Conor MacNeill wrote:
>
> I think this should have been done on the 1.3 branch, so that
> we have it right for the 1.3 release. I can merge it over if
> you like.
>
> Also, on 1.3 the index.html file no longer exists. I have
> already updated tar.html with the longfile attribute.

Sorry, I missed your orignal response, and figured I would press on.  Now
that I found it, I will merge the code over and update the documentation.
(I know that in 1.3 the doc doesn't exist, but I was pressing on in 1.4).

> Finally, you have made truncate work more like omit. When I
> read up on the GNU format, I am pretty sure it said that
> POSIX standard is to truncate the path but include the file
> in the archive. Do we want to have an omit AND truncate
> option.

Take a close look at TarOutputStream.putNextEntry.  See if you come to the
same conclusion that I did - namely before my change truncate meant omit.
What do you think should be done?

> We should probably also only print the statement
>
> "Resulting tar file can only be processed successfully by
> GNU compatible tar commands"

OK, I plead guitly to laziness.  Answer my previous question and then I
will do the necessary cleanup.  I also assume that you are otherwise OK
with the warn behavior I have implemented?

- Sam Ruby

Reply via email to