Issues relating to screen size.
User STEVE GILBERT (STEVE.GILBERT@thales-cs.com.nospam) posted:
My hint is to never use region image vp's - These are notorious for having
problems, we had 7 identical test machines and 1 bloke (not me) tried a
region image record on one and it would fail on all the others...
I have never had a situation where it was not possible to verify using the
Re: Issues relating to screen size.
User DAVID MARSH (DAVID.MARSH@thales-cs.com.nospam) posted:
Unless your functional requiremets actually define precisely the number of
pixels, which might be so for some people, what are you trying to compare?
Whilst the check everything on the screen approach is a very easy way to set
up a test on one machine it will kick you in the teeth in the end.
You might have a series of functions to extract the contents of objects,
window names and so on. You then might use a larger function to compare a
whole page. If you pass in the expected results you could use that function
to test the same screen in different states.
Then you never have to worry about the operating system version, fix pack,
video card, driver, etc etc etc.
An alternative would be to do a funky little function to check out the size
of the screen then compare to the relevant image for that svreen size. Even
that could fail because you could have 2 screens with identical pixel counts
but say one lost a bit off the top and another from the bottom.