>Hello all:
>
>I'm in need of information and advice again.
>
>AUTOMATED TESTING of VFP applications.
>
>In earlier incarnations of FoxPro (read DOS here), we used a product called Vermont Test. It was excellent because it recorded keystrokes and mouse usage. And it recorded screens for later comparison and reporting. We loved it! We'd record our testing
>scripts, indicate which screens to record, and whenever we made changes, we'd run the automated testing scripts. "Look Ma! No hands." Welcome to Regression Testing.
>
>Now for Windows it is another story altogether. They've a product Vermont HighTest, but is does poorly with VFP.
>
>I may not have understood their explanation perfectly, but it seems VFP does not use true Windows objects. It uses bitmaps which the testing software cannot deal with effectively, other than to record the x,y coordinates on the screen.
>
>Question 1: Does VFP really use bitmaps in lieu of true Windows objects? If so, why? These are both Microsoft products.

The original FoxPro for Windows was designed for high compatability with the DOS version. Using bitmaps made this easier. People often request that the Fox team at MS use native Windows controls. This would slow down the product and we'd lose many of the PEMs that we now have. I'll take performance over automated testing anyday.

>
>Question 2: Is there a good automated testing product which will record screens (at tester's selected places in the program) and generate reports?

I'm not aware of any, but that doesn't mean they don't exist.

>
>I am aware of the Coverage Profiler and find it useful, but it doesn't address question 2 very well.
>
>TIA for any help!
>
>David Powell
>WagnerSys@aol.com