From: ts <decoux@...>
Date: 2001-11-01T03:13:18+09:00
Subject: [ruby-talk:24003] Re: RCR: Fun with attribute shortcuts solves RCR #3 and more

>>>>> "K" == Kevin Smith <kevinbsmith@yahoo.com> writes:

K> --- Dan Schmidt <dfan@harmonixmusic.com> wrote:
>> Gunnar Andersson <dff180@yahoo.com> writes:
>> 
>> | # file: AttrInitializer
>> | class Module
>> |    def attr_initializer(*attrs)
>> |       alist = plist = ""
>> |       attrs.each do |v|
>> |          alist += "@#{v},"     # >> @var1,@var2,@var3,
>> ...
>> |          plist += "#{v},"      # >> var1,var2,var3, ...
>> |       end
>> |       alist.chop!; plist.chop! # Remove last comma 
>> |       class_eval "def initialize(#{plist}); #{alist} =
>> #{plist}; end"
>> |    end
>> | end
>> 
>> Perhaps the generated initialize method could also check
>> for the presence of a initialize_more (or a better name,
>> hopefully) method and call it if it exists.

K> I was thinking that it would be nice if you could pass in a
K> block. The last thing attr_initializer would do before
K> returning would be to yield.

 Well, the problem that I see with this that is not really equivalent with
 attr_accessor, attr_writer, ...

  * You can have only one attr_initializer unlike attr_accessor
  * attr_accessor is, at least for me, an optimization. attr_initializer is
    nor a macro nor an optimization
  * attr_initializer call class_eval this mean that it exist the possibilty
    that it can't work with $SAFE 

 
Guy Decoux