[ruby-talk:11231] Re: RCR: Enumerable#hashify

On Wed, 21 Feb 2001, Kevin Smith wrote:

> David Alan Black wrote:
> >  Enumerable#hashify(value)
> >
> >Description: 
> >
> >  Return a hash based on iterating through the receiver (an
> >  enumerable), either setting every value to the parameter "value" or
> >  setting the values via a block.
> 
> Seems reasonable, but I wince every time I read 
> the word "hashify". I'm not even sure why. Could 
> you explain why to_h or to_hash would not be an 
> appropriate name?

I steered away from to_h and to_hash partly because I felt that to do
those names justice, they'd have to behave like to_s/to_str or
to_a/to_ary.  I wasn't sure what that would entail... and I didn't
want the possibility of a simple E-to-H method to be delayed because
of that.

In addition to hashify, I had also considered "hash_with(value)",
though I rejected that because it didn't sound right when doing 
the other thing you don't like, namely...

> One other nit: I would prefer that it require 
> either a parameter or a block. Initializing all 
> the values to true doesn't seem helpful to me.

... calling it without an argument ("h = a.hash_with" looks weird).
So "hashify" was an attempt to find a name that would make sense with
or without an argument.

As for nit #2: personally I like having it default to something.  You
can always send it a value anyway (i.e., the defaulting version is a
superset of the non-defaulting version), and there's something nice, I
find, about just being able to flip an array into a hash very
concisely:

    h = a.hashify


David

-- 
David Alan Black
home: [e-mail]
work: [e-mail]
Web:  http://pirate.shu.edu/~blackdav