Hi Tim, The Servers view displays what has been added to the server but not necessarily what is on the server. When a module/project is added to the Server it 'may' or 'may not' be synchronized. By synchronized we mean that the module/project or files within it, might not yet been installed/published to the server. When the resources of a module/project in the workspace are not synchronized with the resources on the server, the status of the server and module will change to republish and the module that is not synchronized will/should display with italics, but apparently that didn't work all the time as you can see from Comment #2 on bug# 267993.
Another behavior that comes to mind that is similar to the Servers view is the behaviour of the workspace and file system. In this scenario, when a user modifies a file that exists in the workspace but is modified outside of the IDE, the IDE will not pick up the changes until you do a refresh or try to edit the file. In the case of the servers view we use the status label to show that resources are out of sync. In the scope of your question, there is two UI aids that aim to assist the user understand the status of the Server: the server status label and italics fonts on modules. - The server status flag, is a label that aims help users understand that a module/project has changes in the workspace that need to be published to the server. The server status flag in WTP 3.0 and earlier versions will display in a separate column, but in 3.1 and older the status flag displays a label decorator beside the server or module. The status of the flag is republish and synchronized, but in the case of an error (if I recall correctly) adopters can introduce a one lined message. I will need to look into the code to make sure that adopters can put their own one lined error messages, but if I am mistaken on this, it sounds like a neat enhancement for 3.2 :) - The italics fonts on a module, aim to emphasizes the mismatch of the module/project between the Server and the workspace resources. If you have other ideas on how we can improve the scenario, we are open to discussion. We could setup a meeting for later this week and hold an open discussion. Best Regards, -- -- Mr. Angel Vera Server Tools Developer for WTP and Rational Lotus Notes: Angel Vera/Toronto/i...@ibmca Tel: 905-413-5919 - E-Mail: [email protected] ----- Those who think that something is impossible, should not interfere with those who are willing to do it. - Angel Vera From: Tim McConnell <[email protected]> To: "General discussion of project-wide or architectural issues." <[email protected]> Date: 07/24/2009 10:06 PM Subject: Re: [wtp-dev] Another WTP Question Sent by: [email protected] Thanks for the information Rochelle. I see that the bugzilla issue is "Resolved" with "WontFix", which is a little disconcerting to me since it has and will continue to cause many usability issues as deployment exceptions happen all the time. I wonder if anyone from the WTP development team might offer a work-around to mitigate this confusion ?? Thanks. [email protected] wrote: > This is similar to a question I had as well - see issue > https://bugs.eclipse.org/bugs/show_bug.cgi?id=267993 > > Rochelle > > Tim McConnell wrote: >> Hi, I've noticed that whenever I deploy a single module or multiple >> modules from the WTP "Add and Remove" wizard and I get a deployment >> exception from the server, it still shows up in the Server view as if >> it deployed successfully, which is very confusing to our end-users. I >> assume we're doing something incorrect in our BehaviorDelegate but >> it's not obvious to me. When I debug this code in Server.publishImpl() >> below, I see that our BehaviorDelegate CoreException is caught. I also >> see our module contained within modules2 below so I expect that it'll >> get deleted (or at least not added) but instead it gets saved in the >> .metadata\.plugins\org.eclipse.wst.server.core\publish0.dat file even >> though it doesn't really exist in the server. Can someone help explain >> what else I need to do in our BehaviorDelegate when throwing our >> CoreException ?? Or is this a bug or expected behavior ?? Thanks >> much..... >> >> try { >> getBehaviourDelegate(monitor).publish(kind, modules4, >> monitor, info); >> } catch (CoreException ce) { >> Trace.trace(Trace.WARNING, "Error during publishing", >> ce); >> status = ce.getStatus(); >> } >> >> final List<IModule[]> modules2 = new ArrayList<IModule[]>(); >> visit(new IModuleVisitor() { >> public boolean visit(IModule[] module) { >> if (getModulePublishState(module) == >> IServer.PUBLISH_STATE_NONE) >> getServerPublishInfo().fill(module); >> >> modules2.add(module); >> return true; >> } >> }, monitor); >> >> >> getServerPublishInfo().removeDeletedModulePublishInfo(Server.this, >> modules2); >> getServerPublishInfo().clearCache(); >> getServerPublishInfo().save(); >> >> >> >> > _______________________________________________ > wtp-dev mailing list > [email protected] > https://dev.eclipse.org/mailman/listinfo/wtp-dev > -- Thanks, Tim McConnell _______________________________________________ wtp-dev mailing list [email protected] https://dev.eclipse.org/mailman/listinfo/wtp-dev
<<inline: graycol.gif>>
<<inline: ecblank.gif>>
_______________________________________________ wtp-dev mailing list [email protected] https://dev.eclipse.org/mailman/listinfo/wtp-dev
