[plt-scheme] Reimplementation of srfi-26 (cut and cute)

Lauri Alanko wrote:
> On Wed, Jul 20, 2005 at 09:40:04PM +0200, Jens Axel Søgaard wrote:
> 
>>Attached is a reimplementation of srfi-26 with
>>better error messages than the original.
> 
> 
> I just tried this out, and found that the reimplementation does not
> follow the spec. The SRFI allows a <slot-or-expr> as the first argument
> to cut or cute, hence making a slot at the operator position perfectly
> legal.

You are absolutely right. This is bug 7824:

     <http://bugs.plt-scheme.org/>

Which also has a fix - but someone needs to check the
code, and then commit it to SVN.

The test suite for the reference implementation didn't
have any tests for <> in the first position, so somehow
I missed it that case.

 > I also removed the special error handling for (cut <...> ...). Whether
 > it even should be an error is arguable, and if it should, then the
 > macro should check _all_ non-tail occurrences of <...>, not just at
 > the operator position.

The grammar in the specification doesn't allow <...> in the
first position, so why not check for that?

-- 
Jens Axel Søgaard