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.FileDownloader
  17 import youtube_dl.InfoExtractors
  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 def crc32(value):
  42     return '%08x' % (binascii.crc32(value.encode('utf8')) & 0xffffffff)
  43
  44 class FileDownloader(youtube_dl.FileDownloader):
  45     def __init__(self, *args, **kwargs):
  46         self.to_stderr = self.to_screen
  47         self.processed_info_dicts = []
  48         return youtube_dl.FileDownloader.__init__(self, *args, **kwargs)
  49     def report_warning(self, message):
  50         # Don't accept warnings during tests
  51         raise ExtractorError(message)
  52     def process_info(self, info_dict):
  53         self.processed_info_dicts.append(info_dict)
  54         return youtube_dl.FileDownloader.process_info(self, info_dict)
  55
  56 def _file_md5(fn):
  57     with open(fn, 'rb') as f:
  58         return hashlib.md5(f.read()).hexdigest()
  59
  60 with io.open(DEF_FILE, encoding='utf-8') as deff:
  61     defs = json.load(deff)
  62 with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
  63     parameters = json.load(pf)
  64
  65
  66 class TestDownload(unittest.TestCase):
  67     maxDiff = None
  68     def setUp(self):
  69         self.parameters = parameters
  70         self.defs = defs
  71
  72 ### Dynamically generate tests
  73 def generator(test_case):
  74
  75     def test_template(self):
  76         ie = youtube_dl.InfoExtractors.get_info_extractor(test_case['name'])
  77         if not ie._WORKING:
  78             print('Skipping: IE marked as not _WORKING')
  79             return
  80         if 'playlist' not in test_case and not test_case['file']:
  81             print('Skipping: No output file specified')
  82             return
  83         if 'skip' in test_case:
  84             print('Skipping: {0}'.format(test_case['skip']))
  85             return
  86
  87         params = self.parameters.copy()
  88         params.update(test_case.get('params', {}))
  89
  90         fd = FileDownloader(params)
  91         for ie in youtube_dl.InfoExtractors.gen_extractors():
  92             fd.add_info_extractor(ie)
  93         finished_hook_called = set()
  94         def _hook(status):
  95             if status['status'] == 'finished':
  96                 finished_hook_called.add(status['filename'])
  97         fd.add_progress_hook(_hook)
  98
  99         test_cases = test_case.get('playlist', [test_case])
 100         for tc in test_cases:
 101             _try_rm(tc['file'])
 102             _try_rm(tc['file'] + '.part')
 103             _try_rm(tc['file'] + '.info.json')
 104         try:
 105             for retry in range(1, RETRIES + 1):
 106                 try:
 107                     fd.download([test_case['url']])
 108                 except (DownloadError, ExtractorError) as err:
 109                     if retry == RETRIES: raise
 110
 111                     # Check if the exception is not a network related one
 112                     if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
 113                         raise
 114
 115                     print('Retrying: {0} failed tries\n\n##########\n\n'.format(retry))
 116                 else:
 117                     break
 118
 119             for tc in test_cases:
 120                 if not test_case.get('params', {}).get('skip_download', False):
 121                     self.assertTrue(os.path.exists(tc['file']), msg='Missing file ' + tc['file'])
 122                     self.assertTrue(tc['file'] in finished_hook_called)
 123                 self.assertTrue(os.path.exists(tc['file'] + '.info.json'))
 124                 if 'md5' in tc:
 125                     md5_for_file = _file_md5(tc['file'])
 126                     self.assertEqual(md5_for_file, tc['md5'])
 127                 with io.open(tc['file'] + '.info.json', encoding='utf-8') as infof:
 128                     info_dict = json.load(infof)
 129                 for (info_field, value) in tc.get('info_dict', {}).items():
 130                     if isinstance(value, compat_str) and value.startswith('crc32:'):
 131                         self.assertEqual(value, 'crc32:' + crc32(info_dict.get(info_field)))
 132                     else:
 133                         self.assertEqual(value, info_dict.get(info_field))
 134
 135                 # If checkable fields are missing from the test case, print the info_dict
 136                 test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'crc32:' + crc32(value))
 137                     for key, value in info_dict.items()
 138                     if value and key in ('title', 'description', 'uploader', 'upload_date', 'uploader_id', 'location'))
 139                 if not all(key in tc.get('info_dict', {}).keys() for key in test_info_dict.keys()):
 140                     sys.stderr.write(u'\n"info_dict": ' + json.dumps(test_info_dict, ensure_ascii=False, indent=2) + u'\n')
 141
 142                 # Check for the presence of mandatory fields
 143                 for key in ('id', 'url', 'title', 'ext'):
 144                     self.assertTrue(key in info_dict.keys() and info_dict[key])
 145         finally:
 146             for tc in test_cases:
 147                 _try_rm(tc['file'])
 148                 _try_rm(tc['file'] + '.part')
 149                 _try_rm(tc['file'] + '.info.json')
 150
 151     return test_template
 152
 153 ### And add them to TestDownload
 154 for test_case in defs:
 155     test_method = generator(test_case)
 156     test_method.__name__ = "test_{0}".format(test_case["name"])
 157     setattr(TestDownload, test_method.__name__, test_method)
 158     del test_method
 159
 160
 161 if __name__ == '__main__':
 162     unittest.main()