<showname>[^/]+)/(?P<episode>[^/?]+)[/?]?.*$' 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: @@ -2462,14 +2379,7 @@ 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('<meta name="description" content="([^"]*)"', webPage) description = unescapeHTML(descMatch.group(1)) @@ -2480,14 +2390,9 @@ class EscapistIE(InfoExtractor): configUrlMatch = re.search('config=(.*)$', playerUrl) configUrl = compat_urllib_parse.unquote(configUrlMatch.group(1)) - self.report_config_download(showName) - try: - configJSON = compat_urllib_request.urlopen(configUrl) - m = re.match(r'text/html; charset="?([^"]+)"?', configJSON.headers['Content-Type']) - configJSON = configJSON.read().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 configuration: ' + compat_str(err)) - return + configJSON = self._download_webpage(configUrl, showName, + u'Downloading configuration', + u'unable to download configuration') # Technically, it's JavaScript, not JSON configJSON = configJSON.replace("'", '"') @@ -2665,40 +2570,32 @@ class SoundcloudIE(InfoExtractor): # extract simple title (uploader + slug of song title) slug_title = mobj.group(2) simple_title = uploader + u'-' + slug_title + full_title = '%s/%s' % (uploader, slug_title) - self.report_resolve('%s/%s' % (uploader, slug_title)) + self.report_resolve(full_title) url = 'http://soundcloud.com/%s/%s' % (uploader, slug_title) resolv_url = 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=b45b1aa10f1ac2941910a7f0d10f8e28' - request = compat_urllib_request.Request(resolv_url) - try: - info_json_bytes = compat_urllib_request.urlopen(request).read() - info_json = info_json_bytes.decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_error(u'unable to download video webpage: %s' % compat_str(err)) - return + info_json = self._download_webpage(resolv_url, full_title, u'Downloading info JSON') info = json.loads(info_json) video_id = info['id'] - self.report_extraction('%s/%s' % (uploader, slug_title)) + self.report_extraction(full_title) streams_url = 'https://api.sndcdn.com/i1/tracks/' + str(video_id) + '/streams?client_id=b45b1aa10f1ac2941910a7f0d10f8e28' - request = compat_urllib_request.Request(streams_url) - try: - stream_json_bytes = compat_urllib_request.urlopen(request).read() - stream_json = stream_json_bytes.decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_error(u'unable to download stream definitions: %s' % compat_str(err)) - return + stream_json = self._download_webpage(streams_url, full_title, + u'Downloading stream definitions', + u'unable to download stream definitions') streams = json.loads(stream_json) mediaURL = streams['http_mp3_128_url'] + upload_date = unified_strdate(info['created_at']) return [{ 'id': info['id'], 'url': mediaURL, 'uploader': info['user']['username'], - 'upload_date': info['created_at'], + 'upload_date': upload_date, 'title': info['title'], 'ext': u'mp3', 'description': info['description'], @@ -2714,7 +2611,7 @@ class SoundcloudSetIE(InfoExtractor): """ _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)' - IE_NAME = u'soundcloud' + IE_NAME = u'soundcloud:set' def report_resolve(self, video_id): """Report information extraction.""" @@ -2731,18 +2628,13 @@ class SoundcloudSetIE(InfoExtractor): # extract simple title (uploader + slug of song title) slug_title = mobj.group(2) simple_title = uploader + u'-' + slug_title + full_title = '%s/sets/%s' % (uploader, slug_title) - self.report_resolve('%s/sets/%s' % (uploader, slug_title)) + self.report_resolve(full_title) url = 'http://soundcloud.com/%s/sets/%s' % (uploader, slug_title) resolv_url = 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=b45b1aa10f1ac2941910a7f0d10f8e28' - request = compat_urllib_request.Request(resolv_url) - try: - info_json_bytes = compat_urllib_request.urlopen(request).read() - info_json = info_json_bytes.decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_error(u'unable to download video webpage: %s' % compat_str(err)) - return + info_json = self._download_webpage(resolv_url, full_title) videos = [] info = json.loads(info_json) @@ -2751,19 +2643,14 @@ class SoundcloudSetIE(InfoExtractor): self._downloader.report_error(u'unable to download video webpage: %s' % compat_str(err['error_message'])) return + self.report_extraction(full_title) for track in info['tracks']: video_id = track['id'] - self.report_extraction('%s/sets/%s' % (uploader, slug_title)) streams_url = 'https://api.sndcdn.com/i1/tracks/' + str(video_id) + '/streams?client_id=b45b1aa10f1ac2941910a7f0d10f8e28' - request = compat_urllib_request.Request(streams_url) - try: - stream_json_bytes = compat_urllib_request.urlopen(request).read() - stream_json = stream_json_bytes.decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_error(u'unable to download stream definitions: %s' % compat_str(err)) - return + stream_json = self._download_webpage(streams_url, video_id, u'Downloading track info JSON') + self.report_extraction(video_id) streams = json.loads(stream_json) mediaURL = streams['http_mp3_128_url'] @@ -2771,7 +2658,7 @@ class SoundcloudSetIE(InfoExtractor): 'id': video_id, 'url': mediaURL, 'uploader': track['user']['username'], - 'upload_date': track['created_at'], + 'upload_date': unified_strdate(track['created_at']), 'title': track['title'], 'ext': u'mp3', 'description': track['description'], @@ -3158,18 +3045,11 @@ class YoukuIE(InfoExtractor): info_url = 'http://v.youku.com/player/getPlayList/VideoIDS/' + video_id - request = compat_urllib_request.Request(info_url, None, std_headers) - try: - self.report_download_webpage(video_id) - jsondata = compat_urllib_request.urlopen(request).read() - 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 + jsondata = self._download_webpage(info_url, video_id) self.report_extraction(video_id) try: - jsonstr = jsondata.decode('utf-8') - config = json.loads(jsonstr) + config = json.loads(jsondata) video_title = config['data'][0]['title'] seed = config['data'][0]['seed'] @@ -3237,15 +3117,8 @@ class XNXXIE(InfoExtractor): return video_id = mobj.group(1) - self.report_download_webpage(video_id) - # Get webpage content - try: - webpage_bytes = compat_urllib_request.urlopen(url).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 download video webpage: %s' % err) - return + webpage = self._download_webpage(url, video_id) result = re.search(self.VIDEO_URL_RE, webpage) if result is None: @@ -3317,12 +3190,7 @@ class GooglePlusIE(InfoExtractor): # Step 1, Retrieve post webpage to extract further information self.report_extract_entry(post_url) - request = compat_urllib_request.Request(post_url) - try: - webpage = 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 retrieve entry webpage: %s' % compat_str(err)) - return + webpage = self._download_webpage(post_url, video_id, u'Downloading entry webpage') # Extract update date upload_date = None @@ -3359,12 +3227,7 @@ class GooglePlusIE(InfoExtractor): self._downloader.report_error(u'unable to extract video page URL') video_page = mobj.group(1) - request = compat_urllib_request.Request(video_page) - try: - webpage = 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 retrieve video webpage: %s' % compat_str(err)) - return + webpage = self._download_webpage(video_page, video_id, u'Downloading video page') self.report_extract_vid_page(video_page) @@ -3441,7 +3304,13 @@ class JustinTVIE(InfoExtractor): # starts at 1 and increases. Can we treat all parts as one video? _VALID_URL = r"""(?x)^(?:http://)?(?:www\.)?(?:twitch|justin)\.tv/ - ([^/]+)(?:/b/([^/]+))?/?(?:\#.*)?$""" + (?: + (?P<channelid>[^/]+)| + (?:(?:[^/]+)/b/(?P<videoid>[^/]+))| + (?:(?:[^/]+)/c/(?P<chapterid>[^/]+)) + ) + /?(?:\#.*)?$ + """ _JUSTIN_PAGE_LIMIT = 100 IE_NAME = u'justin.tv' @@ -3451,20 +3320,15 @@ class JustinTVIE(InfoExtractor): (channel, offset, offset + self._JUSTIN_PAGE_LIMIT)) # Return count of items, list of *valid* items - def _parse_page(self, url): - try: - urlh = compat_urllib_request.urlopen(url) - webpage_bytes = urlh.read() - webpage = webpage_bytes.decode('utf-8', 'ignore') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_error(u'unable to download video info JSON: %s' % compat_str(err)) - return + def _parse_page(self, url, video_id): + webpage = self._download_webpage(url, video_id, + u'Downloading video info JSON', + u'unable to download video info JSON') response = json.loads(webpage) if type(response) != list: error_text = response.get('error', 'unknown error') - self._downloader.report_error(u'Justin.tv API: %s' % error_text) - return + raise ExtractorError(u'Justin.tv API: %s' % error_text) info = [] for clip in response: video_url = clip['video_file_url'] @@ -3488,18 +3352,67 @@ class JustinTVIE(InfoExtractor): 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 + raise ExtractorError(u'invalid URL: %s' % url) - api = 'http://api.justin.tv' - video_id = mobj.group(mobj.lastindex) + api_base = 'http://api.justin.tv' paged = False - if mobj.lastindex == 1: + if mobj.group('channelid'): paged = True - api += '/channel/archives/%s.json' + video_id = mobj.group('channelid') + api = api_base + '/channel/archives/%s.json' % video_id + elif mobj.group('chapterid'): + chapter_id = mobj.group('chapterid') + + webpage = self._download_webpage(url, chapter_id) + m = re.search(r'PP\.archive_id = "([0-9]+)";', webpage) + if not m: + raise ExtractorError(u'Cannot find archive of a chapter') + archive_id = m.group(1) + + api = api_base + '/broadcast/by_chapter/%s.xml' % chapter_id + chapter_info_xml = self._download_webpage(api, chapter_id, + note=u'Downloading chapter information', + errnote=u'Chapter information download failed') + doc = xml.etree.ElementTree.fromstring(chapter_info_xml) + for a in doc.findall('.//archive'): + if archive_id == a.find('./id').text: + break + else: + raise ExtractorError(u'Could not find chapter in chapter information') + + video_url = a.find('./video_file_url').text + video_ext = video_url.rpartition('.')[2] or u'flv' + + chapter_api_url = u'https://api.twitch.tv/kraken/videos/c' + chapter_id + chapter_info_json = self._download_webpage(chapter_api_url, u'c' + chapter_id, + note='Downloading chapter metadata', + errnote='Download of chapter metadata failed') + chapter_info = json.loads(chapter_info_json) + + bracket_start = int(doc.find('.//bracket_start').text) + bracket_end = int(doc.find('.//bracket_end').text) + + # TODO determine start (and probably fix up file) + # youtube-dl -v http://www.twitch.tv/firmbelief/c/1757457 + #video_url += u'?start=' + TODO:start_timestamp + # bracket_start is 13290, but we want 51670615 + self._downloader.report_warning(u'Chapter detected, but we can just download the whole file. ' + u'Chapter starts at %s and ends at %s' % (formatSeconds(bracket_start), formatSeconds(bracket_end))) + + info = { + 'id': u'c' + chapter_id, + 'url': video_url, + 'ext': video_ext, + 'title': chapter_info['title'], + 'thumbnail': chapter_info['preview'], + 'description': chapter_info['description'], + 'uploader': chapter_info['channel']['display_name'], + 'uploader_id': chapter_info['channel']['name'], + } + return [info] else: - api += '/broadcast/by_archive/%s.json' - api = api % (video_id,) + video_id = mobj.group('videoid') + api = api_base + '/broadcast/by_archive/%s.json' % video_id self.report_extraction(video_id) @@ -3510,7 +3423,7 @@ class JustinTVIE(InfoExtractor): if paged: self.report_download_page(video_id, offset) page_url = api + ('?offset=%d&limit=%d' % (offset, limit)) - page_count, page_info = self._parse_page(page_url) + page_count, page_info = self._parse_page(page_url, video_id) info.extend(page_info) if not paged or page_count != limit: break @@ -3523,8 +3436,7 @@ class FunnyOrDieIE(InfoExtractor): 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 + raise ExtractorError(u'invalid URL: %s' % url) video_id = mobj.group('id') webpage = self._download_webpage(url, video_id) @@ -3557,7 +3469,8 @@ class FunnyOrDieIE(InfoExtractor): return [info] class SteamIE(InfoExtractor): - _VALID_URL = r"""http://store.steampowered.com/ + _VALID_URL = r"""http://store\.steampowered\.com/ + (agecheck/)? (?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 @@ -3623,42 +3536,35 @@ class UstreamIE(InfoExtractor): return [info] class WorldStarHipHopIE(InfoExtractor): - _VALID_URL = r'http://(?:www|m)\.worldstar(?:candy|hiphop)\.com/videos/video\.php\?v=(?P<id>.*)' + _VALID_URL = r'https?://(?:www|m)\.worldstar(?:candy|hiphop)\.com/videos/video\.php\?v=(?P<id>.*)' IE_NAME = u'WorldStarHipHop' def _real_extract(self, url): - _src_url = r"""(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'so\.addVariable\("file","(.*?)"\)' 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() + video_url = mobj.group(1) if 'mp4' in video_url: ext = 'mp4' else: ext = 'flv' else: - self._downloader.report_error(u'Cannot find video url for %s' % video_id) - return + raise ExtractorError(u'Cannot find video url for %s' % video_id) - _title = r"""<title>(.*)</title>""" + mobj = re.search(r"<title>(.*)</title>", webpage_src) - mobj = re.search(_title, webpage_src) - - if mobj is not None: - title = mobj.group(1) - else: - title = 'World Start Hip Hop - %s' % time.ctime() - - _thumbnail = r"""rel="image_src" href="(.*)" />""" - mobj = re.search(_thumbnail, webpage_src) + if mobj is None: + raise ExtractorError(u'Cannot determine title') + title = mobj.group(1) + mobj = re.search(r'rel="image_src" href="(.*)" />', webpage_src) # Getting thumbnail and if not thumbnail sets correct title for WSHH candy video. if mobj is not None: thumbnail = mobj.group(1) @@ -3756,7 +3662,7 @@ 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:</label>(?P<uploader>.*)</li>', webpage) @@ -3863,7 +3769,7 @@ 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, @@ -3990,12 +3896,13 @@ class KeekIE(InfoExtractor): return [info] class TEDIE(InfoExtractor): - _VALID_URL=r'''http://www.ted.com/ + _VALID_URL=r'''http://www\.ted\.com/ ( ((?P<type_playlist>playlists)/(?P<playlist_id>\d+)) # We have a playlist | ((?P<type_talk>talks)) # We have a simple talk ) + (/lang/(.*?))? # The url may contain the language /(?P<name>\w+) # Here goes the name and then ".html" ''' @@ -4246,7 +4153,7 @@ class ARDIE(InfoExtractor): return [info] class TumblrIE(InfoExtractor): - _VALID_URL = r'http://(?P<blog_name>.*?).tumblr.com/((post)|(video))/(?P<id>\d*)/(.*?)' + _VALID_URL = r'http://(?P<blog_name>.*?)\.tumblr\.com/((post)|(video))/(?P<id>\d*)/(.*?)' def _real_extract(self, url): m_url = re.match(self._VALID_URL, url) @@ -4256,7 +4163,7 @@ 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(?P<video_url>http://%s.tumblr.com/video_file/%s/(.*?))\\x22 type=\\x22video/(?P<ext>.*?)\\x22' % (blog, video_id) + re_video = r'src=\\x22(?P<video_url>http://%s\.tumblr\.com/video_file/%s/(.*?))\\x22 type=\\x22video/(?P<ext>.*?)\\x22' % (blog, video_id) video = re.search(re_video, webpage) if video is None: self.to_screen("No video founded") @@ -4269,8 +4176,8 @@ 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'<title>(.*?) - (?P<title>.*?)</title>' - title = unescapeHTML(re.search(re_title, webpage).group('title')) + re_title = r'<title>(?P<title>.*?)</title>' + title = unescapeHTML(re.search(re_title, webpage, re.DOTALL).group('title')) return [{'id': video_id, 'url': video_url, @@ -4279,6 +4186,83 @@ class TumblrIE(InfoExtractor): 'ext': ext }] +class BandcampIE(InfoExtractor): + _VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P<title>.*)' + + 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]+)' + + 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_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 = 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) + + 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. @@ -4334,6 +4318,8 @@ def gen_extractors(): LiveLeakIE(), ARDIE(), TumblrIE(), + BandcampIE(), + RedTubeIE(), GenericIE() ] |