def test_info_json(self):
expected = list(EXPECTED_ANNOTATIONS) #Two annotations could have the same text.
ie = youtube_dl.extractor.YoutubeIE()
ydl = YoutubeDL(params)
ydl.add_info_extractor(ie)
ydl.download([TEST_ID])
self.assertTrue(os.path.exists(ANNOTATIONS_FILE))
annoxml = None
with io.open(ANNOTATIONS_FILE, 'r', encoding='utf-8') as annof:
annoxml = xml.etree.ElementTree.parse(annof)
self.assertTrue(annoxml is not None, 'Failed to parse annotations XML')
root = annoxml.getroot()
self.assertEqual(root.tag, 'document')
annotationsTag = root.find('annotations')
self.assertEqual(annotationsTag.tag, 'annotations')
annotations = annotationsTag.findall('annotation')
#Not all the annotations have TEXT children and the annotations are returned unsorted.
for a in annotations:
self.assertEqual(a.tag, 'annotation')
if a.get('type') == 'text':
textTag = a.find('TEXT')
text = textTag.text
self.assertTrue(text in expected) #assertIn only added in python 2.7
#remove the first occurance, there could be more than one annotation with the same text
expected.remove(text)
#We should have seen (and removed) all the expected annotation texts.
self.assertEqual(len(expected), 0, 'Not all expected annotations were found.')
def test_info_json(self):
expected = list(EXPECTED_ANNOTATIONS) # Two annotations could have the same text.
ie = youtube_dl.extractor.YoutubeIE()
ydl = YoutubeDL(params)
ydl.add_info_extractor(ie)
ydl.download([TEST_ID])
self.assertTrue(os.path.exists(ANNOTATIONS_FILE))
annoxml = None
with io.open(ANNOTATIONS_FILE, 'r', encoding='utf-8') as annof:
annoxml = xml.etree.ElementTree.parse(annof)
self.assertTrue(annoxml is not None, 'Failed to parse annotations XML')
root = annoxml.getroot()
self.assertEqual(root.tag, 'document')
annotationsTag = root.find('annotations')
self.assertEqual(annotationsTag.tag, 'annotations')
annotations = annotationsTag.findall('annotation')
# Not all the annotations have TEXT children and the annotations are returned unsorted.
for a in annotations:
self.assertEqual(a.tag, 'annotation')
if a.get('type') == 'text':
textTag = a.find('TEXT')
text = textTag.text
self.assertTrue(text in expected) # assertIn only added in python 2.7
# remove the first occurrence, there could be more than one annotation with the same text
expected.remove(text)
# We should have seen (and removed) all the expected annotation texts.
self.assertEqual(len(expected), 0, 'Not all expected annotations were found.')
def test_info_json(self):
ie = youtube_dl.extractor.YoutubeIE()
ydl = YoutubeDL(params)
ydl.add_info_extractor(ie)
ydl.download([TEST_ID])
self.assertTrue(os.path.exists(INFO_JSON_FILE))
with io.open(INFO_JSON_FILE, 'r', encoding='utf-8') as jsonf:
jd = json.load(jsonf)
self.assertEqual(jd['upload_date'], u'20121002')
self.assertEqual(jd['description'], EXPECTED_DESCRIPTION)
self.assertEqual(jd['id'], TEST_ID)
self.assertEqual(jd['extractor'], 'youtube')
self.assertEqual(jd['title'], u'''youtube-dl test video "'/\ä↭𝕐''')
self.assertEqual(jd['uploader'], 'Philipp Hagemeister')
self.assertTrue(os.path.exists(DESCRIPTION_FILE))
with io.open(DESCRIPTION_FILE, 'r', encoding='utf-8') as descf:
descr = descf.read()
self.assertEqual(descr, EXPECTED_DESCRIPTION)
def test_info_json(self):
ie = youtube_dl.extractor.YoutubeIE()
ydl = YoutubeDL(params)
ydl.add_info_extractor(ie)
ydl.download([TEST_ID])
self.assertTrue(os.path.exists(INFO_JSON_FILE))
with io.open(INFO_JSON_FILE, 'r', encoding='utf-8') as jsonf:
jd = json.load(jsonf)
self.assertEqual(jd['upload_date'], u'20121002')
self.assertEqual(jd['description'], EXPECTED_DESCRIPTION)
self.assertEqual(jd['id'], TEST_ID)
self.assertEqual(jd['extractor'], 'youtube')
self.assertEqual(jd['title'], u'''youtube-dl test video "'/\ä↭𝕐''')
self.assertEqual(jd['uploader'], 'Philipp Hagemeister')
self.assertTrue(os.path.exists(DESCRIPTION_FILE))
with io.open(DESCRIPTION_FILE, 'r', encoding='utf-8') as descf:
descr = descf.read()
self.assertEqual(descr, EXPECTED_DESCRIPTION)
def test_template(self):
ie = youtube_dl.extractor.get_info_extractor(test_case['name'])
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: {0}'.format(test_case['skip']))
return
params = self.parameters.copy()
params.update(test_case.get('params', {}))
ydl = YoutubeDL(params)
for ie in youtube_dl.extractor.gen_extractors():
ydl.add_info_extractor(ie)
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:'):
self.assertEqual(expected, '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')
