Re: cond-expand feature testing for presence of a Java class

[Jamison Hope <jrh at theptrgroup dot com>]

On Apr 3, 2011, at 8:23 PM, Per Bothner wrote:

> On 04/03/2011 04:56 PM, Jamison Hope wrote:
> > I recently found myself wanting to conditionalize some code on the
> > presence of a Java class. At first, I tried to see if I could have a
> > module call (provide 'some-feature) only when this class was found,
> > but then I realized it would be much simpler to modify the feature
> > testing in IfFeature.java to recognize class names as a last resort.
> > This has proven quite handy for me, letting me do something like:
> >
> > (cond-expand
> > (com.example.AClass ;; AClass is found in classpath, safe to use
> > (define-simple-class MySubclass (com.example.AClass) ...))
> > (else (error "AClass not found.")))
> >
> > It seems unlikely for some unrelated feature's name to exactly  
> > match a
> > fully-qualified class name, so I think this patch is pretty safe.
>
> Seems reasonable - but idea occurs to me:  Perhaps we should
> use the class: namespace prefix, as in:
>
> http://www.gnu.org/software/kawa/Field-operations.html
>
> Your example would be:
>
> (cond-expand
>  (class:com.example.AClass ;; AClass is found in classpath, safe to  
> use
>    (define-simple-class MySubclass (com.example.AClass) ...))
>  (else (error "AClass not found.")))
>
> What do you think?

That's fine with me.

--
Jamison Hope
The PTR Group
www.theptrgroup.com