Merge a57ae3f760 into 0153b387e5

* [VidLii] Add HD support  (yt-dlp backport-ish)

* Also fix a bug with the view count

---------

Co-authored-by: dirkf <fieldhouse@gmx.net>

test/helper.py

@@ -5,9 +5,9 @@ import hashlib
 import json
 import os.path
 import re
-import types
 import ssl
 import sys
+import types
 import unittest
 
 import youtube_dl.extractor
@@ -181,18 +181,18 @@ def expect_value(self, got, expected, field):
             op, _, expected_num = expected.partition(':')
             expected_num = int(expected_num)
             if op == 'mincount':
-                assert_func = assertGreaterEqual
+                assert_func = self.assertGreaterEqual
                 msg_tmpl = 'Expected %d items in field %s, but only got %d'
             elif op == 'maxcount':
-                assert_func = assertLessEqual
+                assert_func = self.assertLessEqual
                 msg_tmpl = 'Expected maximum %d items in field %s, but got %d'
             elif op == 'count':
-                assert_func = assertEqual
+                assert_func = self.assertEqual
                 msg_tmpl = 'Expected exactly %d items in field %s, but got %d'
             else:
                 assert False
             assert_func(
-                self, len(got), expected_num,
+                len(got), expected_num,
                 msg_tmpl % (expected_num, field, len(got)))
             return
         self.assertEqual(
@@ -262,27 +262,6 @@ def assertRegexpMatches(self, text, regexp, msg=None):
             self.assertTrue(m, msg)
 
 
-def assertGreaterEqual(self, got, expected, msg=None):
-    if not (got >= expected):
-        if msg is None:
-            msg = '%r not greater than or equal to %r' % (got, expected)
-        self.assertTrue(got >= expected, msg)
-
-
-def assertLessEqual(self, got, expected, msg=None):
-    if not (got <= expected):
-        if msg is None:
-            msg = '%r not less than or equal to %r' % (got, expected)
-        self.assertTrue(got <= expected, msg)
-
-
-def assertEqual(self, got, expected, msg=None):
-    if not (got == expected):
-        if msg is None:
-            msg = '%r not equal to %r' % (got, expected)
-        self.assertTrue(got == expected, msg)
-
-
 def expect_warnings(ydl, warnings_re):
     real_warning = ydl.report_warning
 

test/test_download.py

@@ -9,8 +9,6 @@ import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 from test.helper import (
-    assertGreaterEqual,
-    assertLessEqual,
     expect_warnings,
     get_params,
     gettestcases,
@@ -36,12 +34,20 @@ from youtube_dl.utils import (
     ExtractorError,
     error_to_compat_str,
     format_bytes,
+    IDENTITY,
+    preferredencoding,
     UnavailableVideoError,
 )
 from youtube_dl.extractor import get_info_extractor
 
 RETRIES = 3
 
+# Some unittest APIs require actual str
+if not isinstance('TEST', str):
+    _encode_str = lambda s: s.encode(preferredencoding())
+else:
+    _encode_str = IDENTITY
+
 
 class YoutubeDL(youtube_dl.YoutubeDL):
     def __init__(self, *args, **kwargs):
@@ -102,7 +108,7 @@ def generator(test_case, tname):
 
         def print_skipping(reason):
             print('Skipping %s: %s' % (test_case['name'], reason))
-            self.skipTest(reason)
+            self.skipTest(_encode_str(reason))
 
         if not ie.working():
             print_skipping('IE marked as not _WORKING')
@@ -187,16 +193,14 @@ def generator(test_case, tname):
                 expect_info_dict(self, res_dict, test_case.get('info_dict', {}))
 
             if 'playlist_mincount' in test_case:
-                assertGreaterEqual(
-                    self,
+                self.assertGreaterEqual(
                     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_maxcount' in test_case:
-                assertLessEqual(
-                    self,
+                self.assertLessEqual(
                     len(res_dict['entries']),
                     test_case['playlist_maxcount'],
                     'Expected at most %d in playlist %s, but got %d' % (
@@ -243,8 +247,8 @@ def generator(test_case, tname):
                         if params.get('test'):
                             expected_minsize = max(expected_minsize, 10000)
                         got_fsize = os.path.getsize(tc_filename)
-                        assertGreaterEqual(
-                            self, got_fsize, expected_minsize,
+                        self.assertGreaterEqual(
+                            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)))

youtube_dl/YoutubeDL.py

@@ -1039,8 +1039,8 @@ class YoutubeDL(object):
         elif result_type in ('playlist', 'multi_video'):
             # Protect from infinite recursion due to recursively nested playlists
             # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
-            webpage_url = ie_result['webpage_url']
-            if webpage_url in self._playlist_urls:
+            webpage_url = ie_result.get('webpage_url')  # not all pl/mv have this
+            if webpage_url and webpage_url in self._playlist_urls:
                 self.to_screen(
                     '[download] Skipping already downloaded playlist: %s'
                     % ie_result.get('title') or ie_result.get('id'))
@@ -1048,6 +1048,10 @@ class YoutubeDL(object):
 
             self._playlist_level += 1
             self._playlist_urls.add(webpage_url)
+            new_result = dict((k, v) for k, v in extra_info.items() if k not in ie_result)
+            if new_result:
+                new_result.update(ie_result)
+                ie_result = new_result
             try:
                 return self.__process_playlist(ie_result, download)
             finally:
@@ -1593,6 +1597,28 @@ class YoutubeDL(object):
         self.cookiejar.add_cookie_header(pr)
         return pr.get_header('Cookie')
 
+    def _fill_common_fields(self, info_dict, final=True):
+
+        for ts_key, date_key in (
+                ('timestamp', 'upload_date'),
+                ('release_timestamp', 'release_date'),
+        ):
+            if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
+                # Working around out-of-range timestamp values (e.g. negative ones on Windows,
+                # see http://bugs.python.org/issue1646728)
+                try:
+                    upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
+                    info_dict[date_key] = compat_str(upload_date.strftime('%Y%m%d'))
+                except (ValueError, OverflowError, OSError):
+                    pass
+
+        # Auto generate title fields corresponding to the *_number fields when missing
+        # in order to always have clean titles. This is very common for TV series.
+        if final:
+            for field in ('chapter', 'season', 'episode'):
+                if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
+                    info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
+
     def process_video_result(self, info_dict, download=True):
         assert info_dict.get('_type', 'video') == 'video'
 
@@ -1660,24 +1686,7 @@ class YoutubeDL(object):
         if 'display_id' not in info_dict and 'id' in info_dict:
             info_dict['display_id'] = info_dict['id']
 
-        for ts_key, date_key in (
-                ('timestamp', 'upload_date'),
-                ('release_timestamp', 'release_date'),
-        ):
-            if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
-                # Working around out-of-range timestamp values (e.g. negative ones on Windows,
-                # see http://bugs.python.org/issue1646728)
-                try:
-                    upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
-                    info_dict[date_key] = compat_str(upload_date.strftime('%Y%m%d'))
-                except (ValueError, OverflowError, OSError):
-                    pass
-
-        # Auto generate title fields corresponding to the *_number fields when missing
-        # in order to always have clean titles. This is very common for TV series.
-        for field in ('chapter', 'season', 'episode'):
-            if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
-                info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
+        self._fill_common_fields(info_dict)
 
         for cc_kind in ('subtitles', 'automatic_captions'):
             cc = info_dict.get(cc_kind)

youtube_dl/extractor/extractors.py

@@ -530,6 +530,7 @@ from .indavideo import IndavideoEmbedIE
 from .infoq import InfoQIE
 from .instagram import (
     InstagramIE,
+    InstagramStoriesIE,
     InstagramUserIE,
     InstagramTagIE,
 )
@@ -898,21 +899,13 @@ from .ooyala import (
 )
 from .ora import OraTVIE
 from .orf import (
-    ORFTVthekIE,
-    ORFFM4IE,
+    ORFONIE,
+    ORFONLiveIE,
     ORFFM4StoryIE,
-    ORFOE1IE,
-    ORFOE3IE,
-    ORFNOEIE,
-    ORFWIEIE,
-    ORFBGLIE,
-    ORFOOEIE,
-    ORFSTMIE,
-    ORFKTNIE,
-    ORFSBGIE,
-    ORFTIRIE,
-    ORFVBGIE,
     ORFIPTVIE,
+    ORFPodcastIE,
+    ORFRadioIE,
+    ORFRadioCollectionIE,
 )
 from .outsidetv import OutsideTVIE
 from .packtpub import (

youtube_dl/extractor/instagram.py

@@ -9,6 +9,7 @@ from .common import InfoExtractor
 from ..compat import (
     compat_str,
     compat_HTTPError,
+    compat_urlparse,
 )
 from ..utils import (
     ExtractorError,
@@ -305,7 +306,7 @@ class InstagramPlaylistIE(InfoExtractor):
                 'first': 12,
                 'after': cursor,
             }
-            variables.update(self._query_vars_for(data))
+            variables.update(self._query_vars_for(data, url))
             variables = json.dumps(variables)
 
             if self._gis_tmpl:
@@ -320,6 +321,10 @@ class InstagramPlaylistIE(InfoExtractor):
 
             # try all of the ways to generate a GIS query, and not only use the
             # first one that works, but cache it for future requests
+            if hasattr(self, '_get_query_hash'):
+                query_hash = self._get_query_hash(url)
+            else:
+                query_hash = self._QUERY_HASH
             for gis_tmpl in gis_tmpls:
                 try:
                     json_data = self._download_json(
@@ -329,10 +334,10 @@ class InstagramPlaylistIE(InfoExtractor):
                             'X-Instagram-GIS': hashlib.md5(
                                 ('%s:%s' % (gis_tmpl, variables)).encode('utf-8')).hexdigest(),
                         }, query={
-                            'query_hash': self._QUERY_HASH,
+                            'query_hash': query_hash,
                             'variables': variables,
                         })
-                    media = self._parse_timeline_from(json_data)
+                    media = self._parse_timeline_from(json_data, url)
                     self._gis_tmpl = gis_tmpl
                     break
                 except ExtractorError as e:
@@ -351,25 +356,71 @@ class InstagramPlaylistIE(InfoExtractor):
                 node = edge.get('node')
                 if not node or not isinstance(node, dict):
                     continue
-                if node.get('__typename') != 'GraphVideo' and node.get('is_video') is not True:
-                    continue
-                video_id = node.get('shortcode')
-                if not video_id:
+                if node.get('__typename') not in ('GraphVideo', 'GraphStoryVideo') and node.get('is_video') is not True:
                     continue
 
-                info = self.url_result(
-                    'https://instagram.com/p/%s/' % video_id,
-                    ie=InstagramIE.ie_key(), video_id=video_id)
+                if node.get('__typename') == 'GraphVideo':
+                    video_id = node.get('shortcode')
+                    if not video_id:
+                        continue
 
-                description = try_get(
-                    node, lambda x: x['edge_media_to_caption']['edges'][0]['node']['text'],
-                    compat_str)
-                thumbnail = node.get('thumbnail_src') or node.get('display_src')
-                timestamp = int_or_none(node.get('taken_at_timestamp'))
+                    info = self.url_result(
+                        'https://instagram.com/p/%s/' % video_id,
+                        ie=InstagramIE.ie_key(), video_id=video_id)
 
-                comment_count = get_count('to_comment')
-                like_count = get_count('preview_like')
-                view_count = int_or_none(node.get('video_view_count'))
+                    description = try_get(
+                        node, lambda x: x['edge_media_to_caption']['edges'][0]['node']['text'],
+                        compat_str)
+                    thumbnail = node.get('thumbnail_src') or node.get('display_src')
+                    timestamp = int_or_none(node.get('taken_at_timestamp'))
+
+                    comment_count = get_count('to_comment')
+                    like_count = get_count('preview_like')
+                    view_count = int_or_none(node.get('video_view_count'))
+                elif node.get('__typename') == 'GraphStoryVideo':
+                    if 'display_resources' not in node:
+                        continue
+                    thumbnails = []
+                    for thumb in node.get('display_resources', []):
+                        if 'src' not in thumb:
+                            continue
+                        thumbnails.append({
+                            'url': thumb['src'],
+                            'width': int_or_none(thumb.get('config_width')),
+                            'height': int_or_none(thumb.get('config_height')),
+                        })
+                    formats = []
+                    owner = node.get('owner', {})
+                    creator = owner.get('username')
+                    title = ('Stories - %s' % creator) if creator else 'Stories'
+                    video_res = node.get('video_resources')
+                    if not video_res:
+                        continue
+                    for res in video_res:
+                        if 'src' not in res:
+                            continue
+                        formats.append({
+                            'url': res['src'],
+                            'format_id': res.get('profile'),
+                            'width': int_or_none(res.get('config_width')),
+                            'height': int_or_none(res.get('config_height')),
+                            'format_note': res.get('mime_type'),
+                        })
+                    thumbnail = thumbnails[0]['url'] if thumbnails else None
+                    description = None
+                    timestamp = int_or_none(node.get('taken_at_timestamp') or node.get('date'))
+                    like_count = view_count = comment_count = None
+                    info = {
+                        'thumbnails': thumbnails,
+                        'id': node['id'],
+                        'title': title,
+                        'uploader': creator,
+                        'creator': creator,
+                        'duration': int_or_none(node.get('video_duration')),
+                        'formats': formats,
+                    }
+                else:
+                    continue
 
                 info.update({
                     'description': description,
@@ -423,15 +474,13 @@ class InstagramUserIE(InstagramPlaylistIE):
         }
     }
 
-    _QUERY_HASH = '42323d64886122307be10013ad2dcc44',
+    _QUERY_HASH = '42323d64886122307be10013ad2dcc44'
 
-    @staticmethod
-    def _parse_timeline_from(data):
+    def _parse_timeline_from(self, data, url):
         # extracts the media timeline data from a GraphQL result
         return data['data']['user']['edge_owner_to_timeline_media']
 
-    @staticmethod
-    def _query_vars_for(data):
+    def _query_vars_for(self, data, url):
         # returns a dictionary of variables to add to the timeline query based
         # on the GraphQL of the original page
         return {
@@ -457,18 +506,152 @@ class InstagramTagIE(InstagramPlaylistIE):
         }
     }
 
-    _QUERY_HASH = 'f92f56d47dc7a55b606908374b43a314',
+    _QUERY_HASH = 'f92f56d47dc7a55b606908374b43a314'
 
-    @staticmethod
-    def _parse_timeline_from(data):
+    def _parse_timeline_from(self, data, url):
         # extracts the media timeline data from a GraphQL result
         return data['data']['hashtag']['edge_hashtag_to_media']
 
-    @staticmethod
-    def _query_vars_for(data):
+    def _query_vars_for(self, data, url):
         # returns a dictionary of variables to add to the timeline query based
         # on the GraphQL of the original page
         return {
             'tag_name':
                 data['entry_data']['TagPage'][0]['graphql']['hashtag']['name']
         }
+
+
+class InstagramStoriesIE(InstagramPlaylistIE):
+    _VALID_URL = r'https?://(?:www\.)?instagram\.com/(?:highlights/)?stories/(?P<id>[^/]+)(/[0-9]+/?)?'
+    IE_DESC = 'Instagram stories'
+    IE_NAME = 'instagram:stories'
+    _TESTS = [{
+        'skip': 'Requires cookies',
+        'url': 'https://www.instagram.com/stories/jillianbmele/',
+        'md5': '29bc2e6a657e9eb4e81d18455e7200fc',
+        'info_dict': {
+            'id': '2284559186280109841',
+            'title': 'Stories - jillianbmele',
+            'ext': 'mp4',
+            'timestamp': 15865606971,
+            'uploader': 'jillianbmele',
+            'upload_date': '20200404',
+        },
+    }, {
+        'skip': 'Requires cookies',
+        'url': 'https://www.instagram.com/stories/highlights/17946082630331720/',
+        'md5': '41515e24cbf3478d71afe408bcdef43d',
+        'info_dict': {
+            'id': '2275290256240448451',
+            'title': 'Stories - lasvegasbarbie',
+            'ext': 'mp4',
+            'timestamp': 1585455755,
+            'uploader': 'lasvegasbarbie',
+            'upload_date': '20200329',
+        },
+    }, {
+        'skip': 'Requires cookies',
+        'url': 'https://www.instagram.com/stories/lisamarieboothe/2288168092151825799/?igshid=1fl8dqdvhla4a',
+        'md5': '08088623250416e23957b564832d20e3',
+        'info_dict': {
+            'id': '2288168092151825799',
+            'title': 'Stories - lisamarieboothe',
+            'ext': 'mp4',
+            'timestamp': 1586990910,
+            'uploader': 'lisamarieboothe',
+            'upload_date': '20200415',
+        },
+    }]
+    _QUERY_HASH = '04334405dbdef91f2c4e207b84c204d7'
+    _QUERY_HASH_HIGHLIGHTS = '45246d3fe16ccc6577e0bd297a5db1ab'
+    _QUERY_HASH_REEL_CONTENT = 'f5dc1457da7a4d3f88762dae127e0238'
+
+    def _get_query_hash(self, url):
+        path = compat_urlparse.urlparse(url).path
+        if path.startswith('/stories/highlights/'):
+            return self._QUERY_HASH_HIGHLIGHTS
+        if path.strip('/').split('/')[-1].isdigit():
+            return self._QUERY_HASH_REEL_CONTENT
+        return self._QUERY_HASH
+
+    def _parse_timeline_from(self, data, url):
+        # extracts the media timeline data from a GraphQL result
+        path = compat_urlparse.urlparse(url).path
+        if path.startswith('/stories/highlights/'):
+            return {'edges': [{'node': x} for x in data['data']['reels_media'][0]['items']]}
+        object_id = re.match(r'([0-9]+)', path.rstrip('/').split('/')[-1])
+        if object_id:
+            object_id = object_id.group(1)
+            return {'edges': [
+                {'node': x} for x in data['data']['reels_media'][0]['items']
+                if x['id'] == object_id
+            ]}
+        username = path.rstrip('/').split('/')[-1]
+        # Get the reel ID
+        reel_id = None
+        for edge in data['data']['user']['feed_reels_tray']['edge_reels_tray_to_reel']['edges']:
+            if edge['node']['owner']['username'] == username:
+                reel_id = edge['node']['id']
+                break  # There should only be one
+        if not reel_id:
+            # No story reel at the moment
+            return {'edges': []}
+        variables = {
+            'reel_ids': [reel_id],
+            'tag_names': [],
+            'location_ids': [],
+            'highlight_reel_ids': [],
+            'precomposed_overlay': False,
+            'show_story_viewer_list': True,
+            'story_viewer_fetch_count': 50,
+            'stories_video_dash_manifest': False
+        }
+        json_data = self._download_json(
+            'https://www.instagram.com/graphql/query/', username,
+            'Downloading JSON', headers={
+                'X-Requested-With': 'XMLHttpRequest',
+                'X-Instagram-GIS': hashlib.md5(
+                    ('%s:%s' % (self._gis_tmpl, variables)).encode('utf-8')).hexdigest(),
+            }, query={
+                'query_hash': self._QUERY_HASH_REEL_CONTENT,
+                'variables': json.dumps(variables),
+            })
+        if not json_data['data']['reels_media']:
+            raise ExtractorError('Matching reel not found')
+        return {'edges': [{'node': x} for x in json_data['data']['reels_media'][0]['items']]}
+
+    def _query_vars_for(self, data, url):
+        # returns a dictionary of variables to add to the timeline query based
+        # on the GraphQL of the original page
+        path = compat_urlparse.urlparse(url).path
+        if path.startswith('/stories/highlights/'):
+            return {
+                'reel_ids': [],
+                'tag_names': [],
+                'location_ids': [],
+                'highlight_reel_ids': [data['entry_data']['StoriesPage'][0]['highlight']['id']],
+                'precomposed_overlay': False,
+            }
+        object_id = re.match(r'[0-9]+', path.rstrip('/').split('/')[-1])
+        if object_id:
+            username = path.strip('/').split('/')[-2]
+            user_id = self._download_json(
+                'https://www.instagram.com/stories/%s' % username, username,
+                'Downloading JSON', headers={'X-Requested-With': 'XMLHttpRequest'},
+                query={'__a': '1'})['user']['id']
+            return {
+                'reel_ids': [user_id],
+                'tag_names': [],
+                'location_ids': [],
+                'highlight_reel_ids': [],
+                'precomposed_overlay': False,
+                'show_story_viewer_list': True,
+                'story_viewer_fetch_count': 50,
+                'story_viewer_cursor': '',
+                'stories_video_dash_manifest': False,
+            }
+        return {
+            'only_stories': True,
+            'stories_prefetch': True,
+            'stories_video_dash_manifest': False,
+        }

youtube_dl/extractor/mixcloud.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import itertools
@@ -10,7 +11,7 @@ from ..compat import (
     compat_ord,
     compat_str,
     compat_urllib_parse_unquote,
-    compat_zip
+    compat_zip as zip,
 )
 from ..utils import (
     int_or_none,
@@ -24,7 +25,7 @@ class MixcloudBaseIE(InfoExtractor):
     def _call_api(self, object_type, object_fields, display_id, username, slug=None):
         lookup_key = object_type + 'Lookup'
         return self._download_json(
-            'https://www.mixcloud.com/graphql', display_id, query={
+            'https://app.mixcloud.com/graphql', display_id, query={
                 'query': '''{
   %s(lookup: {username: "%s"%s}) {
     %s
@@ -44,7 +45,7 @@ class MixcloudIE(MixcloudBaseIE):
             'ext': 'm4a',
             'title': 'Cryptkeeper',
             'description': 'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.',
-            'uploader': 'Daniel Holbach',
+            'uploader': 'dholbach',  # was: 'Daniel Holbach',
             'uploader_id': 'dholbach',
             'thumbnail': r're:https?://.*\.jpg',
             'view_count': int,
@@ -57,7 +58,7 @@ class MixcloudIE(MixcloudBaseIE):
             'id': 'gillespeterson_caribou-7-inch-vinyl-mix-chat',
             'ext': 'mp3',
             'title': 'Caribou 7 inch Vinyl Mix & Chat',
-            'description': 'md5:2b8aec6adce69f9d41724647c65875e8',
+            'description': r're:Last week Dan Snaith aka Caribou swung by the Brownswood.{136}',
             'uploader': 'Gilles Peterson Worldwide',
             'uploader_id': 'gillespeterson',
             'thumbnail': 're:https?://.*',
@@ -65,6 +66,23 @@ class MixcloudIE(MixcloudBaseIE):
             'timestamp': 1422987057,
             'upload_date': '20150203',
         },
+        'params': {
+            'skip_download': '404 not found',
+        },
+    }, {
+        'url': 'https://www.mixcloud.com/gillespeterson/carnival-m%C3%BAsica-popular-brasileira-mix/',
+        'info_dict': {
+            'id': 'gillespeterson_carnival-música-popular-brasileira-mix',
+            'ext': 'm4a',
+            'title': 'Carnival Música Popular Brasileira Mix',
+            'description': r're:Gilles was recently in Brazil to play at Boiler Room.{208}',
+            'timestamp': 1454347174,
+            'upload_date': '20160201',
+            'uploader': 'Gilles Peterson Worldwide',
+            'uploader_id': 'gillespeterson',
+            'thumbnail': 're:https?://.*',
+            'view_count': int,
+        },
     }, {
         'url': 'https://beta.mixcloud.com/RedLightRadio/nosedrip-15-red-light-radio-01-18-2016/',
         'only_matching': True,
@@ -76,10 +94,10 @@ class MixcloudIE(MixcloudBaseIE):
         """Encrypt/Decrypt XOR cipher. Both ways are possible because it's XOR."""
         return ''.join([
             compat_chr(compat_ord(ch) ^ compat_ord(k))
-            for ch, k in compat_zip(ciphertext, itertools.cycle(key))])
+            for ch, k in zip(ciphertext, itertools.cycle(key))])
 
     def _real_extract(self, url):
-        username, slug = re.match(self._VALID_URL, url).groups()
+        username, slug = self._match_valid_url(url).groups()
         username, slug = compat_urllib_parse_unquote(username), compat_urllib_parse_unquote(slug)
         track_id = '%s_%s' % (username, slug)
 

youtube_dl/extractor/vidlii.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+
 from ..utils import (
     float_or_none,
     get_element_by_id,
@@ -11,6 +12,7 @@ from ..utils import (
     strip_or_none,
     unified_strdate,
     urljoin,
+    str_to_int,
 )
 
 
@@ -35,6 +37,26 @@ class VidLiiIE(InfoExtractor):
             'categories': ['News & Politics'],
             'tags': ['Vidlii', 'Jan', 'Videogames'],
         }
+    }, {
+        # HD
+        'url': 'https://www.vidlii.com/watch?v=2Ng8Abj2Fkl',
+        'md5': '450e7da379c884788c3a4fa02a3ce1a4',
+        'info_dict': {
+            'id': '2Ng8Abj2Fkl',
+            'ext': 'mp4',
+            'title': 'test',
+            'description': 'md5:cc55a86032a7b6b3cbfd0f6b155b52e9',
+            'thumbnail': 'https://www.vidlii.com/usfi/thmp/2Ng8Abj2Fkl.jpg',
+            'uploader': 'VidLii',
+            'uploader_url': 'https://www.vidlii.com/user/VidLii',
+            'upload_date': '20200927',
+            'duration': 5,
+            'view_count': int,
+            'comment_count': int,
+            'average_rating': float,
+            'categories': ['Film & Animation'],
+            'tags': list,
+        },
     }, {
         'url': 'https://www.vidlii.com/embed?v=tJluaH4BJ3v&a=0',
         'only_matching': True,
@@ -46,11 +68,32 @@ class VidLiiIE(InfoExtractor):
         webpage = self._download_webpage(
             'https://www.vidlii.com/watch?v=%s' % video_id, video_id)
 
-        video_url = self._search_regex(
-            r'src\s*:\s*(["\'])(?P<url>(?:https?://)?(?:(?!\1).)+)\1', webpage,
-            'video url', group='url')
+        formats = []
 
-        title = self._search_regex(
+        def add_format(format_url, height=None):
+            height = int(self._search_regex(r'(\d+)\.mp4',
+                         format_url, 'height', default=360))
+
+            formats.append({
+                'url': format_url,
+                'format_id': '%dp' % height if height else None,
+                'height': height,
+            })
+
+        sources = re.findall(
+            r'src\s*:\s*(["\'])(?P<url>(?:https?://)?(?:(?!\1).)+)\1',
+            webpage)
+
+        formats = []
+        if len(sources) > 1:
+            add_format(sources[1][1])
+            self._check_formats(formats, video_id)
+        if len(sources) > 0:
+            add_format(sources[0][1])
+
+        self._sort_formats(formats)
+
+        title = self._html_search_regex(
             (r'<h1>([^<]+)</h1>', r'<title>([^<]+) - VidLii<'), webpage,
             'title')
 
@@ -82,9 +125,9 @@ class VidLiiIE(InfoExtractor):
             default=None) or self._search_regex(
             r'duration\s*:\s*(\d+)', webpage, 'duration', fatal=False))
 
-        view_count = int_or_none(self._search_regex(
-            (r'<strong>(\d+)</strong> views',
-             r'Views\s*:\s*<strong>(\d+)</strong>'),
+        view_count = str_to_int(self._html_search_regex(
+            (r'<strong>([\d,.]+)</strong> views',
+             r'Views\s*:\s*<strong>([\d,.]+)</strong>'),
             webpage, 'view count', fatal=False))
 
         comment_count = int_or_none(self._search_regex(
@@ -109,7 +152,7 @@ class VidLiiIE(InfoExtractor):
 
         return {
             'id': video_id,
-            'url': video_url,
+            'formats': formats,
             'title': title,
             'description': description,
             'thumbnail': thumbnail,