userfeb wrote:Why doesn't eValid have "programming" controls in the script language?
The eValid approach to web application testing is based on the assumtion that a test will PASS, but that if errors occur then alternative action has to be taken.
So, eValid does have "flow of control" in the script, but not in the conventional programming language sense. Instead, eValid has the notion of an action flag, an ERROR flag or an overtime ALARM flag or a TIMEOUT flag.
When these kinds of FAILures in a playback occur, then eValid can take corrective action.
But to put that condition on each command -- as you might in a programming language, with each command doing a check with an
if -- you get a pretty messy (too complex) script. So what eValid does is declare in the script the place where control picks up when a flag is thrown.
So if you have a script and it says
OnErrorGoScript <name.evs> then what will happen is that in case ANY subsequent command throws an ERROR flag then control will jump there.
For the purposes of a regression test, or a functional test which performs some time or size measurement, this approach is a lot simpler than the same kind of thing done in a programming language.
eValid Support
by userfeb » Wed Feb 09, 2011 5:24 pm