On Tue, Jun 26, 2012 at 8:35 PM, Paul Belanger
<[email protected]> wrote:
> On 12-06-26 07:08 PM, Ken Barber wrote:
>>
>> I wanted to see what peoples thoughts were around when an acceptance
>> test is required, there was a debate on a ticket recently and two
>> people told me it wasn't necessary to add acceptance tests for a
>> particular case
>
> For me, you need to decided what you are trying to solve.  A problem with
> regressions in release branches, new features going into master or security
> issues.  I think the hardest part is trying to determine which scenario you
> add them in, and which you don't.  Honestly, I'm of the mind set it doesn't
> matter.  Every time you make a change to code, you should be adding a test
> too.

FWIW, this is the policy that the developers here at Puppet follow:
write a test that demonstrates the problem first, then fix the
problem, so that you know you actually have a test for it.

Many of those are unit style tests, though, rather than high level
acceptance tests that exercise the whole product.

> Now, if developers don't see the value in adding them, I think there is a
> bigger issue. Is it because they don't want too or just because of the
> amount of time it takes to add and test them?  Again, ideally it would be
> easy to drop new tests into puppet and have them executed.

You would be absolutely right. :)

-- 
Daniel Pittman
⎋ Puppet Labs Developer – http://puppetlabs.com
♲ Made with 100 percent post-consumer electrons

-- 
You received this message because you are subscribed to the Google Groups 
"Puppet Developers" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/puppet-dev?hl=en.

Reply via email to