]> jfr.im git - yt-dlp.git/blobdiff - test/test_YoutubeDL.py
Make more fields available for `--print` when used with `--flat-playlist`
[yt-dlp.git] / test / test_YoutubeDL.py
index df8994b8427c08ad8848ee4ee137e4f66dd31a70..30c48c78fa002233aebbeea6728965e16418aa45 100644 (file)
@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 # coding: utf-8
 
 from __future__ import unicode_literals
 import copy
 
 from test.helper import FakeYDL, assertRegexpMatches
-from youtube_dl import YoutubeDL
-from youtube_dl.compat import compat_str, compat_urllib_error
-from youtube_dl.extractor import YoutubeIE
-from youtube_dl.extractor.common import InfoExtractor
-from youtube_dl.postprocessor.common import PostProcessor
-from youtube_dl.utils import ExtractorError, match_filter_func
+from yt_dlp import YoutubeDL
+from yt_dlp.compat import compat_str, compat_urllib_error
+from yt_dlp.extractor import YoutubeIE
+from yt_dlp.extractor.common import InfoExtractor
+from yt_dlp.postprocessor.common import PostProcessor
+from yt_dlp.utils import ExtractorError, float_or_none, match_filter_func
 
 TEST_URL = 'http://localhost/sample.mp4'
 
@@ -29,6 +29,7 @@ def __init__(self, *args, **kwargs):
         self.msgs = []
 
     def process_info(self, info_dict):
+        info_dict.pop('__original_infodict', None)
         self.downloaded_info_dicts.append(info_dict)
 
     def to_screen(self, msg):
@@ -42,6 +43,7 @@ def _make_result(formats, **kwargs):
         'title': 'testttitle',
         'extractor': 'testex',
         'extractor_key': 'TestEx',
+        'webpage_url': 'http://example.com/watch?v=shenanigans',
     }
     res.update(**kwargs)
     return res
@@ -77,7 +79,7 @@ def test_prefer_free_formats(self):
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['ext'], 'mp4')
 
-        # No prefer_free_formats => prefer mp4 and flv for greater compatibility
+        # No prefer_free_formats => prefer mp4 and webm
         ydl = YDL()
         ydl.params['prefer_free_formats'] = False
         formats = [
@@ -103,7 +105,7 @@ def test_prefer_free_formats(self):
         yie._sort_formats(info_dict['formats'])
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['ext'], 'flv')
+        self.assertEqual(downloaded['ext'], 'webm')
 
     def test_format_selection(self):
         formats = [
@@ -242,6 +244,7 @@ def test_format_selection_video(self):
     def test_format_selection_string_ops(self):
         formats = [
             {'format_id': 'abc-cba', 'ext': 'mp4', 'url': TEST_URL},
+            {'format_id': 'zxc-cxz', 'ext': 'webm', 'url': TEST_URL},
         ]
         info_dict = _make_result(formats)
 
@@ -253,6 +256,11 @@ def test_format_selection_string_ops(self):
 
         # does not equal (!=)
         ydl = YDL({'format': '[format_id!=abc-cba]'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'zxc-cxz')
+
+        ydl = YDL({'format': '[format_id!=abc-cba][format_id!=zxc-cxz]'})
         self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
 
         # starts with (^=)
@@ -262,7 +270,12 @@ def test_format_selection_string_ops(self):
         self.assertEqual(downloaded['format_id'], 'abc-cba')
 
         # does not start with (!^=)
-        ydl = YDL({'format': '[format_id!^=abc-cba]'})
+        ydl = YDL({'format': '[format_id!^=abc]'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'zxc-cxz')
+
+        ydl = YDL({'format': '[format_id!^=abc][format_id!^=zxc]'})
         self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
 
         # ends with ($=)
@@ -272,20 +285,36 @@ def test_format_selection_string_ops(self):
         self.assertEqual(downloaded['format_id'], 'abc-cba')
 
         # does not end with (!$=)
-        ydl = YDL({'format': '[format_id!$=abc-cba]'})
+        ydl = YDL({'format': '[format_id!$=cba]'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'zxc-cxz')
+
+        ydl = YDL({'format': '[format_id!$=cba][format_id!$=cxz]'})
         self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
 
         # contains (*=)
-        ydl = YDL({'format': '[format_id*=-]'})
+        ydl = YDL({'format': '[format_id*=bc-cb]'})
         ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['format_id'], 'abc-cba')
 
         # does not contain (!*=)
+        ydl = YDL({'format': '[format_id!*=bc-cb]'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'zxc-cxz')
+
+        ydl = YDL({'format': '[format_id!*=abc][format_id!*=zxc]'})
+        self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
+
         ydl = YDL({'format': '[format_id!*=-]'})
         self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
 
     def test_youtube_format_selection(self):
+        # FIXME: Rewrite in accordance with the new format sorting options
+        return
+
         order = [
             '38', '37', '46', '22', '45', '35', '44', '18', '34', '43', '6', '5', '17', '36', '13',
             # Apple HTTP Live Streaming
@@ -323,7 +352,7 @@ def format_info(f_id):
         yie._sort_formats(info_dict['formats'])
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], '137+141')
+        self.assertEqual(downloaded['format_id'], '248+172')
         self.assertEqual(downloaded['ext'], 'mp4')
 
         info_dict = _make_result(list(formats_order), extractor='youtube')
@@ -387,7 +416,7 @@ def test_audio_only_extractor_format_selection(self):
         # For extractors with incomplete formats (all formats are audio-only or
         # video-only) best and worst should fallback to corresponding best/worst
         # video-only or audio-only formats (as per
-        # https://github.com/rg3/youtube-dl/pull/5556)
+        # https://github.com/ytdl-org/youtube-dl/pull/5556)
         formats = [
             {'format_id': 'low', 'ext': 'mp3', 'preference': 1, 'vcodec': 'none', 'url': TEST_URL},
             {'format_id': 'high', 'ext': 'mp3', 'preference': 2, 'vcodec': 'none', 'url': TEST_URL},
@@ -418,7 +447,7 @@ def test_format_not_available(self):
         self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
 
     def test_format_selection_issue_10083(self):
-        # See https://github.com/rg3/youtube-dl/issues/10083
+        # See https://github.com/ytdl-org/youtube-dl/issues/10083
         formats = [
             {'format_id': 'regular', 'height': 360, 'url': TEST_URL},
             {'format_id': 'video', 'height': 720, 'acodec': 'none', 'url': TEST_URL},
@@ -510,19 +539,19 @@ def test_format_filtering(self):
 
     def test_default_format_spec(self):
         ydl = YDL({'simulate': True})
-        self.assertEqual(ydl._default_format_spec({}), 'bestvideo+bestaudio/best')
+        self.assertEqual(ydl._default_format_spec({}), 'bestvideo*+bestaudio/best')
 
         ydl = YDL({})
         self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best/bestvideo+bestaudio')
 
         ydl = YDL({'simulate': True})
-        self.assertEqual(ydl._default_format_spec({'is_live': True}), 'bestvideo+bestaudio/best')
+        self.assertEqual(ydl._default_format_spec({'is_live': True}), 'bestvideo*+bestaudio/best')
 
         ydl = YDL({'outtmpl': '-'})
         self.assertEqual(ydl._default_format_spec({}), 'best/bestvideo+bestaudio')
 
         ydl = YDL({})
-        self.assertEqual(ydl._default_format_spec({}, download=False), 'bestvideo+bestaudio/best')
+        self.assertEqual(ydl._default_format_spec({}, download=False), 'bestvideo*+bestaudio/best')
         self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best/bestvideo+bestaudio')
 
 
@@ -543,6 +572,7 @@ def s_formats(lang, autocaption=False):
             'subtitles': subtitles,
             'automatic_captions': auto_captions,
             'extractor': 'TEST',
+            'webpage_url': 'http://example.com/watch?v=shenanigans',
         }
 
         def get_info(params={}):
@@ -572,6 +602,26 @@ def get_info(params={}):
         self.assertTrue(subs)
         self.assertEqual(set(subs.keys()), set(['es', 'fr']))
 
+        result = get_info({'writesubtitles': True, 'subtitleslangs': ['all', '-en']})
+        subs = result['requested_subtitles']
+        self.assertTrue(subs)
+        self.assertEqual(set(subs.keys()), set(['es', 'fr']))
+
+        result = get_info({'writesubtitles': True, 'subtitleslangs': ['en', 'fr', '-en']})
+        subs = result['requested_subtitles']
+        self.assertTrue(subs)
+        self.assertEqual(set(subs.keys()), set(['fr']))
+
+        result = get_info({'writesubtitles': True, 'subtitleslangs': ['-en', 'en']})
+        subs = result['requested_subtitles']
+        self.assertTrue(subs)
+        self.assertEqual(set(subs.keys()), set(['en']))
+
+        result = get_info({'writesubtitles': True, 'subtitleslangs': ['e.+']})
+        subs = result['requested_subtitles']
+        self.assertTrue(subs)
+        self.assertEqual(set(subs.keys()), set(['es', 'en']))
+
         result = get_info({'writesubtitles': True, 'writeautomaticsub': True, 'subtitleslangs': ['es', 'pt']})
         subs = result['requested_subtitles']
         self.assertTrue(subs)
@@ -598,41 +648,100 @@ def test_add_extra_info(self):
         self.assertEqual(test_dict['extractor'], 'Foo')
         self.assertEqual(test_dict['playlist'], 'funny videos')
 
-    def test_prepare_filename(self):
-        info = {
-            'id': '1234',
-            'ext': 'mp4',
-            'width': None,
-            'height': 1080,
-            'title1': '$PATH',
-            'title2': '%PATH%',
-        }
+    outtmpl_info = {
+        'id': '1234',
+        'ext': 'mp4',
+        'width': None,
+        'height': 1080,
+        'title1': '$PATH',
+        'title2': '%PATH%',
+        'title3': 'foo/bar\\test',
+        'timestamp': 1618488000,
+        'duration': 100000,
+        'playlist_index': 1,
+        '_last_playlist_index': 100,
+        'n_entries': 10,
+        'formats': [{'id': 'id1'}, {'id': 'id2'}, {'id': 'id3'}]
+    }
+
+    def test_prepare_outtmpl(self):
+        def out(tmpl, **params):
+            params['outtmpl'] = tmpl
+            ydl = YoutubeDL(params)
+            ydl._num_downloads = 1
+            outtmpl, tmpl_dict = ydl.prepare_outtmpl(tmpl, self.outtmpl_info)
+            return outtmpl % tmpl_dict
+
+        self.assertEqual(out('%(id)s.%(ext)s'), '1234.mp4')
+        self.assertEqual(out('%(duration_string)s'), '27:46:40')
+        self.assertTrue(float_or_none(out('%(epoch)d')))
+        self.assertEqual(out('%(resolution)s'), '1080p')
+        self.assertEqual(out('%(playlist_index)s'), '001')
+        self.assertEqual(out('%(autonumber)s'), '00001')
+        self.assertEqual(out('%(autonumber+2)03d', autonumber_start=3), '005')
+        self.assertEqual(out('%(autonumber)s', autonumber_size=3), '001')
+
+        self.assertEqual(out('%%'), '%')
+        self.assertEqual(out('%%%%'), '%%')
+        self.assertEqual(out('%(invalid@tmpl|def)s', outtmpl_na_placeholder='none'), 'none')
+        self.assertEqual(out('%()s'), 'NA')
+        self.assertEqual(out('%s'), '%s')
+
+        NA_TEST_OUTTMPL = '%(uploader_date)s-%(width)d-%(x|def)s-%(id)s.%(ext)s'
+        self.assertEqual(out(NA_TEST_OUTTMPL), 'NA-NA-def-1234.mp4')
+        self.assertEqual(out(NA_TEST_OUTTMPL, outtmpl_na_placeholder='none'), 'none-none-def-1234.mp4')
+        self.assertEqual(out(NA_TEST_OUTTMPL, outtmpl_na_placeholder=''), '--def-1234.mp4')
+
+        FMT_TEST_OUTTMPL = '%%(height)%s.%%(ext)s'
+        self.assertEqual(out(FMT_TEST_OUTTMPL % 's'), '1080.mp4')
+        self.assertEqual(out(FMT_TEST_OUTTMPL % 'd'), '1080.mp4')
+        self.assertEqual(out(FMT_TEST_OUTTMPL % '6d'), '  1080.mp4')
+        self.assertEqual(out(FMT_TEST_OUTTMPL % '-6d'), '1080  .mp4')
+        self.assertEqual(out(FMT_TEST_OUTTMPL % '06d'), '001080.mp4')
+        self.assertEqual(out(FMT_TEST_OUTTMPL % ' 06d'), ' 01080.mp4')
+        self.assertEqual(out(FMT_TEST_OUTTMPL % '   06d'), ' 01080.mp4')
+        self.assertEqual(out(FMT_TEST_OUTTMPL % '0 6d'), ' 01080.mp4')
+        self.assertEqual(out(FMT_TEST_OUTTMPL % '0   6d'), ' 01080.mp4')
+        self.assertEqual(out(FMT_TEST_OUTTMPL % '   0   6d'), ' 01080.mp4')
+
+        self.assertEqual(out('%(id)d'), '1234')
+        self.assertEqual(out('%(id)d %(id)r'), "1234 '1234'")
+        self.assertEqual(out('%(ext)s-%(ext|def)d'), 'mp4-def')
+        self.assertEqual(out('%(width|0)04d'), '0000')
+        self.assertEqual(out('%(width|)d', outtmpl_na_placeholder='none'), '')
+
+        FORMATS = self.outtmpl_info['formats']
+        self.assertEqual(out('%(timestamp+-1000>%H-%M-%S)s'), '11-43-20')
+        self.assertEqual(out('%(id+1-height+3)05d'), '00158')
+        self.assertEqual(out('%(width+100)05d'), 'NA')
+        self.assertEqual(out('%(formats.0)s'), str(FORMATS[0]))
+        self.assertEqual(out('%(formats.-1.id)s'), str(FORMATS[-1]['id']))
+        self.assertEqual(out('%(formats.3)s'), 'NA')
+        self.assertEqual(out('%(formats.:2:-1)r'), repr(FORMATS[:2:-1]))
+        self.assertEqual(out('%(formats.0.id.-1+id)f'), '1235.000000')
 
+    def test_prepare_filename(self):
         def fname(templ):
-            ydl = YoutubeDL({'outtmpl': templ})
-            return ydl.prepare_filename(info)
-        self.assertEqual(fname('%(id)s.%(ext)s'), '1234.mp4')
-        self.assertEqual(fname('%(id)s-%(width)s.%(ext)s'), '1234-NA.mp4')
-        # Replace missing fields with 'NA'
-        self.assertEqual(fname('%(uploader_date)s-%(id)s.%(ext)s'), 'NA-1234.mp4')
-        self.assertEqual(fname('%(height)d.%(ext)s'), '1080.mp4')
-        self.assertEqual(fname('%(height)6d.%(ext)s'), '  1080.mp4')
-        self.assertEqual(fname('%(height)-6d.%(ext)s'), '1080  .mp4')
-        self.assertEqual(fname('%(height)06d.%(ext)s'), '001080.mp4')
-        self.assertEqual(fname('%(height) 06d.%(ext)s'), ' 01080.mp4')
-        self.assertEqual(fname('%(height)   06d.%(ext)s'), ' 01080.mp4')
-        self.assertEqual(fname('%(height)0 6d.%(ext)s'), ' 01080.mp4')
-        self.assertEqual(fname('%(height)0   6d.%(ext)s'), ' 01080.mp4')
-        self.assertEqual(fname('%(height)   0   6d.%(ext)s'), ' 01080.mp4')
+            params = {'outtmpl': templ}
+            ydl = YoutubeDL(params)
+            return ydl.prepare_filename(self.outtmpl_info)
+
         self.assertEqual(fname('%%'), '%')
         self.assertEqual(fname('%%%%'), '%%')
-        self.assertEqual(fname('%%(height)06d.%(ext)s'), '%(height)06d.mp4')
+        self.assertEqual(fname('%%(width)06d.%(ext)s'), '%(width)06d.mp4')
         self.assertEqual(fname('%(width)06d.%(ext)s'), 'NA.mp4')
         self.assertEqual(fname('%(width)06d.%%(ext)s'), 'NA.%(ext)s')
         self.assertEqual(fname('%%(width)06d.%(ext)s'), '%(width)06d.mp4')
+
         self.assertEqual(fname('Hello %(title1)s'), 'Hello $PATH')
         self.assertEqual(fname('Hello %(title2)s'), 'Hello %PATH%')
 
+        self.assertEqual(fname('%(title3)s'), 'foo_bar_test')
+        self.assertEqual(fname('%(formats.0)s'), "{'id' - 'id1'}")
+
+        self.assertEqual(fname('%(id)r %(height)r'), "'1234' 1080")
+        self.assertEqual(fname('%(formats.0)r'), "{'id' - 'id1'}")
+
     def test_format_note(self):
         ydl = YoutubeDL()
         self.assertEqual(ydl._format_note({}), '')
@@ -690,7 +799,7 @@ def __init__(self, *args, **kwargs):
             def process_info(self, info_dict):
                 super(YDL, self).process_info(info_dict)
 
-            def _match_entry(self, info_dict, incomplete):
+            def _match_entry(self, info_dict, incomplete=False):
                 res = super(FilterYDL, self)._match_entry(info_dict, incomplete)
                 if res is None:
                     self.downloaded_info_dicts.append(info_dict)
@@ -706,6 +815,7 @@ def _match_entry(self, info_dict, incomplete):
             'playlist_id': '42',
             'uploader': "變態妍字幕版 太妍 тест",
             'creator': "тест ' 123 ' тест--",
+            'webpage_url': 'http://example.com/watch?v=shenanigans',
         }
         second = {
             'id': '2',
@@ -717,6 +827,7 @@ def _match_entry(self, info_dict, incomplete):
             'filesize': 5 * 1024,
             'playlist_id': '43',
             'uploader': "тест 123",
+            'webpage_url': 'http://example.com/watch?v=SHENANIGANS',
         }
         videos = [first, second]
 
@@ -792,11 +903,15 @@ def test_playlist_items_selection(self):
             'webpage_url': 'http://example.com',
         }
 
-        def get_ids(params):
+        def get_downloaded_info_dicts(params):
             ydl = YDL(params)
-            # make a copy because the dictionary can be modified
-            ydl.process_ie_result(playlist.copy())
-            return [int(v['id']) for v in ydl.downloaded_info_dicts]
+            # make a deep copy because the dictionary and nested entries
+            # can be modified
+            ydl.process_ie_result(copy.deepcopy(playlist))
+            return ydl.downloaded_info_dicts
+
+        def get_ids(params):
+            return [int(v['id']) for v in get_downloaded_info_dicts(params)]
 
         result = get_ids({})
         self.assertEqual(result, [1, 2, 3, 4])
@@ -828,8 +943,24 @@ def get_ids(params):
         result = get_ids({'playlist_items': '2-4,3-4,3'})
         self.assertEqual(result, [2, 3, 4])
 
+        # Tests for https://github.com/ytdl-org/youtube-dl/issues/10591
+        # @{
+        result = get_downloaded_info_dicts({'playlist_items': '2-4,3-4,3'})
+        self.assertEqual(result[0]['playlist_index'], 2)
+        self.assertEqual(result[1]['playlist_index'], 3)
+
+        result = get_downloaded_info_dicts({'playlist_items': '2-4,3-4,3'})
+        self.assertEqual(result[0]['playlist_index'], 2)
+        self.assertEqual(result[1]['playlist_index'], 3)
+        self.assertEqual(result[2]['playlist_index'], 4)
+
+        result = get_downloaded_info_dicts({'playlist_items': '4,2'})
+        self.assertEqual(result[0]['playlist_index'], 4)
+        self.assertEqual(result[1]['playlist_index'], 2)
+        # @}
+
     def test_urlopen_no_file_protocol(self):
-        # see https://github.com/rg3/youtube-dl/issues/8227
+        # see https://github.com/ytdl-org/youtube-dl/issues/8227
         ydl = YDL()
         self.assertRaises(compat_urllib_error.URLError, ydl.urlopen, 'file:///etc/passwd')
 
@@ -875,6 +1006,76 @@ def _real_extract(self, url):
         self.assertEqual(downloaded['extractor'], 'testex')
         self.assertEqual(downloaded['extractor_key'], 'TestEx')
 
+    # Test case for https://github.com/ytdl-org/youtube-dl/issues/27064
+    def test_ignoreerrors_for_playlist_with_url_transparent_iterable_entries(self):
+
+        class _YDL(YDL):
+            def __init__(self, *args, **kwargs):
+                super(_YDL, self).__init__(*args, **kwargs)
+
+            def trouble(self, s, tb=None):
+                pass
+
+        ydl = _YDL({
+            'format': 'extra',
+            'ignoreerrors': True,
+        })
+
+        class VideoIE(InfoExtractor):
+            _VALID_URL = r'video:(?P<id>\d+)'
+
+            def _real_extract(self, url):
+                video_id = self._match_id(url)
+                formats = [{
+                    'format_id': 'default',
+                    'url': 'url:',
+                }]
+                if video_id == '0':
+                    raise ExtractorError('foo')
+                if video_id == '2':
+                    formats.append({
+                        'format_id': 'extra',
+                        'url': TEST_URL,
+                    })
+                return {
+                    'id': video_id,
+                    'title': 'Video %s' % video_id,
+                    'formats': formats,
+                }
+
+        class PlaylistIE(InfoExtractor):
+            _VALID_URL = r'playlist:'
+
+            def _entries(self):
+                for n in range(3):
+                    video_id = compat_str(n)
+                    yield {
+                        '_type': 'url_transparent',
+                        'ie_key': VideoIE.ie_key(),
+                        'id': video_id,
+                        'url': 'video:%s' % video_id,
+                        'title': 'Video Transparent %s' % video_id,
+                    }
+
+            def _real_extract(self, url):
+                return self.playlist_result(self._entries())
+
+        ydl.add_info_extractor(VideoIE(ydl))
+        ydl.add_info_extractor(PlaylistIE(ydl))
+        info = ydl.extract_info('playlist:')
+        entries = info['entries']
+        self.assertEqual(len(entries), 3)
+        self.assertTrue(entries[0] is None)
+        self.assertTrue(entries[1] is None)
+        self.assertEqual(len(ydl.downloaded_info_dicts), 1)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(entries[2], downloaded)
+        self.assertEqual(downloaded['url'], TEST_URL)
+        self.assertEqual(downloaded['title'], 'Video Transparent 2')
+        self.assertEqual(downloaded['id'], '2')
+        self.assertEqual(downloaded['extractor'], 'Video')
+        self.assertEqual(downloaded['extractor_key'], 'Video')
+
 
 if __name__ == '__main__':
     unittest.main()