Prev: [ANN] Mail 2.2.0
Next: Text Fighting (Gladitorial) Sim
From: Intransition on 11 Apr 2010 08:47

On Apr 10, 6:38 pm, Intransition <transf...(a)gmail.com> wrote:
> Just released Facets 2.8.3. Release notes are below. This release
> should be the last in the 2.8.x series. The next release, in fact,
> will be 3.0 as Facets will be making the transition of a modular
> design instead of adding extension methods directly into core classes.
> Please let me know if you have any thoughts/concerns about this.

FYI. Looks like I won't be moving Ruby Facets to a modular design
after all! My main reason for doing so was to finally give-in and
conform to ActiveSupport's way of doing things, so that the two libs
could play nice without issue or special conditional clauses. So often
I've see people write extensions emulating how ActiveSupport handled
it with mixin modules (b/c presumably that's where they learned how to
do it), and I would often suggest they not bother with all that and
just insert the extensions directly. But I was always told they wanted
to "do the right thing" and play nice with ActiveSupport. But the
truth is, using modules provides a false sense of security --nothing
of significant import is gained by it. Well, lo and behold, the guys
over at ActiveSupport must have finally seen the light, because I was
just perusing the new 3.0 lib and all the mixin modules are gone!. All
I can say is "Thank the Maker"! While using modules has some benefits
in documenting niceties, both in RDoc and error messages, they were
fraught with issues like the Double Inclusion Problem, the need for
separate modules for metaclass extensions, and worst of all, as I
mentioned, forcing everyone else to do the same thing in order to
coexist.

So whomever is responsible for this change over there at the
"Railroad". Thank You! You just made my life, and ultimately everyone
else's, a whole lot easier.

 | 
Pages: 1
Prev: [ANN] Mail 2.2.0
Next: Text Fighting (Gladitorial) Sim