Make writing user-defined character properties nicer.

p4raw-id: //depot/perl@16054

diff --git a/lib/utf8_heavy.pl b/lib/utf8_heavy.pl
index 29d4ac2..d5a0b33 100644 (file)
--- a/lib/utf8_heavy.pl
+++ b/lib/utf8_heavy.pl
@@ -184,7 +184,13 @@ sub SWASHNEW {
            print STDERR "$1 => $2\n" if DEBUG;
            if ($char =~ /[-+!]/) {
                my ($c,$t) = split(/::/, $name, 2);     # bogus use of ::, really
-               my $subobj = $c->SWASHNEW($t, "", 0, 0, 0);
+               my $subobj;
+               if ($c eq 'utf8') {
+                   $subobj = $c->SWASHNEW($t, "", 0, 0, 0);
+               }
+               elsif ($c =~ /^([0-9a-fA-F]+)/) {
+                   $subobj = utf8->SWASHNEW("", $c, 0, 0, 0);
+               }
                return $subobj unless ref $subobj;
                push @extras, $name => $subobj;
                $bits = $subobj->{BITS} if $bits < $subobj->{BITS};

diff --git a/pod/perlunicode.pod b/pod/perlunicode.pod
index 033c9ac..17e070c 100644 (file)
--- a/pod/perlunicode.pod
+++ b/pod/perlunicode.pod
@@ -631,22 +631,28 @@ newline-separated lines.  Each line must be one of the following:
 =item *
 
 Two hexadecimal numbers separated by a tabulator denoting a range
-of Unicode codepoints.
+of Unicode codepoints to include.
 
 =item *
 
-An existing character property prefixed by "+utf8::" to include
-all the characters in that property.
+Something to include, prefixed by "+": either an built-in character
+property (prefixed by "utf8::"), for all the characters in that
+property; or two hexadecimal codepoints for a range; or a single
+hexadecimal codepoint.
 
 =item *
 
-An existing character property prefixed by "-utf8::" to exclude
-all the characters in that property.
+Something to exclude, prefixed by "-": either an existing character
+property (prefixed by "utf8::"), for all the characters in that
+property; or two hexadecimal codepoints for a range; or a single
+hexadecimal codepoint.
 
 =item *
 
-An existing character property prefixed by "!utf8::" to include
-all except the characters in that property.
+Something to negate, prefixed "!": either an existing character
+property (prefixed by "utf8::") for all the characters except the
+characters in the property; or two hexadecimal codepoints for a range;
+or a single hexadecimal codepoint.
 
 =back
 

diff --git a/t/op/pat.t b/t/op/pat.t
index 905204b..5681d6a 100755 (executable)
--- a/t/op/pat.t
+++ b/t/op/pat.t
@@ -6,7 +6,7 @@
 
 $| = 1;
 
-print "1..908\n";
+print "1..910\n";
 
 BEGIN {
     chdir 't' if -d 't';
@@ -2870,3 +2870,17 @@ print "\x{3040}" =~ /\P{InKana3}/ ? "ok $test\n" : "not ok $test\n"; $test++;
 print "\x{3040}" =~ /\p{InNotKana}/ ? "ok $test\n" : "not ok $test\n"; $test++;
 print "\x{3041}" =~ /\P{InNotKana}/ ? "ok $test\n" : "not ok $test\n"; $test++;
 
+sub InConsonant { # Not EBCDIC-aware.
+    return <<EOF;
+0061   007f
+-0061
+-0065
+-0069
+-006f
+-0075
+EOF
+}
+
+print "d" =~ /\p{InConsonant}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+print "e" =~ /\P{InConsonant}/ ? "ok $test\n" : "not ok $test\n"; $test++;
+