age_restricted,
base_url,
bug_reports_message,
+ classproperty,
clean_html,
determine_ext,
determine_protocol,
For a video, the dictionaries must include the following fields:
id: Video identifier.
- title: Video title, unescaped.
+ title: Video title, unescaped. Set to an empty string if video has
+ no title as opposed to "None" which signifies that the
+ extractor failed to obtain a title
Additionally, it must contain either a formats entry or a url one:
* no_resume The server does not support resuming the
(HTTP or RTMP) download. Boolean.
* has_drm The format has DRM and cannot be downloaded. Boolean
- * downloader_options A dictionary of downloader options as
- described in FileDownloader (For internal use only)
+ * downloader_options A dictionary of downloader options
+ (For internal use only)
+ * http_chunk_size Chunk size for HTTP downloads
+ * ffmpeg_args Extra arguments for ffmpeg downloader
RTMP formats can also have the additional fields: page_url,
app, play_path, tc_url, flash_version, rtmp_live, rtmp_conn,
rtmp_protocol, rtmp_real_time
_WORKING = True
_NETRC_MACHINE = None
IE_DESC = None
+ SEARCH_KEY = None
- _LOGIN_HINTS = {
- 'any': 'Use --cookies, --cookies-from-browser, --username and --password, or --netrc to provide account credentials',
- 'cookies': (
- 'Use --cookies-from-browser or --cookies for the authentication. '
- 'See https://github.com/ytdl-org/youtube-dl#how-do-i-pass-cookies-to-youtube-dl for how to manually pass cookies'),
- 'password': 'Use --username and --password, or --netrc to provide account credentials',
- }
+ def _login_hint(self, method=NO_DEFAULT, netrc=None):
+ password_hint = f'--username and --password, or --netrc ({netrc or self._NETRC_MACHINE}) to provide account credentials'
+ return {
+ None: '',
+ 'any': f'Use --cookies, --cookies-from-browser, {password_hint}',
+ 'password': f'Use {password_hint}',
+ 'cookies': (
+ 'Use --cookies-from-browser or --cookies for the authentication. '
+ 'See https://github.com/ytdl-org/youtube-dl#how-do-i-pass-cookies-to-youtube-dl for how to manually pass cookies'),
+ }[method if method is not NO_DEFAULT else 'any' if self.supports_login() else 'cookies']
def __init__(self, downloader=None):
"""Constructor. Receives an optional downloader (a YoutubeDL instance).
if username:
self._perform_login(username, password)
elif self.get_param('username') and False not in (self.IE_DESC, self._NETRC_MACHINE):
- self.report_warning(f'Login with password is not supported for this website. {self._LOGIN_HINTS["cookies"]}')
+ self.report_warning(f'Login with password is not supported for this website. {self._login_hint("cookies")}')
self._real_initialize()
self._ready = True
"""A string for getting the InfoExtractor with get_info_extractor"""
return cls.__name__[:-2]
- @property
- def IE_NAME(self):
- return compat_str(type(self).__name__[:-2])
+ @classproperty
+ def IE_NAME(cls):
+ return cls.__name__[:-2]
@staticmethod
def __can_accept_status_code(err, expected_status):
self.get_param('ignore_no_formats_error') or self.get_param('wait_for_video')):
self.report_warning(msg)
return
- if method is NO_DEFAULT:
- method = 'any' if self.supports_login() else 'cookies'
- if method is not None:
- assert method in self._LOGIN_HINTS, 'Invalid login method'
- msg = f'{msg}. {self._LOGIN_HINTS[method]}'
+ msg += format_field(self._login_hint(method), template='. %s')
raise ExtractorError(msg, expected=True)
def raise_geo_restricted(
def _sort_formats(self, formats, field_preference=[]):
if not formats:
return
- format_sort = self.FormatSort(self, field_preference)
- formats.sort(key=lambda f: format_sort.calculate_preference(f))
+ formats.sort(key=self.FormatSort(self, field_preference).calculate_preference)
def _check_formats(self, formats, video_id):
if formats:
content_type = 'video'
elif codecs['acodec'] != 'none':
content_type = 'audio'
- elif codecs.get('tcodec', 'none') != 'none':
+ elif codecs.get('scodec', 'none') != 'none':
content_type = 'text'
elif mimetype2ext(mime_type) in ('tt', 'dfxp', 'ttml', 'xml', 'json'):
content_type = 'text'
self._set_cookie(domain, cookie, value)
break
- def get_testcases(self, include_onlymatching=False):
- t = getattr(self, '_TEST', None)
+ @classmethod
+ def get_testcases(cls, include_onlymatching=False):
+ t = getattr(cls, '_TEST', None)
if t:
- assert not hasattr(self, '_TESTS'), \
- '%s has _TEST and _TESTS' % type(self).__name__
+ assert not hasattr(cls, '_TESTS'), f'{cls.ie_key()}IE has _TEST and _TESTS'
tests = [t]
else:
- tests = getattr(self, '_TESTS', [])
+ tests = getattr(cls, '_TESTS', [])
for t in tests:
if not include_onlymatching and t.get('only_matching', False):
continue
- t['name'] = type(self).__name__[:-len('IE')]
+ t['name'] = cls.ie_key()
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 tc.get('playlist', []):
- 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
+ @classproperty
+ def age_limit(cls):
+ """Get age limit from the testcases"""
+ return max(traverse_obj(
+ tuple(cls.get_testcases(include_onlymatching=False)),
+ (..., (('playlist', 0), None), 'info_dict', 'age_limit')) or [0])
+
+ @classmethod
+ def is_suitable(cls, age_limit):
+ """Test whether the extractor is generally suitable for the given age limit"""
+ return not age_restricted(cls.age_limit, age_limit)
+
+ @classmethod
+ def description(cls, *, markdown=True, search_examples=None):
+ """Description of the extractor"""
+ desc = ''
+ if cls._NETRC_MACHINE:
+ if markdown:
+ desc += f' [<abbr title="netrc machine"><em>{cls._NETRC_MACHINE}</em></abbr>]'
+ else:
+ desc += f' [{cls._NETRC_MACHINE}]'
+ if cls.IE_DESC is False:
+ desc += ' [HIDDEN]'
+ elif cls.IE_DESC:
+ desc += f' {cls.IE_DESC}'
+ if cls.SEARCH_KEY:
+ desc += f'; "{cls.SEARCH_KEY}:" prefix'
+ if search_examples:
+ _COUNTS = ('', '5', '10', 'all')
+ desc += f' (Example: "{cls.SEARCH_KEY}{random.choice(_COUNTS)}:{random.choice(search_examples)}")'
+ if not cls.working():
+ desc += ' (**Currently broken**)' if markdown else ' (Currently broken)'
+
+ name = f' - **{cls.IE_NAME}**' if markdown else cls.IE_NAME
+ return f'{name}:{desc}' if desc else name
def extract_subtitles(self, *args, **kwargs):
if (self.get_param('writesubtitles', False)
def _get_automatic_captions(self, *args, **kwargs):
raise NotImplementedError('This method must be implemented by subclasses')
+ @property
+ def _cookies_passed(self):
+ """Whether cookies have been passed to YoutubeDL"""
+ return self.get_param('cookiefile') is not None or self.get_param('cookiesfrombrowser') is not None
+
def mark_watched(self, *args, **kwargs):
if not self.get_param('mark_watched', False):
return
- if (self.supports_login() and self._get_login_info()[0] is not None
- or self.get_param('cookiefile') or self.get_param('cookiesfrombrowser')):
+ if self.supports_login() and self._get_login_info()[0] is not None or self._cookies_passed:
self._mark_watched(*args, **kwargs)
def _mark_watched(self, *args, **kwargs):
"""Returns an iterator of search results"""
raise NotImplementedError('This method must be implemented by subclasses')
- @property
- def SEARCH_KEY(self):
- return self._SEARCH_KEY
+ @classproperty
+ def SEARCH_KEY(cls):
+ return cls._SEARCH_KEY