8/16/2012 - Thursday
By now, its been 3 weeks ~ maybe 100 hours ~ of dog headed attempt to get broad hand's-on perspective of Web/SOA/J2EE development. So far, I've tackled:
Apache Tomcat - for Front-end/View (HTML/JSP) and Controller (Servlets) web framework
J2EE Glassfish - for "Model" or business logic served up via Soap/HTML
Eclipse IDE - with all the plugins to make "stuff" for all of the above
Plus all the Operating System and Browser specific quirks of developing with each of above platforms
Now I was attacking Google Web Toolkit (as an alternate Front-end and Controller web-framework)
I actually started with GWT few days back on Sunday (8/12/2012). And I had triumphantly completed the first section of Google Web Toolkit Tutorial:
BUILDING A SAMPLE GWT APPLICATION
And I had moved on to the next section
COMMUNICATING WITH SERVER
And that's were I got stuck (again) .
I had completed the server side implementation and the client side updates to call the server method (using GWT RPC) to get the (fake) stock quotes being used to populate the browser view.
Everything worked EXCEPT for a bit of code functionality designed to mimic catching and reacting to a GWT RPC exception. Code was functional but the simulated exception throwing/catching was not happening. That Sunday, after working through the tutorials ALL WEEKEND, all I could manage was to review my code, staring and comparing with tutorial for easy mistakes...Nothing. At this point, I was too brain-dead to go into discipliened/debug mode.
Today, I finally picked it back up and I started setting breaking points to step through the code but bewilderingly, breakpoints where not working. The Eclipse IDE was just not doing anything with them. Google search on "eclipse breakpoints not working gwt" showed me that others had the problem but there was no clear RCA. I tried restarting browser, IDE, VM, server, application, and changed my eclipse Properties -> Java Build Path -> library -> JRE System Library from "java-6-openjdk-i386" (base Ubuntu JRE install) to "jdk1.6.0_33" (which I had recently downloaded from Sun to try to solve some other issue on some other project at some other point), also fixed my $JAVA_HOME environement variable to point to the "jdk1.6.0_33" folder (export JAVA_HOME = usr/lib/jdk1.6.0_33)....I'm not sure exactly which combination of advice/actions cured the problem but something worked. Actually it worked two fold...the breakpoint debugging functionality worked and so did the apparent bug with the non-functioning exception trying/catching....phew
No comments:
Post a Comment