From: Mathieu Bouchard <matju@...>
Date: 2000-09-26T12:35:37+09:00
Subject: [ruby-talk:5125] Re: Crazy idea? infix method calls

> Since it mimics an operator, it will have to have a precedence.
> That's all that comes to my mind. If I had had more sleep last
> night, I might have an opinion about high versus low... :)
> Presumably we could fix it so that for all operaors op1 and op2,
> a op1 b in c op2 d  would mean the same as the expression
> a op1 c.include?(b) op2 d.
> Does anyone but me like this proposal at all?

let's say that there are 5 classes of precedence

1. the very high: :: []
2. operators on objects, that return objects
3. operators on objects, that return booleans (predicates: comparisons)
4. operators on booleans, that return booleans (logic)
5. the rest (interval, assignment, the very-low-prec logic)

then 'in' would fit in #3. I'd tend to say at exactly the same level as
the pattern match (=~ !~) operators in particular.

matju