[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: red hat 9 != 9.0
From: |
Andrew Stribblehill |
Subject: |
Re: red hat 9 != 9.0 |
Date: |
Wed, 9 Apr 2003 00:53:16 +0100 |
User-agent: |
Mutt/1.5.4i |
Quoting Paul Heinlein <heinlein@cse.ogi.edu> (2003-04-08 05:56:59 BST):
> On Red Hat 9 systems, /etc/redhat-release reads
>
> Red Hat Linux release 9 (Shrike)
>
> src/misc.c won't parse this correctly because it scans for "%d.%d". I
> guess there are two ways to set the classes:
>
> * just let it be redhat_9
> * by fiat, declare it both redhat_9 and redhat_9_0
>
> The patch below, made against 2.0.6, takes the first approach, though
> I certainly can understand the arguments for the second. (Warning: I'm
> not a C hacker. This looks right to me, and it seems to work as
> expected on Linux/gcc boxes, but any fixes are certainly welcome.)
I hear that there's precedent for RedHat's marketing department to
drop the .0 -- apparently Redhat 7.0 was only retro-actively labelled
that. On this assumption, I think we should declare an implicit .0
for RH 9.
RedHat have always had genuine reasons for bumping the major version
number and I expect that if their next release doesn't have
compatibility problems with Shrike, they'll number it 9.1.
<plug type="shameless">If you want an example of code which groks 0,
1 or 2 numbers in a release name, look at the Debian release
parsing.</plug>
--
MALIN HEBRIDES
SOUTH OR SOUTHEAST 3 OR 4, OCCASIONALLY 5 IN WEST. MAINLY FAIR.
MODERATE OR GOOD
- red hat 9 != 9.0, Paul Heinlein, 2003/04/08
- Re: red hat 9 != 9.0,
Andrew Stribblehill <=