svn commit: r376772 - /forrest/events/forrest-friday/20060210-log.txt

[cr...@apache.org]

Author: crossley
Date: Fri Feb 10 09:30:05 2006
New Revision: 376772

URL: http://svn.apache.org/viewcvs?rev=376772&view=rev
Log:
Update

Modified:
    forrest/events/forrest-friday/20060210-log.txt

Modified: forrest/events/forrest-friday/20060210-log.txt
URL: http://svn.apache.org/viewcvs/forrest/events/forrest-friday/20060210-log.txt?rev=376772&r1=376771&r2=376772&view=diff
==============================================================================
--- forrest/events/forrest-friday/20060210-log.txt (original)
+++ forrest/events/forrest-friday/20060210-log.txt Fri Feb 10 09:30:05 2006
@@ -671,3 +671,23 @@
 [03:20] _Gav_: background-image: url(../themes/images/roundcorner-​t-l-5-bgffffff-stroke4C6C8F-fg4C6C​8F.png);
 [03:21] _Gav_: Just change the colors in the url itself!
 [03:23] _Gav_: back later
+[03:30] rgardler: twilliams: are you about? I would like to ask your opinion on a possible LM caching solution
+[03:34] twilliams_: i'm here for a couple minutes.
+[03:37] rgardler: OK, I think I've found a solution - using the MultSourceValidity class, but...
+[03:38] rgardler: My concern is why doesn't the only inputmoudle that uses caching (the XMLFileInputModule) use "normal" cocoon pipeline cache components
+[03:38] rgardler: DO you have any insights into this qeustion?
+[03:39] twilliams_: it depends on the day i'm asked... sometimes I think i understand it; others, not so much:(
+[03:40] rgardler: :-))
+[03:40] rgardler: I suspect that means I should experiment then compare notes with you cia the mail list?
+[03:41] twilliams_: when i first looked into this, i assumed it didn't use "normal caching" because of it's inability at the module level to do implement CacheableProcessingComponent since it deals with multiple inputs
+[03:42] twilliams_: now I'm not so sure and need to better understand the lifecycle of these input components 
+[03:43] rgardler: Yes, that is my problem too... what if I said I may have an interim solution that does not implement CacheableProcessingComponent, but does use cocoons validity classes
+[03:43] rgardler: i.e. we could later move to implement the interface when we understand how
+[03:44] twilliams_: we're already using SourceValidity, i do like the idea of swapping our m_src out for the MultiSourceValidity though
+[03:44] twilliams_: then add a new source with every mount
+[03:45] rgardler: Oh. I didn't realise we already used SourceValidity - seems an ovcious move to multisourcevalidity then
+[03:45] rgardler: I'll do that we can do the rest on the mail list
+[03:45] twilliams_: ok
+[03:45] rgardler: Cool, thanks for your input
+[03:46] twilliams_: later...
+[03:46] rgardler: bye for now