def _check_params(attrs):
params = get_params()
for attr in attrs:
if attr not in params:
print('Missing %s. Skipping.' % attr)
return
return params
def _check_params(attrs):
params = get_params()
for attr in attrs:
if attr not in params:
print('Missing %s. Skipping.' % attr)
return
return params
def setUp(self):
self.stored_name_1 = None
self.stored_name_2 = None
self.params = get_params({
'skip_download': False,
'writeinfojson': False,
'quiet': True,
'verbose': False,
'cachedir': False,
})
self.files = []
import io
import xml.etree.ElementTree
import ananse_dl.Ananse
import ananse_dl.extractor
class AnanseDl(ananse_dl.AnanseDl):
def __init__(self, *args, **kwargs):
super(AnanseDl, self).__init__(*args, **kwargs)
self.to_stderr = self.to_screen
params = get_params({
'writeannotations': True,
'skip_download': True,
'writeinfojson': False,
'format': 'flv',
})
TEST_ID = 'gr51aVj-mLg'
ANNOTATIONS_FILE = TEST_ID + '.flv.annotations.xml'
EXPECTED_ANNOTATIONS = ['Speech bubble', 'Note', 'Title', 'Spotlight', 'Label']
class TestAnnotations(unittest.TestCase):
def setUp(self):
# Clear old files
self.tearDown()
def test_info_json(self):
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', [])]
def print_skipping(reason):
print('Skipping %s: %s' % (test_case['name'], reason))
if not ie.working():
print_skipping('IE marked as not _WORKING')
return
if 'playlist' not in test_case:
info_dict = test_case.get('info_dict', {})
if not test_case.get('file') and not (info_dict.get('id') and info_dict.get('ext')):
print_skipping('The output file cannot be know, the "file" '
'key is missing or the info_dict is incomplete')
return
if 'skip' in test_case:
print_skipping(test_case['skip'])
return
for other_ie in other_ies:
if not other_ie.working():
print_skipping(u'test depends on %sIE, marked as not WORKING' % other_ie.ie_key())
return
params = get_params(test_case.get('params', {}))
ydl = YoutubeDL(params)
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)
def get_tc_filename(tc):
return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
test_cases = test_case.get('playlist', [test_case])
def try_rm_tcs_files():
for tc in test_cases:
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:
ydl.download([test_case['url']])
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):
raise
if try_num == RETRIES:
report_warning(u'Failed due to network errors, skipping...')
return
print('Retrying: {0} failed tries\n\n##########\n\n'.format(try_num))
try_num += 1
else:
break
for tc in test_cases:
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)
info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json'
self.assertTrue(os.path.exists(info_json_fn))
if 'md5' in tc:
md5_for_file = _file_md5(tc_filename)
self.assertEqual(md5_for_file, tc['md5'])
with io.open(info_json_fn, encoding='utf-8') as infof:
info_dict = json.load(infof)
for (info_field, expected) in tc.get('info_dict', {}).items():
if isinstance(expected, compat_str) and expected.startswith('md5:'):
got = 'md5:' + md5(info_dict.get(info_field))
else:
got = info_dict.get(info_field)
self.assertEqual(expected, got,
u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
# If checkable fields are missing from the test case, print the info_dict
test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'md5:' + md5(value))
for key, value in info_dict.items()
if value and key in ('title', 'description', 'uploader', 'upload_date', 'uploader_id', 'location'))
if not all(key in tc.get('info_dict', {}).keys() for key in test_info_dict.keys()):
sys.stderr.write(u'\n"info_dict": ' + json.dumps(test_info_dict, ensure_ascii=False, indent=4) + u'\n')
# Check for the presence of mandatory fields
for key in ('id', 'url', 'title', 'ext'):
self.assertTrue(key in info_dict.keys() and info_dict[key])
# Check for mandatory fields that are automatically set by YoutubeDL
for key in ['webpage_url', 'extractor', 'extractor_key']:
self.assertTrue(info_dict.get(key), u'Missing field: %s' % key)
finally:
try_rm_tcs_files()
def test_template(self):
ie = yt_dlp.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))
if not ie.working():
print_skipping('IE marked as not _WORKING')
return
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 incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?'
)
if 'skip' in test_case:
print_skipping(test_case['skip'])
return
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())
return
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):
raise
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)
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))
if not ie.working():
print_skipping('IE marked as not _WORKING')
return
for tc in test_cases:
info_dict = tc.get('info_dict', {})
if not tc.get('file') and not (info_dict.get('id') and info_dict.get('ext')):
raise Exception('Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?')
if 'skip' in test_case:
print_skipping(test_case['skip'])
return
for other_ie in other_ies:
if not other_ie.working():
print_skipping(u'test depends on %sIE, marked as not WORKING' % other_ie.ie_key())
return
params = get_params(test_case.get('params', {}))
if is_playlist and 'playlist' not in test_case:
params.setdefault('extract_flat', True)
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 tc.get('file') or 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 sine 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'])
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):
raise
if try_num == RETRIES:
report_warning(u'Failed due to network errors, skipping...')
return
print('Retrying: {0} failed tries\n\n##########\n\n'.format(try_num))
try_num += 1
else:
break
if is_playlist:
self.assertEqual(res_dict['_type'], 'playlist')
self.assertTrue('entries' in res_dict)
expect_info_dict(self, test_case.get('info_dict', {}), res_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)
for tc in test_cases:
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(md5_for_file, tc['md5'])
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, tc.get('info_dict', {}), 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)
def test_template(self):
ie = youtube_dl.extractor.get_info_extractor(test_case['name'])
def print_skipping(reason):
print('Skipping %s: %s' % (test_case['name'], reason))
if not ie._WORKING:
print_skipping('IE marked as not _WORKING')
return
if 'playlist' not in test_case and not test_case['file']:
print_skipping('No output file specified')
return
if 'skip' in test_case:
print_skipping(test_case['skip'])
return
params = get_params(test_case.get('params', {}))
ydl = YoutubeDL(params)
ydl.add_default_info_extractors()
finished_hook_called = set()
def _hook(status):
if status['status'] == 'finished':
finished_hook_called.add(status['filename'])
ydl.fd.add_progress_hook(_hook)
test_cases = test_case.get('playlist', [test_case])
for tc in test_cases:
try_rm(tc['file'])
try_rm(tc['file'] + '.part')
try_rm(tc['file'] + '.info.json')
try:
for retry in range(1, RETRIES + 1):
try:
ydl.download([test_case['url']])
except (DownloadError, ExtractorError) as err:
if retry == RETRIES: raise
# Check if the exception is not a network related one
if not err.exc_info[0] in (compat_urllib_error.URLError,
socket.timeout,
UnavailableVideoError):
raise
print(
'Retrying: {0} failed tries\n\n##########\n\n'.format(
retry))
else:
break
for tc in test_cases:
if not test_case.get('params', {}).get('skip_download', False):
self.assertTrue(os.path.exists(tc['file']),
msg='Missing file ' + tc['file'])
self.assertTrue(tc['file'] in finished_hook_called)
self.assertTrue(os.path.exists(tc['file'] + '.info.json'))
if 'md5' in tc:
md5_for_file = _file_md5(tc['file'])
self.assertEqual(md5_for_file, tc['md5'])
with io.open(tc['file'] + '.info.json',
encoding='utf-8') as infof:
info_dict = json.load(infof)
for (info_field, expected) in tc.get('info_dict', {}).items():
if isinstance(expected,
compat_str) and expected.startswith('md5:'):
got = 'md5:' + md5(info_dict.get(info_field))
else:
got = info_dict.get(info_field)
self.assertEqual(
expected, got,
u'invalid value for field %s, expected %r, got %r' %
(info_field, expected, got))
# If checkable fields are missing from the test case, print the info_dict
test_info_dict = dict(
(key, value if not isinstance(value, compat_str)
or len(value) < 250 else 'md5:' + md5(value))
for key, value in info_dict.items()
if value and key in ('title', 'description', 'uploader',
'upload_date', 'uploader_id',
'location'))
if not all(key in tc.get('info_dict', {}).keys()
for key in test_info_dict.keys()):
sys.stderr.write(u'\n"info_dict": ' + json.dumps(
test_info_dict, ensure_ascii=False, indent=2) + u'\n')
# Check for the presence of mandatory fields
for key in ('id', 'url', 'title', 'ext'):
self.assertTrue(key in info_dict.keys() and info_dict[key])
finally:
for tc in test_cases:
try_rm(tc['file'])
try_rm(tc['file'] + '.part')
try_rm(tc['file'] + '.info.json')
import io
import xml.etree.ElementTree
import youtube_dl.YoutubeDL
import youtube_dl.extractor
class YoutubeDL(youtube_dl.YoutubeDL):
def __init__(self, *args, **kwargs):
super(YoutubeDL, self).__init__(*args, **kwargs)
self.to_stderr = self.to_screen
params = get_params({
'writeannotations': True,
'skip_download': True,
'writeinfojson': False,
'format': 'flv',
})
TEST_ID = 'gr51aVj-mLg'
ANNOTATIONS_FILE = TEST_ID + '.flv.annotations.xml'
EXPECTED_ANNOTATIONS = ['Speech bubble', 'Note', 'Title', 'Spotlight', 'Label']
class TestAnnotations(unittest.TestCase):
def setUp(self):
# Clear old files
self.tearDown()
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', [])]
def print_skipping(reason):
print('Skipping %s: %s' % (test_case['name'], reason))
if not ie.working():
print_skipping('IE marked as not _WORKING')
return
if 'playlist' not in test_case:
info_dict = test_case.get('info_dict', {})
if not test_case.get('file') and not (info_dict.get('id') and info_dict.get('ext')):
raise Exception('Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?')
if 'skip' in test_case:
print_skipping(test_case['skip'])
return
for other_ie in other_ies:
if not other_ie.working():
print_skipping(u'test depends on %sIE, marked as not WORKING' % other_ie.ie_key())
return
params = get_params(test_case.get('params', {}))
ydl = YoutubeDL(params)
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)
def get_tc_filename(tc):
return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
test_cases = test_case.get('playlist', [test_case])
def try_rm_tcs_files():
for tc in test_cases:
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:
ydl.download([test_case['url']])
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):
raise
if try_num == RETRIES:
report_warning(u'Failed due to network errors, skipping...')
return
print('Retrying: {0} failed tries\n\n##########\n\n'.format(try_num))
try_num += 1
else:
break
for tc in test_cases:
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)
info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json'
self.assertTrue(os.path.exists(info_json_fn))
if 'md5' in tc:
md5_for_file = _file_md5(tc_filename)
self.assertEqual(md5_for_file, tc['md5'])
with io.open(info_json_fn, encoding='utf-8') as infof:
info_dict = json.load(infof)
expect_info_dict(self, tc.get('info_dict', {}), info_dict)
finally:
try_rm_tcs_files()
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', [])
]
def print_skipping(reason):
print('Skipping %s: %s' % (test_case['name'], reason))
if not ie.working():
print_skipping('IE marked as not _WORKING')
return
if 'playlist' not in test_case:
info_dict = test_case.get('info_dict', {})
if not test_case.get('file') and not (info_dict.get('id')
and info_dict.get('ext')):
raise Exception(
'Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?'
)
if 'skip' in test_case:
print_skipping(test_case['skip'])
return
for other_ie in other_ies:
if not other_ie.working():
print_skipping(u'test depends on %sIE, marked as not WORKING' %
other_ie.ie_key())
return
params = get_params(test_case.get('params', {}))
ydl = YoutubeDL(params)
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)
def get_tc_filename(tc):
return tc.get('file') or ydl.prepare_filename(
tc.get('info_dict', {}))
test_cases = test_case.get('playlist', [test_case])
def try_rm_tcs_files():
for tc in test_cases:
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:
ydl.download([test_case['url']])
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):
raise
if try_num == RETRIES:
report_warning(
u'Failed due to network errors, skipping...')
return
print(
'Retrying: {0} failed tries\n\n##########\n\n'.format(
try_num))
try_num += 1
else:
break
for tc in test_cases:
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)
info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json'
self.assertTrue(os.path.exists(info_json_fn))
if 'md5' in tc:
md5_for_file = _file_md5(tc_filename)
self.assertEqual(md5_for_file, tc['md5'])
with io.open(info_json_fn, encoding='utf-8') as infof:
info_dict = json.load(infof)
expect_info_dict(self, tc.get('info_dict', {}), info_dict)
finally:
try_rm_tcs_files()
import io
import xml.etree.ElementTree
import youtube_dl.YoutubeDL
import youtube_dl.extractor
class YoutubeDL(youtube_dl.YoutubeDL):
def __init__(self, *args, **kwargs):
super(YoutubeDL, self).__init__(*args, **kwargs)
self.to_stderr = self.to_screen
params = get_params({"writeannotations": True, "skip_download": True, "writeinfojson": False, "format": "flv"})
TEST_ID = "gr51aVj-mLg"
ANNOTATIONS_FILE = TEST_ID + ".annotations.xml"
EXPECTED_ANNOTATIONS = ["Speech bubble", "Note", "Title", "Spotlight", "Label"]
class TestAnnotations(unittest.TestCase):
def setUp(self):
# Clear old files
self.tearDown()
def test_info_json(self):
expected = list(EXPECTED_ANNOTATIONS) # Two annotations could have the same text.
ie = youtube_dl.extractor.YoutubeIE()
import io
import json
import youtube_dl.YoutubeDL
import youtube_dl.extractor
class YoutubeDL(youtube_dl.YoutubeDL):
def __init__(self, *args, **kwargs):
super(YoutubeDL, self).__init__(*args, **kwargs)
self.to_stderr = self.to_screen
params = get_params({
'writeinfojson': True,
'skip_download': True,
'writedescription': True,
})
TEST_ID = 'BaW_jenozKc'
INFO_JSON_FILE = TEST_ID + '.info.json'
DESCRIPTION_FILE = TEST_ID + '.mp4.description'
EXPECTED_DESCRIPTION = u'''test chars: "'/\ä↭𝕐
This is a test video for youtube-dl.
For more information, contact [email protected] .'''
class TestInfoJSON(unittest.TestCase):
def test_template(self):
ie = youtube_dl.extractor.get_info_extractor(test_case['name'])
def print_skipping(reason):
print('Skipping %s: %s' % (test_case['name'], reason))
if not ie._WORKING:
print_skipping('IE marked as not _WORKING')
return
if 'playlist' not in test_case and not test_case['file']:
print_skipping('No output file specified')
return
if 'skip' in test_case:
print_skipping(test_case['skip'])
return
params = get_params(test_case.get('params', {}))
ydl = YoutubeDL(params)
ydl.add_default_info_extractors()
finished_hook_called = set()
def _hook(status):
if status['status'] == 'finished':
finished_hook_called.add(status['filename'])
ydl.fd.add_progress_hook(_hook)
test_cases = test_case.get('playlist', [test_case])
for tc in test_cases:
try_rm(tc['file'])
try_rm(tc['file'] + '.part')
try_rm(tc['file'] + '.info.json')
try:
for retry in range(1, RETRIES + 1):
try:
ydl.download([test_case['url']])
except (DownloadError, ExtractorError) as err:
if retry == RETRIES: raise
# Check if the exception is not a network related one
if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
raise
print('Retrying: {0} failed tries\n\n##########\n\n'.format(retry))
else:
break
for tc in test_cases:
if not test_case.get('params', {}).get('skip_download', False):
self.assertTrue(os.path.exists(tc['file']), msg='Missing file ' + tc['file'])
self.assertTrue(tc['file'] in finished_hook_called)
self.assertTrue(os.path.exists(tc['file'] + '.info.json'))
if 'md5' in tc:
md5_for_file = _file_md5(tc['file'])
self.assertEqual(md5_for_file, tc['md5'])
with io.open(tc['file'] + '.info.json', encoding='utf-8') as infof:
info_dict = json.load(infof)
for (info_field, expected) in tc.get('info_dict', {}).items():
if isinstance(expected, compat_str) and expected.startswith('md5:'):
got = 'md5:' + md5(info_dict.get(info_field))
else:
got = info_dict.get(info_field)
self.assertEqual(expected, got,
u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
# If checkable fields are missing from the test case, print the info_dict
test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'md5:' + md5(value))
for key, value in info_dict.items()
if value and key in ('title', 'description', 'uploader', 'upload_date', 'uploader_id', 'location'))
if not all(key in tc.get('info_dict', {}).keys() for key in test_info_dict.keys()):
sys.stderr.write(u'\n"info_dict": ' + json.dumps(test_info_dict, ensure_ascii=False, indent=2) + u'\n')
# Check for the presence of mandatory fields
for key in ('id', 'url', 'title', 'ext'):
self.assertTrue(key in info_dict.keys() and info_dict[key])
finally:
for tc in test_cases:
try_rm(tc['file'])
try_rm(tc['file'] + '.part')
try_rm(tc['file'] + '.info.json')
import io
import json
import youtube_dl.YoutubeDL
import youtube_dl.extractor
class YoutubeDL(youtube_dl.YoutubeDL):
def __init__(self, *args, **kwargs):
super(YoutubeDL, self).__init__(*args, **kwargs)
self.to_stderr = self.to_screen
params = get_params({
'writeinfojson': True,
'skip_download': True,
'writedescription': True,
})
TEST_ID = 'BaW_jenozKc'
INFO_JSON_FILE = TEST_ID + '.info.json'
DESCRIPTION_FILE = TEST_ID + '.mp4.description'
EXPECTED_DESCRIPTION = u'''test chars: "'/\ä↭𝕐
test URL: https://github.com/rg3/youtube-dl/issues/1892
This is a test video for youtube-dl.
For more information, contact [email protected] .'''
class TestInfoJSON(unittest.TestCase):
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", [])]
def print_skipping(reason):
print("Skipping %s: %s" % (test_case["name"], reason))
if not ie.working():
print_skipping("IE marked as not _WORKING")
return
if "playlist" not in test_case:
info_dict = test_case.get("info_dict", {})
if not test_case.get("file") and not (info_dict.get("id") and info_dict.get("ext")):
print_skipping(
'The output file cannot be know, the "file" ' "key is missing or the info_dict is incomplete"
)
return
if "skip" in test_case:
print_skipping(test_case["skip"])
return
for other_ie in other_ies:
if not other_ie.working():
print_skipping(u"test depends on %sIE, marked as not WORKING" % other_ie.ie_key())
return
params = get_params(test_case.get("params", {}))
ydl = YoutubeDL(params)
ydl.add_default_info_extractors()
finished_hook_called = set()
def _hook(status):
if status["status"] == "finished":
finished_hook_called.add(status["filename"])
ydl.fd.add_progress_hook(_hook)
def get_tc_filename(tc):
return tc.get("file") or ydl.prepare_filename(tc.get("info_dict", {}))
test_cases = test_case.get("playlist", [test_case])
def try_rm_tcs_files():
for tc in test_cases:
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:
ydl.download([test_case["url"]])
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) or (
err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503
):
raise
if try_num == RETRIES:
report_warning(u"Failed due to network errors, skipping...")
return
print("Retrying: {0} failed tries\n\n##########\n\n".format(try_num))
try_num += 1
else:
break
for tc in test_cases:
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)
info_json_fn = os.path.splitext(tc_filename)[0] + ".info.json"
self.assertTrue(os.path.exists(info_json_fn))
if "md5" in tc:
md5_for_file = _file_md5(tc_filename)
self.assertEqual(md5_for_file, tc["md5"])
with io.open(info_json_fn, encoding="utf-8") as infof:
info_dict = json.load(infof)
for (info_field, expected) in tc.get("info_dict", {}).items():
if isinstance(expected, compat_str) and expected.startswith("md5:"):
got = "md5:" + md5(info_dict.get(info_field))
else:
got = info_dict.get(info_field)
self.assertEqual(
expected, got, u"invalid value for field %s, expected %r, got %r" % (info_field, expected, got)
)
# If checkable fields are missing from the test case, print the info_dict
test_info_dict = dict(
(key, value if not isinstance(value, compat_str) or len(value) < 250 else "md5:" + md5(value))
for key, value in info_dict.items()
if value and key in ("title", "description", "uploader", "upload_date", "uploader_id", "location")
)
if not all(key in tc.get("info_dict", {}).keys() for key in test_info_dict.keys()):
sys.stderr.write(
u'\n"info_dict": ' + json.dumps(test_info_dict, ensure_ascii=False, indent=2) + u"\n"
)
# Check for the presence of mandatory fields
for key in ("id", "url", "title", "ext"):
self.assertTrue(key in info_dict.keys() and info_dict[key])
# Check for mandatory fields that are automatically set by YoutubeDL
for key in ["webpage_url", "extractor", "extractor_key"]:
self.assertTrue(info_dict.get(key), u"Missing field: %s" % key)
finally:
try_rm_tcs_files()