On Tue, Mar 01, 2005 at 09:32:28PM -0800, Larry Wall wrote:
> On Tue, Mar 01, 2005 at 11:06:17PM -0600, Rod Adams wrote:
> : Since the line between rules and subs is already blurring significantly,
> : I want to blur it a little more. I want to write rules which can take
> : parameters.
>
> No problem. That's how the arguments to rules like <before foo> are
> already passed. If I recall, we originally specified three basic forms:
> <foo bar> # bar is pattern
> <foo: bar> # bar is string
> <foo(bar)> # bar is Perl expression
Yes, this is written in A05, although it's often hard to spot and
easy to overlook. They're in the large table under "Metacharacter reform":
<name(expr)> # call rule, passing Perl args
{ .name(expr) } # same thing.
<$var(expr)> # call rule indirectly by name
{ .$var(expr) } # same thing.
<name pat> # call rule, passing regex arg
{ .name(/pat/) } # same thing.
# maybe...
<name: text> # call rule, passing string
{ .name(q<text>) } # same thing.
The argument form of subrules is not currently mentioned in S05.
I've been designing and implementing PGE consistent with the
above syntaxes.
Pm
