From jkohl at telusplanet.net  Thu Jan  1 14:51:06 2004
From: jkohl at telusplanet.net (Jonathan Kohl)
Date: Thu, 1 Jan 2004 12:51:06 -0700
Subject: [Wtr-general] Mozilla Control update
In-Reply-To: <5.1.0.14.2.20031230001003.034586c0@127.0.0.1>
Message-ID: <000001c3d0a0$989d5b60$6401a8c0@tintin>


> I have a day-long class where i teach people how to use this stuff. Why
> don't we propose to teach a half-day tutorial? I can share all my examples
> and notes and libraries from my class. In fact, i've booked January to
> merge this into rubyforge. This is all hands-on, where we tell students to
> bring a laptop and they learn how to do it on their own machines.
This sounds like a great idea. If you need any help, let me know and I'll do
what I can.

-Jonathan





From chrismo at clabs.org  Sat Jan  3 02:28:51 2004
From: chrismo at clabs.org (Chris Morris)
Date: Sat, 03 Jan 2004 01:28:51 -0600
Subject: [Wtr-general] iec update: 2004.xxx.x
Message-ID: <3FF66F33.5060401@clabs.org>

Been coding away on the IEC codebase in prep for Austin, while 
simultaneously trying to fulfill some of the need for tests on IEC 
itself. I just committed some new code -- here's the latest changelog 
stuff in progress, though it ain't all necessarily working like I want 
it to.

I'm trying to make sure all changes keep IEC backward compatible ... but 
we don't have the tests to prove it right at the moment, though I do use 
the lib 'in production' for my own personal use. Anyway, as I notify the 
lists of these commits, if any of you want to get the latest from cvs 
and try it out, please do.

====2004.xxx.x
* ClIEController.form now has optional param to automatically wrap the DOM
  form instance in an IEDomFormWrapper instance.
* ClIEController.initialize now takes an options hash to specify, well,
  options. Two options at this point: visible and auto_wrap_form. This
  method is still backwards compatible -- if you pass in a boolean then
  the visible variable is set and all others ignored.
* Big refactoring/improvements in accessing DOM nodes:
  * IEDomFormWrapper used to have more funky code for wrapping widgets, now
    it's a more flexible plug-in-ish, polymorphing structure, with a
    separate subclass to wrap every node type (although only some node types
    are supported yet -- see readme.txt). Supporting additional node types
    will be much easier to add to the code now.
  * There were some limitations because the previous code favored element
    nodes instead of all nodes on the form. That's been fixed so that text
    nodes are not discriminated against.
  * Special code nonsense for ASP.NET labels has been removed, though should
    still be supported.
  * A new method_missing on ClIEController itself allows very easy access
    to DOM nodes (in and out of forms) that have id/name attributes.
    [there is still a little screwiness here using this method to talk to
     some form inputs].
    
    If the page has a link with an id like so:
   
      <a href='http://clabs.org' id=clabs>cLabs.org</a>
     
    You can simply call:

      @iec.clabs

    ... to click the link.

    Buttons with names/ids can be accessed the same way, other node
    types, etc.

    You can even add your own method_missing to forward calls to your @iec
    instance so your test scripts can look something like this:

      home
      browse
      apples
      add; add;
      checkout

    (unfortunately, assigning values to text inputs, for example, doesn't
     work with this method_missing approach, because Ruby interprets this:
   
       first_name = 'chris'
      
     as a local variable assignment, and it's never passed to
     method_missing. So it has to again be:
    
       @iec.first_name = 'chris')
      
* WindowsFormWrapper code for use against .NET Windows Forms that have COM
  interop enabled has been removed. This code lives in a file called
  dotnet.form.rb that's back in the old cLabs scrapware cliecontroller CVS
  repository.

-- 
Chris
http://clabs.org/blogki
http://cministries.org/blogki
-=-=-=-=--=-=-=-=-=-=-
Free solo piano album (mp3)
http://cministries.org/cstudios/blackandwhite.htm




From jkohl at telusplanet.net  Sat Jan  3 14:04:25 2004
From: jkohl at telusplanet.net (Jonathan Kohl)
Date: Sat, 3 Jan 2004 12:04:25 -0700
Subject: [Wtr-general] packages for two controllers on RubyForge
In-Reply-To: <3FF18ADC.8080903@clabs.org>
Message-ID: <000001c3d22c$67728be0$6401a8c0@tintin>

I was just wondering about having a separate package for the controller Paul
wrote on the Rubyforge site. Chris' iec seems to be geared more for
developers while the one Paul has been working on is more for testers.

I've been using the controller Paul wrote, and have added a couple of
methods for my own use. Paul is getting close to having most of the
requirements I outlined on the WTR site for web testing. 
Here is the status so far:
* hyperlinks - can click links by the text name between the <a href=""> and
</a>
* images - can click images based on the name of the image (this needs some
work, but it does the job)
* buttons - done
* check boxes - done but needs a little polishing
* select items in:
	* drop-down boxes - done
	* radio lists - done
* enter text in text fields - done 
* handle popup warning or error message boxes - Paul is working on this
* handle pages that use frames - Paul has a solution; I haven't tried it yet

* read and handle inline messages - easy to handle with Ruby, but no
generic verification method 
* verify that objects, widgets or text exists - simple enough to do with
Ruby, but no generic method yet. It would be nice to do something like
SilkTest's "exists"


I'm using this controller to automate web application functionality in my
current role as a software tester. Paul and I were talking about using CVS
for his controller, and the Rubyforge site would work well. Currently, we've
been emailing or IM'ing code. It would be good to start using some change
control on his controller.

Does having two packages seem reasonable?

Thanks;

-Jonathan






From chrismo at clabs.org  Sat Jan  3 14:39:32 2004
From: chrismo at clabs.org (Chris Morris)
Date: Sat, 03 Jan 2004 13:39:32 -0600
Subject: [Wtr-general] packages for two controllers on RubyForge
In-Reply-To: <000001c3d22c$67728be0$6401a8c0@tintin>
References: <000001c3d22c$67728be0$6401a8c0@tintin>
Message-ID: <3FF71A74.90902@clabs.org>

Jonathan Kohl wrote:

>Does having two packages seem reasonable?
>  
>
In the long run, I don't think so.

Pragmatically, there may be a bit of work merging in two code bases, and 
having two while we work this out is no big deal, especially because the 
efforts here are voluntary and it may take a while to get around to 
putting it together.

But all of the functionality you listed sounds great. I don't see why we 
should be duplicating our efforts here, I'd love to have all of that in IEC.

Bret had some code he'd put together to help his prior presentations 
with Marick, and he emailed me the code yesterday. Some of the stuff 
looks great and I'd like to integrate it into the IEC codebase.

To be frank, it's frustrating to me to hear that you and Paul have been 
working on code on not sharing that code with the group so far. It's 
possible you could've saved me some time these past few days if you were 
already working in CVS committing your code. If no other communication 
happens in the Wiki or on the list here, I can at least see the code 
where I do an update. If we're going to work as a team, then let's work 
as a team.

As far as the argument that my code is for developers and your code is 
for testers -- well, we've already had this conversation. I don't think 
we'll be able to nail down what is most readable and easiest to use for 
everyone -- there will always be differences of opinion. But I think 
it's very early in the project and I'd love to see your code, because I 
believe it could be integrated into my codebase and still maintain 
ease-of-use and readability for the two of you. It's at least worth a shot.

(Apologies if this comes off too cranky, I was up late last night and 
have kids running around my ankles while I'm writing this... :-)

-- 
Chris
http://clabs.org/blogki




From jkohl at telusplanet.net  Sat Jan  3 15:51:04 2004
From: jkohl at telusplanet.net (Jonathan Kohl)
Date: Sat, 3 Jan 2004 13:51:04 -0700
Subject: [Wtr-general] packages for two controllers on RubyForge
In-Reply-To: <3FF71A74.90902@clabs.org>
Message-ID: <000001c3d23b$4dbef710$6401a8c0@tintin>

> >Does having two packages seem reasonable?
> In the long run, I don't think so.
> Pragmatically, there may be a bit of work merging in two code bases, and
> having two while we work this out is no big deal, especially because the
> efforts here are voluntary and it may take a while to get around to
> putting it together.
> But all of the functionality you listed sounds great. I don't see why we
> should be duplicating our efforts here, I'd love to have all of that in
> IEC.
I'm inclined to agree. It would be better to all work on the same code-base.
I was under the impression that we were developing two different controllers
at this point which could be due to my own confusion.

> To be frank, it's frustrating to me to hear that you and Paul have been
> working on code on not sharing that code with the group so far. It's
> possible you could've saved me some time these past few days if you were
> already working in CVS committing your code. If no other communication
> happens in the Wiki or on the list here, I can at least see the code
> where I do an update. If we're going to work as a team, then let's work
> as a team.
Well, that's what I'm trying to do here - get us all on the same page. I can
understand your frustration. I don't think we've achieved a common vision,
and I thought we were continuing with two different controllers at this
point. I apologize if this was due to my own misunderstanding, and I didn't
mean to cause you grief.

Paul has shared updates - he emailed out some code to help with frames, and
his controller from a few weeks ago is in the "contrib." branch.

> As far as the argument that my code is for developers and your code is
> for testers -- well, we've already had this conversation.
We had some difference of opinions a few weeks ago on syntax which I guess
were never resolved:
JK: So do we need to choose a style, agree and move forward at this point? 
JK: Any other thoughts with regard to syntax styles? How do we address 
JK: these concerns and move forward?
BP: If this remains an issue of concern, we should choose a time and IM on
BP: it. 
BP:I  don't think email is really working

> I don't think
> we'll be able to nail down what is most readable and easiest to use for
> everyone -- there will always be differences of opinion. But I think
> it's very early in the project and I'd love to see your code, because I
> believe it could be integrated into my codebase and still maintain
> ease-of-use and readability for the two of you. It's at least worth a
> shot.
I think it's worth a shot as well. 
 
> (Apologies if this comes off too cranky, I was up late last night and
> have kids running around my ankles while I'm writing this... :-)
No prob.







From chrismo at clabs.org  Sat Jan  3 16:51:05 2004
From: chrismo at clabs.org (Chris Morris)
Date: Sat, 03 Jan 2004 15:51:05 -0600
Subject: [Wtr-general] packages for two controllers on RubyForge
In-Reply-To: <000001c3d23b$4dbef710$6401a8c0@tintin>
References: <000001c3d23b$4dbef710$6401a8c0@tintin>
Message-ID: <3FF73949.6030208@clabs.org>


>I apologize if this was due to my own misunderstanding, and I didn't
>mean to cause you grief.
>  
>
Ah, no worries. I probably came off too gruff -- I've had a nap since I
posted (probably should have posted _after_ the nap :-)

> We had some difference of opinions a few weeks ago on syntax which I guess
>
>were never resolved:
>  
>
I think it'd be best to take it one actual piece of code at a time for
now. Some of this may simply be shared unfamiliarity and when we get
things side-by-side, it may go away. Plus, the agile peoples and my own
experience says no matter what direction we decide on beforehand will
probably be changed by actual usage in the future.

I may be able to take some of the Jonathan/Paul methods and integrate
them into the IEC backend, leaving the surface call the same, so any
client code you had would work unchanged, then mold it further down the
road if it needs it.

>I think it's worth a shot as well. 
>  
>
Do you want to email me anything, or get CVS access going (maybe Paul
already has it working? I think he's at least setup in the RubyForge
project)? ... that said, I may not look at it right away simply because
Austin in next week. I dunno, there could be stuff in there I could use
by that time.

-- 
Chris
http://clabs.org/blogki





From paul.rogers at shaw.ca  Sat Jan  3 17:17:42 2004
From: paul.rogers at shaw.ca (Paul Rogers)
Date: Sat, 03 Jan 2004 15:17:42 -0700
Subject: [Wtr-general] packages for two controllers on RubyForge
In-Reply-To: <3FF73949.6030208@clabs.org>
Message-ID: <000001c3d247$68951ea0$6400a8c0@NewDell>

I'll take another look at getting cvs going - I think that will help the
discussion rather than trying to send code as attachments.

Im very interested in getting to one code base - I really want to get
Chriss dom saver into my controller (or my controller into Chriss
stuff). I was under a fair amount of pressure to get some stuff going at
work, and using the familiar was the quickest route

My next plan is to get the Windows GUI test (
http://raa.ruby-lang.org/list.rhtml?name=win32-guitest)  sort of
integrated, so I can use it to click javascript boxes, file upload boxes
etc ( as these cant be accessed with IE methods)

I also need to look into why I cant access new windows with Ruby, In my
earlier tests it appeared to be due to a bug in the Ruby WinOLE library.

Have a good New Year!

Paul


-----Original Message-----
From: wtr-general-bounces at rubyforge.org
[mailto:wtr-general-bounces at rubyforge.org] On Behalf Of Chris Morris
Sent: 03 January 2004 14:51
To: wtr-general at rubyforge.org
Subject: Re: [Wtr-general] packages for two controllers on RubyForge



>I apologize if this was due to my own misunderstanding, and I didn't 
>mean to cause you grief.
>  
>
Ah, no worries. I probably came off too gruff -- I've had a nap since I
posted (probably should have posted _after_ the nap :-)

> We had some difference of opinions a few weeks ago on syntax which I 
> guess
>
>were never resolved:
>  
>
I think it'd be best to take it one actual piece of code at a time for
now. Some of this may simply be shared unfamiliarity and when we get
things side-by-side, it may go away. Plus, the agile peoples and my own
experience says no matter what direction we decide on beforehand will
probably be changed by actual usage in the future.

I may be able to take some of the Jonathan/Paul methods and integrate
them into the IEC backend, leaving the surface call the same, so any
client code you had would work unchanged, then mold it further down the
road if it needs it.

>I think it's worth a shot as well.
>  
>
Do you want to email me anything, or get CVS access going (maybe Paul
already has it working? I think he's at least setup in the RubyForge
project)? ... that said, I may not look at it right away simply because
Austin in next week. I dunno, there could be stuff in there I could use
by that time.

-- 
Chris
http://clabs.org/blogki



_______________________________________________
Wtr-general mailing list
Wtr-general at rubyforge.org
http://rubyforge.org/mailman/listinfo/wtr-general



From jkohl at telusplanet.net  Sat Jan  3 17:37:14 2004
From: jkohl at telusplanet.net (Jonathan Kohl)
Date: Sat, 3 Jan 2004 15:37:14 -0700
Subject: [Wtr-general] packages for two controllers on RubyForge
In-Reply-To: <3FF73949.6030208@clabs.org>
Message-ID: <000001c3d24a$22821870$6401a8c0@tintin>

Good to hear we all want to move together - I was a bit worried there for a
bit.

> I think it'd be best to take it one actual piece of code at a time for
> now. Some of this may simply be shared unfamiliarity and when we get
> things side-by-side, it may go away. Plus, the agile peoples and my own
> experience says no matter what direction we decide on beforehand will
> probably be changed by actual usage in the future.
This sounds reasonable, and yes, I anticipate things changing due to actual
usage. With Paul's version of the controller and a couple methods I've
added, I'm most of the way through automating the booking engine for this
site: http://c0dsp.westjet.com/internet/sky/skylights.cgi  It is quite
difficult in Step 2 since the data is dynamic. I'm already getting a lot of
ideas on how to handle things better. I'm stuck on the popup windows they
use for error messaging, and the SSL window right now. Once Paul has some
code in place to handle these issues, I should be able to automate the whole
booking process. 

 
> I may be able to take some of the Jonathan/Paul methods and integrate
> them into the IEC backend, leaving the surface call the same, so any
> client code you had would work unchanged, then mold it further down the
> road if it needs it.
It would be nice to all work together and contribute/enhance to the same
code-base.

> Do you want to email me anything, or get CVS access going (maybe Paul
> already has it working? I think he's at least setup in the RubyForge
> project)? ... that said, I may not look at it right away simply because
> Austin in next week. I dunno, there could be stuff in there I could use
> by that time.
I have Tortoise CVS on my Win2K box at home and at work, so I could upload
any small things I do into CVS. I think that would be best from my end, or I
could email either Chris or Paul with changes and you guys could CVS them.

I could email you what I have on the booking engine to show how I develop
tests as well.





From chrismo at clabs.org  Sat Jan  3 17:43:44 2004
From: chrismo at clabs.org (Chris Morris)
Date: Sat, 03 Jan 2004 16:43:44 -0600
Subject: [Wtr-general] packages for two controllers on RubyForge
In-Reply-To: <000001c3d247$68951ea0$6400a8c0@NewDell>
References: <000001c3d247$68951ea0$6400a8c0@NewDell>
Message-ID: <3FF745A0.2030204@clabs.org>

Paul Rogers wrote:

>My next plan is to get the Windows GUI test (
>http://raa.ruby-lang.org/list.rhtml?name=win32-guitest)  sort of
>integrated, so I can use it to click javascript boxes, file upload boxes
>etc ( as these cant be accessed with IE methods)
>  
>
Have you done anything with this yet? I'd been thinking along the same 
lines ever since you posted this to the list (I think it was you).

-- 
Chris
http://clabs.org/blogki




From paul.rogers at shaw.ca  Sat Jan  3 17:55:54 2004
From: paul.rogers at shaw.ca (Paul Rogers)
Date: Sat, 03 Jan 2004 15:55:54 -0700
Subject: [Wtr-general] packages for two controllers on RubyForge
In-Reply-To: <3FF745A0.2030204@clabs.org>
Message-ID: <000101c3d24c$be5477a0$6400a8c0@NewDell>

I used some very similar code to do just the things I needed - clicking
javascrip boxes and the file upload box. This at first glance appears to
be a far nicer library than I developed. It looks like a real good
starting point for the methods we would need. Im happy to take over this
bit of the development, as I said to Jonathan I would send him some code
to do this on his site.  Its -25 degrees c in Calgary this weekend so I
potentially hava a lot of time on my hands...

-----Original Message-----
From: wtr-general-bounces at rubyforge.org
[mailto:wtr-general-bounces at rubyforge.org] On Behalf Of Chris Morris
Sent: 03 January 2004 15:44
To: wtr-general at rubyforge.org
Subject: Re: [Wtr-general] packages for two controllers on RubyForge


Paul Rogers wrote:

>My next plan is to get the Windows GUI test (
>http://raa.ruby-lang.org/list.rhtml?name=win32-guitest)  sort of 
>integrated, so I can use it to click javascript boxes, file upload 
>boxes etc ( as these cant be accessed with IE methods)
>  
>
Have you done anything with this yet? I'd been thinking along the same 
lines ever since you posted this to the list (I think it was you).

-- 
Chris
http://clabs.org/blogki


_______________________________________________
Wtr-general mailing list
Wtr-general at rubyforge.org
http://rubyforge.org/mailman/listinfo/wtr-general



From paul.rogers at shaw.ca  Sat Jan  3 19:10:43 2004
From: paul.rogers at shaw.ca (Paul Rogers)
Date: Sat, 03 Jan 2004 17:10:43 -0700
Subject: [Wtr-general] Windows gui testing
In-Reply-To: <000601c3becf$31d8def0$6400a8c0@NewDell>
Message-ID: <000001c3d257$32514340$6400a8c0@NewDell>

I havent managed to make this work so far.

Looks like the default install is for cygwin. I'll send moonwolf (the
author) an email and see if there is a new version, or a readme.

Paul

-----Original Message-----
From: wtr-general-bounces at rubyforge.org
[mailto:wtr-general-bounces at rubyforge.org] On Behalf Of Paul Rogers
Sent: 09 December 2003 20:39
To: wtr-general at rubyforge.org
Subject: [Wtr-general] Windows gui testing


http://raa.ruby-lang.org/list.rhtml?name=win32-guitest

Havent had a chance to look through it much. But I'll let you all know
what I find out

Paul

_______________________________________________
Wtr-general mailing list
Wtr-general at rubyforge.org
http://rubyforge.org/mailman/listinfo/wtr-general



From paul.rogers at shaw.ca  Sat Jan  3 19:27:52 2004
From: paul.rogers at shaw.ca (Paul Rogers)
Date: Sat, 03 Jan 2004 17:27:52 -0700
Subject: [Wtr-general] Windows gui testing
In-Reply-To: <000001c3d257$32514340$6400a8c0@NewDell>
Message-ID: <000401c3d259$9738a4e0$6400a8c0@NewDell>

The version on the RAA is a compiled cygwin version for Ruby 1.8

I recompiled it for ruby 1.8 (cygwin) and it seems to work well. Chris -
are you able to compile this on Windows?

Ive tried the free version of the Visual c compiler suggested by
Nathaniel (
http://groups.google.com/groups?selm=001301c3ab2f%24f5470230%240500a8c0%
40abraham  ) from here: http://www.winprog.org/tutorial/msvc.html

But I cant get anything to work - I probably have something sent wrong.

Paul


-----Original Message-----
From: wtr-general-bounces at rubyforge.org
[mailto:wtr-general-bounces at rubyforge.org] On Behalf Of Paul Rogers
Sent: 03 January 2004 17:11
To: wtr-general at rubyforge.org
Subject: RE: [Wtr-general] Windows gui testing


I havent managed to make this work so far.

Looks like the default install is for cygwin. I'll send moonwolf (the
author) an email and see if there is a new version, or a readme.

Paul

-----Original Message-----
From: wtr-general-bounces at rubyforge.org
[mailto:wtr-general-bounces at rubyforge.org] On Behalf Of Paul Rogers
Sent: 09 December 2003 20:39
To: wtr-general at rubyforge.org
Subject: [Wtr-general] Windows gui testing


http://raa.ruby-lang.org/list.rhtml?name=win32-guitest

Havent had a chance to look through it much. But I'll let you all know
what I find out

Paul

_______________________________________________
Wtr-general mailing list
Wtr-general at rubyforge.org
http://rubyforge.org/mailman/listinfo/wtr-general

_______________________________________________
Wtr-general mailing list
Wtr-general at rubyforge.org
http://rubyforge.org/mailman/listinfo/wtr-general



From chrismo at clabs.org  Sun Jan  4 02:08:42 2004
From: chrismo at clabs.org (Chris Morris)
Date: Sun, 04 Jan 2004 01:08:42 -0600
Subject: [Wtr-general] packages for two controllers on RubyForge
In-Reply-To: <000901c3d24e$3b883490$6401a8c0@tintin>
References: <000901c3d24e$3b883490$6401a8c0@tintin>
Message-ID: <3FF7BBFA.5000100@clabs.org>

Jonathan Kohl wrote:

> I need to get an RSS reader one of these days to keep up with blogs I 
> read. Seems that is how most people access mine and it makes sense.

I use SharpReader -- it works great for me.

Google for a link.

-- 
Chris
http://clabs.org/blogki





From paul.rogers at shaw.ca  Sun Jan  4 17:43:30 2004
From: paul.rogers at shaw.ca (Paul Rogers)
Date: Sun, 04 Jan 2004 15:43:30 -0700
Subject: [Wtr-general] packages for two controllers on RubyForge
In-Reply-To: <3FF7BBFA.5000100@clabs.org>
Message-ID: <000001c3d314$2d6a3e50$6400a8c0@NewDell>

Ive used rubric
http://rubric.rubyforge.org/

-----Original Message-----
From: wtr-general-bounces at rubyforge.org
[mailto:wtr-general-bounces at rubyforge.org] On Behalf Of Chris Morris
Sent: 04 January 2004 00:09
To: wtr-general at rubyforge.org
Subject: Re: [Wtr-general] packages for two controllers on RubyForge


Jonathan Kohl wrote:

> I need to get an RSS reader one of these days to keep up with blogs I
> read. Seems that is how most people access mine and it makes sense.

I use SharpReader -- it works great for me.

Google for a link.

-- 
Chris
http://clabs.org/blogki



_______________________________________________
Wtr-general mailing list
Wtr-general at rubyforge.org
http://rubyforge.org/mailman/listinfo/wtr-general



From paul.rogers at shaw.ca  Sun Jan  4 19:13:45 2004
From: paul.rogers at shaw.ca (Paul Rogers)
Date: Sun, 04 Jan 2004 17:13:45 -0700
Subject: [Wtr-general] ieController with clickers for js boxes
Message-ID: <000101c3d320$c8db8630$6400a8c0@NewDell>

Here it is....

Still no cvs. I think my password is at work so I may be able to get it
going tomorrow

The zip contains a few things that arent relevant to this, they are
things I use at work, but are part of the overall test structure I use
at work.

To try it, unzip to a suitable location
Then run testIEController.rb

I was mistaken before when I said that the clicker needed to be in a
separate process. I know something did though - maybe the filename
clicker - which hasn't been tested.

Paul
-------------- next part --------------
A non-text attachment was scrubbed...
Name: controller_with_win_clicker.zip
Type: application/x-zip-compressed
Size: 12732 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/wtr-general/attachments/20040104/2c2af613/attachment.bin 

From chrismo at clabs.org  Sun Jan  4 22:16:08 2004
From: chrismo at clabs.org (Chris Morris)
Date: Sun, 04 Jan 2004 21:16:08 -0600
Subject: [Wtr-general] ieController with clickers for js boxes
In-Reply-To: <000101c3d320$c8db8630$6400a8c0@NewDell>
References: <000101c3d320$c8db8630$6400a8c0@NewDell>
Message-ID: <3FF8D6F8.9040404@clabs.org>

Paul Rogers wrote:

>Here it is....
>
>Still no cvs. I think my password is at work so I may be able to get it
>going tomorrow
>
>The zip contains a few things that arent relevant to this, they are
>things I use at work, but are part of the overall test structure I use
>at work.
>
>To try it, unzip to a suitable location
>Then run testIEController.rb
>  
>
Is this the latest of all the stuff Jonathan mentioned yesterday, or 
just the stuff related to dialog clicking? It looks more like the 
former, but I haven't looked at it in much detail yet.

-- 
Chris
http://clabs.org/blogki





From paul.rogers at shaw.ca  Mon Jan  5 10:30:44 2004
From: paul.rogers at shaw.ca (paul.rogers at shaw.ca)
Date: Mon, 05 Jan 2004 08:30:44 -0700
Subject: [Wtr-general] ieController with clickers for js boxes
Message-ID: <762e277620ba.7620ba762e27@shaw.ca>

I guess its both :-)


Its the javascript clicker, but I used a recent version of my controller.


----- Original Message -----
From: Chris Morris <chrismo at clabs.org>
Date: Sunday, January 4, 2004 8:16 pm
Subject: Re: [Wtr-general] ieController with clickers for js boxes

> Paul Rogers wrote:
> 
> >Here it is....
> >
> >Still no cvs. I think my password is at work so I may be able to 
> get it
> >going tomorrow
> >
> >The zip contains a few things that arent relevant to this, they are
> >things I use at work, but are part of the overall test structure 
> I use
> >at work.
> >
> >To try it, unzip to a suitable location
> >Then run testIEController.rb
> >  
> >
> Is this the latest of all the stuff Jonathan mentioned yesterday, 
> or 
> just the stuff related to dialog clicking? It looks more like the 
> former, but I haven't looked at it in much detail yet.
> 
> -- 
> Chris
> http://clabs.org/blogki
> 
> 
> 
> _______________________________________________
> Wtr-general mailing list
> Wtr-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/wtr-general
> 



From jkohl at telusplanet.net  Mon Jan  5 11:02:34 2004
From: jkohl at telusplanet.net (Jonathan Kohl)
Date: Mon, 5 Jan 2004 09:02:34 -0700
Subject: [Wtr-general] ieController with clickers for js boxes
In-Reply-To: <762e277620ba.7620ba762e27@shaw.ca>
Message-ID: <000801c3d3a5$5549f280$6401a8c0@tintin>

The demo looks good Paul. I'll have to try it out one day this week on the
booking engine.

Thanks;

-Jonathan

> -----Original Message-----
> From: wtr-general-bounces at rubyforge.org [mailto:wtr-general-
> bounces at rubyforge.org] On Behalf Of paul.rogers at shaw.ca
> Sent: January 5, 2004 8:31 AM
> To: chrismo at clabs.org; wtr-general at rubyforge.org
> Subject: Re: [Wtr-general] ieController with clickers for js boxes
> 
> I guess its both :-)
> 
> 
> Its the javascript clicker, but I used a recent version of my controller.
> 
> 
> ----- Original Message -----
> From: Chris Morris <chrismo at clabs.org>
> Date: Sunday, January 4, 2004 8:16 pm
> Subject: Re: [Wtr-general] ieController with clickers for js boxes
> 
> > Paul Rogers wrote:
> >
> > >Here it is....
> > >
> > >Still no cvs. I think my password is at work so I may be able to
> > get it
> > >going tomorrow
> > >
> > >The zip contains a few things that arent relevant to this, they are
> > >things I use at work, but are part of the overall test structure
> > I use
> > >at work.
> > >
> > >To try it, unzip to a suitable location
> > >Then run testIEController.rb
> > >
> > >
> > Is this the latest of all the stuff Jonathan mentioned yesterday,
> > or
> > just the stuff related to dialog clicking? It looks more like the
> > former, but I haven't looked at it in much detail yet.
> >
> > --
> > Chris
> > http://clabs.org/blogki
> >
> >
> >
> > _______________________________________________
> > Wtr-general mailing list
> > Wtr-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/wtr-general
> >
> 
> _______________________________________________
> Wtr-general mailing list
> Wtr-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/wtr-general




From chrismo at clabs.org  Tue Jan  6 02:07:05 2004
From: chrismo at clabs.org (Chris Morris)
Date: Tue, 06 Jan 2004 01:07:05 -0600
Subject: [Wtr-general] minor iec update
Message-ID: <3FFA5E99.4020409@clabs.org>

Commited some additional code stuffs tonight, change log at the end.

I'd hoped to have more node wrappers in place (esp. for form widgets), 
but I've decided to double-up in my Austin prep and have been piecing 
together a demo website with Borges (a Ruby web app framework that I've 
been wanting to check out for a long time), and have hit a minor snag 
with it ... but hey, it's a good example of a buggy website :-)

I hope to have an IEC release before Austin, as well as getting the 
sample app running on clabs.org so I can show y'all what I've been 
doing. I also plan to get the sample app and accompanying code into 
WTR's CVS, though it probably won't be part of the distribution at this 
point.

* Two additional options: :demo_mode and :demo_wait. Calls that fall through
  iec.method_missing use these options (currently, call that fall through
  iec.form.method_missing do not use these, but these two method_missing's
  need to be refactored and merged). If :demo_mode is true, then selected
  nodes will be highlighted (red text, red border) and will sleep for
  :demo_wait seconds. This is an eye-candy feature for being able to
  optionally visually follow a test script.
* Added fix for the RPC intermittent bug that would occur of an instance
  of ie/iec was created too soon after a previous one was killed.
* new submit method calls form.submit. Call to iec.form is with defaults, so
  there's no control for multi-form pages, but it's a convenience method
  anyway.
* new dump_dom method in iec, takes care of using ClIEDomViewer without
  having to touch it directly. Default output is $stdout, but method accepts
  an io instance.
* minor fix on the Select node wrapper.
* new node wrappers: Header, Italic, Input Text 

-- 
Chris
http://clabs.org/blogki




From chrismo at clabs.org  Tue Jan  6 15:55:18 2004
From: chrismo at clabs.org (Chris Morris)
Date: Tue, 06 Jan 2004 14:55:18 -0600
Subject: [Wtr-general] Mozilla Control update II
In-Reply-To: <3FEE2FDB.6010702@clabs.org>
References: <3FEE2FDB.6010702@clabs.org>
Message-ID: <3FFB20B6.9000209@clabs.org>

Chris Morris wrote:

> I've contacted the MozAxc author, but he's out of pocket until Jan 6th.

Better news (maybe) ... the author clued me into the .object property of 
the IE DOM OBJECT instance which would force me into the MozAxc's world. 
So I've now successfully done this:

  
iec.ie.document.MozAxcEmbeddedBrowser.object.document.write('<html><body>success!</body></html>')

... which is wonderful, we're now another layer deeper than before.

But ... I've been dinking around with trying to use document in the ways 
I'm used to with IE, and there's not nearly the same functionality 
inside, and now we're into the realm of the MozAxc author's stuff. 
Doesn't seem to be nearly the same level of completeness as IE's. He 
claims compatibility with W3C DOM Level 1 (I think -- his web page does, 
but a comment in his last email seemed maybe not even that), so we may 
still be able to do everything we need, just a lot more work for the IEC 
library to be able to seemlessly support both IE and Moz (embedded in IE).

-- 
Chris
http://clabs.org




From chrismo at clabs.org  Wed Jan  7 01:16:55 2004
From: chrismo at clabs.org (Chris Morris)
Date: Wed, 07 Jan 2004 00:16:55 -0600
Subject: [Wtr-general] un-public iec release - 2004.006.0
Message-ID: <3FFBA457.9010401@clabs.org>

I've added all my presentation stuffs to the wtr cvs repository tonight 
(contrib/motunes), and included it in a new IEC distribution. It 
includes a tweaked install of Borges (minus a ton of rdoc files) and has 
really inflated the .zip from 17-ish-k to 190k, though that's still 
overall pretty small.

I've uploaded the .zip to clabs, but not to RubyForge and I don't plan 
on announcing this one right away ... I wanted the .zip ready for me to 
try on a 'clean' machine at work to make sure everything I need for 
Austin is inside -- so until I'm comfortable that the .zip is complete, 
I won't announce this 'un. Zip can be found here: http://clabs.org/dl/iec.

I was hoping to get the sample app running on clabs.org itself, but my 
web server box doesn't have Ruby 1.8 (which Borges 1.0.0-alphaX 
requires) and I'm contacting my server guy to see about getting 1.8 
available. The new method_missing code which relies on id attributes in 
the nodes is pretty slick (IMNSHO), but relies on id attributes -- if 
you know of a site that reliably uses id attributes, you can experiment 
with this version of IEC against it. (If I get some extra free time - 
ha! - I want to update my clWiki to have id attributes, so I/you can 
demo the new stuff against the wtr wiki).

(BTW, Bret, I'll need Ruby 1.8 on your laptop, in case you're running an 
older version).

I won't post the change log since most of it I've posted as I committed 
to cvs, and it's in the doc/changelog.txt file. Only other change was an 
updated license.txt file, plus a Ruby script that auto-updates the top 
of each distributed .rb file with this boilerplate, plus a list of 
contributors (Brian, Bret, Jon, Paul) at the end of the license -- 
technically, I don't think any of your contributions are in there yet, 
but they should be soon, as I've gotten email contributions from y'all.

-- 
Chris
http://clabs.org/blogki




From chrismo at clabs.org  Wed Jan  7 09:37:27 2004
From: chrismo at clabs.org (Chris Morris)
Date: Wed, 07 Jan 2004 08:37:27 -0600
Subject: [Wtr-general] Mozilla Control update II
In-Reply-To: <3FFB20B6.9000209@clabs.org>
References: <3FEE2FDB.6010702@clabs.org> <3FFB20B6.9000209@clabs.org>
Message-ID: <3FFC19A7.20603@clabs.org>

Chris Morris wrote:

> iec.ie.document.MozAxcEmbeddedBrowser.object.document.write('<html><body>success!</body></html>') 
>

... btw, I've updated the scratch .rb file that experiments with this in 
cvs (wtr/cliec/contrib/mozaxc)

-- 
Chris
http://clabs.org




From bret at pettichord.com  Wed Jan  7 12:38:33 2004
From: bret at pettichord.com (Bret Pettichord)
Date: Wed, 07 Jan 2004 11:38:33 -0600
Subject: [Wtr-general] un-public iec release - 2004.006.0
In-Reply-To: <3FFBA457.9010401@clabs.org>
Message-ID: <5.1.0.14.2.20040107113717.07ea0200@127.0.0.1>

At 12:16 AM 1/7/2004, Chris Morris wrote:
>(BTW, Bret, I'll need Ruby 1.8 on your laptop, in case you're running an 
>older version).

I'm still using 1.6.8, but have 1.8 also installed. Which 1.8 version do 
you recommend?


_____________________________________
  Bret Pettichord, Software Tester
    Book - www.testinglessons.com
    Consulting - www.pettichord.com
    Blog - www.io.com/~wazmo/blog
    Hotlist - www.testinghotlist.com



From chrismo at clabs.org  Wed Jan  7 12:53:36 2004
From: chrismo at clabs.org (Chris Morris)
Date: Wed, 07 Jan 2004 11:53:36 -0600
Subject: [Wtr-general] un-public iec release - 2004.006.0
In-Reply-To: <3FFC46D9.3070609@clabs.org>
References: <5.1.0.14.2.20040107113717.07ea0200@127.0.0.1>
	<3FFC46D9.3070609@clabs.org>
Message-ID: <3FFC47A0.3000207@clabs.org>

Chris Morris wrote:

> Bret Pettichord wrote:
>
>> At 12:16 AM 1/7/2004, Chris Morris wrote:
>>
>>> (BTW, Bret, I'll need Ruby 1.8 on your laptop, in case you're 
>>> running an older version).
>>
>>
>>
>> I'm still using 1.6.8, but have 1.8 also installed. Which 1.8 version 
>> do you recommend?
>
>
> Just the 1.8-10 installer from the pragprog guys ... which is 1.8.0.

(IEC doesn't need it, but the demo app I built does because it uses 
Borges which has 1.8 specifics inside it)

-- 
Chris
http://clabs.org




From chrismo at clabs.org  Wed Jan  7 14:08:52 2004
From: chrismo at clabs.org (Chris Morris)
Date: Wed, 07 Jan 2004 13:08:52 -0600
Subject: [Wtr-general] un-public iec release - 2004.006.0
In-Reply-To: <3FFC47A0.3000207@clabs.org>
References: <5.1.0.14.2.20040107113717.07ea0200@127.0.0.1>	<3FFC46D9.3070609@clabs.org>
	<3FFC47A0.3000207@clabs.org>
Message-ID: <3FFC5944.8030600@clabs.org>

Chris Morris wrote:

> (IEC doesn't need it, but the demo app I built does because it uses 
> Borges which has 1.8 specifics inside it)

You can download the iec.2004.006.0.zip and see if you can get it all 
working if you want ... contrib/motunes has the demo app and all that in 
it. There's a readme that should tell you how to get it setup.

-- 
Chris
http://clabs.org




From paul.rogers at shaw.ca  Wed Jan  7 22:01:16 2004
From: paul.rogers at shaw.ca (Paul Rogers)
Date: Wed, 07 Jan 2004 20:01:16 -0700
Subject: [Wtr-general] Testing tool from Microsoft
In-Reply-To: <3FFC5944.8030600@clabs.org>
Message-ID: <000101c3d593$af2ee220$6400a8c0@NewDell>

Its not ruby....

I havent looked into it any detail, but may prove useful


http://msdn.microsoft.com/msdnmag/issues/02/06/bugslayer/toc.asp?frame=t
rue



From chrismo at clabs.org  Fri Jan  9 01:39:09 2004
From: chrismo at clabs.org (Chris Morris)
Date: Fri, 09 Jan 2004 00:39:09 -0600
Subject: [Wtr-general] iec.2004.009.0 release
Message-ID: <3FFE4C8D.2070200@clabs.org>

Mostly minor stuff, but one fix in regard to the new options hash that 
you'll need if you got 2004.006.0:

====2004.009.0
* Fixed visible and auto_form_wrap assignments from options hash. If the
  options were set to false in the hash, they still showed as true in IEC.
* Added a node.focus call inside the highlight method, to ensure the control
  is scrolled to and made visible. (applies to demo mode)
* Updated google example code to reflect newer IEC features.
* Wrapper for A nodes getting used for SPAN nodes because of a regex that 
  wasn't specific enough. Fixed.
* mozaxc scratch contrib files included in distro
* Updated MoTunes sample app and test/demo files with minor tweaks  

-- 

Chris
http://clabs.org/blogki




From jkohl at telusplanet.net  Tue Jan 13 11:17:51 2004
From: jkohl at telusplanet.net (jkohl at telusplanet.net)
Date: Tue, 13 Jan 2004 08:17:51 -0800
Subject: [Wtr-general] AWTA
Message-ID: <1074010671.40041a2f64820@webmail.telusplanet.net>

How did the AWTA go?

Thanks;

-Jonathan



From chrismo at clabs.org  Tue Jan 13 12:49:09 2004
From: chrismo at clabs.org (Chris Morris)
Date: Tue, 13 Jan 2004 11:49:09 -0600
Subject: [Wtr-general] AWTA
In-Reply-To: <1074010671.40041a2f64820@webmail.telusplanet.net>
References: <1074010671.40041a2f64820@webmail.telusplanet.net>
Message-ID: <40042F95.4030404@clabs.org>

jkohl at telusplanet.net wrote:

>How did the AWTA go?
>  
>
It went well. Bret and I did some on the fly pair presenting, which 
worked really well. I drove the code examples and Bret could frame it 
all into a tester POV and field a lot of their questions.

For some participants, it seemed the bulk of the value of IEC was the 
link into MSDN to the docs on the COM interfaces :-)

Bret made an interesting point that I'd heard him make before, but not 
this clearly, that one of the real values of IEC driving IE is its 
visible element helping build trust between testers and customers. The 
customers can trust the interactive element because they can see the 
results (though they are ultimately still trusting the testers 
implicitly because they may not actually sit down and watch every test 
case through ... :-) ... I'm sure Bret will jump in and clarify if I 
mussed this up.

Hopefully, the presentation can bring some people to try it out. 
Accommodating the different styles testers have when testing web stuffs 
is already an issue with just the four of us, and any success from here 
out will depend on this, since that's obviously the only value IEC 
provides. Having more people using it will help shape the different options.

--
Chris
http://clabs.org




From bret at pettichord.com  Tue Jan 13 12:53:38 2004
From: bret at pettichord.com (Bret Pettichord)
Date: Tue, 13 Jan 2004 11:53:38 -0600
Subject: [Wtr-general] Testing tool from Microsoft
In-Reply-To: <000101c3d593$af2ee220$6400a8c0@NewDell>
References: <3FFC5944.8030600@clabs.org>
Message-ID: <5.1.0.14.2.20040113115005.060fb408@127.0.0.1>

I read through this a while ago. I doubt it will prove useful. It's 
basically a case study in how not to create a test tool. The tool uses 
low-level event recording and playback and therefore has serious problems 
running tests reliably. This is, at least, the author's conclusions.

Better would be to take a look at these open-source perl libraries for 
driving Windows GUI's:

Win32-GuiTest, Ernesto Guisado
Strong support for various controls.
http://search.cpan.org/author/ERNGUI/Win32-GuiTest-1.3/

Win32-CtrlGUI, Toby Everett
Strong support for window identification.
http://search.cpan.org/author/TEVERETT/Win32-CtrlGUI-0.30/


At 09:01 PM 1/7/2004, Paul Rogers wrote:
>Its not ruby....
>
>I havent looked into it any detail, but may prove useful
>
>
>http://msdn.microsoft.com/msdnmag/issues/02/06/bugslayer/toc.asp?frame=true
>
>_______________________________________________
>Wtr-general mailing list
>Wtr-general at rubyforge.org
>http://rubyforge.org/mailman/listinfo/wtr-general

_____________________________________
  Bret Pettichord, Software Tester
    Book - www.testinglessons.com
    Consulting - www.pettichord.com
    Blog - www.io.com/~wazmo/blog
    Hotlist - www.testinghotlist.com



From jkohl at telusplanet.net  Tue Jan 13 19:59:35 2004
From: jkohl at telusplanet.net (Jonathan Kohl)
Date: Tue, 13 Jan 2004 17:59:35 -0700
Subject: [Wtr-general] AWTA
In-Reply-To: <40042F95.4030404@clabs.org>
Message-ID: <000a01c3da39$ada5ef70$6401a8c0@tintin>


> It went well. Bret and I did some on the fly pair presenting, which
> worked really well. I drove the code examples and Bret could frame it
> all into a tester POV and field a lot of their questions.
That sounds great!

> Bret made an interesting point that I'd heard him make before, but not
> this clearly, that one of the real values of IEC driving IE is its
> visible element helping build trust between testers and customers. The
> customers can trust the interactive element because they can see the
> results (though they are ultimately still trusting the testers
> implicitly because they may not actually sit down and watch every test
> case through ... :-) ... I'm sure Bret will jump in and clarify if I
> mussed this up.
I tried this on the last project I was on. At the end of each iteration, I
would demo the product to the business testers who did the acceptance tests.
I used the IE Controller to drive the app for the demo, and I would automate
some simple acceptance tests as smoke tests. The project was quite small and
short, so I don't have a lot to go on as for the effectiveness. I'm going to
try it on the project I'm now coordinating the testing efforts on.

> Hopefully, the presentation can bring some people to try it out.
> Accommodating the different styles testers have when testing web stuffs
> is already an issue with just the four of us, and any success from here
> out will depend on this, since that's obviously the only value IEC
> provides. Having more people using it will help shape the different
> options.
I agree. If we can get a good user base, things should really start to get
polished so that they are usable to many. 





From bret at pettichord.com  Wed Jan 14 13:30:24 2004
From: bret at pettichord.com (Bret Pettichord)
Date: Wed, 14 Jan 2004 12:30:24 -0600
Subject: [Wtr-general] AWTA
In-Reply-To: <40042F95.4030404@clabs.org>
References: <1074010671.40041a2f64820@webmail.telusplanet.net>
	<1074010671.40041a2f64820@webmail.telusplanet.net>
Message-ID: <5.1.0.14.2.20040114122419.061bc860@127.0.0.1>

Commercial GUI tools also are easy to trust because you can watch them run. 
But you always have to run them in this mode, which many testers complain 
about it. They can't do other work on a machine that is running tests.

But driving IE through the DOM gives us several other options:
1. We can keep IE visible as tests are running, but put other windows over it.
2. We can make it invisible while running tests.
3. And the latest version has a "debug" mode that actually highlights the 
controls that are being operated or verified.

I don't know if we'll be able to keep these modes all working as we move to 
adding support for popups and such. But i would like to try, because these 
are areas where i think the tool is substantially better than the 
commercial alternatives, and highlights that we aren't just making a baby tool.


_____________________________________
  Bret Pettichord, Software Tester
    Book - www.testinglessons.com
    Consulting - www.pettichord.com
    Blog - www.io.com/~wazmo/blog
    Hotlist - www.testinghotlist.com



From chrismo at clabs.org  Wed Jan 14 13:58:06 2004
From: chrismo at clabs.org (Chris Morris)
Date: Wed, 14 Jan 2004 12:58:06 -0600
Subject: [Wtr-general] AWTA
In-Reply-To: <5.1.0.14.2.20040114122419.061bc860@127.0.0.1>
References: <1074010671.40041a2f64820@webmail.telusplanet.net>	<1074010671.40041a2f64820@webmail.telusplanet.net>
	<5.1.0.14.2.20040114122419.061bc860@127.0.0.1>
Message-ID: <4005913E.9010103@clabs.org>

Bret Pettichord wrote:

> Commercial GUI tools also are easy to trust because you can watch them 
> run. But you always have to run them in this mode, which many testers 
> complain about it. They can't do other work on a machine that is 
> running tests.
>
> But driving IE through the DOM gives us several other options:
> 1. We can keep IE visible as tests are running, but put other windows 
> over it.
> 2. We can make it invisible while running tests.
> 3. And the latest version has a "debug" mode that actually highlights 
> the controls that are being operated or verified.
>
> I don't know if we'll be able to keep these modes all working as we 
> move to adding support for popups and such. But i would like to try, 
> because these are areas where i think the tool is substantially better 
> than the commercial alternatives, and highlights that we aren't just 
> making a baby tool.

Ah - excellent summary - thanks, Bret. I've added this here: 
http://www.clabs.org/wtr/index.cgi?page=/WhatDistinguishesUs

-- 
Chris
http://clabs.org




From paul.rogers at shaw.ca  Wed Jan 14 14:52:14 2004
From: paul.rogers at shaw.ca (paul.rogers at shaw.ca)
Date: Wed, 14 Jan 2004 12:52:14 -0700
Subject: [Wtr-general] AWTA
Message-ID: <3b0fb73b4122.3b41223b0fb7@shaw.ca>

Ive already found that we cant use the machine for 2 things ny longer. I havent ingestigated but I think its probably due to the fact that I fire the onfocus method whenever going to new controls - so that any javascript gets executed.

I used to run my version of the controller on a 100Mhz pentium pc with 64M ram, running win2k. So the hardware is cheap.....

----- Original Message -----
From: Bret Pettichord <bret at pettichord.com>
Date: Wednesday, January 14, 2004 11:30 am
Subject: Re: [Wtr-general] AWTA

> Commercial GUI tools also are easy to trust because you can watch 
> them run. 
> But you always have to run them in this mode, which many testers 
> complain 
> about it. They can't do other work on a machine that is running tests.
> 
> But driving IE through the DOM gives us several other options:
> 1. We can keep IE visible as tests are running, but put other 
> windows over it.
> 2. We can make it invisible while running tests.
> 3. And the latest version has a "debug" mode that actually 
> highlights the 
> controls that are being operated or verified.
> 
> I don't know if we'll be able to keep these modes all working as 
> we move to 
> adding support for popups and such. But i would like to try, 
> because these 
> are areas where i think the tool is substantially better than the 
> commercial alternatives, and highlights that we aren't just making 
> a baby tool.
> 
> 
> _____________________________________
>  Bret Pettichord, Software Tester
>    Book - www.testinglessons.com
>    Consulting - www.pettichord.com
>    Blog - www.io.com/~wazmo/blog
>    Hotlist - www.testinghotlist.com
> 
> _______________________________________________
> Wtr-general mailing list
> Wtr-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/wtr-general
> 



From bret at pettichord.com  Wed Jan 14 18:36:59 2004
From: bret at pettichord.com (Bret Pettichord)
Date: Wed, 14 Jan 2004 17:36:59 -0600
Subject: [Wtr-general] Fwd: Capturing irb input
Message-ID: <5.1.0.14.2.20040114173554.0621a788@127.0.0.1>

I like to teach students to experiment with IRB and then save the result as 
a script. This apparently does this automatically. Haven't tried it yet.

Bret

>To: bret at pettichord.com
>Cc: Rob Sartin <sartin at omniscientturtle.com>
>Subject: Capturing irb input
>From: Rob Sartin <sartin at navanax.org>
>Date: Sat, 10 Jan 2004 15:11:38 -0600
>
>http://blade.nagaokaut.ac.jp/cgi-bin/scat.rb/ruby/ruby-talk/59054
>
>said:
>
>If you just want input, put this in your .irbrc, assuming your irb is
>built with readline support:
>
>
>    def dump_history(file=nil)
>      if file
>        File.open(file, "w") do |f|
>          f.puts IRB::ReadlineInputMethod::HISTORY.to_a
>        end
>      else
>        puts IRB::ReadlineInputMethod::HISTORY.to_a
>      end
>    end
>
>
>Regards,
>
>Rob

_____________________________________
  Bret Pettichord, Software Tester
    Book - www.testinglessons.com
    Consulting - www.pettichord.com
    Blog - www.io.com/~wazmo/blog
    Hotlist - www.testinghotlist.com



From bret at pettichord.com  Wed Jan 14 21:43:29 2004
From: bret at pettichord.com (Bret Pettichord)
Date: Wed, 14 Jan 2004 20:43:29 -0600
Subject: [Wtr-general] AWTA
In-Reply-To: <3b0fb73b4122.3b41223b0fb7@shaw.ca>
Message-ID: <5.1.0.14.2.20040114204241.061d1108@127.0.0.1>

This was discussed at the workshop as a potential limitation of our method. 
Now i hear you've already addressed it.

At 01:52 PM 1/14/2004, you wrote:
>Ive already found that we cant use the machine for 2 things ny longer. I 
>havent ingestigated but I think its probably due to the fact that I fire 
>the onfocus method whenever going to new controls - so that any javascript 
>gets executed.

_____________________________________
  Bret Pettichord, Software Tester
    Book - www.testinglessons.com
    Consulting - www.pettichord.com
    Blog - www.io.com/~wazmo/blog
    Hotlist - www.testinghotlist.com



From paul.rogers at shaw.ca  Wed Jan 14 21:53:02 2004
From: paul.rogers at shaw.ca (Paul Rogers)
Date: Wed, 14 Jan 2004 19:53:02 -0700
Subject: [Wtr-general] AWTA
In-Reply-To: <5.1.0.14.2.20040114204241.061d1108@127.0.0.1>
Message-ID: <000201c3db12$b13460c0$6400a8c0@NewDell>

Yep. When I goto for example a button, I use the fireEvent method of IE
to fire events that may have javascript attached - fireEvent("onFocus")
fireEvent("onClick")

I found this especially important for text fields-where I may have to do

fireEvent("onFocus")
fireEvent("onClick")
# mimic a keypress
fireEvent("onKeyPress")   # maybe need onkeydown and onkeyup depending
upon the javascript
fireEvent("onChange")

On the app I am currently working on, there is a javascript function
that uses the onKeypress event.

Firing all these event sdoes significantly slow down the test. But I
also add each character of a text field one, by one, specifically so I
am able to use the onKeypress and onChange to better mimic a user





-----Original Message-----
From: wtr-general-bounces at rubyforge.org
[mailto:wtr-general-bounces at rubyforge.org] On Behalf Of Bret Pettichord
Sent: 14 January 2004 19:43
To: wtr-general at rubyforge.org
Subject: Re: [Wtr-general] AWTA


This was discussed at the workshop as a potential limitation of our
method. 
Now i hear you've already addressed it.

At 01:52 PM 1/14/2004, you wrote:
>Ive already found that we cant use the machine for 2 things ny longer. 
>I
>havent ingestigated but I think its probably due to the fact that I
fire 
>the onfocus method whenever going to new controls - so that any
javascript 
>gets executed.

_____________________________________
  Bret Pettichord, Software Tester
    Book - www.testinglessons.com
    Consulting - www.pettichord.com
    Blog - www.io.com/~wazmo/blog
    Hotlist - www.testinghotlist.com

_______________________________________________
Wtr-general mailing list
Wtr-general at rubyforge.org
http://rubyforge.org/mailman/listinfo/wtr-general