diff options
Diffstat (limited to 'youtube_dl/extractor')
-rw-r--r-- | youtube_dl/extractor/__init__.py | 3 | ||||
-rw-r--r-- | youtube_dl/extractor/common.py | 3 | ||||
-rw-r--r-- | youtube_dl/extractor/generic.py | 8 | ||||
-rw-r--r-- | youtube_dl/extractor/ro220.py | 42 | ||||
-rw-r--r-- | youtube_dl/extractor/videofyme.py | 7 | ||||
-rw-r--r-- | youtube_dl/extractor/xhamster.py | 18 | ||||
-rw-r--r-- | youtube_dl/extractor/youtube.py | 212 |
7 files changed, 216 insertions, 77 deletions
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index b4db8f0bf..39c530ba3 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -57,6 +57,7 @@ from .pornotube import PornotubeIE from .rbmaradio import RBMARadioIE from .redtube import RedTubeIE from .ringtv import RingTVIE +from .ro220 import Ro220IE from .roxwel import RoxwelIE from .rtlnow import RTLnowIE from .sina import SinaIE @@ -116,12 +117,14 @@ _ALL_CLASSES = [ ] _ALL_CLASSES.append(GenericIE) + 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. """ return [klass() for klass in _ALL_CLASSES] + def get_info_extractor(ie_name): """Returns the info extractor class with the given ie_name""" return globals()[ie_name+'IE'] diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 8009c2d85..52c4483c9 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -47,7 +47,8 @@ class InfoExtractor(object): uploader_id: Nickname or id of the video uploader. location: Physical location of the video. player_url: SWF Player URL (used for rtmpdump). - subtitles: The subtitle file contents. + subtitles: The subtitle file contents as a dictionary in the format + {language: subtitles}. view_count: How many users have watched the video on the platform. urlhandle: [internal] The urlHandle to be used to download the file, like returned by urllib.request.urlopen diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index da016f7ee..8488dca05 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -7,12 +7,14 @@ from .common import InfoExtractor from ..utils import ( compat_urllib_error, compat_urllib_parse, + compat_urllib_parse_urlparse, compat_urllib_request, ExtractorError, ) from .brightcove import BrightcoveIE + class GenericIE(InfoExtractor): IE_DESC = u'Generic downloader that works on some sites' _VALID_URL = r'.*' @@ -23,7 +25,7 @@ class GenericIE(InfoExtractor): u'file': u'13601338388002.mp4', u'md5': u'85b90ccc9d73b4acd9138d3af4c27f89', u'info_dict': { - u"uploader": u"www.hodiho.fr", + u"uploader": u"www.hodiho.fr", u"title": u"R\u00e9gis plante sa Jeep" } }, @@ -161,6 +163,10 @@ class GenericIE(InfoExtractor): raise ExtractorError(u'Invalid URL: %s' % url) video_url = compat_urllib_parse.unquote(mobj.group(1)) + if video_url.startswith('//'): + video_url = compat_urllib_parse_urlparse(url).scheme + ':' + video_url + if '://' not in video_url: + video_url = url + ('' if url.endswith('/') else '/') + video_url video_id = os.path.basename(video_url) # here's a fun little line of code for you: diff --git a/youtube_dl/extractor/ro220.py b/youtube_dl/extractor/ro220.py new file mode 100644 index 000000000..c32f64d99 --- /dev/null +++ b/youtube_dl/extractor/ro220.py @@ -0,0 +1,42 @@ +import re + +from .common import InfoExtractor +from ..utils import ( + clean_html, + compat_parse_qs, +) + + +class Ro220IE(InfoExtractor): + IE_NAME = '220.ro' + _VALID_URL = r'(?x)(?:https?://)?(?:www\.)?220\.ro/(?P<category>[^/]+)/(?P<shorttitle>[^/]+)/(?P<video_id>[^/]+)' + _TEST = { + u"url": u"http://www.220.ro/sport/Luati-Le-Banii-Sez-4-Ep-1/LYV6doKo7f/", + u'file': u'LYV6doKo7f.mp4', + u'md5': u'03af18b73a07b4088753930db7a34add', + u'info_dict': { + u"title": u"Luati-le Banii sez 4 ep 1", + u"description": u"Iata-ne reveniti dupa o binemeritata vacanta. Va astept si pe Facebook cu pareri si comentarii.", + } + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('video_id') + + webpage = self._download_webpage(url, video_id) + flashVars_str = self._search_regex( + r'<param name="flashVars" value="([^"]+)"', + webpage, u'flashVars') + flashVars = compat_parse_qs(flashVars_str) + + info = { + '_type': 'video', + 'id': video_id, + 'ext': 'mp4', + 'url': flashVars['videoURL'][0], + 'title': flashVars['title'][0], + 'description': clean_html(flashVars['desc'][0]), + 'thumbnail': flashVars['preview'][0], + } + return info diff --git a/youtube_dl/extractor/videofyme.py b/youtube_dl/extractor/videofyme.py index 04106672b..94f64ffa5 100644 --- a/youtube_dl/extractor/videofyme.py +++ b/youtube_dl/extractor/videofyme.py @@ -14,7 +14,7 @@ class VideofyMeIE(InfoExtractor): _TEST = { u'url': u'http://www.videofy.me/thisisvideofyme/1100701', u'file': u'1100701.mp4', - u'md5': u'2046dd5758541d630bfa93e741e2fd79', + u'md5': u'c77d700bdc16ae2e9f3c26019bd96143', u'info_dict': { u'title': u'This is VideofyMe', u'description': None, @@ -32,9 +32,8 @@ class VideofyMeIE(InfoExtractor): config = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8')) video = config.find('video') sources = video.find('sources') - url_node = find_xpath_attr(sources, 'source', 'id', 'HQ on') - if url_node is None: - url_node = find_xpath_attr(sources, 'source', 'id', 'HQ off') + url_node = next(node for node in [find_xpath_attr(sources, 'source', 'id', 'HQ %s' % key) + for key in ['on', 'av', 'off']] if node is not None) video_url = url_node.find('url').text return {'id': video_id, diff --git a/youtube_dl/extractor/xhamster.py b/youtube_dl/extractor/xhamster.py index 0f1feeffd..88b8b6be0 100644 --- a/youtube_dl/extractor/xhamster.py +++ b/youtube_dl/extractor/xhamster.py @@ -3,7 +3,8 @@ import re from .common import InfoExtractor from ..utils import ( compat_urllib_parse, - + unescapeHTML, + determine_ext, ExtractorError, ) @@ -36,15 +37,16 @@ class XHamsterIE(InfoExtractor): video_url = compat_urllib_parse.unquote(mobj.group('file')) else: video_url = mobj.group('server')+'/key='+mobj.group('file') - video_extension = video_url.split('.')[-1] video_title = self._html_search_regex(r'<title>(?P<title>.+?) - xHamster\.com</title>', webpage, u'title') - # Can't see the description anywhere in the UI - # video_description = self._html_search_regex(r'<span>Description: </span>(?P<description>[^<]+)', - # webpage, u'description', fatal=False) - # if video_description: video_description = unescapeHTML(video_description) + # Only a few videos have an description + mobj = re.search('<span>Description: </span>(?P<description>[^<]+)', webpage) + if mobj: + video_description = unescapeHTML(mobj.group('description')) + else: + video_description = None mobj = re.search(r'hint=\'(?P<upload_date_Y>[0-9]{4})-(?P<upload_date_m>[0-9]{2})-(?P<upload_date_d>[0-9]{2}) [0-9]{2}:[0-9]{2}:[0-9]{2} [A-Z]{3,4}\'', webpage) if mobj: @@ -62,9 +64,9 @@ class XHamsterIE(InfoExtractor): return [{ 'id': video_id, 'url': video_url, - 'ext': video_extension, + 'ext': determine_ext(video_url), 'title': video_title, - # 'description': video_description, + 'description': video_description, 'upload_date': video_upload_date, 'uploader_id': video_uploader_id, 'thumbnail': video_thumbnail diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 4c9bab459..af01c9da0 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -155,11 +155,22 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # Listed in order of quality _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13', '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', '17', '13', '95', '94', '93', '92', '132', '151', '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_extensions = { '13': '3gp', @@ -181,7 +192,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '100': 'webm', '101': 'webm', '102': 'webm', - + # videos that use m3u8 '92': 'mp4', '93': 'mp4', @@ -190,6 +201,29 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '96': 'mp4', '132': 'mp4', '151': 'mp4', + + # Dash mp4 + '133': 'mp4', + '134': 'mp4', + '135': 'mp4', + '136': 'mp4', + '137': 'mp4', + '138': 'mp4', + '139': 'mp4', + '140': 'mp4', + '141': 'mp4', + '160': 'mp4', + + # Dash webm + '171': 'webm', + '172': 'webm', + '242': 'webm', + '243': 'webm', + '244': 'webm', + '245': 'webm', + '246': 'webm', + '247': 'webm', + '248': 'webm', } _video_dimensions = { '5': '240x400', @@ -217,11 +251,58 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '96': '1080p', '100': '360p', '101': '480p', - '102': '720p', + '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', } - _3d_itags = ['85', '84', '102', '83', '101', '82', '100'] + IE_NAME = u'youtube' _TESTS = [ { @@ -342,17 +423,19 @@ class YoutubeIE(YoutubeBaseInfoExtractor): elif len(s) == 87: return s[6:27] + s[4] + s[28:39] + s[27] + s[40:59] + s[2] + s[60:] elif len(s) == 86: - return s[5:20] + s[2] + s[21:] + return s[5:40] + s[3] + s[41:48] + s[0] + s[49:86] elif len(s) == 85: return s[83:34:-1] + s[0] + s[33:27:-1] + s[3] + s[26:19:-1] + s[34] + s[18:3:-1] + s[27] elif len(s) == 84: - return s[83:27:-1] + s[0] + s[26:5:-1] + s[2:0:-1] + s[27] + return s[5:40] + s[3] + s[41:48] + s[0] + s[49:84] elif len(s) == 83: return s[81:64:-1] + s[82] + s[63:52:-1] + s[45] + s[51:45:-1] + s[1] + s[44:1:-1] + s[0] elif len(s) == 82: - return s[36] + s[79:67:-1] + s[81] + s[66:40:-1] + s[33] + s[39:36:-1] + s[40] + s[35] + s[0] + s[67] + s[32:0:-1] + s[34] + return s[1:19] + s[0] + s[20:68] + s[19] + s[69:82] elif len(s) == 81: return s[56] + s[79:56:-1] + s[41] + s[55:41:-1] + s[80] + s[40:34:-1] + s[0] + s[33:29:-1] + s[34] + s[28:9:-1] + s[29] + s[8:0:-1] + s[9] + elif len(s) == 80: + return s[1:19] + s[0] + s[20:68] + s[19] + s[69:80] elif len(s) == 79: return s[54] + s[77:54:-1] + s[39] + s[53:39:-1] + s[78] + s[38:34:-1] + s[0] + s[33:29:-1] + s[34] + s[28:9:-1] + s[29] + s[8:0:-1] + s[9] @@ -375,11 +458,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor): try: sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8') except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - return (u'unable to download video subtitles: %s' % compat_str(err), None) + self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err)) + return {} sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list) if not sub_lang_list: - return (u'video doesn\'t have subtitles', None) + self._downloader.report_warning(u'video doesn\'t have subtitles') + return {} return sub_lang_list def _list_available_subtitles(self, video_id): @@ -388,8 +473,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): def _request_subtitle(self, sub_lang, sub_name, video_id, format): """ - Return tuple: - (error_message, sub_lang, sub) + Return the subtitle as a string or None if they are not found """ self.report_video_subtitles_request(video_id, sub_lang, format) params = compat_urllib_parse.urlencode({ @@ -402,21 +486,24 @@ class YoutubeIE(YoutubeBaseInfoExtractor): try: sub = compat_urllib_request.urlopen(url).read().decode('utf-8') except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - return (u'unable to download video subtitles: %s' % compat_str(err), None, None) + self._downloader.report_warning(u'unable to download video subtitles for %s: %s' % (sub_lang, compat_str(err))) + return if not sub: - return (u'Did not fetch video subtitles', None, None) - return (None, sub_lang, sub) + self._downloader.report_warning(u'Did not fetch video subtitles') + return + return sub def _request_automatic_caption(self, video_id, webpage): """We need the webpage for getting the captions url, pass it as an argument to speed up the process.""" - sub_lang = self._downloader.params.get('subtitleslang') or 'en' + sub_lang = (self._downloader.params.get('subtitleslangs') or ['en'])[0] sub_format = self._downloader.params.get('subtitlesformat') self.to_screen(u'%s: Looking for automatic captions' % video_id) mobj = re.search(r';ytplayer.config = ({.*?});', webpage) err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang if mobj is None: - return [(err_msg, None, None)] + self._downloader.report_warning(err_msg) + return {} player_config = json.loads(mobj.group(1)) try: args = player_config[u'args'] @@ -431,40 +518,43 @@ class YoutubeIE(YoutubeBaseInfoExtractor): }) subtitles_url = caption_url + '&' + params sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions') - return [(None, sub_lang, sub)] - except KeyError: - return [(err_msg, None, None)] - - def _extract_subtitle(self, video_id): + return {sub_lang: sub} + # An extractor error can be raise by the download process if there are + # no automatic captions but there are subtitles + except (KeyError, ExtractorError): + self._downloader.report_warning(err_msg) + return {} + + def _extract_subtitles(self, video_id): """ - Return a list with a tuple: - [(error_message, sub_lang, sub)] + Return a dictionary: {language: subtitles} or {} if the subtitles + couldn't be found """ - sub_lang_list = self._get_available_subtitles(video_id) + available_subs_list = self._get_available_subtitles(video_id) sub_format = self._downloader.params.get('subtitlesformat') - if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles - return [(sub_lang_list[0], None, None)] - if self._downloader.params.get('subtitleslang', False): - sub_lang = self._downloader.params.get('subtitleslang') - elif 'en' in sub_lang_list: - sub_lang = 'en' + if not available_subs_list: #There was some error, it didn't get the available subtitles + return {} + if self._downloader.params.get('allsubtitles', False): + sub_lang_list = available_subs_list else: - sub_lang = list(sub_lang_list.keys())[0] - if not sub_lang in sub_lang_list: - return [(u'no closed captions found in the specified language "%s"' % sub_lang, None, None)] - - subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) - return [subtitle] - - def _extract_all_subtitles(self, video_id): - sub_lang_list = self._get_available_subtitles(video_id) - sub_format = self._downloader.params.get('subtitlesformat') - if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles - return [(sub_lang_list[0], None, None)] - subtitles = [] + if self._downloader.params.get('subtitleslangs', False): + reqested_langs = self._downloader.params.get('subtitleslangs') + elif 'en' in available_subs_list: + reqested_langs = ['en'] + else: + reqested_langs = [list(available_subs_list.keys())[0]] + + sub_lang_list = {} + for sub_lang in reqested_langs: + if not sub_lang in available_subs_list: + self._downloader.report_warning(u'no closed captions found in the specified language "%s"' % sub_lang) + continue + sub_lang_list[sub_lang] = available_subs_list[sub_lang] + subtitles = {} for sub_lang in sub_lang_list: subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) - subtitles.append(subtitle) + if subtitle: + subtitles[sub_lang] = subtitle return subtitles def _print_formats(self, formats): @@ -472,7 +562,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): for x in formats: print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???'), - ' (3D)' if x in self._3d_itags else '')) + ' ('+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) @@ -655,25 +745,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # subtitles video_subtitles = None - if self._downloader.params.get('writesubtitles', False): - video_subtitles = self._extract_subtitle(video_id) - if video_subtitles: - (sub_error, sub_lang, sub) = video_subtitles[0] - if sub_error: - self._downloader.report_warning(sub_error) - - if self._downloader.params.get('writeautomaticsub', False): + if self._downloader.params.get('writesubtitles', False) or self._downloader.params.get('allsubtitles', False): + video_subtitles = self._extract_subtitles(video_id) + elif self._downloader.params.get('writeautomaticsub', False): video_subtitles = self._request_automatic_caption(video_id, video_webpage) - (sub_error, sub_lang, sub) = video_subtitles[0] - if sub_error: - self._downloader.report_warning(sub_error) - - if self._downloader.params.get('allsubtitles', False): - video_subtitles = self._extract_all_subtitles(video_id) - for video_subtitle in video_subtitles: - (sub_error, sub_lang, sub) = video_subtitle - if sub_error: - self._downloader.report_warning(sub_error) if self._downloader.params.get('listsubtitles', False): self._list_available_subtitles(video_id) @@ -699,6 +774,17 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if m_s is not None: self.to_screen(u'%s: Encrypted signatures detected.' % video_id) video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] + m_s = re.search(r'[&,]s=', args.get('adaptive_fmts', u'')) + if m_s is not None: + if 'url_encoded_fmt_stream_map' in video_info: + video_info['url_encoded_fmt_stream_map'][0] += ',' + args['adaptive_fmts'] + else: + video_info['url_encoded_fmt_stream_map'] = [args['adaptive_fmts']] + elif 'adaptive_fmts' in video_info: + if 'url_encoded_fmt_stream_map' in video_info: + video_info['url_encoded_fmt_stream_map'][0] += ',' + video_info['adaptive_fmts'][0] + else: + video_info['url_encoded_fmt_stream_map'] = video_info['adaptive_fmts'] except ValueError: pass @@ -758,7 +844,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): video_format = '{0} - {1}{2}'.format(format_param if format_param else video_extension, self._video_dimensions.get(format_param, '???'), - ' (3D)' if format_param in self._3d_itags else '') + ' ('+self._special_itags[format_param]+')' if format_param in self._special_itags else '') results.append({ 'id': video_id, |