diff options
Diffstat (limited to 'youtube_dl/extractor')
30 files changed, 836 insertions, 201 deletions
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 8a063648c..cebb8717f 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -8,6 +8,7 @@ from .arte import ( ArteTVPlus7IE, ArteTVCreativeIE, ArteTVFutureIE, + ArteTVDDCIE, ) from .auengine import AUEngineIE from .bambuser import BambuserIE, BambuserChannelIE @@ -19,6 +20,8 @@ from .brightcove import BrightcoveIE from .c56 import C56IE from .canalplus import CanalplusIE from .canalc2 import Canalc2IE +from .cbs import CBSIE +from .channel9 import Channel9IE from .cinemassacre import CinemassacreIE from .clipfish import ClipfishIE from .clipsyndicate import ClipsyndicateIE @@ -100,6 +103,7 @@ from .myvideo import MyVideoIE from .naver import NaverIE from .nba import NBAIE from .nbc import NBCNewsIE +from .ndtv import NDTVIE from .newgrounds import NewgroundsIE from .nhl import NHLIE, NHLVideocenterIE from .niconico import NiconicoIE @@ -110,6 +114,7 @@ from .orf import ORFIE from .pbs import PBSIE from .photobucket import PhotobucketIE from .podomatic import PodomaticIE +from .pornhd import PornHdIE from .pornhub import PornHubIE from .pornotube import PornotubeIE from .pyvideo import PyvideoIE @@ -128,6 +133,7 @@ from .smotri import ( SmotriIE, SmotriCommunityIE, SmotriUserIE, + SmotriBroadcastIE, ) from .sohu import SohuIE from .soundcloud import SoundcloudIE, SoundcloudSetIE, SoundcloudUserIE @@ -210,6 +216,7 @@ from .youtube import ( YoutubeWatchLaterIE, YoutubeFavouritesIE, YoutubeHistoryIE, + YoutubeTopListIE, ) from .zdf import ZDFIE diff --git a/youtube_dl/extractor/appletrailers.py b/youtube_dl/extractor/appletrailers.py index a527f10de..ef5644aa5 100644 --- a/youtube_dl/extractor/appletrailers.py +++ b/youtube_dl/extractor/appletrailers.py @@ -1,5 +1,4 @@ import re -import xml.etree.ElementTree import json from .common import InfoExtractor @@ -65,18 +64,18 @@ class AppleTrailersIE(InfoExtractor): uploader_id = mobj.group('company') playlist_url = compat_urlparse.urljoin(url, u'includes/playlists/itunes.inc') - playlist_snippet = self._download_webpage(playlist_url, movie) - playlist_cleaned = re.sub(r'(?s)<script[^<]*?>.*?</script>', u'', playlist_snippet) - playlist_cleaned = re.sub(r'<img ([^<]*?)>', r'<img \1/>', playlist_cleaned) - # The ' in the onClick attributes are not escaped, it couldn't be parsed - # with xml.etree.ElementTree.fromstring - # like: http://trailers.apple.com/trailers/wb/gravity/ - def _clean_json(m): - return u'iTunes.playURL(%s);' % m.group(1).replace('\'', ''') - playlist_cleaned = re.sub(self._JSON_RE, _clean_json, playlist_cleaned) - playlist_html = u'<html>' + playlist_cleaned + u'</html>' + def fix_html(s): + s = re.sub(r'(?s)<script[^<]*?>.*?</script>', u'', s) + s = re.sub(r'<img ([^<]*?)>', r'<img \1/>', s) + # The ' in the onClick attributes are not escaped, it couldn't be parsed + # like: http://trailers.apple.com/trailers/wb/gravity/ + def _clean_json(m): + return u'iTunes.playURL(%s);' % m.group(1).replace('\'', ''') + s = re.sub(self._JSON_RE, _clean_json, s) + s = u'<html>' + s + u'</html>' + return s + doc = self._download_xml(playlist_url, movie, transform_source=fix_html) - doc = xml.etree.ElementTree.fromstring(playlist_html) playlist = [] for li in doc.findall('./div/ul/li'): on_click = li.find('.//a').attrib['onClick'] diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index 56a5d009f..4b7bef775 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -10,6 +10,7 @@ from ..utils import ( determine_ext, get_element_by_id, compat_str, + get_element_by_attribute, ) # There are different sources of video in arte.tv, the extraction process @@ -142,7 +143,9 @@ class ArteTVPlus7IE(InfoExtractor): def _extract_from_webpage(self, webpage, video_id, lang): json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url') + return self._extract_from_json_url(json_url, video_id, lang) + def _extract_from_json_url(self, json_url, video_id, lang): json_info = self._download_webpage(json_url, video_id, 'Downloading info json') self.report_extraction(video_id) info = json.loads(json_info) @@ -257,3 +260,35 @@ class ArteTVFutureIE(ArteTVPlus7IE): webpage = self._download_webpage(url, anchor_id) row = get_element_by_id(anchor_id, webpage) return self._extract_from_webpage(row, anchor_id, lang) + + +class ArteTVDDCIE(ArteTVPlus7IE): + IE_NAME = u'arte.tv:ddc' + _VALID_URL = r'http?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>.+)' + + _TEST = { + u'url': u'http://ddc.arte.tv/folge/neues-aus-mauretanien', + u'file': u'049881-009_PLUS7-D.flv', + u'info_dict': { + u'title': u'Mit offenen Karten', + u'description': u'md5:57929b0eaeddeb8a0c983f58e9ebd3b6', + u'upload_date': u'20131207', + }, + u'params': { + # rtmp download + u'skip_download': True, + }, + } + + def _real_extract(self, url): + video_id, lang = self._extract_url_info(url) + if lang == 'folge': + lang = 'de' + elif lang == 'emission': + lang = 'fr' + webpage = self._download_webpage(url, video_id) + scriptElement = get_element_by_attribute('class', 'visu_video_block', webpage) + script_url = self._html_search_regex(r'src="(.*?)"', scriptElement, 'script url') + javascriptPlayerGenerator = self._download_webpage(script_url, video_id, 'Download javascript player generator') + json_url = self._search_regex(r"json_url=(.*)&rendering_place.*", javascriptPlayerGenerator, 'json url') + return self._extract_from_json_url(json_url, video_id, lang) diff --git a/youtube_dl/extractor/bliptv.py b/youtube_dl/extractor/bliptv.py index 493504f75..5e33a69df 100644 --- a/youtube_dl/extractor/bliptv.py +++ b/youtube_dl/extractor/bliptv.py @@ -51,8 +51,7 @@ class BlipTVIE(InfoExtractor): url = 'http://blip.tv/play/g_%s' % api_mobj.group('video_id') urlp = compat_urllib_parse_urlparse(url) if urlp.path.startswith('/play/'): - request = compat_urllib_request.Request(url) - response = compat_urllib_request.urlopen(request) + response = self._request_webpage(url, None, False) redirecturl = response.geturl() rurlp = compat_urllib_parse_urlparse(redirecturl) file_id = compat_parse_qs(rurlp.fragment)['file'][0].rpartition('/')[2] @@ -69,25 +68,23 @@ class BlipTVIE(InfoExtractor): request.add_header('User-Agent', 'iTunes/10.6.1') self.report_extraction(mobj.group(1)) info = None - try: - urlh = compat_urllib_request.urlopen(request) - if urlh.headers.get('Content-Type', '').startswith('video/'): # Direct download - basename = url.split('/')[-1] - title,ext = os.path.splitext(basename) - title = title.decode('UTF-8') - ext = ext.replace('.', '') - self.report_direct_download(title) - info = { - 'id': title, - 'url': url, - 'uploader': None, - 'upload_date': None, - 'title': title, - 'ext': ext, - 'urlhandle': urlh - } - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'ERROR: unable to download video info webpage: %s' % compat_str(err)) + urlh = self._request_webpage(request, None, False, + u'unable to download video info webpage') + if urlh.headers.get('Content-Type', '').startswith('video/'): # Direct download + basename = url.split('/')[-1] + title,ext = os.path.splitext(basename) + title = title.decode('UTF-8') + ext = ext.replace('.', '') + self.report_direct_download(title) + info = { + 'id': title, + 'url': url, + 'uploader': None, + 'upload_date': None, + 'title': title, + 'ext': ext, + 'urlhandle': urlh + } if info is None: # Regular URL try: json_code_bytes = urlh.read() diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py index 66fe0ac9a..b1b7526ca 100644 --- a/youtube_dl/extractor/brightcove.py +++ b/youtube_dl/extractor/brightcove.py @@ -55,6 +55,18 @@ class BrightcoveIE(InfoExtractor): u'uploader': u'Mashable', }, }, + { + # test that the default referer works + # from http://national.ballet.ca/interact/video/Lost_in_Motion_II/ + u'url': u'http://link.brightcove.com/services/player/bcpid756015033001?bckey=AQ~~,AAAApYJi_Ck~,GxhXCegT1Dp39ilhXuxMJxasUhVNZiil&bctid=2878862109001', + u'info_dict': { + u'id': u'2878862109001', + u'ext': u'mp4', + u'title': u'Lost in Motion II', + u'description': u'md5:363109c02998fee92ec02211bd8000df', + u'uploader': u'National Ballet of Canada', + }, + }, ] @classmethod @@ -118,17 +130,21 @@ class BrightcoveIE(InfoExtractor): videoPlayer = query.get('@videoPlayer') if videoPlayer: - return self._get_video_info(videoPlayer[0], query_str, query) + return self._get_video_info(videoPlayer[0], query_str, query, + # We set the original url as the default 'Referer' header + referer=url) else: player_key = query['playerKey'] return self._get_playlist_info(player_key[0]) - def _get_video_info(self, video_id, query_str, query): + def _get_video_info(self, video_id, query_str, query, referer=None): request_url = self._FEDERATED_URL_TEMPLATE % query_str req = compat_urllib_request.Request(request_url) linkBase = query.get('linkBaseURL') if linkBase is not None: - req.add_header('Referer', linkBase[0]) + referer = linkBase[0] + if referer is not None: + req.add_header('Referer', referer) webpage = self._download_webpage(req, video_id) self.report_extraction(video_id) diff --git a/youtube_dl/extractor/cbs.py b/youtube_dl/extractor/cbs.py new file mode 100644 index 000000000..ac0315853 --- /dev/null +++ b/youtube_dl/extractor/cbs.py @@ -0,0 +1,30 @@ +import re + +from .common import InfoExtractor + + +class CBSIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?cbs\.com/shows/[^/]+/video/(?P<id>[^/]+)/.*' + + _TEST = { + u'url': u'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/', + u'file': u'4JUVEwq3wUT7.flv', + u'info_dict': { + u'title': u'Connect Chat feat. Garth Brooks', + u'description': u'Connect with country music singer Garth Brooks, as he chats with fans on Wednesday November 27, 2013. Be sure to tune in to Garth Brooks: Live from Las Vegas, Friday November 29, at 9/8c on CBS!', + u'duration': 1495, + }, + u'params': { + # rtmp download + u'skip_download': True, + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + webpage = self._download_webpage(url, video_id) + real_id = self._search_regex( + r"video\.settings\.pid\s*=\s*'([^']+)';", + webpage, u'real video ID') + return self.url_result(u'theplatform:%s' % real_id) diff --git a/youtube_dl/extractor/channel9.py b/youtube_dl/extractor/channel9.py new file mode 100644 index 000000000..ae70ea229 --- /dev/null +++ b/youtube_dl/extractor/channel9.py @@ -0,0 +1,267 @@ +# encoding: utf-8 + +import re + +from .common import InfoExtractor +from ..utils import ExtractorError + +class Channel9IE(InfoExtractor): + ''' + Common extractor for channel9.msdn.com. + + The type of provided URL (video or playlist) is determined according to + meta Search.PageType from web page HTML rather than URL itself, as it is + not always possible to do. + ''' + IE_DESC = u'Channel 9' + IE_NAME = u'channel9' + _VALID_URL = r'^https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+)/?' + + _TESTS = [ + { + u'url': u'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002', + u'file': u'Events_TechEd_Australia_2013_KOS002.mp4', + u'md5': u'bbd75296ba47916b754e73c3a4bbdf10', + u'info_dict': { + u'title': u'Developer Kick-Off Session: Stuff We Love', + u'description': u'md5:c08d72240b7c87fcecafe2692f80e35f', + u'duration': 4576, + u'thumbnail': u'http://media.ch9.ms/ch9/9d51/03902f2d-fc97-4d3c-b195-0bfe15a19d51/KOS002_220.jpg', + u'session_code': u'KOS002', + u'session_day': u'Day 1', + u'session_room': u'Arena 1A', + u'session_speakers': [ u'Ed Blankenship', u'Andrew Coates', u'Brady Gaster', u'Patrick Klug', u'Mads Kristensen' ], + }, + }, + { + u'url': u'http://channel9.msdn.com/posts/Self-service-BI-with-Power-BI-nuclear-testing', + u'file': u'posts_Self-service-BI-with-Power-BI-nuclear-testing.mp4', + u'md5': u'b43ee4529d111bc37ba7ee4f34813e68', + u'info_dict': { + u'title': u'Self-service BI with Power BI - nuclear testing', + u'description': u'md5:d1e6ecaafa7fb52a2cacdf9599829f5b', + u'duration': 1540, + u'thumbnail': u'http://media.ch9.ms/ch9/87e1/0300391f-a455-4c72-bec3-4422f19287e1/selfservicenuk_512.jpg', + u'authors': [ u'Mike Wilmot' ], + }, + } + ] + + _RSS_URL = 'http://channel9.msdn.com/%s/RSS' + + # Sorted by quality + _known_formats = ['MP3', 'MP4', 'Mid Quality WMV', 'Mid Quality MP4', 'High Quality WMV', 'High Quality MP4'] + + def _restore_bytes(self, formatted_size): + if not formatted_size: + return 0 + m = re.match(r'^(?P<size>\d+(?:\.\d+)?)\s+(?P<units>[a-zA-Z]+)', formatted_size) + if not m: + return 0 + units = m.group('units') + try: + exponent = [u'B', u'KB', u'MB', u'GB', u'TB', u'PB', u'EB', u'ZB', u'YB'].index(units.upper()) + except ValueError: + return 0 + size = float(m.group('size')) + return int(size * (1024 ** exponent)) + + def _formats_from_html(self, html): + FORMAT_REGEX = r''' + (?x) + <a\s+href="(?P<url>[^"]+)">(?P<quality>[^<]+)</a>\s* + <span\s+class="usage">\((?P<note>[^\)]+)\)</span>\s* + (?:<div\s+class="popup\s+rounded">\s* + <h3>File\s+size</h3>\s*(?P<filesize>.*?)\s* + </div>)? # File size part may be missing + ''' + # Extract known formats + formats = [{'url': x.group('url'), + 'format_id': x.group('quality'), + 'format_note': x.group('note'), + 'format': '%s (%s)' % (x.group('quality'), x.group('note')), + 'filesize': self._restore_bytes(x.group('filesize')), # File size is approximate + } for x in list(re.finditer(FORMAT_REGEX, html)) if x.group('quality') in self._known_formats] + # Sort according to known formats list + formats.sort(key=lambda fmt: self._known_formats.index(fmt['format_id'])) + return formats + + def _extract_title(self, html): + title = self._html_search_meta(u'title', html, u'title') + if title is None: + title = self._og_search_title(html) + TITLE_SUFFIX = u' (Channel 9)' + if title is not None and title.endswith(TITLE_SUFFIX): + title = title[:-len(TITLE_SUFFIX)] + return title + + def _extract_description(self, html): + DESCRIPTION_REGEX = r'''(?sx) + <div\s+class="entry-content">\s* + <div\s+id="entry-body">\s* + (?P<description>.+?)\s* + </div>\s* + </div> + ''' + m = re.search(DESCRIPTION_REGEX, html) + if m is not None: + return m.group('description') + return self._html_search_meta(u'description', html, u'description') + + def _extract_duration(self, html): + m = re.search(r'data-video_duration="(?P<hours>\d{2}):(?P<minutes>\d{2}):(?P<seconds>\d{2})"', html) + return ((int(m.group('hours')) * 60 * 60) + (int(m.group('minutes')) * 60) + int(m.group('seconds'))) if m else None + + def _extract_slides(self, html): + m = re.search(r'<a href="(?P<slidesurl>[^"]+)" class="slides">Slides</a>', html) + return m.group('slidesurl') if m is not None else None + + def _extract_zip(self, html): + m = re.search(r'<a href="(?P<zipurl>[^"]+)" class="zip">Zip</a>', html) + return m.group('zipurl') if m is not None else None + + def _extract_avg_rating(self, html): + m = re.search(r'<p class="avg-rating">Avg Rating: <span>(?P<avgrating>[^<]+)</span></p>', html) + return float(m.group('avgrating')) if m is not None else 0 + + def _extract_rating_count(self, html): + m = re.search(r'<div class="rating-count">\((?P<ratingcount>[^<]+)\)</div>', html) + return int(self._fix_count(m.group('ratingcount'))) if m is not None else 0 + + def _extract_view_count(self, html): + m = re.search(r'<li class="views">\s*<span class="count">(?P<viewcount>[^<]+)</span> Views\s*</li>', html) + return int(self._fix_count(m.group('viewcount'))) if m is not None else 0 + + def _extract_comment_count(self, html): + m = re.search(r'<li class="comments">\s*<a href="#comments">\s*<span class="count">(?P<commentcount>[^<]+)</span> Comments\s*</a>\s*</li>', html) + return int(self._fix_count(m.group('commentcount'))) if m is not None else 0 + + def _fix_count(self, count): + return int(str(count).replace(',', '')) if count is not None else None + + def _extract_authors(self, html): + m = re.search(r'(?s)<li class="author">(.*?)</li>', html) + if m is None: + return None + return re.findall(r'<a href="/Niners/[^"]+">([^<]+)</a>', m.group(1)) + + def _extract_session_code(self, html): + m = re.search(r'<li class="code">\s*(?P<code>.+?)\s*</li>', html) + return m.group('code') if m is not None else None + + def _extract_session_day(self, html): + m = re.search(r'<li class="day">\s*<a href="/Events/[^"]+">(?P<day>[^<]+)</a>\s*</li>', html) + return m.group('day') if m is not None else None + + def _extract_session_room(self, html): + m = re.search(r'<li class="room">\s*(?P<room>.+?)\s*</li>', html) + return m.group('room') if m is not None else None + + def _extract_session_speakers(self, html): + return re.findall(r'<a href="/Events/Speakers/[^"]+">([^<]+)</a>', html) + + def _extract_content(self, html, content_path): + # Look for downloadable content + formats = self._formats_from_html(html) + slides = self._extract_slides(html) + zip_ = self._extract_zip(html) + + # Nothing to download + if len(formats) == 0 and slides is None and zip_ is None: + self._downloader.report_warning(u'None of recording, slides or zip are available for %s' % content_path) + return + + # Extract meta + title = self._extract_title(html) + description = self._extract_description(html) + thumbnail = self._og_search_thumbnail(html) + duration = self._extract_duration(html) + avg_rating = self._extract_avg_rating(html) + rating_count = self._extract_rating_count(html) + view_count = self._extract_view_count(html) + comment_count = self._extract_comment_count(html) + + common = {'_type': 'video', + 'id': content_path, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'avg_rating': avg_rating, + 'rating_count': rating_count, + 'view_count': view_count, + 'comment_count': comment_count, + } + + result = [] + + if slides is not None: + d = common.copy() + d.update({ 'title': title + '-Slides', 'url': slides }) + result.append(d) + + if zip_ is not None: + d = common.copy() + d.update({ 'title': title + '-Zip', 'url': zip_ }) + result.append(d) + + if len(formats) > 0: + d = common.copy() + d.update({ 'title': title, 'formats': formats }) + result.append(d) + + return result + + def _extract_entry_item(self, html, content_path): + contents = self._extract_content(html, content_path) + if contents is None: + return contents + + authors = self._extract_authors(html) + + for content in contents: + content['authors'] = authors + + return contents + + def _extract_session(self, html, content_path): + contents = self._extract_content(html, content_path) + if contents is None: + return contents + + session_meta = {'session_code': self._extract_session_code(html), + 'session_day': self._extract_session_day(html), + 'session_room': self._extract_session_room(html), + 'session_speakers': self._extract_session_speakers(html), + } + + for content in contents: + content.update(session_meta) + + return contents + + def _extract_list(self, content_path): + rss = self._download_xml(self._RSS_URL % content_path, content_path, u'Downloading RSS') + entries = [self.url_result(session_url.text, 'Channel9') + for session_url in rss.findall('./channel/item/link')] + title_text = rss.find('./channel/title').text + return self.playlist_result(entries, content_path, title_text) + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + content_path = mobj.group('contentpath') + + webpage = self._download_webpage(url, content_path, u'Downloading web page') + + page_type_m = re.search(r'<meta name="Search.PageType" content="(?P<pagetype>[^"]+)"/>', webpage) + if page_type_m is None: + raise ExtractorError(u'Search.PageType not found, don\'t know how to process this page', expected=True) + + page_type = page_type_m.group('pagetype') + if page_type == 'List': # List page, may contain list of 'item'-like objects + return self._extract_list(content_path) + elif page_type == 'Entry.Item': # Any 'item'-like page, may contain downloadable content + return self._extract_entry_item(webpage, content_path) + elif page_type == 'Session': # Event session page, may contain downloadable content + return self._extract_session(webpage, content_path) + else: + raise ExtractorError(u'Unexpected Search.PageType %s' % page_type, expected=True)
\ No newline at end of file diff --git a/youtube_dl/extractor/clipsyndicate.py b/youtube_dl/extractor/clipsyndicate.py index d4fc86973..c60089ad3 100644 --- a/youtube_dl/extractor/clipsyndicate.py +++ b/youtube_dl/extractor/clipsyndicate.py @@ -1,9 +1,9 @@ import re -import xml.etree.ElementTree from .common import InfoExtractor from ..utils import ( find_xpath_attr, + fix_xml_all_ampersand, ) @@ -30,12 +30,10 @@ class ClipsyndicateIE(InfoExtractor): # it includes a required token flvars = self._search_regex(r'flvars: "(.*?)"', js_player, u'flvars') - playlist_page = self._download_webpage( + pdoc = self._download_xml( 'http://eplayer.clipsyndicate.com/osmf/playlist?%s' % flvars, - video_id, u'Downloading video info') - # Fix broken xml - playlist_page = re.sub('&', '&', playlist_page) - pdoc = xml.etree.ElementTree.fromstring(playlist_page.encode('utf-8')) + video_id, u'Downloading video info', + transform_source=fix_xml_all_ampersand) track_doc = pdoc.find('trackList/track') def find_param(name): diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 92a0c5050..fe8ce9e6c 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -34,8 +34,8 @@ class InfoExtractor(object): The dictionaries must include the following fields: id: Video identifier. - url: Final video URL. title: Video title, unescaped. + url: Final video URL. ext: Video filename extension. Instead of url and ext, formats can also specified. @@ -54,6 +54,7 @@ class InfoExtractor(object): player_url: SWF Player URL (used for rtmpdump). subtitles: The subtitle file contents as a dictionary in the format {language: subtitles}. + duration: Length of the video in seconds, as an integer. view_count: How many users have watched the video on the platform. like_count: Number of positive ratings of the video dislike_count: Number of negative ratings of the video @@ -154,27 +155,38 @@ class InfoExtractor(object): def IE_NAME(self): return type(self).__name__[:-2] - def _request_webpage(self, url_or_request, video_id, note=None, errnote=None): + def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): """ Returns the response handle """ if note is None: self.report_download_webpage(video_id) elif note is not False: - self.to_screen(u'%s: %s' % (video_id, note)) + if video_id is None: + self.to_screen(u'%s' % (note,)) + else: + self.to_screen(u'%s: %s' % (video_id, note)) try: return self._downloader.urlopen(url_or_request) except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: if errnote is None: errnote = u'Unable to download webpage' - raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2], cause=err) + errmsg = u'%s: %s' % (errnote, compat_str(err)) + if fatal: + raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) + else: + self._downloader.report_warning(errmsg) + return False - def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None): + def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True): """ Returns a tuple (page content as string, URL handle) """ # Strip hashes from the URL (#1038) if isinstance(url_or_request, (compat_str, str)): url_or_request = url_or_request.partition('#')[0] - urlh = self._request_webpage(url_or_request, video_id, note, errnote) + urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal) + if urlh is False: + assert not fatal + return False content_type = urlh.headers.get('Content-Type', '') webpage_bytes = urlh.read() m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) @@ -209,14 +221,22 @@ class InfoExtractor(object): content = webpage_bytes.decode(encoding, 'replace') return (content, urlh) - def _download_webpage(self, url_or_request, video_id, note=None, errnote=None): + def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): """ Returns the data of the page as a string """ - return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0] + res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal) + if res is False: + return res + else: + content, _ = res + return content def _download_xml(self, url_or_request, video_id, - note=u'Downloading XML', errnote=u'Unable to download XML'): + note=u'Downloading XML', errnote=u'Unable to download XML', + transform_source=None): """Return the xml as an xml.etree.ElementTree.Element""" xml_string = self._download_webpage(url_or_request, video_id, note, errnote) + if transform_source: + xml_string = transform_source(xml_string) return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8')) def to_screen(self, msg): diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index 3bd0b862c..6685c94a3 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -28,7 +28,7 @@ class DailymotionBaseInfoExtractor(InfoExtractor): class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): """Information Extractor for Dailymotion""" - _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/(?:embed/)?video/([^/]+)' + _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)' IE_NAME = u'dailymotion' _FORMATS = [ @@ -81,7 +81,7 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): # Extract id and simplified title from URL mobj = re.match(self._VALID_URL, url) - video_id = mobj.group(1).split('_')[0].split('?')[0] + video_id = mobj.group('id') url = 'http://www.dailymotion.com/video/%s' % video_id @@ -101,10 +101,6 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): self.to_screen(u'Vevo video detected: %s' % vevo_id) return self.url_result(u'vevo:%s' % vevo_id, ie='Vevo') - video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>', - # Looking for official user - r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'], - webpage, 'video uploader', fatal=False) age_limit = self._rta_search(webpage) video_upload_date = None @@ -147,13 +143,15 @@ class DailymotionIE(DailymotionBaseInfoExtractor, SubtitlesInfoExtractor): self._list_available_subtitles(video_id, webpage) return - view_count = str_to_int(self._search_regex( - r'video_views_value[^>]+>([\d\.,]+)<', webpage, u'view count')) + view_count = self._search_regex( + r'video_views_count[^>]+>\s+([\d\.,]+)', webpage, u'view count', fatal=False) + if view_count is not None: + view_count = str_to_int(view_count) return { 'id': video_id, 'formats': formats, - 'uploader': video_uploader, + 'uploader': info['owner_screenname'], 'upload_date': video_upload_date, 'title': self._og_search_title(webpage), 'subtitles': video_subtitles, diff --git a/youtube_dl/extractor/daum.py b/youtube_dl/extractor/daum.py index d418ce4a8..4876ecb48 100644 --- a/youtube_dl/extractor/daum.py +++ b/youtube_dl/extractor/daum.py @@ -9,7 +9,7 @@ from ..utils import ( class DaumIE(InfoExtractor): - _VALID_URL = r'https?://tvpot\.daum\.net/.*?clipid=(?P<id>\d+)' + _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/.*?clipid=(?P<id>\d+)' IE_NAME = u'daum.net' _TEST = { diff --git a/youtube_dl/extractor/ign.py b/youtube_dl/extractor/ign.py index 57b79a336..381af91e4 100644 --- a/youtube_dl/extractor/ign.py +++ b/youtube_dl/extractor/ign.py @@ -44,7 +44,7 @@ class IGNIE(InfoExtractor): { u'file': u'638672ee848ae4ff108df2a296418ee2.mp4', u'info_dict': { - u'title': u'GTA 5\'s Twisted Beauty in Super Slow Motion', + u'title': u'26 Twisted Moments from GTA 5 in Slow Motion', u'description': u'The twisted beauty of GTA 5 in stunning slow motion.', }, }, diff --git a/youtube_dl/extractor/metacafe.py b/youtube_dl/extractor/metacafe.py index e59bdd604..99d3c83a5 100644 --- a/youtube_dl/extractor/metacafe.py +++ b/youtube_dl/extractor/metacafe.py @@ -1,14 +1,10 @@ import re -import socket from .common import InfoExtractor from ..utils import ( - compat_http_client, compat_parse_qs, - compat_urllib_error, compat_urllib_parse, compat_urllib_request, - compat_str, determine_ext, ExtractorError, ) @@ -93,12 +89,8 @@ class MetacafeIE(InfoExtractor): def _real_initialize(self): # Retrieve disclaimer - request = compat_urllib_request.Request(self._DISCLAIMER) - try: - self.report_disclaimer() - compat_urllib_request.urlopen(request).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to retrieve disclaimer: %s' % compat_str(err)) + self.report_disclaimer() + self._download_webpage(self._DISCLAIMER, None, False, u'Unable to retrieve disclaimer') # Confirm age disclaimer_form = { @@ -107,11 +99,8 @@ class MetacafeIE(InfoExtractor): } request = compat_urllib_request.Request(self._FILTER_POST, compat_urllib_parse.urlencode(disclaimer_form)) request.add_header('Content-Type', 'application/x-www-form-urlencoded') - try: - self.report_age_confirmation() - compat_urllib_request.urlopen(request).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) + self.report_age_confirmation() + self._download_webpage(request, None, False, u'Unable to confirm age') def _real_extract(self, url): # Extract id and simplified title from URL diff --git a/youtube_dl/extractor/metacritic.py b/youtube_dl/extractor/metacritic.py index 6b95b4998..e560c1d35 100644 --- a/youtube_dl/extractor/metacritic.py +++ b/youtube_dl/extractor/metacritic.py @@ -1,8 +1,10 @@ import re -import xml.etree.ElementTree import operator from .common import InfoExtractor +from ..utils import ( + fix_xml_all_ampersand, +) class MetacriticIE(InfoExtractor): @@ -23,9 +25,8 @@ class MetacriticIE(InfoExtractor): video_id = mobj.group('id') webpage = self._download_webpage(url, video_id) # The xml is not well formatted, there are raw '&' - info_xml = self._download_webpage('http://www.metacritic.com/video_data?video=' + video_id, - video_id, u'Downloading info xml').replace('&', '&') - info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8')) + info = self._download_xml('http://www.metacritic.com/video_data?video=' + video_id, + video_id, u'Downloading info xml', transform_source=fix_xml_all_ampersand) clip = next(c for c in info.findall('playList/clip') if c.find('id').text == video_id) formats = [] diff --git a/youtube_dl/extractor/mixcloud.py b/youtube_dl/extractor/mixcloud.py index e2baf44d7..125d81551 100644 --- a/youtube_dl/extractor/mixcloud.py +++ b/youtube_dl/extractor/mixcloud.py @@ -1,13 +1,10 @@ import json import re -import socket from .common import InfoExtractor from ..utils import ( - compat_http_client, - compat_urllib_error, - compat_urllib_request, unified_strdate, + ExtractorError, ) @@ -31,13 +28,18 @@ class MixcloudIE(InfoExtractor): """Returns 1st active url from list""" for url in url_list: try: - compat_urllib_request.urlopen(url) + # We only want to know if the request succeed + # don't download the whole file + self._request_webpage(url, None, False) return url - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error): + except ExtractorError: url = None return None + def _get_url(self, template_url): + return self.check_urls(template_url % i for i in range(30)) + def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -53,13 +55,18 @@ class MixcloudIE(InfoExtractor): preview_url = self._search_regex(r'data-preview-url="(.+?)"', webpage, u'preview url') song_url = preview_url.replace('/previews/', '/cloudcasts/originals/') template_url = re.sub(r'(stream\d*)', 'stream%d', song_url) - final_song_url = self.check_urls(template_url % i for i in range(30)) + final_song_url = self._get_url(template_url) + if final_song_url is None: + self.to_screen('Trying with m4a extension') + template_url = template_url.replace('.mp3', '.m4a').replace('originals/', 'm4a/64/') + final_song_url = self._get_url(template_url) + if final_song_url is None: + raise ExtractorError(u'Unable to extract track url') return { 'id': track_id, 'title': info['name'], 'url': final_song_url, - 'ext': 'mp3', 'description': info.get('description'), 'thumbnail': info['pictures'].get('extra_large'), 'uploader': info['user']['name'], diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index 6b3feb560..5b2bd9633 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -82,8 +82,13 @@ class MTVServicesInfoExtractor(InfoExtractor): def _get_videos_info(self, uri): video_id = self._id_from_uri(uri) data = compat_urllib_parse.urlencode({'uri': uri}) - idoc = self._download_xml(self._FEED_URL +'?' + data, video_id, - u'Downloading info') + + def fix_ampersand(s): + """ Fix unencoded ampersand in XML """ + return s.replace(u'& ', '& ') + idoc = self._download_xml( + self._FEED_URL + '?' + data, video_id, + u'Downloading info', transform_source=fix_ampersand) return [self._get_video_info(item) for item in idoc.findall('.//item')] diff --git a/youtube_dl/extractor/naver.py b/youtube_dl/extractor/naver.py index c012ec0cf..4cab30631 100644 --- a/youtube_dl/extractor/naver.py +++ b/youtube_dl/extractor/naver.py @@ -9,7 +9,7 @@ from ..utils import ( class NaverIE(InfoExtractor): - _VALID_URL = r'https?://tvcast\.naver\.com/v/(?P<id>\d+)' + _VALID_URL = r'https?://(?:m\.)?tvcast\.naver\.com/v/(?P<id>\d+)' _TEST = { u'url': u'http://tvcast.naver.com/v/81652', diff --git a/youtube_dl/extractor/ndtv.py b/youtube_dl/extractor/ndtv.py new file mode 100644 index 000000000..2e8501f99 --- /dev/null +++ b/youtube_dl/extractor/ndtv.py @@ -0,0 +1,66 @@ +import json +import re +import time + +from .common import InfoExtractor +from ..utils import month_by_name + + +class NDTVIE(InfoExtractor): + _VALID_URL = r'^https?://(?:www\.)?ndtv\.com/video/player/[^/]*/[^/]*/(?P<id>[a-z0-9]+)' + + _TEST = { + u"url": u"http://www.ndtv.com/video/player/news/ndtv-exclusive-don-t-need-character-certificate-from-rahul-gandhi-says-arvind-kejriwal/300710", + u"file": u"300710.mp4", + u"md5": u"39f992dbe5fb531c395d8bbedb1e5e88", + u"info_dict": { + u"title": u"NDTV exclusive: Don't need character certificate from Rahul Gandhi, says Arvind Kejriwal", + u"description": u"In an exclusive interview to NDTV, Aam Aadmi Party's Arvind Kejriwal says it makes no difference to him that Rahul Gandhi said the Congress needs to learn from his party.", + u"upload_date": u"20131208", + u"duration": 1327, + u"thumbnail": u"http://i.ndtvimg.com/video/images/vod/medium/2013-12/big_300710_1386518307.jpg", + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + webpage = self._download_webpage(url, video_id) + + filename = self._search_regex( + r"__filename='([^']+)'", webpage, u'video filename') + video_url = (u'http://bitcast-b.bitgravity.com/ndtvod/23372/ndtv/%s' % + filename) + + duration_str = filename = self._search_regex( + r"__duration='([^']+)'", webpage, u'duration', fatal=False) + duration = None if duration_str is None else int(duration_str) + + date_m = re.search(r'''(?x) + <p\s+class="vod_dateline">\s* + Published\s+On:\s* + (?P<monthname>[A-Za-z]+)\s+(?P<day>[0-9]+),\s*(?P<year>[0-9]+) + ''', webpage) + upload_date = None + assert date_m + if date_m is not None: + month = month_by_name(date_m.group('monthname')) + if month is not None: + upload_date = '%s%02d%02d' % ( + date_m.group('year'), month, int(date_m.group('day'))) + + description = self._og_search_description(webpage) + READ_MORE = u' (Read more)' + if description.endswith(READ_MORE): + description = description[:-len(READ_MORE)] + + return { + 'id': video_id, + 'url': video_url, + 'title': self._og_search_title(webpage), + 'description': description, + 'thumbnail': self._og_search_thumbnail(webpage), + 'duration': duration, + 'upload_date': upload_date, + } diff --git a/youtube_dl/extractor/pornhd.py b/youtube_dl/extractor/pornhd.py new file mode 100644 index 000000000..71abd5013 --- /dev/null +++ b/youtube_dl/extractor/pornhd.py @@ -0,0 +1,38 @@ +import re + +from .common import InfoExtractor +from ..utils import compat_urllib_parse + + +class PornHdIE(InfoExtractor): + _VALID_URL = r'(?:http://)?(?:www\.)?pornhd\.com/videos/(?P<video_id>[0-9]+)/(?P<video_title>.+)' + _TEST = { + u'url': u'http://www.pornhd.com/videos/1962/sierra-day-gets-his-cum-all-over-herself-hd-porn-video', + u'file': u'1962.flv', + u'md5': u'35272469887dca97abd30abecc6cdf75', + u'info_dict': { + u"title": u"sierra-day-gets-his-cum-all-over-herself-hd-porn-video", + u"age_limit": 18, + } + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + + video_id = mobj.group('video_id') + video_title = mobj.group('video_title') + + webpage = self._download_webpage(url, video_id) + + video_url = self._html_search_regex( + r'&hd=(http.+?)&', webpage, u'video URL') + video_url = compat_urllib_parse.unquote(video_url) + age_limit = 18 + + return { + 'id': video_id, + 'url': video_url, + 'ext': 'flv', + 'title': video_title, + 'age_limit': age_limit, + } diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index 8b3471919..d9135c6b9 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -12,7 +12,7 @@ from ..aes import ( ) class PornHubIE(InfoExtractor): - _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>pornhub\.com/view_video\.php\?viewkey=(?P<videoid>[0-9]+))' + _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>pornhub\.com/view_video\.php\?viewkey=(?P<videoid>[0-9a-f]+))' _TEST = { u'url': u'http://www.pornhub.com/view_video.php?viewkey=648719015', u'file': u'648719015.mp4', diff --git a/youtube_dl/extractor/rtlnow.py b/youtube_dl/extractor/rtlnow.py index 2f238de35..511674d8d 100644 --- a/youtube_dl/extractor/rtlnow.py +++ b/youtube_dl/extractor/rtlnow.py @@ -7,14 +7,15 @@ from ..utils import ( ExtractorError, ) + class RTLnowIE(InfoExtractor): """Information Extractor for RTL NOW, RTL2 NOW, RTL NITRO, SUPER RTL NOW, VOX NOW and n-tv NOW""" - _VALID_URL = r'(?:http://)?(?P<url>(?P<base_url>rtl-now\.rtl\.de/|rtl2now\.rtl2\.de/|(?:www\.)?voxnow\.de/|(?:www\.)?rtlnitronow\.de/|(?:www\.)?superrtlnow\.de/|(?:www\.)?n-tvnow\.de/)[a-zA-Z0-9-]+/[a-zA-Z0-9-]+\.php\?(?:container_id|film_id)=(?P<video_id>[0-9]+)&player=1(?:&season=[0-9]+)?(?:&.*)?)' + _VALID_URL = r'(?:http://)?(?P<url>(?P<base_url>rtl-now\.rtl\.de|rtl2now\.rtl2\.de|(?:www\.)?voxnow\.de|(?:www\.)?rtlnitronow\.de|(?:www\.)?superrtlnow\.de|(?:www\.)?n-tvnow\.de)/+[a-zA-Z0-9-]+/[a-zA-Z0-9-]+\.php\?(?:container_id|film_id)=(?P<video_id>[0-9]+)&player=1(?:&season=[0-9]+)?(?:&.*)?)' _TESTS = [{ u'url': u'http://rtl-now.rtl.de/ahornallee/folge-1.php?film_id=90419&player=1&season=1', u'file': u'90419.flv', u'info_dict': { - u'upload_date': u'20070416', + u'upload_date': u'20070416', u'title': u'Ahornallee - Folge 1 - Der Einzug', u'description': u'Folge 1 - Der Einzug', }, diff --git a/youtube_dl/extractor/smotri.py b/youtube_dl/extractor/smotri.py index 5a28bc820..4ea89bf85 100644 --- a/youtube_dl/extractor/smotri.py +++ b/youtube_dl/extractor/smotri.py @@ -3,10 +3,13 @@ import re import json import hashlib +import uuid from .common import InfoExtractor from ..utils import ( - ExtractorError + compat_urllib_parse, + compat_urllib_request, + ExtractorError, ) @@ -249,3 +252,105 @@ class SmotriUserIE(InfoExtractor): u'user nickname') return self.playlist_result(entries, user_id, user_nickname) + + +class SmotriBroadcastIE(InfoExtractor): + IE_DESC = u'Smotri.com broadcasts' + IE_NAME = u'smotri:broadcast' + _VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/live/(?P<broadcastid>[^/]+))/?.*' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + broadcast_id = mobj.group('broadcastid') + + broadcast_url = 'http://' + mobj.group('url') + broadcast_page = self._download_webpage(broadcast_url, broadcast_id, u'Downloading broadcast page') + + if re.search(u'>Режиссер с логином <br/>"%s"<br/> <span>не существует<' % broadcast_id, broadcast_page) is not None: + raise ExtractorError(u'Broadcast %s does not exist' % broadcast_id, expected=True) + + # Adult content + if re.search(u'EroConfirmText">', broadcast_page) is not None: + + (username, password) = self._get_login_info() + if username is None: + raise ExtractorError(u'Erotic broadcasts allowed only for registered users, ' + u'use --username and --password options to provide account credentials.', expected=True) + + # Log in + login_form_strs = { + u'login-hint53': '1', + u'confirm_erotic': '1', + u'login': username, + u'password': password, + } + # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode + # chokes on unicode + login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) + login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8') + login_url = broadcast_url + '/?no_redirect=1' + request = compat_urllib_request.Request(login_url, login_data) + request.add_header('Content-Type', 'application/x-www-form-urlencoded') + broadcast_page = self._download_webpage( + request, broadcast_id, note=u'Logging in and confirming age') + + if re.search(u'>Неверный логин или пароль<', broadcast_page) is not None: + raise ExtractorError(u'Unable to log in: bad username or password', expected=True) + + adult_content = True + else: + adult_content = False + + ticket = self._html_search_regex( + u'window\.broadcast_control\.addFlashVar\\(\'file\', \'([^\']+)\'\\);', + broadcast_page, u'broadcast ticket') + + url = 'http://smotri.com/broadcast/view/url/?ticket=%s' % ticket + + broadcast_password = self._downloader.params.get('videopassword', None) + if broadcast_password: + url += '&pass=%s' % hashlib.md5(broadcast_password.encode('utf-8')).hexdigest() + + broadcast_json_page = self._download_webpage(url, broadcast_id, u'Downloading broadcast JSON') + + try: + broadcast_json = json.loads(broadcast_json_page) + + protected_broadcast = broadcast_json['_pass_protected'] == 1 + if protected_broadcast and not broadcast_password: + raise ExtractorError(u'This broadcast is protected by a password, use the --video-password option', expected=True) + + broadcast_offline = broadcast_json['is_play'] == 0 + if broadcast_offline: + raise ExtractorError(u'Broadcast %s is offline' % broadcast_id, expected=True) + + rtmp_url = broadcast_json['_server'] + if not rtmp_url.startswith('rtmp://'): + raise ExtractorError(u'Unexpected broadcast rtmp URL') + + broadcast_playpath = broadcast_json['_streamName'] + broadcast_thumbnail = broadcast_json['_imgURL'] + broadcast_title = broadcast_json['title'] + broadcast_description = broadcast_json['description'] + broadcaster_nick = broadcast_json['nick'] + broadcaster_login = broadcast_json['login'] + rtmp_conn = 'S:%s' % uuid.uuid4().hex + except KeyError: + if protected_broadcast: + raise ExtractorError(u'Bad broadcast password', expected=True) + raise ExtractorError(u'Unexpected broadcast JSON') + + return { + 'id': broadcast_id, + 'url': rtmp_url, + 'title': broadcast_title, + 'thumbnail': broadcast_thumbnail, + 'description': broadcast_description, + 'uploader': broadcaster_nick, + 'uploader_id': broadcaster_login, + 'age_limit': 18 if adult_content else 0, + 'ext': 'flv', + 'play_path': broadcast_playpath, + 'rtmp_live': True, + 'rtmp_conn': rtmp_conn + } diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py index cb6dedab7..cbba4094b 100644 --- a/youtube_dl/extractor/soundcloud.py +++ b/youtube_dl/extractor/soundcloud.py @@ -1,3 +1,4 @@ +# encoding: utf-8 import json import re import itertools @@ -23,7 +24,10 @@ class SoundcloudIE(InfoExtractor): """ _VALID_URL = r'''^(?:https?://)? - (?:(?:(?:www\.)?soundcloud\.com/([\w\d-]+)/([\w\d-]+)/?(?:[?].*)?$) + (?:(?:(?:www\.)?soundcloud\.com/ + (?P<uploader>[\w\d-]+)/ + (?!sets/)(?P<title>[\w\d-]+)/? + (?P<token>[^?]+?)?(?:[?].*)?$) |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+)) |(?P<widget>w\.soundcloud\.com/player/?.*?url=.*) ) @@ -56,6 +60,32 @@ class SoundcloudIE(InfoExtractor): u'skip_download': True, }, }, + # private link + { + u'url': u'https://soundcloud.com/jaimemf/youtube-dl-test-video-a-y-baw/s-8Pjrp', + u'md5': u'aa0dd32bfea9b0c5ef4f02aacd080604', + u'info_dict': { + u'id': u'123998367', + u'ext': u'mp3', + u'title': u'Youtube - Dl Test Video \'\' Ä↭', + u'uploader': u'jaimeMF', + u'description': u'test chars: \"\'/\\ä↭', + u'upload_date': u'20131209', + }, + }, + # downloadable song + { + u'url': u'https://soundcloud.com/simgretina/just-your-problem-baby-1', + u'md5': u'56a8b69568acaa967b4c49f9d1d52d19', + u'info_dict': { + u'id': u'105614606', + u'ext': u'wav', + u'title': u'Just Your Problem Baby (Acapella)', + u'description': u'Vocals', + u'uploader': u'Sim Gretina', + u'upload_date': u'20130815', + }, + }, ] _CLIENT_ID = 'b45b1aa10f1ac2941910a7f0d10f8e28' @@ -73,7 +103,7 @@ class SoundcloudIE(InfoExtractor): def _resolv_url(cls, url): return 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=' + cls._CLIENT_ID - def _extract_info_dict(self, info, full_title=None, quiet=False): + def _extract_info_dict(self, info, full_title=None, quiet=False, secret_token=None): track_id = compat_str(info['id']) name = full_title or track_id if quiet: @@ -82,7 +112,7 @@ class SoundcloudIE(InfoExtractor): thumbnail = info['artwork_url'] if thumbnail is not None: thumbnail = thumbnail.replace('-large', '-t500x500') - ext = info.get('original_format', u'mp3') + ext = u'mp3' result = { 'id': track_id, 'uploader': info['user']['username'], @@ -98,14 +128,16 @@ class SoundcloudIE(InfoExtractor): track_id, self._CLIENT_ID)) result['formats'] = [{ 'format_id': 'download', - 'ext': ext, + 'ext': info.get('original_format', u'mp3'), 'url': format_url, 'vcodec': 'none', }] else: # We have to retrieve the url + streams_url = ('http://api.soundcloud.com/i1/tracks/{0}/streams?' + 'client_id={1}&secret_token={2}'.format(track_id, self._IPHONE_CLIENT_ID, secret_token)) stream_json = self._download_webpage( - 'http://api.soundcloud.com/i1/tracks/{0}/streams?client_id={1}'.format(track_id, self._IPHONE_CLIENT_ID), + streams_url, track_id, u'Downloading track url') formats = [] @@ -157,6 +189,7 @@ class SoundcloudIE(InfoExtractor): raise ExtractorError(u'Invalid URL: %s' % url) track_id = mobj.group('track_id') + token = None if track_id is not None: info_json_url = 'http://api.soundcloud.com/tracks/' + track_id + '.json?client_id=' + self._CLIENT_ID full_title = track_id @@ -165,19 +198,22 @@ class SoundcloudIE(InfoExtractor): return self.url_result(query['url'][0], ie='Soundcloud') else: # extract uploader (which is in the url) - uploader = mobj.group(1) + uploader = mobj.group('uploader') # extract simple title (uploader + slug of song title) - slug_title = mobj.group(2) - full_title = '%s/%s' % (uploader, slug_title) + slug_title = mobj.group('title') + token = mobj.group('token') + full_title = resolve_title = '%s/%s' % (uploader, slug_title) + if token: + resolve_title += '/%s' % token self.report_resolve(full_title) - url = 'http://soundcloud.com/%s/%s' % (uploader, slug_title) + url = 'http://soundcloud.com/%s' % resolve_title info_json_url = self._resolv_url(url) info_json = self._download_webpage(info_json_url, full_title, u'Downloading info JSON') info = json.loads(info_json) - return self._extract_info_dict(info, full_title) + return self._extract_info_dict(info, full_title, secret_token=token) class SoundcloudSetIE(SoundcloudIE): _VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)(?:[?].*)?$' diff --git a/youtube_dl/extractor/stanfordoc.py b/youtube_dl/extractor/stanfordoc.py index d54e01a12..44c52c718 100644 --- a/youtube_dl/extractor/stanfordoc.py +++ b/youtube_dl/extractor/stanfordoc.py @@ -1,14 +1,7 @@ import re -import socket -import xml.etree.ElementTree from .common import InfoExtractor from ..utils import ( - compat_http_client, - compat_str, - compat_urllib_error, - compat_urllib_request, - ExtractorError, orderedSet, unescapeHTML, @@ -45,11 +38,7 @@ class StanfordOpenClassroomIE(InfoExtractor): self.report_extraction(info['id']) baseUrl = 'http://openclassroom.stanford.edu/MainFolder/courses/' + course + '/videos/' xmlUrl = baseUrl + video + '.xml' - try: - metaXml = compat_urllib_request.urlopen(xmlUrl).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to download video info XML: %s' % compat_str(err)) - mdoc = xml.etree.ElementTree.fromstring(metaXml) + mdoc = self._download_xml(xmlUrl, info['id']) try: info['title'] = mdoc.findall('./title')[0].text info['url'] = baseUrl + mdoc.findall('./videoFile')[0].text @@ -95,12 +84,9 @@ class StanfordOpenClassroomIE(InfoExtractor): 'upload_date': None, } - self.report_download_webpage(info['id']) rootURL = 'http://openclassroom.stanford.edu/MainFolder/HomePage.php' - try: - rootpage = compat_urllib_request.urlopen(rootURL).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to download course info page: ' + compat_str(err)) + rootpage = self._download_webpage(rootURL, info['id'], + errnote=u'Unable to download course info page') info['title'] = info['id'] diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py index 61452e47d..cec65261b 100644 --- a/youtube_dl/extractor/theplatform.py +++ b/youtube_dl/extractor/theplatform.py @@ -3,6 +3,7 @@ import json from .common import InfoExtractor from ..utils import ( + ExtractorError, xpath_with_ns, ) @@ -32,6 +33,17 @@ class ThePlatformIE(InfoExtractor): smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?' 'format=smil&mbr=true'.format(video_id)) meta = self._download_xml(smil_url, video_id) + + try: + error_msg = next( + n.attrib['abstract'] + for n in meta.findall(_x('.//smil:ref')) + if n.attrib.get('title') == u'Geographic Restriction') + except StopIteration: + pass + else: + raise ExtractorError(error_msg, expected=True) + info_url = 'http://link.theplatform.com/s/dJ5BDC/{0}?format=preview'.format(video_id) info_json = self._download_webpage(info_url, video_id) info = json.loads(info_json) diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index fb2bd225a..ea4409528 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -115,7 +115,7 @@ class VimeoIE(InfoExtractor): def _real_initialize(self): self._login() - def _real_extract(self, url, new_video=True): + def _real_extract(self, url): url, data = unsmuggle_url(url) headers = std_headers if data is not None: @@ -151,8 +151,14 @@ class VimeoIE(InfoExtractor): config = json.loads(config_json) except RegexNotFoundError: # For pro videos or player.vimeo.com urls - config = self._search_regex([r' = {config:({.+?}),assets:', r'(?:c|b)=({.+?});'], - webpage, u'info section', flags=re.DOTALL) + # We try to find out to which variable is assigned the config dic + m_variable_name = re.search('(\w)\.video\.id', webpage) + if m_variable_name is not None: + config_re = r'%s=({.+?});' % re.escape(m_variable_name.group(1)) + else: + config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});'] + config = self._search_regex(config_re, webpage, u'info section', + flags=re.DOTALL) config = json.loads(config) except Exception as e: if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage): diff --git a/youtube_dl/extractor/wimp.py b/youtube_dl/extractor/wimp.py index b9c3b13f9..82a626e0e 100644 --- a/youtube_dl/extractor/wimp.py +++ b/youtube_dl/extractor/wimp.py @@ -11,7 +11,8 @@ class WimpIE(InfoExtractor): u'file': u'deerfence.flv', u'md5': u'8b215e2e0168c6081a1cf84b2846a2b5', u'info_dict': { - u"title": u"Watch Till End: Herd of deer jump over a fence." + u"title": u"Watch Till End: Herd of deer jump over a fence.", + u"description": u"These deer look as fluid as running water when they jump over this fence as a herd. This video is one that needs to be watched until the very end for the true majesty to be witnessed, but once it comes, it's sure to take your breath away.", } } @@ -19,18 +20,14 @@ class WimpIE(InfoExtractor): mobj = re.match(self._VALID_URL, url) video_id = mobj.group(1) webpage = self._download_webpage(url, video_id) - title = self._search_regex(r'<meta name="description" content="(.+?)" />',webpage, 'video title') - thumbnail_url = self._search_regex(r'<meta property="og\:image" content="(.+?)" />', webpage,'video thumbnail') googleString = self._search_regex("googleCode = '(.*?)'", webpage, 'file url') googleString = base64.b64decode(googleString).decode('ascii') - final_url = self._search_regex('","(.*?)"', googleString,'final video url') - ext = final_url.rpartition(u'.')[2] - - return [{ - 'id': video_id, - 'url': final_url, - 'ext': ext, - 'title': title, - 'thumbnail': thumbnail_url, - }] + final_url = self._search_regex('","(.*?)"', googleString, u'final video url') + return { + 'id': video_id, + 'url': final_url, + 'title': self._og_search_title(webpage), + 'thumbnail': self._og_search_thumbnail(webpage), + 'description': self._og_search_description(webpage), + } diff --git a/youtube_dl/extractor/xhamster.py b/youtube_dl/extractor/xhamster.py index 279f75e7a..ef9997ee4 100644 --- a/youtube_dl/extractor/xhamster.py +++ b/youtube_dl/extractor/xhamster.py @@ -26,7 +26,7 @@ class XHamsterIE(InfoExtractor): { u'url': u'http://xhamster.com/movies/2221348/britney_spears_sexy_booty.html?hd', u'file': u'2221348.flv', - u'md5': u'970a94178ca4118c5aa3aaea21211b81', + u'md5': u'e767b9475de189320f691f49c679c4c7', u'info_dict': { u"upload_date": u"20130914", u"uploader_id": u"jojo747400", @@ -46,7 +46,7 @@ class XHamsterIE(InfoExtractor): return mobj.group('server')+'/key='+mobj.group('file') def is_hd(webpage): - return webpage.find('<div class=\'icon iconHD\'>') != -1 + return webpage.find('<div class=\'icon iconHD\'') != -1 mobj = re.match(self._VALID_URL, url) diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 01715024c..a68a214ca 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -7,7 +7,6 @@ import itertools import json import os.path import re -import socket import string import struct import traceback @@ -17,9 +16,7 @@ from .common import InfoExtractor, SearchInfoExtractor from .subtitles import SubtitlesInfoExtractor from ..utils import ( compat_chr, - compat_http_client, compat_parse_qs, - compat_urllib_error, compat_urllib_parse, compat_urllib_request, compat_urlparse, @@ -45,19 +42,11 @@ class YoutubeBaseInfoExtractor(InfoExtractor): # If True it will raise an error if no login info is provided _LOGIN_REQUIRED = False - def report_lang(self): - """Report attempt to set language.""" - self.to_screen(u'Setting language') - def _set_language(self): - request = compat_urllib_request.Request(self._LANG_URL) - try: - self.report_lang() - compat_urllib_request.urlopen(request).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_warning(u'unable to set language: %s' % compat_str(err)) - return False - return True + return bool(self._download_webpage( + self._LANG_URL, None, + note=u'Setting language', errnote='unable to set language', + fatal=False)) def _login(self): (username, password) = self._get_login_info() @@ -67,12 +56,12 @@ class YoutubeBaseInfoExtractor(InfoExtractor): raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True) return False - request = compat_urllib_request.Request(self._LOGIN_URL) - try: - login_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_warning(u'unable to fetch login page: %s' % compat_str(err)) - return False + login_page = self._download_webpage( + self._LOGIN_URL, None, + note=u'Downloading login page', + errnote=u'unable to fetch login page', fatal=False) + if login_page is False: + return galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"', login_page, u'Login GALX parameter') @@ -102,29 +91,28 @@ class YoutubeBaseInfoExtractor(InfoExtractor): # chokes on unicode login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') - request = compat_urllib_request.Request(self._LOGIN_URL, login_data) - try: - self.report_login() - login_results = compat_urllib_request.urlopen(request).read().decode('utf-8') - if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: - self._downloader.report_warning(u'unable to log in: bad username or password') - return False - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) + + req = compat_urllib_request.Request(self._LOGIN_URL, login_data) + login_results = self._download_webpage( + req, None, + note=u'Logging in', errnote=u'unable to log in', fatal=False) + if login_results is False: + return False + if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: + self._downloader.report_warning(u'unable to log in: bad username or password') return False return True def _confirm_age(self): age_form = { - 'next_url': '/', - 'action_confirm': 'Confirm', - } - request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) - try: - self.report_age_confirmation() - compat_urllib_request.urlopen(request).read().decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) + 'next_url': '/', + 'action_confirm': 'Confirm', + } + req = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) + + self._download_webpage( + req, None, + note=u'Confirming age', errnote=u'Unable to confirm age') return True def _real_initialize(self): @@ -1389,9 +1377,9 @@ 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 @@ -1584,6 +1572,9 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): if playlist_id.startswith('RD'): # Mixes require a custom extraction process return self._extract_mix(playlist_id) + if playlist_id.startswith('TL'): + raise ExtractorError(u'For downloading YouTube.com top lists, use ' + u'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True) # Extract the video ids from the playlist pages ids = [] @@ -1606,6 +1597,38 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): return self.playlist_result(url_results, playlist_id, playlist_title) +class YoutubeTopListIE(YoutubePlaylistIE): + IE_NAME = u'youtube:toplist' + IE_DESC = (u'YouTube.com top lists, "yttoplist:{channel}:{list title}"' + u' (Example: "yttoplist:music:Top Tracks")') + _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + channel = mobj.group('chann') + title = mobj.group('title') + query = compat_urllib_parse.urlencode({'title': title}) + 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) + + video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"' + ids = [] + # sometimes the webpage doesn't contain the videos + # retry until we get them + for i in itertools.count(0): + msg = u'Downloading Youtube mix' + if i > 0: + msg += ', retry #%d' % i + webpage = self._download_webpage(url, title, msg) + ids = orderedSet(re.findall(video_re, webpage)) + if ids: + break + url_results = self._ids_to_results(ids) + return self.playlist_result(url_results, playlist_title=title) + + class YoutubeChannelIE(InfoExtractor): IE_DESC = u'YouTube.com channels' _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" @@ -1631,10 +1654,11 @@ class YoutubeChannelIE(InfoExtractor): 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 + autogenerated = re.search(r'''(?x) + class="[^"]*?(?: + channel-header-autogenerated-label| + yt-channel-title-autogenerated + )[^"]*"''', channel_page) is not None if autogenerated: # The videos are contained in a single page @@ -1736,10 +1760,6 @@ class YoutubeSearchIE(SearchInfoExtractor): IE_NAME = u'youtube:search' _SEARCH_KEY = 'ytsearch' - def report_download_page(self, query, pagenum): - """Report attempt to download search page with given number.""" - self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum)) - def _get_n_results(self, query, n): """Get a specified number of results for a query""" @@ -1748,16 +1768,15 @@ class YoutubeSearchIE(SearchInfoExtractor): limit = n while (50 * pagenum) < limit: - self.report_download_page(query, pagenum+1) result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) - request = compat_urllib_request.Request(result_url) - try: - data = compat_urllib_request.urlopen(request).read().decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to download API page: %s' % compat_str(err)) - api_response = json.loads(data)['data'] - - if not 'items' in api_response: + data_json = self._download_webpage( + result_url, video_id=u'query "%s"' % query, + note=u'Downloading page %s' % (pagenum + 1), + errnote=u'Unable to download API page') + data = json.loads(data_json) + api_response = data['data'] + + if 'items' not in api_response: raise ExtractorError(u'[youtube] No video results') new_ids = list(video['id'] for video in api_response['items']) diff --git a/youtube_dl/extractor/zdf.py b/youtube_dl/extractor/zdf.py index 689f19735..35ece354a 100644 --- a/youtube_dl/extractor/zdf.py +++ b/youtube_dl/extractor/zdf.py @@ -73,14 +73,14 @@ class ZDFIE(InfoExtractor): try: proto_pref = -PROTO_ORDER.index(format_m.group('proto')) except ValueError: - proto_pref = 999 + proto_pref = -999 quality = fnode.find('./quality').text QUALITY_ORDER = ['veryhigh', '300', 'high', 'med', 'low'] try: quality_pref = -QUALITY_ORDER.index(quality) except ValueError: - quality_pref = 999 + quality_pref = -999 abr = int(fnode.find('./audioBitrate').text) // 1000 vbr = int(fnode.find('./videoBitrate').text) // 1000 |