import urllib
from .utils import *
-
-
from .extractor.common import InfoExtractor, SearchInfoExtractor
+
+from .extractor.ard import ARDIE
+from .extractor.arte import ArteTvIE
+from .extractor.dailymotion import DailymotionIE
from .extractor.metacafe import MetacafeIE
from .extractor.statigram import StatigramIE
+from .extractor.photobucket import PhotobucketIE
+from .extractor.vimeo import VimeoIE
+from .extractor.yahoo import YahooIE
from .extractor.youtube import YoutubeIE, YoutubePlaylistIE, YoutubeUserIE, YoutubeChannelIE
+from .extractor.zdf import ZDFIE
-class DailymotionIE(InfoExtractor):
- """Information Extractor for Dailymotion"""
-
- _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/video/([^/]+)'
- IE_NAME = u'dailymotion'
-
- def _real_extract(self, url):
- # Extract id and simplified title from URL
- mobj = re.match(self._VALID_URL, url)
- if mobj is None:
- raise ExtractorError(u'Invalid URL: %s' % url)
-
- video_id = mobj.group(1).split('_')[0].split('?')[0]
-
- video_extension = 'mp4'
-
- # Retrieve video webpage to extract further information
- request = compat_urllib_request.Request(url)
- request.add_header('Cookie', 'family_filter=off')
- webpage = self._download_webpage(request, video_id)
-
- # Extract URL, uploader and title from webpage
- self.report_extraction(video_id)
- mobj = re.search(r'\s*var flashvars = (.*)', webpage)
- if mobj is None:
- raise ExtractorError(u'Unable to extract media URL')
- flashvars = compat_urllib_parse.unquote(mobj.group(1))
-
- for key in ['hd1080URL', 'hd720URL', 'hqURL', 'sdURL', 'ldURL', 'video_url']:
- if key in flashvars:
- max_quality = key
- self.to_screen(u'Using %s' % key)
- break
- else:
- raise ExtractorError(u'Unable to extract video URL')
-
- mobj = re.search(r'"' + max_quality + r'":"(.+?)"', flashvars)
- if mobj is None:
- raise ExtractorError(u'Unable to extract video URL')
-
- video_url = compat_urllib_parse.unquote(mobj.group(1)).replace('\\/', '/')
-
- # TODO: support choosing qualities
-
- mobj = re.search(r'<meta property="og:title" content="(?P<title>[^"]*)" />', webpage)
- if mobj is None:
- raise ExtractorError(u'Unable to extract title')
- video_title = unescapeHTML(mobj.group('title'))
-
- video_uploader = None
- video_uploader = self._search_regex([r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>',
- # Looking for official user
- r'<(?:span|a) .*?rel="author".*?>([^<]+?)</'],
- webpage, 'video uploader')
-
- video_upload_date = None
- mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
- if mobj is not None:
- video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
-
- return [{
- 'id': video_id,
- 'url': video_url,
- 'uploader': video_uploader,
- 'upload_date': video_upload_date,
- 'title': video_title,
- 'ext': video_extension,
- }]
-class PhotobucketIE(InfoExtractor):
- """Information extractor for photobucket.com."""
- # TODO: the original _VALID_URL was:
- # r'(?:http://)?(?:[a-z0-9]+\.)?photobucket\.com/.*[\?\&]current=(.*\.flv)'
- # Check if it's necessary to keep the old extracion process
- _VALID_URL = r'(?:http://)?(?:[a-z0-9]+\.)?photobucket\.com/.*(([\?\&]current=)|_)(?P<id>.*)\.(?P<ext>(flv)|(mp4))'
- IE_NAME = u'photobucket'
- def _real_extract(self, url):
- # Extract id from URL
- mobj = re.match(self._VALID_URL, url)
- if mobj is None:
- raise ExtractorError(u'Invalid URL: %s' % url)
- video_id = mobj.group('id')
- video_extension = mobj.group('ext')
-
- # Retrieve video webpage to extract further information
- webpage = self._download_webpage(url, video_id)
-
- # Extract URL, uploader, and title from webpage
- self.report_extraction(video_id)
- # We try first by looking the javascript code:
- mobj = re.search(r'Pb\.Data\.Shared\.put\(Pb\.Data\.Shared\.MEDIA, (?P<json>.*?)\);', webpage)
- if mobj is not None:
- info = json.loads(mobj.group('json'))
- return [{
- 'id': video_id,
- 'url': info[u'downloadUrl'],
- 'uploader': info[u'username'],
- 'upload_date': datetime.date.fromtimestamp(info[u'creationDate']).strftime('%Y%m%d'),
- 'title': info[u'title'],
- 'ext': video_extension,
- 'thumbnail': info[u'thumbUrl'],
- }]
-
- # We try looking in other parts of the webpage
- video_url = self._search_regex(r'<link rel="video_src" href=".*\?file=([^"]+)" />',
- webpage, u'video URL')
-
- mobj = re.search(r'<title>(.*) video by (.*) - Photobucket</title>', webpage)
- if mobj is None:
- raise ExtractorError(u'Unable to extract title')
- video_title = mobj.group(1).decode('utf-8')
- video_uploader = mobj.group(2).decode('utf-8')
-
- return [{
- 'id': video_id.decode('utf-8'),
- 'url': video_url.decode('utf-8'),
- 'uploader': video_uploader,
- 'upload_date': None,
- 'title': video_title,
- 'ext': video_extension.decode('utf-8'),
- }]
-
-
-class YahooIE(InfoExtractor):
- """Information extractor for screen.yahoo.com."""
- _VALID_URL = r'http://screen\.yahoo\.com/.*?-(?P<id>\d*?)\.html'
-
- def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
- if mobj is None:
- raise ExtractorError(u'Invalid URL: %s' % url)
- video_id = mobj.group('id')
- webpage = self._download_webpage(url, video_id)
- m_id = re.search(r'YUI\.namespace\("Media"\)\.CONTENT_ID = "(?P<new_id>.+?)";', webpage)
-
- if m_id is None:
- # TODO: Check which url parameters are required
- info_url = 'http://cosmos.bcst.yahoo.com/rest/v2/pops;lmsoverride=1;outputformat=mrss;cb=974419660;id=%s;rd=news.yahoo.com;datacontext=mdb;lg=KCa2IihxG3qE60vQ7HtyUy' % video_id
- webpage = self._download_webpage(info_url, video_id, u'Downloading info webpage')
- info_re = r'''<title><!\[CDATA\[(?P<title>.*?)\]\]></title>.*
- <description><!\[CDATA\[(?P<description>.*?)\]\]></description>.*
- <media:pubStart><!\[CDATA\[(?P<date>.*?)\ .*\]\]></media:pubStart>.*
- <media:content\ medium="image"\ url="(?P<thumb>.*?)"\ name="LARGETHUMB"
- '''
- self.report_extraction(video_id)
- m_info = re.search(info_re, webpage, re.VERBOSE|re.DOTALL)
- if m_info is None:
- raise ExtractorError(u'Unable to extract video info')
- video_title = m_info.group('title')
- video_description = m_info.group('description')
- video_thumb = m_info.group('thumb')
- video_date = m_info.group('date')
- video_date = datetime.datetime.strptime(video_date, '%m/%d/%Y').strftime('%Y%m%d')
-
- # TODO: Find a way to get mp4 videos
- rest_url = 'http://cosmos.bcst.yahoo.com/rest/v2/pops;element=stream;outputformat=mrss;id=%s;lmsoverride=1;bw=375;dynamicstream=1;cb=83521105;tech=flv,mp4;rd=news.yahoo.com;datacontext=mdb;lg=KCa2IihxG3qE60vQ7HtyUy' % video_id
- webpage = self._download_webpage(rest_url, video_id, u'Downloading video url webpage')
- m_rest = re.search(r'<media:content url="(?P<url>.*?)" path="(?P<path>.*?)"', webpage)
- video_url = m_rest.group('url')
- video_path = m_rest.group('path')
- if m_rest is None:
- raise ExtractorError(u'Unable to extract video url')
-
- else: # We have to use a different method if another id is defined
- long_id = m_id.group('new_id')
- info_url = 'http://video.query.yahoo.com/v1/public/yql?q=SELECT%20*%20FROM%20yahoo.media.video.streams%20WHERE%20id%3D%22' + long_id + '%22%20AND%20format%3D%22mp4%2Cflv%22%20AND%20protocol%3D%22rtmp%2Chttp%22%20AND%20plrs%3D%2286Gj0vCaSzV_Iuf6hNylf2%22%20AND%20acctid%3D%22389%22%20AND%20plidl%3D%22%22%20AND%20pspid%3D%22792700001%22%20AND%20offnetwork%3D%22false%22%20AND%20site%3D%22ivy%22%20AND%20lang%3D%22en-US%22%20AND%20region%3D%22US%22%20AND%20override%3D%22none%22%3B&env=prod&format=json&callback=YUI.Env.JSONP.yui_3_8_1_1_1368368376830_335'
- webpage = self._download_webpage(info_url, video_id, u'Downloading info json')
- json_str = re.search(r'YUI.Env.JSONP.yui.*?\((.*?)\);', webpage).group(1)
- info = json.loads(json_str)
- res = info[u'query'][u'results'][u'mediaObj'][0]
- stream = res[u'streams'][0]
- video_path = stream[u'path']
- video_url = stream[u'host']
- meta = res[u'meta']
- video_title = meta[u'title']
- video_description = meta[u'description']
- video_thumb = meta[u'thumbnail']
- video_date = None # I can't find it
-
- info_dict = {
- 'id': video_id,
- 'url': video_url,
- 'play_path': video_path,
- 'title':video_title,
- 'description': video_description,
- 'thumbnail': video_thumb,
- 'upload_date': video_date,
- 'ext': 'flv',
- }
- return info_dict
-
-class VimeoIE(InfoExtractor):
- """Information extractor for vimeo.com."""
-
- # _VALID_URL matches Vimeo URLs
- _VALID_URL = r'(?P<proto>https?://)?(?:(?:www|player)\.)?vimeo(?P<pro>pro)?\.com/(?:(?:(?:groups|album)/[^/]+)|(?:.*?)/)?(?P<direct_link>play_redirect_hls\?clip_id=)?(?:videos?/)?(?P<id>[0-9]+)'
- IE_NAME = u'vimeo'
-
- def _verify_video_password(self, url, video_id, webpage):
- password = self._downloader.params.get('password', None)
- if password is None:
- raise ExtractorError(u'This video is protected by a password, use the --password option')
- token = re.search(r'xsrft: \'(.*?)\'', webpage).group(1)
- data = compat_urllib_parse.urlencode({'password': password,
- 'token': token})
- # I didn't manage to use the password with https
- if url.startswith('https'):
- pass_url = url.replace('https','http')
- else:
- pass_url = url
- password_request = compat_urllib_request.Request(pass_url+'/password', data)
- password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
- password_request.add_header('Cookie', 'xsrft=%s' % token)
- pass_web = self._download_webpage(password_request, video_id,
- u'Verifying the password',
- u'Wrong password')
-
- def _real_extract(self, url, new_video=True):
- # Extract ID from URL
- mobj = re.match(self._VALID_URL, url)
- if mobj is None:
- raise ExtractorError(u'Invalid URL: %s' % url)
-
- video_id = mobj.group('id')
- if not mobj.group('proto'):
- url = 'https://' + url
- if mobj.group('direct_link') or mobj.group('pro'):
- url = 'https://vimeo.com/' + video_id
-
- # Retrieve video webpage to extract further information
- request = compat_urllib_request.Request(url, None, std_headers)
- webpage = self._download_webpage(request, video_id)
-
- # Now we begin extracting as much information as we can from what we
- # retrieved. First we extract the information common to all extractors,
- # and latter we extract those that are Vimeo specific.
- self.report_extraction(video_id)
-
- # Extract the config JSON
- try:
- config = webpage.split(' = {config:')[1].split(',assets:')[0]
- config = json.loads(config)
- except:
- if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
- raise ExtractorError(u'The author has restricted the access to this video, try with the "--referer" option')
-
- if re.search('If so please provide the correct password.', webpage):
- self._verify_video_password(url, video_id, webpage)
- return self._real_extract(url)
- else:
- raise ExtractorError(u'Unable to extract info section')
-
- # Extract title
- video_title = config["video"]["title"]
-
- # Extract uploader and uploader_id
- video_uploader = config["video"]["owner"]["name"]
- video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
-
- # Extract video thumbnail
- video_thumbnail = config["video"]["thumbnail"]
-
- # Extract video description
- video_description = get_element_by_attribute("itemprop", "description", webpage)
- if video_description: video_description = clean_html(video_description)
- else: video_description = u''
-
- # Extract upload date
- video_upload_date = None
- mobj = re.search(r'<meta itemprop="dateCreated" content="(\d{4})-(\d{2})-(\d{2})T', webpage)
- if mobj is not None:
- video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
-
- # Vimeo specific: extract request signature and timestamp
- sig = config['request']['signature']
- timestamp = config['request']['timestamp']
-
- # Vimeo specific: extract video codec and quality information
- # First consider quality, then codecs, then take everything
- # TODO bind to format param
- codecs = [('h264', 'mp4'), ('vp8', 'flv'), ('vp6', 'flv')]
- files = { 'hd': [], 'sd': [], 'other': []}
- for codec_name, codec_extension in codecs:
- if codec_name in config["video"]["files"]:
- if 'hd' in config["video"]["files"][codec_name]:
- files['hd'].append((codec_name, codec_extension, 'hd'))
- elif 'sd' in config["video"]["files"][codec_name]:
- files['sd'].append((codec_name, codec_extension, 'sd'))
- else:
- files['other'].append((codec_name, codec_extension, config["video"]["files"][codec_name][0]))
-
- for quality in ('hd', 'sd', 'other'):
- if len(files[quality]) > 0:
- video_quality = files[quality][0][2]
- video_codec = files[quality][0][0]
- video_extension = files[quality][0][1]
- self.to_screen(u'%s: Downloading %s file at %s quality' % (video_id, video_codec.upper(), video_quality))
- break
- else:
- raise ExtractorError(u'No known codec found')
-
- video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
- %(video_id, sig, timestamp, video_quality, video_codec.upper())
-
- return [{
- 'id': video_id,
- 'url': video_url,
- 'uploader': video_uploader,
- 'uploader_id': video_uploader_id,
- 'upload_date': video_upload_date,
- 'title': video_title,
- 'ext': video_extension,
- 'thumbnail': video_thumbnail,
- 'description': video_description,
- }]
-
-
-class ArteTvIE(InfoExtractor):
- """arte.tv information extractor."""
-
- _VALID_URL = r'(?:http://)?videos\.arte\.tv/(?:fr|de)/videos/.*'
- _LIVE_URL = r'index-[0-9]+\.html$'
-
- IE_NAME = u'arte.tv'
-
- def fetch_webpage(self, url):
- request = compat_urllib_request.Request(url)
- try:
- self.report_download_webpage(url)
- webpage = 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 video webpage: %s' % compat_str(err))
- except ValueError as err:
- raise ExtractorError(u'Invalid URL: %s' % url)
- return webpage
-
- def grep_webpage(self, url, regex, regexFlags, matchTuples):
- page = self.fetch_webpage(url)
- mobj = re.search(regex, page, regexFlags)
- info = {}
-
- if mobj is None:
- raise ExtractorError(u'Invalid URL: %s' % url)
-
- for (i, key, err) in matchTuples:
- if mobj.group(i) is None:
- raise ExtractorError(err)
- else:
- info[key] = mobj.group(i)
-
- return info
-
- def extractLiveStream(self, url):
- video_lang = url.split('/')[-4]
- info = self.grep_webpage(
- url,
- r'src="(.*?/videothek_js.*?\.js)',
- 0,
- [
- (1, 'url', u'Invalid URL: %s' % url)
- ]
- )
- http_host = url.split('/')[2]
- next_url = 'http://%s%s' % (http_host, compat_urllib_parse.unquote(info.get('url')))
- info = self.grep_webpage(
- next_url,
- r'(s_artestras_scst_geoFRDE_' + video_lang + '.*?)\'.*?' +
- '(http://.*?\.swf).*?' +
- '(rtmp://.*?)\'',
- re.DOTALL,
- [
- (1, 'path', u'could not extract video path: %s' % url),
- (2, 'player', u'could not extract video player: %s' % url),
- (3, 'url', u'could not extract video url: %s' % url)
- ]
- )
- video_url = u'%s/%s' % (info.get('url'), info.get('path'))
-
- def extractPlus7Stream(self, url):
- video_lang = url.split('/')[-3]
- info = self.grep_webpage(
- url,
- r'param name="movie".*?videorefFileUrl=(http[^\'"&]*)',
- 0,
- [
- (1, 'url', u'Invalid URL: %s' % url)
- ]
- )
- next_url = compat_urllib_parse.unquote(info.get('url'))
- info = self.grep_webpage(
- next_url,
- r'<video lang="%s" ref="(http[^\'"&]*)' % video_lang,
- 0,
- [
- (1, 'url', u'Could not find <video> tag: %s' % url)
- ]
- )
- next_url = compat_urllib_parse.unquote(info.get('url'))
-
- info = self.grep_webpage(
- next_url,
- r'<video id="(.*?)".*?>.*?' +
- '<name>(.*?)</name>.*?' +
- '<dateVideo>(.*?)</dateVideo>.*?' +
- '<url quality="hd">(.*?)</url>',
- re.DOTALL,
- [
- (1, 'id', u'could not extract video id: %s' % url),
- (2, 'title', u'could not extract video title: %s' % url),
- (3, 'date', u'could not extract video date: %s' % url),
- (4, 'url', u'could not extract video url: %s' % url)
- ]
- )
-
- return {
- 'id': info.get('id'),
- 'url': compat_urllib_parse.unquote(info.get('url')),
- 'uploader': u'arte.tv',
- 'upload_date': unified_strdate(info.get('date')),
- 'title': info.get('title').decode('utf-8'),
- 'ext': u'mp4',
- 'format': u'NA',
- 'player_url': None,
- }
-
- def _real_extract(self, url):
- video_id = url.split('/')[-1]
- self.report_extraction(video_id)
-
- if re.search(self._LIVE_URL, video_id) is not None:
- self.extractLiveStream(url)
- return
- else:
- info = self.extractPlus7Stream(url)
-
- return [info]
class GenericIE(InfoExtractor):
return [info]
-class ARDIE(InfoExtractor):
- _VALID_URL = r'^(?:https?://)?(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[^/\?]+)(?:\?.*)?'
- _TITLE = r'<h1(?: class="boxTopHeadline")?>(?P<title>.*)</h1>'
- _MEDIA_STREAM = r'mediaCollection\.addMediaStream\((?P<media_type>\d+), (?P<quality>\d+), "(?P<rtmp_url>[^"]*)", "(?P<video_url>[^"]*)", "[^"]*"\)'
-
- def _real_extract(self, url):
- # determine video id from url
- m = re.match(self._VALID_URL, url)
-
- numid = re.search(r'documentId=([0-9]+)', url)
- if numid:
- video_id = numid.group(1)
- else:
- video_id = m.group('video_id')
-
- # determine title and media streams from webpage
- html = self._download_webpage(url, video_id)
- title = re.search(self._TITLE, html).group('title')
- streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)]
- if not streams:
- assert '"fsk"' in html
- raise ExtractorError(u'This video is only available after 8:00 pm')
-
- # choose default media type and highest quality for now
- stream = max([s for s in streams if int(s["media_type"]) == 0],
- key=lambda s: int(s["quality"]))
-
- # there's two possibilities: RTMP stream or HTTP download
- info = {'id': video_id, 'title': title, 'ext': 'mp4'}
- if stream['rtmp_url']:
- self.to_screen(u'RTMP download detected')
- assert stream['video_url'].startswith('mp4:')
- info["url"] = stream["rtmp_url"]
- info["play_path"] = stream['video_url']
- else:
- assert stream["video_url"].endswith('.mp4')
- info["url"] = stream["video_url"]
- return [info]
-
-class ZDFIE(InfoExtractor):
- _VALID_URL = r'^http://www\.zdf\.de\/ZDFmediathek\/(.*beitrag\/video\/)(?P<video_id>[^/\?]+)(?:\?.*)?'
- _TITLE = r'<h1(?: class="beitragHeadline")?>(?P<title>.*)</h1>'
- _MEDIA_STREAM = r'<a href="(?P<video_url>.+(?P<media_type>.streaming).+/zdf/(?P<quality>[^\/]+)/[^"]*)".+class="play".+>'
- _MMS_STREAM = r'href="(?P<video_url>mms://[^"]*)"'
- _RTSP_STREAM = r'(?P<video_url>rtsp://[^"]*.mp4)'
- def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
- if mobj is None:
- raise ExtractorError(u'Invalid URL: %s' % url)
- video_id = mobj.group('video_id')
-
- html = self._download_webpage(url, video_id)
- streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)]
- if streams is None:
- raise ExtractorError(u'No media url found.')
-
- # s['media_type'] == 'wstreaming' -> use 'Windows Media Player' and mms url
- # s['media_type'] == 'hstreaming' -> use 'Quicktime' and rtsp url
- # choose first/default media type and highest quality for now
- for s in streams: #find 300 - dsl1000mbit
- if s['quality'] == '300' and s['media_type'] == 'wstreaming':
- stream_=s
- break
- for s in streams: #find veryhigh - dsl2000mbit
- if s['quality'] == 'veryhigh' and s['media_type'] == 'wstreaming': # 'hstreaming' - rtsp is not working
- stream_=s
- break
- if stream_ is None:
- raise ExtractorError(u'No stream found.')
-
- media_link = self._download_webpage(stream_['video_url'], video_id,'Get stream URL')
-
- self.report_extraction(video_id)
- mobj = re.search(self._TITLE, html)
- if mobj is None:
- raise ExtractorError(u'Cannot extract title')
- title = unescapeHTML(mobj.group('title'))
-
- mobj = re.search(self._MMS_STREAM, media_link)
- if mobj is None:
- mobj = re.search(self._RTSP_STREAM, media_link)
- if mobj is None:
- raise ExtractorError(u'Cannot extract mms:// or rtsp:// URL')
- mms_url = mobj.group('video_url')
-
- mobj = re.search('(.*)[.](?P<ext>[^.]+)', mms_url)
- if mobj is None:
- raise ExtractorError(u'Cannot extract extention')
- ext = mobj.group('ext')
-
- return [{'id': video_id,
- 'url': mms_url,
- 'title': title,
- 'ext': ext
- }]
class TumblrIE(InfoExtractor):
_VALID_URL = r'http://(?P<blog_name>.*?)\.tumblr\.com/((post)|(video))/(?P<id>\d*)/(.*?)'