From: <[EMAIL PROTECTED]>

> On Fri, 26 Jul 2002, Wannheden, Knut wrote:
> 
> > But otherwise sounds good!  This will make a cleaner interface for people
> > writing custom frontends as they can supply their own ProjectHelper and
> > practically call it with any input.  Just the ProjectHelper knows how to
> > deal with it ;-)
> 
> I think you are talking about frontents supplying their ProjectHelper 
> implementations, not replace oat.ant.ProjectHelper. 
> 
> That's how it think it should work.

I agree in part. I think APIs passing "Object" should be avoided as much as 
posible.
They say nothing about the constraints imposed by the API. It makes no sense to
pass things other than those able to produce an stream of input to the current 
implementations
of ProjectHelperImpl.

A more useful functionality would be to able to pass something like a URL from
which these parsers can obtain their imput.

For other types of ProjectHelperImpl implementations, they should be able to 
use a
URL or URI type syntax to describe the input to their implementations. This 
would
at least maintain a strongly typed API.

With respect to having an API to spacify the PHI to use, I think that is 
defenitly good
as one may have the autoload settings to use PHI2 and at the same time have 
other tools
requiring their own PHI.

Jose Alberto



--
To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>

Reply via email to