Include and mandatory-include behaviour in test plans

Zygmunt Krynicki zygmunt.krynicki at
Wed Jun 17 12:19:57 UTC 2015

On Wed, Jun 17, 2015 at 2:09 PM, Maciej Kisielewski
<maciej.kisielewski at> wrote:
> With the advent of a field specifying mandatory jobs for test plan units
> (which I will call 'mandatory-include' from now on), there are a few design
> decisions we have to make.
> If you have better alternatives for my approach below, please, do share :-)
> As the job may be present in the 'include' and/or 'mandatory-include'
> fields, we have 4 scenarios. This is my proposed behaviour for them:
> 1) not included, not mandatory-included
> Job shouldn't be available on job-selection screen. Job should never run.
> Note that if the job is required by other job it may become visible and,
> when selected, might be run.

What it is a dependency. Then it will run simply because it's pulled
into the run list automatically. What you specified is good but let's
be explicit if we're talking about _run_ list or _desired_ list.

> 2) included, not mandatory-included
> Job should be available on job-selection screen, user should be able to
> select and deselect it. It should be run only when selected or required by
> other jobs.
> 3) not included, mandatory-included
> Job should not be listed in job-selection screen and it should ALWAYS run.
> 4) included and mandatory-included
> Job should be listed in job-selection screen, but user should not be able to
> deselect it. It should always run.
> Jobs that are not deselectable should be rendered differently to cue the
> user (e.g. greyed-out)

I think cases 3 and 4 are not necessary. I'd rather see that as a
mistake and treat all cases like 4 (jobs should be visible as this is
simply useful) but all mandatory inclusions should take precedence. If
this happens within one test plan I would issue simple diagnostic
during validation.

> As an alternative, 3) could behave like 4) with advice from validators, that
> when placed in 'mandatory-include' the job doesn't have to be specified in
> the  'include'.

Oh, that :-)

> What do you think?
> --
> Have a good one,
> Maciek
> --
> Checkbox-devel mailing list
> Checkbox-devel at
> Modify settings or unsubscribe at:

More information about the Checkbox-devel mailing list