X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fyoutube.py;h=8e768ea4fbad4cf7bd58703e43e853e478d8c450;hb=dbb94fb044abae465644bb14daa45b0658ef5cf0;hp=874429b78cc4917ca1cbbec7245c85436dd73783;hpb=ffa8f0df0a878463078467709f615b1e57c61ec1;p=youtube-dl diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 874429b78..8e768ea4f 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -27,17 +27,21 @@ from ..utils import ( get_element_by_id, get_element_by_attribute, ExtractorError, + int_or_none, + PagedList, + RegexNotFoundError, unescapeHTML, unified_strdate, orderedSet, write_json_file, + uppercase_escape, ) class YoutubeBaseInfoExtractor(InfoExtractor): """Provide base functions for Youtube extractors""" _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' - _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' + _AGE_URL = 'https://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' _NETRC_MACHINE = 'youtube' # If True it will raise an error if no login info is provided _LOGIN_REQUIRED = False @@ -108,7 +112,8 @@ class YoutubeBaseInfoExtractor(InfoExtractor): 'next_url': '/', 'action_confirm': 'Confirm', } - req = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) + req = compat_urllib_request.Request(self._AGE_URL, + compat_urllib_parse.urlencode(age_form).encode('ascii')) self._download_webpage( req, None, @@ -131,13 +136,16 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): ( (?:https?://|//)? # http(s):// or protocol-independent URL (optional) (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/| + (?:www\.)?deturl\.com/www\.youtube\.com/| + (?:www\.)?pwnyoutube\.com/| + (?:www\.)?yourepeat\.com/| tube\.majestyc\.net/| youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains (?:.*?\#/)? # handle anchor (#/) redirect 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 - (?:(?:watch|movie)(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) + (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) (?:\?|\#!?) # the params delimiter ? or # or #! (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) v= @@ -150,168 +158,78 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): (?(1).+)? # if we found the ID, everything can follow $""" _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' - # Listed in order of quality - _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '36', '17', '13', - # Apple HTTP Live Streaming - '96', '95', '94', '93', '92', '132', '151', - # 3D - '85', '84', '102', '83', '101', '82', '100', - # Dash video - '138', '137', '248', '136', '247', '135', '246', - '245', '244', '134', '243', '133', '242', '160', - # Dash audio - '141', '172', '140', '171', '139', - ] - _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '36', '17', '13', - # Apple HTTP Live Streaming - '96', '95', '94', '93', '92', '132', '151', - # 3D - '85', '102', '84', '101', '83', '100', '82', - # Dash video - '138', '248', '137', '247', '136', '246', '245', - '244', '135', '243', '134', '242', '133', '160', - # Dash audio - '172', '141', '171', '140', '139', - ] - _video_formats_map = { - 'flv': ['35', '34', '6', '5'], - '3gp': ['36', '17', '13'], - 'mp4': ['38', '37', '22', '18'], - 'webm': ['46', '45', '44', '43'], - } - _video_extensions = { - '13': '3gp', - '17': '3gp', - '18': 'mp4', - '22': 'mp4', - '36': '3gp', - '37': 'mp4', - '38': 'mp4', - '43': 'webm', - '44': 'webm', - '45': 'webm', - '46': 'webm', + _formats = { + '5': {'ext': 'flv', 'width': 400, 'height': 240}, + '6': {'ext': 'flv', 'width': 450, 'height': 270}, + '13': {'ext': '3gp'}, + '17': {'ext': '3gp', 'width': 176, 'height': 144}, + '18': {'ext': 'mp4', 'width': 640, 'height': 360}, + '22': {'ext': 'mp4', 'width': 1280, 'height': 720}, + '34': {'ext': 'flv', 'width': 640, 'height': 360}, + '35': {'ext': 'flv', 'width': 854, 'height': 480}, + '36': {'ext': '3gp', 'width': 320, 'height': 240}, + '37': {'ext': 'mp4', 'width': 1920, 'height': 1080}, + '38': {'ext': 'mp4', 'width': 4096, 'height': 3072}, + '43': {'ext': 'webm', 'width': 640, 'height': 360}, + '44': {'ext': 'webm', 'width': 854, 'height': 480}, + '45': {'ext': 'webm', 'width': 1280, 'height': 720}, + '46': {'ext': 'webm', 'width': 1920, 'height': 1080}, + # 3d videos - '82': 'mp4', - '83': 'mp4', - '84': 'mp4', - '85': 'mp4', - '100': 'webm', - '101': 'webm', - '102': 'webm', + '82': {'ext': 'mp4', 'height': 360, 'resolution': '360p', 'format_note': '3D', 'preference': -20}, + '83': {'ext': 'mp4', 'height': 480, 'resolution': '480p', 'format_note': '3D', 'preference': -20}, + '84': {'ext': 'mp4', 'height': 720, 'resolution': '720p', 'format_note': '3D', 'preference': -20}, + '85': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': '3D', 'preference': -20}, + '100': {'ext': 'webm', 'height': 360, 'resolution': '360p', 'format_note': '3D', 'preference': -20}, + '101': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': '3D', 'preference': -20}, + '102': {'ext': 'webm', 'height': 720, 'resolution': '720p', 'format_note': '3D', 'preference': -20}, # Apple HTTP Live Streaming - '92': 'mp4', - '93': 'mp4', - '94': 'mp4', - '95': 'mp4', - '96': 'mp4', - '132': 'mp4', - '151': 'mp4', - - # Dash mp4 - '133': 'mp4', - '134': 'mp4', - '135': 'mp4', - '136': 'mp4', - '137': 'mp4', - '138': 'mp4', - '160': 'mp4', + '92': {'ext': 'mp4', 'height': 240, 'resolution': '240p', 'format_note': 'HLS', 'preference': -10}, + '93': {'ext': 'mp4', 'height': 360, 'resolution': '360p', 'format_note': 'HLS', 'preference': -10}, + '94': {'ext': 'mp4', 'height': 480, 'resolution': '480p', 'format_note': 'HLS', 'preference': -10}, + '95': {'ext': 'mp4', 'height': 720, 'resolution': '720p', 'format_note': 'HLS', 'preference': -10}, + '96': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': 'HLS', 'preference': -10}, + '132': {'ext': 'mp4', 'height': 240, 'resolution': '240p', 'format_note': 'HLS', 'preference': -10}, + '151': {'ext': 'mp4', 'height': 72, 'resolution': '72p', 'format_note': 'HLS', 'preference': -10}, + + # DASH mp4 video + '133': {'ext': 'mp4', 'height': 240, 'resolution': '240p', 'format_note': 'DASH video', 'preference': -40}, + '134': {'ext': 'mp4', 'height': 360, 'resolution': '360p', 'format_note': 'DASH video', 'preference': -40}, + '135': {'ext': 'mp4', 'height': 480, 'resolution': '480p', 'format_note': 'DASH video', 'preference': -40}, + '136': {'ext': 'mp4', 'height': 720, 'resolution': '720p', 'format_note': 'DASH video', 'preference': -40}, + '137': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': 'DASH video', 'preference': -40}, + '138': {'ext': 'mp4', 'height': 1081, 'resolution': '>1080p', 'format_note': 'DASH video', 'preference': -40}, + '160': {'ext': 'mp4', 'height': 192, 'resolution': '192p', 'format_note': 'DASH video', 'preference': -40}, + '264': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': 'DASH video', 'preference': -40}, # Dash mp4 audio - '139': 'm4a', - '140': 'm4a', - '141': 'm4a', + '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50}, + '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50}, + '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50}, # Dash webm - '171': 'webm', - '172': 'webm', - '242': 'webm', - '243': 'webm', - '244': 'webm', - '245': 'webm', - '246': 'webm', - '247': 'webm', - '248': 'webm', - } - _video_dimensions = { - '5': '400x240', - '6': '???', - '13': '???', - '17': '176x144', - '18': '640x360', - '22': '1280x720', - '34': '640x360', - '35': '854x480', - '36': '320x240', - '37': '1920x1080', - '38': '4096x3072', - '43': '640x360', - '44': '854x480', - '45': '1280x720', - '46': '1920x1080', - '82': '360p', - '83': '480p', - '84': '720p', - '85': '1080p', - '92': '240p', - '93': '360p', - '94': '480p', - '95': '720p', - '96': '1080p', - '100': '360p', - '101': '480p', - '102': '720p', - '132': '240p', - '151': '72p', - '133': '240p', - '134': '360p', - '135': '480p', - '136': '720p', - '137': '1080p', - '138': '>1080p', - '139': '48k', - '140': '128k', - '141': '256k', - '160': '192p', - '171': '128k', - '172': '256k', - '242': '240p', - '243': '360p', - '244': '480p', - '245': '480p', - '246': '480p', - '247': '720p', - '248': '1080p', - } - _special_itags = { - '82': '3D', - '83': '3D', - '84': '3D', - '85': '3D', - '100': '3D', - '101': '3D', - '102': '3D', - '133': 'DASH Video', - '134': 'DASH Video', - '135': 'DASH Video', - '136': 'DASH Video', - '137': 'DASH Video', - '138': 'DASH Video', - '139': 'DASH Audio', - '140': 'DASH Audio', - '141': 'DASH Audio', - '160': 'DASH Video', - '171': 'DASH Audio', - '172': 'DASH Audio', - '242': 'DASH Video', - '243': 'DASH Video', - '244': 'DASH Video', - '245': 'DASH Video', - '246': 'DASH Video', - '247': 'DASH Video', - '248': 'DASH Video', + '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40}, + '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40}, + '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40}, + '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40}, + '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40}, + '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40}, + '242': {'ext': 'webm', 'height': 240, 'resolution': '240p', 'format_note': 'DASH webm', 'preference': -40}, + '243': {'ext': 'webm', 'height': 360, 'resolution': '360p', 'format_note': 'DASH webm', 'preference': -40}, + '244': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': 'DASH webm', 'preference': -40}, + '245': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': 'DASH webm', 'preference': -40}, + '246': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': 'DASH webm', 'preference': -40}, + '247': {'ext': 'webm', 'height': 720, 'resolution': '720p', 'format_note': 'DASH webm', 'preference': -40}, + '248': {'ext': 'webm', 'height': 1080, 'resolution': '1080p', 'format_note': 'DASH webm', 'preference': -40}, + + # Dash webm audio + '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH webm audio', 'abr': 48, 'preference': -50}, + '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH webm audio', 'abr': 256, 'preference': -50}, + + # RTMP (unnamed) + '_rtmp': {'protocol': 'rtmp'}, } IE_NAME = u'youtube' @@ -363,6 +281,22 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): u"uploader_id": u"setindia" } }, + { + u"url": u"http://www.youtube.com/watch?v=a9LDPn-MO4I", + u"file": u"a9LDPn-MO4I.m4a", + u"note": u"256k DASH audio (format 141) via DASH manifest", + u"info_dict": { + u"upload_date": "20121002", + u"uploader_id": "8KVIDEO", + u"description": "No description available.", + u"uploader": "8KVIDEO", + u"title": "UHDTV TEST 8K VIDEO.mp4" + }, + u"params": { + u"youtube_include_dash_manifest": True, + u"format": "141", + }, + }, ] @@ -570,7 +504,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): return a % b m = re.match( - r'^(?P[a-zA-Z]+)\((?P[a-z0-9,]+)\)$', expr) + r'^(?P[a-zA-Z$]+)\((?P[a-z0-9,]+)\)$', expr) if m: fname = m.group('func') if fname not in functions: @@ -1083,7 +1017,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): def _get_available_subtitles(self, video_id, webpage): try: sub_list = self._download_webpage( - 'http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id, + 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id, video_id, note=False) except ExtractorError as err: self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err)) @@ -1097,9 +1031,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): 'lang': lang, 'v': video_id, 'fmt': self._downloader.params.get('subtitlesformat', 'srt'), - 'name': l[0].encode('utf-8'), + 'name': unescapeHTML(l[0]).encode('utf-8'), }) - url = u'http://www.youtube.com/api/timedtext?' + params + url = u'https://www.youtube.com/api/timedtext?' + params sub_lang_list[lang] = url if not sub_lang_list: self._downloader.report_warning(u'video doesn\'t have subtitles') @@ -1153,69 +1087,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): self._downloader.report_warning(err_msg) return {} - def _print_formats(self, formats): - print('Available formats:') - for x in formats: - print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'), - self._video_dimensions.get(x, '???'), - ' ('+self._special_itags[x]+')' if x in self._special_itags else '')) - - def _extract_id(self, url): - mobj = re.match(self._VALID_URL, url, re.VERBOSE) + @classmethod + def extract_id(cls, url): + mobj = re.match(cls._VALID_URL, url, re.VERBOSE) if mobj is None: raise ExtractorError(u'Invalid URL: %s' % url) video_id = mobj.group(2) return video_id - def _get_video_url_list(self, url_map): - """ - Transform a dictionary in the format {itag:url} to a list of (itag, url) - with the requested formats. - """ - req_format = self._downloader.params.get('format', None) - format_limit = self._downloader.params.get('format_limit', None) - available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats - if format_limit is not None and format_limit in available_formats: - format_list = available_formats[available_formats.index(format_limit):] - else: - format_list = available_formats - existing_formats = [x for x in format_list if x in url_map] - if len(existing_formats) == 0: - raise ExtractorError(u'no known formats available for video') - if self._downloader.params.get('listformats', None): - self._print_formats(existing_formats) - return - if req_format is None or req_format == 'best': - video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality - elif req_format == 'worst': - video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality - elif req_format in ('-1', 'all'): - video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats - else: - # Specific formats. We pick the first in a slash-delimeted sequence. - # Format can be specified as itag or 'mp4' or 'flv' etc. We pick the highest quality - # available in the specified format. For example, - # if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. - # if '1/mp4/3/4' is requested and '1' and '5' (is a mp4) are available, we pick '1'. - # if '1/mp4/3/4' is requested and '4' and '5' (is a mp4) are available, we pick '5'. - req_formats = req_format.split('/') - video_url_list = None - for rf in req_formats: - if rf in url_map: - video_url_list = [(rf, url_map[rf])] - break - if rf in self._video_formats_map: - for srf in self._video_formats_map[rf]: - if srf in url_map: - video_url_list = [(srf, url_map[srf])] - break - else: - continue - break - if video_url_list is None: - raise ExtractorError(u'requested format not available') - return video_url_list - def _extract_from_m3u8(self, manifest_url, video_id): url_map = {} def _get_urls(_manifest): @@ -1239,7 +1118,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): mobj = re.search(self._NEXT_URL_RE, url) if mobj: url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') - video_id = self._extract_id(url) + video_id = self.extract_id(url) # Get video webpage url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id @@ -1361,7 +1240,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): video_description = u'' def _extract_count(klass): - count = self._search_regex(r'class="%s">([\d,]+)' % re.escape(klass), video_webpage, klass, fatal=False) + count = self._search_regex( + r'class="%s">([\d,]+)' % re.escape(klass), + video_webpage, klass, default=None) if count is not None: return int(count.replace(',', '')) return None @@ -1377,9 +1258,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): if 'length_seconds' not in video_info: self._downloader.report_warning(u'unable to extract video duration') - video_duration = '' + video_duration = None else: - video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]) + video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])) # annotations video_annotations = None @@ -1387,7 +1268,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): video_annotations = self._extract_annotations(video_id) # Decide which formats to download - try: mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage) if not mobj: @@ -1412,9 +1292,27 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): except ValueError: pass + def _map_to_format_list(urlmap): + formats = [] + for itag, video_real_url in urlmap.items(): + dct = { + 'format_id': itag, + 'url': video_real_url, + 'player_url': player_url, + } + if itag in self._formats: + dct.update(self._formats[itag]) + formats.append(dct) + return formats + if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): self.report_rtmp_download() - video_url_list = [(None, video_info['conn'][0])] + formats = [{ + 'format_id': '_rtmp', + 'protocol': 'rtmp', + 'url': video_info['conn'][0], + 'player_url': player_url, + }] elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1: encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts',[''])[0] if 'rtmpe%3Dyes' in encoded_url_map: @@ -1459,55 +1357,75 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): if 'ratebypass' not in url: url += '&ratebypass=yes' url_map[url_data['itag'][0]] = url - video_url_list = self._get_video_url_list(url_map) - if not video_url_list: - return + formats = _map_to_format_list(url_map) elif video_info.get('hlsvp'): manifest_url = video_info['hlsvp'][0] url_map = self._extract_from_m3u8(manifest_url, video_id) - video_url_list = self._get_video_url_list(url_map) - if not video_url_list: - return - + formats = _map_to_format_list(url_map) else: raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info') - results = [] - for itag, video_real_url in video_url_list: - # Extension - video_extension = self._video_extensions.get(itag, 'flv') - - video_format = '{0} - {1}{2}'.format(itag if itag else video_extension, - self._video_dimensions.get(itag, '???'), - ' ('+self._special_itags[itag]+')' if itag in self._special_itags else '') - - results.append({ - 'id': video_id, - 'url': video_real_url, - 'uploader': video_uploader, - 'uploader_id': video_uploader_id, - 'upload_date': upload_date, - 'title': video_title, - 'ext': video_extension, - 'format': video_format, - 'format_id': itag, - 'thumbnail': video_thumbnail, - 'description': video_description, - 'player_url': player_url, - 'subtitles': video_subtitles, - 'duration': video_duration, - 'age_limit': 18 if age_gate else 0, - 'annotations': video_annotations, - 'webpage_url': 'https://www.youtube.com/watch?v=%s' % video_id, - 'view_count': view_count, - 'like_count': like_count, - 'dislike_count': dislike_count, - }) - return results + # Look for the DASH manifest + dash_manifest_url_lst = video_info.get('dashmpd') + if (dash_manifest_url_lst and dash_manifest_url_lst[0] and + self._downloader.params.get('youtube_include_dash_manifest', False)): + try: + dash_doc = self._download_xml( + dash_manifest_url_lst[0], video_id, + note=u'Downloading DASH manifest', + errnote=u'Could not download DASH manifest') + for r in dash_doc.findall(u'.//{urn:mpeg:DASH:schema:MPD:2011}Representation'): + url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL') + if url_el is None: + continue + format_id = r.attrib['id'] + video_url = url_el.text + filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength')) + f = { + 'format_id': format_id, + 'url': video_url, + 'width': int_or_none(r.attrib.get('width')), + 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000), + 'asr': int_or_none(r.attrib.get('audioSamplingRate')), + 'filesize': filesize, + } + try: + existing_format = next( + fo for fo in formats + if fo['format_id'] == format_id) + except StopIteration: + f.update(self._formats.get(format_id, {})) + formats.append(f) + else: + existing_format.update(f) + + except (ExtractorError, KeyError) as e: + self.report_warning(u'Skipping DASH manifest: %s' % e, video_id) + + self._sort_formats(formats) + + return { + 'id': video_id, + 'uploader': video_uploader, + 'uploader_id': video_uploader_id, + 'upload_date': upload_date, + 'title': video_title, + 'thumbnail': video_thumbnail, + 'description': video_description, + 'subtitles': video_subtitles, + 'duration': video_duration, + 'age_limit': 18 if age_gate else 0, + 'annotations': video_annotations, + 'webpage_url': 'https://www.youtube.com/watch?v=%s' % video_id, + 'view_count': view_count, + 'like_count': like_count, + 'dislike_count': dislike_count, + 'formats': formats, + } class YoutubePlaylistIE(YoutubeBaseInfoExtractor): IE_DESC = u'YouTube.com playlists' - _VALID_URL = r"""(?: + _VALID_URL = r"""(?x)(?: (?:https?://)? (?:\w+\.)? youtube\.com/ @@ -1516,21 +1434,20 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): \? (?:.*?&)*? (?:p|a|list)= | p/ ) - ((?:PL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}) + ( + (?:PL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,} + # Top tracks, they can also include dots + |(?:MC)[\w\.]* + ) .* | ((?:PL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,}) )""" - _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&page=%s' + _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s' _MORE_PAGES_INDICATOR = r'data-link-type="next"' - _VIDEO_RE = r'href="/watch\?v=(?P[0-9A-Za-z_-]{11})&[^"]*?index=(?P\d+)' + _VIDEO_RE = r'href="\s*/watch\?v=(?P[0-9A-Za-z_-]{11})&[^"]*?index=(?P\d+)' IE_NAME = u'youtube:playlist' - @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 _real_initialize(self): self._login() @@ -1554,7 +1471,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): def _real_extract(self, url): # Extract playlist id - mobj = re.match(self._VALID_URL, url, re.VERBOSE) + mobj = re.match(self._VALID_URL, url) if mobj is None: raise ExtractorError(u'Invalid URL: %s' % url) playlist_id = mobj.group(1) or mobj.group(2) @@ -1576,22 +1493,31 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): raise ExtractorError(u'For downloading YouTube.com top lists, use ' u'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True) + url = self._TEMPLATE_URL % playlist_id + page = self._download_webpage(url, playlist_id) + more_widget_html = content_html = page + # Extract the video ids from the playlist pages ids = [] for page_num in itertools.count(1): - url = self._TEMPLATE_URL % (playlist_id, page_num) - page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) - matches = re.finditer(self._VIDEO_RE, page) + matches = re.finditer(self._VIDEO_RE, content_html) # We remove the duplicates and the link with index 0 # (it's not the first video of the playlist) new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0') ids.extend(new_ids) - if re.search(self._MORE_PAGES_INDICATOR, page) is None: + mobj = re.search(r'data-uix-load-more-href="/?(?P[^"]+)"', more_widget_html) + if not mobj: break - playlist_title = self._og_search_title(page) + more = self._download_json( + 'https://youtube.com/%s' % mobj.group('more'), playlist_id, 'Downloading page #%s' % page_num) + content_html = more['content_html'] + more_widget_html = more['load_more_widget_html'] + + playlist_title = self._html_search_regex( + r'

\s*(.*?)\s*

', page, u'title') url_results = self._ids_to_results(ids) return self.playlist_result(url_results, playlist_id, playlist_title) @@ -1608,7 +1534,7 @@ class YoutubeTopListIE(YoutubePlaylistIE): channel = mobj.group('chann') title = mobj.group('title') query = compat_urllib_parse.urlencode({'title': title}) - playlist_re = 'href="([^"]+?%s[^"]+?)"' % re.escape(query) + playlist_re = 'href="([^"]+?%s.*?)"' % re.escape(query) channel_page = self._download_webpage('https://www.youtube.com/%s' % channel, title) link = self._html_search_regex(playlist_re, channel_page, u'list') url = compat_urlparse.urljoin('https://www.youtube.com/', link) @@ -1633,7 +1559,7 @@ class YoutubeChannelIE(InfoExtractor): IE_DESC = u'YouTube.com channels' _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" _MORE_PAGES_INDICATOR = 'yt-uix-load-more' - _MORE_PAGES_URL = 'http://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' + _MORE_PAGES_URL = 'https://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' IE_NAME = u'youtube:channel' def extract_videos_from_page(self, page): @@ -1668,11 +1594,10 @@ class YoutubeChannelIE(InfoExtractor): # Download all channel pages using the json-based channel_ajax query for pagenum in itertools.count(1): url = self._MORE_PAGES_URL % (pagenum, channel_id) - page = self._download_webpage(url, channel_id, - u'Downloading page #%s' % pagenum) - - page = json.loads(page) - + page = self._download_json( + url, channel_id, note=u'Downloading page #%s' % pagenum, + transform_source=uppercase_escape) + ids_in_page = self.extract_videos_from_page(page['content_html']) video_ids.extend(ids_in_page) @@ -1689,9 +1614,9 @@ class YoutubeChannelIE(InfoExtractor): class YoutubeUserIE(InfoExtractor): IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)' _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)' - _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s' + _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s' _GDATA_PAGE_SIZE = 50 - _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json' + _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json' IE_NAME = u'youtube:user' @classmethod @@ -1715,41 +1640,36 @@ class YoutubeUserIE(InfoExtractor): # page by page until there are no video ids - it means we got # all of them. - video_ids = [] - - for pagenum in itertools.count(0): + def download_page(pagenum): start_index = pagenum * self._GDATA_PAGE_SIZE + 1 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) - page = self._download_webpage(gdata_url, username, - u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE)) + page = self._download_webpage( + gdata_url, username, + u'Downloading video ids from %d to %d' % ( + start_index, start_index + self._GDATA_PAGE_SIZE)) try: response = json.loads(page) except ValueError as err: raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) if 'entry' not in response['feed']: - # Number of videos is a multiple of self._MAX_RESULTS - break + return # Extract video identifiers - ids_in_page = [] - for entry in response['feed']['entry']: - ids_in_page.append(entry['id']['$t'].split('/')[-1]) - video_ids.extend(ids_in_page) - - # A little optimization - if current page is not - # "full", ie. does not contain PAGE_SIZE video ids then - # we can assume that this page is the last one - there - # are no more ids on further pages - no need to query - # again. - - if len(ids_in_page) < self._GDATA_PAGE_SIZE: - break + entries = response['feed']['entry'] + for entry in entries: + title = entry['title']['$t'] + video_id = entry['id']['$t'].split('/')[-1] + yield { + '_type': 'url', + 'url': video_id, + 'ie_key': 'Youtube', + 'id': video_id, + 'title': title, + } + url_results = PagedList(download_page, self._GDATA_PAGE_SIZE) - url_results = [ - self.url_result(video_id, 'Youtube', video_id=video_id) - for video_id in video_ids] return self.playlist_result(url_results, playlist_title=username) @@ -1777,7 +1697,8 @@ class YoutubeSearchIE(SearchInfoExtractor): api_response = data['data'] if 'items' not in api_response: - raise ExtractorError(u'[youtube] No video results') + raise ExtractorError( + u'[youtube] No video results', expected=True) new_ids = list(video['id'] for video in api_response['items']) video_ids += new_ids @@ -1827,7 +1748,7 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): action = 'action_load_system_feed' if self._PERSONAL_FEED: action = 'action_load_personal_feed' - return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME) + return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME) @property def IE_NAME(self): @@ -1896,13 +1817,16 @@ class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): class YoutubeTruncatedURLIE(InfoExtractor): IE_NAME = 'youtube:truncated_url' IE_DESC = False # Do not list - _VALID_URL = r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$' + _VALID_URL = r'''(?x) + (?:https?://)?[^/]+/watch\?(?:feature=[a-z_]+)?$| + (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$ + ''' def _real_extract(self, url): raise ExtractorError( u'Did you forget to quote the URL? Remember that & is a meta ' u'character in most shells, so you want to put the URL in quotes, ' u'like youtube-dl ' - u'\'http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc\'' - u' (or simply youtube-dl BaW_jenozKc ).', + u'"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" ' + u' or simply youtube-dl BaW_jenozKc .', expected=True)