Re: [Gnumed-devel] signalling with passing of parameters

[Karsten Hilbert]

> I think it is. Passing parameters with the message makes sense in all 
> cases where the parameters are likely to be reused by a large number of 
> client widgets, and where the number of parameters is limited.
> It makes a noticeable difference in performance.
Sure. We shouldn't let this practice proliferate all too
profusely, though. Eventually, if properly cached, does it
really make such a big difference ? I mean, when A sends off
"patient_changed" the cached patient will have been changed
and cached (that's the point of sending the signal after all).
Now, whether B accesses gmCachedPerson.CachedPerson.dob or
kwds['dob'] doesn't seem much different to me in terms of
overhead ?

> I need to rewrite gmCachedPerson completely first. In it's current 
That's what I thought. Just wanted to make sure. Reason being
that for some of the work we are doing (using GnuMed as a
companion to other software) I need to implement
"patient_locked" to prevent changing the patient in this
instance of the frontend.

Karsten
-- 
GPG key ID E4071346 @ wwwkeys.pgp.net
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346