fix warning + carp interaction

[p5sagit/p5-mst-13.2.git] / lib / charnames.pm

diff --git a/lib/charnames.pm b/lib/charnames.pm
index ec200ec..0241534 100644 (file)
--- a/lib/charnames.pm
+++ b/lib/charnames.pm
@@ -55,7 +55,10 @@ sub charnames
   }
 
   ## If we don't have it by now, give up.
-  die "Unknown charname '$name'" unless @off;
+  unless (@off) {
+      carp "Unknown charname '$name'";
+      return "\x{FFFD}";
+  }
 
   ##
   ## Now know where in the string the name starts.
@@ -78,10 +81,11 @@ sub charnames
   if ($^H & $bytes::hint_bits) {       # "use bytes" in effect?
     use bytes;
     return chr $ord if $ord <= 255;
-    my $hex = sprintf '%X=0%o', $ord, $ord;
+    my $hex = sprintf "%04x", $ord;
     my $fname = substr $txt, $off[0] + 2, $off[1] - $off[0] - 2;
-    die "Character 0x$hex with name '$fname' is above 0xFF";
+    croak "Character 0x$hex with name '$fname' is above 0xFF";
   }
+
   return pack "U", $ord;
 }
 
@@ -123,6 +127,8 @@ sub import
   }
 }
 
+require Unicode::UCD; # for Unicode::UCD::_getcode()
+
 my %viacode;
 
 sub viacode
@@ -131,16 +137,24 @@ sub viacode
         carp "charnames::viacode() expects one numeric argument";
         return ()
     }
+
     my $arg = shift;
+    my $code = Unicode::UCD::_getcode($arg);
 
     my $hex;
-    if ($arg =~ m/^[0-9]+$/) {
+
+    if (defined $code) {
         $hex = sprintf "%04X", $arg;
     } else {
         carp("unexpected arg \"$arg\" to charnames::viacode()");
         return;
     }
 
+    if ($code > 0x10FFFF) {
+       carp "Unicode characters only allocated up to 0x10FFFF (you asked for $hex)";
+       return "\x{FFFD}";
+    }
+
     return $viacode{$hex} if exists $viacode{$hex};
 
     $txt = do "unicore/Name.pl" unless $txt;
@@ -224,6 +238,13 @@ constant used inside double-quoted strings: in other words, you cannot
 use variables inside the C<\N{...}>.  If you want similar run-time
 functionality, use charnames::vianame().
 
+For the C0 and C1 control characters (U+0000..U+001F, U+0080..U+009F)
+as of Unicode 3.1, there are no official Unicode names but you can
+use instead the ISO 6429 names (LINE FEED, ESCAPE, and so forth).
+In Unicode 3.2 some naming changes will happen since ISO 6429 has been
+updated.  Also note that the U+UU80, U+0081, U+0084, and U+0099
+do not have names even in ISO 6429.
+
 =head1 CUSTOM TRANSLATORS
 
 The mechanism of translation of C<\N{...}> escapes is general and not
@@ -282,6 +303,11 @@ Returns undef if no name is known for the name.
 This works only for the standard names, and does not yet aply 
 to custom translators.
 
+=head1 ILLEGAL CHARACTERS
+
+If you ask for a character that does not exist, a warning is given
+and the special Unicode I<replacement character> "\x{FFFD}" is returned.
+
 =head1 BUGS
 
 Since evaluation of the translation function happens in a middle of