Hi All,
Which unit testing framework would you recommend among those available
for Common Lisp ?
My needs are simple, and not (yet) related to XP:
Test a non-object API (actually a binding to a C library).
I could do it without a test framework, but I'm trying to take
the habit to use tests so I think structuring them wouldn't be bad.
BTW, I'm using ACL.
Thanks,
Rodolphe, Lisp newby
Rodolphe Saugier <················@ircam.fr> writes:
> Hi All,
>
> Which unit testing framework would you recommend among those available
> for Common Lisp ?
> My needs are simple, and not (yet) related to XP:
> Test a non-object API (actually a binding to a C library).
> I could do it without a test framework, but I'm trying to take
> the habit to use tests so I think structuring them wouldn't be bad.
> BTW, I'm using ACL.
Take a look at
http://www.franz.com/support/documentation/6.2/doc/test-harness.htm
--
Duane Rettig ·····@franz.com Franz Inc. http://www.franz.com/
555 12th St., Suite 1450 http://www.555citycenter.com/
Oakland, Ca. 94607 Phone: (510) 452-2000; Fax: (510) 452-0182
On Thu, 18 Mar 2004 12:18:56 +0100, Rodolphe Saugier wrote:
> Which unit testing framework would you recommend among those available
> for Common Lisp ?
Check out CLUnit at http://www.ancar.org/CLUnit.html.
faa
According to Rodolphe Saugier <················@ircam.fr>:
> Which unit testing framework would you recommend among those available
> for Common Lisp ?
I use clos-unit. Cliki says it is "basically a port of JUnit." I've not
used JUnit. I found clos-unit to be similar enough to PyUnit, which I have
used.
http://www.lme.die.supsi.ch/~pedrazz/clos-unit/
--
Ng Pheng Siong <····@netmemetic.com>
http://firewall.rulemaker.net -+- Firewall Change Management & Version Control
http://sandbox.rulemaker.net/ngps -+- Open Source Python Crypto & SSL
Rodolphe Saugier <················@ircam.fr> writes:
> Hi All,
>
> Which unit testing framework would you recommend among those available
> for Common Lisp ?
> My needs are simple, and not (yet) related to XP:
> Test a non-object API (actually a binding to a C library).
> I could do it without a test framework, but I'm trying to take
> the habit to use tests so I think structuring them wouldn't be bad.
> BTW, I'm using ACL.
If I were you, I'd probably use the test harness built into ACL.
However, I don't use ACL, and I've been happily using rt. They're
both quite simple.
--
/|_ .-----------------------.
,' .\ / | No to Imperialist war |
,--' _,' | Wage class war! |
/ / `-----------------------'
( -. |
| ) |
(`-. '--.)
`. )----'
I've always wondered something about unit testing. What if the unit
testing suite is buggy?
--
Those who do not remember the history of Lisp are doomed to repeat it,
badly.
> (dwim x)
NIL
David Steuber wrote:
> I've always wondered something about unit testing. What if the unit
> testing suite is buggy?
>
You fix the tests.
I know it seems like a catch-22, but in practice the fact that the test
code and the code being tested change somewhat independently means that
each body of code tends to expose bugs in the other, so things work
better than if the units tests didn't exist. Another nice things is that
when you make some giant galactic change, the tests enable you to
quickly find a state where you are reasonably confident that the new
code works about as well as the old code.
mikel <·····@evins.net> writes:
> David Steuber wrote:
>> I've always wondered something about unit testing. What if the unit
>> testing suite is buggy?
>>
>
> You fix the tests.
>
> I know it seems like a catch-22, but in practice the fact that the
> test code and the code being tested change somewhat independently
> means that each body of code tends to expose bugs in the other, so
> things work better than if the units tests didn't exist. Another nice
> things is that when you make some giant galactic change, the tests
> enable you to quickly find a state where you are reasonably confident
> that the new code works about as well as the old code.
Also, if the unit-testing framework is extensively used, it
will also be extensively tested, probably more so than any of the code
it is used to test.
--
Raymond Wiker Mail: ·············@fast.no
Senior Software Engineer Web: http://www.fast.no/
Fast Search & Transfer ASA Phone: +47 23 01 11 60
P.O. Box 1677 Vika Fax: +47 35 54 87 99
NO-0120 Oslo, NORWAY Mob: +47 48 01 11 60
Try FAST Search: http://alltheweb.com/
David Steuber <·············@verizon.net> writes:
> I've always wondered something about unit testing. What if the unit
> testing suite is buggy?
Did you notice that I pointed in the direction of simple test suites? :-)
rt is 150-250 LOC, depending on what version you use. This is kind of
like proof generation -- code to generate proofs is hard to write, and
you generally aren't 100% confident in the proofs it generates.
That's why you have it emit proofs in a logical framework that is very
easy to write proof checkers for. At some point, it comes down to
being able to hand-verify your code. 200 lines of simple code is easy
enough to verify.
--
/|_ .-----------------------.
,' .\ / | No to Imperialist war |
,--' _,' | Wage class war! |
/ / `-----------------------'
( -. |
| ) |
(`-. '--.)
`. )----'
Thomas F. Burdick wrote:
> David Steuber <·············@verizon.net> writes:
>
>
>>I've always wondered something about unit testing. What if the unit
>>testing suite is buggy?
>
>
> Did you notice that I pointed in the direction of simple test suites? :-)
>
> rt is 150-250 LOC, depending on what version you use. This is kind of
> like proof generation -- code to generate proofs is hard to write, and
> you generally aren't 100% confident in the proofs it generates.
> That's why you have it emit proofs in a logical framework that is very
> easy to write proof checkers for. At some point, it comes down to
> being able to hand-verify your code. 200 lines of simple code is easy
> enough to verify.
Only as a little note:
you could write in one hour a theorem prover that is capable to prove
anything that can be proved and which is always correct under given
axioms...
Thomas F. Burdick wrote:
> Rodolphe Saugier <················@ircam.fr> writes:
>
>
>>Hi All,
>>
>>Which unit testing framework would you recommend among those available
>>for Common Lisp ?
>>My needs are simple, and not (yet) related to XP:
>>Test a non-object API (actually a binding to a C library).
>>I could do it without a test framework, but I'm trying to take
>>the habit to use tests so I think structuring them wouldn't be bad.
>>BTW, I'm using ACL.
>
>
> If I were you, I'd probably use the test harness built into ACL.
> However, I don't use ACL, and I've been happily using rt. They're
> both quite simple.
>
I've used xptest with some success.