We've been going around a bit about how PMCs should be set and
assigned to and suchlike things. So, here's the scoop.
There are three sorts of assignment:
1) Pointer assignment
2) Requests to get a value
3) Forced value sets
#1 is essentially a sort of bookkeeping. A pointer to one sort of PMC
is replaced with another pointer. This happens, for example, with the
SET opcode. It does *not* affect the contents of the destination PMC.
#2 is what you get with normal assignment. $foo = $bar, for example.
$foo's assign vtable method is called with $bar as a parameter. $foo
figures out what it should do--if it's a tied variable of some sort
it should perform its assign action. (This includes throwing an
exception if the assignment isn't valid) Otherwise it should
typecheck the RHS and morph itself into the RHS's type.
#3 calls the destination PMC's destroy method if it has one. Then the
source's clone method is called with the destination as the PMC
passed in.
So, what about the ramifications of:
$foo = @bar
@bar's get_integer method is called, and $foo's set_integer method is
called. No temp PMCs.
$foo = %bar
calls %bar's get_string method and $foo's set_string method.
$foo = keys %bar (or whatever we're using in perl 6 for this)
calls %bar's get_integer method and $foo's set_integer method.
The only issue we really have is how do we get a scalar PMC for an
aggregate PMC. Working on that.
--
Dan
--------------------------------------"it's like this"-------------------
Dan Sugalski even samurai
[EMAIL PROTECTED] have teddy bears and even
teddy bears get drunk
