X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/6d835d4d6903cf5e18ad01844736929e06d16004..acb27260e91c30efdb413d0a6fa6279adcc98469:/youtube_dl/extractor/common.py
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 7e41132..e80a2da 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -12,13 +12,15 @@ import sys
import time
import xml.etree.ElementTree
-from ..utils import (
+from ..compat import (
+ compat_cookiejar,
compat_http_client,
compat_urllib_error,
compat_urllib_parse_urlparse,
compat_urlparse,
compat_str,
-
+)
+from ..utils import (
clean_html,
compiled_regex_type,
ExtractorError,
@@ -42,7 +44,11 @@ class InfoExtractor(object):
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 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.
@@ -86,6 +92,11 @@ class InfoExtractor(object):
by this field, regardless of all other values.
-1 for default (order by other properties),
-2 or smaller for less than default.
+ * 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),
@@ -145,6 +156,38 @@ class InfoExtractor(object):
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 or a PagedList object, each
+ element of which is a valid dictionary under this specfication.
+
+ 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.
+
+ Additionally, it may have properties believed to be identical to the
+ resolved entity, 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.
@@ -254,9 +297,11 @@ class InfoExtractor(object):
content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal)
return (content, urlh)
- def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True):
+ def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None):
content_type = urlh.headers.get('Content-Type', '')
webpage_bytes = urlh.read()
+ if prefix is not None:
+ webpage_bytes = prefix + 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)
@@ -381,17 +426,18 @@ class InfoExtractor(object):
"""Report attempt to log in."""
self.to_screen('Logging in')
- #Methods for following #608
+ # Methods for following #608
@staticmethod
def url_result(url, ie=None, video_id=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
return video_info
+
@staticmethod
def playlist_result(entries, playlist_id=None, playlist_title=None):
"""Returns a playlist"""
@@ -403,7 +449,7 @@ class InfoExtractor(object):
video_info['title'] = playlist_title
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.
@@ -424,22 +470,25 @@ class InfoExtractor(object):
_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('Unable to extract %s' % _name)
else:
self._downloader.report_warning('unable to extract %s; '
- 'please report this issue on http://yt-dl.org/bug' % _name)
+ 'please report this issue on http://yt-dl.org/bug' % _name)
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:
@@ -472,7 +521,7 @@ class InfoExtractor(object):
raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
except (IOError, netrc.NetrcParseError) as err:
self._downloader.report_warning('parsing .netrc: %s' % compat_str(err))
-
+
return (username, password)
def _get_tfa_info(self):
@@ -533,9 +582,9 @@ class InfoExtractor(object):
display_name = name
return self._html_search_regex(
r'''(?ix)]+(?:itemprop|name|property)=["\']?%s["\']?)
- [^>]+content=["\']([^"\']+)["\']''' % re.escape(name),
- html, display_name, fatal=fatal, **kwargs)
+ (?=[^>]+(?:itemprop|name|property)=(["\']?)%s\1)
+ [^>]+content=(["\'])(?P.*?)\1''' % 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')
@@ -566,7 +615,7 @@ class InfoExtractor(object):
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):
if not formats:
@@ -611,6 +660,7 @@ 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('height') if f.get('height') is not None else -1,
f.get('width') if f.get('width') is not None else -1,
@@ -768,6 +818,11 @@ class InfoExtractor(object):
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)
+
class SearchInfoExtractor(InfoExtractor):
"""