diff options
Diffstat (limited to 'youtube_dl/extractor')
-rw-r--r-- | youtube_dl/extractor/__init__.py | 8 | ||||
-rw-r--r-- | youtube_dl/extractor/arte.py | 35 | ||||
-rw-r--r-- | youtube_dl/extractor/bild.py | 39 | ||||
-rw-r--r-- | youtube_dl/extractor/cinemassacre.py | 17 | ||||
-rw-r--r-- | youtube_dl/extractor/cnn.py | 2 | ||||
-rw-r--r-- | youtube_dl/extractor/common.py | 6 | ||||
-rw-r--r-- | youtube_dl/extractor/crunchyroll.py | 11 | ||||
-rw-r--r-- | youtube_dl/extractor/francetv.py | 4 | ||||
-rw-r--r-- | youtube_dl/extractor/funnyordie.py | 2 | ||||
-rw-r--r-- | youtube_dl/extractor/generic.py | 29 | ||||
-rw-r--r-- | youtube_dl/extractor/justintv.py | 155 | ||||
-rw-r--r-- | youtube_dl/extractor/mitele.py | 13 | ||||
-rw-r--r-- | youtube_dl/extractor/mixcloud.py | 20 | ||||
-rw-r--r-- | youtube_dl/extractor/pbs.py | 20 | ||||
-rw-r--r-- | youtube_dl/extractor/sexykarma.py | 117 | ||||
-rw-r--r-- | youtube_dl/extractor/ted.py | 23 | ||||
-rw-r--r-- | youtube_dl/extractor/telecinco.py | 19 | ||||
-rw-r--r-- | youtube_dl/extractor/twitch.py | 187 | ||||
-rw-r--r-- | youtube_dl/extractor/vidzi.py | 33 | ||||
-rw-r--r-- | youtube_dl/extractor/vimeo.py | 14 | ||||
-rw-r--r-- | youtube_dl/extractor/youtube.py | 5 |
21 files changed, 546 insertions, 213 deletions
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index c0b26c5b3..ab4af2079 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -26,6 +26,7 @@ from .bandcamp import BandcampIE, BandcampAlbumIE from .bbccouk import BBCCoUkIE from .beeg import BeegIE from .behindkink import BehindKinkIE +from .bild import BildIE from .bilibili import BiliBiliIE from .blinkx import BlinkxIE from .bliptv import BlipTVIE, BlipTVUserIE @@ -174,7 +175,6 @@ from .jadorecettepub import JadoreCettePubIE from .jeuxvideo import JeuxVideoIE from .jove import JoveIE from .jukebox import JukeboxIE -from .justintv import JustinTVIE from .jpopsukitv import JpopsukiIE from .kankan import KankanIE from .keezmovies import KeezMoviesIE @@ -317,6 +317,7 @@ from .sbs import SBSIE from .scivee import SciVeeIE from .screencast import ScreencastIE from .servingsys import ServingSysIE +from .sexykarma import SexyKarmaIE from .shared import SharedIE from .sharesix import ShareSixIE from .sina import SinaIE @@ -368,6 +369,7 @@ from .teachingchannel import TeachingChannelIE from .teamcoco import TeamcocoIE from .techtalks import TechTalksIE from .ted import TEDIE +from .telecinco import TelecincoIE from .telemb import TeleMBIE from .tenplay import TenPlayIE from .testurl import TestURLIE @@ -396,6 +398,7 @@ from .tutv import TutvIE from .tvigle import TvigleIE from .tvp import TvpIE from .tvplay import TVPlayIE +from .twitch import TwitchIE from .ubu import UbuIE from .udemy import ( UdemyIE, @@ -421,6 +424,7 @@ from .videopremium import VideoPremiumIE from .videott import VideoTtIE from .videoweed import VideoWeedIE from .vidme import VidmeIE +from .vidzi import VidziIE from .vimeo import ( VimeoIE, VimeoAlbumIE, @@ -489,10 +493,8 @@ from .youtube import ( YoutubeUserIE, YoutubeWatchLaterIE, ) - from .zdf import ZDFIE - _ALL_CLASSES = [ klass for name, klass in globals().items() diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index c3d02f85e..b9a9440c0 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -10,8 +10,8 @@ from ..utils import ( unified_strdate, determine_ext, get_element_by_id, - compat_str, get_element_by_attribute, + int_or_none, ) # There are different sources of video in arte.tv, the extraction process @@ -90,15 +90,24 @@ class ArteTVPlus7IE(InfoExtractor): if not upload_date_str: upload_date_str = player_info.get('VDA', '').split(' ')[0] + title = player_info['VTI'].strip() + subtitle = player_info.get('VSU', '').strip() + if subtitle: + title += ' - %s' % subtitle + info_dict = { 'id': player_info['VID'], - 'title': player_info['VTI'], + 'title': title, 'description': player_info.get('VDE'), 'upload_date': unified_strdate(upload_date_str), 'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'), } - all_formats = player_info['VSR'].values() + all_formats = [] + for format_id, format_dict in player_info['VSR'].items(): + fmt = dict(format_dict) + fmt['format_id'] = format_id + all_formats.append(fmt) # Some formats use the m3u8 protocol all_formats = list(filter(lambda f: f.get('videoFormat') != 'M3U8', all_formats)) def _match_lang(f): @@ -149,22 +158,12 @@ class ArteTVPlus7IE(InfoExtractor): ) formats = sorted(formats, key=sort_key) def _format(format_info): - quality = '' - height = format_info.get('height') - if height is not None: - quality = compat_str(height) - bitrate = format_info.get('bitrate') - if bitrate is not None: - quality += '-%d' % bitrate - if format_info.get('versionCode') is not None: - format_id = '%s-%s' % (quality, format_info['versionCode']) - else: - format_id = quality info = { - 'format_id': format_id, - 'format_note': format_info.get('versionLibelle'), - 'width': format_info.get('width'), - 'height': height, + 'format_id': format_info['format_id'], + 'format_note': '%s, %s' % (format_info.get('versionCode'), format_info.get('versionLibelle')), + 'width': int_or_none(format_info.get('width')), + 'height': int_or_none(format_info.get('height')), + 'tbr': int_or_none(format_info.get('bitrate')), } if format_info['mediaType'] == 'rtmp': info['url'] = format_info['streamer'] diff --git a/youtube_dl/extractor/bild.py b/youtube_dl/extractor/bild.py new file mode 100644 index 000000000..0269d1174 --- /dev/null +++ b/youtube_dl/extractor/bild.py @@ -0,0 +1,39 @@ +#coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import int_or_none + + +class BildIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?bild\.de/(?:[^/]+/)+(?P<display_id>[^/]+)-(?P<id>\d+)(?:,auto=true)?\.bild\.html' + IE_DESC = 'Bild.de' + _TEST = { + 'url': 'http://www.bild.de/video/clip/apple-ipad-air/das-koennen-die-neuen-ipads-38184146.bild.html', + 'md5': 'dd495cbd99f2413502a1713a1156ac8a', + 'info_dict': { + 'id': '38184146', + 'ext': 'mp4', + 'title': 'BILD hat sie getestet', + 'thumbnail': 'http://bilder.bild.de/fotos/stand-das-koennen-die-neuen-ipads-38184138/Bild/1.bild.jpg', + 'duration': 196, + 'description': 'Mit dem iPad Air 2 und dem iPad Mini 3 hat Apple zwei neue Tablet-Modelle präsentiert. BILD-Reporter Sven Stein durfte die Geräte bereits testen. ', + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + xml_url = url.split(".bild.html")[0] + ",view=xml.bild.xml" + doc = self._download_xml(xml_url, video_id) + + duration = int_or_none(doc.attrib.get('duration'), scale=1000) + + return { + 'id': video_id, + 'title': doc.attrib['ueberschrift'], + 'description': doc.attrib.get('text'), + 'url': doc.attrib['src'], + 'thumbnail': doc.attrib.get('img'), + 'duration': duration, + } diff --git a/youtube_dl/extractor/cinemassacre.py b/youtube_dl/extractor/cinemassacre.py index 496271be4..d064a28f9 100644 --- a/youtube_dl/extractor/cinemassacre.py +++ b/youtube_dl/extractor/cinemassacre.py @@ -42,7 +42,7 @@ class CinemassacreIE(InfoExtractor): webpage = self._download_webpage(url, display_id) video_date = mobj.group('date_Y') + mobj.group('date_m') + mobj.group('date_d') - mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?id=(?:Cinemassacre-)?(?P<video_id>.+?))"', webpage) + mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?[^"]*\bid=(?:Cinemassacre-)?(?P<video_id>.+?))"', webpage) if not mobj: raise ExtractorError('Can\'t extract embed url and video id') playerdata_url = mobj.group('embed_url') @@ -53,17 +53,22 @@ class CinemassacreIE(InfoExtractor): video_description = self._html_search_regex( r'<div class="entry-content">(?P<description>.+?)</div>', webpage, 'description', flags=re.DOTALL, fatal=False) + video_thumbnail = self._og_search_thumbnail(webpage) playerdata = self._download_webpage(playerdata_url, video_id, 'Downloading player webpage') - video_thumbnail = self._search_regex( - r'image: \'(?P<thumbnail>[^\']+)\'', playerdata, 'thumbnail', fatal=False) - sd_url = self._search_regex(r'file: \'([^\']+)\', label: \'SD\'', playerdata, 'sd_file') - videolist_url = self._search_regex(r'file: \'([^\']+\.smil)\'}', playerdata, 'videolist_url') + vidurl = self._search_regex( + r'\'vidurl\'\s*:\s*"([^\']+)"', playerdata, 'vidurl').replace('\\/', '/') + vidid = self._search_regex( + r'\'vidid\'\s*:\s*"([^\']+)"', playerdata, 'vidid') + videoserver = self._html_search_regex( + r"'videoserver'\s*:\s*'([^']+)'", playerdata, 'videoserver') + + videolist_url = 'http://%s/vod/smil:%s.smil/jwplayer.smil' % (videoserver, vidid) videolist = self._download_xml(videolist_url, video_id, 'Downloading videolist XML') formats = [] - baseurl = sd_url[:sd_url.rfind('/')+1] + baseurl = vidurl[:vidurl.rfind('/')+1] for video in videolist.findall('.//video'): src = video.get('src') if not src: diff --git a/youtube_dl/extractor/cnn.py b/youtube_dl/extractor/cnn.py index dae40c136..78877b1cf 100644 --- a/youtube_dl/extractor/cnn.py +++ b/youtube_dl/extractor/cnn.py @@ -12,7 +12,7 @@ from ..utils import ( class CNNIE(InfoExtractor): _VALID_URL = r'''(?x)https?://((edition|www)\.)?cnn\.com/video/(data/.+?|\?)/ - (?P<path>.+?/(?P<title>[^/]+?)(?:\.cnn|(?=&)))''' + (?P<path>.+?/(?P<title>[^/]+?)(?:\.cnn(-ap)?|(?=&)))''' _TESTS = [{ 'url': 'http://edition.cnn.com/video/?/video/sports/2013/06/09/nadal-1-on-1.cnn', diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 450c7dfd6..e8366f7f9 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -281,6 +281,12 @@ class InfoExtractor(object): raw_filename = basen + '.dump' filename = sanitize_filename(raw_filename, restricted=True) self.to_screen('Saving request to ' + filename) + # Working around MAX_PATH limitation on Windows (see + # http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx) + if os.name == 'nt': + absfilepath = os.path.abspath(filename) + if len(absfilepath) > 259: + filename = '\\\\?\\' + absfilepath with open(filename, 'wb') as outf: outf.write(webpage_bytes) diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py index f99888ecc..e3057d900 100644 --- a/youtube_dl/extractor/crunchyroll.py +++ b/youtube_dl/extractor/crunchyroll.py @@ -39,6 +39,7 @@ class CrunchyrollIE(SubtitlesInfoExtractor): 'thumbnail': 'http://img1.ak.crunchyroll.com/i/spire1-tmb/20c6b5e10f1a47b10516877d3c039cae1380951166_full.jpg', 'uploader': 'Yomiuri Telecasting Corporation (YTV)', 'upload_date': '20131013', + 'url': 're:(?!.*&)', }, 'params': { # rtmp @@ -237,12 +238,14 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text streamdata_req.data = 'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality='+stream_quality+'&media%5Fid='+stream_id+'&video%5Fformat='+stream_format streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded') streamdata_req.add_header('Content-Length', str(len(streamdata_req.data))) - streamdata = self._download_webpage(streamdata_req, video_id, note='Downloading media info for '+video_format) - video_url = self._search_regex(r'<host>([^<]+)', streamdata, 'video_url') - video_play_path = self._search_regex(r'<file>([^<]+)', streamdata, 'video_play_path') + streamdata = self._download_xml( + streamdata_req, video_id, + note='Downloading media info for %s' % video_format) + video_url = streamdata.find('.//host').text + video_play_path = streamdata.find('.//file').text formats.append({ 'url': video_url, - 'play_path': video_play_path, + 'play_path': video_play_path, 'ext': 'flv', 'format': video_format, 'format_id': video_format, diff --git a/youtube_dl/extractor/francetv.py b/youtube_dl/extractor/francetv.py index 0b3374d97..566e20d76 100644 --- a/youtube_dl/extractor/francetv.py +++ b/youtube_dl/extractor/francetv.py @@ -46,7 +46,7 @@ class FranceTVBaseInfoExtractor(InfoExtractor): f4m_format['preference'] = 1 formats.extend(f4m_formats) elif video_url.endswith('.m3u8'): - formats.extend(self._extract_m3u8_formats(video_url, video_id)) + formats.extend(self._extract_m3u8_formats(video_url, video_id, 'mp4')) elif video_url.startswith('rtmp'): formats.append({ 'url': video_url, @@ -58,7 +58,7 @@ class FranceTVBaseInfoExtractor(InfoExtractor): formats.append({ 'url': video_url, 'format_id': format_id, - 'preference': 2, + 'preference': -1, }) self._sort_formats(formats) diff --git a/youtube_dl/extractor/funnyordie.py b/youtube_dl/extractor/funnyordie.py index d966e8403..ec6d96ada 100644 --- a/youtube_dl/extractor/funnyordie.py +++ b/youtube_dl/extractor/funnyordie.py @@ -37,7 +37,7 @@ class FunnyOrDieIE(InfoExtractor): video_id = mobj.group('id') webpage = self._download_webpage(url, video_id) - links = re.findall(r'<source src="([^"]+/v)\d+\.([^"]+)" type=\'video', webpage) + links = re.findall(r'<source src="([^"]+/v)[^"]+\.([^"]+)" type=\'video', webpage) if not links: raise ExtractorError('No media links available for %s' % video_id) diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 122763a23..9b6498894 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -380,6 +380,17 @@ class GenericIE(InfoExtractor): 'uploader': 'education-portal.com', }, }, + { + 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz', + 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4', + 'info_dict': { + 'id': 'uxjb0lwrcz', + 'ext': 'mp4', + 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks', + 'duration': 1715.0, + 'uploader': 'thoughtworks.wistia.com', + }, + }, ] def report_following_redirect(self, new_url): @@ -476,7 +487,8 @@ class GenericIE(InfoExtractor): 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube' ) % (url, url), expected=True) else: - assert ':' in default_search + if ':' not in default_search: + default_search += ':' return self.url_result(default_search + url) url, smuggled_data = unsmuggle_url(url) @@ -609,13 +621,13 @@ class GenericIE(InfoExtractor): if mobj: player_url = unescapeHTML(mobj.group('url')) surl = smuggle_url(player_url, {'Referer': url}) - return self.url_result(surl, 'Vimeo') + return self.url_result(surl) # Look for embedded (swf embed) Vimeo player mobj = re.search( - r'<embed[^>]+?src="(https?://(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage) + r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage) if mobj: - return self.url_result(mobj.group(1), 'Vimeo') + return self.url_result(mobj.group(1)) # Look for embedded YouTube player matches = re.findall(r'''(?x) @@ -652,17 +664,20 @@ class GenericIE(InfoExtractor): # Look for embedded Wistia player match = re.search( - r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage) + r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage) if match: + embed_url = self._proto_relative_url( + unescapeHTML(match.group('url'))) return { '_type': 'url_transparent', - 'url': unescapeHTML(match.group('url')), + 'url': embed_url, 'ie_key': 'Wistia', 'uploader': video_uploader, 'title': video_title, 'id': video_id, } - match = re.search(r'(?:id=["\']wistia_|data-wistiaid=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage) + + match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage) if match: return { '_type': 'url_transparent', diff --git a/youtube_dl/extractor/justintv.py b/youtube_dl/extractor/justintv.py deleted file mode 100644 index 27017e89f..000000000 --- a/youtube_dl/extractor/justintv.py +++ /dev/null @@ -1,155 +0,0 @@ -from __future__ import unicode_literals - -import itertools -import json -import os -import re - -from .common import InfoExtractor -from ..utils import ( - compat_str, - ExtractorError, - formatSeconds, -) - - -class JustinTVIE(InfoExtractor): - """Information extractor for justin.tv and twitch.tv""" - # TODO: One broadcast may be split into multiple videos. The key - # 'broadcast_id' is the same for all parts, and 'broadcast_part' - # starts at 1 and increases. Can we treat all parts as one video? - - _VALID_URL = r"""(?x)^(?:http://)?(?:www\.)?(?:twitch|justin)\.tv/ - (?: - (?P<channelid>[^/]+)| - (?:(?:[^/]+)/b/(?P<videoid>[^/]+))| - (?:(?:[^/]+)/c/(?P<chapterid>[^/]+)) - ) - /?(?:\#.*)?$ - """ - _JUSTIN_PAGE_LIMIT = 100 - IE_NAME = 'justin.tv' - IE_DESC = 'justin.tv and twitch.tv' - _TEST = { - 'url': 'http://www.twitch.tv/thegamedevhub/b/296128360', - 'md5': 'ecaa8a790c22a40770901460af191c9a', - 'info_dict': { - 'id': '296128360', - 'ext': 'flv', - 'upload_date': '20110927', - 'uploader_id': 25114803, - 'uploader': 'thegamedevhub', - 'title': 'Beginner Series - Scripting With Python Pt.1' - } - } - - # Return count of items, list of *valid* items - def _parse_page(self, url, video_id, counter): - info_json = self._download_webpage( - url, video_id, - 'Downloading video info JSON on page %d' % counter, - 'Unable to download video info JSON %d' % counter) - - response = json.loads(info_json) - if type(response) != list: - error_text = response.get('error', 'unknown error') - raise ExtractorError('Justin.tv API: %s' % error_text) - info = [] - for clip in response: - video_url = clip['video_file_url'] - if video_url: - video_extension = os.path.splitext(video_url)[1][1:] - video_date = re.sub('-', '', clip['start_time'][:10]) - video_uploader_id = clip.get('user_id', clip.get('channel_id')) - video_id = clip['id'] - video_title = clip.get('title', video_id) - info.append({ - 'id': compat_str(video_id), - 'url': video_url, - 'title': video_title, - 'uploader': clip.get('channel_name', video_uploader_id), - 'uploader_id': video_uploader_id, - 'upload_date': video_date, - 'ext': video_extension, - }) - return (len(response), info) - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - - api_base = 'http://api.justin.tv' - paged = False - if mobj.group('channelid'): - paged = True - video_id = mobj.group('channelid') - api = api_base + '/channel/archives/%s.json' % video_id - elif mobj.group('chapterid'): - chapter_id = mobj.group('chapterid') - - webpage = self._download_webpage(url, chapter_id) - m = re.search(r'PP\.archive_id = "([0-9]+)";', webpage) - if not m: - raise ExtractorError('Cannot find archive of a chapter') - archive_id = m.group(1) - - api = api_base + '/broadcast/by_chapter/%s.xml' % chapter_id - doc = self._download_xml( - api, chapter_id, - note='Downloading chapter information', - errnote='Chapter information download failed') - for a in doc.findall('.//archive'): - if archive_id == a.find('./id').text: - break - else: - raise ExtractorError('Could not find chapter in chapter information') - - video_url = a.find('./video_file_url').text - video_ext = video_url.rpartition('.')[2] or 'flv' - - chapter_api_url = 'https://api.twitch.tv/kraken/videos/c' + chapter_id - chapter_info = self._download_json( - chapter_api_url, 'c' + chapter_id, - note='Downloading chapter metadata', - errnote='Download of chapter metadata failed') - - bracket_start = int(doc.find('.//bracket_start').text) - bracket_end = int(doc.find('.//bracket_end').text) - - # TODO determine start (and probably fix up file) - # youtube-dl -v http://www.twitch.tv/firmbelief/c/1757457 - #video_url += '?start=' + TODO:start_timestamp - # bracket_start is 13290, but we want 51670615 - self._downloader.report_warning('Chapter detected, but we can just download the whole file. ' - 'Chapter starts at %s and ends at %s' % (formatSeconds(bracket_start), formatSeconds(bracket_end))) - - info = { - 'id': 'c' + chapter_id, - 'url': video_url, - 'ext': video_ext, - 'title': chapter_info['title'], - 'thumbnail': chapter_info['preview'], - 'description': chapter_info['description'], - 'uploader': chapter_info['channel']['display_name'], - 'uploader_id': chapter_info['channel']['name'], - } - return info - else: - video_id = mobj.group('videoid') - api = api_base + '/broadcast/by_archive/%s.json' % video_id - - entries = [] - offset = 0 - limit = self._JUSTIN_PAGE_LIMIT - for counter in itertools.count(1): - page_url = api + ('?offset=%d&limit=%d' % (offset, limit)) - page_count, page_info = self._parse_page( - page_url, video_id, counter) - entries.extend(page_info) - if not paged or page_count != limit: - break - offset += limit - return { - '_type': 'playlist', - 'id': video_id, - 'entries': entries, - } diff --git a/youtube_dl/extractor/mitele.py b/youtube_dl/extractor/mitele.py index 979f3d692..6691521e5 100644 --- a/youtube_dl/extractor/mitele.py +++ b/youtube_dl/extractor/mitele.py @@ -6,6 +6,7 @@ import json from .common import InfoExtractor from ..utils import ( compat_urllib_parse, + compat_urlparse, get_element_by_attribute, parse_duration, strip_jsonp, @@ -39,13 +40,21 @@ class MiTeleIE(InfoExtractor): ).replace('\'', '"') embed_data = json.loads(embed_data_json) - info_url = embed_data['flashvars']['host'] + domain = embed_data['mediaUrl'] + if not domain.startswith('http'): + # only happens in telecinco.es videos + domain = 'http://' + domain + info_url = compat_urlparse.urljoin( + domain, + compat_urllib_parse.unquote(embed_data['flashvars']['host']) + ) info_el = self._download_xml(info_url, episode).find('./video/info') video_link = info_el.find('videoUrl/link').text token_query = compat_urllib_parse.urlencode({'id': video_link}) token_info = self._download_json( - 'http://token.mitele.es/?' + token_query, episode, + embed_data['flashvars']['ov_tk'] + '?' + token_query, + episode, transform_source=strip_jsonp ) diff --git a/youtube_dl/extractor/mixcloud.py b/youtube_dl/extractor/mixcloud.py index a4564d3de..bb8937c4d 100644 --- a/youtube_dl/extractor/mixcloud.py +++ b/youtube_dl/extractor/mixcloud.py @@ -33,22 +33,22 @@ class MixcloudIE(InfoExtractor): }, } - def check_urls(self, url_list): - """Returns 1st active url from list""" - for url in url_list: + def _get_url(self, track_id, template_url): + server_count = 30 + for i in range(server_count): + url = template_url % i try: # We only want to know if the request succeed # don't download the whole file - self._request_webpage(HEADRequest(url), None, False) + self._request_webpage( + HEADRequest(url), track_id, + 'Checking URL %d/%d ...' % (i + 1, server_count + 1)) return url except ExtractorError: - url = None + pass 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) uploader = mobj.group(1) @@ -61,11 +61,11 @@ class MixcloudIE(InfoExtractor): r'\s(?:data-preview-url|m-preview)="(.+?)"', webpage, 'preview url') song_url = preview_url.replace('/previews/', '/c/originals/') template_url = re.sub(r'(stream\d*)', 'stream%d', song_url) - final_song_url = self._get_url(template_url) + final_song_url = self._get_url(track_id, 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) + final_song_url = self._get_url(track_id, template_url) if final_song_url is None: raise ExtractorError('Unable to extract track url') diff --git a/youtube_dl/extractor/pbs.py b/youtube_dl/extractor/pbs.py index 8f140d626..6118ed5c2 100644 --- a/youtube_dl/extractor/pbs.py +++ b/youtube_dl/extractor/pbs.py @@ -80,8 +80,14 @@ class PBSIE(InfoExtractor): 'thumbnail': 're:^https?://.*\.jpg$', 'upload_date': '20140122', } + }, + { + 'url': 'http://www.pbs.org/wgbh/pages/frontline/united-states-of-secrets/', + 'info_dict': { + 'id': 'united-states-of-secrets', + }, + 'playlist_count': 2, } - ] def _extract_webpage(self, url): @@ -96,6 +102,12 @@ class PBSIE(InfoExtractor): r'<input type="hidden" id="air_date_[0-9]+" value="([^"]+)"', webpage, 'upload date', default=None)) + # tabbed frontline videos + tabbed_videos = re.findall( + r'<div[^>]+class="videotab[^"]*"[^>]+vid="(\d+)"', webpage) + if tabbed_videos: + return tabbed_videos, presumptive_id, upload_date + MEDIA_ID_REGEXES = [ r"div\s*:\s*'videoembed'\s*,\s*mediaid\s*:\s*'(\d+)'", # frontline video embed r'class="coveplayerid">([^<]+)<', # coveplayer @@ -130,6 +142,12 @@ class PBSIE(InfoExtractor): def _real_extract(self, url): video_id, display_id, upload_date = self._extract_webpage(url) + if isinstance(video_id, list): + entries = [self.url_result( + 'http://video.pbs.org/video/%s' % vid_id, 'PBS', vid_id) + for vid_id in video_id] + return self.playlist_result(entries, display_id) + info_url = 'http://video.pbs.org/videoInfo/%s?format=json' % video_id info = self._download_json(info_url, display_id) diff --git a/youtube_dl/extractor/sexykarma.py b/youtube_dl/extractor/sexykarma.py new file mode 100644 index 000000000..c833fc8ee --- /dev/null +++ b/youtube_dl/extractor/sexykarma.py @@ -0,0 +1,117 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + unified_strdate, + parse_duration, + int_or_none, +) + + +class SexyKarmaIE(InfoExtractor): + IE_DESC = 'Sexy Karma and Watch Indian Porn' + _VALID_URL = r'https?://(?:www\.)?(?:sexykarma\.com|watchindianporn\.net)/(?:[^/]+/)*video/(?P<display_id>[^/]+)-(?P<id>[a-zA-Z0-9]+)\.html' + _TESTS = [{ + 'url': 'http://www.sexykarma.com/gonewild/video/taking-a-quick-pee-yHI70cOyIHt.html', + 'md5': 'b9798e7d1ef1765116a8f516c8091dbd', + 'info_dict': { + 'id': 'yHI70cOyIHt', + 'display_id': 'taking-a-quick-pee', + 'ext': 'mp4', + 'title': 'Taking a quick pee.', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'wildginger7', + 'upload_date': '20141007', + 'duration': 22, + 'view_count': int, + 'comment_count': int, + 'categories': list, + } + }, { + 'url': 'http://www.sexykarma.com/gonewild/video/pot-pixie-tribute-8Id6EZPbuHf.html', + 'md5': 'dd216c68d29b49b12842b9babe762a5d', + 'info_dict': { + 'id': '8Id6EZPbuHf', + 'display_id': 'pot-pixie-tribute', + 'ext': 'mp4', + 'title': 'pot_pixie tribute', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'banffite', + 'upload_date': '20141013', + 'duration': 16, + 'view_count': int, + 'comment_count': int, + 'categories': list, + } + }, { + 'url': 'http://www.watchindianporn.net/video/desi-dancer-namrata-stripping-completely-nude-and-dancing-on-a-hot-number-dW2mtctxJfs.html', + 'md5': '9afb80675550406ed9a63ac2819ef69d', + 'info_dict': { + 'id': 'dW2mtctxJfs', + 'display_id': 'desi-dancer-namrata-stripping-completely-nude-and-dancing-on-a-hot-number', + 'ext': 'mp4', + 'title': 'Desi dancer namrata stripping completely nude and dancing on a hot number', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'Don', + 'upload_date': '20140213', + 'duration': 83, + 'view_count': int, + 'comment_count': int, + 'categories': list, + } + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + display_id = mobj.group('display_id') + + webpage = self._download_webpage(url, display_id) + + video_url = self._html_search_regex( + r"url: escape\('([^']+)'\)", webpage, 'url') + + title = self._html_search_regex( + r'<h2 class="he2"><span>(.*?)</span>', + webpage, 'title') + thumbnail = self._html_search_regex( + r'<span id="container"><img\s+src="([^"]+)"', + webpage, 'thumbnail', fatal=False) + + uploader = self._html_search_regex( + r'class="aupa">\s*(.*?)</a>', + webpage, 'uploader') + upload_date = unified_strdate(self._html_search_regex( + r'Added: <strong>(.+?)</strong>', webpage, 'upload date', fatal=False)) + + duration = parse_duration(self._search_regex( + r'<td>Time:\s*</td>\s*<td align="right"><span>\s*(.+?)\s*</span>', + webpage, 'duration', fatal=False)) + + view_count = int_or_none(self._search_regex( + r'<td>Views:\s*</td>\s*<td align="right"><span>\s*(\d+)\s*</span>', + webpage, 'view count', fatal=False)) + comment_count = int_or_none(self._search_regex( + r'<td>Comments:\s*</td>\s*<td align="right"><span>\s*(\d+)\s*</span>', + webpage, 'comment count', fatal=False)) + + categories = re.findall( + r'<a href="[^"]+/search/video/desi"><span>([^<]+)</span></a>', + webpage) + + return { + 'id': video_id, + 'display_id': display_id, + 'url': video_url, + 'title': title, + 'thumbnail': thumbnail, + 'uploader': uploader, + 'upload_date': upload_date, + 'duration': duration, + 'view_count': view_count, + 'comment_count': comment_count, + 'categories': categories, + } diff --git a/youtube_dl/extractor/ted.py b/youtube_dl/extractor/ted.py index d5e28efad..cd4af96fd 100644 --- a/youtube_dl/extractor/ted.py +++ b/youtube_dl/extractor/ted.py @@ -65,6 +65,22 @@ class TEDIE(SubtitlesInfoExtractor): 'title': 'Who are the hackers?', }, 'playlist_mincount': 6, + }, { + # contains a youtube video + 'url': 'https://www.ted.com/talks/douglas_adams_parrots_the_universe_and_everything', + 'add_ie': ['Youtube'], + 'info_dict': { + 'id': '_ZG8HBuDjgc', + 'ext': 'mp4', + 'title': 'Douglas Adams: Parrots the Universe and Everything', + 'description': 'md5:01ad1e199c49ac640cb1196c0e9016af', + 'uploader': 'University of California Television (UCTV)', + 'uploader_id': 'UCtelevision', + 'upload_date': '20080522', + }, + 'params': { + 'skip_download': True, + }, }] _NATIVE_FORMATS = { @@ -114,6 +130,13 @@ class TEDIE(SubtitlesInfoExtractor): talk_info = self._extract_info(webpage)['talks'][0] + if talk_info.get('external') is not None: + self.to_screen('Found video from %s' % talk_info['external']['service']) + return { + '_type': 'url', + 'url': talk_info['external']['uri'], + } + formats = [{ 'url': format_url, 'format_id': format_id, diff --git a/youtube_dl/extractor/telecinco.py b/youtube_dl/extractor/telecinco.py new file mode 100644 index 000000000..db9788c18 --- /dev/null +++ b/youtube_dl/extractor/telecinco.py @@ -0,0 +1,19 @@ +#coding: utf-8 +from __future__ import unicode_literals + +from .mitele import MiTeleIE + + +class TelecincoIE(MiTeleIE): + IE_NAME = 'telecinco.es' + _VALID_URL = r'https?://www\.telecinco\.es/[^/]+/[^/]+/[^/]+/(?P<episode>.*?)\.html' + + _TEST = { + 'url': 'http://www.telecinco.es/robinfood/temporada-01/t01xp14/Bacalao-cocochas-pil-pil_0_1876350223.html', + 'info_dict': { + 'id': 'MDSVID20141015_0058', + 'ext': 'mp4', + 'title': 'Con Martín Berasategui, hacer un bacalao al ...', + 'duration': 662, + }, + } diff --git a/youtube_dl/extractor/twitch.py b/youtube_dl/extractor/twitch.py new file mode 100644 index 000000000..36aa1ad6e --- /dev/null +++ b/youtube_dl/extractor/twitch.py @@ -0,0 +1,187 @@ +from __future__ import unicode_literals + +import itertools +import re + +from .common import InfoExtractor +from ..utils import ( + ExtractorError, + parse_iso8601, +) + + +class TwitchIE(InfoExtractor): + # TODO: One broadcast may be split into multiple videos. The key + # 'broadcast_id' is the same for all parts, and 'broadcast_part' + # starts at 1 and increases. Can we treat all parts as one video? + _VALID_URL = r"""(?x)^(?:http://)?(?:www\.)?twitch\.tv/ + (?: + (?P<channelid>[^/]+)| + (?:(?:[^/]+)/b/(?P<videoid>[^/]+))| + (?:(?:[^/]+)/c/(?P<chapterid>[^/]+)) + ) + /?(?:\#.*)?$ + """ + _PAGE_LIMIT = 100 + _API_BASE = 'https://api.twitch.tv' + _TESTS = [{ + 'url': 'http://www.twitch.tv/riotgames/b/577357806', + 'info_dict': { + 'id': 'a577357806', + 'title': 'Worlds Semifinals - Star Horn Royal Club vs. OMG', + }, + 'playlist_mincount': 12, + }, { + 'url': 'http://www.twitch.tv/acracingleague/c/5285812', + 'info_dict': { + 'id': 'c5285812', + 'title': 'ACRL Off Season - Sports Cars @ Nordschleife', + }, + 'playlist_mincount': 3, + }, { + 'url': 'http://www.twitch.tv/vanillatv', + 'info_dict': { + 'id': 'vanillatv', + 'title': 'VanillaTV', + }, + 'playlist_mincount': 412, + }] + + def _handle_error(self, response): + if not isinstance(response, dict): + return + error = response.get('error') + if error: + raise ExtractorError( + '%s returned error: %s - %s' % (self.IE_NAME, error, response.get('message')), + expected=True) + + def _download_json(self, url, video_id, note='Downloading JSON metadata'): + response = super(TwitchIE, self)._download_json(url, video_id, note) + self._handle_error(response) + return response + + def _extract_media(self, item, item_id): + ITEMS = { + 'a': 'video', + 'c': 'chapter', + } + info = self._extract_info(self._download_json( + '%s/kraken/videos/%s%s' % (self._API_BASE, item, item_id), item_id, + 'Downloading %s info JSON' % ITEMS[item])) + response = self._download_json( + '%s/api/videos/%s%s' % (self._API_BASE, item, item_id), item_id, + 'Downloading %s playlist JSON' % ITEMS[item]) + entries = [] + chunks = response['chunks'] + qualities = list(chunks.keys()) + for num, fragment in enumerate(zip(*chunks.values()), start=1): + formats = [] + for fmt_num, fragment_fmt in enumerate(fragment): + format_id = qualities[fmt_num] + fmt = { + 'url': fragment_fmt['url'], + 'format_id': format_id, + 'quality': 1 if format_id == 'live' else 0, + } + m = re.search(r'^(?P<height>\d+)[Pp]', format_id) + if m: + fmt['height'] = int(m.group('height')) + formats.append(fmt) + self._sort_formats(formats) + entry = dict(info) + entry['id'] = '%s_%d' % (entry['id'], num) + entry['title'] = '%s part %d' % (entry['title'], num) + entry['formats'] = formats + entries.append(entry) + return self.playlist_result(entries, info['id'], info['title']) + + def _extract_info(self, info): + return { + 'id': info['_id'], + 'title': info['title'], + 'description': info['description'], + 'duration': info['length'], + 'thumbnail': info['preview'], + 'uploader': info['channel']['display_name'], + 'uploader_id': info['channel']['name'], + 'timestamp': parse_iso8601(info['recorded_at']), + 'view_count': info['views'], + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + if mobj.group('chapterid'): + return self._extract_media('c', mobj.group('chapterid')) + + """ + webpage = self._download_webpage(url, chapter_id) + m = re.search(r'PP\.archive_id = "([0-9]+)";', webpage) + if not m: + raise ExtractorError('Cannot find archive of a chapter') + archive_id = m.group(1) + + api = api_base + '/broadcast/by_chapter/%s.xml' % chapter_id + doc = self._download_xml( + api, chapter_id, + note='Downloading chapter information', + errnote='Chapter information download failed') + for a in doc.findall('.//archive'): + if archive_id == a.find('./id').text: + break + else: + raise ExtractorError('Could not find chapter in chapter information') + + video_url = a.find('./video_file_url').text + video_ext = video_url.rpartition('.')[2] or 'flv' + + chapter_api_url = 'https://api.twitch.tv/kraken/videos/c' + chapter_id + chapter_info = self._download_json( + chapter_api_url, 'c' + chapter_id, + note='Downloading chapter metadata', + errnote='Download of chapter metadata failed') + + bracket_start = int(doc.find('.//bracket_start').text) + bracket_end = int(doc.find('.//bracket_end').text) + + # TODO determine start (and probably fix up file) + # youtube-dl -v http://www.twitch.tv/firmbelief/c/1757457 + #video_url += '?start=' + TODO:start_timestamp + # bracket_start is 13290, but we want 51670615 + self._downloader.report_warning('Chapter detected, but we can just download the whole file. ' + 'Chapter starts at %s and ends at %s' % (formatSeconds(bracket_start), formatSeconds(bracket_end))) + + info = { + 'id': 'c' + chapter_id, + 'url': video_url, + 'ext': video_ext, + 'title': chapter_info['title'], + 'thumbnail': chapter_info['preview'], + 'description': chapter_info['description'], + 'uploader': chapter_info['channel']['display_name'], + 'uploader_id': chapter_info['channel']['name'], + } + return info + """ + elif mobj.group('videoid'): + return self._extract_media('a', mobj.group('videoid')) + elif mobj.group('channelid'): + channel_id = mobj.group('channelid') + info = self._download_json( + '%s/kraken/channels/%s' % (self._API_BASE, channel_id), + channel_id, 'Downloading channel info JSON') + channel_name = info.get('display_name') or info.get('name') + entries = [] + offset = 0 + limit = self._PAGE_LIMIT + for counter in itertools.count(1): + response = self._download_json( + '%s/kraken/channels/%s/videos/?offset=%d&limit=%d' + % (self._API_BASE, channel_id, offset, limit), + channel_id, 'Downloading channel videos JSON page %d' % counter) + videos = response['videos'] + if not videos: + break + entries.extend([self.url_result(video['url'], 'Twitch') for video in videos]) + offset += limit + return self.playlist_result(entries, channel_id, channel_name) diff --git a/youtube_dl/extractor/vidzi.py b/youtube_dl/extractor/vidzi.py new file mode 100644 index 000000000..669979e13 --- /dev/null +++ b/youtube_dl/extractor/vidzi.py @@ -0,0 +1,33 @@ +#coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor + + +class VidziIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?vidzi\.tv/(?P<id>\w+)' + _TEST = { + 'url': 'http://vidzi.tv/cghql9yq6emu.html', + 'md5': '4f16c71ca0c8c8635ab6932b5f3f1660', + 'info_dict': { + 'id': 'cghql9yq6emu', + 'ext': 'mp4', + 'title': 'youtube-dl test video 1\\\\2\'3/4<5\\\\6ä7↭', + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + video_url = self._html_search_regex( + r'{\s*file\s*:\s*"([^"]+)"\s*}', webpage, 'video url') + title = self._html_search_regex( + r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title') + + return { + 'id': video_id, + 'title': title, + 'url': video_url, + } +
\ No newline at end of file diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index e6a86f18e..07959d3fe 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -157,6 +157,18 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor): 'duration': 62, } }, + { + # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/ + 'url': 'https://player.vimeo.com/video/98044508', + 'note': 'The js code contains assignments to the same variable as the config', + 'info_dict': { + 'id': '98044508', + 'ext': 'mp4', + 'title': 'Pier Solar OUYA Official Trailer', + 'uploader': 'Tulio Gonçalves', + 'uploader_id': 'user28849593', + }, + }, ] def _verify_video_password(self, url, video_id, webpage): @@ -244,7 +256,7 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor): # 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)) + 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, 'info section', diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index cfae2de89..4ab56e0ac 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -191,8 +191,9 @@ class YoutubeBaseInfoExtractor(InfoExtractor): def _real_initialize(self): if self._downloader is None: return - if not self._set_language(): - return + if self._get_login_info()[0] is not None: + if not self._set_language(): + return if not self._login(): return self._confirm_age() |