[racket] plai test/exn

From: Eric Dobson (endobson at cs.brown.edu)
Date: Mon Sep 19 18:24:44 EDT 2011

I found it immediately by searching for error on docs.racket-lang.org,
and going to the plai entry.

http://docs.racket-lang.org/plai/plai-scheme.html?q=error#(def._((lib._plai/main..rkt)._error))

Is there some other place you would expect it to be?

(It does need some scheme->racket cleanup though)

-Eric

On Mon, Sep 19, 2011 at 3:21 PM, Jeremy Kun <kun.jeremy at gmail.com> wrote:
> Ah! Why don't they state that they shadow "error" in the documentation?
> Isn't that considered bad form? And worse, undocumented bad form!
> Everything I was taught in school by idealistic Racket programmers is
> crashing down before my eyes! I might as well be coding in C.
> But seriously, this should be in the docs.
>
> Jeremy
>
>
> On Mon, Sep 19, 2011 at 5:12 PM, Jay McCarthy <jay.mccarthy at gmail.com>
> wrote:
>>
>> And that's what it means by "user code"
>>
>> Jay
>>
>> On Tue, Sep 20, 2011 at 6:57 AM, Eric Dobson <endobson at cs.brown.edu>
>> wrote:
>> > Also IIRC test/exn only works on errors raised using 'error' from
>> > PLAI, which is different from 'error' from racket/base.
>> >
>> > -Eric
>> >
>> > On Mon, Sep 19, 2011 at 2:49 PM, John Clements
>> > <clements at brinckerhoff.org> wrote:
>> >>
>> >> On Sep 19, 2011, at 1:47 PM, Jeremy Kun wrote:
>> >>
>> >>> Not sure if this is a bug, but it doesn't match the documentation
>> >>> here. It says test/exn only succeeds when the error is explicitly raised by
>> >>> the user, but I try something as simple as:
>> >>>
>> >>> "file1.rkt"
>> >>> (provide foo)
>> >>> (define (foo) (error "hello"))
>> >>>
>> >>> "file2.rkt"
>> >>> (require plai "file1.rkt")
>> >>> (test/exn (foo) "hello")
>> >>>
>> >>> and the test fails with "no expected value". I can even replace
>> >>> "hello" with a call to error, and see that that argument is never evaluated.
>> >>> So by user-written code, I'm assuming they mean code contained in the same
>> >>> module. Or am I missing something fundamental about Racket's idea of
>> >>> "user-written code"?
>> >>
>> >> Just a quick look, but it looks like you're failing to thunk the
>> >> arguments to test/exn, so it never even gets control.
>> >>
>> >> John
>> >>
>> >>
>> >> _________________________________________________
>> >>  For list-related administrative tasks:
>> >>  http://lists.racket-lang.org/listinfo/users
>> >>
>> >
>> > _________________________________________________
>> >  For list-related administrative tasks:
>> >  http://lists.racket-lang.org/listinfo/users
>>
>>
>>
>> --
>> Jay McCarthy <jay at cs.byu.edu>
>> Assistant Professor / Brigham Young University
>> http://faculty.cs.byu.edu/~jay
>>
>> "The glory of God is Intelligence" - D&C 93
>
>



Posted on the users mailing list.