diff options
Diffstat (limited to 'youtube_dl/extractor')
| -rw-r--r-- | youtube_dl/extractor/__init__.py | 9 | ||||
| -rw-r--r-- | youtube_dl/extractor/arte.py | 4 | ||||
| -rw-r--r-- | youtube_dl/extractor/auengine.py | 31 | ||||
| -rw-r--r-- | youtube_dl/extractor/bambuser.py | 3 | ||||
| -rw-r--r-- | youtube_dl/extractor/canalplus.py | 3 | ||||
| -rw-r--r-- | youtube_dl/extractor/common.py | 48 | ||||
| -rw-r--r-- | youtube_dl/extractor/d8.py | 22 | ||||
| -rw-r--r-- | youtube_dl/extractor/dailymotion.py | 2 | ||||
| -rw-r--r-- | youtube_dl/extractor/gamekings.py | 6 | ||||
| -rw-r--r-- | youtube_dl/extractor/generic.py | 31 | ||||
| -rw-r--r-- | youtube_dl/extractor/livestream.py | 44 | ||||
| -rw-r--r-- | youtube_dl/extractor/mtv.py | 2 | ||||
| -rw-r--r-- | youtube_dl/extractor/redtube.py | 4 | ||||
| -rw-r--r-- | youtube_dl/extractor/rtlnow.py | 12 | ||||
| -rw-r--r-- | youtube_dl/extractor/soundcloud.py | 2 | ||||
| -rw-r--r-- | youtube_dl/extractor/southparkstudios.py | 25 | ||||
| -rw-r--r-- | youtube_dl/extractor/spiegel.py | 50 | ||||
| -rw-r--r-- | youtube_dl/extractor/ted.py | 38 | ||||
| -rw-r--r-- | youtube_dl/extractor/toutv.py | 75 | ||||
| -rw-r--r-- | youtube_dl/extractor/tvp.py | 5 | ||||
| -rw-r--r-- | youtube_dl/extractor/vevo.py | 7 | ||||
| -rw-r--r-- | youtube_dl/extractor/youtube.py | 67 |
22 files changed, 366 insertions, 124 deletions
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 0594a3666..b0df1cef7 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -25,6 +25,7 @@ from .comedycentral import ComedyCentralIE from .condenast import CondeNastIE from .criterion import CriterionIE from .cspan import CSpanIE +from .d8 import D8IE from .dailymotion import ( DailymotionIE, DailymotionPlaylistIE, @@ -80,7 +81,7 @@ from .keezmovies import KeezMoviesIE from .kickstarter import KickStarterIE from .keek import KeekIE from .liveleak import LiveLeakIE -from .livestream import LivestreamIE +from .livestream import LivestreamIE, LivestreamOriginalIE from .metacafe import MetacafeIE from .metacritic import MetacriticIE from .mit import TechTVMITIE, MITIE @@ -116,7 +117,10 @@ from .slashdot import SlashdotIE from .slideshare import SlideshareIE from .sohu import SohuIE from .soundcloud import SoundcloudIE, SoundcloudSetIE, SoundcloudUserIE -from .southparkstudios import SouthParkStudiosIE +from .southparkstudios import ( + SouthParkStudiosIE, + SouthparkDeIE, +) from .space import SpaceIE from .spankwire import SpankwireIE from .spiegel import SpiegelIE @@ -129,6 +133,7 @@ from .techtalks import TechTalksIE from .ted import TEDIE from .tf1 import TF1IE from .thisav import ThisAVIE +from .toutv import TouTvIE from .traileraddict import TrailerAddictIE from .trilulilu import TriluliluIE from .tube8 import Tube8IE diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index b35a679e3..44d0b5d70 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -69,7 +69,7 @@ class ArteTvIE(InfoExtractor): lang = mobj.group('lang') return self._extract_liveweb(url, name, lang) - if re.search(self._LIVE_URL, video_id) is not None: + if re.search(self._LIVE_URL, url) is not None: raise ExtractorError(u'Arte live streams are not yet supported, sorry') # self.extractLiveStream(url) # return @@ -115,7 +115,7 @@ class ArteTvIE(InfoExtractor): event_doc = config_doc.find('event') url_node = event_doc.find('video').find('urlHd') if url_node is None: - url_node = video_doc.find('urlSd') + url_node = event_doc.find('urlSd') return {'id': video_id, 'title': event_doc.find('name%s' % lang.capitalize()).text, diff --git a/youtube_dl/extractor/auengine.py b/youtube_dl/extractor/auengine.py index 0febbff4f..95c038003 100644 --- a/youtube_dl/extractor/auengine.py +++ b/youtube_dl/extractor/auengine.py @@ -1,10 +1,10 @@ -import os.path import re from .common import InfoExtractor from ..utils import ( compat_urllib_parse, - compat_urllib_parse_urlparse, + determine_ext, + ExtractorError, ) class AUEngineIE(InfoExtractor): @@ -25,22 +25,25 @@ class AUEngineIE(InfoExtractor): title = self._html_search_regex(r'<title>(?P<title>.+?)</title>', webpage, u'title') title = title.strip() - links = re.findall(r'[^A-Za-z0-9]?(?:file|url):\s*["\'](http[^\'"&]*)', webpage) - links = [compat_urllib_parse.unquote(l) for l in links] + links = re.findall(r'\s(?:file|url):\s*["\']([^\'"]+)["\']', webpage) + links = map(compat_urllib_parse.unquote, links) + + thumbnail = None + video_url = None for link in links: - root, pathext = os.path.splitext(compat_urllib_parse_urlparse(link).path) - if pathext == '.png': + if link.endswith('.png'): thumbnail = link - elif pathext == '.mp4': - url = link - ext = pathext + elif '/videos/' in link: + video_url = link + if not video_url: + raise ExtractorError(u'Could not find video URL') + ext = u'.' + determine_ext(video_url) if ext == title[-len(ext):]: title = title[:-len(ext)] - ext = ext[1:] - return [{ + + return { 'id': video_id, - 'url': url, - 'ext': ext, + 'url': video_url, 'title': title, 'thumbnail': thumbnail, - }] + } diff --git a/youtube_dl/extractor/bambuser.py b/youtube_dl/extractor/bambuser.py index f3b36f473..967568c4a 100644 --- a/youtube_dl/extractor/bambuser.py +++ b/youtube_dl/extractor/bambuser.py @@ -15,7 +15,8 @@ class BambuserIE(InfoExtractor): _TEST = { u'url': u'http://bambuser.com/v/4050584', - u'md5': u'fba8f7693e48fd4e8641b3fd5539a641', + # MD5 seems to be flaky, see https://travis-ci.org/rg3/youtube-dl/jobs/14051016#L388 + #u'md5': u'fba8f7693e48fd4e8641b3fd5539a641', u'info_dict': { u'id': u'4050584', u'ext': u'flv', diff --git a/youtube_dl/extractor/canalplus.py b/youtube_dl/extractor/canalplus.py index 1db9b24cf..bfa2a8b40 100644 --- a/youtube_dl/extractor/canalplus.py +++ b/youtube_dl/extractor/canalplus.py @@ -5,6 +5,7 @@ import xml.etree.ElementTree from .common import InfoExtractor from ..utils import unified_strdate + class CanalplusIE(InfoExtractor): _VALID_URL = r'https?://(www\.canalplus\.fr/.*?/(?P<path>.*)|player\.canalplus\.fr/#/(?P<id>\d+))' _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/cplus/%s' @@ -25,7 +26,7 @@ class CanalplusIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + video_id = mobj.groupdict().get('id') if video_id is None: webpage = self._download_webpage(url, mobj.group('path')) video_id = self._search_regex(r'videoId = "(\d+)";', webpage, u'video id') diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 9c20d30b4..eb3435c77 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -71,6 +71,10 @@ class InfoExtractor(object): ("3D" or "DASH video") * width Width of the video, if known * height Height of the video, if known + * abr Average audio bitrate in KBit/s + * acodec Name of the audio codec in use + * vbr Average video bitrate in KBit/s + * vcodec Name of the video codec in use webpage_url: The url to the video webpage, if given to youtube-dl it should allow to get the same result again. (It will be set by YoutubeDL if it's missing) @@ -315,13 +319,19 @@ class InfoExtractor(object): # Helper functions for extracting OpenGraph info @staticmethod - def _og_regex(prop): - return r'<meta.+?property=[\'"]og:%s[\'"].+?content=(?:"(.+?)"|\'(.+?)\')' % re.escape(prop) + def _og_regexes(prop): + content_re = r'content=(?:"([^>]+?)"|\'(.+?)\')' + property_re = r'property=[\'"]og:%s[\'"]' % re.escape(prop) + template = r'<meta[^>]+?%s[^>]+?%s' + return [ + template % (property_re, content_re), + template % (content_re, property_re), + ] def _og_search_property(self, prop, html, name=None, **kargs): if name is None: name = 'OpenGraph %s' % prop - escaped = self._search_regex(self._og_regex(prop), html, name, flags=re.DOTALL, **kargs) + escaped = self._search_regex(self._og_regexes(prop), html, name, flags=re.DOTALL, **kargs) if escaped is None: return None return unescapeHTML(escaped) @@ -336,10 +346,21 @@ class InfoExtractor(object): return self._og_search_property('title', html, **kargs) def _og_search_video_url(self, html, name='video url', secure=True, **kargs): - regexes = [self._og_regex('video')] - if secure: regexes.insert(0, self._og_regex('video:secure_url')) + regexes = self._og_regexes('video') + if secure: regexes = self._og_regexes('video:secure_url') + regexes return self._html_search_regex(regexes, html, name, **kargs) + def _html_search_meta(self, name, html, display_name=None): + if display_name is None: + display_name = name + return self._html_search_regex( + r'''(?ix)<meta(?=[^>]+(?:name|property)=["\']%s["\']) + [^>]+content=["\']([^"\']+)["\']''' % re.escape(name), + html, display_name, fatal=False) + + def _dc_search_uploader(self, html): + return self._html_search_meta('dc.creator', html, 'uploader') + def _rta_search(self, html): # See http://www.rtalabel.org/index.php?content=howtofaq#single if re.search(r'(?ix)<meta\s+name="rating"\s+' @@ -348,6 +369,23 @@ class InfoExtractor(object): return 18 return 0 + def _media_rating_search(self, html): + # See http://www.tjg-designs.com/WP/metadata-code-examples-adding-metadata-to-your-web-pages/ + rating = self._html_search_meta('rating', html) + + if not rating: + return None + + RATING_TABLE = { + 'safe for kids': 0, + 'general': 8, + '14 years': 14, + 'mature': 17, + 'restricted': 19, + } + return RATING_TABLE.get(rating.lower(), None) + + class SearchInfoExtractor(InfoExtractor): """ diff --git a/youtube_dl/extractor/d8.py b/youtube_dl/extractor/d8.py new file mode 100644 index 000000000..a56842b16 --- /dev/null +++ b/youtube_dl/extractor/d8.py @@ -0,0 +1,22 @@ +# encoding: utf-8 +from .canalplus import CanalplusIE + + +class D8IE(CanalplusIE): + _VALID_URL = r'https?://www\.d8\.tv/.*?/(?P<path>.*)' + _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/d8/%s' + IE_NAME = u'd8.tv' + + _TEST = { + u'url': u'http://www.d8.tv/d8-docs-mags/pid6589-d8-campagne-intime.html', + u'file': u'966289.flv', + u'info_dict': { + u'title': u'Campagne intime - Documentaire exceptionnel', + u'description': u'md5:d2643b799fb190846ae09c61e59a859f', + u'upload_date': u'20131108', + }, + u'params': { + # rtmp + u'skip_download': True, + }, + } diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index e87690f9d..71f5e03ee 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -186,7 +186,7 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor): webpage = self._download_webpage(request, id, u'Downloading page %s' % pagenum) - playlist_el = get_element_by_attribute(u'class', u'video_list', webpage) + playlist_el = get_element_by_attribute(u'class', u'row video_list', webpage) video_ids.extend(re.findall(r'data-id="(.+?)"', playlist_el)) if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None: diff --git a/youtube_dl/extractor/gamekings.py b/youtube_dl/extractor/gamekings.py index 6aa657ef4..c91669b0e 100644 --- a/youtube_dl/extractor/gamekings.py +++ b/youtube_dl/extractor/gamekings.py @@ -1,9 +1,6 @@ import re from .common import InfoExtractor -from ..utils import ( - determine_ext, -) class GamekingsIE(InfoExtractor): @@ -11,7 +8,8 @@ class GamekingsIE(InfoExtractor): _TEST = { u"url": u"http://www.gamekings.tv/videos/phoenix-wright-ace-attorney-dual-destinies-review/", u'file': u'20130811.mp4', - u'md5': u'2f32b1f7b80fdc5cb616efb4f387f8a3', + # MD5 is flaky, seems to change regularly + #u'md5': u'2f32b1f7b80fdc5cb616efb4f387f8a3', u'info_dict': { u"title": u"Phoenix Wright: Ace Attorney \u2013 Dual Destinies Review", u"description": u"Melle en Steven hebben voor de review een week in de rechtbank doorbracht met Phoenix Wright: Ace Attorney - Dual Destinies.", diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index c7552fddb..e1d6a2a01 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -162,6 +162,16 @@ class GenericIE(InfoExtractor): raise ExtractorError(u'Failed to download URL: %s' % url) self.report_extraction(video_id) + + # it's tempting to parse this further, but you would + # have to take into account all the variations like + # Video Title - Site Name + # Site Name | Video Title + # Video Title - Tagline | Site Name + # and so on and so forth; it's just not practical + video_title = self._html_search_regex(r'<title>(.*)</title>', + webpage, u'video title', default=u'video', flags=re.DOTALL) + # Look for BrightCove: bc_url = BrightcoveIE._extract_brightcove_url(webpage) if bc_url is not None: @@ -177,11 +187,13 @@ class GenericIE(InfoExtractor): return self.url_result(surl, 'Vimeo') # Look for embedded YouTube player - mobj = re.search( - r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?youtube.com/embed/.+?)\1', webpage) - if mobj: - surl = unescapeHTML(mobj.group(u'url')) - return self.url_result(surl, 'Youtube') + matches = re.findall( + r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?youtube.com/embed/.+?)\1', webpage) + if matches: + urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Youtube') + for tuppl in matches] + return self.playlist_result( + urlrs, playlist_id=video_id, playlist_title=video_title) # Look for Bandcamp pages with custom domain mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage) @@ -226,15 +238,6 @@ class GenericIE(InfoExtractor): video_extension = os.path.splitext(video_id)[1][1:] video_id = os.path.splitext(video_id)[0] - # it's tempting to parse this further, but you would - # have to take into account all the variations like - # Video Title - Site Name - # Site Name | Video Title - # Video Title - Tagline | Site Name - # and so on and so forth; it's just not practical - video_title = self._html_search_regex(r'<title>(.*)</title>', - webpage, u'video title', default=u'video', flags=re.DOTALL) - # video uploader is domain name video_uploader = self._search_regex(r'(?:https?://)?([^/]*)/.*', url, u'video uploader') diff --git a/youtube_dl/extractor/livestream.py b/youtube_dl/extractor/livestream.py index 4531fd6ab..1a3e0ae6b 100644 --- a/youtube_dl/extractor/livestream.py +++ b/youtube_dl/extractor/livestream.py @@ -1,16 +1,19 @@ import re import json +import xml.etree.ElementTree from .common import InfoExtractor from ..utils import ( compat_urllib_parse_urlparse, compat_urlparse, get_meta_content, + xpath_with_ns, ExtractorError, ) class LivestreamIE(InfoExtractor): + IE_NAME = u'livestream' _VALID_URL = r'http://new.livestream.com/.*?/(?P<event_name>.*?)(/videos/(?P<id>\d+))?/?$' _TEST = { u'url': u'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370', @@ -54,3 +57,44 @@ class LivestreamIE(InfoExtractor): info = json.loads(self._download_webpage(api_url, video_id, u'Downloading video info')) return self._extract_video_info(info) + + +# The original version of Livestream uses a different system +class LivestreamOriginalIE(InfoExtractor): + IE_NAME = u'livestream:original' + _VALID_URL = r'https?://www\.livestream\.com/(?P<user>[^/]+)/video\?.*?clipId=(?P<id>.*?)(&|$)' + _TEST = { + u'url': u'http://www.livestream.com/dealbook/video?clipId=pla_8aa4a3f1-ba15-46a4-893b-902210e138fb', + u'info_dict': { + u'id': u'pla_8aa4a3f1-ba15-46a4-893b-902210e138fb', + u'ext': u'flv', + u'title': u'Spark 1 (BitCoin) with Cameron Winklevoss & Tyler Winklevoss of Winklevoss Capital', + }, + u'params': { + # rtmp + u'skip_download': True, + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + user = mobj.group('user') + api_url = 'http://x{0}x.api.channel.livestream.com/2.0/clipdetails?extendedInfo=true&id={1}'.format(user, video_id) + + api_response = self._download_webpage(api_url, video_id) + info = xml.etree.ElementTree.fromstring(api_response.encode('utf-8')) + item = info.find('channel').find('item') + ns = {'media': 'http://search.yahoo.com/mrss'} + thumbnail_url = item.find(xpath_with_ns('media:thumbnail', ns)).attrib['url'] + # Remove the extension and number from the path (like 1.jpg) + path = self._search_regex(r'(user-files/.+)_.*?\.jpg$', thumbnail_url, u'path') + + return { + 'id': video_id, + 'title': item.find('title').text, + 'url': 'rtmp://extondemand.livestream.com/ondemand', + 'play_path': 'mp4:trans/dv15/mogulus-{0}.mp4'.format(path), + 'ext': 'flv', + 'thumbnail': thumbnail_url, + } diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index 24a79ae13..3df7f9b85 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -48,7 +48,7 @@ class MTVIE(InfoExtractor): def _transform_rtmp_url(rtmp_video_url): m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url) if not m: - raise ExtractorError(u'Cannot transform RTMP url') + return rtmp_video_url base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/' return base + m.group('finalid') diff --git a/youtube_dl/extractor/redtube.py b/youtube_dl/extractor/redtube.py index 994778e16..3bbda128e 100644 --- a/youtube_dl/extractor/redtube.py +++ b/youtube_dl/extractor/redtube.py @@ -8,7 +8,9 @@ class RedTubeIE(InfoExtractor): _TEST = { u'url': u'http://www.redtube.com/66418', u'file': u'66418.mp4', - u'md5': u'7b8c22b5e7098a3e1c09709df1126d2d', + # md5 varies from time to time, as in + # https://travis-ci.org/rg3/youtube-dl/jobs/14052463#L295 + #u'md5': u'7b8c22b5e7098a3e1c09709df1126d2d', u'info_dict': { u"title": u"Sucked on a toilet", u"age_limit": 18, diff --git a/youtube_dl/extractor/rtlnow.py b/youtube_dl/extractor/rtlnow.py index 9ac7c3be8..2f238de35 100644 --- a/youtube_dl/extractor/rtlnow.py +++ b/youtube_dl/extractor/rtlnow.py @@ -63,18 +63,6 @@ class RTLnowIE(InfoExtractor): }, }, { - u'url': u'http://www.rtlnitronow.de/recht-ordnung/stadtpolizei-frankfurt-gerichtsvollzieher-leipzig.php?film_id=129679&player=1&season=1', - u'file': u'129679.flv', - u'info_dict': { - u'upload_date': u'20131016', - u'title': u'Recht & Ordnung - Stadtpolizei Frankfurt/ Gerichtsvollzieher...', - u'description': u'Stadtpolizei Frankfurt/ Gerichtsvollzieher Leipzig', - }, - u'params': { - u'skip_download': True, - }, - }, - { u'url': u'http://www.n-tvnow.de/top-gear/episode-1-2013-01-01-00-00-00.php?film_id=124903&player=1&season=10', u'file': u'124903.flv', u'info_dict': { diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py index 4717fbb77..83e1f055f 100644 --- a/youtube_dl/extractor/soundcloud.py +++ b/youtube_dl/extractor/soundcloud.py @@ -87,7 +87,7 @@ class SoundcloudIE(InfoExtractor): 'uploader': info['user']['username'], 'upload_date': unified_strdate(info['created_at']), 'title': info['title'], - 'ext': u'mp3', + 'ext': info.get('original_format', u'mp3'), 'description': info['description'], 'thumbnail': thumbnail, } diff --git a/youtube_dl/extractor/southparkstudios.py b/youtube_dl/extractor/southparkstudios.py index b1e96b679..a711531e6 100644 --- a/youtube_dl/extractor/southparkstudios.py +++ b/youtube_dl/extractor/southparkstudios.py @@ -5,21 +5,19 @@ from .mtv import MTVIE, _media_xml_tag class SouthParkStudiosIE(MTVIE): IE_NAME = u'southparkstudios.com' - _VALID_URL = r'https?://www\.southparkstudios\.com/(clips|full-episodes)/(?P<id>.+?)(\?|#|$)' + _VALID_URL = r'(https?://)?(www\.)?(?P<url>southparkstudios\.com/(clips|full-episodes)/(?P<id>.+?)(\?|#|$))' _FEED_URL = 'http://www.southparkstudios.com/feeds/video-player/mrss' - _TEST = { + # Overwrite MTVIE properties we don't want + _TESTS = [{ u'url': u'http://www.southparkstudios.com/clips/104437/bat-daded#tab=featured', u'file': u'a7bff6c2-ed00-11e0-aca6-0026b9414f30.mp4', u'info_dict': { u'title': u'Bat Daded', u'description': u'Randy disqualifies South Park by getting into a fight with Bat Dad.', }, - } - - # Overwrite MTVIE properties we don't want - _TESTS = [] + }] def _get_thumbnail_url(self, uri, itemdoc): search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail')) @@ -31,8 +29,23 @@ class SouthParkStudiosIE(MTVIE): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) + url = u'http://www.' + mobj.group(u'url') video_id = mobj.group('id') webpage = self._download_webpage(url, video_id) mgid = self._search_regex(r'swfobject.embedSWF\(".*?(mgid:.*?)"', webpage, u'mgid') return self._get_videos_info(mgid) + +class SouthparkDeIE(SouthParkStudiosIE): + IE_NAME = u'southpark.de' + _VALID_URL = r'(https?://)?(www\.)?(?P<url>southpark\.de/(clips|alle-episoden)/(?P<id>.+?)(\?|#|$))' + _FEED_URL = 'http://www.southpark.de/feeds/video-player/mrss/' + + _TESTS = [{ + u'url': u'http://www.southpark.de/clips/uygssh/the-government-wont-respect-my-privacy#tab=featured', + u'file': u'85487c96-b3b9-4e39-9127-ad88583d9bf2.mp4', + u'info_dict': { + u'title': u'The Government Won\'t Respect My Privacy', + u'description': u'Cartman explains the benefits of "Shitter" to Stan, Kyle and Craig.', + }, + }] diff --git a/youtube_dl/extractor/spiegel.py b/youtube_dl/extractor/spiegel.py index 13c86401c..6dc2eda6d 100644 --- a/youtube_dl/extractor/spiegel.py +++ b/youtube_dl/extractor/spiegel.py @@ -2,18 +2,27 @@ import re import xml.etree.ElementTree from .common import InfoExtractor +from ..utils import determine_ext class SpiegelIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<videoID>[0-9]+)(?:\.html)?(?:#.*)?$' - _TEST = { + _TESTS = [{ u'url': u'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html', u'file': u'1259285.mp4', u'md5': u'2c2754212136f35fb4b19767d242f66e', u'info_dict': { u"title": u"Vulkanausbruch in Ecuador: Der \"Feuerschlund\" ist wieder aktiv" } - } + }, + { + u'url': u'http://www.spiegel.de/video/schach-wm-videoanalyse-des-fuenften-spiels-video-1309159.html', + u'file': u'1309159.mp4', + u'md5': u'f2cdf638d7aa47654e251e1aee360af1', + u'info_dict': { + u'title': u'Schach-WM in der Videoanalyse: Carlsen nutzt die Fehlgriffe des Titelverteidigers' + } + }] def _real_extract(self, url): m = re.match(self._VALID_URL, url) @@ -21,25 +30,38 @@ class SpiegelIE(InfoExtractor): webpage = self._download_webpage(url, video_id) - video_title = self._html_search_regex(r'<div class="module-title">(.*?)</div>', - webpage, u'title') + video_title = self._html_search_regex( + r'<div class="module-title">(.*?)</div>', webpage, u'title') xml_url = u'http://video2.spiegel.de/flash/' + video_id + u'.xml' - xml_code = self._download_webpage(xml_url, video_id, - note=u'Downloading XML', errnote=u'Failed to download XML') + xml_code = self._download_webpage( + xml_url, video_id, + note=u'Downloading XML', errnote=u'Failed to download XML') idoc = xml.etree.ElementTree.fromstring(xml_code) - last_type = idoc[-1] - filename = last_type.findall('./filename')[0].text - duration = float(last_type.findall('./duration')[0].text) - video_url = 'http://video2.spiegel.de/flash/' + filename - video_ext = filename.rpartition('.')[2] + formats = [ + { + 'format_id': n.tag.rpartition('type')[2], + 'url': u'http://video2.spiegel.de/flash/' + n.find('./filename').text, + 'width': int(n.find('./width').text), + 'height': int(n.find('./height').text), + 'abr': int(n.find('./audiobitrate').text), + 'vbr': int(n.find('./videobitrate').text), + 'vcodec': n.find('./codec').text, + 'acodec': 'MP4A', + } + for n in list(idoc) + # Blacklist type 6, it's extremely LQ and not available on the same server + if n.tag.startswith('type') and n.tag != 'type6' + ] + formats.sort(key=lambda f: f['vbr']) + duration = float(idoc[0].findall('./duration')[0].text) + info = { 'id': video_id, - 'url': video_url, - 'ext': video_ext, 'title': video_title, 'duration': duration, + 'formats': formats, } - return [info] + return info diff --git a/youtube_dl/extractor/ted.py b/youtube_dl/extractor/ted.py index 76cfdfb90..2e497c86e 100644 --- a/youtube_dl/extractor/ted.py +++ b/youtube_dl/extractor/ted.py @@ -43,26 +43,25 @@ class TEDIE(SubtitlesInfoExtractor): self.to_screen(u'Getting info of playlist %s: "%s"' % (playlist_id,name)) return [self._playlist_videos_info(url,name,playlist_id)] - def _playlist_videos_info(self,url,name,playlist_id=0): + + def _playlist_videos_info(self, url, name, playlist_id): '''Returns the videos of the playlist''' - video_RE=r''' - <li\ id="talk_(\d+)"([.\s]*?)data-id="(?P<video_id>\d+)" - ([.\s]*?)data-playlist_item_id="(\d+)" - ([.\s]*?)data-mediaslug="(?P<mediaSlug>.+?)" - ''' - video_name_RE=r'<p\ class="talk-title"><a href="(?P<talk_url>/talks/(.+).html)">(?P<fullname>.+?)</a></p>' - webpage=self._download_webpage(url, playlist_id, 'Downloading playlist webpage') - m_videos=re.finditer(video_RE,webpage,re.VERBOSE) - m_names=re.finditer(video_name_RE,webpage) + + webpage = self._download_webpage( + url, playlist_id, u'Downloading playlist webpage') + matches = re.finditer( + r'<p\s+class="talk-title[^"]*"><a\s+href="(?P<talk_url>/talks/[^"]+\.html)">[^<]*</a></p>', + webpage) playlist_title = self._html_search_regex(r'div class="headline">\s*?<h1>\s*?<span>(.*?)</span>', webpage, 'playlist title') - playlist_entries = [] - for m_video, m_name in zip(m_videos,m_names): - talk_url='http://www.ted.com%s' % m_name.group('talk_url') - playlist_entries.append(self.url_result(talk_url, 'TED')) - return self.playlist_result(playlist_entries, playlist_id = playlist_id, playlist_title = playlist_title) + playlist_entries = [ + self.url_result(u'http://www.ted.com' + m.group('talk_url'), 'TED') + for m in matches + ] + return self.playlist_result( + playlist_entries, playlist_id=playlist_id, playlist_title=playlist_title) def _talk_info(self, url, video_id=0): """Return the video for the talk in the url""" @@ -85,7 +84,7 @@ class TEDIE(SubtitlesInfoExtractor): 'ext': 'mp4', 'url': stream['file'], 'format': stream['id'] - } for stream in info['htmlStreams']] + } for stream in info['htmlStreams']] video_id = info['id'] @@ -95,7 +94,7 @@ class TEDIE(SubtitlesInfoExtractor): self._list_available_subtitles(video_id, webpage) return - info = { + return { 'id': video_id, 'title': title, 'thumbnail': thumbnail, @@ -104,11 +103,6 @@ class TEDIE(SubtitlesInfoExtractor): 'formats': formats, } - # TODO: Remove when #980 has been merged - info.update(info['formats'][-1]) - - return info - def _get_available_subtitles(self, video_id, webpage): try: options = self._search_regex(r'(?:<select name="subtitles_language_select" id="subtitles_language_select">)(.*?)(?:</select>)', webpage, 'subtitles_language_select', flags=re.DOTALL) diff --git a/youtube_dl/extractor/toutv.py b/youtube_dl/extractor/toutv.py new file mode 100644 index 000000000..73ea67da9 --- /dev/null +++ b/youtube_dl/extractor/toutv.py @@ -0,0 +1,75 @@ +# coding: utf-8 +import re +import xml.etree.ElementTree + +from .common import InfoExtractor +from ..utils import ( + ExtractorError, + unified_strdate, +) + + +class TouTvIE(InfoExtractor): + IE_NAME = u'tou.tv' + _VALID_URL = r'https?://www\.tou\.tv/(?P<id>[a-zA-Z0-9_-]+(?:/(?P<episode>S[0-9]+E[0-9]+)))' + + _TEST = { + u'url': u'http://www.tou.tv/30-vies/S04E41', + u'file': u'30-vies_S04E41.mp4', + u'info_dict': { + u'title': u'30 vies Saison 4 / Épisode 41', + u'description': u'md5:da363002db82ccbe4dafeb9cab039b09', + u'age_limit': 8, + u'uploader': u'Groupe des Nouveaux Médias', + u'duration': 1296, + u'upload_date': u'20131118', + u'thumbnail': u'http://static.tou.tv/medias/images/2013-11-18_19_00_00_30VIES_0341_01_L.jpeg', + }, + u'params': { + u'skip_download': True, # Requires rtmpdump + }, + u'xskip': 'Only available in Canada' + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + webpage = self._download_webpage(url, video_id) + + mediaId = self._search_regex( + r'"idMedia":\s*"([^"]+)"', webpage, u'media ID') + + # TODO test from de + streams_url = u'http://release.theplatform.com/content.select?pid=' + mediaId + streams_webpage = self._download_webpage( + streams_url, video_id, note=u'Downloading stream list') + + streams_doc = xml.etree.ElementTree.fromstring( + streams_webpage.encode('utf-8')) + video_url = next(n.text + for n in streams_doc.findall('.//choice/url') + if u'//ad.doubleclick' not in n.text) + if video_url.endswith('/Unavailable.flv'): + raise ExtractorError( + u'Access to this video is blocked from outside of Canada', + expected=True) + + duration_str = self._html_search_meta( + 'video:duration', webpage, u'duration') + duration = int(duration_str) if duration_str else None + upload_date_str = self._html_search_meta( + 'video:release_date', webpage, u'upload date') + upload_date = unified_strdate(upload_date_str) if upload_date_str else None + + return { + 'id': video_id, + 'title': self._og_search_title(webpage), + 'url': video_url, + 'description': self._og_search_description(webpage), + 'uploader': self._dc_search_uploader(webpage), + 'thumbnail': self._og_search_thumbnail(webpage), + 'age_limit': self._media_rating_search(webpage), + 'duration': duration, + 'upload_date': upload_date, + 'ext': 'mp4', + } diff --git a/youtube_dl/extractor/tvp.py b/youtube_dl/extractor/tvp.py index 32e0f5037..bfed9dd04 100644 --- a/youtube_dl/extractor/tvp.py +++ b/youtube_dl/extractor/tvp.py @@ -13,9 +13,10 @@ class TvpIE(InfoExtractor): u'md5': u'148408967a6a468953c0a75cbdaf0d7a', u'file': u'12878238.wmv', u'info_dict': { - u'title': u'31.10.2013', - u'description': u'31.10.2013', + u'title': u'31.10.2013 - Odcinek 2', + u'description': u'31.10.2013 - Odcinek 2', }, + u'skip': u'Download has to use same server IP as extraction. Therefore, a good (load-balancing) DNS resolver will make the download fail.' } def _real_extract(self, url): diff --git a/youtube_dl/extractor/vevo.py b/youtube_dl/extractor/vevo.py index 3f6020f74..4378b1780 100644 --- a/youtube_dl/extractor/vevo.py +++ b/youtube_dl/extractor/vevo.py @@ -78,12 +78,13 @@ class VevoIE(InfoExtractor): continue format_url = self._SMIL_BASE_URL + m.group('path') - format_note = ('%(vcodec)s@%(vbr)4sk, %(acodec)s@%(abr)3sk' % - m.groupdict()) formats.append({ 'url': format_url, 'format_id': u'SMIL_' + m.group('cbr'), - 'format_note': format_note, + 'vcodec': m.group('vcodec'), + 'acodec': m.group('acodec'), + 'vbr': int(m.group('vbr')), + 'abr': int(m.group('abr')), 'ext': m.group('ext'), 'width': int(m.group('width')), 'height': int(m.group('height')), diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 56d30f52a..41838237c 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -139,10 +139,10 @@ class YoutubeBaseInfoExtractor(InfoExtractor): class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): IE_DESC = u'YouTube.com' - _VALID_URL = r"""^ + _VALID_URL = r"""(?x)^ ( - (?:https?://)? # http(s):// (optional) - (?:(?:(?:(?:\w+\.)?youtube(?:-nocookie)?\.com/| + (?:https?://|//)? # http(s):// or protocol-independent URL (optional) + (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/| tube\.majestyc\.net/| youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains (?:.*?\#/)? # handle anchor (#/) redirect urls @@ -363,6 +363,18 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): u"uploader_id": u"justintimberlakeVEVO" } }, + { + u"url": u"//www.YouTube.com/watch?v=yZIXLfi8CZQ", + u"file": u"yZIXLfi8CZQ.mp4", + u"note": u"Embed-only video (#1746)", + u"info_dict": { + u"upload_date": u"20120608", + u"title": u"Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012", + u"description": u"md5:09b78bd971f1e3e289601dfba15ca4f7", + u"uploader": u"SET India", + u"uploader_id": u"setindia" + } + }, ] @@ -370,7 +382,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): def suitable(cls, url): """Receives a URL and returns True if suitable for this IE.""" if YoutubePlaylistIE.suitable(url): return False - return re.match(cls._VALID_URL, url, re.VERBOSE) is not None + return re.match(cls._VALID_URL, url) is not None def __init__(self, *args, **kwargs): super(YoutubeIE, self).__init__(*args, **kwargs) @@ -1019,6 +1031,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): """Turn the encrypted s field into a working signature""" if player_url is not None: + if player_url.startswith(u'//'): + player_url = u'https:' + player_url try: player_id = (player_url, len(s)) if player_id not in self._player_cache: @@ -1270,7 +1284,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): # We simulate the access to the video from www.youtube.com/v/{video_id} # this can be viewed without login into Youtube data = compat_urllib_parse.urlencode({'video_id': video_id, - 'el': 'embedded', + 'el': 'player_embedded', 'gl': 'US', 'hl': 'en', 'eurl': 'https://youtube.googleapis.com/v/' + video_id, @@ -1299,6 +1313,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): else: raise ExtractorError(u'"token" parameter not in video info for unknown reason') + if 'view_count' in video_info: + view_count = int(video_info['view_count'][0]) + else: + view_count = None + # Check for "rental" videos if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: raise ExtractorError(u'"rental" videos not supported') @@ -1487,6 +1506,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): '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, }) return results @@ -1592,20 +1612,31 @@ class YoutubeChannelIE(InfoExtractor): # Download channel page channel_id = mobj.group(1) video_ids = [] + url = 'https://www.youtube.com/channel/%s/videos' % channel_id + channel_page = self._download_webpage(url, channel_id) + if re.search(r'channel-header-autogenerated-label', channel_page) is not None: + autogenerated = True + else: + autogenerated = False - # 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) - - ids_in_page = self.extract_videos_from_page(page['content_html']) - video_ids.extend(ids_in_page) - - if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: - break + if autogenerated: + # The videos are contained in a single page + # the ajax pages can't be used, they are empty + video_ids = self.extract_videos_from_page(channel_page) + else: + # 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) + + ids_in_page = self.extract_videos_from_page(page['content_html']) + video_ids.extend(ids_in_page) + + if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: + break self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) |