On Sun, Aug 27, 2006, Robert Collins wrote:

> I'll hold it in a branch indefinately then. If we get some good traction
> this coming weekend - the bug-fixing-fest - I will be much happier about
> it sitting in a branch for a month, maybe two, than if we dont.

Oh, it'll happen. It'll be useful if we can pull in a few more C++-cluey
people into the fold.

If this one is even moderately successful I think it would be useful to
extend the invitation to squid-users. As long as people who join are willing
to code and not complain. :)

> With regards to the sit-down-and-nut-out - refactoring is /defined/ as
> 'improving the design of existing code'. I really think its fine to
> refactor to make incremental improvements. This is in no way
> incompatible with:
>  * doing a design overhaul and setting a new target for the refactoring 
>  * Doing a rewrite of some chunks of the code base.
> 
> The primary benefit I'm aiming at is a squid which is easily amenable to
> writing unit tests: so that we can be confident regressions are not
> occuring.

Its a good goal. The trouble is that you're creating new APIs just by
refactoring.  It might be easy to change the API(s) at the point you've
done the refactoring but I don't think it'll be so easy to do later on
as other code is brought into line to use the new API(s).





Adrian

Reply via email to