[plt-scheme] [scribble] syntax quasi/unquote

To make use `lambda' in the right-and side of a `define-syntax', it you
start with just `scheme/base' as in the default evaluator, then you
need to also require `(for-syntax scheme/base)':

 @(define ex-eval (make-base-eval))
 @interaction-eval[#:eval ex-eval (require (for-syntax scheme/base))]

 @(let ([unsyntax 'hide]) 
  @def+int[
 #:eval ex-eval
 (define-syntax FOO
   (lambda (stx)
     (syntax-case stx ()
       ((FOO val)
        #`(+ 1 #,(+ 1 2))))))
 (FOO 2) ])


At Thu, 29 Apr 2010 14:05:44 -0400, Eric Tanter wrote:
> > I thought that something like this might work
> > 
> > (define-syntax-rule @INTERACTION[form ...]
> >   @interaction[(eval:alts @#,SCHEMEBLOCK[form] form) ...])
> > 
> > but the problem is that the first part o `eval:lats' must be an
> > element, instead of a block.
> > 
> > So, here's a different idea. The `def+int', etc. forms recognize
> > `unsyntax' by its binding, so you can just shadow that binding, like
> > this:
> > 
> > @(let ([unsyntax 'hide]) 
> >    @def+int[
> >     (define l 
> >      #`(#,(+ 1 2)))
> >     #`(#,l)
> >    ])
> 
> it does not work with a macro, for instance
> 
> #lang scheme
> (define-syntax FOO
>   (lambda (stx)
>     (syntax-case stx ()
>       ((FOO val)
>        #`(+ 1 #,(+ 1 2))))))
> (FOO 2) 
> --> 4
> 
> and:
> 
> @(let ([unsyntax 'hide]) 
>  @def+int[
> (define-syntax FOO
>   (lambda (stx)
>     (syntax-case stx ()
>       ((FOO val)
>        #`(+ 1 #,(+ 1 2))))))
> (FOO 2) ])
> 
> gives 
> eval:1:0: compile: unbound identifier in the transformer environment (and no 
> #%app syntax transformer is bound) at: lambda in: (lambda (stx) (syntax-case 
> stx () ((FOO val) (quasisyntax (+ 1 (unsyntax (+ 1 2)))))))
> 
>  === context ===
> /Applications/_Extra/Devel/PLT Scheme 
> v4.2.5/collects/scheme/private/more-scheme.ss:158:2: 
> call-with-break-parameterization
> /Applications/_Extra/Devel/PLT Scheme v4.2.5/collects/scheme/sandbox.ss:700:9: 
> loop
> 
> I get the same error if I shadow both unsyntax and quasisyntax.
> 
> (side question: why having chosen the same escape characters for the @... than 
> the ones for syntax objects?)
> 
> Thanks,
> 
> -- Éric