yt-dlp/test/test_download.py

165 lines
6.2 KiB
Python
Raw Normal View History

#!/usr/bin/env python
import errno
2012-09-28 09:34:56 -04:00
import hashlib
import io
2012-09-28 09:34:56 -04:00
import os
2012-10-15 07:01:36 -04:00
import json
2012-11-28 09:09:56 -05:00
import unittest
import sys
import socket
import binascii
# Allow direct execution
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
2012-11-28 09:09:56 -05:00
import youtube_dl.FileDownloader
import youtube_dl.InfoExtractors
from youtube_dl.utils import *
2012-12-11 21:55:06 -05:00
DEF_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), 'tests.json')
PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json")
RETRIES = 3
# General configuration (from __init__, not very elegant...)
jar = compat_cookiejar.CookieJar()
cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar)
proxy_handler = compat_urllib_request.ProxyHandler()
opener = compat_urllib_request.build_opener(proxy_handler, cookie_processor, YoutubeDLHandler())
compat_urllib_request.install_opener(opener)
socket.setdefaulttimeout(10)
def _try_rm(filename):
""" Remove a file if it exists """
try:
os.remove(filename)
except OSError as ose:
if ose.errno != errno.ENOENT:
raise
md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
class FileDownloader(youtube_dl.FileDownloader):
def __init__(self, *args, **kwargs):
self.to_stderr = self.to_screen
2012-12-20 08:14:43 -05:00
self.processed_info_dicts = []
return youtube_dl.FileDownloader.__init__(self, *args, **kwargs)
def report_warning(self, message):
# Don't accept warnings during tests
raise ExtractorError(message)
2012-12-20 08:14:43 -05:00
def process_info(self, info_dict):
self.processed_info_dicts.append(info_dict)
return youtube_dl.FileDownloader.process_info(self, info_dict)
2012-12-11 21:55:06 -05:00
def _file_md5(fn):
with open(fn, 'rb') as f:
return hashlib.md5(f.read()).hexdigest()
with io.open(DEF_FILE, encoding='utf-8') as deff:
defs = json.load(deff)
with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
parameters = json.load(pf)
2012-12-11 21:55:06 -05:00
2012-12-20 08:14:43 -05:00
2012-12-11 21:55:06 -05:00
class TestDownload(unittest.TestCase):
2013-04-11 12:38:43 -04:00
maxDiff = None
def setUp(self):
self.parameters = parameters
self.defs = defs
2013-01-01 13:07:01 -05:00
### Dynamically generate tests
def generator(test_case):
2012-12-11 21:55:06 -05:00
def test_template(self):
ie = youtube_dl.InfoExtractors.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
2012-12-20 08:14:43 -05:00
2013-01-01 13:34:54 -05:00
params = self.parameters.copy()
params.update(test_case.get('params', {}))
2012-12-20 08:14:43 -05:00
fd = FileDownloader(params)
for ie in youtube_dl.InfoExtractors.gen_extractors():
fd.add_info_extractor(ie)
2013-01-12 14:34:50 -05:00
finished_hook_called = set()
def _hook(status):
if status['status'] == 'finished':
finished_hook_called.add(status['filename'])
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:
fd.download([test_case['url']])
except (DownloadError, ExtractorError) as err:
if retry == RETRIES: raise
# Check if the exception is not a network related one
2013-03-30 21:29:34 -04:00
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:
2013-01-01 15:01:49 -05:00
if not test_case.get('params', {}).get('skip_download', False):
self.assertTrue(os.path.exists(tc['file']), msg='Missing file ' + tc['file'])
2013-01-12 14:34:50 -05:00
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)
2013-06-23 15:33:11 -04:00
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:
2013-06-23 15:33:11 -04:00
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')
2012-12-11 21:55:06 -05:00
return test_template
### And add them to TestDownload
for test_case in defs:
test_method = generator(test_case)
test_method.__name__ = "test_{0}".format(test_case["name"])
setattr(TestDownload, test_method.__name__, test_method)
del test_method
2012-11-28 09:09:56 -05:00
if __name__ == '__main__':
unittest.main()