git.shiar.nl
/
sheet.git
/ blobdiff
commit
grep
author
committer
pickaxe
?
search:
re
summary
|
shortlog
|
log
|
commit
|
commitdiff
|
tree
raw
|
inline
| side by side
keys: move key definitions into translation includes
[sheet.git]
/
tools
/
mkcharinfo
diff --git
a/tools/mkcharinfo
b/tools/mkcharinfo
index 27c11d93bcf53b34c50fdc3d895900cc1579c16c..64738ae80acc41c11b135b412ff0afdd940f377e 100755
(executable)
--- a/
tools/mkcharinfo
+++ b/
tools/mkcharinfo
@@
-1,6
+1,5
@@
#!/usr/bin/env perl
#!/usr/bin/env perl
-use 5.010;
-use strict;
+use 5.014;
use warnings;
use utf8;
no if $] >= 5.018, warnings => 'experimental::smartmatch';
use warnings;
use utf8;
no if $] >= 5.018, warnings => 'experimental::smartmatch';
@@
-8,12
+7,16
@@
no if $] >= 5.018, warnings => 'experimental::smartmatch';
use open OUT => ':utf8', ':std';
use Data::Dump 'pp';
use open OUT => ':utf8', ':std';
use Data::Dump 'pp';
-our $VERSION = '1.0
0
';
+our $VERSION = '1.0
2
';
my %info = (
my %info = (
+ # prepare presentational string for some control(lish) entries
"\xAD" => {string => '-'},
"\x{200E}" => {string => '→'},
"\x{200F}" => {string => '←'},
"\xAD" => {string => '-'},
"\x{200E}" => {string => '→'},
"\x{200F}" => {string => '←'},
+ "\x{200B}" => {string => '␣'}, # nbsp: ~ in TeX
+ "\x{200C}" => {string => '|'}, # ISO-9995-7-081 lookalike (alt: ∣ ⊺ ⟙)
+ "\x{200D}" => {string => '⁀'}, # join (alt: ∤ |ͯ ⨝)
);
$info{chr $_} //= {} for 32 .. 126;
);
$info{chr $_} //= {} for 32 .. 126;
@@
-43,19
+46,17
@@
eval {
} or warn "Failed importing html entities: $@";
my %diinc = (
} or warn "Failed importing html entities: $@";
my %diinc = (
- 'digraphs.inc.pl' => 'u-di',
+ 'data/digraphs-rfc.inc.pl' => 'u-di',
+ 'data/digraphs-shiar.inc.pl' => 'u-prop',
+ 'data/digraphs-vim.inc.pl' => 'u-vim',
);
);
-for (keys %diinc) {
+for (
sort
keys %diinc) {
-e $_ or next;
my $di = do $_ or die "Error reading digraphs file $_: ", $@ || $!;
-e $_ or next;
my $di = do $_ or die "Error reading digraphs file $_: ", $@ || $!;
- while (my ($mnem, $cp) = each %$di) {
+ for my $mnem (sort keys %{$di}) {
+ my $cp = $di->{$mnem};
length $mnem == 2 or next; # limit to digraphs
my $class = $diinc{$_};
length $mnem == 2 or next; # limit to digraphs
my $class = $diinc{$_};
- if (ref $cp) {
- # old style array
- $class = 'u-prop' if $cp->[2] and $cp->[2] =~ m/\bXz\b/;
- $cp = chr $cp->[0];
- }
$info{$cp}->{di} //= $mnem;
$info{$cp}->{class}->{$class}++;
}
$info{$cp}->{di} //= $mnem;
$info{$cp}->{class}->{$class}++;
}
@@
-63,7
+64,7
@@
for (keys %diinc) {
eval {
# read introducing unicode versions for known characters
eval {
# read introducing unicode versions for known characters
- my $agemap = do 'unicode-age.inc.pl' or die $@ || $!;
+ my $agemap = do '
data/
unicode-age.inc.pl' or die $@ || $!;
for my $chr (keys %info) {
my $version = $agemap->{ord $chr} or next;
$info{$chr}->{class}->{'u-v'.$version}++
for my $chr (keys %info) {
my $version = $agemap->{ord $chr} or next;
$info{$chr}->{class}->{'u-v'.$version}++
@@
-82,6
+83,10
@@
for my $chr (keys %info) {
$info->{$_} = $info{$chr}->{$_} for keys %{ $info{$chr} };
$info->{$_} = $info{$chr}->{$_} for keys %{ $info{$chr} };
+ # ignore vim flag in addition to rfc support, replace otherwise
+ $info->{class}->{'u-di'} or $info->{class}->{'u-prop'}++
+ if delete $info->{class}->{'u-vim'};
+
# categorise by unicode types and writing script
$info->{class}->{$_}++ for $info->{category};
$info->{class}->{$_}++ for $info->{script} || ();
# categorise by unicode types and writing script
$info->{class}->{$_}++ for $info->{category};
$info->{class}->{$_}++ for $info->{script} || ();
@@
-90,15
+95,15
@@
for my $chr (keys %info) {
$info->{class}->{Xa}++ if $info->{block} eq 'Basic Latin';
$info->{class}->{Xl}++ if $info->{block} eq 'Latin-1 Supplement';
$info->{class}->{Xa}++ if $info->{block} eq 'Basic Latin';
$info->{class}->{Xl}++ if $info->{block} eq 'Latin-1 Supplement';
-
given ($cp)
{
-
when
($info->{string}) {
+ {
+
if
($info->{string}) {
# keep predefined presentational string
}
# keep predefined presentational string
}
-
when
($info->{combining}) {
+
elsif
($info->{combining}) {
# overlay combining accents
$info->{string} = chr(9676) . $chr;
}
# overlay combining accents
$info->{string} = chr(9676) . $chr;
}
-
when
(($cp & ~0b1001_1111) == 0 or $cp == 127) {
+
elsif
(($cp & ~0b1001_1111) == 0 or $cp == 127) {
# control characters (first 32 chars from 0 and 128)
# rename to something more descriptive
$info->{name} = $info->{unicode10}
# control characters (first 32 chars from 0 and 128)
# rename to something more descriptive
$info->{name} = $info->{unicode10}
@@
-115,6
+120,7
@@
for my $chr (keys %info) {
}
# output perl code of hash
}
# output perl code of hash
+say "# automatically generated by $0";
say 'use utf8;';
say '+{';
for my $cp (sort keys %info) {
say 'use utf8;';
say '+{';
for my $cp (sort keys %info) {