[Yum] Re: Weird yum behaviour maybe?

[Tom Diehl]

On 22 May 2003, seth vidal wrote:

> well don't expect that :)

Ok, I guess.

> 
> prior to rpm 4.2-0.66 rpm did alpha to numeric token comparisons in a
> nondeterministic way.
> 
> ie:
> 
> if you're comparing 'a' to 1 and you want to see which is greater then
> rpm used to say:
> a vs 1
> a is newer
> 1 vs a
> 1 is newer
> 
> see the problem :)

So you are saying it is totally dependant on the order??

> now it says:
> a vs 1
> 1 is newer
> 1 vs a
> 1 is newer
> 
> numbers always win vs letters
> so:
> z vs 1
> 1 is newer
> 1 vs z
> 1 is newer.

That makes sense I guess. I suspected it was something like this but what
confused me was the fact the rpm on the command line would do the upgrade.

> 
> I know jeff johnson reads this list occasionally, I wonder if the
> deterministic numeric-alpha comparison patch went into rpm 4.1.1
> 
> in short - yum can't do anything more than what rpm does here.
> 
> And the reason why rpm on the command line is doing it is mostly luck.
> It just matters which one goes into the rpm comparison.

I assume that given above you do not need the -d10 yum output?? :-)

FWIW 
(tigger pts2) $ rpm -q rpm
rpm-4.1-1.06
(tigger pts2) $

In the previous message you say: 
Also - if you can - update 8.0 to rpm 4.1.1 and just use yum 1.97(or one
of the dailies)

Are there any disadvantages to doing this?

Thanks for the info, I will simply upgrade by hand.

-- 
.............Tom	"Nothing would please me more than being able to 
tdiehl at rogueind.com	hire ten programmers and deluge the hobby market 
			with good software." -- Bill Gates 1976

   			We are still waiting ....