[qmtest] passing data from one test to another

[Scott Lowrey]

Mark Mitchell wrote:

> J C wrote:
>
>> Are there any mechanisms in QMTest to pass data
>> between 2 different tests during run-time?
>>
> We don't make that easy because we don't think it is good test design.
>

Have to disagree somewhat.  After considerable experience with QMTest, 
we've run into some brain twisters that don't seem to lend themselves to 
tidy, compartmentalized design.   Although I tell my team that data 
independence is important and that they should look at the problem from 
several angles to find the best way to structure a suite, it ain't 
always easy.

An alternative is to pack several tests into one with multiple 
opportunities to annotate and fail along the way, but that sort of 
defeats the whole purpose of using QMTest, right?

Maybe it's just me.  :)