[YoutubeDL] guard against strange sys.stdouts

[youtube-dl] / test / test_download.py

diff --git a/test/test_download.py b/test/test_download.py
index 8fccdaf9e079e23de57856457fbc865be844fecd..87aced97cfe6e25750183e2fd94d4c2392703d9b 100644 (file)
--- a/test/test_download.py
+++ b/test/test_download.py
@@ -7,28 +7,31 @@ import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 from test.helper import (
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 from test.helper import (

+    assertGreaterEqual,
+    expect_warnings,

     get_params,
     gettestcases,
     get_params,
     gettestcases,

+    expect_info_dict,

     try_rm,
     try_rm,

-    md5,
-    report_warning
+    report_warning,

 )
 
 
 import hashlib
 import io
 import json
 )
 
 
 import hashlib
 import io
 import json

-import re

 import socket
 
 import youtube_dl.YoutubeDL
 import socket
 
 import youtube_dl.YoutubeDL

-from youtube_dl.utils import (
+from youtube_dl.compat import (

     compat_http_client,
     compat_http_client,

-    compat_str,

     compat_urllib_error,
     compat_HTTPError,
     compat_urllib_error,
     compat_HTTPError,

+)
+from youtube_dl.utils import (

     DownloadError,
     ExtractorError,
     DownloadError,
     ExtractorError,

+    format_bytes,

     UnavailableVideoError,
 )
 from youtube_dl.extractor import get_info_extractor
     UnavailableVideoError,
 )
 from youtube_dl.extractor import get_info_extractor


@@ -65,15 +68,21 @@ def generator(test_case):
     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 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', [])]

+        is_playlist = any(k.startswith('playlist') for k in test_case)
+        test_cases = test_case.get(
+            'playlist', [] if is_playlist else [test_case])
+

         def print_skipping(reason):
             print('Skipping %s: %s' % (test_case['name'], reason))
         if not ie.working():
             print_skipping('IE marked as not _WORKING')
             return
         def print_skipping(reason):
             print('Skipping %s: %s' % (test_case['name'], reason))
         if not ie.working():
             print_skipping('IE marked as not _WORKING')
             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')):
+
+        for tc in test_cases:
+            info_dict = tc.get('info_dict', {})
+            if not tc.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?')
                 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


@@ -83,21 +92,27 @@ def generator(test_case):
                 return
 
         params = get_params(test_case.get('params', {}))
                 return
 
         params = get_params(test_case.get('params', {}))

+        if is_playlist and 'playlist' not in test_case:
+            params.setdefault('extract_flat', True)
+            params.setdefault('skip_download', True)

 
 

-        ydl = YoutubeDL(params)
+        ydl = YoutubeDL(params, auto_init=False)

         ydl.add_default_info_extractors()
         finished_hook_called = set()
         def _hook(status):
             if status['status'] == 'finished':
                 finished_hook_called.add(status['filename'])
         ydl.add_progress_hook(_hook)
         ydl.add_default_info_extractors()
         finished_hook_called = set()
         def _hook(status):
             if status['status'] == 'finished':
                 finished_hook_called.add(status['filename'])
         ydl.add_progress_hook(_hook)

+        expect_warnings(ydl, test_case.get('expected_warnings', []))

 
         def get_tc_filename(tc):
             return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
 
 
         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])
-        def try_rm_tcs_files():
-            for tc in test_cases:
+        res_dict = None
+        def try_rm_tcs_files(tcs=None):
+            if tcs is None:
+                tcs = test_cases
+            for tc in tcs:

                 tc_filename = get_tc_filename(tc)
                 try_rm(tc_filename)
                 try_rm(tc_filename + '.part')
                 tc_filename = get_tc_filename(tc)
                 try_rm(tc_filename)
                 try_rm(tc_filename + '.part')


@@ -107,7 +122,10 @@ def generator(test_case):
             try_num = 1
             while True:
                 try:
             try_num = 1
             while True:
                 try:

-                    ydl.download([test_case['url']])
+                    # We're not using .download here sine that is just a shim
+                    # for outside error handling, and returns the exit code
+                    # instead of the result dict.
+                    res_dict = ydl.extract_info(test_case['url'])

                 except (DownloadError, ExtractorError) as err:
                     # Check if the exception is not a network related one
                     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):
                 except (DownloadError, ExtractorError) as err:
                     # Check if the exception is not a network related one
                     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):


@@ -123,54 +141,66 @@ def generator(test_case):
                 else:
                     break
 
                 else:
                     break
 

+            if is_playlist:
+                self.assertEqual(res_dict['_type'], 'playlist')
+                self.assertTrue('entries' in res_dict)
+                expect_info_dict(self, test_case.get('info_dict', {}), res_dict)
+
+            if 'playlist_mincount' in test_case:
+                assertGreaterEqual(
+                    self,
+                    len(res_dict['entries']),
+                    test_case['playlist_mincount'],
+                    'Expected at least %d in playlist %s, but got only %d' % (
+                        test_case['playlist_mincount'], test_case['url'],
+                        len(res_dict['entries'])))
+            if 'playlist_count' in test_case:
+                self.assertEqual(
+                    len(res_dict['entries']),
+                    test_case['playlist_count'],
+                    'Expected %d entries in playlist %s, but got %d.' % (
+                        test_case['playlist_count'],
+                        test_case['url'],
+                        len(res_dict['entries']),
+                    ))
+            if 'playlist_duration_sum' in test_case:
+                got_duration = sum(e['duration'] for e in res_dict['entries'])
+                self.assertEqual(
+                    test_case['playlist_duration_sum'], got_duration)
+

             for tc in test_cases:
                 tc_filename = get_tc_filename(tc)
                 if not test_case.get('params', {}).get('skip_download', False):
                     self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename)
                     self.assertTrue(tc_filename in finished_hook_called)
             for tc in test_cases:
                 tc_filename = get_tc_filename(tc)
                 if not test_case.get('params', {}).get('skip_download', False):
                     self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename)
                     self.assertTrue(tc_filename in finished_hook_called)

+                    expected_minsize = tc.get('file_minsize', 10000)
+                    if expected_minsize is not None:
+                        if params.get('test'):
+                            expected_minsize = max(expected_minsize, 10000)
+                        got_fsize = os.path.getsize(tc_filename)
+                        assertGreaterEqual(
+                            self, got_fsize, expected_minsize,
+                            'Expected %s to be at least %s, but it\'s only %s ' %
+                            (tc_filename, format_bytes(expected_minsize),
+                                format_bytes(got_fsize)))
+                    if 'md5' in tc:
+                        md5_for_file = _file_md5(tc_filename)
+                        self.assertEqual(md5_for_file, tc['md5'])

                 info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json'
                 info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json'

-                self.assertTrue(os.path.exists(info_json_fn))
-                if 'md5' in tc:
-                    md5_for_file = _file_md5(tc_filename)
-                    self.assertEqual(md5_for_file, tc['md5'])
+                self.assertTrue(
+                    os.path.exists(info_json_fn),
+                    'Missing info file %s' % info_json_fn)

                 with io.open(info_json_fn, encoding='utf-8') as infof:
                     info_dict = json.load(infof)
                 with io.open(info_json_fn, encoding='utf-8') as infof:
                     info_dict = json.load(infof)

-                for (info_field, expected) in tc.get('info_dict', {}).items():
-                    if isinstance(expected, compat_str) and expected.startswith('re:'):
-                        got = info_dict.get(info_field)
-                        match_str = expected[len('re:'):]
-                        match_rex = re.compile(match_str)
-
-                        self.assertTrue(
-                            isinstance(got, compat_str) and match_rex.match(got),
-                            u'field %s (value: %r) should match %r' % (info_field, got, match_str))
-                    elif isinstance(expected, type):
-                        got = info_dict.get(info_field)
-                        self.assertTrue(isinstance(got, expected),
-                            u'Expected type %r, but got value %r of type %r' % (expected, got, type(got)))
-                    else:
-                        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))
-
-                # 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])
-                # Check for mandatory fields that are automatically set by YoutubeDL
-                for key in ['webpage_url', 'extractor', 'extractor_key']:
-                    self.assertTrue(info_dict.get(key), u'Missing field: %s' % key)
-
-                # 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', 'timestamp', '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=4) + u'\n')
+
+                expect_info_dict(self, tc.get('info_dict', {}), info_dict)

         finally:
             try_rm_tcs_files()
         finally:
             try_rm_tcs_files()

+            if is_playlist and res_dict is not None and res_dict.get('entries'):
+                # Remove all other files that may have been extracted if the
+                # extractor returns full results even with extract_flat
+                res_tcs = [{'info_dict': e} for e in res_dict['entries']]
+                try_rm_tcs_files(res_tcs)

 
     return test_template
 
 
     return test_template