2012-12-12 21:15:21 +08:00
|
|
|
#!/usr/bin/env python
|
|
|
|
|
2013-10-15 08:00:53 +08:00
|
|
|
# Allow direct execution
|
|
|
|
import os
|
|
|
|
import sys
|
|
|
|
import unittest
|
|
|
|
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
|
|
|
|
|
2013-10-29 06:03:26 +08:00
|
|
|
from test.helper import (
|
|
|
|
get_params,
|
|
|
|
get_testcases,
|
|
|
|
try_rm,
|
|
|
|
md5,
|
|
|
|
report_warning
|
|
|
|
)
|
2013-10-15 08:00:53 +08:00
|
|
|
|
|
|
|
|
2012-09-28 21:34:56 +08:00
|
|
|
import hashlib
|
2012-12-12 21:15:21 +08:00
|
|
|
import io
|
2012-10-15 19:01:36 +08:00
|
|
|
import json
|
2012-12-20 23:30:55 +08:00
|
|
|
import socket
|
2012-11-28 22:09:56 +08:00
|
|
|
|
2013-06-19 04:14:21 +08:00
|
|
|
import youtube_dl.YoutubeDL
|
2013-10-15 08:00:53 +08:00
|
|
|
from youtube_dl.utils import (
|
2014-02-06 11:19:55 +08:00
|
|
|
compat_http_client,
|
2013-10-15 08:00:53 +08:00
|
|
|
compat_str,
|
|
|
|
compat_urllib_error,
|
2013-10-29 07:07:16 +08:00
|
|
|
compat_HTTPError,
|
2013-10-15 08:00:53 +08:00
|
|
|
DownloadError,
|
|
|
|
ExtractorError,
|
|
|
|
UnavailableVideoError,
|
|
|
|
)
|
2013-11-04 02:14:53 +08:00
|
|
|
from youtube_dl.extractor import get_info_extractor
|
2012-12-12 21:15:21 +08:00
|
|
|
|
2013-03-09 17:05:43 +08:00
|
|
|
RETRIES = 3
|
|
|
|
|
2013-06-19 04:14:21 +08:00
|
|
|
class YoutubeDL(youtube_dl.YoutubeDL):
|
2012-12-12 21:15:21 +08:00
|
|
|
def __init__(self, *args, **kwargs):
|
|
|
|
self.to_stderr = self.to_screen
|
2012-12-20 21:14:43 +08:00
|
|
|
self.processed_info_dicts = []
|
2013-06-19 04:14:21 +08:00
|
|
|
super(YoutubeDL, self).__init__(*args, **kwargs)
|
2013-06-06 21:07:05 +08:00
|
|
|
def report_warning(self, message):
|
2013-06-07 17:19:27 +08:00
|
|
|
# Don't accept warnings during tests
|
|
|
|
raise ExtractorError(message)
|
2012-12-20 21:14:43 +08:00
|
|
|
def process_info(self, info_dict):
|
|
|
|
self.processed_info_dicts.append(info_dict)
|
2013-06-19 04:14:21 +08:00
|
|
|
return super(YoutubeDL, self).process_info(info_dict)
|
2012-12-12 10:55:06 +08:00
|
|
|
|
2012-12-12 21:15:21 +08:00
|
|
|
def _file_md5(fn):
|
|
|
|
with open(fn, 'rb') as f:
|
|
|
|
return hashlib.md5(f.read()).hexdigest()
|
|
|
|
|
2013-06-28 03:15:16 +08:00
|
|
|
defs = get_testcases()
|
2013-06-28 00:28:45 +08:00
|
|
|
|
2012-12-20 21:14:43 +08:00
|
|
|
|
2012-12-12 10:55:06 +08:00
|
|
|
class TestDownload(unittest.TestCase):
|
2013-04-12 00:38:43 +08:00
|
|
|
maxDiff = None
|
2012-12-12 21:15:21 +08:00
|
|
|
def setUp(self):
|
|
|
|
self.defs = defs
|
|
|
|
|
2013-01-02 02:07:01 +08:00
|
|
|
### Dynamically generate tests
|
2012-12-12 22:14:58 +08:00
|
|
|
def generator(test_case):
|
|
|
|
|
2012-12-12 10:55:06 +08:00
|
|
|
def test_template(self):
|
2013-06-24 04:53:42 +08:00
|
|
|
ie = youtube_dl.extractor.get_info_extractor(test_case['name'])
|
2013-11-04 02:14:53 +08:00
|
|
|
other_ies = [get_info_extractor(ie_key) for ie_key in test_case.get('add_ie', [])]
|
2013-06-28 17:20:00 +08:00
|
|
|
def print_skipping(reason):
|
|
|
|
print('Skipping %s: %s' % (test_case['name'], reason))
|
2013-11-04 02:14:53 +08:00
|
|
|
if not ie.working():
|
2013-06-28 17:20:00 +08:00
|
|
|
print_skipping('IE marked as not _WORKING')
|
2012-12-12 21:15:21 +08:00
|
|
|
return
|
2013-10-29 05:01:37 +08:00
|
|
|
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
|
2012-12-12 21:15:21 +08:00
|
|
|
if 'skip' in test_case:
|
2013-06-28 17:20:00 +08:00
|
|
|
print_skipping(test_case['skip'])
|
2012-12-12 21:15:21 +08:00
|
|
|
return
|
2013-11-04 02:14:53 +08:00
|
|
|
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
|
2012-12-20 21:14:43 +08:00
|
|
|
|
2013-10-15 08:00:53 +08:00
|
|
|
params = get_params(test_case.get('params', {}))
|
2012-12-20 21:14:43 +08:00
|
|
|
|
2013-06-19 04:14:21 +08:00
|
|
|
ydl = YoutubeDL(params)
|
2013-06-28 05:51:06 +08:00
|
|
|
ydl.add_default_info_extractors()
|
2013-01-13 03:34:50 +08:00
|
|
|
finished_hook_called = set()
|
|
|
|
def _hook(status):
|
|
|
|
if status['status'] == 'finished':
|
|
|
|
finished_hook_called.add(status['filename'])
|
2013-12-23 17:37:27 +08:00
|
|
|
ydl.add_progress_hook(_hook)
|
2013-01-02 02:30:29 +08:00
|
|
|
|
2013-10-29 05:01:37 +08:00
|
|
|
def get_tc_filename(tc):
|
|
|
|
return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
|
|
|
|
|
2013-01-02 02:30:29 +08:00
|
|
|
test_cases = test_case.get('playlist', [test_case])
|
2013-10-29 05:01:37 +08:00
|
|
|
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')
|
2013-11-20 13:34:48 +08:00
|
|
|
try_rm(os.path.splitext(tc_filename)[0] + '.info.json')
|
2013-10-29 05:01:37 +08:00
|
|
|
try_rm_tcs_files()
|
2013-01-02 02:30:29 +08:00
|
|
|
try:
|
2013-10-29 06:03:26 +08:00
|
|
|
try_num = 1
|
|
|
|
while True:
|
2013-03-09 17:05:43 +08:00
|
|
|
try:
|
2013-06-19 04:14:21 +08:00
|
|
|
ydl.download([test_case['url']])
|
2013-03-09 17:05:43 +08:00
|
|
|
except (DownloadError, ExtractorError) as err:
|
|
|
|
# Check if the exception is not a network related one
|
2014-02-06 11:19:55 +08:00
|
|
|
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):
|
2013-03-09 17:05:43 +08:00
|
|
|
raise
|
|
|
|
|
2013-10-29 06:03:26 +08:00
|
|
|
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
|
2013-03-09 17:05:43 +08:00
|
|
|
else:
|
|
|
|
break
|
2013-01-02 02:30:29 +08:00
|
|
|
|
|
|
|
for tc in test_cases:
|
2013-10-29 05:01:37 +08:00
|
|
|
tc_filename = get_tc_filename(tc)
|
2013-01-02 04:01:49 +08:00
|
|
|
if not test_case.get('params', {}).get('skip_download', False):
|
2013-10-29 05:01:37 +08:00
|
|
|
self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename)
|
|
|
|
self.assertTrue(tc_filename in finished_hook_called)
|
2013-11-20 13:34:48 +08:00
|
|
|
info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json'
|
|
|
|
self.assertTrue(os.path.exists(info_json_fn))
|
2013-01-02 02:30:29 +08:00
|
|
|
if 'md5' in tc:
|
2013-10-29 05:01:37 +08:00
|
|
|
md5_for_file = _file_md5(tc_filename)
|
2013-01-02 02:30:29 +08:00
|
|
|
self.assertEqual(md5_for_file, tc['md5'])
|
2013-11-20 13:34:48 +08:00
|
|
|
with io.open(info_json_fn, encoding='utf-8') as infof:
|
2013-01-02 02:30:29 +08:00
|
|
|
info_dict = json.load(infof)
|
2013-06-24 03:33:11 +08:00
|
|
|
for (info_field, expected) in tc.get('info_dict', {}).items():
|
|
|
|
if isinstance(expected, compat_str) and expected.startswith('md5:'):
|
2013-08-29 18:15:45 +08:00
|
|
|
got = 'md5:' + md5(info_dict.get(info_field))
|
2013-06-09 20:21:42 +08:00
|
|
|
else:
|
2013-06-24 03:33:11 +08:00
|
|
|
got = info_dict.get(info_field)
|
2013-08-29 18:15:45 +08:00
|
|
|
self.assertEqual(expected, got,
|
|
|
|
u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
|
2013-06-09 20:21:42 +08:00
|
|
|
|
|
|
|
# If checkable fields are missing from the test case, print the info_dict
|
2013-06-09 21:03:54 +08:00
|
|
|
test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'md5:' + md5(value))
|
2013-06-09 20:21:42 +08:00
|
|
|
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()):
|
2014-01-17 09:53:34 +08:00
|
|
|
sys.stderr.write(u'\n"info_dict": ' + json.dumps(test_info_dict, ensure_ascii=False, indent=4) + u'\n')
|
2013-06-09 20:21:42 +08:00
|
|
|
|
|
|
|
# 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])
|
2013-11-03 19:11:13 +08:00
|
|
|
# Check for mandatory fields that are automatically set by YoutubeDL
|
2013-11-03 19:14:44 +08:00
|
|
|
for key in ['webpage_url', 'extractor', 'extractor_key']:
|
2013-11-03 19:11:13 +08:00
|
|
|
self.assertTrue(info_dict.get(key), u'Missing field: %s' % key)
|
2013-01-02 02:30:29 +08:00
|
|
|
finally:
|
2013-10-29 05:01:37 +08:00
|
|
|
try_rm_tcs_files()
|
2012-12-12 21:15:21 +08:00
|
|
|
|
2012-12-12 10:55:06 +08:00
|
|
|
return test_template
|
2012-12-12 21:15:21 +08:00
|
|
|
|
2012-12-12 22:14:58 +08:00
|
|
|
### And add them to TestDownload
|
2013-06-25 20:38:00 +08:00
|
|
|
for n, test_case in enumerate(defs):
|
2012-12-12 22:14:58 +08:00
|
|
|
test_method = generator(test_case)
|
2013-06-28 01:13:11 +08:00
|
|
|
tname = 'test_' + str(test_case['name'])
|
|
|
|
i = 1
|
|
|
|
while hasattr(TestDownload, tname):
|
2013-06-28 01:43:43 +08:00
|
|
|
tname = 'test_' + str(test_case['name']) + '_' + str(i)
|
2013-06-28 01:13:11 +08:00
|
|
|
i += 1
|
|
|
|
test_method.__name__ = tname
|
2012-12-12 21:15:21 +08:00
|
|
|
setattr(TestDownload, test_method.__name__, test_method)
|
2012-12-12 22:14:58 +08:00
|
|
|
del test_method
|
2012-11-28 22:09:56 +08:00
|
|
|
|
|
|
|
|
|
|
|
if __name__ == '__main__':
|
|
|
|
unittest.main()
|