Author: lwall
Date: 2009-05-28 18:55:52 +0200 (Thu, 28 May 2009)
New Revision: 26953
Modified:
docs/Perl6/Spec/S05-regex.pod
Log:
[S05] document use of #= tags in {*} actions
Modified: docs/Perl6/Spec/S05-regex.pod
===================================================================
--- docs/Perl6/Spec/S05-regex.pod 2009-05-28 07:40:29 UTC (rev 26952)
+++ docs/Perl6/Spec/S05-regex.pod 2009-05-28 16:55:52 UTC (rev 26953)
@@ -3723,21 +3723,33 @@
Whenever a closure within the grammar returns a C<Whatever> object, the
grammar engine tries to call a method of the same name as the name of the
-current regex on the action object, passing along the current Match object as
-the first positional argument.
+current regex on the action object, passing along the current C<Match> object
as
+the first positional argument, and the tag of the reduction (if any) as the
second
+argument. The tag is supplied via a C<#=> comment on the same line as the
C<{*}>.
grammar Integer {
token TOP {
- \d+ {*}
+ | 0b<[01]>+ {*} #= binary
+ | \d+ {*} #= decimal
}
}
class Twice {
- method TOP($/) {
- make 2 * $/;
+ multi method TOP($match, $tag) {
+ my $text = ~$match;
+ $text = :2($text) if $tag eq 'binary'
+ make $text;
}
+ multi method TOP($match) {
+ make 2 * $match.ast;
+ }
}
Integer.parse('21', :action(Twice.new)).ast # 42
+A C<{*}> is assumed at the end of every rule, and the method is
+called with no tag argument. Note that the implicit C<{*}> is
+I<outside> the alternation in the C<TOP> rule above, despite
+the fact that no explicit square brackets were used.
+
=back
=head1 Syntactic categories
