From: Gavin Sinclair <gsinclair@...>
Date: 2002-12-09T12:26:54+09:00
Subject: Re: [YANQ] -yet another (ruby) newbie question -string#concat

From: "Mike Campbell" <michael_s_campbell@yahoo.com>


>
> > String#concat is a synonym for String#<<, and since it modifies the
> > object, a concat! version wouldn't make sense.  You can indeed use
> > String#+ if you want to create a new string instead of modifying the
> > old one.
>
> To be fair here, it might be argued that the String#concat alias is misnamed
and
> *IT* doesn't make sense, given what it does.  Seems to me (and I'm guessing
> Botp) that the alias SHOULD be String#concat!.


I'd agree with that, FWIW.  IMO, botp's confusion is quite understandable.  I
never use #concat, but if I did I'm sure I would have raised this issue before.

I see no reason why str1.concat(str2) should modify str1, any more than the
Unix command "cat file1 file2" should modify file1 (it doesn't).

The Mirriam-Webster online dictionary lists "concatenate [v.t.]" as

   to link together in a series or chain

This suggests that String#concat should indeed be destruvtive.  However, since
there is ample room for confusion, and there are many String#xxx! methods, I
conclude that the ideal solution is for String to have the following methods:

   String#concat
   String#concat!

with their operations being (now) obvious.

I wonder how much existing code such a change will break.  After all, who, in a
published program or library, would use #concat ? :)

Cheers,
Gavin