[jira] [Commented] (COMPRESS-338) fill some gaps in tar implementation

["Stefan Bodewig (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/COMPRESS-338?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15208728#comment-15208728 ] 

Stefan Bodewig commented on COMPRESS-338:
-----------------------------------------

I've added an improved Lister class with commit 02e0f9a - we seem to detect (sym)links as well as GNU tar does.

> fill some gaps in tar implementation
> ------------------------------------
>
>                 Key: COMPRESS-338
>                 URL: https://issues.apache.org/jira/browse/COMPRESS-338
>             Project: Commons Compress
>          Issue Type: Improvement
>    Affects Versions: 1.10
>            Reporter: Stefan Bodewig
>
> The test archive linked to in COMPRESS-336 shows our tar code isn't completely on par with the python lib that uses it for its tests:
> * we don't seem to identify all kinds of links (./ustar/linktest2/symtype -> ../linktest1/regtype is identified as two entries, as is ./ustar/linktest2/lnktype link to ./ustar/linktest1/regtype)
> * we only seem to parse gnu/sparse-0.0 properly, not 0.1 or 1.0
> * {{Lister}} shows parts of the Pax-Headers as extra entries for the pax/ headers
> * I'm not sure about the non-ASCII characters in a bunch of file names
> It is probably best to pick the issues independently as subtasks of this.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)