I got caching of the navigation stuff working nicely. It isn't as efficient as it could be because it constructs the "tree" for every URL, but at least it doesn't re-render for the same URL thanks to caching.
The nav_tree in general is not very efficient, and the sooner it gets rewritten, the better. This one is from Alan Runyan (runyaga) <tal:checkperms tal:define="global checkPermission nocall:here/portal_membership/checkPermission; global wtool nocall:here/portal_workflow;"> <metal:portlet use-macro="here/portlet_navigation/macros/portlet" /> </tal:checkperms>
Here is another question for you. I notice in my profiling that two things are just killing me. They are:
here/getAllowedTypes portal.list_filtered_actions(here)
I have associated both of these with a RamCache that is not very restrictive (in terms of REQUEST contents, i.e. AUTHENTICATED_USER, URL1) and yet they never get cached. Any idea why? They are python scripts (in case that matters). Any more clever ideas?
No, I don't know why they aren't getting cached. Maybe one of the caching gurus on the list can answer that one. Nate
On Thu, May 27, 2004 at 03:14:52PM +0200, Nate Aune wrote:
Here is another question for you. I notice in my profiling that two things are just killing me. They are:
here/getAllowedTypes portal.list_filtered_actions(here)
I have associated both of these with a RamCache that is not very restrictive (in terms of REQUEST contents, i.e. AUTHENTICATED_USER, URL1) and yet they never get cached. Any idea why? They are python scripts (in case that matters). Any more clever ideas?
Python scripts are usually fine candidates for RAM caching. What do you mean by "they never get cached"? What do you see in the cache statistics? What exactly *is* in your REQUEST variable settings for this cache manager? I'd suspect something's not right there. -- Paul Winkler http://www.slinkp.com
participants (2)
-
Nate Aune -
Paul Winkler