Jan Rychter wrote:
> It would need to be measured. I have suspicions that building widget
> trees on every request is expensive, especially as one usually makes
> database calls when building widgets. And in case of a bounce request or
> a bot request all that stuff sits around in memory until the session
> gets expired.
You're right of course, but what I said was meant to only apply to
low-level session init, not the generation and storage of the whole
widget tree.
So I think we should separate the issue of session generation from
the issue of eager widget tree creation (which a proper lazy navigation
macro would solve).
> Sounds good. My problem with (1) is that I don't understand why there
> are three calls to redirect in handle-client-request and what purpose
> they serve.
My analysis of those three redirects, in order of occurence:
(1) seems to be unnecessary.
(2) seems slightly off too -- why would we need to prune the
session id from the URI if a cookie came in? Might warrant
some quick testing of what's going on and what removal
would change.
(3) is the only one that makes sense to me: when we're talking
to a non-AJAX client (or in some other way get passed the
action via normal GET) we want to remove it from the URI
to avoid it being called again on refresh.
In my opinion we should just remove the first two.
(3) is more of an issue; we should ensure that actions will
get called only once (or find some other solution for this)
before removing it.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"weblocks" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/weblocks?hl=en
-~----------~----~----~----~------~----~------~--~---
