From: Trevor Andrade <trevor.andrade@...>
Date: 2005-01-09T08:26:56+09:00
Subject: Re: Type inference in ruby

In this case I did not mean redefine integers and integer addition I meant
define in the first place.  What I was trying to say is that if you had no
built in construct in Ruby of integers then there would be absolutely no way
in which to define integers or integer addition in Ruby.  You would have to
make a c extension to the language.  Of course it is possible to redefine
integers but that is not what I was talking about.

What I was originally trying to get at is that there are only 2 things you
do in Ruby 1) send messages to objects 2) Perform built in operations on
built on things like integers and strings.  In 1) you are not really doing
anything instead you are sending messages to objects and telling them to do
things.  In 2) you are doing something like adding integers, concatenating
strings, displaying output, getting input etc.  Type checking for 1)
involves checking if an object can accept a certain kind of message.  Type
checking for 2) is more involved and depends on the built-in types
themselves, like for instance the fact that you can't add a string to an
integer although you can add a float to an integer.  Even for 2) type
inference is still possible because there are only a finite number of built
in constructs and their properties are known.  Of course if you add in new c
extensions to the language then these cannot be checked. However I don't
think type checking has to be exhaustive or strong.

-----Original Message-----
From: Alexander Kellett [mailto:ruby-lists@lypanov.net] 
Sent: Saturday, January 08, 2005 5:26 PM
To: ruby-talk ML
Subject: Re: Type inference in ruby

On Jan 8, 2005, at 9:28 PM, Trevor Andrade wrote:
> You cannot define how to add integers in Ruby and in fact you cannot 
> even define
> integers themselves.  Integers are low level constructs that must be 
> built
> into the language.  [snip]

its possible to redefine integer additions in ruby.
however yes you're correct in saying that its not possible
to redefine the integer constructor. (tho i'm working
on making that possible ;))

> There is only one thing wrong with type inference and that is that new
> methods can be defined at runtime in Ruby which means it is possible 
> that an
> object does not know how to respond at compile time to a message but 
> it does
> know how to respond at runtime.  However I think this scenario happens 
> very
> rarely and if it was happening the programmer could just turn off type
> inference.

i think personally that test suites should guide typing metadata.
if a testcase only has a test for a single data type. then the
assumption can be made that the type itself is required. if however
it has a test for two different data types each of which have a given
method, which happens to be part of an included module, to ease test
case writing it should be possible via comment hints to show the
inferrer / document-type-metadata-hinter that it is duck typed not
static typed.

Alex