X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/a4f82408d9a3d921d9c2af7e4d757f44737dc7ea..d904ce9fca517ea1b6310e731acddcdd05dda618:/youtube_dl/extractor/common.py?ds=inline

diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 342bfb8..65bb770 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -1,4 +1,7 @@
+from __future__ import unicode_literals
+
 import base64
+import datetime
 import hashlib
 import json
 import netrc
@@ -9,15 +12,22 @@ import sys
 import time
 import xml.etree.ElementTree
 
-from ..utils import (
+from ..compat import (
+    compat_cookiejar,
+    compat_HTTPError,
     compat_http_client,
     compat_urllib_error,
     compat_urllib_parse_urlparse,
+    compat_urlparse,
     compat_str,
-
+)
+from ..utils import (
+    age_restricted,
+    bug_reports_message,
     clean_html,
     compiled_regex_type,
     ExtractorError,
+    float_or_none,
     int_or_none,
     RegexNotFoundError,
     sanitize_filename,
@@ -33,11 +43,15 @@ class InfoExtractor(object):
     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
+    passed to the YoutubeDL. The YoutubeDL processes this
     information possibly downloading the video to the file system, among
     other possible outcomes.
 
-    The dictionaries must include the following fields:
+    The type field determines the type of the result.
+    By far the most common value (and the default if _type is missing) is
+    "video", which indicates a single video.
+
+    For a video, the dictionaries must include the following fields:
 
     id:             Video identifier.
     title:          Video title, unescaped.
@@ -67,6 +81,7 @@ class InfoExtractor(object):
                     * acodec     Name of the audio codec in use
                     * asr        Audio sampling rate in Hertz
                     * vbr        Average video bitrate in KBit/s
+                    * fps        Frame rate
                     * vcodec     Name of the video codec in use
                     * container  Name of the container format
                     * filesize   The number of bytes, if known in advance
@@ -74,16 +89,36 @@ class InfoExtractor(object):
                     * player_url SWF Player URL (used for rtmpdump).
                     * protocol   The protocol that will be used for the actual
                                  download, lower-case.
-                                 "http", "https", "rtsp", "rtmp", "m3u8" or so.
+                                 "http", "https", "rtsp", "rtmp", "rtmpe",
+                                 "m3u8", or "m3u8_native".
                     * preference Order number of this format. If this field is
                                  present and not None, the formats get sorted
                                  by this field, regardless of all other values.
                                  -1 for default (order by other properties),
                                  -2 or smaller for less than default.
+                                 < -1000 to hide the format (if there is
+                                    another one which is strictly better)
+                    * language_preference  Is this in the correct requested
+                                 language?
+                                 10 if it's what the URL is about,
+                                 -1 for default (don't know),
+                                 -10 otherwise, other values reserved for now.
                     * quality    Order number of the video quality of this
                                  format, irrespective of the file format.
                                  -1 for default (order by other properties),
                                  -2 or smaller for less than default.
+                    * source_preference  Order number for this video source
+                                  (quality takes higher priority)
+                                 -1 for default (order by other properties),
+                                 -2 or smaller for less than default.
+                    * http_headers  A dictionary of additional HTTP headers
+                                 to add to the request.
+                    * stretched_ratio  If given and not 1, indicates that the
+                                 video's pixels are not square.
+                                 width : height ratio as float.
+                    * no_resume  The server does not support resuming the
+                                 (HTTP or RTMP) download. Boolean.
+
     url:            Final video URL.
     ext:            Video filename extension.
     format:         The video format, defaults to ext (used for --get-format)
@@ -91,40 +126,99 @@ class InfoExtractor(object):
 
     The following fields are optional:
 
+    alt_title:      A secondary title of the video.
     display_id      An alternative identifier for the video, not necessarily
                     unique, but available before title. Typically, id is
                     something like "4234987", title "Dancing naked mole rats",
                     and display_id "dancing-naked-mole-rats"
     thumbnails:     A list of dictionaries, with the following entries:
+                        * "id" (optional, string) - Thumbnail format ID
                         * "url"
+                        * "preference" (optional, int) - quality of the image
                         * "width" (optional, int)
                         * "height" (optional, int)
                         * "resolution" (optional, string "{width}x{height"},
                                         deprecated)
     thumbnail:      Full URL to a video thumbnail image.
-    description:    One-line video description.
+    description:    Full video description.
     uploader:       Full name of the video uploader.
+    creator:        The main artist who created the video.
     timestamp:      UNIX timestamp of the moment the video became available.
     upload_date:    Video upload date (YYYYMMDD).
                     If not explicitly set, calculated from timestamp.
     uploader_id:    Nickname or id of the video uploader.
-    location:       Physical location of the video.
-    subtitles:      The subtitle file contents as a dictionary in the format
-                    {language: subtitles}.
+    location:       Physical location where the video was filmed.
+    subtitles:      The available subtitles as a dictionary in the format
+                    {language: subformats}. "subformats" is a list sorted from
+                    lower to higher preference, each element is a dictionary
+                    with the "ext" entry and one of:
+                        * "data": The subtitles file contents
+                        * "url": A url pointing to the subtitles file
+    automatic_captions: Like 'subtitles', used by the YoutubeIE for
+                    automatically generated captions
     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
+    average_rating: Average rating give by users, the scale used depends on the webpage
     comment_count:  Number of comments on the video
+    comments:       A list of comments, each with one or more of the following
+                    properties (all but one of text or html optional):
+                        * "author" - human-readable name of the comment author
+                        * "author_id" - user ID of the comment author
+                        * "id" - Comment ID
+                        * "html" - Comment as HTML
+                        * "text" - Plain text of the comment
+                        * "timestamp" - UNIX timestamp of comment
+                        * "parent" - ID of the comment this one is replying to.
+                                     Set to "root" to indicate that this is a
+                                     comment to the original video.
     age_limit:      Age restriction for the video, as an integer (years)
     webpage_url:    The url to the video webpage, if given to youtube-dl it
                     should allow to get the same result again. (It will be set
                     by YoutubeDL if it's missing)
     categories:     A list of categories that the video falls in, for example
                     ["Sports", "Berlin"]
+    is_live:        True, False, or None (=unknown). Whether this video is a
+                    live stream that goes on instead of a fixed-length video.
 
     Unless mentioned otherwise, the fields should be Unicode strings.
 
+    Unless mentioned otherwise, None is equivalent to absence of information.
+
+
+    _type "playlist" indicates multiple videos.
+    There must be a key "entries", which is a list, an iterable, or a PagedList
+    object, each element of which is a valid dictionary by this specification.
+
+    Additionally, playlists can have "title" and "id" attributes with the same
+    semantics as videos (see above).
+
+
+    _type "multi_video" indicates that there are multiple videos that
+    form a single show, for examples multiple acts of an opera or TV episode.
+    It must have an entries key like a playlist and contain all the keys
+    required for a video at the same time.
+
+
+    _type "url" indicates that the video must be extracted from another
+    location, possibly by a different extractor. Its only required key is:
+    "url" - the next URL to extract.
+    The key "ie_key" can be set to the class name (minus the trailing "IE",
+    e.g. "Youtube") if the extractor class is known in advance.
+    Additionally, the dictionary may have any properties of the resolved entity
+    known in advance, for example "title" if the title of the referred video is
+    known ahead of time.
+
+
+    _type "url_transparent" entities have the same specification as "url", but
+    indicate that the given additional information is more precise than the one
+    associated with the resolved URL.
+    This is useful when a site employs a video service that hosts the video and
+    its technical metadata, but that video service does not embed a useful
+    title, description etc.
+
+
     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.
@@ -153,6 +247,14 @@ class InfoExtractor(object):
             cls._VALID_URL_RE = re.compile(cls._VALID_URL)
         return cls._VALID_URL_RE.match(url) is not None
 
+    @classmethod
+    def _match_id(cls, url):
+        if '_VALID_URL_RE' not in cls.__dict__:
+            cls._VALID_URL_RE = re.compile(cls._VALID_URL)
+        m = cls._VALID_URL_RE.match(url)
+        assert m
+        return m.group('id')
+
     @classmethod
     def working(cls):
         """Getter method for _WORKING."""
@@ -166,8 +268,15 @@ class InfoExtractor(object):
 
     def extract(self, url):
         """Extracts URL information and returns it in list of dicts."""
-        self.initialize()
-        return self._real_extract(url)
+        try:
+            self.initialize()
+            return self._real_extract(url)
+        except ExtractorError:
+            raise
+        except compat_http_client.IncompleteRead as e:
+            raise ExtractorError('A network error has occured.', cause=e, expected=True)
+        except (KeyError, StopIteration) as e:
+            raise ExtractorError('An extractor error has occured.', cause=e)
 
     def set_downloader(self, downloader):
         """Sets the downloader for this IE."""
@@ -196,26 +305,25 @@ class InfoExtractor(object):
             self.report_download_webpage(video_id)
         elif note is not False:
             if video_id is None:
-                self.to_screen(u'%s' % (note,))
+                self.to_screen('%s' % (note,))
             else:
-                self.to_screen(u'%s: %s' % (video_id, note))
+                self.to_screen('%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 False:
                 return False
             if errnote is None:
-                errnote = u'Unable to download webpage'
-            errmsg = u'%s: %s' % (errnote, compat_str(err))
+                errnote = 'Unable to download webpage'
+            errmsg = '%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, fatal=True):
+    def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True, encoding=None):
         """ 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]
@@ -224,8 +332,11 @@ class InfoExtractor(object):
         if urlh is False:
             assert not fatal
             return False
-        content_type = urlh.headers.get('Content-Type', '')
-        webpage_bytes = urlh.read()
+        content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal, encoding=encoding)
+        return (content, urlh)
+
+    @staticmethod
+    def _guess_encoding_from_content(content_type, webpage_bytes):
         m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
         if m:
             encoding = m.group(1)
@@ -238,12 +349,22 @@ class InfoExtractor(object):
                 encoding = 'utf-16'
             else:
                 encoding = 'utf-8'
+
+        return encoding
+
+    def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None, encoding=None):
+        content_type = urlh.headers.get('Content-Type', '')
+        webpage_bytes = urlh.read()
+        if prefix is not None:
+            webpage_bytes = prefix + webpage_bytes
+        if not encoding:
+            encoding = self._guess_encoding_from_content(content_type, webpage_bytes)
         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)
+            self.to_screen('Dumping request to ' + url)
             dump = base64.b64encode(webpage_bytes).decode('ascii')
             self._downloader.to_screen(dump)
         if self._downloader.params.get('write_pages', False):
@@ -253,11 +374,17 @@ class InfoExtractor(object):
                 url = url_or_request
             basen = '%s_%s' % (video_id, url)
             if len(basen) > 240:
-                h = u'___' + hashlib.md5(basen.encode('utf-8')).hexdigest()
+                h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest()
                 basen = basen[:240 - len(h)] + h
             raw_filename = basen + '.dump'
             filename = sanitize_filename(raw_filename, restricted=True)
-            self.to_screen(u'Saving request to ' + filename)
+            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)
 
@@ -266,21 +393,41 @@ class InfoExtractor(object):
         except LookupError:
             content = webpage_bytes.decode('utf-8', 'replace')
 
-        if (u'<title>Access to this site is blocked</title>' in content and
-                u'Websense' in content[:512]):
-            msg = u'Access to this webpage has been blocked by Websense filtering software in your network.'
+        if ('<title>Access to this site is blocked</title>' in content and
+                'Websense' in content[:512]):
+            msg = 'Access to this webpage has been blocked by Websense filtering software in your network.'
             blocked_iframe = self._html_search_regex(
                 r'<iframe src="([^"]+)"', content,
-                u'Websense information URL', default=None)
+                'Websense information URL', default=None)
             if blocked_iframe:
-                msg += u' Visit %s for more details' % blocked_iframe
+                msg += ' Visit %s for more details' % blocked_iframe
+            raise ExtractorError(msg, expected=True)
+        if '<title>The URL you requested has been blocked</title>' in content[:512]:
+            msg = (
+                'Access to this webpage has been blocked by Indian censorship. '
+                'Use a VPN or proxy server (with --proxy) to route around it.')
+            block_msg = self._html_search_regex(
+                r'</h1><p>(.*?)</p>',
+                content, 'block message', default=None)
+            if block_msg:
+                msg += ' (Message: "%s")' % block_msg.replace('\n', ' ')
             raise ExtractorError(msg, expected=True)
 
-        return (content, urlh)
+        return content
 
-    def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
+    def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5, encoding=None):
         """ Returns the data of the page as a string """
-        res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal)
+        success = False
+        try_count = 0
+        while success is False:
+            try:
+                res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal, encoding=encoding)
+                success = True
+            except compat_http_client.IncompleteRead as e:
+                try_count += 1
+                if try_count >= tries:
+                    raise e
+                self._sleep(timeout, video_id)
         if res is False:
             return res
         else:
@@ -288,11 +435,11 @@ class InfoExtractor(object):
             return content
 
     def _download_xml(self, url_or_request, video_id,
-                      note=u'Downloading XML', errnote=u'Unable to download XML',
-                      transform_source=None, fatal=True):
+                      note='Downloading XML', errnote='Unable to download XML',
+                      transform_source=None, fatal=True, encoding=None):
         """Return the xml as an xml.etree.ElementTree.Element"""
         xml_string = self._download_webpage(
-            url_or_request, video_id, note, errnote, fatal=fatal)
+            url_or_request, video_id, note, errnote, fatal=fatal, encoding=encoding)
         if xml_string is False:
             return xml_string
         if transform_source:
@@ -300,59 +447,71 @@ class InfoExtractor(object):
         return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8'))
 
     def _download_json(self, url_or_request, video_id,
-                       note=u'Downloading JSON metadata',
-                       errnote=u'Unable to download JSON metadata',
+                       note='Downloading JSON metadata',
+                       errnote='Unable to download JSON metadata',
                        transform_source=None,
-                       fatal=True):
+                       fatal=True, encoding=None):
         json_string = self._download_webpage(
-            url_or_request, video_id, note, errnote, fatal=fatal)
+            url_or_request, video_id, note, errnote, fatal=fatal,
+            encoding=encoding)
         if (not fatal) and json_string is False:
             return None
+        return self._parse_json(
+            json_string, video_id, transform_source=transform_source, fatal=fatal)
+
+    def _parse_json(self, json_string, video_id, transform_source=None, fatal=True):
         if transform_source:
             json_string = transform_source(json_string)
         try:
             return json.loads(json_string)
         except ValueError as ve:
-            raise ExtractorError('Failed to download JSON', cause=ve)
+            errmsg = '%s: Failed to parse JSON ' % video_id
+            if fatal:
+                raise ExtractorError(errmsg, cause=ve)
+            else:
+                self.report_warning(errmsg + str(ve))
 
     def report_warning(self, msg, video_id=None):
-        idstr = u'' if video_id is None else u'%s: ' % video_id
+        idstr = '' if video_id is None else '%s: ' % video_id
         self._downloader.report_warning(
-            u'[%s] %s%s' % (self.IE_NAME, idstr, msg))
+            '[%s] %s%s' % (self.IE_NAME, idstr, msg))
 
     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))
+        self._downloader.to_screen('[%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)
+        self.to_screen('%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)
+        self.to_screen('%s: Downloading webpage' % video_id)
 
     def report_age_confirmation(self):
         """Report attempt to confirm age."""
-        self.to_screen(u'Confirming age')
+        self.to_screen('Confirming age')
 
     def report_login(self):
         """Report attempt to log in."""
-        self.to_screen(u'Logging in')
+        self.to_screen('Logging in')
 
-    #Methods for following #608
+    # Methods for following #608
     @staticmethod
-    def url_result(url, ie=None, video_id=None):
+    def url_result(url, ie=None, video_id=None, video_title=None):
         """Returns a url that points to a page that should be processed"""
-        #TODO: ie should be the class used for getting the info
+        # TODO: ie should be the class used for getting the info
         video_info = {'_type': 'url',
                       'url': url,
                       'ie_key': ie}
         if video_id is not None:
             video_info['id'] = video_id
+        if video_title is not None:
+            video_info['title'] = video_title
         return video_info
+
     @staticmethod
-    def playlist_result(entries, playlist_id=None, playlist_title=None):
+    def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None):
         """Returns a playlist"""
         video_info = {'_type': 'playlist',
                       'entries': entries}
@@ -360,9 +519,11 @@ class InfoExtractor(object):
             video_info['id'] = playlist_id
         if playlist_title:
             video_info['title'] = playlist_title
+        if playlist_description:
+            video_info['description'] = playlist_description
         return video_info
 
-    def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0):
+    def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None):
         """
         Perform a regex search on the given string, using a single or a list of
         patterns returning the first matching group.
@@ -377,28 +538,30 @@ class InfoExtractor(object):
                 if mobj:
                     break
 
-        if os.name != 'nt' and sys.stderr.isatty():
-            _name = u'\033[0;34m%s\033[0m' % name
+        if not self._downloader.params.get('no_color') and os.name != 'nt' and sys.stderr.isatty():
+            _name = '\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)
+            if group is None:
+                # return the first matching group
+                return next(g for g in mobj.groups() if g is not None)
+            else:
+                return mobj.group(group)
         elif default is not _NO_DEFAULT:
             return default
         elif fatal:
-            raise RegexNotFoundError(u'Unable to extract %s' % _name)
+            raise RegexNotFoundError('Unable to extract %s' % _name)
         else:
-            self._downloader.report_warning(u'unable to extract %s; '
-                u'please report this issue on http://yt-dl.org/bug' % _name)
+            self._downloader.report_warning('unable to extract %s' % _name + bug_reports_message())
             return None
 
-    def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0):
+    def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None):
         """
         Like _search_regex, but strips HTML tags and unescapes entities.
         """
-        res = self._search_regex(pattern, string, name, default, fatal, flags)
+        res = self._search_regex(pattern, string, name, default, fatal, flags, group)
         if res:
             return clean_html(res).strip()
         else:
@@ -406,7 +569,7 @@ class InfoExtractor(object):
 
     def _get_login_info(self):
         """
-        Get the the login info as (username, password)
+        Get the login info as (username, password)
         It will look in the netrc file using the _NETRC_MACHINE value
         If there's no info available, return (None, None)
         """
@@ -430,10 +593,26 @@ class InfoExtractor(object):
                 else:
                     raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
             except (IOError, netrc.NetrcParseError) as err:
-                self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err))
-        
+                self._downloader.report_warning('parsing .netrc: %s' % compat_str(err))
+
         return (username, password)
 
+    def _get_tfa_info(self):
+        """
+        Get the two-factor authentication info
+        TODO - asking the user will be required for sms/phone verify
+        currently just uses the command line option
+        If there's no info available, return None
+        """
+        if self._downloader is None:
+            return None
+        downloader_params = self._downloader.params
+
+        if downloader_params.get('twofactor', None) is not None:
+            return downloader_params['twofactor']
+
+        return None
+
     # Helper functions for extracting OpenGraph info
     @staticmethod
     def _og_regexes(prop):
@@ -454,7 +633,7 @@ class InfoExtractor(object):
         return unescapeHTML(escaped)
 
     def _og_search_thumbnail(self, html, **kargs):
-        return self._og_search_property('image', html, u'thumbnail url', fatal=False, **kargs)
+        return self._og_search_property('image', html, 'thumbnail url', fatal=False, **kargs)
 
     def _og_search_description(self, html, **kargs):
         return self._og_search_property('description', html, fatal=False, **kargs)
@@ -463,8 +642,9 @@ class InfoExtractor(object):
         return self._og_search_property('title', html, **kargs)
 
     def _og_search_video_url(self, html, name='video url', secure=True, **kargs):
-        regexes = self._og_regexes('video')
-        if secure: regexes = self._og_regexes('video:secure_url') + regexes
+        regexes = self._og_regexes('video') + self._og_regexes('video:url')
+        if secure:
+            regexes = self._og_regexes('video:secure_url') + regexes
         return self._html_search_regex(regexes, html, name, **kargs)
 
     def _og_search_url(self, html, **kargs):
@@ -474,10 +654,10 @@ class InfoExtractor(object):
         if display_name is None:
             display_name = name
         return self._html_search_regex(
-            r'''(?ix)<meta
-                    (?=[^>]+(?:itemprop|name|property)=["\']?%s["\']?)
-                    [^>]+content=["\']([^"\']+)["\']''' % re.escape(name),
-            html, display_name, fatal=fatal, **kwargs)
+            r'''(?isx)<meta
+                    (?=[^>]+(?:itemprop|name|property)=(["\']?)%s\1)
+                    [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(name),
+            html, display_name, fatal=fatal, group='content', **kwargs)
 
     def _dc_search_uploader(self, html):
         return self._html_search_meta('dc.creator', html, 'uploader')
@@ -506,13 +686,28 @@ class InfoExtractor(object):
         }
         return RATING_TABLE.get(rating.lower(), None)
 
+    def _family_friendly_search(self, html):
+        # See http://schema.org/VideoObject
+        family_friendly = self._html_search_meta('isFamilyFriendly', html)
+
+        if not family_friendly:
+            return None
+
+        RATING_TABLE = {
+            '1': 0,
+            'true': 0,
+            '0': 18,
+            'false': 18,
+        }
+        return RATING_TABLE.get(family_friendly.lower(), None)
+
     def _twitter_search_player(self, html):
         return self._html_search_meta('twitter:player', html,
-            'twitter card player')
+                                      'twitter card player')
 
-    def _sort_formats(self, formats):
+    def _sort_formats(self, formats, field_preference=None):
         if not formats:
-            raise ExtractorError(u'No video formats found')
+            raise ExtractorError('No video formats found')
 
         def _formats_key(f):
             # TODO remove the following workaround
@@ -520,6 +715,9 @@ class InfoExtractor(object):
             if not f.get('ext') and 'url' in f:
                 f['ext'] = determine_ext(f['url'])
 
+            if isinstance(field_preference, (list, tuple)):
+                return tuple(f.get(field) if f.get(field) is not None else -1 for field in field_preference)
+
             preference = f.get('preference')
             if preference is None:
                 proto = f.get('protocol')
@@ -532,9 +730,9 @@ class InfoExtractor(object):
 
             if f.get('vcodec') == 'none':  # audio only
                 if self._downloader.params.get('prefer_free_formats'):
-                    ORDER = [u'aac', u'mp3', u'm4a', u'webm', u'ogg', u'opus']
+                    ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus']
                 else:
-                    ORDER = [u'webm', u'opus', u'ogg', u'mp3', u'aac', u'm4a']
+                    ORDER = ['webm', 'opus', 'ogg', 'mp3', 'aac', 'm4a']
                 ext_preference = 0
                 try:
                     audio_ext_preference = ORDER.index(f['ext'])
@@ -542,9 +740,9 @@ class InfoExtractor(object):
                     audio_ext_preference = -1
             else:
                 if self._downloader.params.get('prefer_free_formats'):
-                    ORDER = [u'flv', u'mp4', u'webm']
+                    ORDER = ['flv', 'mp4', 'webm']
                 else:
-                    ORDER = [u'webm', u'flv', u'mp4']
+                    ORDER = ['webm', 'flv', 'mp4']
                 try:
                     ext_preference = ORDER.index(f['ext'])
                 except ValueError:
@@ -553,22 +751,48 @@ class InfoExtractor(object):
 
             return (
                 preference,
+                f.get('language_preference') if f.get('language_preference') is not None else -1,
                 f.get('quality') if f.get('quality') is not None else -1,
+                f.get('tbr') if f.get('tbr') is not None else -1,
+                f.get('filesize') if f.get('filesize') is not None else -1,
+                f.get('vbr') if f.get('vbr') is not None else -1,
                 f.get('height') if f.get('height') is not None else -1,
                 f.get('width') if f.get('width') is not None else -1,
                 ext_preference,
-                f.get('tbr') if f.get('tbr') is not None else -1,
-                f.get('vbr') if f.get('vbr') is not None else -1,
                 f.get('abr') if f.get('abr') is not None else -1,
                 audio_ext_preference,
-                f.get('filesize') if f.get('filesize') is not None else -1,
+                f.get('fps') if f.get('fps') is not None else -1,
                 f.get('filesize_approx') if f.get('filesize_approx') is not None else -1,
-                f.get('format_id'),
+                f.get('source_preference') if f.get('source_preference') is not None else -1,
+                f.get('format_id') if f.get('format_id') is not None else '',
             )
         formats.sort(key=_formats_key)
 
+    def _check_formats(self, formats, video_id):
+        if formats:
+            formats[:] = filter(
+                lambda f: self._is_valid_url(
+                    f['url'], video_id,
+                    item='%s video format' % f.get('format_id') if f.get('format_id') else 'video'),
+                formats)
+
+    def _is_valid_url(self, url, video_id, item='video'):
+        url = self._proto_relative_url(url, scheme='http:')
+        # For now assume non HTTP(S) URLs always valid
+        if not (url.startswith('http://') or url.startswith('https://')):
+            return True
+        try:
+            self._request_webpage(url, video_id, 'Checking %s URL' % item)
+            return True
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError):
+                self.report_warning(
+                    '%s URL is invalid, skipping' % item, video_id)
+                return False
+            raise
+
     def http_scheme(self):
-        """ Either "https:" or "https:", depending on the user's preferences """
+        """ Either "http:" or "https:", depending on the user's preferences """
         return (
             'http:'
             if self._downloader.params.get('prefer_insecure', False)
@@ -586,29 +810,268 @@ class InfoExtractor(object):
 
     def _sleep(self, timeout, video_id, msg_template=None):
         if msg_template is None:
-            msg_template = u'%(video_id)s: Waiting for %(timeout)s seconds'
+            msg_template = '%(video_id)s: Waiting for %(timeout)s seconds'
         msg = msg_template % {'video_id': video_id, 'timeout': timeout}
         self.to_screen(msg)
         time.sleep(timeout)
 
-    def _extract_f4m_formats(self, manifest_url, video_id):
+    def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None):
         manifest = self._download_xml(
             manifest_url, video_id, 'Downloading f4m manifest',
             'Unable to download f4m manifest')
 
         formats = []
-        for media_el in manifest.findall('{http://ns.adobe.com/f4m/1.0}media'):
+        manifest_version = '1.0'
+        media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media')
+        if not media_nodes:
+            manifest_version = '2.0'
+            media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media')
+        for i, media_el in enumerate(media_nodes):
+            if manifest_version == '2.0':
+                manifest_url = ('/'.join(manifest_url.split('/')[:-1]) + '/' +
+                                (media_el.attrib.get('href') or media_el.attrib.get('url')))
+            tbr = int_or_none(media_el.attrib.get('bitrate'))
             formats.append({
+                'format_id': '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)])),
                 'url': manifest_url,
                 'ext': 'flv',
-                'tbr': int_or_none(media_el.attrib.get('bitrate')),
+                'tbr': tbr,
                 'width': int_or_none(media_el.attrib.get('width')),
                 'height': int_or_none(media_el.attrib.get('height')),
+                'preference': preference,
             })
         self._sort_formats(formats)
 
         return formats
 
+    def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None,
+                              entry_protocol='m3u8', preference=None,
+                              m3u8_id=None):
+
+        formats = [{
+            'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])),
+            'url': m3u8_url,
+            'ext': ext,
+            'protocol': 'm3u8',
+            'preference': preference - 1 if preference else -1,
+            'resolution': 'multiple',
+            'format_note': 'Quality selection URL',
+        }]
+
+        format_url = lambda u: (
+            u
+            if re.match(r'^https?://', u)
+            else compat_urlparse.urljoin(m3u8_url, u))
+
+        m3u8_doc = self._download_webpage(
+            m3u8_url, video_id,
+            note='Downloading m3u8 information',
+            errnote='Failed to download m3u8 information')
+        last_info = None
+        last_media = None
+        kv_rex = re.compile(
+            r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)')
+        for line in m3u8_doc.splitlines():
+            if line.startswith('#EXT-X-STREAM-INF:'):
+                last_info = {}
+                for m in kv_rex.finditer(line):
+                    v = m.group('val')
+                    if v.startswith('"'):
+                        v = v[1:-1]
+                    last_info[m.group('key')] = v
+            elif line.startswith('#EXT-X-MEDIA:'):
+                last_media = {}
+                for m in kv_rex.finditer(line):
+                    v = m.group('val')
+                    if v.startswith('"'):
+                        v = v[1:-1]
+                    last_media[m.group('key')] = v
+            elif line.startswith('#') or not line.strip():
+                continue
+            else:
+                if last_info is None:
+                    formats.append({'url': format_url(line)})
+                    continue
+                tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000)
+                format_id = []
+                if m3u8_id:
+                    format_id.append(m3u8_id)
+                last_media_name = last_media.get('NAME') if last_media and last_media.get('TYPE') != 'SUBTITLES' else None
+                format_id.append(last_media_name if last_media_name else '%d' % (tbr if tbr else len(formats)))
+                f = {
+                    'format_id': '-'.join(format_id),
+                    'url': format_url(line.strip()),
+                    'tbr': tbr,
+                    'ext': ext,
+                    'protocol': entry_protocol,
+                    'preference': preference,
+                }
+                codecs = last_info.get('CODECS')
+                if codecs:
+                    # TODO: looks like video codec is not always necessarily goes first
+                    va_codecs = codecs.split(',')
+                    if va_codecs[0]:
+                        f['vcodec'] = va_codecs[0].partition('.')[0]
+                    if len(va_codecs) > 1 and va_codecs[1]:
+                        f['acodec'] = va_codecs[1].partition('.')[0]
+                resolution = last_info.get('RESOLUTION')
+                if resolution:
+                    width_str, height_str = resolution.split('x')
+                    f['width'] = int(width_str)
+                    f['height'] = int(height_str)
+                if last_media is not None:
+                    f['m3u8_media'] = last_media
+                    last_media = None
+                formats.append(f)
+                last_info = {}
+        self._sort_formats(formats)
+        return formats
+
+    # TODO: improve extraction
+    def _extract_smil_formats(self, smil_url, video_id, fatal=True):
+        smil = self._download_xml(
+            smil_url, video_id, 'Downloading SMIL file',
+            'Unable to download SMIL file', fatal=fatal)
+        if smil is False:
+            assert not fatal
+            return []
+
+        base = smil.find('./head/meta').get('base')
+
+        formats = []
+        rtmp_count = 0
+        if smil.findall('./body/seq/video'):
+            video = smil.findall('./body/seq/video')[0]
+            fmts, rtmp_count = self._parse_smil_video(video, video_id, base, rtmp_count)
+            formats.extend(fmts)
+        else:
+            for video in smil.findall('./body/switch/video'):
+                fmts, rtmp_count = self._parse_smil_video(video, video_id, base, rtmp_count)
+                formats.extend(fmts)
+
+        self._sort_formats(formats)
+
+        return formats
+
+    def _parse_smil_video(self, video, video_id, base, rtmp_count):
+        src = video.get('src')
+        if not src:
+            return ([], rtmp_count)
+        bitrate = int_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
+        width = int_or_none(video.get('width'))
+        height = int_or_none(video.get('height'))
+        proto = video.get('proto')
+        if not proto:
+            if base:
+                if base.startswith('rtmp'):
+                    proto = 'rtmp'
+                elif base.startswith('http'):
+                    proto = 'http'
+        ext = video.get('ext')
+        if proto == 'm3u8':
+            return (self._extract_m3u8_formats(src, video_id, ext), rtmp_count)
+        elif proto == 'rtmp':
+            rtmp_count += 1
+            streamer = video.get('streamer') or base
+            return ([{
+                'url': streamer,
+                'play_path': src,
+                'ext': 'flv',
+                'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate),
+                'tbr': bitrate,
+                'width': width,
+                'height': height,
+            }], rtmp_count)
+        elif proto.startswith('http'):
+            return ([{
+                'url': base + src,
+                'ext': ext or 'flv',
+                'tbr': bitrate,
+                'width': width,
+                'height': height,
+            }], rtmp_count)
+
+    def _live_title(self, name):
+        """ Generate the title for a live video """
+        now = datetime.datetime.now()
+        now_str = now.strftime("%Y-%m-%d %H:%M")
+        return name + ' ' + now_str
+
+    def _int(self, v, name, fatal=False, **kwargs):
+        res = int_or_none(v, **kwargs)
+        if 'get_attr' in kwargs:
+            print(getattr(v, kwargs['get_attr']))
+        if res is None:
+            msg = 'Failed to extract %s: Could not parse value %r' % (name, v)
+            if fatal:
+                raise ExtractorError(msg)
+            else:
+                self._downloader.report_warning(msg)
+        return res
+
+    def _float(self, v, name, fatal=False, **kwargs):
+        res = float_or_none(v, **kwargs)
+        if res is None:
+            msg = 'Failed to extract %s: Could not parse value %r' % (name, v)
+            if fatal:
+                raise ExtractorError(msg)
+            else:
+                self._downloader.report_warning(msg)
+        return res
+
+    def _set_cookie(self, domain, name, value, expire_time=None):
+        cookie = compat_cookiejar.Cookie(
+            0, name, value, None, None, domain, None,
+            None, '/', True, False, expire_time, '', None, None, None)
+        self._downloader.cookiejar.set_cookie(cookie)
+
+    def get_testcases(self, include_onlymatching=False):
+        t = getattr(self, '_TEST', None)
+        if t:
+            assert not hasattr(self, '_TESTS'), \
+                '%s has _TEST and _TESTS' % type(self).__name__
+            tests = [t]
+        else:
+            tests = getattr(self, '_TESTS', [])
+        for t in tests:
+            if not include_onlymatching and t.get('only_matching', False):
+                continue
+            t['name'] = type(self).__name__[:-len('IE')]
+            yield t
+
+    def is_suitable(self, age_limit):
+        """ Test whether the extractor is generally suitable for the given
+        age limit (i.e. pornographic sites are not, all others usually are) """
+
+        any_restricted = False
+        for tc in self.get_testcases(include_onlymatching=False):
+            if 'playlist' in tc:
+                tc = tc['playlist'][0]
+            is_restricted = age_restricted(
+                tc.get('info_dict', {}).get('age_limit'), age_limit)
+            if not is_restricted:
+                return True
+            any_restricted = any_restricted or is_restricted
+        return not any_restricted
+
+    def extract_subtitles(self, *args, **kwargs):
+        if (self._downloader.params.get('writesubtitles', False) or
+                self._downloader.params.get('listsubtitles')):
+            return self._get_subtitles(*args, **kwargs)
+        return {}
+
+    def _get_subtitles(self, *args, **kwargs):
+        raise NotImplementedError("This method must be implemented by subclasses")
+
+    def extract_automatic_captions(self, *args, **kwargs):
+        if (self._downloader.params.get('writeautomaticsub', False) or
+                self._downloader.params.get('listsubtitles')):
+            return self._get_automatic_captions(*args, **kwargs)
+        return {}
+
+    def _get_automatic_captions(self, *args, **kwargs):
+        raise NotImplementedError("This method must be implemented by subclasses")
+
 
 class SearchInfoExtractor(InfoExtractor):
     """
@@ -628,7 +1091,7 @@ class SearchInfoExtractor(InfoExtractor):
     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)
+            raise ExtractorError('Invalid search query "%s"' % query)
 
         prefix = mobj.group('prefix')
         query = mobj.group('query')
@@ -639,9 +1102,9 @@ class SearchInfoExtractor(InfoExtractor):
         else:
             n = int(prefix)
             if n <= 0:
-                raise ExtractorError(u'invalid download number %s for query "%s"' % (n, query))
+                raise ExtractorError('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))
+                self._downloader.report_warning('%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)