From: Michael Campbell <michael.campbell@...>
Date: 2005-08-05T04:51:06+09:00
Subject: Re: Cascading <=> comparisons

"Robert Klemme" <bob.news@gmx.net> writes:

> Michael Campbell <michael.campbell@gmail.com> wrote:
> > "Robert Klemme" <bob.news@gmx.net> writes:
> >
> >> David A. Black <dblack@wobblini.net> wrote:
> >>> Hi --
> >>> On Sat, 16 Jul 2005, Robert Klemme wrote:
> >>>
> >>>> This technique can be applied to sort_by, too:
> >>>>
> >>>>>> hash.sort_by {|a| a.reverse}
> >>>> => [["a", 1], ["d", 1], ["c", 3], ["b", 4]]
> >>>> But this shows a strange anomaly - this seems like a bug in 1.8.2.
> >>>>
> >>>>>> hash.sort_by {|a| a.reverse!}
> >>>> => [[1, "a"], [1, "d"], [3, "c"], [4, "b"]]
> >>> I'll probably figure this out as soon as I hit 'send'... but what's
> >>> the anomaly here?
> >> And, did you figure in the meantime? It's the different result -
> >> IMHO anything that sort_by does should have no such effect on the
> >> result.

> > Even if the operation is a "bang" one?
> 
> IMO yes.

Wow.  How do you tell ruby to not do what it normally would?  How do
reconcile that sort_by {|a| <x>} is different than sort_by {|a| <x!> }
for any given <x>?  What about OTHER side effects of <X> (like, a
database call or something)?



-- 
I tend to view "truly flexible" by another term: "Make everything
equally hard". -- DHH