A wise old hermit known only as Vincent Harcq
<[EMAIL PROTECTED]> once said:
> ifHasMethod does not call setCurrentMethod.
> I propose to change that and the corresponding J files.
Wasn't that the point of the hasMethod tag we discussed recently (thread
with subject "entitycmp.j") which Ara removed because "the code was the
same"? I assumed that ifHasMethod was therefore already doing this, since
that was the point of the removed code...
> The issue I have is that XDtMethod:ifIsNotAbstract does not work in case
> a setter method is not abstract while a getter is.
>
> Another solution could be to parametrized XDtMethod:ifIsNotAbstract, but
> I think setting the current method will makes things clearer.
I also prefer setting the current method over adding another parameter to
ifIsNotAbstract. Having it in ifHasMethod does lose some flexibility,
though. It's fine for the BMP getters & setters, where we then just have
forAllPersistentFields
ifHasMethod(getter)
ifIsNotAbstract
<getter stuff>
ifHasMethod(setter)
ifIsNotAbstract
<setter stuff>
but if you're doing some weird custom template where code in one generated
method varies depending just on the existence of some other method(s), but
you still want to use the method level tag information from the original
method, then having ifHasMethod change the current method makes this more
difficult. Having said that, I'm struggling to come up with a decent
example that needs you to do this. If ifHasMethod isn't already changing
the current method, though, then I don't see anything wrong with how we
had it before i.e. one tag to check the existence, and another to change
the context to the new method.
Andrew.
_______________________________________________
Xdoclet-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/xdoclet-devel
