Union/intersection types confusion (was: Re: Difficulty deriving a general type for combinator) - comp.lang.haskell

Допис з розмови Union/intersection types confusion (was: Re: Difficulty deriving a general type for combinator)

Simon Richard Clarkstone

Переглянути профіль

Більше налаштувань 20 Вер 2008, 01:02

Групи новин: comp.lang.haskell

Від: Simon Richard Clarkstone <s.r.clarkst...@dunelm.org.uk>

Дата: Fri, 19 Sep 2008 23:02:44 +0100

Локально: Сб 20 Вер 2008 01:02

Тема: Union/intersection types confusion (was: Re: Difficulty deriving a general type for combinator)

  Відповісти автору
  |
  Переслати | Друкувати | Дивитися гілку
  | Показати оригінал
  | Повідомити адміністраторів про цей допис | Знайти повідлмлення від цього автора
  

Simon Richard Clarkstone wrote:
> Well, these particular examples could both be typed if there were
> something like "union types". Using Java's syntax for interface unions:

> app2 :: (a -> b && c -> d) -> (a,b) -> (c,d)
> twice :: (a -> b && b -> c) -> a -> c

I worded that very badly. "Type1 && Type2" in Java means an item from
the intersection of the sets of items of each type, which AIUI is called
an intersection type. When I said "union", I was thinking of the set of
operations and contracts known to be present by examining the type: the
union of the sets of method signatures of the two types.

Conversely, given a union type, the set of operations and contracts
known to be present is the *intersection* of those for all the united
types. I would guess there is a good reason for this reversal, and it
is not just a coincidence.

--
Simon Richard Clarkstone:
s.r.cl?rkst...@dunelm.org.uk / s?m?n_cl?rkst...@yahoo.co.uk
| My half-daughter went to the GMH riots |
| But all I got was this stupid ¥-shirt. |

Відповісти автору Переслати