This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
libgcj/7854: fastjar update (-u) should be implemented
- From: 116145 at bugs dot debian dot org
- To: gcc-gnats at gcc dot gnu dot org
- Date: 8 Sep 2002 07:27:47 -0000
- Subject: libgcj/7854: fastjar update (-u) should be implemented
- Reply-to: 116145 at bugs dot debian dot org
>Number: 7854
>Category: libgcj
>Synopsis: fastjar update (-u) should be implemented
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: unassigned
>State: open
>Class: change-request
>Submitter-Id: net
>Arrival-Date: Sun Sep 08 00:36:01 PDT 2002
>Closed-Date:
>Last-Modified:
>Originator: 116145@bugs.debian.org
>Release: 3.2 / HEAD
>Organization:
>Environment:
>Description:
[known, but not in GNATS yet, see http://bugs.debian.org/116145]
fastjar's -u option is disabled due to corrupted archives.
Per Bothner responded:
Yes, I ran into the same problem. Basically the implementation of 'u' is pretty much not there. I looked at trying to fix it, but decided I didn't have time. Until somebody fixes it, we should change fastjar to exit with an error on 'u'.
The first question is whether fastjar should try to update in place or create a temporary file and copy it over. The former is faster, especially when just adding files, and is less liable to run out of disk space.
However, it is tricky when files are replaced that are not at the end. In that case I guess the solution is to read the archive members that need to be moved into main memory.
Then it is just a small matter of programming ...
>How-To-Repeat:
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted: