[elephant-devel] 5am tests

Ian Eslick eslick at media.mit.edu
Tue Jul 29 13:57:14 UTC 2008


My vote is that we stick with 5am - it may be easy to factor our what  
we need from arnesi into a separate 5am-util package.  There can't be  
that many dependencies, it's a rather small amount of code.

Ian

On Jul 29, 2008, at 8:57 AM, Henrik Hjelte wrote:

> On Sun, Jul 27, 2008 at 9:32 AM, Leslie P. Polzer <leslie.polzer at gmx.net 
> > wrote:
>>
>>> How about using lift for testing. It is well supported, and  
>>> creates nice reports.
>>
>> I think in the context of Elephant either 5am and Lift would be fine.
>>
>> All or most tests are in the form that rt expects, with only a thin
>> layer of 5am adapter above.
>
> Yes most but not all. One of the main problems with rt is that the
> list of tests needed an
> implicit order of execution, so for some tests I have used the elegant
> 5am feature of
> providing a list of dependencies directly in the test definition. This
> enables you to rerun
> a single test without running all of the testcases, and it makes you
> document the implicit
> ordering of the tests.
>
> This great feature is something I believe Lift lacks.
>
> Also, I think I have some random test that 5am provides, but I think
> Lift has something
> similar.
>
>>
>> How about you hacking Lift support?
> Then on my wishlist add the dependency tracking option that 5am has,
> it would be a good
> contribution to Lift.
>
> Or, maybe you can help fix arnesi for windows or remove the dependency
> of arnesi
> for 5am, I don't know how much it uses arnesi, I don't think it is  
> much.
>
> Is Lift without difficult dependencies? When I used cl-markdown also
> by Gary King I know it
> changed a lot of things in the running Lisp with some MOP tweakings.
>
> /Henrik Hjelte
> _______________________________________________
> elephant-devel site list
> elephant-devel at common-lisp.net
> http://common-lisp.net/mailman/listinfo/elephant-devel




More information about the elephant-devel mailing list