Test::Harness in Core

[Andy Armstrong]

On 8 Jun 2009, at 17:46, Ovid wrote:
> As most of us are aware, there's a problem with dual-life modules in  
> the core.  It might be nice if Test::Harness was not one of those  
> problems.
>
> What would it take for TH to be split into two distributions?  One  
> of those would reside in core and only contain core features.  The  
> other (Test::Harness::Extended?) would be where new features are  
> developed.

I guess we could make it so that any class in Test::Harness would  
automatically delegate to the corresponding Test::Harness::Extended  
class if present.

Just for clarity what kind of problems do you anticipate if we keep  
developing new features against the code that's in core?

> Obviously there are all sorts of stumbling blocks with this, but if  
> we could figure out a clean way of making this a seamless  
> separation, we could keep a stable TH in core for years  
> (presumably), while still developing.

We could just fork it again :)

-- 
Andy Armstrong, Hexten