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

   1 #!/usr/bin/env python
   2
   3 # Allow direct execution
   4 import os
   5 import sys
   6 import unittest
   7 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
   8
   9 from test.helper import (
  10     get_params,
  11     get_testcases,
  12     global_setup,
  13     try_rm,
  14     md5,
  15     report_warning
  16 )
  17 global_setup()
  18
  19
  20 import hashlib
  21 import io
  22 import json
  23 import socket
  24
  25 import youtube_dl.YoutubeDL
  26 from youtube_dl.utils import (
  27     compat_str,
  28     compat_urllib_error,
  29     compat_HTTPError,
  30     DownloadError,
  31     ExtractorError,
  32     UnavailableVideoError,
  33 )
  34
  35 RETRIES = 3
  36
  37 class YoutubeDL(youtube_dl.YoutubeDL):
  38     def __init__(self, *args, **kwargs):
  39         self.to_stderr = self.to_screen
  40         self.processed_info_dicts = []
  41         super(YoutubeDL, self).__init__(*args, **kwargs)
  42     def report_warning(self, message):
  43         # Don't accept warnings during tests
  44         raise ExtractorError(message)
  45     def process_info(self, info_dict):
  46         self.processed_info_dicts.append(info_dict)
  47         return super(YoutubeDL, self).process_info(info_dict)
  48
  49 def _file_md5(fn):
  50     with open(fn, 'rb') as f:
  51         return hashlib.md5(f.read()).hexdigest()
  52
  53 defs = get_testcases()
  54
  55
  56 class TestDownload(unittest.TestCase):
  57     maxDiff = None
  58     def setUp(self):
  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:
  72             info_dict = test_case.get('info_dict', {})
  73             if not test_case.get('file') and not (info_dict.get('id') and info_dict.get('ext')):
  74                 print_skipping('The output file cannot be know, the "file" '
  75                     'key is missing or the info_dict is incomplete')
  76                 return
  77         if 'skip' in test_case:
  78             print_skipping(test_case['skip'])
  79             return
  80
  81         params = get_params(test_case.get('params', {}))
  82
  83         ydl = YoutubeDL(params)
  84         ydl.add_default_info_extractors()
  85         finished_hook_called = set()
  86         def _hook(status):
  87             if status['status'] == 'finished':
  88                 finished_hook_called.add(status['filename'])
  89         ydl.fd.add_progress_hook(_hook)
  90
  91         def get_tc_filename(tc):
  92             return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
  93
  94         test_cases = test_case.get('playlist', [test_case])
  95         def try_rm_tcs_files():
  96             for tc in test_cases:
  97                 tc_filename = get_tc_filename(tc)
  98                 try_rm(tc_filename)
  99                 try_rm(tc_filename + '.part')
 100                 try_rm(tc_filename + '.info.json')
 101         try_rm_tcs_files()
 102         try:
 103             try_num = 1
 104             while True:
 105                 try:
 106                     ydl.download([test_case['url']])
 107                 except (DownloadError, ExtractorError) as err:
 108                     # Check if the exception is not a network related one
 109                     if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503):
 110                         raise
 111
 112                     if try_num == RETRIES:
 113                         report_warning(u'Failed due to network errors, skipping...')
 114                         return
 115
 116                     print('Retrying: {0} failed tries\n\n##########\n\n'.format(try_num))
 117
 118                     try_num += 1
 119                 else:
 120                     break
 121
 122             for tc in test_cases:
 123                 tc_filename = get_tc_filename(tc)
 124                 if not test_case.get('params', {}).get('skip_download', False):
 125                     self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename)
 126                     self.assertTrue(tc_filename in finished_hook_called)
 127                 self.assertTrue(os.path.exists(tc_filename + '.info.json'))
 128                 if 'md5' in tc:
 129                     md5_for_file = _file_md5(tc_filename)
 130                     self.assertEqual(md5_for_file, tc['md5'])
 131                 with io.open(tc_filename + '.info.json', encoding='utf-8') as infof:
 132                     info_dict = json.load(infof)
 133                 for (info_field, expected) in tc.get('info_dict', {}).items():
 134                     if isinstance(expected, compat_str) and expected.startswith('md5:'):
 135                         got = 'md5:' + md5(info_dict.get(info_field))
 136                     else:
 137                         got = info_dict.get(info_field)
 138                     self.assertEqual(expected, got,
 139                         u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
 140
 141                 # If checkable fields are missing from the test case, print the info_dict
 142                 test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'md5:' + md5(value))
 143                     for key, value in info_dict.items()
 144                     if value and key in ('title', 'description', 'uploader', 'upload_date', 'uploader_id', 'location'))
 145                 if not all(key in tc.get('info_dict', {}).keys() for key in test_info_dict.keys()):
 146                     sys.stderr.write(u'\n"info_dict": ' + json.dumps(test_info_dict, ensure_ascii=False, indent=2) + u'\n')
 147
 148                 # Check for the presence of mandatory fields
 149                 for key in ('id', 'url', 'title', 'ext'):
 150                     self.assertTrue(key in info_dict.keys() and info_dict[key])
 151                 # Check for mandatory fields that are automatically set by YoutubeDL
 152                 for key in ['webpage_url', 'extractor', 'extractor_key']:
 153                     self.assertTrue(info_dict.get(key), u'Missing field: %s' % key)
 154         finally:
 155             try_rm_tcs_files()
 156
 157     return test_template
 158
 159 ### And add them to TestDownload
 160 for n, test_case in enumerate(defs):
 161     test_method = generator(test_case)
 162     tname = 'test_' + str(test_case['name'])
 163     i = 1
 164     while hasattr(TestDownload, tname):
 165         tname = 'test_'  + str(test_case['name']) + '_' + str(i)
 166         i += 1
 167     test_method.__name__ = tname
 168     setattr(TestDownload, test_method.__name__, test_method)
 169     del test_method
 170
 171
 172 if __name__ == '__main__':
 173     unittest.main()