def refresh(self, url, channel_url, max_episodes):
"""
Fetch a channel or playlist contents.
Doesn't yet fetch video entry informations, so we only get the video id and title.
"""
# Duplicate a bit of the YoutubeDL machinery here because we only
# want to parse the channel/playlist first, not to fetch video entries.
# We call YoutubeDL.extract_info(process=False), so we
# have to call extract_info again ourselves when we get a result of type 'url'.
def extract_type(ie_result):
result_type = ie_result.get('_type', 'video')
if result_type not in ('url', 'playlist', 'multi_video'):
raise Exception(
'Unsuported result_type: {}'.format(result_type))
has_playlist = result_type in ('playlist', 'multi_video')
return result_type, has_playlist
opts = {
'youtube_include_dash_manifest':
False, # only interested in video title and id
}
opts.update(self._ydl_opts)
with youtube_dl.YoutubeDL(opts) as ydl:
ie_result = ydl.extract_info(url, download=False, process=False)
result_type, has_playlist = extract_type(ie_result)
while not has_playlist:
if result_type in ('url', 'url_transparent'):
ie_result['url'] = sanitize_url(ie_result['url'])
if result_type == 'url':
logger.debug("extract_info(%s) to get the video list",
ie_result['url'])
# We have to add extra_info to the results because it may be
# contained in a playlist
ie_result = ydl.extract_info(
ie_result['url'],
download=False,
process=False,
ie_key=ie_result.get('ie_key'))
result_type, has_playlist = extract_type(ie_result)
cover_url = youtube.get_cover(
channel_url) # youtube-dl doesn't provide the cover url!
description = youtube.get_channel_desc(
channel_url) # youtube-dl doesn't provide the description!
return feedcore.Result(
feedcore.UPDATED_FEED,
YoutubeFeed(url, cover_url, description, max_episodes, ie_result,
self))
def extractChannelSubscription(self, url, user_name=None):
entries = None
ydl_opts = {}
youtube_dl.utils.std_headers['User-Agent'] = self._user_agent
with youtube_dl.YoutubeDL(ydl_opts) as ydl:
ie = self._selectExtractor(ydl._ies, url)
if ie is None:
return None
ie = ydl.get_info_extractor(ie.ie_key())
if not ie.working():
ydl.report_warning(
'The program functionality for this site has been marked as broken, '
'and will probably not work.')
try:
ie_result = ie.extract(url)
if ie_result is None:
return None
ydl.add_default_extra_info(ie_result, ie, url)
# ydl.process_ie_result(ie_result)
# here the url type changed, so we hava to select extractor again
url = sanitize_url(ie_result['url'])
ie = self._selectExtractor(ydl._ies, ie_result['url'])
ie = ydl.get_info_extractor(ie.ie_key())
ie_result = ie.extract(url)
ie_entries = ie_result['entries']
entries = list(itertools.islice(ie_entries, 0, None))
except GeoRestrictedError as e:
msg = e.msg
if e.countries:
msg += '\nThis video is available in %s.' % ', '.join(
map(ISO3166Utils.short2full, e.countries))
msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
ydl.report_error(msg)
except ExtractorError as e: # An error we somewhat expected
ydl.report_error(compat_str(e), e.format_traceback())
except MaxDownloadsReached:
raise
except Exception as e:
if ydl.params.get('ignoreerrors', False):
ydl.report_error(error_to_compat_str(e),
tb=encode_compat_str(
traceback.format_exc()))
return None
else:
raise
return entries
def add(self, url):
from youtube_dl import utils as u
safe_url = u.sanitize_url(url) # great helper method, thanks OP
self.queue.put(safe_url)
print("QueueManager added url {} to queue".format(safe_url))
def test_sanitize_url_slash_expansion(self):
self.assertEqual(sanitize_url('//google.com'), 'http://google.com')
def test_full_prefix_sanitize_url(self):
self.assertEqual(sanitize_url('www.google.com'),
'http://www.google.com')
def test_no_colon_prefix_sanitize_url(self):
self.assertEqual(sanitize_url('http//google.com'), 'http://google.com')
def test_sanitize_url_base_case(self):
self.assertEqual(sanitize_url('http://google.com'),
'http://google.com')
