]> jfr.im git - yt-dlp.git/blobdiff - test/test_YoutubeDL.py
[utils] Add temporary shim for logging
[yt-dlp.git] / test / test_YoutubeDL.py
index 426e52305d72f6751c786b6bbf7a703317e8c34f..c15c7704c54ea405acb4d1c84b418bfc6201cc8f 100644 (file)
@@ -10,7 +10,6 @@
 
 import copy
 import json
-import urllib.error
 
 from test.helper import FakeYDL, assertRegexpMatches
 from yt_dlp import YoutubeDL
@@ -68,8 +67,7 @@ def test_prefer_free_formats(self):
             {'ext': 'mp4', 'height': 460, 'url': TEST_URL},
         ]
         info_dict = _make_result(formats)
-        yie = YoutubeIE(ydl)
-        yie._sort_formats(info_dict['formats'])
+        ydl.sort_formats(info_dict)
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['ext'], 'webm')
@@ -82,8 +80,7 @@ def test_prefer_free_formats(self):
             {'ext': 'mp4', 'height': 1080, 'url': TEST_URL},
         ]
         info_dict['formats'] = formats
-        yie = YoutubeIE(ydl)
-        yie._sort_formats(info_dict['formats'])
+        ydl.sort_formats(info_dict)
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['ext'], 'mp4')
@@ -97,8 +94,7 @@ def test_prefer_free_formats(self):
             {'ext': 'flv', 'height': 720, 'url': TEST_URL},
         ]
         info_dict['formats'] = formats
-        yie = YoutubeIE(ydl)
-        yie._sort_formats(info_dict['formats'])
+        ydl.sort_formats(info_dict)
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['ext'], 'mp4')
@@ -110,15 +106,14 @@ def test_prefer_free_formats(self):
             {'ext': 'webm', 'height': 720, 'url': TEST_URL},
         ]
         info_dict['formats'] = formats
-        yie = YoutubeIE(ydl)
-        yie._sort_formats(info_dict['formats'])
+        ydl.sort_formats(info_dict)
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['ext'], 'webm')
 
     def test_format_selection(self):
         formats = [
-            {'format_id': '35', 'ext': 'mp4', 'preference': 1, 'url': TEST_URL},
+            {'format_id': '35', 'ext': 'mp4', 'preference': 0, 'url': TEST_URL},
             {'format_id': 'example-with-dashes', 'ext': 'webm', 'preference': 1, 'url': TEST_URL},
             {'format_id': '45', 'ext': 'webm', 'preference': 2, 'url': TEST_URL},
             {'format_id': '47', 'ext': 'webm', 'preference': 3, 'url': TEST_URL},
@@ -186,22 +181,19 @@ def test_format_selection_audio_exts(self):
 
         info_dict = _make_result(formats)
         ydl = YDL({'format': 'best'})
-        ie = YoutubeIE(ydl)
-        ie._sort_formats(info_dict['formats'])
+        ydl.sort_formats(info_dict)
         ydl.process_ie_result(copy.deepcopy(info_dict))
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['format_id'], 'aac-64')
 
         ydl = YDL({'format': 'mp3'})
-        ie = YoutubeIE(ydl)
-        ie._sort_formats(info_dict['formats'])
+        ydl.sort_formats(info_dict)
         ydl.process_ie_result(copy.deepcopy(info_dict))
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['format_id'], 'mp3-64')
 
         ydl = YDL({'prefer_free_formats': True})
-        ie = YoutubeIE(ydl)
-        ie._sort_formats(info_dict['formats'])
+        ydl.sort_formats(info_dict)
         ydl.process_ie_result(copy.deepcopy(info_dict))
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['format_id'], 'ogg-64')
@@ -346,8 +338,7 @@ def format_info(f_id):
 
         info_dict = _make_result(list(formats_order), extractor='youtube')
         ydl = YDL({'format': 'bestvideo+bestaudio'})
-        yie = YoutubeIE(ydl)
-        yie._sort_formats(info_dict['formats'])
+        ydl.sort_formats(info_dict)
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['format_id'], '248+172')
@@ -355,40 +346,35 @@ def format_info(f_id):
 
         info_dict = _make_result(list(formats_order), extractor='youtube')
         ydl = YDL({'format': 'bestvideo[height>=999999]+bestaudio/best'})
-        yie = YoutubeIE(ydl)
-        yie._sort_formats(info_dict['formats'])
+        ydl.sort_formats(info_dict)
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['format_id'], '38')
 
         info_dict = _make_result(list(formats_order), extractor='youtube')
         ydl = YDL({'format': 'bestvideo/best,bestaudio'})
-        yie = YoutubeIE(ydl)
-        yie._sort_formats(info_dict['formats'])
+        ydl.sort_formats(info_dict)
         ydl.process_ie_result(info_dict)
         downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
         self.assertEqual(downloaded_ids, ['137', '141'])
 
         info_dict = _make_result(list(formats_order), extractor='youtube')
         ydl = YDL({'format': '(bestvideo[ext=mp4],bestvideo[ext=webm])+bestaudio'})
-        yie = YoutubeIE(ydl)
-        yie._sort_formats(info_dict['formats'])
+        ydl.sort_formats(info_dict)
         ydl.process_ie_result(info_dict)
         downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
         self.assertEqual(downloaded_ids, ['137+141', '248+141'])
 
         info_dict = _make_result(list(formats_order), extractor='youtube')
         ydl = YDL({'format': '(bestvideo[ext=mp4],bestvideo[ext=webm])[height<=720]+bestaudio'})
-        yie = YoutubeIE(ydl)
-        yie._sort_formats(info_dict['formats'])
+        ydl.sort_formats(info_dict)
         ydl.process_ie_result(info_dict)
         downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
         self.assertEqual(downloaded_ids, ['136+141', '247+141'])
 
         info_dict = _make_result(list(formats_order), extractor='youtube')
         ydl = YDL({'format': '(bestvideo[ext=none]/bestvideo[ext=webm])+bestaudio'})
-        yie = YoutubeIE(ydl)
-        yie._sort_formats(info_dict['formats'])
+        ydl.sort_formats(info_dict)
         ydl.process_ie_result(info_dict)
         downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
         self.assertEqual(downloaded_ids, ['248+141'])
@@ -396,16 +382,14 @@ def format_info(f_id):
         for f1, f2 in zip(formats_order, formats_order[1:]):
             info_dict = _make_result([f1, f2], extractor='youtube')
             ydl = YDL({'format': 'best/bestvideo'})
-            yie = YoutubeIE(ydl)
-            yie._sort_formats(info_dict['formats'])
+            ydl.sort_formats(info_dict)
             ydl.process_ie_result(info_dict)
             downloaded = ydl.downloaded_info_dicts[0]
             self.assertEqual(downloaded['format_id'], f1['format_id'])
 
             info_dict = _make_result([f2, f1], extractor='youtube')
             ydl = YDL({'format': 'best/bestvideo'})
-            yie = YoutubeIE(ydl)
-            yie._sort_formats(info_dict['formats'])
+            ydl.sort_formats(info_dict)
             ydl.process_ie_result(info_dict)
             downloaded = ydl.downloaded_info_dicts[0]
             self.assertEqual(downloaded['format_id'], f1['format_id'])
@@ -480,7 +464,7 @@ def test_format_filtering(self):
         for f in formats:
             f['url'] = 'http://_/'
             f['ext'] = 'unknown'
-        info_dict = _make_result(formats)
+        info_dict = _make_result(formats, _format_sort_fields=('id', ))
 
         ydl = YDL({'format': 'best[filesize<3000]'})
         ydl.process_ie_result(info_dict)
@@ -646,6 +630,7 @@ def test_add_extra_info(self):
         self.assertEqual(test_dict['playlist'], 'funny videos')
 
     outtmpl_info = {
+        'id': '1234',
         'id': '1234',
         'ext': 'mp4',
         'width': None,
@@ -662,7 +647,11 @@ def test_add_extra_info(self):
         'playlist_autonumber': 2,
         '__last_playlist_index': 100,
         'n_entries': 10,
-        'formats': [{'id': 'id 1'}, {'id': 'id 2'}, {'id': 'id 3'}]
+        'formats': [
+            {'id': 'id 1', 'height': 1080, 'width': 1920},
+            {'id': 'id 2', 'height': 720},
+            {'id': 'id 3'}
+        ]
     }
 
     def test_prepare_outtmpl_and_filename(self):
@@ -680,7 +669,7 @@ def test(tmpl, expected, *, info=None, **params):
             for (name, got), expect in zip((('outtmpl', out), ('filename', fname)), expected):
                 if callable(expect):
                     self.assertTrue(expect(got), f'Wrong {name} from {tmpl}')
-                else:
+                elif expect is not None:
                     self.assertEqual(got, expect, f'Wrong {name} from {tmpl}')
 
         # Side-effects
@@ -695,7 +684,7 @@ def test(tmpl, expected, *, info=None, **params):
         test('%(id)s.%(ext)s', '1234.mp4')
         test('%(duration_string)s', ('27:46:40', '27-46-40'))
         test('%(resolution)s', '1080p')
-        test('%(playlist_index)s', '001')
+        test('%(playlist_index|)s', '001')
         test('%(playlist_autonumber)s', '02')
         test('%(autonumber)s', '00001')
         test('%(autonumber+2)03d', '005', autonumber_start=3)
@@ -729,6 +718,7 @@ def test(tmpl, expected, *, info=None, **params):
         self.assertTrue(isinstance(YoutubeDL.validate_outtmpl('%(title)'), ValueError))
         test('%(invalid@tmpl|def)s', 'none', outtmpl_na_placeholder='none')
         test('%(..)s', 'NA')
+        test('%(formats.{id)s', 'NA')
 
         # Entire info_dict
         def expect_same_infodict(out):
@@ -765,20 +755,23 @@ def expect_same_infodict(out):
         test('%(ext)c', 'm')
         test('%(id)d %(id)r', "1234 '1234'")
         test('%(id)r %(height)r', "'1234' 1080")
+        test('%(title5)a %(height)a', (R"'\xe1\xe9\xed \U0001d400' 1080", None))
         test('%(ext)s-%(ext|def)d', 'mp4-def')
-        test('%(width|0)04d', '0000')
-        test('a%(width|)d', 'a', outtmpl_na_placeholder='none')
+        test('%(width|0)04d', '0')
+        test('a%(width|b)d', 'ab', outtmpl_na_placeholder='none')
 
         FORMATS = self.outtmpl_info['formats']
-        sanitize = lambda x: x.replace(':', ':').replace('"', """).replace('\n', ' ')
 
         # Custom type casting
         test('%(formats.:.id)l', 'id 1, id 2, id 3')
         test('%(formats.:.id)#l', ('id 1\nid 2\nid 3', 'id 1 id 2 id 3'))
         test('%(ext)l', 'mp4')
         test('%(formats.:.id) 18l', '  id 1, id 2, id 3')
-        test('%(formats)j', (json.dumps(FORMATS), sanitize(json.dumps(FORMATS))))
-        test('%(formats)#j', (json.dumps(FORMATS, indent=4), sanitize(json.dumps(FORMATS, indent=4))))
+        test('%(formats)j', (json.dumps(FORMATS), None))
+        test('%(formats)#j', (
+            json.dumps(FORMATS, indent=4),
+            json.dumps(FORMATS, indent=4).replace(':', ':').replace('"', """).replace('\n', ' ')
+        ))
         test('%(title5).3B', 'á')
         test('%(title5)U', 'áéí 𝐀')
         test('%(title5)#U', 'a\u0301e\u0301i\u0301 𝐀')
@@ -803,8 +796,8 @@ def expect_same_infodict(out):
         test('%(title|%)s %(title|%%)s', '% %%')
         test('%(id+1-height+3)05d', '00158')
         test('%(width+100)05d', 'NA')
-        test('%(formats.0) 15s', ('% 15s' % FORMATS[0], '% 15s' % sanitize(str(FORMATS[0]))))
-        test('%(formats.0)r', (repr(FORMATS[0]), sanitize(repr(FORMATS[0]))))
+        test('%(formats.0) 15s', ('% 15s' % FORMATS[0], None))
+        test('%(formats.0)r', (repr(FORMATS[0]), None))
         test('%(height.0)03d', '001')
         test('%(-height.0)04d', '-001')
         test('%(formats.-1.id)s', FORMATS[-1]['id'])
@@ -813,6 +806,12 @@ def expect_same_infodict(out):
         test('%(formats.:2:-1)r', repr(FORMATS[:2:-1]))
         test('%(formats.0.id.-1+id)f', '1235.000000')
         test('%(formats.0.id.-1+formats.1.id.-1)d', '3')
+        out = json.dumps([{'id': f['id'], 'height.:2': str(f['height'])[:2]}
+                          if 'height' in f else {'id': f['id']}
+                          for f in FORMATS])
+        test('%(formats.:.{id,height.:2})j', (out, None))
+        test('%(formats.:.{id,height}.id)l', ', '.join(f['id'] for f in FORMATS))
+        test('%(.{id,title})j', ('{"id": "1234"}', '{"id": "1234"}'))
 
         # Alternates
         test('%(title,id)s', '1234')
@@ -826,6 +825,10 @@ def expect_same_infodict(out):
         test('%(title&foo|baz)s.bar', 'baz.bar')
         test('%(x,id&foo|baz)s.bar', 'foo.bar')
         test('%(x,title&foo|baz)s.bar', 'baz.bar')
+        test('%(id&a\nb|)s', ('a\nb', 'a b'))
+        test('%(id&hi {:>10} {}|)s', 'hi       1234 1234')
+        test(R'%(id&{0} {}|)s', 'NA')
+        test(R'%(id&{0.1}|)s', 'NA')
 
         # Laziness
         def gen():
@@ -871,12 +874,12 @@ def test_postprocessors(self):
 
         class SimplePP(PostProcessor):
             def run(self, info):
-                with open(audiofile, 'wt') as f:
+                with open(audiofile, 'w') as f:
                     f.write('EXAMPLE')
                 return [info['filepath']], info
 
         def run_pp(params, PP):
-            with open(filename, 'wt') as f:
+            with open(filename, 'w') as f:
                 f.write('EXAMPLE')
             ydl = YoutubeDL(params)
             ydl.add_post_processor(PP())
@@ -895,7 +898,7 @@ def run_pp(params, PP):
 
         class ModifierPP(PostProcessor):
             def run(self, info):
-                with open(info['filepath'], 'wt') as f:
+                with open(info['filepath'], 'w') as f:
                     f.write('MODIFIED')
                 return [], info
 
@@ -1097,11 +1100,6 @@ def test_selection(params, expected_ids, evaluate_all=False):
         test_selection({'playlist_items': '-15::2'}, INDICES[1::2], True)
         test_selection({'playlist_items': '-15::15'}, [], True)
 
-    def test_urlopen_no_file_protocol(self):
-        # see https://github.com/ytdl-org/youtube-dl/issues/8227
-        ydl = YDL()
-        self.assertRaises(urllib.error.URLError, ydl.urlopen, 'file:///etc/passwd')
-
     def test_do_not_override_ie_key_in_url_transparent(self):
         ydl = YDL()
 
@@ -1215,6 +1213,62 @@ def _real_extract(self, url):
         self.assertEqual(downloaded['extractor'], 'Video')
         self.assertEqual(downloaded['extractor_key'], 'Video')
 
+    def test_header_cookies(self):
+        from http.cookiejar import Cookie
+
+        ydl = FakeYDL()
+        ydl.report_warning = lambda *_, **__: None
+
+        def cookie(name, value, version=None, domain='', path='', secure=False, expires=None):
+            return Cookie(
+                version or 0, name, value, None, False,
+                domain, bool(domain), bool(domain), path, bool(path),
+                secure, expires, False, None, None, rest={})
+
+        _test_url = 'https://yt.dlp/test'
+
+        def test(encoded_cookies, cookies, headers=False, round_trip=None, error=None):
+            def _test():
+                ydl.cookiejar.clear()
+                ydl._load_cookies(encoded_cookies, from_headers=headers)
+                if headers:
+                    ydl._apply_header_cookies(_test_url)
+                data = {'url': _test_url}
+                ydl._calc_headers(data)
+                self.assertCountEqual(
+                    map(vars, ydl.cookiejar), map(vars, cookies),
+                    'Extracted cookiejar.Cookie is not the same')
+                if not headers:
+                    self.assertEqual(
+                        data.get('cookies'), round_trip or encoded_cookies,
+                        'Cookie is not the same as round trip')
+                ydl.__dict__['_YoutubeDL__header_cookies'] = []
+
+            with self.subTest(msg=encoded_cookies):
+                if not error:
+                    _test()
+                    return
+                with self.assertRaisesRegex(Exception, error):
+                    _test()
+
+        test('test=value; Domain=.yt.dlp', [cookie('test', 'value', domain='.yt.dlp')])
+        test('test=value', [cookie('test', 'value')], error='Unscoped cookies are not allowed')
+        test('cookie1=value1; Domain=.yt.dlp; Path=/test; cookie2=value2; Domain=.yt.dlp; Path=/', [
+            cookie('cookie1', 'value1', domain='.yt.dlp', path='/test'),
+            cookie('cookie2', 'value2', domain='.yt.dlp', path='/')])
+        test('test=value; Domain=.yt.dlp; Path=/test; Secure; Expires=9999999999', [
+            cookie('test', 'value', domain='.yt.dlp', path='/test', secure=True, expires=9999999999)])
+        test('test="value; "; path=/test; domain=.yt.dlp', [
+            cookie('test', 'value; ', domain='.yt.dlp', path='/test')],
+            round_trip='test="value\\073 "; Domain=.yt.dlp; Path=/test')
+        test('name=; Domain=.yt.dlp', [cookie('name', '', domain='.yt.dlp')],
+             round_trip='name=""; Domain=.yt.dlp')
+
+        test('test=value', [cookie('test', 'value', domain='.yt.dlp')], headers=True)
+        test('cookie1=value; Domain=.yt.dlp; cookie2=value', [], headers=True, error='Invalid syntax')
+        ydl.deprecated_feature = ydl.report_error
+        test('test=value', [], headers=True, error='Passing cookies as a header is a potential security risk')
+
 
 if __name__ == '__main__':
     unittest.main()