2012-12-12 06:15:21 -07:00
|
|
|
#!/usr/bin/env python
|
|
|
|
|
2014-11-16 07:17:48 -07:00
|
|
|
from __future__ import unicode_literals
|
|
|
|
|
2013-10-14 18:00:53 -06:00
|
|
|
# Allow direct execution
|
|
|
|
import os
|
|
|
|
import sys
|
|
|
|
import unittest
|
|
|
|
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
|
|
|
|
|
2013-10-28 16:03:26 -06:00
|
|
|
from test.helper import (
|
2014-08-25 10:03:01 -06:00
|
|
|
assertGreaterEqual,
|
2014-10-26 13:49:51 -06:00
|
|
|
expect_warnings,
|
2013-10-28 16:03:26 -06:00
|
|
|
get_params,
|
2014-03-17 07:30:13 -06:00
|
|
|
gettestcases,
|
2014-03-23 08:52:21 -06:00
|
|
|
expect_info_dict,
|
|
|
|
try_rm,
|
|
|
|
report_warning,
|
2013-10-28 16:03:26 -06:00
|
|
|
)
|
2013-10-14 18:00:53 -06:00
|
|
|
|
|
|
|
|
2012-09-28 07:34:56 -06:00
|
|
|
import hashlib
|
2012-12-12 06:15:21 -07:00
|
|
|
import io
|
2012-10-15 05:01:36 -06:00
|
|
|
import json
|
2012-12-20 08:30:55 -07:00
|
|
|
import socket
|
2012-11-28 07:09:56 -07:00
|
|
|
|
2013-06-18 14:14:21 -06:00
|
|
|
import youtube_dl.YoutubeDL
|
2014-11-02 03:46:12 -07:00
|
|
|
from youtube_dl.compat import (
|
2014-02-05 20:19:55 -07:00
|
|
|
compat_http_client,
|
2013-10-14 18:00:53 -06:00
|
|
|
compat_urllib_error,
|
2013-10-28 17:07:16 -06:00
|
|
|
compat_HTTPError,
|
2014-11-02 03:46:12 -07:00
|
|
|
)
|
|
|
|
from youtube_dl.utils import (
|
2013-10-14 18:00:53 -06:00
|
|
|
DownloadError,
|
|
|
|
ExtractorError,
|
2014-08-28 05:59:42 -06:00
|
|
|
format_bytes,
|
2013-10-14 18:00:53 -06:00
|
|
|
UnavailableVideoError,
|
|
|
|
)
|
2013-11-03 11:14:53 -07:00
|
|
|
from youtube_dl.extractor import get_info_extractor
|
2012-12-12 06:15:21 -07:00
|
|
|
|
2013-03-09 02:05:43 -07:00
|
|
|
RETRIES = 3
|
|
|
|
|
2014-11-23 12:41:03 -07:00
|
|
|
|
2013-06-18 14:14:21 -06:00
|
|
|
class YoutubeDL(youtube_dl.YoutubeDL):
|
2012-12-12 06:15:21 -07:00
|
|
|
def __init__(self, *args, **kwargs):
|
|
|
|
self.to_stderr = self.to_screen
|
2012-12-20 06:14:43 -07:00
|
|
|
self.processed_info_dicts = []
|
2013-06-18 14:14:21 -06:00
|
|
|
super(YoutubeDL, self).__init__(*args, **kwargs)
|
2014-11-23 12:41:03 -07:00
|
|
|
|
2013-06-06 07:07:05 -06:00
|
|
|
def report_warning(self, message):
|
2013-06-07 03:19:27 -06:00
|
|
|
# Don't accept warnings during tests
|
|
|
|
raise ExtractorError(message)
|
2014-11-23 12:41:03 -07:00
|
|
|
|
2012-12-20 06:14:43 -07:00
|
|
|
def process_info(self, info_dict):
|
|
|
|
self.processed_info_dicts.append(info_dict)
|
2013-06-18 14:14:21 -06:00
|
|
|
return super(YoutubeDL, self).process_info(info_dict)
|
2012-12-11 19:55:06 -07:00
|
|
|
|
2014-11-23 12:41:03 -07:00
|
|
|
|
2012-12-12 06:15:21 -07:00
|
|
|
def _file_md5(fn):
|
|
|
|
with open(fn, 'rb') as f:
|
|
|
|
return hashlib.md5(f.read()).hexdigest()
|
|
|
|
|
2014-03-17 07:30:13 -06:00
|
|
|
defs = gettestcases()
|
2013-06-27 10:28:45 -06:00
|
|
|
|
2012-12-20 06:14:43 -07:00
|
|
|
|
2012-12-11 19:55:06 -07:00
|
|
|
class TestDownload(unittest.TestCase):
|
2013-04-11 10:38:43 -06:00
|
|
|
maxDiff = None
|
2014-11-23 12:41:03 -07:00
|
|
|
|
2012-12-12 06:15:21 -07:00
|
|
|
def setUp(self):
|
|
|
|
self.defs = defs
|
|
|
|
|
2014-11-23 12:41:03 -07:00
|
|
|
# Dynamically generate tests
|
|
|
|
|
|
|
|
|
2012-12-12 07:14:58 -07:00
|
|
|
def generator(test_case):
|
|
|
|
|
2012-12-11 19:55:06 -07:00
|
|
|
def test_template(self):
|
2013-06-23 14:53:42 -06:00
|
|
|
ie = youtube_dl.extractor.get_info_extractor(test_case['name'])
|
2013-11-03 11:14:53 -07:00
|
|
|
other_ies = [get_info_extractor(ie_key) for ie_key in test_case.get('add_ie', [])]
|
2014-08-21 03:52:07 -06:00
|
|
|
is_playlist = any(k.startswith('playlist') for k in test_case)
|
|
|
|
test_cases = test_case.get(
|
|
|
|
'playlist', [] if is_playlist else [test_case])
|
|
|
|
|
2013-06-28 03:20:00 -06:00
|
|
|
def print_skipping(reason):
|
|
|
|
print('Skipping %s: %s' % (test_case['name'], reason))
|
2013-11-03 11:14:53 -07:00
|
|
|
if not ie.working():
|
2013-06-28 03:20:00 -06:00
|
|
|
print_skipping('IE marked as not _WORKING')
|
2012-12-12 06:15:21 -07:00
|
|
|
return
|
2014-08-21 03:52:07 -06:00
|
|
|
|
|
|
|
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')):
|
2014-02-25 16:12:02 -07:00
|
|
|
raise Exception('Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?')
|
2014-08-21 03:52:07 -06:00
|
|
|
|
2012-12-12 06:15:21 -07:00
|
|
|
if 'skip' in test_case:
|
2013-06-28 03:20:00 -06:00
|
|
|
print_skipping(test_case['skip'])
|
2012-12-12 06:15:21 -07:00
|
|
|
return
|
2013-11-03 11:14:53 -07:00
|
|
|
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 06:14:43 -07:00
|
|
|
|
2013-10-14 18:00:53 -06:00
|
|
|
params = get_params(test_case.get('params', {}))
|
2014-08-21 03:52:07 -06:00
|
|
|
if is_playlist and 'playlist' not in test_case:
|
|
|
|
params.setdefault('extract_flat', True)
|
|
|
|
params.setdefault('skip_download', True)
|
2012-12-20 06:14:43 -07:00
|
|
|
|
2014-11-02 09:53:12 -07:00
|
|
|
ydl = YoutubeDL(params, auto_init=False)
|
2013-06-27 15:51:06 -06:00
|
|
|
ydl.add_default_info_extractors()
|
2013-01-12 12:34:50 -07:00
|
|
|
finished_hook_called = set()
|
2014-11-23 12:41:03 -07:00
|
|
|
|
2013-01-12 12:34:50 -07:00
|
|
|
def _hook(status):
|
|
|
|
if status['status'] == 'finished':
|
|
|
|
finished_hook_called.add(status['filename'])
|
2013-12-23 02:37:27 -07:00
|
|
|
ydl.add_progress_hook(_hook)
|
2014-10-26 13:49:51 -06:00
|
|
|
expect_warnings(ydl, test_case.get('expected_warnings', []))
|
2013-01-01 11:30:29 -07:00
|
|
|
|
2013-10-28 15:01:37 -06:00
|
|
|
def get_tc_filename(tc):
|
|
|
|
return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
|
|
|
|
|
2014-08-27 09:11:45 -06:00
|
|
|
res_dict = None
|
2014-11-23 12:41:03 -07:00
|
|
|
|
2014-08-27 09:11:45 -06:00
|
|
|
def try_rm_tcs_files(tcs=None):
|
|
|
|
if tcs is None:
|
|
|
|
tcs = test_cases
|
|
|
|
for tc in tcs:
|
2013-10-28 15:01:37 -06:00
|
|
|
tc_filename = get_tc_filename(tc)
|
|
|
|
try_rm(tc_filename)
|
|
|
|
try_rm(tc_filename + '.part')
|
2013-11-19 22:34:48 -07:00
|
|
|
try_rm(os.path.splitext(tc_filename)[0] + '.info.json')
|
2013-10-28 15:01:37 -06:00
|
|
|
try_rm_tcs_files()
|
2013-01-01 11:30:29 -07:00
|
|
|
try:
|
2013-10-28 16:03:26 -06:00
|
|
|
try_num = 1
|
|
|
|
while True:
|
2013-03-09 02:05:43 -07:00
|
|
|
try:
|
2014-08-21 03:52:07 -06:00
|
|
|
# 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'])
|
2013-03-09 02:05:43 -07:00
|
|
|
except (DownloadError, ExtractorError) as err:
|
|
|
|
# Check if the exception is not a network related one
|
2014-02-05 20:19:55 -07:00
|
|
|
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 02:05:43 -07:00
|
|
|
raise
|
|
|
|
|
2013-10-28 16:03:26 -06:00
|
|
|
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 02:05:43 -07:00
|
|
|
else:
|
|
|
|
break
|
2013-01-01 11:30:29 -07:00
|
|
|
|
2014-08-21 03:52:07 -06:00
|
|
|
if is_playlist:
|
|
|
|
self.assertEqual(res_dict['_type'], 'playlist')
|
2014-09-28 04:14:16 -06:00
|
|
|
self.assertTrue('entries' in res_dict)
|
2014-08-21 03:52:07 -06:00
|
|
|
expect_info_dict(self, test_case.get('info_dict', {}), res_dict)
|
2014-09-28 04:14:16 -06:00
|
|
|
|
2014-08-21 03:52:07 -06:00
|
|
|
if 'playlist_mincount' in test_case:
|
2014-08-25 10:03:01 -06:00
|
|
|
assertGreaterEqual(
|
|
|
|
self,
|
2014-08-21 03:52:07 -06:00
|
|
|
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'])))
|
2014-08-25 09:02:52 -06:00
|
|
|
if 'playlist_count' in test_case:
|
|
|
|
self.assertEqual(
|
|
|
|
len(res_dict['entries']),
|
|
|
|
test_case['playlist_count'],
|
2014-08-27 09:11:45 -06:00
|
|
|
'Expected %d entries in playlist %s, but got %d.' % (
|
2014-08-27 16:58:24 -06:00
|
|
|
test_case['playlist_count'],
|
2014-08-27 09:11:45 -06:00
|
|
|
test_case['url'],
|
2014-08-27 16:58:24 -06:00
|
|
|
len(res_dict['entries']),
|
|
|
|
))
|
2014-08-27 09:11:45 -06:00
|
|
|
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)
|
2014-08-21 03:52:07 -06:00
|
|
|
|
2013-01-01 11:30:29 -07:00
|
|
|
for tc in test_cases:
|
2013-10-28 15:01:37 -06:00
|
|
|
tc_filename = get_tc_filename(tc)
|
2013-01-01 13:01:49 -07:00
|
|
|
if not test_case.get('params', {}).get('skip_download', False):
|
2013-10-28 15:01:37 -06:00
|
|
|
self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename)
|
|
|
|
self.assertTrue(tc_filename in finished_hook_called)
|
2014-08-29 05:42:42 -06:00
|
|
|
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'])
|
2013-11-19 22:34:48 -07:00
|
|
|
info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json'
|
2014-10-26 17:39:39 -06:00
|
|
|
self.assertTrue(
|
|
|
|
os.path.exists(info_json_fn),
|
|
|
|
'Missing info file %s' % info_json_fn)
|
2013-11-19 22:34:48 -07:00
|
|
|
with io.open(info_json_fn, encoding='utf-8') as infof:
|
2013-01-01 11:30:29 -07:00
|
|
|
info_dict = json.load(infof)
|
2014-03-23 08:52:21 -06:00
|
|
|
|
|
|
|
expect_info_dict(self, tc.get('info_dict', {}), info_dict)
|
2013-01-01 11:30:29 -07:00
|
|
|
finally:
|
2013-10-28 15:01:37 -06:00
|
|
|
try_rm_tcs_files()
|
2014-09-28 04:14:16 -06:00
|
|
|
if is_playlist and res_dict is not None and res_dict.get('entries'):
|
2014-08-27 09:11:45 -06:00
|
|
|
# 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)
|
2012-12-12 06:15:21 -07:00
|
|
|
|
2012-12-11 19:55:06 -07:00
|
|
|
return test_template
|
2012-12-12 06:15:21 -07:00
|
|
|
|
2014-11-23 12:41:03 -07:00
|
|
|
# And add them to TestDownload
|
2013-06-25 06:38:00 -06:00
|
|
|
for n, test_case in enumerate(defs):
|
2012-12-12 07:14:58 -07:00
|
|
|
test_method = generator(test_case)
|
2013-06-27 11:13:11 -06:00
|
|
|
tname = 'test_' + str(test_case['name'])
|
|
|
|
i = 1
|
|
|
|
while hasattr(TestDownload, tname):
|
2014-11-16 07:17:48 -07:00
|
|
|
tname = 'test_%s_%d' % (test_case['name'], i)
|
2013-06-27 11:13:11 -06:00
|
|
|
i += 1
|
2014-11-16 07:17:48 -07:00
|
|
|
test_method.__name__ = str(tname)
|
2012-12-12 06:15:21 -07:00
|
|
|
setattr(TestDownload, test_method.__name__, test_method)
|
2012-12-12 07:14:58 -07:00
|
|
|
del test_method
|
2012-11-28 07:09:56 -07:00
|
|
|
|
|
|
|
|
|
|
|
if __name__ == '__main__':
|
|
|
|
unittest.main()
|