From: "Sam Ruby" <[EMAIL PROTECTED]> wrote:
> Adam R. B. Jack wrote:
>
> > (it is slow to load 600 projects)
>
> why?
Funny you should ask, it oughtn't be. Simple answer, I don't know.
I've had suspicions, I've even did a quick performance analysis (distance
past), but I've never been able to track it down. I know that in the early
days of Gumpy, Nicola would enjoy how fast it was to load metadata, but the
trouble was that was from a local cache of the HTTP|viewcvs files (it did
not check with the original source, if the metadata had changed).
I wondered if it was the fact that I was mapping the raw XML metadata over
to some peer objects, but when I did an analysis I didn't think that was
expensive. I did some profiling,
#print 'Profiling....'
#import profile
#profile.run('irun()', 'iprof')
irun()
and found inordinately high numbers of some dictionary related method down
in the XML classes, but got out of my depth at that point.
I would really appreciate some assistence with debuggin that, 'cos I know it
wasn't originally felt to be expensive & would like to see it return to
that.
regards
Adam
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
