from .utils import *
-class InfoExtractor(object):
- """Information Extractor class.
-
- Information extractors are the classes that, given a URL, extract
- information about the video (or videos) the URL refers to. This
- information includes the real video URL, the video title, author and
- others. The information is stored in a dictionary which is then
- passed to the FileDownloader. The FileDownloader processes this
- information possibly downloading the video to the file system, among
- other possible outcomes.
-
- The dictionaries must include the following fields:
-
- id: Video identifier.
- url: Final video URL.
- title: Video title, unescaped.
- ext: Video filename extension.
-
- The following fields are optional:
-
- format: The video format, defaults to ext (used for --get-format)
- thumbnail: Full URL to a video thumbnail image.
- description: One-line video description.
- uploader: Full name of the video uploader.
- upload_date: Video upload date (YYYYMMDD).
- uploader_id: Nickname or id of the video uploader.
- location: Physical location of the video.
- player_url: SWF Player URL (used for rtmpdump).
- subtitles: The subtitle file contents.
- urlhandle: [internal] The urlHandle to be used to download the file,
- like returned by urllib.request.urlopen
-
- The fields should all be Unicode strings.
-
- Subclasses of this one should re-define the _real_initialize() and
- _real_extract() methods and define a _VALID_URL regexp.
- Probably, they should also be added to the list of extractors.
-
- _real_extract() must return a *list* of information dictionaries as
- described above.
-
- Finally, the _WORKING attribute should be set to False for broken IEs
- in order to warn the users and skip the tests.
- """
-
- _ready = False
- _downloader = None
- _WORKING = True
-
- def __init__(self, downloader=None):
- """Constructor. Receives an optional downloader."""
- self._ready = False
- self.set_downloader(downloader)
-
- @classmethod
- def suitable(cls, url):
- """Receives a URL and returns True if suitable for this IE."""
- return re.match(cls._VALID_URL, url) is not None
-
- @classmethod
- def working(cls):
- """Getter method for _WORKING."""
- return cls._WORKING
-
- def initialize(self):
- """Initializes an instance (authentication, etc)."""
- if not self._ready:
- self._real_initialize()
- self._ready = True
-
- def extract(self, url):
- """Extracts URL information and returns it in list of dicts."""
- self.initialize()
- return self._real_extract(url)
-
- def set_downloader(self, downloader):
- """Sets the downloader for this IE."""
- self._downloader = downloader
-
- def _real_initialize(self):
- """Real initialization process. Redefine in subclasses."""
- pass
-
- def _real_extract(self, url):
- """Real extraction process. Redefine in subclasses."""
- pass
-
- @property
- def IE_NAME(self):
- return type(self).__name__[:-2]
-
- def _request_webpage(self, url_or_request, video_id, note=None, errnote=None):
- """ 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))
- try:
- return compat_urllib_request.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])
-
- def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None):
- """ Returns a tuple (page content as string, URL handle) """
- urlh = self._request_webpage(url_or_request, video_id, note, errnote)
- content_type = urlh.headers.get('Content-Type', '')
- m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
- if m:
- encoding = m.group(1)
- else:
- encoding = 'utf-8'
- webpage_bytes = urlh.read()
- if self._downloader.params.get('dump_intermediate_pages', False):
- try:
- url = url_or_request.get_full_url()
- except AttributeError:
- url = url_or_request
- self.to_screen(u'Dumping request to ' + url)
- dump = base64.b64encode(webpage_bytes).decode('ascii')
- self._downloader.to_screen(dump)
- content = webpage_bytes.decode(encoding, 'replace')
- return (content, urlh)
-
- def _download_webpage(self, url_or_request, video_id, note=None, errnote=None):
- """ Returns the data of the page as a string """
- return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0]
-
- def to_screen(self, msg):
- """Print msg to screen, prefixing it with '[ie_name]'"""
- self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg))
-
- def report_extraction(self, id_or_name):
- """Report information extraction."""
- self.to_screen(u'%s: Extracting information' % id_or_name)
-
- def report_download_webpage(self, video_id):
- """Report webpage download."""
- self.to_screen(u'%s: Downloading webpage' % video_id)
-
- def report_age_confirmation(self):
- """Report attempt to confirm age."""
- self.to_screen(u'Confirming age')
-
- #Methods for following #608
- #They set the correct value of the '_type' key
- def video_result(self, video_info):
- """Returns a video"""
- video_info['_type'] = 'video'
- return video_info
- def url_result(self, url, ie=None):
- """Returns a url that points to a page that should be processed"""
- #TODO: ie should be the class used for getting the info
- video_info = {'_type': 'url',
- 'url': url,
- 'ie_key': ie}
- return video_info
- def playlist_result(self, entries, playlist_id=None, playlist_title=None):
- """Returns a playlist"""
- video_info = {'_type': 'playlist',
- 'entries': entries}
- if playlist_id:
- video_info['id'] = playlist_id
- if playlist_title:
- video_info['title'] = playlist_title
- return video_info
-
- def _search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
- """
- Perform a regex search on the given string, using a single or a list of
- patterns returning the first matching group.
- In case of failure return a default value or raise a WARNING or a
- ExtractorError, depending on fatal, specifying the field name.
- """
- if isinstance(pattern, (str, compat_str, compiled_regex_type)):
- mobj = re.search(pattern, string, flags)
- else:
- for p in pattern:
- mobj = re.search(p, string, flags)
- if mobj: break
-
- if sys.stderr.isatty() and os.name != 'nt':
- _name = u'\033[0;34m%s\033[0m' % name
- else:
- _name = name
-
- if mobj:
- # return the first matching group
- return next(g for g in mobj.groups() if g is not None)
- elif default is not None:
- return default
- elif fatal:
- raise ExtractorError(u'Unable to extract %s' % _name)
- else:
- self._downloader.report_warning(u'unable to extract %s; '
- u'please report this issue on GitHub.' % _name)
- return None
-
- def _html_search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
- """
- Like _search_regex, but strips HTML tags and unescapes entities.
- """
- res = self._search_regex(pattern, string, name, default, fatal, flags)
- if res:
- return clean_html(res).strip()
- else:
- return res
-
-class SearchInfoExtractor(InfoExtractor):
- """
- Base class for paged search queries extractors.
- They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query}
- Instances should define _SEARCH_KEY and _MAX_RESULTS.
- """
-
- @classmethod
- def _make_valid_url(cls):
- return r'%s(?P<prefix>|[1-9][0-9]*|all):(?P<query>[\s\S]+)' % cls._SEARCH_KEY
-
- @classmethod
- def suitable(cls, url):
- return re.match(cls._make_valid_url(), url) is not None
-
- def _real_extract(self, query):
- mobj = re.match(self._make_valid_url(), query)
- if mobj is None:
- raise ExtractorError(u'Invalid search query "%s"' % query)
-
- prefix = mobj.group('prefix')
- query = mobj.group('query')
- if prefix == '':
- return self._get_n_results(query, 1)
- elif prefix == 'all':
- return self._get_n_results(query, self._MAX_RESULTS)
- else:
- n = int(prefix)
- if n <= 0:
- raise ExtractorError(u'invalid download number %s for query "%s"' % (n, query))
- elif n > self._MAX_RESULTS:
- self._downloader.report_warning(u'%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n))
- n = self._MAX_RESULTS
- return self._get_n_results(query, n)
-
- def _get_n_results(self, query, n):
- """Get a specified number of results for a query"""
- raise NotImplementedError("This method must be implemented by sublclasses")
+from .extractor.common import InfoExtractor, SearchInfoExtractor
class YoutubeIE(InfoExtractor):
"""Indicate the download will use the RTMP protocol."""
self.to_screen(u'RTMP download detected')
+ @staticmethod
+ def _decrypt_signature(s):
+ """Decrypt the key the two subkeys must have a length of 43"""
+ (a,b) = s.split('.')
+ if len(a) != 43 or len(b) != 43:
+ raise ExtractorError(u'Unable to decrypt signature, subkeys lengths not valid')
+ b = ''.join([b[:8],a[0],b[9:18],b[-4],b[19:39], b[18]])[0:40]
+ a = a[-40:]
+ s_dec = '.'.join((a,b))[::-1]
+ return s_dec
+
def _get_available_subtitles(self, video_id):
self.report_video_subtitles_download(video_id)
request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
def _request_automatic_caption(self, video_id, webpage):
"""We need the webpage for getting the captions url, pass it as an
argument to speed up the process."""
- sub_lang = self._downloader.params.get('subtitleslang')
+ sub_lang = self._downloader.params.get('subtitleslang') or 'en'
sub_format = self._downloader.params.get('subtitlesformat')
self.to_screen(u'%s: Looking for automatic captions' % video_id)
mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
pass
else:
# We report the original error
- self._downloader.report_error(sub_error)
+ self._downloader.report_warning(sub_error)
if self._downloader.params.get('allsubtitles', False):
video_subtitles = self._extract_all_subtitles(video_id)
for video_subtitle in video_subtitles:
(sub_error, sub_lang, sub) = video_subtitle
if sub_error:
- self._downloader.report_error(sub_error)
+ self._downloader.report_warning(sub_error)
if self._downloader.params.get('listsubtitles', False):
sub_lang_list = self._list_available_subtitles(video_id)
# Decide which formats to download
req_format = self._downloader.params.get('format', None)
+ try:
+ mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage)
+ info = json.loads(mobj.group(1))
+ args = info['args']
+ if args.get('ptk','') == 'vevo' or 'dashmpd':
+ # Vevo videos with encrypted signatures
+ self.to_screen(u'%s: Vevo video detected.' % video_id)
+ video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
+ except ValueError:
+ pass
+
if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
self.report_rtmp_download()
video_url_list = [(None, video_info['conn'][0])]
for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','):
url_data = compat_parse_qs(url_data_str)
if 'itag' in url_data and 'url' in url_data:
- url = url_data['url'][0] + '&signature=' + url_data['sig'][0]
- if not 'ratebypass' in url: url += '&ratebypass=yes'
+ url = url_data['url'][0]
+ if 'sig' in url_data:
+ url += '&signature=' + url_data['sig'][0]
+ elif 's' in url_data:
+ signature = self._decrypt_signature(url_data['s'][0])
+ url += '&signature=' + signature
+ if 'ratebypass' not in url:
+ url += '&ratebypass=yes'
url_map[url_data['itag'][0]] = url
format_limit = self._downloader.params.get('format_limit', None)
video_title = unescapeHTML(mobj.group('title'))
video_uploader = None
- mobj = re.search(r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>', webpage)
- if mobj is None:
- # lookin for official user
- mobj_official = re.search(r'<span rel="author"[^>]+?>([^<]+?)</span>', webpage)
- if mobj_official is None:
- self._downloader.report_warning(u'unable to extract uploader nickname')
- else:
- video_uploader = mobj_official.group(1)
- else:
- video_uploader = mobj.group(1)
+ 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)
_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)
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')
if mobj is None:
# Try to find twitter cards info
mobj = re.search(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)
+ if mobj is None:
+ # We look for Open Graph info:
+ # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
+ m_video_type = re.search(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
+ # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
+ if m_video_type is not None:
+ mobj = re.search(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
|
((?:PL|EC|UU)[0-9A-Za-z-_]{10,})
)"""
- _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json'
+ _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
_MAX_RESULTS = 50
IE_NAME = u'youtube:playlist'
# Number of videos is a multiple of self._MAX_RESULTS
break
- videos += [ (entry['yt$position']['$t'], entry['content']['src'])
- for entry in response['feed']['entry']
- if 'content' in entry ]
+ for entry in response['feed']['entry']:
+ index = entry['yt$position']['$t']
+ if 'media$group' in entry and 'media$player' in entry['media$group']:
+ videos.append((index, entry['media$group']['media$player']['url']))
if len(response['feed']['entry']) < self._MAX_RESULTS:
break
webpage, 'title', default=u'NA')
# Step 2, Stimulate clicking the image box to launch video
- video_page = self._search_regex('href="(https\://plus\.google\.com/photos/.*?)"',
+ video_page = self._search_regex('"(https\://plus\.google\.com/photos/.*?)",,"image/jpeg","video"\]',
webpage, u'video page URL')
webpage = self._download_webpage(video_page, video_id, u'Downloading video page')
(?P<gameID>\d+)/?
(?P<videoID>\d*)(?P<extra>\??) #For urltype == video we sometimes get the videoID
"""
+ _VIDEO_PAGE_TEMPLATE = 'http://store.steampowered.com/video/%s/'
+ _AGECHECK_TEMPLATE = 'http://store.steampowered.com/agecheck/video/%s/?snr=1_agecheck_agecheck__age-gate&ageDay=1&ageMonth=January&ageYear=1970'
@classmethod
def suitable(cls, url):
def _real_extract(self, url):
m = re.match(self._VALID_URL, url, re.VERBOSE)
gameID = m.group('gameID')
- videourl = 'http://store.steampowered.com/agecheck/video/%s/?snr=1_agecheck_agecheck__age-gate&ageDay=1&ageMonth=January&ageYear=1970' % gameID
- self.report_age_confirmation()
+
+ videourl = self._VIDEO_PAGE_TEMPLATE % gameID
webpage = self._download_webpage(videourl, gameID)
- game_title = re.search(r'<h2 class="pageheader">(?P<game_title>.*?)</h2>', webpage).group('game_title')
-
+
+ if re.search('<h2>Please enter your birth date to continue:</h2>', webpage) is not None:
+ videourl = self._AGECHECK_TEMPLATE % gameID
+ self.report_age_confirmation()
+ webpage = self._download_webpage(videourl, gameID)
+
+ self.report_extraction(gameID)
+ game_title = self._html_search_regex(r'<h2 class="pageheader">(.*?)</h2>',
+ webpage, 'game title')
+
urlRE = r"'movie_(?P<videoID>\d+)': \{\s*FILENAME: \"(?P<videoURL>[\w:/\.\?=]+)\"(,\s*MOVIE_NAME: \"(?P<videoName>[\w:/\.\?=\+-]+)\")?\s*\},"
mweb = re.finditer(urlRE, webpage)
namesRE = r'<span class="title">(?P<videoName>.+?)</span>'
webpage = self._download_webpage(url, video_id)
- json_data = self._search_regex(r'<script>window.gon = {.*?};gon\.show=(.+?);</script>',
- webpage, u'json data')
+ json_data = self._search_regex(r'window\.gon.*?gon\.show=(.+?);$',
+ webpage, u'json data', flags=re.MULTILINE)
try:
data = json.loads(json_data)
self.to_screen(u'Getting info of playlist %s: "%s"' % (playlist_id,name))
return [self._playlist_videos_info(url,name,playlist_id)]
- def _talk_video_link(self,mediaSlug):
- '''Returns the video link for that mediaSlug'''
- return 'http://download.ted.com/talks/%s.mp4' % mediaSlug
-
def _playlist_videos_info(self,url,name,playlist_id=0):
'''Returns the videos of the playlist'''
video_RE=r'''
m_videos=re.finditer(video_RE,webpage,re.VERBOSE)
m_names=re.finditer(video_name_RE,webpage)
- playlist_RE = r'div class="headline">(\s*?)<h1>(\s*?)<span>(?P<playlist_title>.*?)</span>'
- m_playlist = re.search(playlist_RE, webpage)
- playlist_title = m_playlist.group('playlist_title')
+ playlist_title = self._html_search_regex(r'div class="headline">\s*?<h1>\s*?<span>(.*?)</span>',
+ webpage, 'playlist title')
playlist_entries = []
for m_video, m_name in zip(m_videos,m_names):
def _talk_info(self, url, video_id=0):
"""Return the video for the talk in the url"""
- m=re.match(self._VALID_URL, url,re.VERBOSE)
- videoName=m.group('name')
- webpage=self._download_webpage(url, video_id, 'Downloading \"%s\" page' % videoName)
+ m = re.match(self._VALID_URL, url,re.VERBOSE)
+ video_name = m.group('name')
+ webpage = self._download_webpage(url, video_id, 'Downloading \"%s\" page' % video_name)
+ self.report_extraction(video_name)
# If the url includes the language we get the title translated
- title_RE=r'<span id="altHeadline" >(?P<title>.*)</span>'
- title=re.search(title_RE, webpage).group('title')
- info_RE=r'''<script\ type="text/javascript">var\ talkDetails\ =(.*?)
- "id":(?P<videoID>[\d]+).*?
- "mediaSlug":"(?P<mediaSlug>[\w\d]+?)"'''
- thumb_RE=r'</span>[\s.]*</div>[\s.]*<img src="(?P<thumbnail>.*?)"'
- thumb_match=re.search(thumb_RE,webpage)
- info_match=re.search(info_RE,webpage,re.VERBOSE)
- video_id=info_match.group('videoID')
- mediaSlug=info_match.group('mediaSlug')
- video_url=self._talk_video_link(mediaSlug)
+ title = self._html_search_regex(r'<span id="altHeadline" >(?P<title>.*)</span>',
+ webpage, 'title')
+ json_data = self._search_regex(r'<script.*?>var talkDetails = ({.*?})</script>',
+ webpage, 'json data')
+ info = json.loads(json_data)
+ desc = self._html_search_regex(r'<div class="talk-intro">.*?<p.*?>(.*?)</p>',
+ webpage, 'description', flags = re.DOTALL)
+
+ thumbnail = self._search_regex(r'</span>[\s.]*</div>[\s.]*<img src="(.*?)"',
+ webpage, 'thumbnail')
info = {
- 'id': video_id,
- 'url': video_url,
+ 'id': info['id'],
+ 'url': info['htmlStreams'][-1]['file'],
'ext': 'mp4',
'title': title,
- 'thumbnail': thumb_match.group('thumbnail')
+ 'thumbnail': thumbnail,
+ 'description': desc,
}
return info
'description': video_description,
}
+class StatigramIE(InfoExtractor):
+ _VALID_URL = r'(?:http://)?(?:www\.)?statigr\.am/p/([^/]+)'
+
+ def _real_extract(self, url):
+ mobj = re.match(self._VALID_URL, url)
+
+ video_id = mobj.group(1)
+ webpage = self._download_webpage(url, video_id)
+ video_url = self._html_search_regex(
+ r'<meta property="og:video:secure_url" content="(.+?)">',
+ webpage, u'video URL')
+ thumbnail_url = self._html_search_regex(
+ r'<meta property="og:image" content="(.+?)" />',
+ webpage, u'thumbnail URL', fatal=False)
+ html_title = self._html_search_regex(
+ r'<title>(.+?)</title>',
+ webpage, u'title')
+ title = html_title.rpartition(u' | Statigram')[0]
+ uploader_id = self._html_search_regex(
+ r'@([^ ]+)', title, u'uploader name', fatal=False)
+ ext = 'mp4'
+
+ return [{
+ 'id': video_id,
+ 'url': video_url,
+ 'ext': ext,
+ 'title': title,
+ 'thumbnail': thumbnail_url,
+ 'uploader_id' : uploader_id
+ }]
+
def gen_extractors():
""" Return a list of an instance of every supported extractor.
The order does matter; the first extractor matched is the one handling the URL.
HypemIE(),
Vbox7IE(),
GametrailersIE(),
+ StatigramIE(),
GenericIE()
]