yumming up down to for loop

[seth vidal]

On Mon, 2004-09-06 at 23:47 -0500, Harry Putnam wrote:
> seth vidal <skvidal at phy.duke.edu> writes:
> 
> > I'd like to remind everyone the PURPOSE of a test release is to get
> > tests and bugs.
> >
> > If you encounter a problem like this it is either:
> > - a packaging bug
> > - a bug in yum
> > - a bug in rpm
> 
> Seth, not to put too fine a point on it, but so far you have
> redirected the subject twice... fussed about a typo that was very
> clear in intent, and have yet to say ONE WORD about the subject.

I got back from a short break and had 30 email about yum in my mailbox.
I wasn't trying to redirect anything. I was trying to get through my
email.


> If you don't know a way to cache the read that happens on each
> invocation of yum, its ok to say so.  But why keep dancing around
> injecting some new subject.  You could as well post reminders and the
> like to your own thread.  Or if you have other subject to discuss,
> start a thread about them.  Why do you insist on derailing my subject?

Why do you insist on assuming it's malevolent.

Caching that data is not all that wise right now. The time is better
spent making sure what's happening in the depresolution and installation
is correct.

remember:
make it work
make it work correctly
make it work fast

in that order.

You want to cache that data. Yah it can be done, but from the tests I've
done so far loading the data off disk from a cPickle is going to take
90% of the time it will take to read it from the xml.

starts to look a little less valuable to add complexity, then.


-sv