Hi,
> this upstream fix (commit 7d85774) could also allow gentoo to drop a patch
> that I wasn't willing to merge when I was still maintainer.
>
> https://bugs.gentoo.org/show_bug.cgi?id=309683
My primary motivation for making this change is consistency with gzip, bzip2
and xz. I didn't test it yet, but as far as I remember it should make
it more consistent.
It shouldn't break any valid lbzip2 usage because I am relaxing
requirements only,
but it should increase compatibility with bzip2 and ease migration to lbzip2.
If I remember correctly the Gentoo patch was doing something different
and I am not
going to adopt Gentoo behavior because it's neither symetric in terms
of compression/
decompression nor consistent with other utilities of this kind (i.e.
gzip, bzip2 and xz).
Your table from the Gentoo bug, extended with -t column:
(I may be wrong somewhere here because now I am writing enterily from what
I remember, without verifying anything.)
bzip2
-z -zc -zf -zcf -d -dc -df -dcf -t -tf
regular file 1 1 1 1 1 1 1 1 1 1
symlink to regular file 1 1 1 1 1 1 1 1
device 1 1 1 1 1 1 1 1
symlink to device 1 1 1 1 1 1 1 1
lbzip2 released (incl. Debian sid)
-z -zc -zf -zcf -d -dc -df -dcf -t -tf
regular file 1 1 1 1 1 1 1 1 1 1
symlink to regular file 1 1 1 1 1
device 1 1 1 1 1
symlink to device 1 1 1 1 1
lbzip2 development (git master)
-z -zc -zf -zcf -d -dc -df -dcf -t -tf
regular file 1 1 1 1 1 1 1 1 1 1
symlink to regular file M 1 1 M 1 1 M 1
device M 1 1 M 1 1 M 1
symlink to device M 1 1 M 1 1 M 1
Gentoo lbzip2
-z -zc -zf -zcf -d -dc -df -dcf -t -tf
regular file 1 1 1 1 1 1 1 1 1 1
symlink to regular file 1 1 G G 1 1 G 1
device 1 1 G G 1 1 G 1
symlink to device 1 1 G G 1 1 G 1
--
Mikolaj
--
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]
