Merge 7fb0a87d7c into 4d05f84325

* no space after @ in decorator

test/test_jsinterp.py

@@ -577,9 +577,11 @@ class TestJSInterpreter(unittest.TestCase):
     def test_unary_operators(self):
         jsi = JSInterpreter('function f(){return 2  -  - - 2;}')
         self.assertEqual(jsi.call_function('f'), 0)
-        # fails
-        # jsi = JSInterpreter('function f(){return 2 + - + - - 2;}')
-        # self.assertEqual(jsi.call_function('f'), 0)
+        jsi = JSInterpreter('function f(){return 2 + - + - - 2;}')
+        self.assertEqual(jsi.call_function('f'), 0)
+        # https://github.com/ytdl-org/youtube-dl/issues/32815
+        jsi = JSInterpreter('function f(){return 0  - 7 * - 6;}')
+        self.assertEqual(jsi.call_function('f'), 42)
 
     """ # fails so far
     def test_packed(self):

test/test_youtube_signature.py

@@ -158,6 +158,10 @@ _NSIG_TESTS = [
         'https://www.youtube.com/s/player/b7910ca8/player_ias.vflset/en_US/base.js',
         '_hXMCwMt9qE310D', 'LoZMgkkofRMCZQ',
     ),
+    (
+        'https://www.youtube.com/s/player/590f65a6/player_ias.vflset/en_US/base.js',
+        '1tm7-g_A9zsI8_Lay_', 'xI4Vem4Put_rOg',
+    ),
 ]
 
 

youtube_dl/extractor/bitchute.py

@@ -6,25 +6,105 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
+    clean_html,
+    ExtractorError,
+    GeoRestrictedError,
+    get_element_by_class,
+    get_element_by_id,
+    int_or_none,
+    merge_dicts,
     orderedSet,
-    unified_strdate,
+    unified_timestamp,
     urlencode_postdata,
+    urljoin,
 )
 
 
-class BitChuteIE(InfoExtractor):
+class BitChuteBaseIE(InfoExtractor):
+    _USER_AGENT = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.57 Safari/537.36'
+
+    def _list_entries(self, list_id):
+        TOKEN = 'zyG6tQcGPE5swyAEFLqKUwMuMMuF6IO2DZ6ZDQjGfsL0e4dcTLwqkTTul05Jdve7'
+        list_url = self._API_URL + list_id
+        offset = 0
+        query = {'showall': '1'} if 18 <= int_or_none(self._downloader.params.get('age_limit')) or 0 else None
+        for page_num in itertools.count(1):
+            data = self._download_json(
+                list_url + '/extend/', list_id,
+                'Downloading list page %d' % (page_num, ),
+                data=urlencode_postdata({
+                    'csrfmiddlewaretoken': TOKEN,
+                    'name': '',
+                    'offset': offset,
+                }), headers={
+                    'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
+                    'Referer': list_url,
+                    'X-Requested-With': 'XMLHttpRequest',
+                    'Cookie': 'csrftoken=' + TOKEN,
+                }, query=query)
+            if data.get('success') is False:
+                break
+            html = data.get('html')
+            if not html:
+                break
+            video_ids = re.findall(
+                r'''class\s*=\s*["'](?:channel-videos-)?image-container[^>]+>\s*<a\b[^>]+\bhref\s*=\s*["']/video/([^"'/]+)''',
+                html)
+            if not video_ids:
+                break
+            offset += len(video_ids)
+            for video_id in video_ids:
+                yield self.url_result(
+                    'https://www.bitchute.com/video/' + video_id,
+                    ie=BitChuteIE.ie_key(), video_id=video_id)
+
+    def _search_title(self, html, title_id, **kwargs):
+        return (
+            clean_html(get_element_by_id(title_id, html)) or None
+            or self._og_search_title(html, default=None)
+            or self._html_search_regex(r'(?s)<title\b[^>]*>.*?</title', html, 'title', **kwargs))
+
+    def _search_description(self, html, descr_id):
+        return (
+            self._og_search_description(html)
+            or (descr_id and clean_html(get_element_by_id(descr_id, html)))
+            or self._html_search_regex(
+                r'(?s)<div\b[^>]+\bclass=["\']full hidden[^>]+>(.+?)</div>',
+                html, 'description', fatal=False))
+
+
+class BitChuteIE(BitChuteBaseIE):
     _VALID_URL = r'https?://(?:www\.)?bitchute\.com/(?:video|embed|torrent/[^/]+)/(?P<id>[^/?#&]+)'
     _TESTS = [{
-        'url': 'https://www.bitchute.com/video/szoMrox2JEI/',
-        'md5': '66c4a70e6bfc40dcb6be3eb1d74939eb',
+        'url': 'https://www.bitchute.com/video/UGlrF9o9b-Q/',
+        'md5': '7e427d7ed7af5a75b5855705ec750e2b',
         'info_dict': {
-            'id': 'szoMrox2JEI',
+            'id': 'UGlrF9o9b-Q',
             'ext': 'mp4',
-            'title': 'Fuck bitches get money',
-            'description': 'md5:3f21f6fb5b1d17c3dee9cf6b5fe60b3a',
+            'title': 'This is the first video on #BitChute !',
+            'timestamp': 1483425420,
+            'upload_date': '20170103',
+            'description': 'md5:a0337e7b1fe39e32336974af8173a034',
             'thumbnail': r're:^https?://.*\.jpg$',
-            'uploader': 'Victoria X Rave',
-            'upload_date': '20170813',
+            'uploader': 'BitChute',
+            'age_limit': None,
+            'channel_url': 'https://www.bitchute.com/channel/bitchute/',
+        },
+    }, {
+        # NSFW (#24419)
+        'url': 'https://www.bitchute.com/video/wrTrKp7PmFZC/',
+        'md5': '4ef880ce8d24e322172d41a0cf6f8096',
+        'info_dict': {
+            'id': 'wrTrKp7PmFZC',
+            'ext': 'mp4',
+            'title': "You Can't Stop Progress | Episode 2",
+            'timestamp': 1541476920,
+            'upload_date': '20181106',
+            'description': 'md5:f191b538a2c4d8f57540141a6bfd7eb0',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': "You Can't Stop Progress",
+            'age_limit': 18,
+            'channel_url': 'https://www.bitchute.com/channel/ycsp/',
         },
     }, {
         'url': 'https://www.bitchute.com/embed/lbb5G1hjPhw/',
@@ -34,51 +114,90 @@ class BitChuteIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    @staticmethod
+    def _extract_urls(webpage):
+        urls = re.finditer(
+            r'''<(?:script|iframe)\b[^>]+\bsrc\s*=\s*("|')(?P<url>%s)''' % (BitChuteIE._VALID_URL, ),
+            webpage)
+        return (mobj.group('url') for mobj in urls)
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(
-            'https://www.bitchute.com/video/%s' % video_id, video_id, headers={
-                'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.57 Safari/537.36',
-            })
+        def get_error_title(html):
+            return clean_html(get_element_by_class('page-title', html)) or None
 
-        title = self._html_search_regex(
-            (r'<[^>]+\bid=["\']video-title[^>]+>([^<]+)', r'<title>([^<]+)'),
-            webpage, 'title', default=None) or self._html_search_meta(
-            'description', webpage, 'title',
-            default=None) or self._og_search_description(webpage)
+        def get_error_text(html):
+            return clean_html(get_element_by_id('page-detail', html)) or None
 
-        format_urls = []
-        for mobj in re.finditer(
-                r'addWebSeed\s*\(\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage):
-            format_urls.append(mobj.group('url'))
-        format_urls.extend(re.findall(r'as=(https?://[^&"\']+)', webpage))
+        webpage, urlh = self._download_webpage_handle(
+            'https://www.bitchute.com/video/' + video_id, video_id,
+            headers={
+                'User-Agent': self._USER_AGENT,
+            }, expected_status=404)
+
+        if urlh.getcode() == 404:
+            raise ExtractorError(get_error_title(webpage) or 'Cannot find video', expected=True)
+
+        title = self._search_title(webpage, 'video-title')
+
+        format_urls = [
+            mobj.group('url')
+            for mobj in re.finditer(
+                r'''\baddWebSeed\s*\(\s*("|')(?P<url>(?:(?!\1).)+)\1''', webpage)]
+        format_urls.extend(re.findall(r'''as=(https?://[^&"']+)''', webpage))
 
         formats = [
             {'url': format_url}
             for format_url in orderedSet(format_urls)]
 
         if not formats:
-            formats = self._parse_html5_media_entries(
-                url, webpage, video_id)[0]['formats']
+            entries = self._parse_html5_media_entries(
+                url, webpage, video_id)
+            if not entries:
+                error = clean_html(get_element_by_id('video-title', webpage)) or None
+                if error == 'Video Unavailable':
+                    raise GeoRestrictedError(error, expected=True)
+                error = get_error_title(webpage)
+                if error:
+                    reason = get_error_text(webpage)
+                    if reason:
+                        self.to_screen(reason)
+                    raise ExtractorError(error, expected=True)
+                raise ExtractorError('Cannot find video', )
+            formats = entries[0]['formats']
 
         self._check_formats(formats, video_id)
         self._sort_formats(formats)
 
-        description = self._html_search_regex(
-            r'(?s)<div\b[^>]+\bclass=["\']full hidden[^>]+>(.+?)</div>',
-            webpage, 'description', fatal=False)
-        thumbnail = self._og_search_thumbnail(
-            webpage, default=None) or self._html_search_meta(
-            'twitter:image:src', webpage, 'thumbnail')
+        description = (
+            self._search_description(webpage, 'video-description')
+            or self._html_search_regex(
+                r'(?s)<div\b[^>]+\bclass=["\']full hidden[^>]+>(.+?)</div>',
+                webpage, 'description', fatal=False))
+        thumbnail = self._html_search_meta(
+            ('og:image', 'twitter:image:src'), webpage, 'thumbnail', fatal=False)
         uploader = self._html_search_regex(
-            (r'(?s)<div class=["\']channel-banner.*?<p\b[^>]+\bclass=["\']name[^>]+>(.+?)</p>',
-             r'(?s)<p\b[^>]+\bclass=["\']video-author[^>]+>(.+?)</p>'),
+            (r'''(?s)<div\b[^>]+?\bclass\s*=\s*["']channel-banner.*?<p\b[^>]+\bclass\s*=\s*["']name\b[^>]+>(.+?)</p>''',
+             r'''(?s)<p\b[^>]+\bclass\s*=\s*["']video-author\b[^>]+>(.+?)</p>'''),
             webpage, 'uploader', fatal=False)
 
-        upload_date = unified_strdate(self._search_regex(
-            r'class=["\']video-publish-date[^>]+>[^<]+ at \d+:\d+ UTC on (.+?)\.',
-            webpage, 'upload date', fatal=False))
+        def more_unified_timestamp(x):
+            # ... at hh:mm TZ on month nth.
+            y = re.split(r'\s+at\s+', x or '')[-1]
+            y = re.sub(r'(?:^\s+|\s+$|\.+$|(?<=\d)(?:st|nd|rd|th))', '', y)
+            y = ' '.join(reversed(re.split(r'\s+on\s+', y, 1)))
+            return unified_timestamp(y) or unified_timestamp(x)
+
+        timestamp = more_unified_timestamp(get_element_by_class('video-publish-date', webpage))
+
+        # TODO: remove this work-around for class matching bug
+        webpage = re.split(r'''('|")channel-banner\1''', webpage, 1)[-1]
+        channel_details = get_element_by_class('details', webpage)
+        channel_details = channel_details and get_element_by_class('name', channel_details)
+        channel_url = urljoin(url, self._search_regex(
+            r'''<a\b[^>]*?\bhref\s*=\s*('|")(?P<url>(?:(?!\1).)+)''',
+            channel_details or '', 'channel url', group='url', default=None))
 
         return {
             'id': video_id,
@@ -86,57 +205,97 @@ class BitChuteIE(InfoExtractor):
             'description': description,
             'thumbnail': thumbnail,
             'uploader': uploader,
-            'upload_date': upload_date,
+            'timestamp': timestamp,
             'formats': formats,
+            'age_limit': 18 if '>This video has been marked as Not Safe For Work' in webpage else None,
+            'channel_url': channel_url,
         }
 
 
-class BitChuteChannelIE(InfoExtractor):
+class BitChuteChannelIE(BitChuteBaseIE):
     _VALID_URL = r'https?://(?:www\.)?bitchute\.com/channel/(?P<id>[^/?#&]+)'
-    _TEST = {
-        'url': 'https://www.bitchute.com/channel/victoriaxrave/',
-        'playlist_mincount': 185,
+    _TESTS = [{
+        'url': 'https://www.bitchute.com/channel/livesonnet/',
+        'playlist_mincount': 135,
         'info_dict': {
-            'id': 'victoriaxrave',
+            'id': 'livesonnet',
+            'title': 'livesonnet_vidz',
+            'description': 'md5:b0017be20656a1347eeb84f1049fc424',
+        }
+    }, {
+        # channel with NSFW content, not listed
+        'url': 'https://www.bitchute.com/channel/hQl9oMSgUyMX/',
+        'playlist_maxcount': 150,
+        'info_dict': {
+            'id': 'hQl9oMSgUyMX',
+            'title': "You Can't Stop Progress",
+            'description': 'md5:a7e3fd8cf02e96ddcc73e9f13d2ce768',
         },
-    }
-
-    _TOKEN = 'zyG6tQcGPE5swyAEFLqKUwMuMMuF6IO2DZ6ZDQjGfsL0e4dcTLwqkTTul05Jdve7'
-
-    def _entries(self, channel_id):
-        channel_url = 'https://www.bitchute.com/channel/%s/' % channel_id
-        offset = 0
-        for page_num in itertools.count(1):
-            data = self._download_json(
-                '%sextend/' % channel_url, channel_id,
-                'Downloading channel page %d' % page_num,
-                data=urlencode_postdata({
-                    'csrfmiddlewaretoken': self._TOKEN,
-                    'name': '',
-                    'offset': offset,
-                }), headers={
-                    'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
-                    'Referer': channel_url,
-                    'X-Requested-With': 'XMLHttpRequest',
-                    'Cookie': 'csrftoken=%s' % self._TOKEN,
-                })
-            if data.get('success') is False:
-                break
-            html = data.get('html')
-            if not html:
-                break
-            video_ids = re.findall(
-                r'class=["\']channel-videos-image-container[^>]+>\s*<a\b[^>]+\bhref=["\']/video/([^"\'/]+)',
-                html)
-            if not video_ids:
-                break
-            offset += len(video_ids)
-            for video_id in video_ids:
-                yield self.url_result(
-                    'https://www.bitchute.com/video/%s' % video_id,
-                    ie=BitChuteIE.ie_key(), video_id=video_id)
+    }, {
+        # channel with NSFW content, listed with adult age limit
+        'url': 'https://www.bitchute.com/channel/hQl9oMSgUyMX/',
+        'playlist_mincount': 160,
+        'info_dict': {
+            'id': 'hQl9oMSgUyMX',
+            'title': "You Can't Stop Progress",
+            'description': 'md5:a7e3fd8cf02e96ddcc73e9f13d2ce768',
+        },
+        'params': {
+            'age_limit': 18,
+        }
+    }]
+    _API_URL = 'https://www.bitchute.com/channel/'
 
     def _real_extract(self, url):
         channel_id = self._match_id(url)
-        return self.playlist_result(
-            self._entries(channel_id), playlist_id=channel_id)
+
+        webpage = self._download_webpage(
+            'https://www.bitchute.com/channel/' + channel_id, channel_id,
+            headers={
+                'User-Agent': self._USER_AGENT,
+            })
+
+        title = self._search_title(webpage, 'channel-title', default=None)
+
+        result = self.playlist_result(
+            self._list_entries(channel_id), playlist_id=channel_id, playlist_title=title)
+
+        return merge_dicts(
+            result,
+            {'description': self._search_description(webpage, 'channel-description')})
+
+
+class BitChutePlaylistIE(BitChuteBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?bitchute\.com/playlist/(?P<id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'https://www.bitchute.com/playlist/g4WTfWTdYEQa/',
+        'playlist_mincount': 1,
+        'info_dict': {
+            'id': 'g4WTfWTdYEQa',
+            'title': 'Podcasts',
+            'description': 'Podcast Playlist',
+        },
+    }]
+    _API_URL = 'https://www.bitchute.com/playlist/'
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+
+        webpage = self._download_webpage(
+            'https://www.bitchute.com/playlist/' + playlist_id, playlist_id,
+            headers={
+                'User-Agent': self._USER_AGENT,
+            })
+
+        title = self._search_title(webpage, 'playlist-title', default=None)
+
+        result = self.playlist_result(
+            self._list_entries(playlist_id), playlist_id=playlist_id, playlist_title=title)
+
+        description = (
+            clean_html(get_element_by_class('description', webpage))
+            or self._search_description(webpage, None))
+
+        return merge_dicts(
+            result,
+            {'description': description})

youtube_dl/extractor/common.py

@@ -3033,7 +3033,6 @@ class InfoExtractor(object):
             transform_source=transform_source, default=None)
 
     def _extract_jwplayer_data(self, webpage, video_id, *args, **kwargs):
-
         # allow passing `transform_source` through to _find_jwplayer_data()
         transform_source = kwargs.pop('transform_source', None)
         kwfind = compat_kwargs({'transform_source': transform_source}) if transform_source else {}

youtube_dl/extractor/extractors.py

@@ -132,6 +132,7 @@ from .biobiochiletv import BioBioChileTVIE
 from .bitchute import (
     BitChuteIE,
     BitChuteChannelIE,
+    BitChutePlaylistIE,
 )
 from .biqle import BIQLEIE
 from .bleacherreport import (

youtube_dl/extractor/palcomp3.py

@@ -8,7 +8,7 @@ from ..compat import compat_str
 from ..utils import (
     int_or_none,
     str_or_none,
-    try_get,
+    traverse_obj,
 )
 
 
@@ -109,7 +109,7 @@ class PalcoMP3ArtistIE(PalcoMP3BaseIE):
     }
     name'''
 
-    @ classmethod
+    @classmethod
     def suitable(cls, url):
         return False if re.match(PalcoMP3IE._VALID_URL, url) else super(PalcoMP3ArtistIE, cls).suitable(url)
 
@@ -118,7 +118,8 @@ class PalcoMP3ArtistIE(PalcoMP3BaseIE):
         artist = self._call_api(artist_slug, self._ARTIST_FIELDS_TMPL)['artist']
 
         def entries():
-            for music in (try_get(artist, lambda x: x['musics']['nodes'], list) or []):
+            for music in traverse_obj(artist, (
+                    'musics', 'nodes', lambda _, m: m['musicID'])):
                 yield self._parse_music(music)
 
         return self.playlist_result(
@@ -137,7 +138,7 @@ class PalcoMP3VideoIE(PalcoMP3BaseIE):
             'title': 'Maiara e Maraisa - Você Faz Falta Aqui - DVD Ao Vivo Em Campo Grande',
             'description': 'md5:7043342c09a224598e93546e98e49282',
             'upload_date': '20161107',
-            'uploader_id': 'maiaramaraisaoficial',
+            'uploader_id': '@maiaramaraisaoficial',
             'uploader': 'Maiara e Maraisa',
         }
     }]

youtube_dl/jsinterp.py

@@ -14,6 +14,7 @@ from .utils import (
     remove_quotes,
     unified_timestamp,
     variadic,
+    write_string,
 )
 from .compat import (
     compat_basestring,
@@ -53,15 +54,16 @@ def wraps_op(op):
 
 # NB In principle NaN cannot be checked by membership.
 # Here all NaN values are actually this one, so _NaN is _NaN,
-# although _NaN != _NaN.
+# although _NaN != _NaN. Ditto Infinity.
 
 _NaN = float('nan')
+_Infinity = float('inf')
 
 
 def _js_bit_op(op):
 
     def zeroise(x):
-        return 0 if x in (None, JS_Undefined, _NaN) else x
+        return 0 if x in (None, JS_Undefined, _NaN, _Infinity) else x
 
     @wraps_op(op)
     def wrapped(a, b):
@@ -84,7 +86,7 @@ def _js_arith_op(op):
 def _js_div(a, b):
     if JS_Undefined in (a, b) or not (a or b):
         return _NaN
-    return operator.truediv(a or 0, b) if b else float('inf')
+    return operator.truediv(a or 0, b) if b else _Infinity
 
 
 def _js_mod(a, b):
@@ -220,6 +222,42 @@ class LocalNameSpace(ChainMap):
         return 'LocalNameSpace%s' % (self.maps, )
 
 
+class Debugger(object):
+    ENABLED = False
+
+    @staticmethod
+    def write(*args, **kwargs):
+        level = kwargs.get('level', 100)
+
+        def truncate_string(s, left, right=0):
+            if s is None or len(s) <= left + right:
+                return s
+            return '...'.join((s[:left - 3], s[-right:] if right else ''))
+
+        write_string('[debug] JS: {0}{1}\n'.format(
+            '  ' * (100 - level),
+            ' '.join(truncate_string(compat_str(x), 50, 50) for x in args)))
+
+    @classmethod
+    def wrap_interpreter(cls, f):
+        def interpret_statement(self, stmt, local_vars, allow_recursion, *args, **kwargs):
+            if cls.ENABLED and stmt.strip():
+                cls.write(stmt, level=allow_recursion)
+            try:
+                ret, should_ret = f(self, stmt, local_vars, allow_recursion, *args, **kwargs)
+            except Exception as e:
+                if cls.ENABLED:
+                    if isinstance(e, ExtractorError):
+                        e = e.orig_msg
+                    cls.write('=> Raises:', e, '<-|', stmt, level=allow_recursion)
+                raise
+            if cls.ENABLED and stmt.strip():
+                if should_ret or not repr(ret) == stmt:
+                    cls.write(['->', '=>'][should_ret], repr(ret), '<-|', stmt, level=allow_recursion)
+            return ret, should_ret
+        return interpret_statement
+
+
 class JSInterpreter(object):
     __named_object_counter = 0
 
@@ -307,8 +345,7 @@ class JSInterpreter(object):
     def __op_chars(cls):
         op_chars = set(';,[')
         for op in cls._all_operators():
-            for c in op[0]:
-                op_chars.add(c)
+            op_chars.update(op[0])
         return op_chars
 
     def _named_object(self, namespace, obj):
@@ -326,9 +363,8 @@ class JSInterpreter(object):
         # collections.Counter() is ~10% slower in both 2.7 and 3.9
         counters = dict((k, 0) for k in _MATCHING_PARENS.values())
         start, splits, pos, delim_len = 0, 0, 0, len(delim) - 1
-        in_quote, escaping, skipping = None, False, 0
-        after_op, in_regex_char_group = True, False
-
+        in_quote, escaping, after_op, in_regex_char_group = None, False, True, False
+        skipping = 0
         for idx, char in enumerate(expr):
             paren_delta = 0
             if not in_quote:
@@ -382,10 +418,12 @@ class JSInterpreter(object):
         return separated[0][1:].strip(), separated[1].strip()
 
     @staticmethod
-    def _all_operators():
-        return itertools.chain(
-            # Ref: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/Operator_Precedence
-            _SC_OPERATORS, _LOG_OPERATORS, _COMP_OPERATORS, _OPERATORS)
+    def _all_operators(_cached=[]):
+        if not _cached:
+            _cached.extend(itertools.chain(
+                # Ref: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/Operator_Precedence
+                _SC_OPERATORS, _LOG_OPERATORS, _COMP_OPERATORS, _OPERATORS))
+        return _cached
 
     def _operator(self, op, left_val, right_expr, expr, local_vars, allow_recursion):
         if op in ('||', '&&'):
@@ -416,7 +454,7 @@ class JSInterpreter(object):
         except Exception as e:
             if allow_undefined:
                 return JS_Undefined
-            raise self.Exception('Cannot get index {idx:.100}'.format(**locals()), expr=repr(obj), cause=e)
+            raise self.Exception('Cannot get index {idx!r:.100}'.format(**locals()), expr=repr(obj), cause=e)
 
     def _dump(self, obj, namespace):
         try:
@@ -438,6 +476,7 @@ class JSInterpreter(object):
     _FINALLY_RE = re.compile(r'finally\s*\{')
     _SWITCH_RE = re.compile(r'switch\s*\(')
 
+    @Debugger.wrap_interpreter
     def interpret_statement(self, stmt, local_vars, allow_recursion=100):
         if allow_recursion < 0:
             raise self.Exception('Recursion limit reached')
@@ -511,7 +550,6 @@ class JSInterpreter(object):
                 expr = self._dump(inner, local_vars) + outer
 
         if expr.startswith('('):
-
             m = re.match(r'\((?P<d>[a-z])%(?P<e>[a-z])\.length\+(?P=e)\.length\)%(?P=e)\.length', expr)
             if m:
                 # short-cut eval of frequently used `(d%e.length+e.length)%e.length`, worth ~6% on `pytest -k test_nsig`
@@ -693,7 +731,7 @@ class JSInterpreter(object):
                 (?P<op>{_OPERATOR_RE})?
                 =(?!=)(?P<expr>.*)$
             )|(?P<return>
-                (?!if|return|true|false|null|undefined)(?P<name>{_NAME_RE})$
+                (?!if|return|true|false|null|undefined|NaN|Infinity)(?P<name>{_NAME_RE})$
             )|(?P<indexing>
                 (?P<in>{_NAME_RE})\[(?P<idx>.+)\]$
             )|(?P<attribute>
@@ -727,11 +765,12 @@ class JSInterpreter(object):
             raise JS_Break()
         elif expr == 'continue':
             raise JS_Continue()
-
         elif expr == 'undefined':
             return JS_Undefined, should_return
         elif expr == 'NaN':
             return _NaN, should_return
+        elif expr == 'Infinity':
+            return _Infinity, should_return
 
         elif md.get('return'):
             return local_vars[m.group('name')], should_return
@@ -760,18 +799,28 @@ class JSInterpreter(object):
             right_expr = separated.pop()
             # handle operators that are both unary and binary, minimal BODMAS
             if op in ('+', '-'):
+                # simplify/adjust consecutive instances of these operators
                 undone = 0
                 while len(separated) > 1 and not separated[-1].strip():
                     undone += 1
                     separated.pop()
                 if op == '-' and undone % 2 != 0:
                     right_expr = op + right_expr
+                elif op == '+':
+                    while len(separated) > 1 and separated[-1].strip() in self.OP_CHARS:
+                        right_expr = separated.pop() + right_expr
+                # hanging op at end of left => unary + (strip) or - (push right)
                 left_val = separated[-1]
                 for dm_op in ('*', '%', '/', '**'):
                     bodmas = tuple(self._separate(left_val, dm_op, skip_delims=skip_delim))
                     if len(bodmas) > 1 and not bodmas[-1].strip():
                         expr = op.join(separated) + op + right_expr
-                        right_expr = None
+                        if len(separated) > 1:
+                            separated.pop()
+                            right_expr = op.join((left_val, right_expr))
+                        else:
+                            separated = [op.join((left_val, right_expr))]
+                            right_expr = None
                         break
                 if right_expr is None:
                     continue
@@ -797,6 +846,8 @@ class JSInterpreter(object):
 
             def eval_method():
                 if (variable, member) == ('console', 'debug'):
+                    if Debugger.ENABLED:
+                        Debugger.write(self.interpret_expression('[{}]'.format(arg_str), local_vars, allow_recursion))
                     return
                 types = {
                     'String': compat_str,

youtube_dl/utils.py

@@ -2406,7 +2406,7 @@ class ExtractorError(YoutubeDLError):
         """ tb, if given, is the original traceback (so that it can be printed out).
         If expected is set, this is a normal error message and most likely not a bug in youtube-dl.
         """
-
+        self.orig_msg = msg
         if sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
             expected = True
         if video_id is not None: