_ Git - youtube-dl/blob - test/test_download.py

   1 #!/usr/bin/env python
   2
   3 import hashlib
   4 import io
   5 import os
   6 import json
   7 import unittest
   8 import sys
   9 import socket
  10 import binascii
  11
  12 # Allow direct execution
  13 sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
  14
  15 import youtube_dl.YoutubeDL
  16 from youtube_dl.utils import *
  17
  18 PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json")
  19
  20 RETRIES = 3
  21
  22 # General configuration (from __init__, not very elegant...)
  23 jar = compat_cookiejar.CookieJar()
  24 cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar)
  25 proxy_handler = compat_urllib_request.ProxyHandler()
  26 opener = compat_urllib_request.build_opener(proxy_handler, cookie_processor, YoutubeDLHandler())
  27 compat_urllib_request.install_opener(opener)
  28 socket.setdefaulttimeout(10)
  29
  30 md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
  31
  32 class YoutubeDL(youtube_dl.YoutubeDL):
  33     def __init__(self, *args, **kwargs):
  34         self.to_stderr = self.to_screen
  35         self.processed_info_dicts = []
  36         super(YoutubeDL, self).__init__(*args, **kwargs)
  37     def report_warning(self, message):
  38         # Don't accept warnings during tests
  39         raise ExtractorError(message)
  40     def process_info(self, info_dict):
  41         self.processed_info_dicts.append(info_dict)
  42         return super(YoutubeDL, self).process_info(info_dict)
  43
  44 def _file_md5(fn):
  45     with open(fn, 'rb') as f:
  46         return hashlib.md5(f.read()).hexdigest()
  47
  48 from helper import get_testcases, try_rm
  49 defs = get_testcases()
  50
  51 with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
  52     parameters = json.load(pf)
  53
  54
  55 class TestDownload(unittest.TestCase):
  56     maxDiff = None
  57     def setUp(self):
  58         self.parameters = parameters
  59         self.defs = defs
  60
  61 ### Dynamically generate tests
  62 def generator(test_case):
  63
  64     def test_template(self):
  65         ie = youtube_dl.extractor.get_info_extractor(test_case['name'])
  66         def print_skipping(reason):
  67             print('Skipping %s: %s' % (test_case['name'], reason))
  68         if not ie._WORKING:
  69             print_skipping('IE marked as not _WORKING')
  70             return
  71         if 'playlist' not in test_case and not test_case['file']:
  72             print_skipping('No output file specified')
  73             return
  74         if 'skip' in test_case:
  75             print_skipping(test_case['skip'])
  76             return
  77
  78         params = self.parameters.copy()
  79         params.update(test_case.get('params', {}))
  80
  81         ydl = YoutubeDL(params)
  82         ydl.add_default_info_extractors()
  83         finished_hook_called = set()
  84         def _hook(status):
  85             if status['status'] == 'finished':
  86                 finished_hook_called.add(status['filename'])
  87         ydl.fd.add_progress_hook(_hook)
  88
  89         test_cases = test_case.get('playlist', [test_case])
  90         for tc in test_cases:
  91             try_rm(tc['file'])
  92             try_rm(tc['file'] + '.part')
  93             try_rm(tc['file'] + '.info.json')
  94         try:
  95             for retry in range(1, RETRIES + 1):
  96                 try:
  97                     ydl.download([test_case['url']])
  98                 except (DownloadError, ExtractorError) as err:
  99                     if retry == RETRIES: raise
 100
 101                     # Check if the exception is not a network related one
 102                     if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
 103                         raise
 104
 105                     print('Retrying: {0} failed tries\n\n##########\n\n'.format(retry))
 106                 else:
 107                     break
 108
 109             for tc in test_cases:
 110                 if not test_case.get('params', {}).get('skip_download', False):
 111                     self.assertTrue(os.path.exists(tc['file']), msg='Missing file ' + tc['file'])
 112                     self.assertTrue(tc['file'] in finished_hook_called)
 113                 self.assertTrue(os.path.exists(tc['file'] + '.info.json'))
 114                 if 'md5' in tc:
 115                     md5_for_file = _file_md5(tc['file'])
 116                     self.assertEqual(md5_for_file, tc['md5'])
 117                 with io.open(tc['file'] + '.info.json', encoding='utf-8') as infof:
 118                     info_dict = json.load(infof)
 119                 for (info_field, expected) in tc.get('info_dict', {}).items():
 120                     if isinstance(expected, compat_str) and expected.startswith('md5:'):
 121                         got = 'md5:' + md5(info_dict.get(info_field))
 122                     else:
 123                         got = info_dict.get(info_field)
 124                     self.assertEqual(expected, got,
 125                         u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
 126
 127                 # If checkable fields are missing from the test case, print the info_dict
 128                 test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'md5:' + md5(value))
 129                     for key, value in info_dict.items()
 130                     if value and key in ('title', 'description', 'uploader', 'upload_date', 'uploader_id', 'location'))
 131                 if not all(key in tc.get('info_dict', {}).keys() for key in test_info_dict.keys()):
 132                     sys.stderr.write(u'\n"info_dict": ' + json.dumps(test_info_dict, ensure_ascii=False, indent=2) + u'\n')
 133
 134                 # Check for the presence of mandatory fields
 135                 for key in ('id', 'url', 'title', 'ext'):
 136                     self.assertTrue(key in info_dict.keys() and info_dict[key])
 137         finally:
 138             for tc in test_cases:
 139                 try_rm(tc['file'])
 140                 try_rm(tc['file'] + '.part')
 141                 try_rm(tc['file'] + '.info.json')
 142
 143     return test_template
 144
 145 ### And add them to TestDownload
 146 for n, test_case in enumerate(defs):
 147     test_method = generator(test_case)
 148     tname = 'test_' + str(test_case['name'])
 149     i = 1
 150     while hasattr(TestDownload, tname):
 151         tname = 'test_'  + str(test_case['name']) + '_' + str(i)
 152         i += 1
 153     test_method.__name__ = tname
 154     setattr(TestDownload, test_method.__name__, test_method)
 155     del test_method
 156
 157
 158 if __name__ == '__main__':
 159     unittest.main()