1677Re: silktest vs using win32::GuiTest
- Feb 20, 2007Let me start off by saying that I have not had any formal training in
testing. I have no documentation on any kinds of testing methodology.
Anything I do have was found on the net somewhere. I started my
career off as a programmer, and my interests lay mostly in that area.
Although I do get a kick out of tearing apart someone else's program,
so I've grown to like testing. But basically, I looked at the task
that needed to be done from a programmer's point of view.
Many of us have created our own framework for the application we are
testing. I wouldn't say mine is any better then anyone else's because
I haven't seen anyone else's, but I am proud of it and where it's
going. I didn't even realize what I was making until the basic code
was written. But here are the things I went through.
I was asked to write regression testing scripts, to make sure that the
steps we've laid out in the manual will work. The basic idea I had
was to make emulate a normal user as closely as possible. That means
instead of using code to push a button, I actually move the mouse
cursor there and click. So the first function I wrote was a Move
Mouse command. Since the one in GuiTest instantly moves the cursor to
where you say, I made a loop that moved it incrementally to provide a
delay and to make it look a little more normal. From there, I moved
on to filling fields. I decided I didn't want to blindly click with
mouse coordinates, so I made a function that looked for a field based
on it's label. Once I find the field, I grab it's screen coordinates
and feed that into the Move Mouse function. This way, they can move
the field where ever they want to and I don't have to change my code.
I think after that I starting coding functions that set a combobox to
a specific value, grabbing it's location in the drop-down and moving
the mouse to click on it. After that, things really started getting
out of hand. I've added code to wait for a window based on it's title
or it's contents, added logging features, grab screenshots, added code
to manipulate objects through SendMessage, added XML files to hold
data that drives the scripts, added database code to interact with the
data our software uses, created a system to run multiple scripts, and
probably a few other things I can't think of right now. Next is a
blackbox testing system, where the scripts intentionally try to make
error messages show. Without a doubt, it is a lot of work to get a
system up and running. But the results are worth it in my eyes. I
don't think I would have gotten anywhere near the results I've gotten
if I went with the top commercial choice I had.
But again, this is just my system. It's pretty geared towards the
software I'm testing, although I've tried to make it as generic as
possible. I'm sure others have systems that do things I haven't even
thought of, or provide better ways of doing things. Anyways, this is
the process I've gone through. It's taken me about a year and a half
script about a third of our application, and it'll probably take me
another year or so to script the rest. Well, I hope this helps.
--- In email@example.com, gs rao <gsrao_ggt@...> wrote:
> Hi Brent,
> I am new to the win32::GuiTest, I want to do the functionality
testing of an application using win32::GuiTest. Can you please tell me
what are the basic steps, we need to take while designing the
automation test frame work for that application using win32::GuiTest ?
> Is there any kind of frame work for the win32::GuiTest ?
> Or can you please guide me how to start the process of automating
the functionality testing for an application using win32::GuiTest ?
> if you have any specific documents related to the functionality
testing using win32::GuiTest, please send it to me ?
> Thanks for your help
> G.Subba Rao
- << Previous post in topic Next post in topic >>