Annotate changelog with bugs closed.

[youtubedl] / test / test_download.py

diff --git a/test/test_download.py b/test/test_download.py
index b9a9be11d9686243ed2a1d5b748db4bc04712c54..d6540588c130f6bafacd4ef7d077e6debf8d911d 100644 (file)
--- a/test/test_download.py
+++ b/test/test_download.py
@@ -6,8 +6,13 @@ import sys
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 

-from test.helper import get_params, get_testcases, global_setup, try_rm, md5
-global_setup()
+from test.helper import (
+    get_params,
+    gettestcases,
+    expect_info_dict,
+    try_rm,
+    report_warning,
+)

 
 
 import hashlib
 
 
 import hashlib


@@ -17,12 +22,14 @@ import socket
 
 import youtube_dl.YoutubeDL
 from youtube_dl.utils import (
 
 import youtube_dl.YoutubeDL
 from youtube_dl.utils import (

-    compat_str,
+    compat_http_client,

     compat_urllib_error,
     compat_urllib_error,

+    compat_HTTPError,

     DownloadError,
     ExtractorError,
     UnavailableVideoError,
 )
     DownloadError,
     ExtractorError,
     UnavailableVideoError,
 )

+from youtube_dl.extractor import get_info_extractor

 
 RETRIES = 3
 
 
 RETRIES = 3
 


@@ -42,7 +49,7 @@ def _file_md5(fn):
     with open(fn, 'rb') as f:
         return hashlib.md5(f.read()).hexdigest()
 
     with open(fn, 'rb') as f:
         return hashlib.md5(f.read()).hexdigest()
 

-defs = get_testcases()
+defs = gettestcases()

 
 
 class TestDownload(unittest.TestCase):
 
 
 class TestDownload(unittest.TestCase):


@@ -55,17 +62,23 @@ def generator(test_case):
 
     def test_template(self):
         ie = youtube_dl.extractor.get_info_extractor(test_case['name'])
 
     def test_template(self):
         ie = youtube_dl.extractor.get_info_extractor(test_case['name'])

+        other_ies = [get_info_extractor(ie_key) for ie_key in test_case.get('add_ie', [])]

         def print_skipping(reason):
             print('Skipping %s: %s' % (test_case['name'], reason))
         def print_skipping(reason):
             print('Skipping %s: %s' % (test_case['name'], reason))

-        if not ie._WORKING:
+        if not ie.working():

             print_skipping('IE marked as not _WORKING')
             return
             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 'playlist' not in test_case:
+            info_dict = test_case.get('info_dict', {})
+            if not test_case.get('file') and not (info_dict.get('id') and info_dict.get('ext')):
+                raise Exception('Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?')

         if 'skip' in test_case:
             print_skipping(test_case['skip'])
             return
         if 'skip' in test_case:
             print_skipping(test_case['skip'])
             return

+        for other_ie in other_ies:
+            if not other_ie.working():
+                print_skipping(u'test depends on %sIE, marked as not WORKING' % other_ie.ie_key())
+                return

 
         params = get_params(test_case.get('params', {}))
 
 
         params = get_params(test_case.get('params', {}))
 


@@ -75,61 +88,55 @@ def generator(test_case):
         def _hook(status):
             if status['status'] == 'finished':
                 finished_hook_called.add(status['filename'])
         def _hook(status):
             if status['status'] == 'finished':
                 finished_hook_called.add(status['filename'])

-        ydl.fd.add_progress_hook(_hook)
+        ydl.add_progress_hook(_hook)
+
+        def get_tc_filename(tc):
+            return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))

 
         test_cases = test_case.get('playlist', [test_case])
 
         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')
+        def try_rm_tcs_files():
+            for tc in test_cases:
+                tc_filename = get_tc_filename(tc)
+                try_rm(tc_filename)
+                try_rm(tc_filename + '.part')
+                try_rm(os.path.splitext(tc_filename)[0] + '.info.json')
+        try_rm_tcs_files()

         try:
         try:

-            for retry in range(1, RETRIES + 1):
+            try_num = 1
+            while True:

                 try:
                     ydl.download([test_case['url']])
                 except (DownloadError, ExtractorError) as err:
                 try:
                     ydl.download([test_case['url']])
                 except (DownloadError, ExtractorError) as err:

-                    if retry == RETRIES: raise
-

                     # Check if the exception is not a network related one
                     # Check if the exception is not a network related one

-                    if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
+                    if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError, compat_http_client.BadStatusLine) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503):

                         raise
 
                         raise
 

-                    print('Retrying: {0} failed tries\n\n##########\n\n'.format(retry))
+                    if try_num == RETRIES:
+                        report_warning(u'Failed due to network errors, skipping...')
+                        return
+
+                    print('Retrying: {0} failed tries\n\n##########\n\n'.format(try_num))
+
+                    try_num += 1

                 else:
                     break
 
             for tc in test_cases:
                 else:
                     break
 
             for tc in test_cases:

+                tc_filename = get_tc_filename(tc)

                 if not test_case.get('params', {}).get('skip_download', False):
                 if not test_case.get('params', {}).get('skip_download', False):

-                    self.assertTrue(os.path.exists(tc['file']), msg='Missing file ' + tc['file'])
-                    self.assertTrue(tc['file'] in finished_hook_called)
-                self.assertTrue(os.path.exists(tc['file'] + '.info.json'))
+                    self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename)
+                    self.assertTrue(tc_filename in finished_hook_called)
+                info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json'
+                self.assertTrue(os.path.exists(info_json_fn))

                 if 'md5' in tc:
                 if 'md5' in tc:

-                    md5_for_file = _file_md5(tc['file'])
+                    md5_for_file = _file_md5(tc_filename)

                     self.assertEqual(md5_for_file, tc['md5'])
                     self.assertEqual(md5_for_file, tc['md5'])

-                with io.open(tc['file'] + '.info.json', encoding='utf-8') as infof:
+                with io.open(info_json_fn, encoding='utf-8') as infof:

                     info_dict = json.load(infof)
                     info_dict = json.load(infof)

-                for (info_field, expected) in tc.get('info_dict', {}).items():
-                    if isinstance(expected, compat_str) and expected.startswith('md5:'):
-                        got = 'md5:' + md5(info_dict.get(info_field))
-                    else:
-                        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])
+
+                expect_info_dict(self, tc.get('info_dict', {}), info_dict)

         finally:
         finally:

-            for tc in test_cases:
-                try_rm(tc['file'])
-                try_rm(tc['file'] + '.part')
-                try_rm(tc['file'] + '.info.json')
+            try_rm_tcs_files()

 
     return test_template
 
 
     return test_template