diff options
| author | Matthew <coletdjnz@protonmail.com> | 2022-11-10 19:35:22 +1300 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2022-11-10 06:35:22 +0000 |
| commit | e72e48c53f16771ea7d786deb6b65a40d82a14c4 (patch) | |
| tree | f3fb45a1896f3854c9d8a75def3e47311d30a236 | |
| parent | 0cf643b234ff2f4d017a980dbaefdb14ed6e4db6 (diff) | |
| download | hypervideo-pre-e72e48c53f16771ea7d786deb6b65a40d82a14c4.tar.lz hypervideo-pre-e72e48c53f16771ea7d786deb6b65a40d82a14c4.tar.xz hypervideo-pre-e72e48c53f16771ea7d786deb6b65a40d82a14c4.zip | |
[extractor/youtube] Ignore incomplete data error for comment replies (#5490)
When --ignore-errors is used.
Closes https://github.com/yt-dlp/yt-dlp/issues/4669
Authored by: coletdjnz
| -rw-r--r-- | yt_dlp/extractor/youtube.py | 20 |
1 files changed, 15 insertions, 5 deletions
diff --git a/yt_dlp/extractor/youtube.py b/yt_dlp/extractor/youtube.py index 5b7c94c4e..5b39f9765 100644 --- a/yt_dlp/extractor/youtube.py +++ b/yt_dlp/extractor/youtube.py @@ -3237,11 +3237,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor): note_prefix = '%sDownloading comment%s API JSON page %d %s' % ( ' ' if parent else '', ' replies' if parent else '', page_num, comment_prog_str) - - response = self._extract_response( - item_id=None, query=continuation, - ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix, - check_get_keys='onResponseReceivedEndpoints' if not is_forced_continuation else None) + try: + response = self._extract_response( + item_id=None, query=continuation, + ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix, + check_get_keys='onResponseReceivedEndpoints' if not is_forced_continuation else None) + except ExtractorError as e: + # Ignore incomplete data error for replies if retries didn't work. + # This is to allow any other parent comments and comment threads to be downloaded. + # See: https://github.com/yt-dlp/yt-dlp/issues/4669 + if 'incomplete data' in str(e).lower() and parent and self.get_param('ignoreerrors') is True: + self.report_warning( + 'Received incomplete data for a comment reply thread and retrying did not help. ' + 'Ignoring to let other comments be downloaded.') + else: + raise is_forced_continuation = False continuation_contents = traverse_obj( response, 'onResponseReceivedEndpoints', expected_type=list, default=[]) |