Organization of tests...
I've made it to the point I'm dangerous. I know I can automate most elements on our product's UI as well as pulling info from most objects on screen for testing purpose. (The developers may need to help me with some of their UI elements.)
I've begun to think about how to organize the scripts and tests for as much reuse as possible.
I currently have scripts broken down into a number of folders based on where in the sw workflow they can be accessed.
Now, I think the way I need to organize the tests is to add groups and workflow tests to the Test Items page. The tests thereon call workflows and subroutines from the scripts folders. The nice thing about this is that it is easy to create the groups of tests and turn them on and off.
I wonder about a few things:
- initially I had a number of tests called from main and there was a sendlog performed at the end of main. With the restructure, should I just added a final "test" item on the Test Items Page which only calls the sendlog (for the entire set of tests that were run.) For that matter, most of the tests assume a particular intial state (login has occurred); should I just call this setup routine prior to running the group of tests that require it?
- in perusing the Help files, I think I saw an example where main was used to iteratively call a number of tests. Is this a more common structuring?
- finally does it sound like I'm approaching the structuring in a reasonable fashion? Being a novice, I'm sure I'm overlooking lots of things. Can someone provide a little feedback about test structuring or point to an existing example or help topic or training clip, ...
Thanks in advance,
chris
I've begun to think about how to organize the scripts and tests for as much reuse as possible.
I currently have scripts broken down into a number of folders based on where in the sw workflow they can be accessed.
Now, I think the way I need to organize the tests is to add groups and workflow tests to the Test Items page. The tests thereon call workflows and subroutines from the scripts folders. The nice thing about this is that it is easy to create the groups of tests and turn them on and off.
I wonder about a few things:
- initially I had a number of tests called from main and there was a sendlog performed at the end of main. With the restructure, should I just added a final "test" item on the Test Items Page which only calls the sendlog (for the entire set of tests that were run.) For that matter, most of the tests assume a particular intial state (login has occurred); should I just call this setup routine prior to running the group of tests that require it?
- in perusing the Help files, I think I saw an example where main was used to iteratively call a number of tests. Is this a more common structuring?
- finally does it sound like I'm approaching the structuring in a reasonable fashion? Being a novice, I'm sure I'm overlooking lots of things. Can someone provide a little feedback about test structuring or point to an existing example or help topic or training clip, ...
Thanks in advance,
chris
