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,
)
))
|youtu\.be/ # just youtu.be/xxxx
+ |https?://(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
)
)? # all until now is optional -> you can pass the naked ID
([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
'141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50},
# Dash webm
- '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},
+ '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
+ '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
+ '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
+ '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
+ '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
+ '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
+ '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+ '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+ '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+ '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+ '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+ '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+ '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+ '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+ '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
# Dash webm audio
- '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH webm audio', 'abr': 48, 'preference': -50},
- '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH webm audio', 'abr': 256, 'preference': -50},
+ '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 48, 'preference': -50},
+ '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
# RTMP (unnamed)
'_rtmp': {'protocol': 'rtmp'},
u"uploader": u"Philipp Hagemeister",
u"uploader_id": u"phihag",
u"upload_date": u"20121002",
- u"description": u"test chars: \"'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ."
+ u"description": u"test chars: \"'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .",
+ u"categories": [u'Science & Technology'],
}
},
{
u"info_dict": {
u"upload_date": u"20120506",
u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
- u"description": u"md5:5b292926389560516e384ac437c0ec07",
+ u"description": u"md5:fea86fda2d5a5784273df5c7cc994d9f",
u"uploader": u"Icona Pop",
u"uploader_id": u"IconaPop"
}
u'id': u'IB3lcPjvWLA',
u'ext': u'm4a',
u'title': u'Afrojack - The Spark ft. Spree Wilson',
- u'description': u'md5:3199ed45ee8836572865580804d7ac0f',
+ u'description': u'md5:9717375db5a9a3992be4668bbf3bc0a8',
u'uploader': u'AfrojackVEVO',
u'uploader_id': u'AfrojackVEVO',
u'upload_date': u'20131011',
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)
+ r'signature=([$a-zA-Z]+)', jscode,
+ 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):
break
if 'token' not in video_info:
if 'reason' in video_info:
- raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True)
+ raise ExtractorError(
+ u'YouTube said: %s' % video_info['reason'][0],
+ expected=True, video_id=video_id)
else:
- raise ExtractorError(u'"token" parameter not in video info for unknown reason')
+ raise ExtractorError(
+ u'"token" parameter not in video info for unknown reason',
+ video_id=video_id)
if 'view_count' in video_info:
view_count = int(video_info['view_count'][0])
# title
if 'title' in video_info:
- video_title = compat_urllib_parse.unquote_plus(video_info['title'][0])
+ video_title = video_info['title'][0]
else:
self._downloader.report_warning(u'Unable to extract video title')
video_title = u'_'
# upload date
upload_date = None
- mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL)
+ mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
+ if mobj is None:
+ mobj = re.search(
+ r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
+ video_webpage)
if mobj is not None:
upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
upload_date = unified_strdate(upload_date)
+ m_cat_container = get_element_by_id("eow-category", video_webpage)
+ if m_cat_container:
+ category = self._html_search_regex(
+ r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
+ default=None)
+ video_categories = None if category is None else [category]
+ else:
+ video_categories = None
+
# description
video_description = get_element_by_id("eow-description", video_webpage)
if video_description:
'title': video_title,
'thumbnail': video_thumbnail,
'description': video_description,
+ 'categories': video_categories,
'subtitles': video_subtitles,
'duration': video_duration,
'age_limit': 18 if age_gate else 0,
| p/
)
(
- (?:PL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
+ (?:PL|LL|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,})
+ ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
)"""
_TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
_MORE_PAGES_INDICATOR = r'data-link-type="next"'
title_span = (search_title('playlist-title') or
search_title('title long-title') or search_title('title'))
title = clean_html(title_span)
- video_re = r'''(?x)data-video-username="(.*?)".*?
+ 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]
+ ids = orderedSet(re.findall(video_re, webpage, flags=re.DOTALL))
url_results = self._ids_to_results(ids)
return self.playlist_result(url_results, playlist_id, title)
self.to_screen(u'Downloading just video %s because of --no-playlist' % video_id)
return self.url_result(video_id, 'Youtube', video_id=video_id)
else:
- self.to_screen(u'Downloading playlist PL%s - add --no-playlist to just download video %s' % (playlist_id, video_id))
+ self.to_screen(u'Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
if playlist_id.startswith('RD'):
# Mixes require a custom extraction process
page = self._download_webpage(url, playlist_id)
more_widget_html = content_html = page
+ # Check if the playlist exists or is private
+ if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None:
+ raise ExtractorError(
+ u'The playlist doesn\'t exist or is private, use --username or '
+ '--netrc to access it.',
+ expected=True)
+
# Extract the video ids from the playlist pages
ids = []
break
more = self._download_json(
- 'https://youtube.com/%s' % mobj.group('more'), playlist_id, 'Downloading page #%s' % page_num)
+ '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'<h1 class="pl-header-title">\s*(.*?)\s*</h1>', page, u'title')
+ 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)
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),
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(
self.url_result(video_id, 'Youtube', video_id=video_id)
for video_id in ids)
- if info['paging'] is None:
+ mobj = re.search(
+ r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
+ feed_html)
+ if mobj is None:
break
- paging = info['paging']
+ paging = mobj.group('paging')
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'
IE_NAME = 'youtube:truncated_url'
IE_DESC = False # Do not list
_VALID_URL = r'''(?x)
- (?:https?://)?[^/]+/watch\?(?:feature=[a-z_]+)?$|
+ (?:https?://)?[^/]+/watch\?(?:
+ feature=[a-z_]+|
+ annotation_id=annotation_[^&]+
+ )?$|
(?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
'''
+ _TESTS = [{
+ 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
+ 'only_matching': True,
+ }, {
+ 'url': 'http://www.youtube.com/watch?',
+ 'only_matching': True,
+ }]
+
def _real_extract(self, url):
raise ExtractorError(
u'Did you forget to quote the URL? Remember that & is a meta '
projects / youtube-dl / blobdiff