[
https://issues.apache.org/jira/browse/COMPRESS-229?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13675659#comment-13675659
]
Christoph Gysin commented on COMPRESS-229:
------------------------------------------
On further investigation, they seem to have copied it back in 2009 to implement
UTF-8 encoded filenames:
https://github.com/jenkinsci/jenkins/commit/fcc75143941c49d94536ed8db90472b9633f8631
Since then, Ant's TarInputStream has been modified to support using different
encodings, so that hack should be undone.
I'll send them a patch to use the current upstream versions, and if possible
use commons-compress instead of Ant.
> incorrect handling of GNU longlink entries
> ------------------------------------------
>
> Key: COMPRESS-229
> URL: https://issues.apache.org/jira/browse/COMPRESS-229
> Project: Commons Compress
> Issue Type: Bug
> Components: Archivers
> Affects Versions: 1.4.1, 1.5
> Reporter: Christoph Gysin
> Fix For: 1.6
>
> Attachments:
> 0001-allow-creating-testsuites-over-different-archives.patch,
> 0002-add-unittest-to-expose-bug-with-long-symlinks.patch,
> 0003-split-out-function-to-retrieve-longname-data.patch,
> 0004-handle-GNU-long-link-name.patch
>
>
> Apache commons-compress handles GNU LongLink entries for long filenames. But
> it fails to handle LongLink entries for long linknames[1] correctly.
> [1] http://git.savannah.gnu.org/cgit/tar.git/tree/src/tar.h#n176
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
