Re: [udig-devel] double click problem

[Jesse Eichar <jesse.eichar@xxxxxxxxxxxxxx>]

For ENTER...  it is a key event.  I don't know if we have an api where you
can trigger events programatically.  You may need to add one.  But given
that a button can access the event manager of a viewport pane you should be
able to make the button issue a ENTER key event.

Jesse

On Thu, Oct 28, 2010 at 2:39 PM, andrea antonello <
andrea.antonello@xxxxxxxxx> wrote:

> > I was mostly referring to adding gestures.  That is not part of
> traditional
> > SWT.  It might be interesting at some point to look into it, I think
> there
> > is some mobile stuff out there but I don't know much about that.
>
> Oh, ok, I didn't understand correctly.
> In fact I ended up backing out the long-click option as well. It was
> conflicting with the long click that tablet pens use to customize
> actions.
> At the moment I am handling this like CAD software does. I am going to
> add the possibility to issue and ENTER event from the right click
> menu. That is nice small and doesn't hurt anyone for sure.
> Regarding this issue, you have any idea about how to issue and ENTER
> key event? I am trying the approach through AWT Robot, not sure if it
> will work though.
>
> Andrea
>
>
>
> > On Thu, Oct 28, 2010 at 10:54 AM, andrea antonello
> > <andrea.antonello@xxxxxxxxx> wrote:
> >>
> >> > You have to look into the SWT API.  I don't think it can be done
> without
> >> > hacking SWT directly for the tablet.  But maybe there is a tablet SWT
> >>
> >> Why you think that? What speak against trapping the time between press
> >> and release and in case issue a double click event?
> >> It is an easy way out and doesn't interfere with any event.
> >>
> >> I am testing it right now and it works well.
> >>
> >> > abstraction that can be used.  I am afraid you are in new territory
> >> > here.
> >>
> >> I guess so, but I would not want to add strange tablet pc thing while
> >> I am able to get on without it. There are too many option and you
> >> never know where it will go in near future. So it would be better to
> >> translate the system0translated-events to something usefull. In fact
> >> the system uses the long click on tablets to supply the right click.
> >> On top of it, I couldn't find anything about swt (but even swing) and
> >> tablet pen. I am guessing this all comes to applications as mouse
> >> events translated by the system.
> >>
> >> Andrea
> >>
> >>
> >> > Jesse
> >> >
> >> > On Thu, Oct 28, 2010 at 8:49 AM, andrea antonello
> >> > <andrea.antonello@xxxxxxxxx> wrote:
> >> >>
> >> >> Alright, since I didn't hear back from anyone on idea about how to
> >> >> bypass the engine, I am assuming it is not really possible (code is
> >> >> quite complex, not really documented and I am too afraid to break the
> >> >> use of the mouse :)).
> >> >>
> >> >> I decided therefore to implement a long click as Jody suggested. I
> >> >> made some test and it works well and it doesn't conflict with any
> >> >> other mouse event. It will require to add a preference for how long a
> >> >> long click should be.
> >> >> What will that mean? Whenever you hold the left button down for more
> >> >> than the defined time preference, it will issue a double-click.
> >> >> In fact it would be then possible to issue any command. Is this the
> >> >> moment we could start to thing about gestures? :) I love gestures.
> >> >>
> >> >> Comments welcome,
> >> >> Andrea
> >> >>
> >> >>
> >> >> On Tue, Oct 26, 2010 at 10:35 AM, andrea antonello
> >> >> <andrea.antonello@xxxxxxxxx> wrote:
> >> >> > We are not yet there. I hacked around in that classes for some time
> >> >> > trying to find a solution to the pen tap but wasn't able to.
> >> >> >
> >> >> > The fact that the tap is translated into a move event makes it
> >> >> > impossible for me to create the logic to trap the event.
> >> >> >
> >> >> > That is why I am tempted to try to bypass ti to see what happens.
> >> >> > Obviously I am not aware of what the consequences are and even on
> how
> >> >> > to best do it.
> >> >> >
> >> >> > Andrea
> >> >> >
> >> >> >
> >> >> >
> >> >> > On Tue, Oct 26, 2010 at 10:25 AM, Jesse Eichar
> >> >> > <jesse.eichar@xxxxxxxxxxxxxx> wrote:
> >> >> >> Sorry I misunderstood.  SO the classes you want are:
> >> >> >> EventHandler and EventJob
> >> >> >> Jesse
> >> >> >>
> >> >> >> On Tue, Oct 26, 2010 at 10:18 AM, andrea antonello
> >> >> >> <andrea.antonello@xxxxxxxxx> wrote:
> >> >> >>>
> >> >> >>> Hi Jesse,
> >> >> >>> that is not the problem I am having.
> >> >> >>> I remember us having the preference added.
> >> >> >>> The problem is the fact that tablet pens give different events
> due
> >> >> >>> to
> >> >> >>> the user not being precise enough in tapping on the screen.
> >> >> >>> I debugged through the engine, and what happens is mainly that
> >> >> >>> "move"
> >> >> >>> events come in the engine. SO it is not possible to handle the
> >> >> >>> timing
> >> >> >>> as we did throught the preferences time.
> >> >> >>>
> >> >> >>> Andrea
> >> >> >>>
> >> >> >>>
	for <udig-devel@xxxxxxxxxxxxxxxxxxxxx>;
	Sat, 06 Nov 2010 16:22:50 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma;
	h=domainkey-signature:mime-version:received:received:in-reply-to
	:references:date:message-id:subject:from:to:content-type;
	bh=fy1WFton/pA0isu6/QfaaKFzd3E0PqfD/ta+tzo7hJc=;
	b=JO43bpG7zz84Pqni3w8+mqGE2Mm7GmVxCHgFvrak7oKO9HkKpE76FRxGLJjAZGdZD/
	sg/1HlxOpxfDS9o6r64BMZV0dbqv8jRMTfS+jSfS+2v4tw8v48dznp6OyQJ5fMqOVY1b
	FefS4Ytx/nh8Ah9RRj31EiuUjkWvmuDIp4fXM=
DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma;
	h=mime-version:in-reply-to:references:date:message-id:subject:from:to
	:content-type;
	b=gzZ2NkGRTiyspmjA/LK+MmEWAPfL/hLQIW2QU9ftz88oHmvveDcLlKqesXP2niQ5Hv
	GWGbBPmm0G8eVA4kMO6sKND/+7gQHoOQPXpUCwhvmINVessgr4Ss4la2J9EYS8gvD6iS
	gqcwBziL4m5YKS/+g7nfoRTJa7WZ2hCkfZx0c=
MIME-Version: 1.0
Received: by 10.223.115.130 with SMTP id i2mr2330130faq.1.1289085769812; Sat,
	06 Nov 2010 16:22:49 -0700 (PDT)
Received: by 10.223.96.202 with HTTP; Sat, 6 Nov 2010 16:22:49 -0700 (PDT)
In-Reply-To: <AANLkTikSSqv1BTCjD31i--gHrVmBvB3SpkEs8cnV8c5r@xxxxxxxxxxxxxx>
References: <AANLkTikSSqv1BTCjD31i--gHrVmBvB3SpkEs8cnV8c5r@xxxxxxxxxxxxxx>
Date: Sun, 7 Nov 2010 09:22:49 +1000
Message-ID: <AANLkTimbgfC7-EjR=G5R7UP-DQ=N68ujFf41p9aCRUXu@xxxxxxxxxxxxxx>
From: Jody Garnett <jody.garnett@xxxxxxxxx>
To: User-friendly Desktop Internet GIS <udig-devel@xxxxxxxxxxxxxxxxxxxxx>
Content-Type: text/plain; charset=ISO-8859-1
Subject: Re: [udig-devel] SWTBot
X-BeenThere: udig-devel@xxxxxxxxxxxxxxxxxxxxx
X-Mailman-Version: 2.1.11
Precedence: list
Reply-To: User-friendly Desktop Internet GIS <udig-devel@xxxxxxxxxxxxxxxxxxxxx>
List-Id: User-friendly Desktop Internet GIS <udig-devel.lists.refractions.net>
List-Unsubscribe: <http://lists.refractions.net/mailman/options/udig-devel>,
	<mailto:udig-devel-request@xxxxxxxxxxxxxxxxxxxxx?subject=unsubscribe>
List-Archive: <http://lists.refractions.net/pipermail/udig-devel>
List-Post: <mailto:udig-devel@xxxxxxxxxxxxxxxxxxxxx>
List-Help: <mailto:udig-devel-request@xxxxxxxxxxxxxxxxxxxxx?subject=help>
List-Subscribe: <http://lists.refractions.net/mailman/listinfo/udig-devel>,
	<mailto:udig-devel-request@xxxxxxxxxxxxxxxxxxxxx?subject=subscribe>
X-List-Received-Date: Sat, 06 Nov 2010 23:20:55 -0000

That looks very cool! I like the fact that it produces ant tasks.

Right now in udig we have three levels of tests:

1) Some pure JUnit tests that can be run to test out java data
structures and so forth. These are very fast; and should probably be
run prior to committing.


2) Some RCP JUnit tests. These need the eclipse environment to start
up a copy of uDig; connect to the copy and run the tests, and then
shut it down. Because of the overhead involved in starting and
stopping a copy of uDig we usually run all the tests for a plugin in
one go (there is a right click option to do just that). These would be
great as "systems integration tests" to be run from a nightly build.

In truth these don't get run very much because it is a trouble for developers.

3) We have a manual testing process used before each release
(basically going through walkthrough 1 and walkthrough 2 to confirm
the application runs as documented).

If SWTBot could automate some of these activities it would be great; I
would think something along the lines of an ant script for each
section of walkthrough 1 and walkthrough 2 (that way if there is still
a few sections we need to do by hand we could do so).

I have been watching some of the ui testing tools and am happy to see
one emerge for SWT apps. All the good ones I have seen previously have
been windows specific.

Jody


On Wed, Nov 3, 2010 at 12:34 AM, Devrim Baris Acar
<devrimbaris@xxxxxxxxx> wrote:
> Hi,
>
> I have been testing out SWTBot[1] for a couple of days to test some of out
> udig related ui functionality and so far I like it.
>
> What I wonder is : what is the best/proper way of ui functionality testing
> with regard to RCP and udig?
> Any experiences?
>
> Best regards,
> [1] http://www.eclipse.org/swtbot/
>
>
> //baris
>
>
> _______________________________________________
> User-friendly Desktop Internet GIS (uDig)
> http://udig.refractions.net
> http://lists.refractions.net/mailman/listinfo/udig-devel
>
>