[
https://issues.apache.org/jira/browse/VFS-626?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17572661#comment-17572661
]
Gary D. Gregory commented on VFS-626:
-------------------------------------
Hello [~asieira]
I think we'll need an example of the current code causing a problem, so a PR
with a failing unit test is key, otherwise, the change, one-liner or not, can
just be changed back at any time and cause a regression.
> Cannot open large bzip2 files
> -----------------------------
>
> Key: VFS-626
> URL: https://issues.apache.org/jira/browse/VFS-626
> Project: Commons VFS
> Issue Type: Bug
> Affects Versions: 2.1
> Reporter: Alexandre Sieira
> Priority: Major
> Labels: easyfix
>
> I was having problems reading a large bzip2 file in which only a part of the
> data was being read. So Googling a bit I came across this Apache Commons
> Compress issue: https://issues.apache.org/jira/browse/COMPRESS-224
> When looking at
> https://github.com/apache/commons-vfs/blob/7b42b5598774a40487dd59d699f5f865962822d9/core/src/main/java/org/apache/commons/vfs2/provider/bzip2/Bzip2FileObject.java#L61
> we can see that the BZip2CompressorInputStream constructor option
> `decompressConcatenated` is not defined and defaults to false, which can
> cause the observed behavior. This manifests itself in multi-stream bzip2
> files, which tends to be the case with larger files.
> I would suggest that this BZip2CompressorInputStream constructor option is
> explicitly set to true. It is a one-liner fix.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
