June was a pretty dead month for JSFUnit development. I was out of town
for three of the last four weeks. Most of the development time I had
available was spent on JBoss AS.
Now that June is almost over, it's time to ramp back up and get a beta
release out the door. If JBoss World really happens in September then I
think that will be a great time to kick it off. Otherwise, I still see
no reason that we shouldn't be able to release a beta in that time
frame, if not sooner.
There are three milestones that I think we need to meet before a beta:
1) Ant and Maven integration: The biggest hurdle to adoption of JSFUnit
will be the ease with which it integrates into a development
environment. We must have something equal to or better than "cactify"
Matt, are you still on board to help with this feature?
2) Build a comprehensive demo: We need to prove that JSFUnit can
effectively test a robust JSF application including AJAX components and
Seam. We have a couple of Red Hat interns who have signed up for this
project. Welcome Jonathan Prindiville and Clive Saldanha to the JSFUnit
3) General feedback: In reaction to some feedback in the forum, I just
spent the last two days doing a huge refactoring of the facade API that
makes it much more robust and easier to use (See Jira JSFUNIT-5). But
I'm sure there are other obvious improvements to be identified.
Please, if you haven't tried out JSFUnit for yourself yet, SPEND AN HOUR
AND GIVE US SOME FEEDBACK.
Lastly, I'm really excited about the potential of this project. JSF is
becoming a huge part of the Java market and JSFUnit stands to become the
de facto choice for testing JSF applications.
Thanks to all of you for your help and continued support.
Jonathan Prindiville wrote:
> Stan Silvert wrote:
>> Just thought I should check in and see how things are going.
> Not too bad. As a way to get familiar with some of the components that
> you wanted to go into the demo, I've been trying to cobble a simple IRC
> server-side Java (I picked up an IRC lib called Martyr ), and Seam
> > Is there anything you need from me?
> Any ideas as to the size/complexity of this demo? Timeframe?
I'm not looking for complexity, I'm looking for coverage. For
RichFaces, AJAX4JSF, and Seam, I think we should start with some app
that covers the most widely-used features of each. As we go along, we
use JSFUnit to do test-driven development of the app.
After you get familiar with everything, I'll leave it up to you to
determine what the widely-used features are. Guessing works. But
asking the community or the developers might be better. It just depends
on how much time you want to spend on that phase.
BTW, I doubt that Seam Remoting falls into the "widely used" category,
but I could be wrong about that.
So I think that should narrow it down for these three technologies.
I'll also leave it to you to determine how much time it will take. This
is the kind of project that could last a few months to a few years
depending on how much you want to throw in. :-)
> Also: someone mentioned that you don't do IRC and that I should try to
> goad you into it. This is me goading. :P
My problem with IRC is that it creates pockets of unshared knowledge.
Discussions should be as open as possible, which is why I prefer email
lists. With email lists, subscribers never miss a message and the
audience tends to be broader, which leads to more informed decisions.
The only advantage I see to IRC is that it is a faster medium. But
email is almost always "fast enough".
I could be convinced otherwise, but that's my current thinking.
P.S. If you really want/need a quick, private conversation with me, I'm
on AIM and Skype.
AIM: stansilver (note the missing "t")
Jonathan Prindiville wrote:
> Hi there,
> I'm not too sure whether you've been made aware of this or not yet, but
> it seems as though Clive (copied on this email) and I will be working
> with you on JSFUnit. We've just begun our 16-month internships -- about
> a month ago -- at the Toronto office.
> At the moment we don't have very much direction from any of the other
> JBoss folks around the office because of the frenzy around this upcoming
> release, so we've just been exploring a bit on our own (building AS,
> deployed the hellojsf-jsfunit demo, reading up on JBoss in general). Is
> there anything that we should be doing at the moment?
Yes, I knew about it. I'm very glad to have the help.
For now, you guys need to be learning JSF and its supporting
technologies such as servlets and JSP's. You may prefer to use Facelets
instead of JSP, so you might want to look at that as well.
* See http://labs.jboss.com/jsfunit/
* Join the JSFUnit public mailing list:
* Get a Jira ID for http://jira.jboss.com
* Find out what you need to do to get comitter rights on the JSFUnit project
* Learn to build JSFUnit and run the demo from the command line
Lastly, I've created a Jira task for your project:
Right now it is assigned to me, but once you are added as a comitter we
can assign the task to one of you. I'll be on vacation all next week,
but feel free to discuss the project on the mailing list or the JSFUnit
forum. Then I can see everything when I get back.
Is that enough to keep you busy until then? :-)
Thanks again for the help,
it's one of the reasons why I am not in favor of HTML Unit
"It doesn't find the form element because HtmlUnit (the underlying
browser implementation for Canoo) doesn't understand/render Dojo's
JBoss Portal Project Lead
Red Hat Inc.
Red Hat France SARL, 171 Avenue Georges Clemenceau, 92024 Nanterre
Siret n° 421 199 464 00056