Richard Frith-Macdonald schrieb:
>
> On 24 Dec 2006, at 10:16, Matt Rice wrote:
>
>> i'm not going to go so far as to object to this, but
>> it seems as though there are already alot of
>> unimplemented methods and classes in GNUstep, and it
>> would be nice if they were easier to find when one
>> wanted to work on something,
>
> It would be quite easy to mark all unimplemented stubs consistently so
> that autogsdoc would document them as unimplemented, and they would
> raise an exception at runtime...
>
> - (id) aMethod
> {
> /** not implemented */
> [self notImplemented: _cmd];
> return nil.
> }
>
> We could even extend autogsdoc so that '/** <notimplemented /> */'
> could be auto-expanded to a big red warning in the documentation along
> with a request that someone work on it and instruction to them abbout
> how they can get started.
>
> Worth doing?
>
Yes, this is surely worth to have it done. I normally place a //FIXME
comment into any method that I write, where I think that I did not add
the full code, still it is better to have a runtime feature showing such
limitations in a proper way.
_______________________________________________
Discuss-gnustep mailing list
Discuss-gnustep@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnustep
