def _write_slides(self, slides_infos: {}, ydl: YoutubeDL):
for slide in slides_infos:
if os.path.exists(encodeFilename(slide.path)):
self.to_screen('Slide %s is already present' %
(slide.filename))
else:
self.to_screen('Downloading slide %s...' % (slide.filename))
try_num = 1
try:
url_f = ydl.urlopen(slide.url)
with open(encodeFilename(slide.path), 'wb') as slide_f:
shutil.copyfileobj(url_f, slide_f)
self.to_screen('Successfully downloaded to: %s' %
(slide.path))
except (compat_urllib_error.URLError,
compat_http_client.HTTPException, socket.error) as err:
self.report_warning(
'(Try %s of 3) Unable to download slide "%s": %s' %
(try_num, slide.url, error_to_compat_str(err)))
if try_num == 3:
self.report_warning(
'Slides are essential. Abort! Please try again later!'
)
exit(1)
try_num += 1
def _create_tmp_dir(self, video_id):
try:
if not os.path.exists(video_id):
os.makedirs(video_id)
except (OSError, IOError) as err:
self.ydl.report_error('unable to create directory ' +
error_to_compat_str(err))
def _remove_tmp_dir(self, video_id):
try:
if os.path.exists(video_id):
shutil.rmtree(video_id)
except (OSError, IOError) as err:
self.ydl.report_error('unable to remove directory ' + error_to_compat_str(err))
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 test_template(self):
ie = youtube_dl.extractor.get_info_extractor(test_case['name'])()
other_ies = [
get_info_extractor(ie_key)()
for ie_key in test_case.get('add_ie', [])
]
is_playlist = any(k.startswith('playlist') for k in test_case)
test_cases = test_case.get('playlist',
[] if is_playlist else [test_case])
def print_skipping(reason):
print('Skipping %s: %s' % (test_case['name'], reason))
self.skipTest(reason)
if not ie.working():
print_skipping('IE marked as not _WORKING')
for tc in test_cases:
info_dict = tc.get('info_dict', {})
if not (info_dict.get('id') and info_dict.get('ext')):
raise Exception(
'Test definition (%s) requires both \'id\' and \'ext\' keys present to define the output file'
% (tname, ))
if 'skip' in test_case:
print_skipping(test_case['skip'])
for other_ie in other_ies:
if not other_ie.working():
print_skipping('test depends on %sIE, marked as not WORKING' %
other_ie.ie_key())
params = get_params(test_case.get('params', {}))
params['outtmpl'] = tname + '_' + params['outtmpl']
if is_playlist and 'playlist' not in test_case:
params.setdefault('extract_flat', 'in_playlist')
params.setdefault('playlistend',
test_case.get('playlist_mincount'))
params.setdefault('skip_download', True)
ydl = YoutubeDL(params, auto_init=False)
ydl.add_default_info_extractors()
finished_hook_called = set()
def _hook(status):
if status['status'] == 'finished':
finished_hook_called.add(status['filename'])
ydl.add_progress_hook(_hook)
expect_warnings(ydl, test_case.get('expected_warnings', []))
def get_tc_filename(tc):
return ydl.prepare_filename(tc.get('info_dict', {}))
res_dict = None
def try_rm_tcs_files(tcs=None):
if tcs is None:
tcs = test_cases
for tc in tcs:
tc_filename = get_tc_filename(tc)
try_rm(tc_filename)
try_rm(tc_filename + '.part')
try_rm(os.path.splitext(tc_filename)[0] + '.info.json')
try_rm_tcs_files()
try:
try_num = 1
while True:
try:
# We're not using .download here since that is just a shim
# for outside error handling, and returns the exit code
# instead of the result dict.
res_dict = ydl.extract_info(
test_case['url'],
force_generic_extractor=params.get(
'force_generic_extractor', False))
except (DownloadError, ExtractorError) as err:
# Check if the exception is not a network related one
if not err.exc_info[0] in (
compat_urllib_error.URLError, socket.timeout,
UnavailableVideoError,
compat_http_client.BadStatusLine) or (
err.exc_info[0] == compat_HTTPError
and err.exc_info[1].code == 503):
msg = getattr(err, 'msg', error_to_compat_str(err))
err.msg = '%s (%s)' % (
msg,
tname,
)
raise err
if try_num == RETRIES:
report_warning(
'%s failed due to network errors, skipping...' %
tname)
return
print(
'Retrying: {0} failed tries\n\n##########\n\n'.format(
try_num))
try_num += 1
else:
break
if is_playlist:
self.assertTrue(
res_dict['_type'] in ['playlist', 'multi_video'])
self.assertTrue('entries' in res_dict)
expect_info_dict(self, res_dict,
test_case.get('info_dict', {}))
if 'playlist_mincount' in test_case:
assertGreaterEqual(
self, len(res_dict['entries']),
test_case['playlist_mincount'],
'Expected at least %d in playlist %s, but got only %d' %
(test_case['playlist_mincount'], test_case['url'],
len(res_dict['entries'])))
if 'playlist_count' in test_case:
self.assertEqual(
len(res_dict['entries']), test_case['playlist_count'],
'Expected %d entries in playlist %s, but got %d.' % (
test_case['playlist_count'],
test_case['url'],
len(res_dict['entries']),
))
if 'playlist_duration_sum' in test_case:
got_duration = sum(e['duration'] for e in res_dict['entries'])
self.assertEqual(test_case['playlist_duration_sum'],
got_duration)
# Generalize both playlists and single videos to unified format for
# simplicity
if 'entries' not in res_dict:
res_dict['entries'] = [res_dict]
for tc_num, tc in enumerate(test_cases):
tc_res_dict = res_dict['entries'][tc_num]
# First, check test cases' data against extracted data alone
expect_info_dict(self, tc_res_dict, tc.get('info_dict', {}))
# Now, check downloaded file consistency
tc_filename = get_tc_filename(tc)
if not test_case.get('params', {}).get('skip_download', False):
self.assertTrue(os.path.exists(tc_filename),
msg='Missing file ' + tc_filename)
self.assertTrue(tc_filename in finished_hook_called)
expected_minsize = tc.get('file_minsize', 10000)
if expected_minsize is not None:
if params.get('test'):
expected_minsize = max(expected_minsize, 10000)
got_fsize = os.path.getsize(tc_filename)
assertGreaterEqual(
self, got_fsize, expected_minsize,
'Expected %s to be at least %s, but it\'s only %s '
% (tc_filename, format_bytes(expected_minsize),
format_bytes(got_fsize)))
if 'md5' in tc:
md5_for_file = _file_md5(tc_filename)
self.assertEqual(tc['md5'], md5_for_file)
# Finally, check test cases' data again but this time against
# extracted data from info JSON file written during processing
info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json'
self.assertTrue(os.path.exists(info_json_fn),
'Missing info file %s' % info_json_fn)
with io.open(info_json_fn, encoding='utf-8') as infof:
info_dict = json.load(infof)
expect_info_dict(self, info_dict, tc.get('info_dict', {}))
finally:
try_rm_tcs_files()
if is_playlist and res_dict is not None and res_dict.get(
'entries'):
# Remove all other files that may have been extracted if the
# extractor returns full results even with extract_flat
res_tcs = [{'info_dict': e} for e in res_dict['entries']]
try_rm_tcs_files(res_tcs)
