Hi,
I'm researching the history of proposals to add types to Python, partly
to improve my own code, and secondly to be able to make an informed PEP
one day.
The Types-SIG archives link
(http://mail.python.org/pipermail/types-sig/) gives a 404 error. Does
anyone know how to get the archive back?
While working on type expressions I am rather stuck for a
way to express recursive types. A simple example of this is a
singly-linked list of integers. In some languages you have compiler
syntax
which suspends evaluation so you can have recursive types. e.g.
typedef Linked_List := int, Linked_
How about mutual recursion?
class LinkedListA(TypeDef):
typedef = (int, LinkedListB)
class LinkedListB(TypeDef):
typedef = (int, LinkedListA)
--
http://mail.python.org/mailman/listinfo/python-list
That's a good fix. But I have misgivngs about needing a global name
registry. I need to support anonymous types and types with local
(lexical) scope. For example:
def makeAnonymousRecursiveType(T):
# anonymous type expression with local scope
LinkedList = (T, lambda: LinkedList)
return
If we -are- limited to lambdas, I see two options. Either embed lambdas
for each circular link, or have the whole expression in one lambda. ie
LinkedList3 = (int, lambda: LinkedList3, lambda: TypeNameX)
vs
LinkedList2 = lambda: (int, LinkedList2, TypeNameX)
The second option looks neate
Guido has proposed a syntax for type annotations in Python-3000.
Example:
def foo(x: t1, y: t2) -> t3:
...body...
http://www.artima.com/weblogs/viewpost.jsp?thread=87182
The types are dynamic and there is significant execution of code prior
to the function body being called. Which tends to
