import json
import os.path
import re
-import socket
import string
import struct
import traceback
-import xml.etree.ElementTree
import zlib
from .common import InfoExtractor, SearchInfoExtractor
from .subtitles import SubtitlesInfoExtractor
from ..utils import (
compat_chr,
- compat_http_client,
compat_parse_qs,
- compat_urllib_error,
compat_urllib_parse,
compat_urllib_request,
compat_urlparse,
clean_html,
get_cachedir,
get_element_by_id,
+ get_element_by_attribute,
ExtractorError,
unescapeHTML,
unified_strdate,
# If True it will raise an error if no login info is provided
_LOGIN_REQUIRED = False
- def report_lang(self):
- """Report attempt to set language."""
- self.to_screen(u'Setting language')
-
def _set_language(self):
- request = compat_urllib_request.Request(self._LANG_URL)
- try:
- self.report_lang()
- compat_urllib_request.urlopen(request).read()
- except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
- self._downloader.report_warning(u'unable to set language: %s' % compat_str(err))
- return False
- return True
+ return bool(self._download_webpage(
+ self._LANG_URL, None,
+ note=u'Setting language', errnote='unable to set language',
+ fatal=False))
def _login(self):
(username, password) = self._get_login_info()
raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True)
return False
- request = compat_urllib_request.Request(self._LOGIN_URL)
- try:
- login_page = compat_urllib_request.urlopen(request).read().decode('utf-8')
- except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
- self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err))
- return False
+ login_page = self._download_webpage(
+ self._LOGIN_URL, None,
+ note=u'Downloading login page',
+ errnote=u'unable to fetch login page', fatal=False)
+ if login_page is False:
+ return
galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
login_page, u'Login GALX parameter')
# chokes on unicode
login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
- request = compat_urllib_request.Request(self._LOGIN_URL, login_data)
- try:
- self.report_login()
- login_results = compat_urllib_request.urlopen(request).read().decode('utf-8')
- if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
- self._downloader.report_warning(u'unable to log in: bad username or password')
- return False
- except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
- self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
+
+ req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
+ login_results = self._download_webpage(
+ req, None,
+ note=u'Logging in', errnote=u'unable to log in', fatal=False)
+ if login_results is False:
+ return False
+ if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
+ self._downloader.report_warning(u'unable to log in: bad username or password')
return False
return True
def _confirm_age(self):
age_form = {
- 'next_url': '/',
- 'action_confirm': 'Confirm',
- }
- request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
- try:
- self.report_age_confirmation()
- compat_urllib_request.urlopen(request).read().decode('utf-8')
- except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
- raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err))
+ 'next_url': '/',
+ 'action_confirm': 'Confirm',
+ }
+ req = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
+
+ self._download_webpage(
+ req, None,
+ note=u'Confirming age', errnote=u'Unable to confirm age')
return True
def _real_initialize(self):
'248': 'webm',
}
_video_dimensions = {
- '5': '240x400',
+ '5': '400x240',
'6': '???',
'13': '???',
- '17': '144x176',
- '18': '360x640',
- '22': '720x1280',
- '34': '360x640',
- '35': '480x854',
- '36': '240x320',
- '37': '1080x1920',
- '38': '3072x4096',
- '43': '360x640',
- '44': '480x854',
- '45': '720x1280',
- '46': '1080x1920',
+ '17': '176x144',
+ '18': '640x360',
+ '22': '1280x720',
+ '34': '640x360',
+ '35': '854x480',
+ '36': '320x240',
+ '37': '1920x1080',
+ '38': '4096x3072',
+ '43': '640x360',
+ '44': '854x480',
+ '45': '1280x720',
+ '46': '1920x1080',
'82': '360p',
'83': '480p',
'84': '720p',
u"uploader": u"Philipp Hagemeister",
u"uploader_id": u"phihag",
u"upload_date": u"20121002",
- u"description": u"test chars: \"'/\\ä↭𝕐\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 ."
}
},
{
super(YoutubeIE, self).__init__(*args, **kwargs)
self._player_cache = {}
- def report_video_webpage_download(self, video_id):
- """Report attempt to download video webpage."""
- self.to_screen(u'%s: Downloading video webpage' % video_id)
-
def report_video_info_webpage_download(self, video_id):
"""Report attempt to download video info webpage."""
self.to_screen(u'%s: Downloading video info webpage' % video_id)
'asrs': 1,
})
list_url = caption_url + '&' + list_params
- list_page = self._download_webpage(list_url, video_id)
- caption_list = xml.etree.ElementTree.fromstring(list_page.encode('utf-8'))
+ caption_list = self._download_xml(list_url, video_id)
original_lang_node = caption_list.find('track')
if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr' :
self._downloader.report_warning(u'Video doesn\'t have automatic captions')
video_id = self._extract_id(url)
# Get video webpage
- self.report_video_webpage_download(video_id)
url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
- request = compat_urllib_request.Request(url)
- try:
- video_webpage_bytes = compat_urllib_request.urlopen(request).read()
- except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
- raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err))
-
- video_webpage = video_webpage_bytes.decode('utf-8', 'ignore')
+ video_webpage = self._download_webpage(url, video_id)
# Attempt to extract SWF player URL
mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
# description
video_description = get_element_by_id("eow-description", video_webpage)
if video_description:
+ video_description = re.sub(r'''(?x)
+ <a\s+
+ (?:[a-zA-Z-]+="[^"]+"\s+)*?
+ title="([^"]+)"\s+
+ (?:[a-zA-Z-]+="[^"]+"\s+)*?
+ class="yt-uix-redirect-link"\s*>
+ [^<]+
+ </a>
+ ''', r'\1', video_description)
video_description = clean_html(video_description)
else:
fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
else:
video_description = u''
+ def _extract_count(klass):
+ count = self._search_regex(r'class="%s">([\d,]+)</span>' % re.escape(klass), video_webpage, klass, fatal=False)
+ if count is not None:
+ return int(count.replace(',', ''))
+ return None
+ like_count = _extract_count(u'likes-count')
+ dislike_count = _extract_count(u'dislikes-count')
+
# subtitles
video_subtitles = self.extract_subtitles(video_id, video_webpage)
if 'length_seconds' not in video_info:
self._downloader.report_warning(u'unable to extract video duration')
- video_duration = ''
+ video_duration = None
else:
- video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])
+ video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
# annotations
video_annotations = None
'annotations': video_annotations,
'webpage_url': 'https://www.youtube.com/watch?v=%s' % video_id,
'view_count': view_count,
+ 'like_count': like_count,
+ 'dislike_count': dislike_count,
})
return results
\? (?:.*?&)*? (?:p|a|list)=
| p/
)
- ((?:PL|EC|UU|FL)?[0-9A-Za-z-_]{10,})
+ ((?:PL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,})
.*
|
- ((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,})
+ ((?:PL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
)"""
_TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&page=%s'
_MORE_PAGES_INDICATOR = r'data-link-type="next"'
- _VIDEO_RE = r'href="/watch\?v=([0-9A-Za-z_-]{11})&'
+ _VIDEO_RE = r'href="/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
IE_NAME = u'youtube:playlist'
@classmethod
def _real_initialize(self):
self._login()
+ def _ids_to_results(self, ids):
+ return [self.url_result(vid_id, 'Youtube', video_id=vid_id)
+ for vid_id in ids]
+
+ def _extract_mix(self, playlist_id):
+ # The mixes are generated from a a single video
+ # 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))
+ 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))
+ 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)
else:
self.to_screen(u'Downloading playlist PL%s - add --no-playlist to just download video %s' % (playlist_id, video_id))
+ if playlist_id.startswith('RD'):
+ # Mixes require a custom extraction process
+ return self._extract_mix(playlist_id)
+ if playlist_id.startswith('TL'):
+ raise ExtractorError(u'For downloading YouTube.com top lists, use '
+ u'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True)
+
# 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)
- # The ids are duplicated
- new_ids = orderedSet(re.findall(self._VIDEO_RE, page))
+ matches = re.finditer(self._VIDEO_RE, page)
+ # 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:
playlist_title = self._og_search_title(page)
- url_results = [self.url_result(vid_id, 'Youtube', video_id=vid_id)
- for vid_id in ids]
+ url_results = self._ids_to_results(ids)
return self.playlist_result(url_results, playlist_id, playlist_title)
+class YoutubeTopListIE(YoutubePlaylistIE):
+ IE_NAME = u'youtube:toplist'
+ IE_DESC = (u'YouTube.com top lists, "yttoplist:{channel}:{list title}"'
+ u' (Example: "yttoplist:music:Top Tracks")')
+ _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$'
+
+ def _real_extract(self, url):
+ mobj = re.match(self._VALID_URL, url)
+ channel = mobj.group('chann')
+ title = mobj.group('title')
+ query = compat_urllib_parse.urlencode({'title': title})
+ 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)
+
+ video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"'
+ ids = []
+ # sometimes the webpage doesn't contain the videos
+ # retry until we get them
+ for i in itertools.count(0):
+ msg = u'Downloading Youtube mix'
+ if i > 0:
+ msg += ', retry #%d' % i
+ webpage = self._download_webpage(url, title, msg)
+ ids = orderedSet(re.findall(video_re, webpage))
+ if ids:
+ break
+ url_results = self._ids_to_results(ids)
+ return self.playlist_result(url_results, playlist_title=title)
+
+
class YoutubeChannelIE(InfoExtractor):
IE_DESC = u'YouTube.com channels'
_VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
video_ids = []
url = 'https://www.youtube.com/channel/%s/videos' % channel_id
channel_page = self._download_webpage(url, channel_id)
- if re.search(r'channel-header-autogenerated-label', channel_page) is not None:
- autogenerated = True
- else:
- autogenerated = False
+ autogenerated = re.search(r'''(?x)
+ class="[^"]*?(?:
+ channel-header-autogenerated-label|
+ yt-channel-title-autogenerated
+ )[^"]*"''', channel_page) is not None
if autogenerated:
# The videos are contained in a single page
IE_NAME = u'youtube:search'
_SEARCH_KEY = 'ytsearch'
- def report_download_page(self, query, pagenum):
- """Report attempt to download search page with given number."""
- self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
-
def _get_n_results(self, query, n):
"""Get a specified number of results for a query"""
limit = n
while (50 * pagenum) < limit:
- self.report_download_page(query, pagenum+1)
result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
- request = compat_urllib_request.Request(result_url)
- try:
- data = compat_urllib_request.urlopen(request).read().decode('utf-8')
- except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
- raise ExtractorError(u'Unable to download API page: %s' % compat_str(err))
- api_response = json.loads(data)['data']
-
- if not 'items' in api_response:
+ data_json = self._download_webpage(
+ result_url, video_id=u'query "%s"' % query,
+ note=u'Downloading page %s' % (pagenum + 1),
+ errnote=u'Unable to download API page')
+ data = json.loads(data_json)
+ api_response = data['data']
+
+ if 'items' not in api_response:
raise ExtractorError(u'[youtube] No video results')
new_ids = list(video['id'] for video in api_response['items'])
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'
Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
"""
_LOGIN_REQUIRED = True
- _PAGING_STEP = 30
# use action_load_personal_feed instead of action_load_system_feed
_PERSONAL_FEED = False
def _real_extract(self, url):
feed_entries = []
- # The step argument is available only in 2.7 or higher
- for i in itertools.count(0):
- paging = i*self._PAGING_STEP
+ paging = 0
+ for i in itertools.count(1):
info = self._download_webpage(self._FEED_TEMPLATE % paging,
u'%s feed' % self._FEED_NAME,
u'Downloading page %s' % i)
for video_id in ids)
if info['paging'] is None:
break
+ paging = info['paging']
return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
_VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
_FEED_NAME = 'watch_later'
_PLAYLIST_TITLE = u'Youtube Watch Later'
- _PAGING_STEP = 100
_PERSONAL_FEED = True
class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
_PERSONAL_FEED = True
_PLAYLIST_TITLE = u'Youtube Watch History'
- def _real_extract(self, url):
- webpage = self._download_webpage('https://www.youtube.com/feed/history', u'History')
- data_paging = self._search_regex(r'data-paging="(\d+)"', webpage, u'data-paging')
- # The step is actually a ridiculously big number (like 1374343569725646)
- self._PAGING_STEP = int(data_paging)
- return super(YoutubeHistoryIE, self)._real_extract(url)
-
class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
IE_NAME = u'youtube:favorites'
IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'