Attention is currently required from: fixeria, laforge, pespin.

View Change

1 comment:

• File pcu/PCU_Tests_SNS.ttcn:

  • Patch Set #1, Line 327: integer idx := 0, integer idx_del := 1, float tout := 20.0

    @pespin@sysmocom.de this is our test code. If there are unknown and unforseen side effect, we should know it.

    I'm sorry but I don't think it's good to start using parameters in testcase with no real good reason or improvement. This may actually cause more unforeseen problems than good things.

    This sounds to me "we've never done it this way, never change a running system", which I don't see as valid argument.

    Well, then the point is: if parameters are not changed, why passing parameters instead of declaring them as variables...

    For me it makes the code more clearer/readable, defining essential variables as test case parameters and "promote" these variable to testcase parameters.
    Further it would have make it easier to split it into a testcase and a function.

    I'm not arguing here where to use testcase parameter in control() or not.

To view, visit change 39011. To unsubscribe, or for help writing mail filters, visit settings.