From: Conrad Schneiker <schneik@...>
Date: 2001-02-20T05:42:08+09:00
Subject: [ruby-talk:11131] Re: Summary: RCR #U002 - proper new name fo r indexes

Robert Feldt wrote:

# On Tue, 20 Feb 2001, Jean Michel wrote:
# 
# > I like to consider an array conceptually as a special case of hash 
where
# > the keys are restricted to be positive integers. So I am very opposed 
to
# > choices which  make them differ  more than  necessary (in the  spirit 
of
# >
# I fully agree with this and would like to add that its easier to
# learn one method name than two...

Likewise.

# > >> >   (a) Hash#values, Array#values
# > >> >   (b) Hash#values_for, Array#values_for
# > >> >   (g) Hash#selections, Array#selections
# > >> >   (h) Hash#collect_indexes, Array#collect_indexes
# > >> >   (f) Hash#subset, Array#subset
# > 
# but I'd vote for Hash#elements_at/Array#elements_at or
# Hash#elements/Array#elements.

I'd vote for using *#values* (1st * --> "Hash", "Array"; 2nd * --> "", 
"_at", "_for") rather than *#elements*, which I think would be more 
natural and more consistent with wider usage elsewhere (read: prospective 
future Ruby users). 

For example, the "Programming Perl" authors describe the above 
generalization as "[A] has is just a funny kind of array in which you look 
values up using key strings instead of numbers. They also speak of 
"key/value" pairs. 

As one might expect, practice is mixed. Here are some additional 
semi-random samples: a ksh manual uses "elements", a Python manual uses 
"items" and "values", and a Java manual uses "entries", "objects", "values 
of ... type", etc. 

Overall, *#values* seems like the more natural generic term, which seems 
more likely to accommodate any subsequent future generalizations that 
people might discover.

Conrad Schneiker
(This note is unofficial and subject to improvement without notice.)