diff options
Diffstat (limited to 'yt_dlp/extractor')
| -rw-r--r-- | yt_dlp/extractor/soundcloud.py | 34 | ||||
| -rw-r--r-- | yt_dlp/extractor/youtube.py | 15 |
2 files changed, 19 insertions, 30 deletions
diff --git a/yt_dlp/extractor/soundcloud.py b/yt_dlp/extractor/soundcloud.py index 0c6f0b0..afb512d 100644 --- a/yt_dlp/extractor/soundcloud.py +++ b/yt_dlp/extractor/soundcloud.py @@ -314,23 +314,11 @@ class SoundcloudBaseIE(InfoExtractor): self.write_debug(f'"{identifier}" is not a requested format, skipping') continue - stream = None - for retry in self.RetryManager(fatal=False): - try: - stream = self._call_api( - format_url, track_id, f'Downloading {identifier} format info JSON', - query=query, headers=self._HEADERS) - except ExtractorError as e: - if isinstance(e.cause, HTTPError) and e.cause.status == 429: - self.report_warning( - 'You have reached the API rate limit, which is ~600 requests per ' - '10 minutes. Use the --extractor-retries and --retry-sleep options ' - 'to configure an appropriate retry count and wait time', only_once=True) - retry.error = e.cause - else: - self.report_warning(e.msg) - - stream_url = traverse_obj(stream, ('url', {url_or_none})) + # XXX: if not extract_flat, 429 error must be caught where _extract_info_dict is called + stream_url = traverse_obj(self._call_api( + format_url, track_id, f'Downloading {identifier} format info JSON', + query=query, headers=self._HEADERS), ('url', {url_or_none})) + if invalid_url(stream_url): continue format_urls.add(stream_url) @@ -647,7 +635,17 @@ class SoundcloudIE(SoundcloudBaseIE): info = self._call_api( info_json_url, full_title, 'Downloading info JSON', query=query, headers=self._HEADERS) - return self._extract_info_dict(info, full_title, token) + for retry in self.RetryManager(): + try: + return self._extract_info_dict(info, full_title, token) + except ExtractorError as e: + if not isinstance(e.cause, HTTPError) or not e.cause.status == 429: + raise + self.report_warning( + 'You have reached the API rate limit, which is ~600 requests per ' + '10 minutes. Use the --extractor-retries and --retry-sleep options ' + 'to configure an appropriate retry count and wait time', only_once=True) + retry.error = e.cause class SoundcloudPlaylistBaseIE(SoundcloudBaseIE): diff --git a/yt_dlp/extractor/youtube.py b/yt_dlp/extractor/youtube.py index 18e0ee9..1c0a70d 100644 --- a/yt_dlp/extractor/youtube.py +++ b/yt_dlp/extractor/youtube.py @@ -3130,7 +3130,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): def _extract_n_function_name(self, jscode): funcname, idx = self._search_regex( - r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)', + r'''(?x)(?:\.get\("n"\)\)&&\(b=|b=String\.fromCharCode\(110\),c=a\.get\(b\)\)&&\(c=) + (?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)''', jscode, 'Initial JS player n function name', group=('nfunc', 'idx')) if not idx: return funcname @@ -3150,17 +3151,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): func_name = self._extract_n_function_name(jscode) - # For redundancy - func_code = self._search_regex( - rf'''(?xs){func_name}\s*=\s*function\s*\((?P<var>[\w$]+)\)\s* - # NB: The end of the regex is intentionally kept strict - {{(?P<code>.+?}}\s*return\ [\w$]+.join\(""\))}};''', - jscode, 'nsig function', group=('var', 'code'), default=None) - if func_code: - func_code = ([func_code[0]], func_code[1]) - else: - self.write_debug('Extracting nsig function with jsinterp') - func_code = jsi.extract_function_code(func_name) + func_code = jsi.extract_function_code(func_name) self.cache.store('youtube-nsig', player_id, func_code) return jsi, player_id, func_code |