On Tue, May 22, 2007 at 01:25:33PM +0100, Nicholas Clark wrote:
>
> And how often does the type of a PMC change, such that its internal
> data layout changes? In Perl 5 this morphing happens everywhere,
> but in Parrot?
Actually, until/unless we have a "scalar container" or "reference"
PMC of some sort, it appears to occur a lot. The example I
keep coming back to is something like (Perl 6):
my @a = (1, 2, 3);
my $b := @a[2];
@a[2] = foo();
In PIR this becomes something like (simplified):
## my @a = (1, 2, 3);
$P1 = '_buildlist'(1, 2, 3) # create a list
.lex '@a', $P1 # bind as @a
## my $b := @a[2];
find_lex $P2, '@a' # look up @a
set $P3, $P2[2] # get reference to @a[2]
.lex '$b', $P3 # bind as $b
## @a[2] = foo();
$P4 = 'foo'() # $P4 could be any type
find_lex $P5, '@a' # look up @a
set $P6, $P5[2] # get reference to @a[2]
assign $P6, $P4 # $P6 (Integer) has to morph
# to whatever type $P4 is
If we try to use a set opcode instead, as in:
## @a[2] = foo();
$P4 = 'foo'() # $P4 could be any type
find_lex $P5, '@a' # look up @a
set $P5[2], $P4 # set @a[2] to $P4
then we end up losing the binding between $b and @a[2] that
was established earlier ($b still refers to the Integer).
Pm
