diff options
Diffstat (limited to 'yt_dlp/extractor/common.py')
| -rw-r--r-- | yt_dlp/extractor/common.py | 613 |
1 files changed, 306 insertions, 307 deletions
diff --git a/yt_dlp/extractor/common.py b/yt_dlp/extractor/common.py index ebeca4395..4fbcfe203 100644 --- a/yt_dlp/extractor/common.py +++ b/yt_dlp/extractor/common.py @@ -1,6 +1,10 @@ import base64 import collections +import getpass import hashlib +import http.client +import http.cookiejar +import http.cookies import itertools import json import math @@ -9,24 +13,12 @@ import os import random import sys import time +import urllib.parse +import urllib.request import xml.etree.ElementTree -from ..compat import ( - compat_cookiejar_Cookie, - compat_cookies_SimpleCookie, - compat_etree_fromstring, - compat_expanduser, - compat_getpass, - compat_http_client, - compat_os_name, - compat_str, - compat_urllib_error, - compat_urllib_parse_unquote, - compat_urllib_parse_urlencode, - compat_urllib_request, - compat_urlparse, - re, -) +from ..compat import functools, re # isort: split +from ..compat import compat_etree_fromstring, compat_expanduser, compat_os_name from ..downloader import FileDownloader from ..downloader.f4m import get_base_url, remove_encrypted_media from ..utils import ( @@ -35,6 +27,7 @@ from ..utils import ( ExtractorError, GeoRestrictedError, GeoUtils, + LenientJSONDecoder, RegexNotFoundError, UnsupportedError, age_restricted, @@ -384,6 +377,11 @@ class InfoExtractor: release_year: Year (YYYY) when the album was released. composer: Composer of the piece + The following fields should only be set for clips that should be cut from the original video: + + section_start: Start time of the section in seconds + section_end: End time of the section in seconds + Unless mentioned otherwise, the fields should be Unicode strings. Unless mentioned otherwise, None is equivalent to absence of information. @@ -610,8 +608,7 @@ class InfoExtractor: if ip_block: self._x_forwarded_for_ip = GeoUtils.random_ipv4(ip_block) - self._downloader.write_debug( - '[debug] Using fake IP %s as X-Forwarded-For' % self._x_forwarded_for_ip) + self.write_debug(f'Using fake IP {self._x_forwarded_for_ip} as X-Forwarded-For') return # Path 2: bypassing based on country code @@ -666,7 +663,7 @@ class InfoExtractor: if hasattr(e, 'countries'): kwargs['countries'] = e.countries raise type(e)(e.orig_msg, **kwargs) - except compat_http_client.IncompleteRead as e: + except http.client.IncompleteRead as e: 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, video_id=self.get_temp_id(url)) @@ -690,6 +687,14 @@ class InfoExtractor: """Sets a YoutubeDL instance as the downloader for this IE.""" self._downloader = downloader + @property + def cache(self): + return self._downloader.cache + + @property + def cookiejar(self): + return self._downloader.cookiejar + def _initialize_pre_login(self): """ Intialization before login. Redefine in subclasses.""" pass @@ -717,7 +722,7 @@ class InfoExtractor: @staticmethod def __can_accept_status_code(err, expected_status): - assert isinstance(err, compat_urllib_error.HTTPError) + assert isinstance(err, urllib.error.HTTPError) if expected_status is None: return False elif callable(expected_status): @@ -725,7 +730,14 @@ class InfoExtractor: else: return err.code in variadic(expected_status) - def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, data=None, headers={}, query={}, expected_status=None): + def _create_request(self, url_or_request, data=None, headers=None, query=None): + if isinstance(url_or_request, urllib.request.Request): + return update_Request(url_or_request, data=data, headers=headers, query=query) + if query: + url_or_request = update_url_query(url_or_request, query) + return sanitized_Request(url_or_request, data, headers or {}) + + def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, data=None, headers=None, query=None, expected_status=None): """ Return the response handle. @@ -753,21 +765,13 @@ class InfoExtractor: # geo unrestricted country. We will do so once we encounter any # geo restriction error. if self._x_forwarded_for_ip: - if 'X-Forwarded-For' not in headers: - headers['X-Forwarded-For'] = self._x_forwarded_for_ip + headers = (headers or {}).copy() + headers.setdefault('X-Forwarded-For', self._x_forwarded_for_ip) - if isinstance(url_or_request, compat_urllib_request.Request): - url_or_request = update_Request( - url_or_request, data=data, headers=headers, query=query) - else: - if query: - url_or_request = update_url_query(url_or_request, query) - if data is not None or headers: - url_or_request = sanitized_Request(url_or_request, data, headers) try: - return self._downloader.urlopen(url_or_request) + return self._downloader.urlopen(self._create_request(url_or_request, data, headers, query)) except network_exceptions as err: - if isinstance(err, compat_urllib_error.HTTPError): + if isinstance(err, urllib.error.HTTPError): if self.__can_accept_status_code(err, expected_status): # Retain reference to error to prevent file object from # being closed before it can be read. Works around the @@ -788,14 +792,42 @@ class InfoExtractor: self.report_warning(errmsg) return False - def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True, encoding=None, data=None, headers={}, query={}, expected_status=None): + def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True, + encoding=None, data=None, headers={}, query={}, expected_status=None): """ Return a tuple (page content as string, URL handle). - See _download_webpage docstring for arguments specification. + Arguments: + url_or_request -- plain text URL as a string or + a urllib.request.Request object + video_id -- Video/playlist/item identifier (string) + + Keyword arguments: + note -- note printed before downloading (string) + errnote -- note printed in case of an error (string) + fatal -- flag denoting whether error should be considered fatal, + i.e. whether it should cause ExtractionError to be raised, + otherwise a warning will be reported and extraction continued + encoding -- encoding for a page content decoding, guessed automatically + when not explicitly specified + data -- POST data (bytes) + headers -- HTTP headers (dict) + query -- URL query (dict) + expected_status -- allows to accept failed HTTP requests (non 2xx + status code) by explicitly specifying a set of accepted status + codes. Can be any of the following entities: + - an integer type specifying an exact failed status code to + accept + - a list or a tuple of integer types specifying a list of + failed status codes to accept + - a callable accepting an actual failed status code and + returning True if it should be accepted + Note that this argument does not affect success status codes (2xx) + which are always accepted. """ + # Strip hashes from the URL (#1038) - if isinstance(url_or_request, (compat_str, str)): + if isinstance(url_or_request, str): url_or_request = url_or_request.partition('#')[0] urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal, data=data, headers=headers, query=query, expected_status=expected_status) @@ -850,140 +882,48 @@ class InfoExtractor: 'Visit http://blocklist.rkn.gov.ru/ for a block reason.', expected=True) + def _request_dump_filename(self, url, video_id): + basen = f'{video_id}_{url}' + 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[:trim_length - len(h)] + h + filename = sanitize_filename(f'{basen}.dump', restricted=True) + # Working around MAX_PATH limitation on Windows (see + # http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx) + if compat_os_name == 'nt': + absfilepath = os.path.abspath(filename) + if len(absfilepath) > 259: + filename = fR'\\?\{absfilepath}' + return filename + + def __decode_webpage(self, webpage_bytes, encoding, headers): + if not encoding: + encoding = self._guess_encoding_from_content(headers.get('Content-Type', ''), webpage_bytes) + try: + return webpage_bytes.decode(encoding, 'replace') + except LookupError: + return webpage_bytes.decode('utf-8', 'replace') + 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.get_param('dump_intermediate_pages', False): self.to_screen('Dumping request to ' + urlh.geturl()) dump = base64.b64encode(webpage_bytes).decode('ascii') self._downloader.to_screen(dump) - if self.get_param('write_pages', False): - 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[:trim_length - len(h)] + h - raw_filename = basen + '.dump' - filename = sanitize_filename(raw_filename, restricted=True) - 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 compat_os_name == 'nt': - absfilepath = os.path.abspath(filename) - if len(absfilepath) > 259: - filename = '\\\\?\\' + absfilepath + if self.get_param('write_pages'): + filename = self._request_dump_filename(urlh.geturl(), video_id) + self.to_screen(f'Saving request to {filename}') with open(filename, 'wb') as outf: outf.write(webpage_bytes) - try: - content = webpage_bytes.decode(encoding, 'replace') - except LookupError: - content = webpage_bytes.decode('utf-8', 'replace') - + content = self.__decode_webpage(webpage_bytes, encoding, urlh.headers) self.__check_blocked(content) return content - def _download_webpage( - self, url_or_request, video_id, note=None, errnote=None, - fatal=True, tries=1, timeout=5, encoding=None, data=None, - headers={}, query={}, expected_status=None): - """ - Return the data of the page as a string. - - Arguments: - url_or_request -- plain text URL as a string or - a compat_urllib_request.Requestobject - video_id -- Video/playlist/item identifier (string) - - Keyword arguments: - note -- note printed before downloading (string) - errnote -- note printed in case of an error (string) - fatal -- flag denoting whether error should be considered fatal, - i.e. whether it should cause ExtractionError to be raised, - otherwise a warning will be reported and extraction continued - tries -- number of tries - timeout -- sleep interval between tries - encoding -- encoding for a page content decoding, guessed automatically - when not explicitly specified - data -- POST data (bytes) - headers -- HTTP headers (dict) - query -- URL query (dict) - expected_status -- allows to accept failed HTTP requests (non 2xx - status code) by explicitly specifying a set of accepted status - codes. Can be any of the following entities: - - an integer type specifying an exact failed status code to - accept - - a list or a tuple of integer types specifying a list of - failed status codes to accept - - a callable accepting an actual failed status code and - returning True if it should be accepted - Note that this argument does not affect success status codes (2xx) - which are always accepted. - """ - - 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, data=data, headers=headers, query=query, - expected_status=expected_status) - 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: - content, _ = res - return content - - def _download_xml_handle( - self, url_or_request, video_id, note='Downloading XML', - errnote='Unable to download XML', transform_source=None, - fatal=True, encoding=None, data=None, headers={}, query={}, - expected_status=None): - """ - Return a tuple (xml as an xml.etree.ElementTree.Element, URL handle). - - See _download_webpage docstring for arguments specification. - """ - res = self._download_webpage_handle( - url_or_request, video_id, note, errnote, fatal=fatal, - encoding=encoding, data=data, headers=headers, query=query, - expected_status=expected_status) - if res is False: - return res - xml_string, urlh = res - return self._parse_xml( - xml_string, video_id, transform_source=transform_source, - fatal=fatal), urlh - - def _download_xml( - self, url_or_request, video_id, - note='Downloading XML', errnote='Unable to download XML', - transform_source=None, fatal=True, encoding=None, - data=None, headers={}, query={}, expected_status=None): - """ - Return the xml as an xml.etree.ElementTree.Element. - - See _download_webpage docstring for arguments specification. - """ - res = self._download_xml_handle( - url_or_request, video_id, note=note, errnote=errnote, - transform_source=transform_source, fatal=fatal, encoding=encoding, - data=data, headers=headers, query=query, - expected_status=expected_status) - return res if res is False else res[0] - def _parse_xml(self, xml_string, video_id, transform_source=None, fatal=True): if transform_source: xml_string = transform_source(xml_string) @@ -996,101 +936,126 @@ class InfoExtractor: else: self.report_warning(errmsg + str(ve)) - def _download_json_handle( - self, url_or_request, video_id, note='Downloading JSON metadata', - errnote='Unable to download JSON metadata', transform_source=None, - fatal=True, encoding=None, data=None, headers={}, query={}, - expected_status=None): - """ - Return a tuple (JSON object, URL handle). - - See _download_webpage docstring for arguments specification. - """ - res = self._download_webpage_handle( - url_or_request, video_id, note, errnote, fatal=fatal, - encoding=encoding, data=data, headers=headers, query=query, - expected_status=expected_status) - if res is False: - return res - json_string, urlh = res - return self._parse_json( - json_string, video_id, transform_source=transform_source, - fatal=fatal), urlh - - def _download_json( - self, url_or_request, video_id, note='Downloading JSON metadata', - errnote='Unable to download JSON metadata', transform_source=None, - fatal=True, encoding=None, data=None, headers={}, query={}, - expected_status=None): - """ - Return the JSON object as a dict. - - See _download_webpage docstring for arguments specification. - """ - res = self._download_json_handle( - url_or_request, video_id, note=note, errnote=errnote, - transform_source=transform_source, fatal=fatal, encoding=encoding, - data=data, headers=headers, query=query, - expected_status=expected_status) - return res if res is False else res[0] - - def _parse_json(self, json_string, video_id, transform_source=None, fatal=True): - if transform_source: - json_string = transform_source(json_string) + def _parse_json(self, json_string, video_id, transform_source=None, fatal=True, **parser_kwargs): try: - return json.loads(json_string, strict=False) + return json.loads( + json_string, cls=LenientJSONDecoder, strict=False, transform_source=transform_source, **parser_kwargs) except ValueError as ve: - errmsg = '%s: Failed to parse JSON ' % video_id + errmsg = f'{video_id}: Failed to parse JSON' if fatal: raise ExtractorError(errmsg, cause=ve) else: - self.report_warning(errmsg + str(ve)) + self.report_warning(f'{errmsg}: {ve}') def _parse_socket_response_as_json(self, data, video_id, transform_source=None, fatal=True): return self._parse_json( data[data.find('{'):data.rfind('}') + 1], video_id, transform_source, fatal) - def _download_socket_json_handle( - self, url_or_request, video_id, note='Polling socket', - errnote='Unable to poll socket', transform_source=None, - fatal=True, encoding=None, data=None, headers={}, query={}, - expected_status=None): - """ - Return a tuple (JSON object, URL handle). + def __create_download_methods(name, parser, note, errnote, return_value): + + def parse(ie, content, *args, **kwargs): + if parser is None: + return content + # parser is fetched by name so subclasses can override it + return getattr(ie, parser)(content, *args, **kwargs) + + def download_handle(self, url_or_request, video_id, note=note, errnote=errnote, transform_source=None, + fatal=True, encoding=None, data=None, headers={}, query={}, expected_status=None): + res = self._download_webpage_handle( + url_or_request, video_id, note=note, errnote=errnote, fatal=fatal, encoding=encoding, + data=data, headers=headers, query=query, expected_status=expected_status) + if res is False: + return res + content, urlh = res + return parse(self, content, video_id, transform_source=transform_source, fatal=fatal), urlh + + def download_content(self, url_or_request, video_id, note=note, errnote=errnote, transform_source=None, + fatal=True, encoding=None, data=None, headers={}, query={}, expected_status=None): + if self.get_param('load_pages'): + url_or_request = self._create_request(url_or_request, data, headers, query) + filename = self._request_dump_filename(url_or_request.full_url, video_id) + self.to_screen(f'Loading request from {filename}') + try: + with open(filename, 'rb') as dumpf: + webpage_bytes = dumpf.read() + except OSError as e: + self.report_warning(f'Unable to load request from disk: {e}') + else: + content = self.__decode_webpage(webpage_bytes, encoding, url_or_request.headers) + return parse(self, content, video_id, transform_source, fatal) + kwargs = { + 'note': note, + 'errnote': errnote, + 'transform_source': transform_source, + 'fatal': fatal, + 'encoding': encoding, + 'data': data, + 'headers': headers, + 'query': query, + 'expected_status': expected_status, + } + if parser is None: + kwargs.pop('transform_source') + # The method is fetched by name so subclasses can override _download_..._handle + res = getattr(self, download_handle.__name__)(url_or_request, video_id, **kwargs) + return res if res is False else res[0] + + def impersonate(func, name, return_value): + func.__name__, func.__qualname__ = name, f'InfoExtractor.{name}' + func.__doc__ = f''' + @param transform_source Apply this transformation before parsing + @returns {return_value} + + See _download_webpage_handle docstring for other arguments specification + ''' + + impersonate(download_handle, f'_download_{name}_handle', f'({return_value}, URL handle)') + impersonate(download_content, f'_download_{name}', f'{return_value}') + return download_handle, download_content + + _download_xml_handle, _download_xml = __create_download_methods( + 'xml', '_parse_xml', 'Downloading XML', 'Unable to download XML', 'xml as an xml.etree.ElementTree.Element') + _download_json_handle, _download_json = __create_download_methods( + 'json', '_parse_json', 'Downloading JSON metadata', 'Unable to download JSON metadata', 'JSON object as a dict') + _download_socket_json_handle, _download_socket_json = __create_download_methods( + 'socket_json', '_parse_socket_response_as_json', 'Polling socket', 'Unable to poll socket', 'JSON object as a dict') + __download_webpage = __create_download_methods('webpage', None, None, None, 'data of the page as a string')[1] - See _download_webpage docstring for arguments specification. - """ - res = self._download_webpage_handle( - url_or_request, video_id, note, errnote, fatal=fatal, - encoding=encoding, data=data, headers=headers, query=query, - expected_status=expected_status) - if res is False: - return res - webpage, urlh = res - return self._parse_socket_response_as_json( - webpage, video_id, transform_source=transform_source, - fatal=fatal), urlh - - def _download_socket_json( - self, url_or_request, video_id, note='Polling socket', - errnote='Unable to poll socket', transform_source=None, - fatal=True, encoding=None, data=None, headers={}, query={}, - expected_status=None): + def _download_webpage( + self, url_or_request, video_id, note=None, errnote=None, + fatal=True, tries=1, timeout=NO_DEFAULT, *args, **kwargs): """ - Return the JSON object as a dict. + Return the data of the page as a string. - See _download_webpage docstring for arguments specification. + Keyword arguments: + tries -- number of tries + timeout -- sleep interval between tries + + See _download_webpage_handle docstring for other arguments specification. """ - res = self._download_socket_json_handle( - url_or_request, video_id, note=note, errnote=errnote, - transform_source=transform_source, fatal=fatal, encoding=encoding, - data=data, headers=headers, query=query, - expected_status=expected_status) - return res if res is False else res[0] + + R''' # NB: These are unused; should they be deprecated? + if tries != 1: + self._downloader.deprecation_warning('tries argument is deprecated in InfoExtractor._download_webpage') + if timeout is NO_DEFAULT: + timeout = 5 + else: + self._downloader.deprecation_warning('timeout argument is deprecated in InfoExtractor._download_webpage') + ''' + + try_count = 0 + while True: + try: + return self.__download_webpage(url_or_request, video_id, note, errnote, None, fatal, *args, **kwargs) + except http.client.IncompleteRead as e: + try_count += 1 + if try_count >= tries: + raise e + self._sleep(timeout, video_id) def report_warning(self, msg, video_id=None, *args, only_once=False, **kwargs): - idstr = format_field(video_id, template='%s: ') + idstr = format_field(video_id, None, '%s: ') msg = f'[{self.IE_NAME}] {idstr}{msg}' if only_once: if f'WARNING: {msg}' in self._printed_messages: @@ -1136,7 +1101,7 @@ class InfoExtractor: self.get_param('ignore_no_formats_error') or self.get_param('wait_for_video')): self.report_warning(msg) return - msg += format_field(self._login_hint(method), template='. %s') + msg += format_field(self._login_hint(method), None, '. %s') raise ExtractorError(msg, expected=True) def raise_geo_restricted( @@ -1228,6 +1193,33 @@ class InfoExtractor: self.report_warning('unable to extract %s' % _name + bug_reports_message()) return None + def _search_json(self, start_pattern, string, name, video_id, *, end_pattern='', + contains_pattern='(?s:.+)', fatal=True, default=NO_DEFAULT, **kwargs): + """Searches string for the JSON object specified by start_pattern""" + # NB: end_pattern is only used to reduce the size of the initial match + if default is NO_DEFAULT: + default, has_default = {}, False + else: + fatal, has_default = False, True + + json_string = self._search_regex( + rf'{start_pattern}\s*(?P<json>{{\s*{contains_pattern}\s*}})\s*{end_pattern}', + string, name, group='json', fatal=fatal, default=None if has_default else NO_DEFAULT) + if not json_string: + return default + + _name = self._downloader._format_err(name, self._downloader.Styles.EMPHASIS) + try: + return self._parse_json(json_string, video_id, ignore_extra=True, **kwargs) + except ExtractorError as e: + if fatal: + raise ExtractorError( + f'Unable to extract {_name} - Failed to parse JSON', cause=e.cause, video_id=video_id) + elif not has_default: + self.report_warning( + f'Unable to extract {_name} - Failed to parse JSON: {e}', video_id=video_id) + return default + 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. @@ -1292,7 +1284,7 @@ class InfoExtractor: if tfa is not None: return tfa - return compat_getpass('Type %s and press [Return]: ' % note) + return getpass.getpass('Type %s and press [Return]: ' % note) # Helper functions for extracting OpenGraph info @staticmethod @@ -1343,7 +1335,7 @@ class InfoExtractor: 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) + return self._html_search_regex(r'(?s)<title\b[^>]*>([^<]+)</title>', html, name, fatal=fatal, **kwargs) def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs): name = variadic(name) @@ -1400,27 +1392,25 @@ class InfoExtractor: return self._html_search_meta('twitter:player', html, 'twitter card player') - def _search_json_ld(self, html, video_id, expected_type=None, **kwargs): - json_ld_list = list(re.finditer(JSON_LD_RE, html)) - default = kwargs.get('default', NO_DEFAULT) - # JSON-LD may be malformed and thus `fatal` should be respected. - # At the same time `default` may be passed that assumes `fatal=False` - # for _search_regex. Let's simulate the same behavior here as well. - fatal = kwargs.get('fatal', True) if default is NO_DEFAULT else False - json_ld = [] - for mobj in json_ld_list: - json_ld_item = self._parse_json( - mobj.group('json_ld'), video_id, fatal=fatal) - if not json_ld_item: - continue - if isinstance(json_ld_item, dict): - json_ld.append(json_ld_item) - elif isinstance(json_ld_item, (list, tuple)): - json_ld.extend(json_ld_item) - if json_ld: - json_ld = self._json_ld(json_ld, video_id, fatal=fatal, expected_type=expected_type) - if json_ld: - return json_ld + def _yield_json_ld(self, html, video_id, *, fatal=True, default=NO_DEFAULT): + """Yield all json ld objects in the html""" + if default is not NO_DEFAULT: + fatal = False + for mobj in re.finditer(JSON_LD_RE, html): + json_ld_item = self._parse_json(mobj.group('json_ld'), video_id, fatal=fatal) + for json_ld in variadic(json_ld_item): + if isinstance(json_ld, dict): + yield json_ld + + def _search_json_ld(self, html, video_id, expected_type=None, *, fatal=True, default=NO_DEFAULT): + """Search for a video in any json ld in the html""" + if default is not NO_DEFAULT: + fatal = False + info = self._json_ld( + list(self._yield_json_ld(html, video_id, fatal=fatal, default=default)), + video_id, fatal=fatal, expected_type=expected_type) + if info: + return info if default is not NO_DEFAULT: return default elif fatal: @@ -1430,7 +1420,7 @@ class InfoExtractor: return {} def _json_ld(self, json_ld, video_id, fatal=True, expected_type=None): - if isinstance(json_ld, compat_str): + if isinstance(json_ld, str): json_ld = self._parse_json(json_ld, video_id, fatal=fatal) if not json_ld: return {} @@ -1451,6 +1441,10 @@ class InfoExtractor: 'ViewAction': 'view', } + def is_type(e, *expected_types): + type = variadic(traverse_obj(e, '@type')) + return any(x in type for x in expected_types) + def extract_interaction_type(e): interaction_type = e.get('interactionType') if isinstance(interaction_type, dict): @@ -1464,9 +1458,7 @@ class InfoExtractor: if not isinstance(interaction_statistic, list): return for is_e in interaction_statistic: - if not isinstance(is_e, dict): - continue - if is_e.get('@type') != 'InteractionCounter': + if not is_type(is_e, 'InteractionCounter'): continue interaction_type = extract_interaction_type(is_e) if not interaction_type: @@ -1503,22 +1495,23 @@ class InfoExtractor: info['chapters'] = chapters def extract_video_object(e): - assert e['@type'] == 'VideoObject' + assert is_type(e, 'VideoObject') author = e.get('author') info.update({ 'url': url_or_none(e.get('contentUrl')), 'title': unescapeHTML(e.get('name')), 'description': unescapeHTML(e.get('description')), - 'thumbnails': [{'url': url_or_none(url)} - for url in variadic(traverse_obj(e, 'thumbnailUrl', 'thumbnailURL'))], + 'thumbnails': [{'url': url} + for url in variadic(traverse_obj(e, 'thumbnailUrl', 'thumbnailURL')) + if url_or_none(url)], 'duration': parse_duration(e.get('duration')), 'timestamp': unified_timestamp(e.get('uploadDate')), # author can be an instance of 'Organization' or 'Person' types. # both types can have 'name' property(inherited from 'Thing' type). [1] # however some websites are using 'Text' type instead. # 1. https://schema.org/VideoObject - 'uploader': author.get('name') if isinstance(author, dict) else author if isinstance(author, compat_str) else None, - 'filesize': float_or_none(e.get('contentSize')), + 'uploader': author.get('name') if isinstance(author, dict) else author if isinstance(author, str) else None, + 'filesize': int_or_none(float_or_none(e.get('contentSize'))), 'tbr': int_or_none(e.get('bitrate')), 'width': int_or_none(e.get('width')), 'height': int_or_none(e.get('height')), @@ -1534,13 +1527,12 @@ class InfoExtractor: 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: + if expected_type is not None and not is_type(e, expected_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'): + if is_type(e, 'TVEpisode', 'Episode'): episode_name = unescapeHTML(e.get('name')) info.update({ 'episode': episode_name, @@ -1550,37 +1542,39 @@ class InfoExtractor: if not info.get('title') and episode_name: info['title'] = episode_name part_of_season = e.get('partOfSeason') - if isinstance(part_of_season, dict) and part_of_season.get('@type') in ('TVSeason', 'Season', 'CreativeWorkSeason'): + if is_type(part_of_season, 'TVSeason', 'Season', 'CreativeWorkSeason'): info.update({ 'season': unescapeHTML(part_of_season.get('name')), 'season_number': int_or_none(part_of_season.get('seasonNumber')), }) part_of_series = e.get('partOfSeries') or e.get('partOfTVSeries') - if isinstance(part_of_series, dict) and part_of_series.get('@type') in ('TVSeries', 'Series', 'CreativeWorkSeries'): + if is_type(part_of_series, 'TVSeries', 'Series', 'CreativeWorkSeries'): info['series'] = unescapeHTML(part_of_series.get('name')) - elif item_type == 'Movie': + elif is_type(e, 'Movie'): info.update({ 'title': unescapeHTML(e.get('name')), 'description': unescapeHTML(e.get('description')), 'duration': parse_duration(e.get('duration')), 'timestamp': unified_timestamp(e.get('dateCreated')), }) - elif item_type in ('Article', 'NewsArticle'): + elif is_type(e, 'Article', 'NewsArticle'): info.update({ 'timestamp': parse_iso8601(e.get('datePublished')), 'title': unescapeHTML(e.get('headline')), 'description': unescapeHTML(e.get('articleBody') or e.get('description')), }) - if traverse_obj(e, ('video', 0, '@type')) == 'VideoObject': + if is_type(traverse_obj(e, ('video', 0)), 'VideoObject'): extract_video_object(e['video'][0]) - elif item_type == 'VideoObject': + elif is_type(traverse_obj(e, ('subjectOf', 0)), 'VideoObject'): + extract_video_object(e['subjectOf'][0]) + elif is_type(e, 'VideoObject'): extract_video_object(e) if expected_type is None: continue else: break video = e.get('video') - if isinstance(video, dict) and video.get('@type') == 'VideoObject': + if is_type(video, 'VideoObject'): extract_video_object(video) if expected_type is None: continue @@ -1597,15 +1591,13 @@ class InfoExtractor: 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 + def _search_nuxt_data(self, webpage, video_id, context_name='__NUXT__', *, fatal=True, traverse=('data', 0)): + """Parses Nuxt.js metadata. This works as long as the function __NUXT__ invokes is a pure function""" rectx = re.escape(context_name) + FUNCTION_RE = r'\(function\((?P<arg_keys>.*?)\){return\s+(?P<js>{.*?})\s*;?\s*}\((?P<arg_vals>.*?)\)' 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']) + (rf'<script>\s*window\.{rectx}={FUNCTION_RE}\s*\)\s*;?\s*</script>', rf'{rectx}\(.*?{FUNCTION_RE}'), + webpage, context_name, group=('js', 'arg_keys', 'arg_vals'), fatal=fatal) args = dict(zip(arg_keys.split(','), arg_vals.split(','))) @@ -1613,7 +1605,8 @@ class InfoExtractor: if val in ('undefined', 'void 0'): args[key] = 'null' - return self._parse_json(js_to_json(js, args), video_id)['data'][0] + ret = self._parse_json(js, video_id, transform_source=functools.partial(js_to_json, vars=args), fatal=fatal) + return traverse_obj(ret, traverse) or {} @staticmethod def _hidden_inputs(html): @@ -2166,7 +2159,7 @@ class InfoExtractor: ]), m3u8_doc) def format_url(url): - return url if re.match(r'^https?://', url) else compat_urlparse.urljoin(m3u8_url, url) + return url if re.match(r'^https?://', url) else urllib.parse.urljoin(m3u8_url, url) if self.get_param('hls_split_discontinuity', False): def _extract_m3u8_playlist_indices(manifest_url=None, m3u8_doc=None): @@ -2539,7 +2532,7 @@ class InfoExtractor: }) continue - src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src) + src_url = src if src.startswith('http') else urllib.parse.urljoin(base, src) src_url = src_url.strip() if proto == 'm3u8' or src_ext == 'm3u8': @@ -2562,7 +2555,7 @@ class InfoExtractor: 'plugin': 'flowplayer-3.2.0.1', } f4m_url += '&' if '?' in f4m_url else '?' - f4m_url += compat_urllib_parse_urlencode(f4m_params) + f4m_url += urllib.parse.urlencode(f4m_params) formats.extend(self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False)) elif src_ext == 'mpd': formats.extend(self._extract_mpd_formats( @@ -2803,13 +2796,18 @@ class InfoExtractor: mime_type = representation_attrib['mimeType'] content_type = representation_attrib.get('contentType', mime_type.split('/')[0]) - codecs = parse_codecs(representation_attrib.get('codecs', '')) + codec_str = representation_attrib.get('codecs', '') + # Some kind of binary subtitle found in some youtube livestreams + if mime_type == 'application/x-rawcc': + codecs = {'scodec': codec_str} + else: + codecs = parse_codecs(codec_str) if content_type not in ('video', 'audio', 'text'): if mime_type == 'image/jpeg': content_type = mime_type - elif codecs['vcodec'] != 'none': + elif codecs.get('vcodec', 'none') != 'none': content_type = 'video' - elif codecs['acodec'] != 'none': + elif codecs.get('acodec', 'none') != 'none': content_type = 'audio' elif codecs.get('scodec', 'none') != 'none': content_type = 'text' @@ -2827,7 +2825,7 @@ class InfoExtractor: if re.match(r'^https?://', base_url): break if mpd_base_url and base_url.startswith('/'): - base_url = compat_urlparse.urljoin(mpd_base_url, base_url) + base_url = urllib.parse.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 += '/' @@ -3097,7 +3095,7 @@ class InfoExtractor: sampling_rate = int_or_none(track.get('SamplingRate')) track_url_pattern = re.sub(r'{[Bb]itrate}', track.attrib['Bitrate'], url_pattern) - track_url_pattern = compat_urlparse.urljoin(ism_url, track_url_pattern) + track_url_pattern = urllib.parse.urljoin(ism_url, track_url_pattern) fragments = [] fragment_ctx = { @@ -3116,7 +3114,7 @@ class InfoExtractor: fragment_ctx['duration'] = (next_fragment_time - fragment_ctx['time']) / fragment_repeat for _ in range(fragment_repeat): fragments.append({ - 'url': re.sub(r'{start[ _]time}', compat_str(fragment_ctx['time']), track_url_pattern), + 'url': re.sub(r'{start[ _]time}', str(fragment_ctx['time']), track_url_pattern), 'duration': fragment_ctx['duration'] / stream_timescale, }) fragment_ctx['time'] += fragment_ctx['duration'] @@ -3184,7 +3182,8 @@ class InfoExtractor: return f return {} - def _media_formats(src, cur_media_type, type_info={}): + def _media_formats(src, cur_media_type, type_info=None): + type_info = type_info or {} full_url = absolute_url(src) ext = type_info.get('ext') or determine_ext(full_url) if ext == 'm3u8': @@ -3202,6 +3201,7 @@ class InfoExtractor: formats = [{ 'url': full_url, 'vcodec': 'none' if cur_media_type == 'audio' else None, + 'ext': ext, }] return is_plain_url, formats @@ -3228,7 +3228,8 @@ class InfoExtractor: media_attributes = extract_attributes(media_tag) src = strip_or_none(media_attributes.get('src')) if src: - _, formats = _media_formats(src, media_type) + f = parse_content_type(media_attributes.get('type')) + _, formats = _media_formats(src, media_type, f) media_info['formats'].extend(formats) media_info['thumbnail'] = absolute_url(media_attributes.get('poster')) if media_content: @@ -3357,7 +3358,7 @@ class InfoExtractor: return formats, subtitles def _extract_wowza_formats(self, url, video_id, m3u8_entry_protocol='m3u8_native', skip_protocols=[]): - query = compat_urlparse.urlparse(url).query + query = urllib.parse.urlparse(url).query url = re.sub(r'/(?:manifest|playlist|jwplayer)\.(?:m3u8|f4m|mpd|smil)', '', url) mobj = re.search( r'(?:(?:http|rtmp|rtsp)(?P<s>s)?:)?(?P<url>//[^?]+)', url) @@ -3463,7 +3464,7 @@ class InfoExtractor: if not isinstance(track, dict): continue track_kind = track.get('kind') - if not track_kind or not isinstance(track_kind, compat_str): + if not track_kind or not isinstance(track_kind, str): continue if track_kind.lower() not in ('captions', 'subtitles'): continue @@ -3536,7 +3537,7 @@ class InfoExtractor: # Often no height is provided but there is a label in # format like "1080p", "720p SD", or 1080. height = int_or_none(self._search_regex( - r'^(\d{3,4})[pP]?(?:\b|$)', compat_str(source.get('label') or ''), + r'^(\d{3,4})[pP]?(?:\b|$)', str(source.get('label') or ''), 'height', default=None)) a_format = { 'url': source_url, @@ -3588,17 +3589,15 @@ class InfoExtractor: def _set_cookie(self, domain, name, value, expire_time=None, port=None, path='/', secure=False, discard=False, rest={}, **kwargs): - cookie = compat_cookiejar_Cookie( + cookie = http.cookiejar.Cookie( 0, name, value, port, port is not None, domain, True, domain.startswith('.'), path, True, secure, expire_time, discard, None, None, rest) - self._downloader.cookiejar.set_cookie(cookie) + self.cookiejar.set_cookie(cookie) def _get_cookies(self, url): - """ Return a compat_cookies_SimpleCookie with the cookies for the url """ - req = sanitized_Request(url) - self._downloader.cookiejar.add_cookie_header(req) - return compat_cookies_SimpleCookie(req.get_header('Cookie')) + """ Return a http.cookies.SimpleCookie with the cookies for the url """ + return http.cookies.SimpleCookie(self._downloader._calc_cookies(url)) def _apply_first_set_cookie_header(self, url_handle, cookie): """ @@ -3742,7 +3741,7 @@ class InfoExtractor: def _get_automatic_captions(self, *args, **kwargs): raise NotImplementedError('This method must be implemented by subclasses') - @property + @functools.cached_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 @@ -3764,10 +3763,10 @@ class InfoExtractor: return headers def _generic_id(self, url): - return compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0]) + return urllib.parse.unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0]) def _generic_title(self, url): - return compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]) + return urllib.parse.unquote(os.path.splitext(url_basename(url))[0]) @staticmethod def _availability(is_private=None, needs_premium=None, needs_subscription=None, needs_auth=None, is_unlisted=None): |