Yukihiro Matsumoto wrote:

> It's intentional, so that it shouldn't be a bug.   But your message
> made me think having field names with "?" can be valid if field
> accessor is disabled for such cases.

Though I wonder if having "foo?" and "foo=" for boolean accessors like 
flags would not be useful.

For example I'm frequently using things like Contract.check_fulfills and 
.check_fulfills= and I think that

   if Contract.check_fulfills? then
     ...
   end

just looks nicer than

   if Contract.check_fulfills
     ...
   end

OTOH I totally dislike .foo?= because the double combinator just looks 
ugly...

So if this were to be implemented I think attr_accessor :foo? should 
create .foo? and .foo= -- and Struct.new(:foo?) then would also be similar.