[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [wsrp-interfaces] Why do we have an onLoad event and no onUnload?
yes, I think I tend to agree. Either we should rethink lifecycle more detailed or drop an uncomplete solution. Mit freundlichen Gruessen / best regards, Richard Jacob ______________________________________________________ IBM Lab Boeblingen, Germany Dept.8288, WebSphere Portal Server Development WSRP Team Lead & Technical Lead WSRP Standardization Phone: ++49 7031 16-3469 - Fax: ++49 7031 16-4888 Email: mailto:richard.jacob@de.ibm.com Michael Freedman <michael.freedman @oracle.com> To wsrp <wsrp@lists.oasis-open.org> 07/12/2005 06:19 cc PM WSRP Interfaces subgroup <wsrp-interfaces@lists.oasis-open.o rg> Subject Re: [wsrp-interfaces] Why do we have an onLoad event and no onUnload? Well I actually had a slightly different use case in mind for OnLoad. In the end there are a variety of lifecycle events it may be interesting to pass to the Portlet -- these include consumer [application] init/load and corresponding destroy/unload, user login/logout, initial "page" render [per user]. Though my use case pertains to the that later, our current wording allows for a variety of interpretations that likely makes it meaningless [i.e. can't get interoperability] to the producer. I wonder if we should just remvoe this from 2.0 and let consumer/producers experiment until we have more time to talk about consumer lifecycle events more thoroughly? -Mike- Richard Jacob wrote: >I haven't seen an answer to that so far. >I think Stefan is right here. If we have an onLoad, we should also have an >onUnload. >Please note that the onUnload does not serve the same purposes as >destroyPortlets() does. > >I tend to agree with Stefan here, the symmetry makes sense here, especially >when we pick up the use case mentioned below (comming from Mike initially I >think). >Here is a use justifying onUnload using the same arguments: > >OnLoad (try to recap Mike's use case) >1. user logs in (and will be routed to his initial page) >2. Consumer fires on load event to allow portlets on the initial page to >prepare the portlet usage i.e. establish DB connections or whatever -> this >minimizes the wait time if the user finally sees the page >3. user invokes initial page (or the Consumer routes him there) -> >getMarkup is called for all portlets. Portlets which were pre-initialized >can return markup faster > >OnUnload (similar arguments) >1. user logs out from Consumer portal. In this case this doesn't mean that >the portlet is destoyed! The CCP has still to be maintained, user might log >in again and still use the portlet. >So here we need to seperate destoy and OnUnload. >2. Consumer sends OnUnload event. This allows the portlets to release >resources like DB connections, etc. Note: these are not necessarirly stored >in the session? > >Mit freundlichen Gruessen / best regards, > > Richard Jacob >______________________________________________________ >IBM Lab Boeblingen, Germany >Dept.8288, WebSphere Portal Server Development >WSRP Team Lead & Technical Lead >WSRP Standardization >Phone: ++49 7031 16-3469 - Fax: ++49 7031 16-4888 >Email: mailto:richard.jacob@de.ibm.com > > > > Stefan Hepper > <sthepper@hursley > .ibm.com> To > WSRP Interfaces subgroup > 06/15/2005 05:46 <wsrp-interfaces@lists.oasis-open.o > PM rg> > cc > > Subject > [wsrp-interfaces] Why do we have an > onLoad event and no onUnload? > > > > > > > > > > >Hi all, >sorry to ask something that you may have discussed already some time ago >before I joined, but I don't quite understand what the use case for the >onLoad event is. > >If it is to initialize resources on behalf of the user why is there no >onUnload event to release these resources? > >Are these events useful at all given that the delivery is not guaranteed? > >Stefan > > > > >
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]