-# coding: utf-8
-from __future__ import unicode_literals
-
import base64
-import datetime
+import collections
import hashlib
+import itertools
import json
+import math
import netrc
import os
import random
-import re
import sys
import time
-import math
+import xml.etree.ElementTree
from ..compat import (
compat_cookiejar_Cookie,
compat_cookies_SimpleCookie,
- compat_etree_Element,
compat_etree_fromstring,
+ compat_expanduser,
compat_getpass,
compat_http_client,
compat_os_name,
compat_urllib_parse_urlencode,
compat_urllib_request,
compat_urlparse,
- compat_xml_parse_error,
+ re,
)
from ..downloader import FileDownloader
-from ..downloader.f4m import (
- get_base_url,
- remove_encrypted_media,
-)
+from ..downloader.f4m import get_base_url, remove_encrypted_media
from ..utils import (
+ JSON_LD_RE,
+ NO_DEFAULT,
+ ExtractorError,
+ GeoRestrictedError,
+ GeoUtils,
+ RegexNotFoundError,
+ UnsupportedError,
age_restricted,
base_url,
bug_reports_message,
+ classproperty,
clean_html,
- compiled_regex_type,
determine_ext,
determine_protocol,
dict_get,
+ encode_data_uri,
error_to_compat_str,
extract_attributes,
- ExtractorError,
+ filter_dict,
fix_xml_ampersands,
float_or_none,
format_field,
- GeoRestrictedError,
- GeoUtils,
int_or_none,
+ join_nonempty,
js_to_json,
- JSON_LD_RE,
mimetype2ext,
network_exceptions,
- NO_DEFAULT,
orderedSet,
parse_bitrate,
parse_codecs,
parse_iso8601,
parse_m3u8_attributes,
parse_resolution,
- RegexNotFoundError,
sanitize_filename,
sanitized_Request,
str_or_none,
str_to_int,
strip_or_none,
traverse_obj,
+ try_get,
unescapeHTML,
unified_strdate,
unified_timestamp,
)
-class InfoExtractor(object):
+class InfoExtractor:
"""Information Extractor class.
Information extractors are the classes that, given a URL, extract
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:
for HDS - URL of the F4M manifest,
for DASH - URL of the MPD manifest,
for MSS - URL of the ISM manifest.
+ * manifest_stream_number (For internal use only)
+ The index of the stream in the manifest file
* ext Will be calculated from URL if missing
* format A human-readable description of the format
("mp4 container with h264/opus").
* width Width of the video, if known
* height Height of the video, if known
* resolution Textual description of width and height
+ * dynamic_range The dynamic range of the video. One of:
+ "SDR" (None), "HDR10", "HDR10+, "HDR12", "HLG, "DV"
* tbr Average bitrate of audio and video in KBit/s
* abr Average audio bitrate in KBit/s
* acodec Name of the audio codec in use
* filesize_approx An estimate for the number of bytes
* 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", "rtmp_ffmpeg", "rtmpe",
- "m3u8", "m3u8_native" or "http_dash_segments".
+ download, lower-case. One of "http", "https" or
+ one of the protocols defined in downloader.PROTOCOL_MAP
* fragment_base_url
Base URL for fragments. Each fragment's path
value (if present) will be relative to
fragment_base_url
* "duration" (optional, int or float)
* "filesize" (optional, int)
+ * is_from_start Is a live format that can be downloaded
+ from the start. Boolean
* 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.
width : height ratio as float.
* no_resume The server does not support resuming the
(HTTP or RTMP) download. Boolean.
- * downloader_options A dictionary of downloader options as
- described in FileDownloader
+ * has_drm The format has DRM and cannot be downloaded. Boolean
+ * 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
The following fields are optional:
+ direct: True if a direct video file was given (must only be set by GenericIE)
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
* "resolution" (optional, string "{width}x{height}",
deprecated)
* "filesize" (optional, int)
- * "_test_url" (optional, bool) - If true, test the URL
+ * "http_headers" (dict) - HTTP headers for the request
thumbnail: Full URL to a video thumbnail image.
description: Full video description.
uploader: Full name of the video uploader.
license: License name the video is licensed under.
creator: The creator of the video.
- release_timestamp: UNIX timestamp of the moment the video was released.
- release_date: The date (YYYYMMDD) when the video was released.
timestamp: UNIX timestamp of the moment the video was uploaded
- upload_date: Video upload date (YYYYMMDD).
- If not explicitly set, calculated from timestamp.
+ upload_date: Video upload date in UTC (YYYYMMDD).
+ If not explicitly set, calculated from timestamp
+ release_timestamp: UNIX timestamp of the moment the video was released.
+ If it is not clear whether to use timestamp or this, use the former
+ release_date: The date (YYYYMMDD) when the video was released in UTC.
+ If not explicitly set, calculated from release_timestamp
+ modified_timestamp: UNIX timestamp of the moment the video was last modified.
+ modified_date: The date (YYYYMMDD) when the video was last modified in UTC.
+ If not explicitly set, calculated from modified_timestamp
uploader_id: Nickname or id of the video uploader.
uploader_url: Full URL to a personal webpage of the video uploader.
channel: Full name of the channel the video is uploaded on.
fields. This depends on a particular extractor.
channel_id: Id of the channel.
channel_url: Full URL to a channel webpage.
+ channel_follower_count: Number of followers of the channel.
location: Physical location where the video was filmed.
subtitles: The available subtitles as a dictionary in the format
{tag: subformats}. "tag" is usually a language code, and
* "url": A URL pointing to the subtitles file
It can optionally also have:
* "name": Name or description of the subtitles
+ * "http_headers": A dictionary of additional HTTP headers
+ to add to the request.
"ext" will be calculated from URL if missing
automatic_captions: Like 'subtitles'; contains automatically generated
captions instead of normal subtitles
series, programme or podcast:
series: Title of the series or programme the video episode belongs to.
+ series_id: Id of the series or programme the video episode belongs to, as a unicode string.
season: Title of the season the video episode belongs to.
season_number: Number of the season the video episode belongs to, as an integer.
season_id: Id of the season the video episode belongs to, as a unicode string.
disc_number: Number of the disc or other physical medium the track belongs to,
as an integer.
release_year: Year (YYYY) when the album was released.
+ composer: Composer of the piece
Unless mentioned otherwise, the fields should be Unicode strings.
Additionally, playlists can have "id", "title", and any other relevent
attributes with the same semantics as videos (see above).
+ It can also have the following optional fields:
+
+ playlist_count: The total number of videos in a playlist. If not given,
+ YoutubeDL tries to calculate it from "entries"
+
_type "multi_video" indicates that there are multiple videos that
form a single show, for examples multiple acts of an opera or TV episode.
title, description etc.
- Subclasses of this one should re-define the _real_initialize() and
- _real_extract() methods and define a _VALID_URL regexp.
+ Subclasses of this should define a _VALID_URL regexp and, re-define the
+ _real_extract() and (optionally) _real_initialize() methods.
Probably, they should also be added to the list of extractors.
+ Subclasses may also override suitable() if necessary, but ensure the function
+ signature is preserved and that this function imports everything it needs
+ (except other extractors), so that lazy_extractors works correctly.
+
+ To support username + password (or netrc) login, the extractor must define a
+ _NETRC_MACHINE and re-define _perform_login(username, password) and
+ (optionally) _initialize_pre_login() methods. The _perform_login method will
+ be called between _initialize_pre_login and _real_initialize if credentials
+ are passed by the user. In cases where it is necessary to have the login
+ process as part of the extraction rather than initialization, _perform_login
+ can be left undefined.
+
_GEO_BYPASS attribute may be set to False in order to disable
geo restriction bypass mechanisms for a particular extractor.
Though it won't disable explicit geo restriction bypass based on
will be used by geo restriction bypass mechanism similarly
to _GEO_COUNTRIES.
- Finally, the _WORKING attribute should be set to False for broken IEs
+ The _WORKING attribute should be set to False for broken IEs
in order to warn the users and skip the tests.
"""
_GEO_COUNTRIES = None
_GEO_IP_BLOCKS = None
_WORKING = True
+ _NETRC_MACHINE = None
+ IE_DESC = None
+ SEARCH_KEY = None
- _LOGIN_HINTS = {
- 'any': 'Use --cookies, --username and --password or --netrc to provide account credentials',
- 'cookies': (
- 'Use --cookies for the authentication. '
- 'See https://github.com/ytdl-org/youtube-dl#how-do-i-pass-cookies-to-youtube-dl for how to 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."""
+ """Constructor. Receives an optional downloader (a YoutubeDL instance).
+ If a downloader is not passed during initialization,
+ it must be set using "set_downloader()" before "extract()" is called"""
self._ready = False
self._x_forwarded_for_ip = None
self._printed_messages = set()
self.set_downloader(downloader)
@classmethod
- def suitable(cls, url):
- """Receives a URL and returns True if suitable for this IE."""
-
+ def _match_valid_url(cls, url):
# This does not use has/getattr intentionally - we want to know whether
# we have cached the regexp for *this* class, whereas getattr would also
# match the superclass
if '_VALID_URL_RE' not in cls.__dict__:
+ if '_VALID_URL' not in cls.__dict__:
+ cls._VALID_URL = cls._make_valid_url()
cls._VALID_URL_RE = re.compile(cls._VALID_URL)
- return cls._VALID_URL_RE.match(url) is not None
+ return cls._VALID_URL_RE.match(url)
+
+ @classmethod
+ def suitable(cls, url):
+ """Receives a URL and returns True if suitable for this IE."""
+ # This function must import everything it needs (except other extractors),
+ # so that lazy_extractors works correctly
+ return cls._match_valid_url(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 compat_str(m.group('id'))
+ return cls._match_valid_url(url).group('id')
+
+ @classmethod
+ def get_temp_id(cls, url):
+ try:
+ return cls._match_id(url)
+ except (IndexError, AttributeError):
+ return None
@classmethod
def working(cls):
"""Getter method for _WORKING."""
return cls._WORKING
+ @classmethod
+ def supports_login(cls):
+ return bool(cls._NETRC_MACHINE)
+
def initialize(self):
"""Initializes an instance (authentication, etc)."""
self._printed_messages = set()
'ip_blocks': self._GEO_IP_BLOCKS,
})
if not self._ready:
+ self._initialize_pre_login()
+ if self.supports_login():
+ username, password = self._get_login_info()
+ 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_hint("cookies")}')
self._real_initialize()
self._ready = True
if country:
self._x_forwarded_for_ip = GeoUtils.random_ipv4(country)
self._downloader.write_debug(
- 'Using fake IP %s (%s) as X-Forwarded-For' % (self._x_forwarded_for_ip, country.upper()))
+ f'Using fake IP {self._x_forwarded_for_ip} ({country.upper()}) as X-Forwarded-For')
def extract(self, url):
"""Extracts URL information and returns it in list of dicts."""
if self.__maybe_fake_ip_and_retry(e.countries):
continue
raise
- except ExtractorError:
+ except UnsupportedError:
raise
+ except ExtractorError as e:
+ kwargs = {
+ 'video_id': e.video_id or self.get_temp_id(url),
+ 'ie': self.IE_NAME,
+ 'tb': e.traceback or sys.exc_info()[2],
+ 'expected': e.expected,
+ 'cause': e.cause
+ }
+ if hasattr(e, 'countries'):
+ kwargs['countries'] = e.countries
+ raise type(e)(e.orig_msg, **kwargs)
except compat_http_client.IncompleteRead as e:
- raise ExtractorError('A network error has occurred.', cause=e, expected=True)
+ raise ExtractorError('A network error has occurred.', cause=e, expected=True, video_id=self.get_temp_id(url))
except (KeyError, StopIteration) as e:
- raise ExtractorError('An extractor error has occurred.', cause=e)
+ raise ExtractorError('An extractor error has occurred.', cause=e, video_id=self.get_temp_id(url))
def __maybe_fake_ip_and_retry(self, countries):
if (not self.get_param('geo_bypass_country', None)
return False
def set_downloader(self, downloader):
- """Sets the downloader for this IE."""
+ """Sets a YoutubeDL instance as the downloader for this IE."""
self._downloader = downloader
+ def _initialize_pre_login(self):
+ """ Intialization before login. Redefine in subclasses."""
+ pass
+
+ def _perform_login(self, username, password):
+ """ Login with username and password. Redefine in subclasses."""
+ pass
+
def _real_initialize(self):
"""Real initialization process. Redefine in subclasses."""
pass
def _real_extract(self, url):
"""Real extraction process. Redefine in subclasses."""
- pass
+ raise NotImplementedError('This method must be implemented by subclasses')
@classmethod
def ie_key(cls):
"""A string for getting the InfoExtractor with get_info_extractor"""
- return compat_str(cls.__name__[:-2])
+ 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):
See _download_webpage docstring for arguments specification.
"""
if not self._downloader._first_webpage_request:
- sleep_interval = float_or_none(self.get_param('sleep_interval_requests')) or 0
+ sleep_interval = self.get_param('sleep_interval_requests') or 0
if sleep_interval > 0:
self.to_screen('Sleeping %s seconds ...' % sleep_interval)
time.sleep(sleep_interval)
self.report_download_webpage(video_id)
elif note is not False:
if video_id is None:
- self.to_screen('%s' % (note,))
+ self.to_screen(str(note))
else:
- self.to_screen('%s: %s' % (video_id, note))
+ self.to_screen(f'{video_id}: {note}')
# Some sites check X-Forwarded-For HTTP header in order to figure out
# the origin of the client behind proxy. This allows bypassing geo
if errnote is None:
errnote = 'Unable to download webpage'
- errmsg = '%s: %s' % (errnote, error_to_compat_str(err))
+ errmsg = f'{errnote}: {error_to_compat_str(err)}'
if fatal:
- raise ExtractorError(errmsg, sys.exc_info()[2], cause=err)
+ raise ExtractorError(errmsg, cause=err)
else:
self.report_warning(errmsg)
return False
dump = base64.b64encode(webpage_bytes).decode('ascii')
self._downloader.to_screen(dump)
if self.get_param('write_pages', False):
- basen = '%s_%s' % (video_id, urlh.geturl())
- if len(basen) > 240:
+ basen = f'{video_id}_{urlh.geturl()}'
+ trim_length = self.get_param('trim_file_name') or 240
+ if len(basen) > trim_length:
h = '___' + hashlib.md5(basen.encode('utf-8')).hexdigest()
- basen = basen[:240 - len(h)] + h
+ basen = basen[:trim_length - len(h)] + h
raw_filename = basen + '.dump'
filename = sanitize_filename(raw_filename, restricted=True)
self.to_screen('Saving request to ' + filename)
fatal=True, encoding=None, data=None, headers={}, query={},
expected_status=None):
"""
- Return a tuple (xml as an compat_etree_Element, URL handle).
+ Return a tuple (xml as an xml.etree.ElementTree.Element, URL handle).
See _download_webpage docstring for arguments specification.
"""
transform_source=None, fatal=True, encoding=None,
data=None, headers={}, query={}, expected_status=None):
"""
- Return the xml as an compat_etree_Element.
+ Return the xml as an xml.etree.ElementTree.Element.
See _download_webpage docstring for arguments specification.
"""
xml_string = transform_source(xml_string)
try:
return compat_etree_fromstring(xml_string.encode('utf-8'))
- except compat_xml_parse_error as ve:
+ except xml.etree.ElementTree.ParseError as ve:
errmsg = '%s: Failed to parse XML ' % video_id
if fatal:
raise ExtractorError(errmsg, cause=ve)
if transform_source:
json_string = transform_source(json_string)
try:
- return json.loads(json_string)
+ return json.loads(json_string, strict=False)
except ValueError as ve:
errmsg = '%s: Failed to parse JSON ' % video_id
if fatal:
def to_screen(self, msg, *args, **kwargs):
"""Print msg to screen, prefixing it with '[ie_name]'"""
- self._downloader.to_screen('[%s] %s' % (self.IE_NAME, msg), *args, **kwargs)
+ self._downloader.to_screen(f'[{self.IE_NAME}] {msg}', *args, **kwargs)
def write_debug(self, msg, *args, **kwargs):
- self._downloader.write_debug('[%s] %s' % (self.IE_NAME, msg), *args, **kwargs)
+ self._downloader.write_debug(f'[{self.IE_NAME}] {msg}', *args, **kwargs)
def get_param(self, name, default=None, *args, **kwargs):
if self._downloader:
return self._downloader.params.get(name, default, *args, **kwargs)
return default
+ def report_drm(self, video_id, partial=False):
+ self.raise_no_formats('This video is DRM protected', expected=True, video_id=video_id)
+
def report_extraction(self, id_or_name):
"""Report information extraction."""
self.to_screen('%s: Extracting information' % id_or_name)
def raise_login_required(
self, msg='This video is only available for registered users',
- metadata_available=False, method='any'):
- if metadata_available and self.get_param('ignore_no_formats_error'):
+ metadata_available=False, method=NO_DEFAULT):
+ if metadata_available and (
+ self.get_param('ignore_no_formats_error') or self.get_param('wait_for_video')):
self.report_warning(msg)
- if method is not None:
- msg = '%s. %s' % (msg, self._LOGIN_HINTS[method])
+ return
+ msg += format_field(self._login_hint(method), template='. %s')
raise ExtractorError(msg, expected=True)
def raise_geo_restricted(
self, msg='This video is not available from your location due to geo restriction',
countries=None, metadata_available=False):
- if metadata_available and self.get_param('ignore_no_formats_error'):
+ if metadata_available and (
+ self.get_param('ignore_no_formats_error') or self.get_param('wait_for_video')):
self.report_warning(msg)
else:
raise GeoRestrictedError(msg, countries=countries)
def raise_no_formats(self, msg, expected=False, video_id=None):
- if expected and self.get_param('ignore_no_formats_error'):
+ if expected and (
+ self.get_param('ignore_no_formats_error') or self.get_param('wait_for_video')):
self.report_warning(msg, video_id)
elif isinstance(msg, ExtractorError):
raise msg
# Methods for following #608
@staticmethod
- def url_result(url, ie=None, video_id=None, video_title=None):
+ def url_result(url, ie=None, video_id=None, video_title=None, *, url_transparent=False, **kwargs):
"""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}
+ if ie is not None:
+ kwargs['ie_key'] = ie if isinstance(ie, str) else ie.ie_key()
if video_id is not None:
- video_info['id'] = video_id
+ kwargs['id'] = video_id
if video_title is not None:
- video_info['title'] = video_title
- return video_info
+ kwargs['title'] = video_title
+ return {
+ **kwargs,
+ '_type': 'url_transparent' if url_transparent else 'url',
+ 'url': url,
+ }
- def playlist_from_matches(self, matches, playlist_id=None, playlist_title=None, getter=None, ie=None):
- urls = orderedSet(
- self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
- for m in matches)
- return self.playlist_result(
- urls, playlist_id=playlist_id, playlist_title=playlist_title)
+ def playlist_from_matches(self, matches, playlist_id=None, playlist_title=None, getter=None, ie=None, video_kwargs=None, **kwargs):
+ urls = (self.url_result(self._proto_relative_url(m), ie, **(video_kwargs or {}))
+ for m in orderedSet(map(getter, matches) if getter else matches))
+ return self.playlist_result(urls, playlist_id, playlist_title, **kwargs)
@staticmethod
- def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None, **kwargs):
+ def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None, *, multi_video=False, **kwargs):
"""Returns a playlist"""
- video_info = {'_type': 'playlist',
- 'entries': entries}
- video_info.update(kwargs)
if playlist_id:
- video_info['id'] = playlist_id
+ kwargs['id'] = playlist_id
if playlist_title:
- video_info['title'] = playlist_title
+ kwargs['title'] = playlist_title
if playlist_description is not None:
- video_info['description'] = playlist_description
- return video_info
+ kwargs['description'] = playlist_description
+ return {
+ **kwargs,
+ '_type': 'multi_video' if multi_video else 'playlist',
+ 'entries': entries,
+ }
def _search_regex(self, pattern, string, name, default=NO_DEFAULT, fatal=True, flags=0, group=None):
"""
In case of failure return a default value or raise a WARNING or a
RegexNotFoundError, depending on fatal, specifying the field name.
"""
- if isinstance(pattern, (str, compat_str, compiled_regex_type)):
+ if string is None:
+ mobj = None
+ elif isinstance(pattern, (str, re.Pattern)):
mobj = re.search(pattern, string, flags)
else:
for p in pattern:
if mobj:
break
- if not self.get_param('no_color') and compat_os_name != 'nt' and sys.stderr.isatty():
- _name = '\033[0;34m%s\033[0m' % name
- else:
- _name = name
+ _name = self._downloader._format_err(name, self._downloader.Styles.EMPHASIS)
if mobj:
if group is None:
if self.get_param('usenetrc', False):
try:
- info = netrc.netrc().authenticators(netrc_machine)
+ netrc_file = compat_expanduser(self.get_param('netrc_location') or '~')
+ if os.path.isdir(netrc_file):
+ netrc_file = os.path.join(netrc_file, '.netrc')
+ info = netrc.netrc(file=netrc_file).authenticators(netrc_machine)
if info is not None:
username = info[0]
password = info[2]
else:
raise netrc.NetrcParseError(
'No authenticators for %s' % netrc_machine)
- except (IOError, netrc.NetrcParseError) as err:
+ except (OSError, netrc.NetrcParseError) as err:
self.report_warning(
'parsing .netrc: %s' % error_to_compat_str(err))
@staticmethod
def _og_regexes(prop):
content_re = r'content=(?:"([^"]+?)"|\'([^\']+?)\'|\s*([^\s"\'=<>`]+?))'
- property_re = (r'(?:name|property)=(?:\'og[:-]%(prop)s\'|"og[:-]%(prop)s"|\s*og[:-]%(prop)s\b)'
- % {'prop': re.escape(prop)})
+ property_re = (r'(?:name|property)=(?:\'og%(sep)s%(prop)s\'|"og%(sep)s%(prop)s"|\s*og%(sep)s%(prop)s\b)'
+ % {'prop': re.escape(prop), 'sep': '(?::|[:-])'})
template = r'<meta[^>]+?%s[^>]+?%s'
return [
template % (property_re, content_re),
def _og_search_description(self, html, **kargs):
return self._og_search_property('description', html, fatal=False, **kargs)
- def _og_search_title(self, html, **kargs):
- return self._og_search_property('title', html, **kargs)
+ def _og_search_title(self, html, *, fatal=False, **kargs):
+ return self._og_search_property('title', html, fatal=fatal, **kargs)
def _og_search_video_url(self, html, name='video url', secure=True, **kargs):
regexes = self._og_regexes('video') + self._og_regexes('video:url')
def _og_search_url(self, html, **kargs):
return self._og_search_property('url', html, **kargs)
+ def _html_extract_title(self, html, name='title', *, fatal=False, **kwargs):
+ return self._html_search_regex(r'(?s)<title>([^<]+)</title>', html, name, fatal=fatal, **kwargs)
+
def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs):
name = variadic(name)
if display_name is None:
continue
info[count_key] = interaction_count
+ def extract_chapter_information(e):
+ chapters = [{
+ 'title': part.get('name'),
+ 'start_time': part.get('startOffset'),
+ 'end_time': part.get('endOffset'),
+ } for part in variadic(e.get('hasPart') or []) if part.get('@type') == 'Clip']
+ for idx, (last_c, current_c, next_c) in enumerate(zip(
+ [{'end_time': 0}] + chapters, chapters, chapters[1:])):
+ current_c['end_time'] = current_c['end_time'] or next_c['start_time']
+ current_c['start_time'] = current_c['start_time'] or last_c['end_time']
+ if None in current_c.values():
+ self.report_warning(f'Chapter {idx} contains broken data. Not extracting chapters')
+ return
+ if chapters:
+ chapters[-1]['end_time'] = chapters[-1]['end_time'] or info['duration']
+ info['chapters'] = chapters
+
def extract_video_object(e):
assert e['@type'] == 'VideoObject'
author = e.get('author')
'url': url_or_none(e.get('contentUrl')),
'title': unescapeHTML(e.get('name')),
'description': unescapeHTML(e.get('description')),
- 'thumbnail': url_or_none(e.get('thumbnailUrl') or e.get('thumbnailURL')),
+ 'thumbnails': [{'url': url_or_none(url)}
+ for url in variadic(traverse_obj(e, 'thumbnailUrl', 'thumbnailURL'))],
'duration': parse_duration(e.get('duration')),
'timestamp': unified_timestamp(e.get('uploadDate')),
# author can be an instance of 'Organization' or 'Person' types.
'view_count': int_or_none(e.get('interactionCount')),
})
extract_interaction_statistic(e)
+ extract_chapter_information(e)
- for e in json_ld:
- if '@context' in e:
+ def traverse_json_ld(json_ld, at_top_level=True):
+ for e in json_ld:
+ if at_top_level and '@context' not in e:
+ continue
+ if at_top_level and set(e.keys()) == {'@context', '@graph'}:
+ traverse_json_ld(variadic(e['@graph'], allowed_types=(dict,)), at_top_level=False)
+ break
item_type = e.get('@type')
if expected_type is not None and expected_type != item_type:
continue
+ rating = traverse_obj(e, ('aggregateRating', 'ratingValue'), expected_type=float_or_none)
+ if rating is not None:
+ info['average_rating'] = rating
if item_type in ('TVEpisode', 'Episode'):
episode_name = unescapeHTML(e.get('name'))
info.update({
info.update({
'timestamp': parse_iso8601(e.get('datePublished')),
'title': unescapeHTML(e.get('headline')),
- 'description': unescapeHTML(e.get('articleBody')),
+ 'description': unescapeHTML(e.get('articleBody') or e.get('description')),
})
+ if traverse_obj(e, ('video', 0, '@type')) == 'VideoObject':
+ extract_video_object(e['video'][0])
elif item_type == 'VideoObject':
extract_video_object(e)
if expected_type is None:
continue
else:
break
- return dict((k, v) for k, v in info.items() if v is not None)
+ traverse_json_ld(json_ld)
+
+ return filter_dict(info)
+
+ def _search_nextjs_data(self, webpage, video_id, *, transform_source=None, fatal=True, **kw):
+ return self._parse_json(
+ self._search_regex(
+ r'(?s)<script[^>]+id=[\'"]__NEXT_DATA__[\'"][^>]*>([^<]+)</script>',
+ webpage, 'next.js data', fatal=fatal, **kw),
+ video_id, transform_source=transform_source, fatal=fatal)
+
+ def _search_nuxt_data(self, webpage, video_id, context_name='__NUXT__'):
+ ''' Parses Nuxt.js metadata. This works as long as the function __NUXT__ invokes is a pure function. '''
+ # not all website do this, but it can be changed
+ # https://stackoverflow.com/questions/67463109/how-to-change-or-hide-nuxt-and-nuxt-keyword-in-page-source
+ rectx = re.escape(context_name)
+ js, arg_keys, arg_vals = self._search_regex(
+ (r'<script>window\.%s=\(function\((?P<arg_keys>.*?)\)\{return\s(?P<js>\{.*?\})\}\((?P<arg_vals>.+?)\)\);?</script>' % rectx,
+ r'%s\(.*?\(function\((?P<arg_keys>.*?)\)\{return\s(?P<js>\{.*?\})\}\((?P<arg_vals>.*?)\)' % rectx),
+ webpage, context_name, group=['js', 'arg_keys', 'arg_vals'])
+
+ args = dict(zip(arg_keys.split(','), arg_vals.split(',')))
+
+ for key, val in args.items():
+ if val in ('undefined', 'void 0'):
+ args[key] = 'null'
+
+ return self._parse_json(js_to_json(js, args), video_id)['data'][0]
@staticmethod
def _hidden_inputs(html):
regex = r' *((?P<reverse>\+)?(?P<field>[a-zA-Z0-9_]+)((?P<separator>[~:])(?P<limit>.*?))?)? *$'
default = ('hidden', 'aud_or_vid', 'hasvid', 'ie_pref', 'lang', 'quality',
- 'res', 'fps', 'codec:vp9.2', 'size', 'br', 'asr',
- 'proto', 'ext', 'hasaud', 'source', 'format_id') # These must not be aliases
- ytdl_default = ('hasaud', 'quality', 'tbr', 'filesize', 'vbr',
+ 'res', 'fps', 'hdr:12', 'codec:vp9.2', 'size', 'br', 'asr',
+ 'proto', 'ext', 'hasaud', 'source', 'id') # These must not be aliases
+ ytdl_default = ('hasaud', 'lang', 'quality', 'tbr', 'filesize', 'vbr',
'height', 'width', 'proto', 'vext', 'abr', 'aext',
- 'fps', 'fs_approx', 'source', 'format_id')
+ 'fps', 'fs_approx', 'source', 'id')
settings = {
'vcodec': {'type': 'ordered', 'regex': True,
'order': ['av0?1', 'vp0?9.2', 'vp0?9', '[hx]265|he?vc?', '[hx]264|avc', 'vp0?8', 'mp4v|h263', 'theora', '', None, 'none']},
'acodec': {'type': 'ordered', 'regex': True,
- 'order': ['opus', 'vorbis', 'aac', 'mp?4a?', 'mp3', 'e?a?c-?3', 'dts', '', None, 'none']},
+ 'order': ['[af]lac', 'wav|aiff', 'opus', 'vorbis|ogg', 'aac', 'mp?4a?', 'mp3', 'e-?a?c-?3', 'ac-?3', 'dts', '', None, 'none']},
+ 'hdr': {'type': 'ordered', 'regex': True, 'field': 'dynamic_range',
+ 'order': ['dv', '(hdr)?12', r'(hdr)?10\+', '(hdr)?10', 'hlg', '', 'sdr', None]},
'proto': {'type': 'ordered', 'regex': True, 'field': 'protocol',
- 'order': ['(ht|f)tps', '(ht|f)tp$', 'm3u8.+', '.*dash', 'ws|websocket', '', 'mms|rtsp', 'none', 'f4']},
+ 'order': ['(ht|f)tps', '(ht|f)tp$', 'm3u8.*', '.*dash', 'websocket_frag', 'rtmpe?', '', 'mms|rtsp', 'ws|websocket', 'f4']},
'vext': {'type': 'ordered', 'field': 'video_ext',
'order': ('mp4', 'webm', 'flv', '', 'none'),
'order_free': ('webm', 'mp4', 'flv', '', 'none')},
'ie_pref': {'priority': True, 'type': 'extractor'},
'hasvid': {'priority': True, 'field': 'vcodec', 'type': 'boolean', 'not_in_list': ('none',)},
'hasaud': {'field': 'acodec', 'type': 'boolean', 'not_in_list': ('none',)},
- 'lang': {'priority': True, 'convert': 'ignore', 'field': 'language_preference'},
- 'quality': {'convert': 'float_none', 'default': -1},
+ 'lang': {'convert': 'float', 'field': 'language_preference', 'default': -1},
+ 'quality': {'convert': 'float', 'default': -1},
'filesize': {'convert': 'bytes'},
'fs_approx': {'convert': 'bytes', 'field': 'filesize_approx'},
'id': {'convert': 'string', 'field': 'format_id'},
'vbr': {'convert': 'float_none'},
'abr': {'convert': 'float_none'},
'asr': {'convert': 'float_none'},
- 'source': {'convert': 'ignore', 'field': 'source_preference'},
+ 'source': {'convert': 'float', 'field': 'source_preference', 'default': -1},
'codec': {'type': 'combined', 'field': ('vcodec', 'acodec')},
'br': {'type': 'combined', 'field': ('tbr', 'vbr', 'abr'), 'same_limit': True},
'res': {'type': 'multiple', 'field': ('height', 'width'),
'function': lambda it: (lambda l: min(l) if l else 0)(tuple(filter(None, it)))},
- # Most of these exist only for compatibility reasons
- 'dimension': {'type': 'alias', 'field': 'res'},
- 'resolution': {'type': 'alias', 'field': 'res'},
- 'extension': {'type': 'alias', 'field': 'ext'},
- 'bitrate': {'type': 'alias', 'field': 'br'},
- 'total_bitrate': {'type': 'alias', 'field': 'tbr'},
- 'video_bitrate': {'type': 'alias', 'field': 'vbr'},
- 'audio_bitrate': {'type': 'alias', 'field': 'abr'},
- 'framerate': {'type': 'alias', 'field': 'fps'},
- 'language_preference': {'type': 'alias', 'field': 'lang'}, # not named as 'language' because such a field exists
- 'protocol': {'type': 'alias', 'field': 'proto'},
+ # For compatibility with youtube-dl
+ 'format_id': {'type': 'alias', 'field': 'id'},
+ 'preference': {'type': 'alias', 'field': 'ie_pref'},
+ 'language_preference': {'type': 'alias', 'field': 'lang'},
'source_preference': {'type': 'alias', 'field': 'source'},
+ 'protocol': {'type': 'alias', 'field': 'proto'},
'filesize_approx': {'type': 'alias', 'field': 'fs_approx'},
- 'filesize_estimate': {'type': 'alias', 'field': 'size'},
- 'samplerate': {'type': 'alias', 'field': 'asr'},
- 'video_ext': {'type': 'alias', 'field': 'vext'},
- 'audio_ext': {'type': 'alias', 'field': 'aext'},
- 'video_codec': {'type': 'alias', 'field': 'vcodec'},
- 'audio_codec': {'type': 'alias', 'field': 'acodec'},
- 'video': {'type': 'alias', 'field': 'hasvid'},
- 'has_video': {'type': 'alias', 'field': 'hasvid'},
- 'audio': {'type': 'alias', 'field': 'hasaud'},
- 'has_audio': {'type': 'alias', 'field': 'hasaud'},
- 'extractor': {'type': 'alias', 'field': 'ie_pref'},
- 'preference': {'type': 'alias', 'field': 'ie_pref'},
- 'extractor_preference': {'type': 'alias', 'field': 'ie_pref'},
- 'format_id': {'type': 'alias', 'field': 'id'},
+
+ # Deprecated
+ 'dimension': {'type': 'alias', 'field': 'res', 'deprecated': True},
+ 'resolution': {'type': 'alias', 'field': 'res', 'deprecated': True},
+ 'extension': {'type': 'alias', 'field': 'ext', 'deprecated': True},
+ 'bitrate': {'type': 'alias', 'field': 'br', 'deprecated': True},
+ 'total_bitrate': {'type': 'alias', 'field': 'tbr', 'deprecated': True},
+ 'video_bitrate': {'type': 'alias', 'field': 'vbr', 'deprecated': True},
+ 'audio_bitrate': {'type': 'alias', 'field': 'abr', 'deprecated': True},
+ 'framerate': {'type': 'alias', 'field': 'fps', 'deprecated': True},
+ 'filesize_estimate': {'type': 'alias', 'field': 'size', 'deprecated': True},
+ 'samplerate': {'type': 'alias', 'field': 'asr', 'deprecated': True},
+ 'video_ext': {'type': 'alias', 'field': 'vext', 'deprecated': True},
+ 'audio_ext': {'type': 'alias', 'field': 'aext', 'deprecated': True},
+ 'video_codec': {'type': 'alias', 'field': 'vcodec', 'deprecated': True},
+ 'audio_codec': {'type': 'alias', 'field': 'acodec', 'deprecated': True},
+ 'video': {'type': 'alias', 'field': 'hasvid', 'deprecated': True},
+ 'has_video': {'type': 'alias', 'field': 'hasvid', 'deprecated': True},
+ 'audio': {'type': 'alias', 'field': 'hasaud', 'deprecated': True},
+ 'has_audio': {'type': 'alias', 'field': 'hasaud', 'deprecated': True},
+ 'extractor': {'type': 'alias', 'field': 'ie_pref', 'deprecated': True},
+ 'extractor_preference': {'type': 'alias', 'field': 'ie_pref', 'deprecated': True},
}
- _order = []
+ def __init__(self, ie, field_preference):
+ self._order = []
+ self.ydl = ie._downloader
+ self.evaluate_params(self.ydl.params, field_preference)
+ if ie.get_param('verbose'):
+ self.print_verbose_info(self.ydl.write_debug)
def _get_field_setting(self, field, key):
if field not in self.settings:
+ if key in ('forced', 'priority'):
+ return False
+ self.ydl.deprecation_warning(
+ f'Using arbitrary fields ({field}) for format sorting is deprecated '
+ 'and may be removed in a future version')
self.settings[field] = {}
propObj = self.settings[field]
if key not in propObj:
if field is None:
continue
if self._get_field_setting(field, 'type') == 'alias':
- field = self._get_field_setting(field, 'field')
+ alias, field = field, self._get_field_setting(field, 'field')
+ if self._get_field_setting(alias, 'deprecated'):
+ self.ydl.deprecation_warning(
+ f'Format sorting alias {alias} is deprecated '
+ f'and may be removed in a future version. Please use {field} instead')
reverse = match.group('reverse') is not None
closest = match.group('separator') == '~'
limit_text = match.group('limit')
has_multiple_limits = has_limit and has_multiple_fields and not self._get_field_setting(field, 'same_limit')
fields = self._get_field_setting(field, 'field') if has_multiple_fields else (field,)
- limits = limit_text.split(":") if has_multiple_limits else (limit_text,) if has_limit else tuple()
+ limits = limit_text.split(':') if has_multiple_limits else (limit_text,) if has_limit else tuple()
limit_count = len(limits)
for (i, f) in enumerate(fields):
add_item(f, reverse, closest,
if format.get('vbr') is not None and format.get('abr') is not None:
format['tbr'] = format.get('vbr', 0) + format.get('abr', 0)
else:
- if format.get('vcodec') != "none" and format.get('vbr') is None:
+ if format.get('vcodec') != 'none' and format.get('vbr') is None:
format['vbr'] = format.get('tbr') - format.get('abr', 0)
- if format.get('acodec') != "none" and format.get('abr') is None:
+ if format.get('acodec') != 'none' and format.get('abr') is None:
format['abr'] = format.get('tbr') - format.get('vbr', 0)
return tuple(self._calculate_field_preference(format, field) for field in self._order)
def _sort_formats(self, formats, field_preference=[]):
if not formats:
- if self.get_param('ignore_no_formats_error'):
- return
- raise ExtractorError('No video formats found')
- format_sort = self.FormatSort() # params and to_screen are taken from the downloader
- format_sort.evaluate_params(self._downloader.params, field_preference)
- if self.get_param('verbose', False):
- format_sort.print_verbose_info(self._downloader.write_debug)
- formats.sort(key=lambda f: format_sort.calculate_preference(f))
+ return
+ formats.sort(key=self.FormatSort(self, field_preference).calculate_preference)
def _check_formats(self, formats, video_id):
if formats:
def _extract_f4m_formats(self, manifest_url, video_id, preference=None, quality=None, f4m_id=None,
transform_source=lambda s: fix_xml_ampersands(s).strip(),
fatal=True, m3u8_id=None, data=None, headers={}, query={}):
- manifest = self._download_xml(
+ res = self._download_xml_handle(
manifest_url, video_id, 'Downloading f4m manifest',
'Unable to download f4m manifest',
# Some manifests may be malformed, e.g. prosiebensat1 generated manifests
# (see https://github.com/ytdl-org/youtube-dl/issues/6215#issuecomment-121704244)
transform_source=transform_source,
fatal=fatal, data=data, headers=headers, query=query)
-
- if manifest is False:
+ if res is False:
return []
+ manifest, urlh = res
+ manifest_url = urlh.geturl()
+
return self._parse_f4m_formats(
manifest, manifest_url, video_id, preference=preference, quality=quality, f4m_id=f4m_id,
transform_source=transform_source, fatal=fatal, m3u8_id=m3u8_id)
def _parse_f4m_formats(self, manifest, manifest_url, video_id, preference=None, quality=None, f4m_id=None,
transform_source=lambda s: fix_xml_ampersands(s).strip(),
fatal=True, m3u8_id=None):
- if not isinstance(manifest, compat_etree_Element) and not fatal:
+ if not isinstance(manifest, xml.etree.ElementTree.Element) and not fatal:
return []
# currently yt-dlp cannot decode the playerVerificationChallenge as Akamai uses Adobe Alchemy
tbr = int_or_none(media_el.attrib.get('bitrate'))
width = int_or_none(media_el.attrib.get('width'))
height = int_or_none(media_el.attrib.get('height'))
- format_id = '-'.join(filter(None, [f4m_id, compat_str(i if tbr is None else tbr)]))
+ format_id = join_nonempty(f4m_id, tbr or i)
# If <bootstrapInfo> is present, the specified f4m is a
# stream-level manifest, and only set-level manifests may refer to
# external resources. See section 11.4 and section 4 of F4M spec
def _m3u8_meta_format(self, m3u8_url, ext=None, preference=None, quality=None, m3u8_id=None):
return {
- 'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])),
+ 'format_id': join_nonempty(m3u8_id, 'meta'),
'url': m3u8_url,
'ext': ext,
'protocol': 'm3u8',
'format_note': 'Quality selection URL',
}
+ def _report_ignoring_subs(self, name):
+ self.report_warning(bug_reports_message(
+ f'Ignoring subtitle tracks found in the {name} manifest; '
+ 'if any subtitle tracks are missing,'
+ ), only_once=True)
+
def _extract_m3u8_formats(self, *args, **kwargs):
fmts, subs = self._extract_m3u8_formats_and_subtitles(*args, **kwargs)
if subs:
- self.report_warning(bug_reports_message(
- "Ignoring subtitle tracks found in the HLS manifest; "
- "if any subtitle tracks are missing,"
- ), only_once=True)
+ self._report_ignoring_subs('HLS')
return fmts
def _extract_m3u8_formats_and_subtitles(
headers=headers, query=query, video_id=video_id)
def _parse_m3u8_formats_and_subtitles(
- self, m3u8_doc, m3u8_url, ext=None, entry_protocol='m3u8_native',
+ self, m3u8_doc, m3u8_url=None, ext=None, entry_protocol='m3u8_native',
preference=None, quality=None, m3u8_id=None, live=False, note=None,
errnote=None, fatal=True, data=None, headers={}, query={},
video_id=None):
formats, subtitles = [], {}
- if '#EXT-X-FAXS-CM:' in m3u8_doc: # Adobe Flash Access
- return formats, subtitles
-
- if (not self.get_param('allow_unplayable_formats')
- and re.search(r'#EXT-X-SESSION-KEY:.*?URI="skd://', m3u8_doc)): # Apple FairPlay
- return formats, subtitles
+ has_drm = re.search('|'.join([
+ r'#EXT-X-FAXS-CM:', # Adobe Flash Access
+ r'#EXT-X-(?:SESSION-)?KEY:.*?URI="skd://', # Apple FairPlay
+ ]), m3u8_doc)
def format_url(url):
return url if re.match(r'^https?://', url) else compat_urlparse.urljoin(m3u8_url, url)
if '#EXT-X-TARGETDURATION' in m3u8_doc: # media playlist, return as is
formats = [{
- 'format_id': '-'.join(map(str, filter(None, [m3u8_id, idx]))),
+ 'format_id': join_nonempty(m3u8_id, idx),
'format_index': idx,
- 'url': m3u8_url,
+ 'url': m3u8_url or encode_data_uri(m3u8_doc.encode('utf-8'), 'application/x-mpegurl'),
'ext': ext,
'protocol': entry_protocol,
'preference': preference,
'quality': quality,
+ 'has_drm': has_drm,
} for idx in _extract_m3u8_playlist_indices(m3u8_doc=m3u8_doc)]
return formats, subtitles
if media_url:
manifest_url = format_url(media_url)
formats.extend({
- 'format_id': '-'.join(map(str, filter(None, (m3u8_id, group_id, name, idx)))),
+ 'format_id': join_nonempty(m3u8_id, group_id, name, idx),
'format_note': name,
'format_index': idx,
'url': manifest_url,
# format_id intact.
if not live:
stream_name = build_stream_name()
- format_id[1] = stream_name if stream_name else '%d' % (tbr if tbr else len(formats))
+ format_id[1] = stream_name or '%d' % (tbr or len(formats))
f = {
- 'format_id': '-'.join(map(str, filter(None, format_id))),
+ 'format_id': join_nonempty(*format_id),
'format_index': idx,
'url': manifest_url,
'manifest_url': m3u8_url,
last_stream_inf = {}
return formats, subtitles
+ def _extract_m3u8_vod_duration(
+ self, m3u8_vod_url, video_id, note=None, errnote=None, data=None, headers={}, query={}):
+
+ m3u8_vod = self._download_webpage(
+ m3u8_vod_url, video_id,
+ note='Downloading m3u8 VOD manifest' if note is None else note,
+ errnote='Failed to download VOD manifest' if errnote is None else errnote,
+ fatal=False, data=data, headers=headers, query=query)
+
+ return self._parse_m3u8_vod_duration(m3u8_vod or '', video_id)
+
+ def _parse_m3u8_vod_duration(self, m3u8_vod, video_id):
+ if '#EXT-X-PLAYLIST-TYPE:VOD' not in m3u8_vod:
+ return None
+
+ return int(sum(
+ float(line[len('#EXTINF:'):].split(',')[0])
+ for line in m3u8_vod.splitlines() if line.startswith('#EXTINF:'))) or None
+
@staticmethod
def _xpath_ns(path, namespace=None):
if not namespace:
return '/'.join(out)
def _extract_smil_formats_and_subtitles(self, smil_url, video_id, fatal=True, f4m_params=None, transform_source=None):
- smil = self._download_smil(smil_url, video_id, fatal=fatal, transform_source=transform_source)
-
- if smil is False:
+ res = self._download_smil(smil_url, video_id, fatal=fatal, transform_source=transform_source)
+ if res is False:
assert not fatal
- return []
+ return [], {}
+
+ smil, urlh = res
+ smil_url = urlh.geturl()
namespace = self._parse_smil_namespace(smil)
def _extract_smil_formats(self, *args, **kwargs):
fmts, subs = self._extract_smil_formats_and_subtitles(*args, **kwargs)
if subs:
- self.report_warning(bug_reports_message(
- "Ignoring subtitle tracks found in the SMIL manifest; "
- "if any subtitle tracks are missing,"
- ), only_once=True)
+ self._report_ignoring_subs('SMIL')
return fmts
def _extract_smil_info(self, smil_url, video_id, fatal=True, f4m_params=None):
- smil = self._download_smil(smil_url, video_id, fatal=fatal)
- if smil is False:
+ res = self._download_smil(smil_url, video_id, fatal=fatal)
+ if res is False:
return {}
+
+ smil, urlh = res
+ smil_url = urlh.geturl()
+
return self._parse_smil(smil, smil_url, video_id, f4m_params=f4m_params)
def _download_smil(self, smil_url, video_id, fatal=True, transform_source=None):
- return self._download_xml(
+ return self._download_xml_handle(
smil_url, video_id, 'Downloading SMIL file',
'Unable to download SMIL file', fatal=fatal, transform_source=transform_source)
rtmp_count = 0
http_count = 0
m3u8_count = 0
+ imgs_count = 0
- srcs = []
+ srcs = set()
media = smil.findall(self._xpath_ns('.//video', namespace)) + smil.findall(self._xpath_ns('.//audio', namespace))
for medium in media:
src = medium.get('src')
if not src or src in srcs:
continue
- srcs.append(src)
+ srcs.add(src)
bitrate = float_or_none(medium.get('system-bitrate') or medium.get('systemBitrate'), 1000)
filesize = int_or_none(medium.get('size') or medium.get('fileSize'))
'height': height,
})
+ for medium in smil.findall(self._xpath_ns('.//imagestream', namespace)):
+ src = medium.get('src')
+ if not src or src in srcs:
+ continue
+ srcs.add(src)
+
+ imgs_count += 1
+ formats.append({
+ 'format_id': 'imagestream-%d' % (imgs_count),
+ 'url': src,
+ 'ext': mimetype2ext(medium.get('type')),
+ 'acodec': 'none',
+ 'vcodec': 'none',
+ 'width': int_or_none(medium.get('width')),
+ 'height': int_or_none(medium.get('height')),
+ 'format_note': 'SMIL storyboards',
+ })
+
return formats
def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
return subtitles
def _extract_xspf_playlist(self, xspf_url, playlist_id, fatal=True):
- xspf = self._download_xml(
+ res = self._download_xml_handle(
xspf_url, playlist_id, 'Downloading xpsf playlist',
'Unable to download xspf manifest', fatal=fatal)
- if xspf is False:
+ if res is False:
return []
+
+ xspf, urlh = res
+ xspf_url = urlh.geturl()
+
return self._parse_xspf(
xspf, playlist_id, xspf_url=xspf_url,
xspf_base_url=base_url(xspf_url))
def _extract_mpd_formats(self, *args, **kwargs):
fmts, subs = self._extract_mpd_formats_and_subtitles(*args, **kwargs)
if subs:
- self.report_warning(bug_reports_message(
- "Ignoring subtitle tracks found in the DASH manifest; "
- "if any subtitle tracks are missing,"
- ), only_once=True)
+ self._report_ignoring_subs('DASH')
return fmts
def _extract_mpd_formats_and_subtitles(
mpd_doc, urlh = res
if mpd_doc is None:
return [], {}
- mpd_base_url = base_url(urlh.geturl())
+
+ # We could have been redirected to a new url when we retrieved our mpd file.
+ mpd_url = urlh.geturl()
+ mpd_base_url = base_url(mpd_url)
return self._parse_mpd_formats_and_subtitles(
mpd_doc, mpd_id, mpd_base_url, mpd_url)
def _parse_mpd_formats(self, *args, **kwargs):
fmts, subs = self._parse_mpd_formats_and_subtitles(*args, **kwargs)
if subs:
- self.report_warning(bug_reports_message(
- "Ignoring subtitle tracks found in the DASH manifest; "
- "if any subtitle tracks are missing,"
- ), only_once=True)
+ self._report_ignoring_subs('DASH')
return fmts
def _parse_mpd_formats_and_subtitles(
extract_Initialization(segment_template)
return ms_info
- skip_unplayable = not self.get_param('allow_unplayable_formats')
-
mpd_duration = parse_duration(mpd_doc.get('mediaPresentationDuration'))
- formats = []
- subtitles = {}
+ formats, subtitles = [], {}
+ stream_numbers = collections.defaultdict(int)
for period in mpd_doc.findall(_add_ns('Period')):
period_duration = parse_duration(period.get('duration')) or mpd_duration
period_ms_info = extract_multisegment_info(period, {
'timescale': 1,
})
for adaptation_set in period.findall(_add_ns('AdaptationSet')):
- if skip_unplayable and is_drm_protected(adaptation_set):
- continue
adaption_set_ms_info = extract_multisegment_info(adaptation_set, period_ms_info)
for representation in adaptation_set.findall(_add_ns('Representation')):
- if skip_unplayable and is_drm_protected(representation):
- continue
representation_attrib = adaptation_set.attrib.copy()
representation_attrib.update(representation.attrib)
# According to [1, 5.3.7.2, Table 9, page 41], @mimeType is mandatory
mime_type = representation_attrib['mimeType']
content_type = representation_attrib.get('contentType', mime_type.split('/')[0])
- codecs = representation_attrib.get('codecs', '')
+ codecs = parse_codecs(representation_attrib.get('codecs', ''))
if content_type not in ('video', 'audio', 'text'):
if mime_type == 'image/jpeg':
content_type = mime_type
- elif codecs.split('.')[0] == 'stpp':
+ elif codecs['vcodec'] != 'none':
+ content_type = 'video'
+ elif codecs['acodec'] != 'none':
+ content_type = 'audio'
+ elif codecs.get('scodec', 'none') != 'none':
+ content_type = 'text'
+ elif mimetype2ext(mime_type) in ('tt', 'dfxp', 'ttml', 'xml', 'json'):
content_type = 'text'
else:
self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
base_url = base_url_e.text + base_url
if re.match(r'^https?://', base_url):
break
- if mpd_base_url and not re.match(r'^https?://', base_url):
- if not mpd_base_url.endswith('/') and not base_url.startswith('/'):
+ if mpd_base_url and base_url.startswith('/'):
+ base_url = compat_urlparse.urljoin(mpd_base_url, base_url)
+ elif mpd_base_url and not re.match(r'^https?://', base_url):
+ if not mpd_base_url.endswith('/'):
mpd_base_url += '/'
base_url = mpd_base_url + base_url
representation_id = representation_attrib.get('id')
'format_note': 'DASH %s' % content_type,
'filesize': filesize,
'container': mimetype2ext(mime_type) + '_dash',
+ **codecs
}
- f.update(parse_codecs(codecs))
elif content_type == 'text':
f = {
'ext': mimetype2ext(mime_type),
'acodec': 'none',
'vcodec': 'none',
}
+ if is_drm_protected(adaptation_set) or is_drm_protected(representation):
+ f['has_drm'] = True
representation_ms_info = extract_multisegment_info(representation, adaption_set_ms_info)
def prepare_template(template_name, identifiers):
segment_duration = None
if 'total_number' not in representation_ms_info and 'segment_duration' in representation_ms_info:
segment_duration = float_or_none(representation_ms_info['segment_duration'], representation_ms_info['timescale'])
- representation_ms_info['total_number'] = int(math.ceil(float(period_duration) / segment_duration))
+ representation_ms_info['total_number'] = int(math.ceil(
+ float_or_none(period_duration, segment_duration, default=0)))
representation_ms_info['fragments'] = [{
media_location_key: media_template % {
'Number': segment_number,
f['url'] = initialization_url
f['fragments'].append({location_key(initialization_url): initialization_url})
f['fragments'].extend(representation_ms_info['fragments'])
+ if not period_duration:
+ period_duration = try_get(
+ representation_ms_info,
+ lambda r: sum(frag['duration'] for frag in r['fragments']), float)
else:
# Assuming direct URL to unfragmented media.
f['url'] = base_url
- if content_type in ('video', 'audio') or mime_type == 'image/jpeg':
+ if content_type in ('video', 'audio', 'image/jpeg'):
+ f['manifest_stream_number'] = stream_numbers[f['url']]
+ stream_numbers[f['url']] += 1
formats.append(f)
elif content_type == 'text':
subtitles.setdefault(lang or 'und', []).append(f)
def _extract_ism_formats(self, *args, **kwargs):
fmts, subs = self._extract_ism_formats_and_subtitles(*args, **kwargs)
if subs:
- self.report_warning(bug_reports_message(
- "Ignoring subtitle tracks found in the ISM manifest; "
- "if any subtitle tracks are missing,"
- ))
+ self._report_ignoring_subs('ISM')
return fmts
def _extract_ism_formats_and_subtitles(self, ism_url, video_id, ism_id=None, note=None, errnote=None, fatal=True, data=None, headers={}, query={}):
"""
if ism_doc.get('IsLive') == 'TRUE':
return [], {}
- if (not self.get_param('allow_unplayable_formats')
- and ism_doc.find('Protection') is not None):
- return [], {}
duration = int(ism_doc.attrib['Duration'])
timescale = int_or_none(ism_doc.get('TimeScale')) or 10000000
})
fragment_ctx['time'] += fragment_ctx['duration']
- format_id = []
- if ism_id:
- format_id.append(ism_id)
- if stream_name:
- format_id.append(stream_name)
- format_id.append(compat_str(tbr))
-
if stream_type == 'text':
subtitles.setdefault(stream_language, []).append({
'ext': 'ismt',
})
elif stream_type in ('video', 'audio'):
formats.append({
- 'format_id': '-'.join(format_id),
+ 'format_id': join_nonempty(ism_id, stream_name, tbr),
'url': ism_url,
'manifest_url': ism_url,
'ext': 'ismv' if stream_type == 'video' else 'isma',
'acodec': 'none' if stream_type == 'video' else fourcc,
'protocol': 'ism',
'fragments': fragments,
+ 'has_drm': ism_doc.find('Protection') is not None,
'_download_params': {
'stream_type': stream_type,
'duration': duration,
})
return formats, subtitles
- def _parse_html5_media_entries(self, base_url, webpage, video_id, m3u8_id=None, m3u8_entry_protocol='m3u8', mpd_id=None, preference=None, quality=None):
+ def _parse_html5_media_entries(self, base_url, webpage, video_id, m3u8_id=None, m3u8_entry_protocol='m3u8_native', mpd_id=None, preference=None, quality=None):
def absolute_url(item_url):
return urljoin(base_url, item_url)
def _extract_akamai_formats(self, *args, **kwargs):
fmts, subs = self._extract_akamai_formats_and_subtitles(*args, **kwargs)
if subs:
- self.report_warning(bug_reports_message(
- "Ignoring subtitle tracks found in the manifests; "
- "if any subtitle tracks are missing,"
- ))
+ self._report_ignoring_subs('akamai')
return fmts
def _extract_akamai_formats_and_subtitles(self, manifest_url, video_id, hosts={}):
http_f = f.copy()
del http_f['manifest_url']
http_url = re.sub(
- REPL_REGEX, protocol + r'://%s/\g<1>%s\3' % (http_host, qualities[i]), f['url'])
+ REPL_REGEX, protocol + fr'://{http_host}/\g<1>{qualities[i]}\3', f['url'])
http_f.update({
'format_id': http_f['format_id'].replace('hls-', protocol + '-'),
'url': http_url,
formats = []
def manifest_url(manifest):
- m_url = '%s/%s' % (http_base_url, manifest)
+ m_url = f'{http_base_url}/{manifest}'
if query:
m_url += '?%s' % query
return m_url
for protocol in ('rtmp', 'rtsp'):
if protocol not in skip_protocols:
formats.append({
- 'url': '%s:%s' % (protocol, url_base),
+ 'url': f'{protocol}:{url_base}',
'format_id': protocol,
'protocol': protocol,
})
return formats
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
+ self._downloader.deprecation_warning('yt_dlp.InfoExtractor._live_title is deprecated and does not work as expected')
+ return name
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)
+ msg = f'Failed to extract {name}: Could not parse value {v!r}'
if fatal:
raise ExtractorError(msg)
else:
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)
+ msg = f'Failed to extract {name}: Could not parse value {v!r}'
if fatal:
raise ExtractorError(msg)
else:
for header, cookies in url_handle.headers.items():
if header.lower() != 'set-cookie':
continue
- if sys.version_info[0] >= 3:
- cookies = cookies.encode('iso-8859-1')
- cookies = cookies.decode('utf-8')
+ cookies = cookies.encode('iso-8859-1').decode('utf-8')
cookie_value = re.search(
r'%s=(.+?);.*?\b[Dd]omain=(.+?)(?:[,;]|$)' % cookie, cookies)
if cookie_value:
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_subtitles(self, *args, **kwargs):
raise NotImplementedError('This method must be implemented by subclasses')
+ def extract_comments(self, *args, **kwargs):
+ if not self.get_param('getcomments'):
+ return None
+ generator = self._get_comments(*args, **kwargs)
+
+ def extractor():
+ comments = []
+ interrupted = True
+ try:
+ while True:
+ comments.append(next(generator))
+ except StopIteration:
+ interrupted = False
+ except KeyboardInterrupt:
+ self.to_screen('Interrupted by user')
+ except Exception as e:
+ if self.get_param('ignoreerrors') is not True:
+ raise
+ self._downloader.report_error(e)
+ comment_count = len(comments)
+ self.to_screen(f'Extracted {comment_count} comments')
+ return {
+ 'comments': comments,
+ 'comment_count': None if interrupted else comment_count
+ }
+ return extractor
+
+ def _get_comments(self, *args, **kwargs):
+ raise NotImplementedError('This method must be implemented by subclasses')
+
@staticmethod
def _merge_subtitle_items(subtitle_list1, subtitle_list2):
- """ Merge subtitle items for one language. Items with duplicated URLs
+ """ Merge subtitle items for one language. Items with duplicated URLs/data
will be dropped. """
- list1_urls = set([item['url'] for item in subtitle_list1])
+ list1_data = {(item.get('url'), item.get('data')) for item in subtitle_list1}
ret = list(subtitle_list1)
- ret.extend([item for item in subtitle_list2 if item['url'] not in list1_urls])
+ ret.extend(item for item in subtitle_list2 if (item.get('url'), item.get('data')) not in list1_data)
return ret
@classmethod
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 (self.get_param('mark_watched', False)
- and (self._get_login_info()[0] is not None
- or self.get_param('cookiefile') is not None)):
+ if not self.get_param('mark_watched', False):
+ return
+ 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):
else 'public' if all_known
else None)
- def _configuration_arg(self, key, default=NO_DEFAULT, casesense=False):
+ def _configuration_arg(self, key, default=NO_DEFAULT, *, ie_key=None, casesense=False):
'''
@returns A list of values for the extractor argument given by "key"
or "default" if no such key is present
@param casesense When false, the values are converted to lower case
'''
val = traverse_obj(
- self._downloader.params, ('extractor_args', self.ie_key().lower(), key))
+ self._downloader.params, ('extractor_args', (ie_key or self.ie_key()).lower(), key))
if val is None:
return [] if default is NO_DEFAULT else default
return list(val) if casesense else [x.lower() for x in val]
+ def _yes_playlist(self, playlist_id, video_id, smuggled_data=None, *, playlist_label='playlist', video_label='video'):
+ if not playlist_id or not video_id:
+ return not video_id
+
+ no_playlist = (smuggled_data or {}).get('force_noplaylist')
+ if no_playlist is not None:
+ return not no_playlist
+
+ video_id = '' if video_id is True else f' {video_id}'
+ playlist_id = '' if playlist_id is True else f' {playlist_id}'
+ if self.get_param('noplaylist'):
+ self.to_screen(f'Downloading just the {video_label}{video_id} because of --no-playlist')
+ return False
+ self.to_screen(f'Downloading {playlist_label}{playlist_id} - add --no-playlist to download just the {video_label}{video_id}')
+ return True
+
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.
+ Instances should define _SEARCH_KEY and optionally _MAX_RESULTS
"""
+ _MAX_RESULTS = float('inf')
+
@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('Invalid search query "%s"' % query)
-
- prefix = mobj.group('prefix')
- query = mobj.group('query')
+ prefix, query = self._match_valid_url(query).group('prefix', 'query')
if prefix == '':
return self._get_n_results(query, 1)
elif prefix == 'all':
else:
n = int(prefix)
if n <= 0:
- raise ExtractorError('invalid download number %s for query "%s"' % (n, query))
+ raise ExtractorError(f'invalid download number {n} for query "{query}"')
elif n > self._MAX_RESULTS:
self.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)
def _get_n_results(self, query, n):
- """Get a specified number of results for a query"""
+ """Get a specified number of results for a query.
+ Either this function or _search_results must be overridden by subclasses """
+ return self.playlist_result(
+ itertools.islice(self._search_results(query), 0, None if n == float('inf') else n),
+ query, query)
+
+ def _search_results(self, query):
+ """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