import json
import os.path
import re
-import string
import struct
import traceback
import zlib
from .common import InfoExtractor, SearchInfoExtractor
from .subtitles import SubtitlesInfoExtractor
+from ..jsinterp import JSInterpreter
from ..utils import (
compat_chr,
compat_parse_qs,
get_element_by_id,
get_element_by_attribute,
ExtractorError,
+ int_or_none,
PagedList,
- RegexNotFoundError,
unescapeHTML,
unified_strdate,
orderedSet,
write_json_file,
+ uppercase_escape,
)
class YoutubeBaseInfoExtractor(InfoExtractor):
"""Provide base functions for Youtube extractors"""
_LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
_LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
- _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
+ _AGE_URL = 'https://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
_NETRC_MACHINE = 'youtube'
# If True it will raise an error if no login info is provided
_LOGIN_REQUIRED = False
'next_url': '/',
'action_confirm': 'Confirm',
}
- req = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
+ req = compat_urllib_request.Request(self._AGE_URL,
+ compat_urllib_parse.urlencode(age_form).encode('ascii'))
self._download_webpage(
req, None,
(?:https?://|//)? # http(s):// or protocol-independent URL (optional)
(?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
(?:www\.)?deturl\.com/www\.youtube\.com/|
- (?:www\.)?pwnyoutube\.com|
+ (?:www\.)?pwnyoutube\.com/|
+ (?:www\.)?yourepeat\.com/|
tube\.majestyc\.net/|
youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
(?:.*?\#/)? # handle anchor (#/) redirect urls
(?: # the various things that can precede the ID:
(?:(?:v|embed|e)/) # v/ or embed/ or e/
|(?: # or the v= param in all its forms
- (?:(?:watch|movie)(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
+ (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
(?:\?|\#!?) # the params delimiter ? or # or #!
(?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
v=
# 3d videos
- '82': {'ext': 'mp4', 'height': 360, 'resolution': '360p', 'format_note': '3D', 'preference': -20},
- '83': {'ext': 'mp4', 'height': 480, 'resolution': '480p', 'format_note': '3D', 'preference': -20},
- '84': {'ext': 'mp4', 'height': 720, 'resolution': '720p', 'format_note': '3D', 'preference': -20},
- '85': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': '3D', 'preference': -20},
- '100': {'ext': 'webm', 'height': 360, 'resolution': '360p', 'format_note': '3D', 'preference': -20},
- '101': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': '3D', 'preference': -20},
- '102': {'ext': 'webm', 'height': 720, 'resolution': '720p', 'format_note': '3D', 'preference': -20},
+ '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
+ '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
+ '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
+ '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
+ '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
+ '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
+ '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
# Apple HTTP Live Streaming
- '92': {'ext': 'mp4', 'height': 240, 'resolution': '240p', 'format_note': 'HLS', 'preference': -10},
- '93': {'ext': 'mp4', 'height': 360, 'resolution': '360p', 'format_note': 'HLS', 'preference': -10},
- '94': {'ext': 'mp4', 'height': 480, 'resolution': '480p', 'format_note': 'HLS', 'preference': -10},
- '95': {'ext': 'mp4', 'height': 720, 'resolution': '720p', 'format_note': 'HLS', 'preference': -10},
- '96': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': 'HLS', 'preference': -10},
- '132': {'ext': 'mp4', 'height': 240, 'resolution': '240p', 'format_note': 'HLS', 'preference': -10},
- '151': {'ext': 'mp4', 'height': 72, 'resolution': '72p', 'format_note': 'HLS', 'preference': -10},
+ '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
+ '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
+ '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
+ '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
+ '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
+ '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
+ '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
# DASH mp4 video
- '133': {'ext': 'mp4', 'height': 240, 'resolution': '240p', 'format_note': 'DASH video', 'preference': -40},
- '134': {'ext': 'mp4', 'height': 360, 'resolution': '360p', 'format_note': 'DASH video', 'preference': -40},
- '135': {'ext': 'mp4', 'height': 480, 'resolution': '480p', 'format_note': 'DASH video', 'preference': -40},
- '136': {'ext': 'mp4', 'height': 720, 'resolution': '720p', 'format_note': 'DASH video', 'preference': -40},
- '137': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': 'DASH video', 'preference': -40},
- '138': {'ext': 'mp4', 'height': 1081, 'resolution': '>1080p', 'format_note': 'DASH video', 'preference': -40},
- '160': {'ext': 'mp4', 'height': 192, 'resolution': '192p', 'format_note': 'DASH video', 'preference': -40},
- '264': {'ext': 'mp4', 'height': 1080, 'resolution': '1080p', 'format_note': 'DASH video', 'preference': -40},
+ '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+ '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+ '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+ '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+ '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+ '138': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+ '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+ '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
# Dash mp4 audio
'139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50},
'141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50},
# Dash webm
- '242': {'ext': 'webm', 'height': 240, 'resolution': '240p', 'format_note': 'DASH webm', 'preference': -40},
- '243': {'ext': 'webm', 'height': 360, 'resolution': '360p', 'format_note': 'DASH webm', 'preference': -40},
- '244': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': 'DASH webm', 'preference': -40},
- '245': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': 'DASH webm', 'preference': -40},
- '246': {'ext': 'webm', 'height': 480, 'resolution': '480p', 'format_note': 'DASH webm', 'preference': -40},
- '247': {'ext': 'webm', 'height': 720, 'resolution': '720p', 'format_note': 'DASH webm', 'preference': -40},
- '248': {'ext': 'webm', 'height': 1080, 'resolution': '1080p', 'format_note': 'DASH webm', 'preference': -40},
+ '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40},
+ '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40},
+ '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40},
+ '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40},
+ '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40},
+ '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'acodec': 'none', 'preference': -40},
+ '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH webm', 'preference': -40},
+ '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH webm', 'preference': -40},
+ '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH webm', 'preference': -40},
+ '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH webm', 'preference': -40},
+ '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH webm', 'preference': -40},
+ '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH webm', 'preference': -40},
+ '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH webm', 'preference': -40},
# Dash webm audio
'171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH webm audio', 'abr': 48, 'preference': -50},
u"uploader_id": u"setindia"
}
},
+ {
+ u"url": u"http://www.youtube.com/watch?v=a9LDPn-MO4I",
+ u"file": u"a9LDPn-MO4I.m4a",
+ u"note": u"256k DASH audio (format 141) via DASH manifest",
+ u"info_dict": {
+ u"upload_date": "20121002",
+ u"uploader_id": "8KVIDEO",
+ u"description": "No description available.",
+ u"uploader": "8KVIDEO",
+ u"title": "UHDTV TEST 8K VIDEO.mp4"
+ },
+ u"params": {
+ u"youtube_include_dash_manifest": True,
+ u"format": "141",
+ },
+ },
+ # DASH manifest with encrypted signature
+ {
+ u'url': u'https://www.youtube.com/watch?v=IB3lcPjvWLA',
+ u'info_dict': {
+ u'id': u'IB3lcPjvWLA',
+ u'ext': u'm4a',
+ u'title': u'Afrojack - The Spark ft. Spree Wilson',
+ u'description': u'md5:3199ed45ee8836572865580804d7ac0f',
+ u'uploader': u'AfrojackVEVO',
+ u'uploader_id': u'AfrojackVEVO',
+ u'upload_date': u'20131011',
+ },
+ u"params": {
+ u'youtube_include_dash_manifest': True,
+ u'format': '141',
+ },
+ },
]
def _parse_sig_js(self, jscode):
funcname = self._search_regex(
r'signature=([a-zA-Z]+)', jscode,
- u'Initial JS player signature function name')
-
- functions = {}
-
- def argidx(varname):
- return string.lowercase.index(varname)
-
- def interpret_statement(stmt, local_vars, allow_recursion=20):
- if allow_recursion < 0:
- raise ExtractorError(u'Recursion limit reached')
-
- if stmt.startswith(u'var '):
- stmt = stmt[len(u'var '):]
- ass_m = re.match(r'^(?P<out>[a-z]+)(?:\[(?P<index>[^\]]+)\])?' +
- r'=(?P<expr>.*)$', stmt)
- if ass_m:
- if ass_m.groupdict().get('index'):
- def assign(val):
- lvar = local_vars[ass_m.group('out')]
- idx = interpret_expression(ass_m.group('index'),
- local_vars, allow_recursion)
- assert isinstance(idx, int)
- lvar[idx] = val
- return val
- expr = ass_m.group('expr')
- else:
- def assign(val):
- local_vars[ass_m.group('out')] = val
- return val
- expr = ass_m.group('expr')
- elif stmt.startswith(u'return '):
- assign = lambda v: v
- expr = stmt[len(u'return '):]
- else:
- raise ExtractorError(
- u'Cannot determine left side of statement in %r' % stmt)
-
- v = interpret_expression(expr, local_vars, allow_recursion)
- return assign(v)
-
- def interpret_expression(expr, local_vars, allow_recursion):
- if expr.isdigit():
- return int(expr)
-
- if expr.isalpha():
- return local_vars[expr]
-
- m = re.match(r'^(?P<in>[a-z]+)\.(?P<member>.*)$', expr)
- if m:
- member = m.group('member')
- val = local_vars[m.group('in')]
- if member == 'split("")':
- return list(val)
- if member == 'join("")':
- return u''.join(val)
- if member == 'length':
- return len(val)
- if member == 'reverse()':
- return val[::-1]
- slice_m = re.match(r'slice\((?P<idx>.*)\)', member)
- if slice_m:
- idx = interpret_expression(
- slice_m.group('idx'), local_vars, allow_recursion-1)
- return val[idx:]
-
- m = re.match(
- r'^(?P<in>[a-z]+)\[(?P<idx>.+)\]$', expr)
- if m:
- val = local_vars[m.group('in')]
- idx = interpret_expression(m.group('idx'), local_vars,
- allow_recursion-1)
- return val[idx]
-
- m = re.match(r'^(?P<a>.+?)(?P<op>[%])(?P<b>.+?)$', expr)
- if m:
- a = interpret_expression(m.group('a'),
- local_vars, allow_recursion)
- b = interpret_expression(m.group('b'),
- local_vars, allow_recursion)
- return a % b
-
- m = re.match(
- r'^(?P<func>[a-zA-Z]+)\((?P<args>[a-z0-9,]+)\)$', expr)
- if m:
- fname = m.group('func')
- if fname not in functions:
- functions[fname] = extract_function(fname)
- argvals = [int(v) if v.isdigit() else local_vars[v]
- for v in m.group('args').split(',')]
- return functions[fname](argvals)
- raise ExtractorError(u'Unsupported JS expression %r' % expr)
-
- def extract_function(funcname):
- func_m = re.search(
- r'function ' + re.escape(funcname) +
- r'\((?P<args>[a-z,]+)\){(?P<code>[^}]+)}',
- jscode)
- argnames = func_m.group('args').split(',')
-
- def resf(args):
- local_vars = dict(zip(argnames, args))
- for stmt in func_m.group('code').split(';'):
- res = interpret_statement(stmt, local_vars)
- return res
- return resf
-
- initial_function = extract_function(funcname)
+ u'Initial JS player signature function name')
+
+ jsi = JSInterpreter(jscode)
+ initial_function = jsi.extract_function(funcname)
return lambda s: initial_function([s])
def _parse_sig_swf(self, file_contents):
def _get_available_subtitles(self, video_id, webpage):
try:
sub_list = self._download_webpage(
- 'http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
+ 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
video_id, note=False)
except ExtractorError as err:
self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
'fmt': self._downloader.params.get('subtitlesformat', 'srt'),
'name': unescapeHTML(l[0]).encode('utf-8'),
})
- url = u'http://www.youtube.com/api/timedtext?' + params
+ url = u'https://www.youtube.com/api/timedtext?' + params
sub_lang_list[lang] = url
if not sub_lang_list:
self._downloader.report_warning(u'video doesn\'t have subtitles')
self._downloader.report_warning(err_msg)
return {}
- def _extract_id(self, url):
- mobj = re.match(self._VALID_URL, url, re.VERBOSE)
+ @classmethod
+ def extract_id(cls, url):
+ mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
video_id = mobj.group(2)
return video_id
- def _get_video_url_list(self, url_map):
- """
- Transform a dictionary in the format {itag:url} to a list of (itag, url)
- with the requested formats.
- """
- existing_formats = [x for x in self._formats if x in url_map]
- if len(existing_formats) == 0:
- raise ExtractorError(u'no known formats available for video')
- video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
- video_url_list.reverse() # order worst to best
- return video_url_list
-
def _extract_from_m3u8(self, manifest_url, video_id):
url_map = {}
def _get_urls(_manifest):
return self._download_webpage(url, video_id, note=u'Searching for annotations.', errnote=u'Unable to download video annotations.')
def _real_extract(self, url):
+ proto = (
+ u'http' if self._downloader.params.get('prefer_insecure', False)
+ else u'https')
+
# Extract original video URL from URL with redirection, like age verification, using next_url parameter
mobj = re.search(self._NEXT_URL_RE, url)
if mobj:
- url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
- video_id = self._extract_id(url)
+ url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
+ video_id = self.extract_id(url)
# Get video webpage
- url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
+ url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
video_webpage = self._download_webpage(url, video_id)
# Attempt to extract SWF player URL
'asv': 3,
'sts':'1588',
})
- video_info_url = 'https://www.youtube.com/get_video_info?' + data
+ video_info_url = proto + '://www.youtube.com/get_video_info?' + data
video_info_webpage = self._download_webpage(video_info_url, video_id,
note=False,
errnote='unable to download video info webpage')
else:
age_gate = False
for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
- video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
+ video_info_url = (proto + '://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
% (video_id, el_type))
video_info_webpage = self._download_webpage(video_info_url, video_id,
note=False,
video_annotations = self._extract_annotations(video_id)
# Decide which formats to download
-
try:
- mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage)
+ mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
if not mobj:
raise ValueError('Could not find vevo ID')
- info = json.loads(mobj.group(1))
- args = info['args']
+ json_code = uppercase_escape(mobj.group(1))
+ ytplayer_config = json.loads(json_code)
+ args = ytplayer_config['args']
# Easy way to know if the 's' value is in url_encoded_fmt_stream_map
# this signatures are encrypted
if 'url_encoded_fmt_stream_map' not in args:
except ValueError:
pass
+ def _map_to_format_list(urlmap):
+ formats = []
+ for itag, video_real_url in urlmap.items():
+ dct = {
+ 'format_id': itag,
+ 'url': video_real_url,
+ 'player_url': player_url,
+ }
+ if itag in self._formats:
+ dct.update(self._formats[itag])
+ formats.append(dct)
+ return formats
+
if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
self.report_rtmp_download()
- video_url_list = [('_rtmp', video_info['conn'][0])]
+ formats = [{
+ 'format_id': '_rtmp',
+ 'protocol': 'rtmp',
+ 'url': video_info['conn'][0],
+ 'player_url': player_url,
+ }]
elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1:
encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts',[''])[0]
if 'rtmpe%3Dyes' in encoded_url_map:
if 'ratebypass' not in url:
url += '&ratebypass=yes'
url_map[url_data['itag'][0]] = url
- video_url_list = self._get_video_url_list(url_map)
+ formats = _map_to_format_list(url_map)
elif video_info.get('hlsvp'):
manifest_url = video_info['hlsvp'][0]
url_map = self._extract_from_m3u8(manifest_url, video_id)
- video_url_list = self._get_video_url_list(url_map)
+ formats = _map_to_format_list(url_map)
else:
raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
- formats = []
- for itag, video_real_url in video_url_list:
- dct = {
- 'format_id': itag,
- 'url': video_real_url,
- 'player_url': player_url,
- }
- dct.update(self._formats[itag])
- formats.append(dct)
+ # Look for the DASH manifest
+ if (self._downloader.params.get('youtube_include_dash_manifest', False)):
+ try:
+ # The DASH manifest used needs to be the one from the original video_webpage.
+ # The one found in get_video_info seems to be using different signatures.
+ # However, in the case of an age restriction there won't be any embedded dashmpd in the video_webpage.
+ # Luckily, it seems, this case uses some kind of default signature (len == 86), so the
+ # combination of get_video_info and the _static_decrypt_signature() decryption fallback will work here.
+ if age_gate:
+ dash_manifest_url = video_info.get('dashmpd')[0]
+ else:
+ dash_manifest_url = ytplayer_config['args']['dashmpd']
+ def decrypt_sig(mobj):
+ s = mobj.group(1)
+ dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
+ return '/signature/%s' % dec_s
+ dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
+ dash_doc = self._download_xml(
+ dash_manifest_url, video_id,
+ note=u'Downloading DASH manifest',
+ errnote=u'Could not download DASH manifest')
+ for r in dash_doc.findall(u'.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
+ url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
+ if url_el is None:
+ continue
+ format_id = r.attrib['id']
+ video_url = url_el.text
+ filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
+ f = {
+ 'format_id': format_id,
+ 'url': video_url,
+ 'width': int_or_none(r.attrib.get('width')),
+ 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
+ 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
+ 'filesize': filesize,
+ }
+ try:
+ existing_format = next(
+ fo for fo in formats
+ if fo['format_id'] == format_id)
+ except StopIteration:
+ f.update(self._formats.get(format_id, {}))
+ formats.append(f)
+ else:
+ existing_format.update(f)
+
+ except (ExtractorError, KeyError) as e:
+ self.report_warning(u'Skipping DASH manifest: %s' % e, video_id)
self._sort_formats(formats)
'duration': video_duration,
'age_limit': 18 if age_gate else 0,
'annotations': video_annotations,
- 'webpage_url': 'https://www.youtube.com/watch?v=%s' % video_id,
+ 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
'view_count': view_count,
'like_count': like_count,
'dislike_count': dislike_count,
class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
IE_DESC = u'YouTube.com playlists'
- _VALID_URL = r"""(?:
+ _VALID_URL = r"""(?x)(?:
(?:https?://)?
(?:\w+\.)?
youtube\.com/
\? (?:.*?&)*? (?:p|a|list)=
| p/
)
- ((?:PL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,})
+ (
+ (?:PL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
+ # Top tracks, they can also include dots
+ |(?:MC)[\w\.]*
+ )
.*
|
((?:PL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
)"""
- _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&page=%s'
+ _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
_MORE_PAGES_INDICATOR = r'data-link-type="next"'
- _VIDEO_RE = r'href="/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
+ _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
IE_NAME = u'youtube:playlist'
- @classmethod
- def suitable(cls, url):
- """Receives a URL and returns True if suitable for this IE."""
- return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
-
def _real_initialize(self):
self._login()
# the id of the playlist is just 'RD' + video_id
url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
webpage = self._download_webpage(url, playlist_id, u'Downloading Youtube mix')
- title_span = (get_element_by_attribute('class', 'title long-title', webpage) or
- get_element_by_attribute('class', 'title ', webpage))
+ search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
+ title_span = (search_title('playlist-title') or
+ search_title('title long-title') or search_title('title'))
title = clean_html(title_span)
- video_re = r'data-index="\d+".*?href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s' % re.escape(playlist_id)
- ids = orderedSet(re.findall(video_re, webpage))
+ video_re = r'''(?x)data-video-username="(.*?)".*?
+ href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id)
+ matches = orderedSet(re.findall(video_re, webpage, flags=re.DOTALL))
+ # Some of the videos may have been deleted, their username field is empty
+ ids = [video_id for (username, video_id) in matches if username]
url_results = self._ids_to_results(ids)
return self.playlist_result(url_results, playlist_id, title)
def _real_extract(self, url):
# Extract playlist id
- mobj = re.match(self._VALID_URL, url, re.VERBOSE)
+ mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
playlist_id = mobj.group(1) or mobj.group(2)
raise ExtractorError(u'For downloading YouTube.com top lists, use '
u'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True)
+ url = self._TEMPLATE_URL % playlist_id
+ page = self._download_webpage(url, playlist_id)
+ more_widget_html = content_html = page
+
# Extract the video ids from the playlist pages
ids = []
for page_num in itertools.count(1):
- url = self._TEMPLATE_URL % (playlist_id, page_num)
- page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
- matches = re.finditer(self._VIDEO_RE, page)
+ matches = re.finditer(self._VIDEO_RE, content_html)
# We remove the duplicates and the link with index 0
# (it's not the first video of the playlist)
new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
ids.extend(new_ids)
- if re.search(self._MORE_PAGES_INDICATOR, page) is None:
+ mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
+ if not mobj:
break
- try:
- playlist_title = self._og_search_title(page)
- except RegexNotFoundError:
- self.report_warning(
- u'Playlist page is missing OpenGraph title, falling back ...',
- playlist_id)
- playlist_title = self._html_search_regex(
- r'<h1 class="pl-header-title">(.*?)</h1>', page, u'title')
+ more = self._download_json(
+ 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
+ 'Downloading page #%s' % page_num,
+ transform_source=uppercase_escape)
+ content_html = more['content_html']
+ more_widget_html = more['load_more_widget_html']
+
+ playlist_title = self._html_search_regex(
+ r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
+ page, u'title')
url_results = self._ids_to_results(ids)
return self.playlist_result(url_results, playlist_id, playlist_title)
channel = mobj.group('chann')
title = mobj.group('title')
query = compat_urllib_parse.urlencode({'title': title})
- playlist_re = 'href="([^"]+?%s[^"]+?)"' % re.escape(query)
+ playlist_re = 'href="([^"]+?%s.*?)"' % re.escape(query)
channel_page = self._download_webpage('https://www.youtube.com/%s' % channel, title)
link = self._html_search_regex(playlist_re, channel_page, u'list')
url = compat_urlparse.urljoin('https://www.youtube.com/', link)
IE_DESC = u'YouTube.com channels'
_VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
_MORE_PAGES_INDICATOR = 'yt-uix-load-more'
- _MORE_PAGES_URL = 'http://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s'
+ _MORE_PAGES_URL = 'https://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s'
IE_NAME = u'youtube:channel'
def extract_videos_from_page(self, page):
# Download all channel pages using the json-based channel_ajax query
for pagenum in itertools.count(1):
url = self._MORE_PAGES_URL % (pagenum, channel_id)
- page = self._download_webpage(url, channel_id,
- u'Downloading page #%s' % pagenum)
-
- page = json.loads(page)
-
+ page = self._download_json(
+ url, channel_id, note=u'Downloading page #%s' % pagenum,
+ transform_source=uppercase_escape)
+
ids_in_page = self.extract_videos_from_page(page['content_html'])
video_ids.extend(ids_in_page)
class YoutubeUserIE(InfoExtractor):
IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
- _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
- _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
+ _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
+ _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s'
_GDATA_PAGE_SIZE = 50
- _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
+ _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
IE_NAME = u'youtube:user'
@classmethod
'_type': 'url',
'url': video_id,
'ie_key': 'Youtube',
- 'id': 'video_id',
+ 'id': video_id,
'title': title,
}
url_results = PagedList(download_page, self._GDATA_PAGE_SIZE)
class YoutubeSearchIE(SearchInfoExtractor):
IE_DESC = u'YouTube.com searches'
- _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
+ _API_URL = u'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
_MAX_RESULTS = 1000
IE_NAME = u'youtube:search'
_SEARCH_KEY = 'ytsearch'
video_ids = []
pagenum = 0
limit = n
+ PAGE_SIZE = 50
- while (50 * pagenum) < limit:
- result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
+ while (PAGE_SIZE * pagenum) < limit:
+ result_url = self._API_URL % (
+ compat_urllib_parse.quote_plus(query.encode('utf-8')),
+ (PAGE_SIZE * pagenum) + 1)
data_json = self._download_webpage(
result_url, video_id=u'query "%s"' % query,
note=u'Downloading page %s' % (pagenum + 1),
api_response = data['data']
if 'items' not in api_response:
- raise ExtractorError(u'[youtube] No video results')
+ raise ExtractorError(
+ u'[youtube] No video results', expected=True)
new_ids = list(video['id'] for video in api_response['items'])
video_ids += new_ids
for video_id in video_ids]
return self.playlist_result(videos, query)
+
class YoutubeSearchDateIE(YoutubeSearchIE):
IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
_API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
_SEARCH_KEY = 'ytsearchdate'
IE_DESC = u'YouTube.com searches, newest videos first'
+
+class YoutubeSearchURLIE(InfoExtractor):
+ IE_DESC = u'YouTube.com search URLs'
+ IE_NAME = u'youtube:search_url'
+ _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
+
+ def _real_extract(self, url):
+ mobj = re.match(self._VALID_URL, url)
+ query = compat_urllib_parse.unquote_plus(mobj.group('query'))
+
+ webpage = self._download_webpage(url, query)
+ result_code = self._search_regex(
+ r'(?s)<ol id="search-results"(.*?)</ol>', webpage, u'result HTML')
+
+ part_codes = re.findall(
+ r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
+ entries = []
+ for part_code in part_codes:
+ part_title = self._html_search_regex(
+ r'(?s)title="([^"]+)"', part_code, 'item title', fatal=False)
+ part_url_snippet = self._html_search_regex(
+ r'(?s)href="([^"]+)"', part_code, 'item URL')
+ part_url = compat_urlparse.urljoin(
+ 'https://www.youtube.com/', part_url_snippet)
+ entries.append({
+ '_type': 'url',
+ 'url': part_url,
+ 'title': part_title,
+ })
+
+ return {
+ '_type': 'playlist',
+ 'entries': entries,
+ 'title': query,
+ }
+
+
class YoutubeShowIE(InfoExtractor):
IE_DESC = u'YouTube.com (multi-season) shows'
_VALID_URL = r'https?://www\.youtube\.com/show/(.*)'
action = 'action_load_system_feed'
if self._PERSONAL_FEED:
action = 'action_load_personal_feed'
- return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
+ return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
@property
def IE_NAME(self):
feed_entries = []
paging = 0
for i in itertools.count(1):
- info = self._download_webpage(self._FEED_TEMPLATE % paging,
+ info = self._download_json(self._FEED_TEMPLATE % paging,
u'%s feed' % self._FEED_NAME,
u'Downloading page %s' % i)
- info = json.loads(info)
- feed_html = info['feed_html']
+ feed_html = info.get('feed_html') or info.get('content_html')
m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
ids = orderedSet(m.group(1) for m in m_ids)
feed_entries.extend(
return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
- IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
+ IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
_VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
_FEED_NAME = 'subscriptions'
_PLAYLIST_TITLE = u'Youtube Subscriptions'
class YoutubeTruncatedURLIE(InfoExtractor):
IE_NAME = 'youtube:truncated_url'
IE_DESC = False # Do not list
- _VALID_URL = r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$'
+ _VALID_URL = r'''(?x)
+ (?:https?://)?[^/]+/watch\?(?:feature=[a-z_]+)?$|
+ (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
+ '''
def _real_extract(self, url):
raise ExtractorError(
projects / youtube-dl / blobdiff