sub DESTROY {}
-sub croak { require Carp; Carp::croak(@_) }
-
my %Cache;
+our (%PropertyAlias, %PA_reverse, %PropValueAlias, %PVA_reverse, %PVA_abbr_map);
+
+sub croak { require Carp; Carp::croak(@_) }
+
##
-## "SWASH" == "SWATCH HASH". A "swatch" is a swatch of the Unicode landscape
+## "SWASH" == "SWATCH HASH". A "swatch" is a swatch of the Unicode landscape.
+## It's a data structure that encodes a set of Unicode characters.
##
sub SWASHNEW {
## ranges.
##
## To make the parsing of $type clear, this code takes the a rather
- ## unorthadox approach of last'ing out of the block once we have the
+ ## unorthodox approach of last'ing out of the block once we have the
## info we need. Were this to be a subroutine, the 'last' would just
## be a 'return'.
##
GETFILE:
{
- ##
- ## 'Is' is always optional, so if it's there, remove it.
- ## Same with 'Category=' and 'Script='.
- ##
- ## 'Block=' is replaced by 'In'.
- ##
- $type =~ s/^Is(?:\s+|[-_])?//i
+ ##
+ ## It could be a user-defined property.
+ ##
+
+ my $caller1 = $type =~ s/(.+)::// ? $1 : caller(1);
+
+ if (defined $caller1 && $type =~ /^(?:\w+)$/) {
+ my $prop = "${caller1}::$type";
+ if (exists &{$prop}) {
+ no strict 'refs';
+
+ $list = &{$prop};
+ last GETFILE;
+ }
+ }
+
+ my $wasIs;
+
+ ($wasIs = $type =~ s/^Is(?:\s+|[-_])?//i)
or
- $type =~ s/^Category\s*=\s*//i
+ $type =~ s/^(?:(?:General(?:\s+|_)?)?Category|gc)\s*[:=]\s*//i
or
- $type =~ s/^Script\s*=\s*//i
+ $type =~ s/^(?:Script|sc)\s*[:=]\s*//i
or
- $type =~ s/^Block\s*=\s*/In/i;
+ $type =~ s/^Block\s*[:=]\s*/In/i;
+
+
+ ##
+ ## See if it's in some enumeration.
+ ##
+ require "unicore/PVA.pl";
+ if ($type =~ /^([\w\s]+)[:=]\s*(.*)/) {
+ my ($enum, $val) = (lc $1, lc $2);
+ $enum =~ tr/ _-//d;
+ $val =~ tr/ _-//d;
+
+ my $pa = $PropertyAlias{$enum} ? $enum : $PA_reverse{$enum};
+ my $f = $PropValueAlias{$pa}{$val} ? $val : $PVA_reverse{$pa}{lc $val};
+
+ if ($pa and $f) {
+ $pa = "gc_sc" if $pa eq "gc" or $pa eq "sc";
+ $file = "unicore/lib/$pa/$PVA_abbr_map{$pa}{lc $f}.pl";
+ last GETFILE;
+ }
+ }
+ else {
+ my $t = lc $type;
+ $t =~ tr/ _-//d;
+
+ if ($PropValueAlias{gc}{$t} or $PropValueAlias{sc}{$t}) {
+ $file = "unicore/lib/gc_sc/$PVA_abbr_map{gc_sc}{$t}.pl";
+ last GETFILE;
+ }
+ }
##
## See if it's in the direct mapping table.
##
require "unicore/Exact.pl";
if (my $base = $utf8::Exact{$type}) {
- $file = "unicore/lib/$base.pl";
+ $file = "unicore/lib/gc_sc/$base.pl";
last GETFILE;
}
print "canonical = $canonical\n" if DEBUG;
require "unicore/Canonical.pl";
- if (my $base = $utf8::Canonical{$canonical}) {
- $file = "unicore/lib/$base.pl";
+ if (my $base = ($utf8::Canonical{$canonical} || $utf8::Canonical{ lc $utf8::PropertyAlias{$canonical} })) {
+ $file = "unicore/lib/gc_sc/$base.pl";
last GETFILE;
}
##
- ## It could be a user-defined property.
+ ## See if it's a user-level "To".
##
- if ($type =~ /^I[ns](\w+)$/) {
- my @caller = caller(1);
-
- if (defined $caller[0]) {
- my $prop = $caller[0] . "::" . $type;
+ my $caller0 = caller(0);
- if (exists &{$prop}) {
- no strict 'refs';
+ if (defined $caller0 && $type =~ /^To(?:\w+)$/) {
+ my $map = $caller0 . "::" . $type;
- $list = &{$prop};
- last GETFILE;
- }
- }
- }
+ if (exists &{$map}) {
+ no strict 'refs';
+
+ $list = &{$map};
+ last GETFILE;
+ }
+ }
##
- ## Last attempt -- see if it's a "To" name (e.g. "ToLower")
+ ## Last attempt -- see if it's a standard "To" name
+ ## (e.g. "ToLower") ToTitle is used by ucfirst().
+ ## The user-level way to access ToDigit() and ToFold()
+ ## is to use Unicode::UCD.
##
- if ($type =~ /^To([A-Z][A-Za-z]+)$/)
+ if ($type =~ /^To(Digit|Fold|Lower|Title|Upper)$/)
{
$file = "unicore/To/$1.pl";
## would like to test to see if $file actually exists....
##
## If we reach here, it was due to a 'last GETFILE' above
- ## (exception: user-defined properties), so we
+ ## (exception: user-defined properties and mappings), so we
## have a filename, so now we load it if we haven't already.
## If we have, return the cached results. The cache key is the
## file to load.
}
my $extras;
- my $bits;
+ my $bits = 0;
my $ORIG = $list;
if ($list) {
no warnings;
$extras = join '', grep /^[^0-9a-fA-F]/, @tmp;
$list = join '',
- sort { hex $a <=> hex $b }
- grep {/^([0-9a-fA-F]+)/ and not $seen{$1}++} @tmp; # XXX doesn't do ranges right
+ map { $_->[1] }
+ sort { $a->[0] <=> $b->[0] }
+ map { /^([0-9a-fA-F]+)/; [ hex($1), $_ ] }
+ grep { /^([0-9a-fA-F]+)/ and not $seen{$1}++ } @tmp; # XXX doesn't do ranges right
}
if ($none) {
if ($minbits < 32) {
my $top = 0;
- while ($list =~ /^([0-9a-fA-F]+)(?:\t([0-9a-fA-F]+)?)(?:\t([0-9a-fA-F]+))?/mg) {
+ while ($list =~ /^([0-9a-fA-F]+)(?:[\t]([0-9a-fA-F]+)?)(?:[ \t]([0-9a-fA-F]+))?/mg) {
my $min = hex $1;
- my $max = hex(defined $2 ? $2 : $1);
- my $val = hex(defined $3 ? $3 : "");
+ my $max = defined $2 ? hex $2 : $min;
+ my $val = defined $3 ? hex $3 : 0;
$val += $max - $min if defined $3;
$top = $val if $val > $top;
}
my $char = $1;
my $name = $2;
print STDERR "$1 => $2\n" if DEBUG;
- if ($char =~ /[-+!]/) {
+ if ($char =~ /[-+!&]/) {
my ($c,$t) = split(/::/, $name, 2); # bogus use of ::, really
my $subobj;
if ($c eq 'utf8') {
- $subobj = $c->SWASHNEW($t, "", 0, 0, 0);
+ $subobj = utf8->SWASHNEW($t, "", 0, 0, 0);
+ }
+ elsif (exists &$name) {
+ $subobj = utf8->SWASHNEW($name, "", 0, 0, 0);
}
elsif ($c =~ /^([0-9a-fA-F]+)/) {
$subobj = utf8->SWASHNEW("", $c, 0, 0, 0);
pos $_ = 0;
if ($bits > 1) {
LINE:
- while (/^([0-9a-fA-F]+)(?:\t([0-9a-fA-F]+)?)(?:\t([0-9a-fA-F]+))?/mg) {
- my $min = hex $1;
- my $max = (defined $2 ? hex $2 : $min);
- my $val = hex $3;
+ while (/^([0-9a-fA-F]+)(?:[ \t]([0-9a-fA-F]+)?)?(?:[ \t]([0-9a-fA-F]+))?/mg) {
+ chomp;
+ my ($a, $b, $c) = ($1, $2, $3);
+ croak "$type: illegal mapping '$_'"
+ if $type =~ /^To/ &&
+ !(defined $a && defined $c);
+ my $min = hex $a;
+ my $max = defined $b ? hex $b : $min;
+ my $val = defined $c ? hex $c : 0;
next if $max < $start;
print "$min $max $val\n" if DEBUG;
if ($none) {
else {
LINE:
while (/^([0-9a-fA-F]+)(?:[ \t]+([0-9a-fA-F]+))?/mg) {
+ chomp;
my $min = hex $1;
- my $max = (defined $2 ? hex $2 : $min);
+ my $max = defined $2 ? hex $2 : $min;
next if $max < $start;
if ($min < $start) {
$min = $start;
}
for my $x ($self->{EXTRAS}) {
pos $x = 0;
- while ($x =~ /^([-+!])(.*)/mg) {
+ while ($x =~ /^([-+!&])(.*)/mg) {
my $char = $1;
my $name = $2;
print STDERR "INDIRECT $1 $2\n" if DEBUG;
}
}
}
+ elsif ($char eq '&') {
+ if ($bits == 1 and $otherbits == 1) {
+ $swatch &= $other;
+ }
+ else {
+ for ($key = 0; $key < $len; $key++) {
+ if (!vec($other, $key, $otherbits)) {
+ vec($swatch, $key, $bits) = 0;
+ }
+ }
+ }
+ }
}
}
if (DEBUG) {
https://perl5.git.perl.org / perl5.git / blobdiff