Re: Is this the Debian Philosophy? (or.... $#@!@#$ bash 2.0!)
"Jens B. Jorgensen" <jjorgens@bdsinc.com> writes:
> This really annoys me. (No, not you Jason.) I agree that we have to
> move on the the next distribution, but given the fact that
> <speculation> *most* </speculation> debian 1.3.1 users use netscape
> and that <speculation> *most* </speculation> of those folks will
> want to use a plug-in, I think this bug merits an upgrade to 2.01
> bash for 1.3.1?
I don't... Here's my slant on Bash 2.01 (since we seem to be into
rants)
* 2.01 is a lot of new and untested code
Already, afaik, at least two bugs have been found, one is the segfault
on declare -p and the other a really nasty bug in libreadline which
causes a segfault on tab completion in a number of the programs which
use libreadline. The Debian policy of "no new code, expect for urgent
security fixes" in stable, is IMHO, a good one. If it weren't for
that, these bugs might have been discovered in the "stable" Debian
1.3.x tree.
* 2.01-0.1 is a *non-maintainer* release
About when 2.01 was released the real maintainer, Guy Maor, was off on
a month along (announced) holiday. I did 2.01 because it fixed the
set -a; set +a man bug and because I urgently wanted a libc6
libreadline. (I don't use netscape and barely knew about the ``bug'',
never mind cared about it). I am not the real maintainer for a very
good reason (apart from that there already is a maintainer), I wasn't
able to port all the changes Guy had done to bash 2.0's libreadline to
bash 2.01's libreadline or adapt one of his security fixes for bash
itself. What I did do in the end worked, and would do for a while *in
unstable*, until Guy got back and could fix my kludged solution.
Guy is of course back, but he's been even more busy than usual with
the move of master, so he hasn't done a proper release yet.
The two reasons above are two good reasons why there is no bash 2.01
in stable. Even if the second one is fixed by Guy doing a proper
release, I'm still of the opinion that there is too much new and
untested code in bash 2.01 for it to go into stable (the two bugs
found so far being excellent arguments for that position).
You just cannot put untested code into a stable tree and have it cause
gdb, es, etc. to segfault on <tab>. That is _not_ stable.
--
James
--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
debian-user-request@lists.debian.org .
Trouble? e-mail to templin@bucknell.edu .
Reply to: