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

   1 #!/usr/bin/env python
   2
   3 import errno
   4 import hashlib
   5 import io
   6 import os
   7 import json
   8 import unittest
   9 import sys
  10 import socket
  11 import binascii
  12
  13 # Allow direct execution
  14 sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
  15
  16 import youtube_dl.YoutubeDL
  17 import youtube_dl.extractor
  18 from youtube_dl.utils import *
  19
  20 DEF_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), 'tests.json')
  21 PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json")
  22
  23 RETRIES = 3
  24
  25 # General configuration (from __init__, not very elegant...)
  26 jar = compat_cookiejar.CookieJar()
  27 cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar)
  28 proxy_handler = compat_urllib_request.ProxyHandler()
  29 opener = compat_urllib_request.build_opener(proxy_handler, cookie_processor, YoutubeDLHandler())
  30 compat_urllib_request.install_opener(opener)
  31 socket.setdefaulttimeout(10)
  32
  33 def _try_rm(filename):
  34     """ Remove a file if it exists """
  35     try:
  36         os.remove(filename)
  37     except OSError as ose:
  38         if ose.errno != errno.ENOENT:
  39             raise
  40
  41 md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
  42
  43 class YoutubeDL(youtube_dl.YoutubeDL):
  44     def __init__(self, *args, **kwargs):
  45         self.to_stderr = self.to_screen
  46         self.processed_info_dicts = []
  47         super(YoutubeDL, self).__init__(*args, **kwargs)
  48     def report_warning(self, message):
  49         # Don't accept warnings during tests
  50         raise ExtractorError(message)
  51     def process_info(self, info_dict):
  52         self.processed_info_dicts.append(info_dict)
  53         return super(YoutubeDL, self).process_info(info_dict)
  54
  55 def _file_md5(fn):
  56     with open(fn, 'rb') as f:
  57         return hashlib.md5(f.read()).hexdigest()
  58
  59 with io.open(DEF_FILE, encoding='utf-8') as deff:
  60     defs = json.load(deff)
  61 for ie in youtube_dl.extractor.gen_extractors():
  62     t = getattr(ie, '_TEST', None)
  63     if t:
  64         t['name'] = type(ie).__name__[:-len('IE')]
  65         defs.append(t)
  66     for t in getattr(ie, '_TESTS', []):
  67         t['name'] = type(ie).__name__[:-len('IE')]
  68         defs.append(t)
  69
  70
  71 with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
  72     parameters = json.load(pf)
  73
  74
  75 class TestDownload(unittest.TestCase):
  76     maxDiff = None
  77     def setUp(self):
  78         self.parameters = parameters
  79         self.defs = defs
  80
  81 ### Dynamically generate tests
  82 def generator(test_case):
  83
  84     def test_template(self):
  85         ie = youtube_dl.extractor.get_info_extractor(test_case['name'])
  86         if not ie._WORKING:
  87             print('Skipping: IE marked as not _WORKING')
  88             return
  89         if 'playlist' not in test_case and not test_case['file']:
  90             print('Skipping: No output file specified')
  91             return
  92         if 'skip' in test_case:
  93             print('Skipping: {0}'.format(test_case['skip']))
  94             return
  95
  96         params = self.parameters.copy()
  97         params.update(test_case.get('params', {}))
  98
  99         ydl = YoutubeDL(params)
 100         for ie in youtube_dl.extractor.gen_extractors():
 101             ydl.add_info_extractor(ie)
 102         finished_hook_called = set()
 103         def _hook(status):
 104             if status['status'] == 'finished':
 105                 finished_hook_called.add(status['filename'])
 106         ydl.fd.add_progress_hook(_hook)
 107
 108         test_cases = test_case.get('playlist', [test_case])
 109         for tc in test_cases:
 110             _try_rm(tc['file'])
 111             _try_rm(tc['file'] + '.part')
 112             _try_rm(tc['file'] + '.info.json')
 113         try:
 114             for retry in range(1, RETRIES + 1):
 115                 try:
 116                     ydl.download([test_case['url']])
 117                 except (DownloadError, ExtractorError) as err:
 118                     if retry == RETRIES: raise
 119
 120                     # Check if the exception is not a network related one
 121                     if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
 122                         raise
 123
 124                     print('Retrying: {0} failed tries\n\n##########\n\n'.format(retry))
 125                 else:
 126                     break
 127
 128             for tc in test_cases:
 129                 if not test_case.get('params', {}).get('skip_download', False):
 130                     self.assertTrue(os.path.exists(tc['file']), msg='Missing file ' + tc['file'])
 131                     self.assertTrue(tc['file'] in finished_hook_called)
 132                 self.assertTrue(os.path.exists(tc['file'] + '.info.json'))
 133                 if 'md5' in tc:
 134                     md5_for_file = _file_md5(tc['file'])
 135                     self.assertEqual(md5_for_file, tc['md5'])
 136                 with io.open(tc['file'] + '.info.json', encoding='utf-8') as infof:
 137                     info_dict = json.load(infof)
 138                 for (info_field, expected) in tc.get('info_dict', {}).items():
 139                     if isinstance(expected, compat_str) and expected.startswith('md5:'):
 140                         self.assertEqual(expected, 'md5:' + md5(info_dict.get(info_field)))
 141                     else:
 142                         got = info_dict.get(info_field)
 143                         self.assertEqual(
 144                             expected, got,
 145                             u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
 146
 147                 # If checkable fields are missing from the test case, print the info_dict
 148                 test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'md5:' + md5(value))
 149                     for key, value in info_dict.items()
 150                     if value and key in ('title', 'description', 'uploader', 'upload_date', 'uploader_id', 'location'))
 151                 if not all(key in tc.get('info_dict', {}).keys() for key in test_info_dict.keys()):
 152                     sys.stderr.write(u'\n"info_dict": ' + json.dumps(test_info_dict, ensure_ascii=False, indent=2) + u'\n')
 153
 154                 # Check for the presence of mandatory fields
 155                 for key in ('id', 'url', 'title', 'ext'):
 156                     self.assertTrue(key in info_dict.keys() and info_dict[key])
 157         finally:
 158             for tc in test_cases:
 159                 _try_rm(tc['file'])
 160                 _try_rm(tc['file'] + '.part')
 161                 _try_rm(tc['file'] + '.info.json')
 162
 163     return test_template
 164
 165 ### And add them to TestDownload
 166 for n, test_case in enumerate(defs):
 167     test_method = generator(test_case)
 168     tname = 'test_' + str(test_case['name'])
 169     i = 1
 170     while hasattr(TestDownload, tname):
 171         tname = 'test_'  + str(test_case['name']) + '_' + str(i)
 172         i += 1
 173     test_method.__name__ = tname
 174     setattr(TestDownload, test_method.__name__, test_method)
 175     del test_method
 176
 177
 178 if __name__ == '__main__':
 179     unittest.main()