Sweden-Number/tools/make_unicode

2549 lines
78 KiB
Plaintext
Raw Normal View History

#!/usr/bin/perl -w
#
# Generate code page .c files from ftp.unicode.org descriptions
#
# Copyright 2000 Alexandre Julliard
#
# This library is free software; you can redistribute it and/or
# modify it under the terms of the GNU Lesser General Public
# License as published by the Free Software Foundation; either
# version 2.1 of the License, or (at your option) any later version.
#
# This library is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
# Lesser General Public License for more details.
#
# You should have received a copy of the GNU Lesser General Public
# License along with this library; if not, write to the Free Software
# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA
#
use strict;
# base URLs for www.unicode.org files
my $UNIVERSION = "12.1.0";
my $MAPPINGS = "http://www.unicode.org/Public/MAPPINGS";
my $UNIDATA = "http://www.unicode.org/Public/$UNIVERSION/ucd/UCD.zip";
my $IDNADATA = "https://www.unicode.org/Public/idna/$UNIVERSION";
my $REPORTS = "http://www.unicode.org/reports";
my $RFCS = "http://www.rfc-editor.org/rfc";
my $MSDATA = "https://download.microsoft.com/download/C/F/7/CF713A5E-9FBC-4FD6-9246-275F65C0E498";
my $MSCODEPAGES = "$MSDATA/Windows Supported Code Page Data Files.zip";
# Sort keys file
my $SORTKEYS = "tr10/allkeys.txt";
# RFC3454 (stringprep data)
my $STRINGPREP = "rfc3454.txt";
# Default char for undefined mappings
my $DEF_CHAR = ord '?';
# Last valid Unicode character
my $MAX_CHAR = 0x10ffff;
my @allfiles =
(
"CodpageFiles/037.txt",
"CodpageFiles/437.txt",
"CodpageFiles/500.txt",
"CodpageFiles/737.txt",
"CodpageFiles/775.txt",
"CodpageFiles/850.txt",
"CodpageFiles/852.txt",
"CodpageFiles/855.txt",
"CodpageFiles/857.txt",
"CodpageFiles/860.txt",
"CodpageFiles/861.txt",
"CodpageFiles/862.txt",
"CodpageFiles/863.txt",
"CodpageFiles/864.txt",
"CodpageFiles/865.txt",
"CodpageFiles/866.txt",
"CodpageFiles/869.txt",
"CodpageFiles/874.txt",
"CodpageFiles/875.txt",
"CodpageFiles/932.txt",
"CodpageFiles/936.txt",
"CodpageFiles/949.txt",
"CodpageFiles/950.txt",
"CodpageFiles/1026.txt",
"CodpageFiles/1250.txt",
"CodpageFiles/1251.txt",
"CodpageFiles/1252.txt",
"CodpageFiles/1253.txt",
"CodpageFiles/1254.txt",
"CodpageFiles/1255.txt",
"CodpageFiles/1256.txt",
"CodpageFiles/1257.txt",
"CodpageFiles/1258.txt",
"CodpageFiles/1361.txt",
"CodpageFiles/10000.txt",
"CodpageFiles/10001.txt",
"CodpageFiles/10002.txt",
"CodpageFiles/10003.txt",
"CodpageFiles/10004.txt",
"CodpageFiles/10005.txt",
"CodpageFiles/10006.txt",
"CodpageFiles/10007.txt",
"CodpageFiles/10008.txt",
"CodpageFiles/10010.txt",
"CodpageFiles/10017.txt",
"CodpageFiles/10021.txt",
"CodpageFiles/10029.txt",
"CodpageFiles/10079.txt",
"CodpageFiles/10081.txt",
"CodpageFiles/10082.txt",
"CodpageFiles/20127.txt",
"CodpageFiles/20866.txt",
"CodpageFiles/21866.txt",
"CodpageFiles/28591.txt",
"CodpageFiles/28592.txt",
"CodpageFiles/28593.txt",
"CodpageFiles/28594.txt",
"CodpageFiles/28595.txt",
"CodpageFiles/28596.txt",
"CodpageFiles/28597.txt",
"CodpageFiles/28598.txt",
"CodpageFiles/28599.txt",
"CodpageFiles/28603.txt",
"CodpageFiles/28605.txt",
);
2000-08-11 22:53:40 +02:00
my %ctype =
2000-08-11 22:53:40 +02:00
(
"upper" => 0x0001,
"lower" => 0x0002,
"digit" => 0x0004,
"space" => 0x0008,
"punct" => 0x0010,
"cntrl" => 0x0020,
"blank" => 0x0040,
"xdigit" => 0x0080,
"alpha" => 0x0100,
"defin" => 0x0200
2000-08-11 22:53:40 +02:00
);
my %bracket_types =
(
"o" => 0x0000,
"c" => 0x0001,
);
my %indic_types =
(
"Other" => 0x0000,
"Bindu" => 0x0001,
"Visarga" => 0x0002,
"Avagraha" => 0x0003,
"Nukta" => 0x0004,
"Virama" => 0x0005,
"Vowel_Independent" => 0x0006,
"Vowel_Dependent" => 0x0007,
"Vowel" => 0x0008,
"Consonant_Placeholder" => 0x0009,
"Consonant" => 0x000a,
"Consonant_Dead" => 0x000b,
"Consonant_Succeeding_Repha" => 0x000c,
"Consonant_Subjoined" => 0x000d,
"Consonant_Medial" => 0x000e,
"Consonant_Final" => 0x000f,
"Consonant_Head_Letter" => 0x0010,
"Modifying_Letter" => 0x0011,
"Tone_Letter" => 0x0012,
"Tone_Mark" => 0x0013,
"Register_Shifter" => 0x0014,
"Consonant_Preceding_Repha" => 0x0015,
"Pure_Killer" => 0x0016,
"Invisible_Stacker" => 0x0017,
"Gemination_Mark" => 0x0018,
"Cantillation_Mark" => 0x0019,
"Non_Joiner" => 0x001a,
"Joiner" => 0x001b,
"Number_Joiner" => 0x001c,
"Number" => 0x001d,
2015-10-01 20:55:56 +02:00
"Brahmi_Joining_Number" => 0x001e,
"Consonant_With_Stacker" => 0x001f,
"Consonant_Prefixed" => 0x0020,
"Syllable_Modifier" => 0x0021,
"Consonant_Killer" => 0x0022,
"Consonant_Initial_Postfixed" => 0x0023,
);
my %matra_types =
(
"Right" => 0x01,
"Left" => 0x02,
"Visual_Order_Left" => 0x03,
"Left_And_Right" => 0x04,
"Top" => 0x05,
"Bottom" => 0x06,
"Top_And_Bottom" => 0x07,
"Top_And_Right" => 0x08,
"Top_And_Left" => 0x09,
"Top_And_Left_And_Right" => 0x0a,
"Bottom_And_Right" => 0x0b,
"Top_And_Bottom_And_Right" => 0x0c,
"Overstruck" => 0x0d,
"Invisible" => 0x0e,
"Bottom_And_Left" => 0x0f
);
my %nameprep_flags =
(
"unassigned" => 0x01,
"prohibited" => 0x02,
"bidi_ral" => 0x04,
"bidi_l" => 0x08
);
my %break_types =
(
"BK" => 0x0001,
"CR" => 0x0002,
"LF" => 0x0003,
"CM" => 0x0004,
"SG" => 0x0005,
"GL" => 0x0006,
"CB" => 0x0007,
"SP" => 0x0008,
"ZW" => 0x0009,
"NL" => 0x000a,
"WJ" => 0x000b,
"JL" => 0x000c,
"JV" => 0x000d,
"JT" => 0x000e,
"H2" => 0x000f,
"H3" => 0x0010,
"XX" => 0x0011,
"OP" => 0x0012,
"CL" => 0x0013,
"CP" => 0x0014,
"QU" => 0x0015,
"NS" => 0x0016,
"EX" => 0x0017,
"SY" => 0x0018,
"IS" => 0x0019,
"PR" => 0x001a,
"PO" => 0x001b,
"NU" => 0x001c,
"AL" => 0x001d,
"ID" => 0x001e,
"IN" => 0x001f,
"HY" => 0x0020,
"BB" => 0x0021,
"BA" => 0x0022,
"SA" => 0x0023,
"AI" => 0x0024,
"B2" => 0x0025,
"HL" => 0x0026,
"CJ" => 0x0027,
"RI" => 0x0028,
"EB" => 0x0029,
"EM" => 0x002a,
"ZWJ" => 0x002b,
);
my %vertical_types =
(
"R" => 0x0000,
"U" => 0x0001,
"Tr" => 0x0002,
"Tu" => 0x0003,
);
my %categories =
2000-08-11 22:53:40 +02:00
(
"Lu" => $ctype{"defin"}|$ctype{"alpha"}|$ctype{"upper"}, # Letter, Uppercase
"Ll" => $ctype{"defin"}|$ctype{"alpha"}|$ctype{"lower"}, # Letter, Lowercase
"Lt" => $ctype{"defin"}|$ctype{"alpha"}|$ctype{"upper"}|$ctype{"lower"}, # Letter, Titlecase
"Mn" => $ctype{"defin"}, # Mark, Non-Spacing
"Mc" => $ctype{"defin"}, # Mark, Spacing Combining
"Me" => $ctype{"defin"}, # Mark, Enclosing
"Nd" => $ctype{"defin"}|$ctype{"digit"}, # Number, Decimal Digit
"Nl" => $ctype{"defin"}|$ctype{"alpha"}, # Number, Letter
"No" => $ctype{"defin"}, # Number, Other
"Zs" => $ctype{"defin"}|$ctype{"space"}, # Separator, Space
"Zl" => $ctype{"defin"}|$ctype{"space"}, # Separator, Line
"Zp" => $ctype{"defin"}|$ctype{"space"}, # Separator, Paragraph
"Cc" => $ctype{"defin"}|$ctype{"cntrl"}, # Other, Control
"Cf" => $ctype{"defin"}|$ctype{"cntrl"}, # Other, Format
"Cs" => $ctype{"defin"}, # Other, Surrogate
"Co" => $ctype{"defin"}, # Other, Private Use
"Cn" => $ctype{"defin"}, # Other, Not Assigned
"Lm" => $ctype{"defin"}|$ctype{"alpha"}, # Letter, Modifier
"Lo" => $ctype{"defin"}|$ctype{"alpha"}, # Letter, Other
"Pc" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Connector
"Pd" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Dash
"Ps" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Open
"Pe" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Close
"Pi" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Initial quote
"Pf" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Final quote
"Po" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Other
"Sm" => $ctype{"defin"}, # Symbol, Math
"Sc" => $ctype{"defin"}, # Symbol, Currency
"Sk" => $ctype{"defin"}, # Symbol, Modifier
"So" => $ctype{"defin"} # Symbol, Other
2000-08-11 22:53:40 +02:00
);
# a few characters need additional categories that cannot be determined automatically
my %special_categories =
2000-08-11 22:53:40 +02:00
(
"xdigit" => [ ord('0')..ord('9'),ord('A')..ord('F'),ord('a')..ord('f'),
0xff10..0xff19, 0xff21..0xff26, 0xff41..0xff46 ],
"space" => [ 0x09..0x0d, 0x85 ],
"blank" => [ 0x09, 0x20, 0xa0, 0x3000, 0xfeff ],
"cntrl" => [ 0x070f, 0x200c, 0x200d,
0x200e, 0x200f, 0x202a, 0x202b, 0x202c, 0x202d, 0x202e,
0x206a, 0x206b, 0x206c, 0x206d, 0x206e, 0x206f, 0xfeff,
0xfff9, 0xfffa, 0xfffb ],
"punct" => [ 0x24, 0x2b, 0x3c..0x3e, 0x5e, 0x60, 0x7c, 0x7e, 0xa2..0xbe,
0xd7, 0xf7 ],
"digit" => [ 0xb2, 0xb3, 0xb9 ],
"lower" => [ 0x2071, 0x207f ]
2000-08-11 22:53:40 +02:00
);
my %directions =
2000-08-11 22:53:40 +02:00
(
"L" => 1, # Left-to-Right
"R" => 2, # Right-to-Left
"AL" => 12, # Right-to-Left Arabic
2000-08-11 22:53:40 +02:00
"EN" => 3, # European Number
"ES" => 4, # European Number Separator
"ET" => 5, # European Number Terminator
"AN" => 6, # Arabic Number
"CS" => 7, # Common Number Separator
"NSM" => 13, # Non-Spacing Mark
"BN" => 14, # Boundary Neutral
2000-08-11 22:53:40 +02:00
"B" => 8, # Paragraph Separator
"S" => 9, # Segment Separator
"WS" => 10, # Whitespace
"ON" => 11, # Other Neutrals
"LRE" => 15, # Left-to-Right Embedding
"LRO" => 15, # Left-to-Right Override
"RLE" => 15, # Right-to-Left Embedding
"RLO" => 15, # Right-to-Left Override
"PDF" => 15, # Pop Directional Format
"LRI" => 15, # Left-to-Right Isolate
"RLI" => 15, # Right-to-Left Isolate
"FSI" => 15, # First Strong Isolate
"PDI" => 15 # Pop Directional Isolate
2000-08-11 22:53:40 +02:00
);
my %c2_types =
(
"L" => 1, # C2_LEFTTORIGHT
"R" => 2, # C2_RIGHTTOLEFT
"AL" => 2, # C2_RIGHTTOLEFT
"EN" => 3, # C2_EUROPENUMBER
"ES" => 4, # C2_EUROPESEPARATOR
"ET" => 5, # C2_EUROPETERMINATOR
"AN" => 6, # C2_ARABICNUMBER
"CS" => 7, # C2_COMMONSEPARATOR
"NSM" => 0, # C2_NOTAPPLICABLE
"BN" => 0, # C2_NOTAPPLICABLE
"B" => 8, # C2_BLOCKSEPARATOR
"S" => 9, # C2_SEGMENTSEPARATOR
"WS" => 10, # C2_WHITESPACE
"ON" => 11, # C2_OTHERNEUTRAL
"LRE" => 11, # C2_OTHERNEUTRAL
"LRO" => 11, # C2_OTHERNEUTRAL
"RLE" => 11, # C2_OTHERNEUTRAL
"RLO" => 11, # C2_OTHERNEUTRAL
"PDF" => 11, # C2_OTHERNEUTRAL
"LRI" => 11, # C2_OTHERNEUTRAL
"RLI" => 11, # C2_OTHERNEUTRAL
"FSI" => 11, # C2_OTHERNEUTRAL
"PDI" => 11 # C2_OTHERNEUTRAL
);
my %bidi_types =
(
"ON" => 0, # Other Neutrals
"L" => 1, # Left-to-Right
"R" => 2, # Right-to-Left
"AN" => 3, # Arabic Number
"EN" => 4, # European Number
"AL" => 5, # Right-to-Left Arabic
"NSM" => 6, # Non-Spacing Mark
"CS" => 7, # Common Number Separator
"ES" => 8, # European Number Separator
"ET" => 9, # European Number Terminator
"BN" => 10, # Boundary Neutral
"S" => 11, # Segment Separator
"WS" => 12, # Whitespace
"B" => 13, # Paragraph Separator
"RLO" => 14, # Right-to-Left Override
"RLE" => 15, # Right-to-Left Embedding
"LRO" => 16, # Left-to-Right Override
"LRE" => 17, # Left-to-Right Embedding
"PDF" => 18, # Pop Directional Format
"LRI" => 19, # Left-to-Right Isolate
"RLI" => 20, # Right-to-Left Isolate
"FSI" => 21, # First Strong Isolate
"PDI" => 22 # Pop Directional Isolate
);
my %joining_types =
(
"U" => 0, # Non_Joining
"T" => 1, # Transparent
"R" => 2, # Right_Joining
"L" => 3, # Left_Joining
"D" => 4, # Dual_Joining
"C" => 5, # Join_Causing
);
my @cp2uni = ();
my @glyph2uni = ();
my @lead_bytes = ();
my @uni2cp = ();
my @tolower_table = ();
my @toupper_table = ();
my @digitmap_table = ();
my @category_table = ();
my @joining_table = ();
my @direction_table = ();
my @decomp_table = ();
my @combining_class_table = ();
my @decomp_compat_table = ();
my @comp_exclusions = ();
my @idna_decomp_table = ();
my @idna_disallowed = ();
my $default_char;
my $default_wchar;
my %joining_forms =
(
"isolated" => [],
"final" => [],
"initial" => [],
"medial" => []
);
sub to_utf16(@)
{
my @ret;
foreach my $ch (@_)
{
if ($ch < 0x10000)
{
push @ret, $ch;
}
else
{
my $val = $ch - 0x10000;
push @ret, 0xd800 | ($val >> 10), 0xdc00 | ($val & 0x3ff);
}
}
return @ret;
}
################################################################
# fetch a unicode.org file and open it
sub open_data_file($$)
{
my ($base, $name) = @_;
(my $dir = "data/$name") =~ s/\/[^\/]+$//;
local *FILE;
unless (-f "data/$name")
{
system "mkdir", "-p", $dir;
if ($base =~ /.*\/([^\/]+\.zip)$/)
{
my $zip = $1;
unless (-f "data/$zip")
{
print "Fetching $base...\n";
!system "wget", "-q", "-O", "data/$zip", "$base" or die "cannot fetch $base";
}
!system "unzip", "-q", "-d", "data", "data/$zip", $name or die "cannot extract $name from $zip";
}
else
{
print "Fetching $base/$name...\n";
!system "wget", "-q", "-O", "data/$name", "$base/$name" or die "cannot fetch $base/$name";
}
}
open FILE, "<data/$name" or die "cannot open data/$name";
return *FILE;
}
################################################################
# recursively get the decomposition for a character
sub get_decomposition($$);
sub get_decomposition($$)
{
my ($char, $table) = @_;
my @ret;
return $char unless defined $table->[$char];
foreach my $ch (@{$table->[$char]})
{
push @ret, get_decomposition( $ch, $table );
}
return @ret;
}
################################################################
# get the composition that results in a given character
sub get_composition($$)
{
my ($ch, $compat) = @_;
return () unless defined $decomp_table[$ch]; # no decomposition
my @ret = @{$decomp_table[$ch]};
return () if @ret < 2; # singleton decomposition
return () if $comp_exclusions[$ch]; # composition exclusion
return () if $combining_class_table[$ch]; # non-starter
return () if $combining_class_table[$ret[0]]; # first char is non-starter
return () if $compat == 1 && !defined $decomp_table[$ret[0]] &&
defined $decomp_compat_table[$ret[0]]; # first char has compat decomposition
return () if $compat == 2 && !defined $decomp_table[$ret[0]] &&
defined $idna_decomp_table[$ret[0]]; # first char has IDNA decomposition
return () if $compat == 2 && defined $idna_decomp_table[$ret[0]] &&
defined $idna_decomp_table[$idna_decomp_table[$ret[0]]->[0]]; # first char's decomposition has IDNA decomposition
return () if $compat == 2 && defined $idna_decomp_table[$ret[1]]; # second char has IDNA decomposition
return @ret;
}
################################################################
# recursively build decompositions
sub build_decompositions(@)
{
my @src = @_;
my @dst;
for (my $i = 0; $i < @src; $i++)
{
next unless defined $src[$i];
my @decomp = to_utf16( get_decomposition( $i, \@src ));
$dst[$i] = \@decomp;
}
return @dst;
}
################################################################
# compose Hangul sequences
sub compose_hangul(@)
{
my $SBASE = 0xac00;
my $LBASE = 0x1100;
my $VBASE = 0x1161;
my $TBASE = 0x11a7;
my $LCOUNT = 19;
my $VCOUNT = 21;
my $TCOUNT = 28;
my $NCOUNT = $VCOUNT * $TCOUNT;
my $SCOUNT = $LCOUNT * $NCOUNT;
my @seq = @_;
my @ret;
my $i;
for ($i = 0; $i < @seq; $i++)
{
my $ch = $seq[$i];
if ($ch >= $LBASE && $ch < $LBASE + $LCOUNT && $i < @seq - 1 &&
$seq[$i+1] >= $VBASE && $seq[$i+1] < $VBASE + $VCOUNT)
{
$ch = $SBASE + (($seq[$i] - $LBASE) * $VCOUNT + ($seq[$i+1] - $VBASE)) * $TCOUNT;
$i++;
}
if ($ch >= $SBASE && $ch < $SBASE + $SCOUNT && !(($ch - $SBASE) % $TCOUNT) && $i < @seq - 1 &&
$seq[$i+1] > $TBASE && $seq[$i+1] < $TBASE + $TCOUNT)
{
$ch += $seq[$i+1] - $TBASE;
$i++;
}
push @ret, $ch;
}
return @ret;
}
################################################################
# read in the Unicode database files
sub load_data()
{
my $start;
# now build mappings from the decomposition field of the Unicode database
my $UNICODE_DATA = open_data_file( $UNIDATA, "UnicodeData.txt" );
while (<$UNICODE_DATA>)
{
# Decode the fields ...
my ($code, $name, $cat, $comb, $bidi,
$decomp, $dec, $dig, $num, $mirror,
$oldname, $comment, $upper, $lower, $title) = split /;/;
my $dst;
2000-08-11 22:53:40 +02:00
my $src = hex $code;
2000-08-11 22:53:40 +02:00
die "unknown category $cat" unless defined $categories{$cat};
die "unknown directionality $bidi" unless defined $directions{$bidi};
$category_table[$src] = $categories{$cat};
$direction_table[$src] = $bidi;
$joining_table[$src] = $joining_types{"T"} if $cat eq "Mn" || $cat eq "Me" || $cat eq "Cf";
2000-08-11 22:53:40 +02:00
if ($lower ne "")
{
$tolower_table[$src] = hex $lower;
}
if ($upper ne "")
{
$toupper_table[$src] = hex $upper;
}
if ($dec ne "")
{
$category_table[$src] |= $ctype{"digit"};
}
if ($dig ne "")
{
$digitmap_table[$src] = ord $dig;
}
$combining_class_table[$src] = ($cat ne "Co") ? $comb : 0x100; # Private Use
2000-08-11 22:53:40 +02:00
# copy the category and direction for everything between First/Last pairs
if ($name =~ /, First>/) { $start = $src; }
if ($name =~ /, Last>/)
{
while ($start < $src)
{
$category_table[$start] = $category_table[$src];
$direction_table[$start] = $direction_table[$src];
$combining_class_table[$start] = $combining_class_table[$src];
2000-08-11 22:53:40 +02:00
$start++;
}
}
2002-06-01 04:55:48 +02:00
2000-08-11 22:53:40 +02:00
next if $decomp eq ""; # no decomposition, skip it
if ($decomp =~ /^<([a-zA-Z]+)>\s+([0-9a-fA-F]+)/)
{
my @seq = map { hex $_; } (split /\s+/, (split /\s+/, $decomp, 2)[1]);
$decomp_compat_table[$src] = \@seq;
}
if ($decomp =~ /^<([a-zA-Z]+)>\s+([0-9a-fA-F]+)$/)
{
# decomposition of the form "<foo> 1234" -> use char if type is known
if ($1 eq "isolated" || $1 eq "final" || $1 eq "initial" || $1 eq "medial")
{
${joining_forms{$1}}[hex $2] = $src;
next;
}
next unless ($1 eq "font" ||
$1 eq "noBreak" ||
$1 eq "circle" ||
$1 eq "super" ||
$1 eq "sub" ||
$1 eq "wide" ||
$1 eq "narrow" ||
$1 eq "compat" ||
$1 eq "small");
$dst = hex $2;
}
elsif ($decomp =~ /^<compat>\s+0020\s+([0-9a-fA-F]+)/)
{
# decomposition "<compat> 0020 1234" -> combining accent
$dst = hex $1;
}
elsif ($decomp =~ /^([0-9a-fA-F]+)/)
{
# decomposition contains only char values without prefix -> use first char
$dst = hex $1;
$category_table[$src] |= $category_table[$dst] if defined $category_table[$dst];
# store decomposition if it contains two chars
if ($decomp =~ /^([0-9a-fA-F]+)\s+([0-9a-fA-F]+)$/)
{
$decomp_table[$src] = $decomp_compat_table[$src] = [ hex $1, hex $2 ];
}
elsif ($decomp =~ /^([0-9a-fA-F]+)$/)
{
# Single char decomposition
$decomp_table[$src] = $decomp_compat_table[$src] = [ hex $1 ];
}
}
else
{
next;
}
}
close $UNICODE_DATA;
2000-08-11 22:53:40 +02:00
# patch the category of some special characters
foreach my $cat (keys %special_categories)
2000-08-11 22:53:40 +02:00
{
my $flag = $ctype{$cat};
foreach my $i (@{$special_categories{$cat}}) { $category_table[$i] |= $flag; }
2000-08-11 22:53:40 +02:00
}
# remove case mappings that don't round-trip
for (my $i = 0; $i < @toupper_table; $i++)
{
next unless defined $toupper_table[$i];
my $ch = $toupper_table[$i];
$toupper_table[$i] = undef unless defined $tolower_table[$ch] && $tolower_table[$ch] == $i;
}
for (my $i = 0; $i < @tolower_table; $i++)
{
next unless defined $tolower_table[$i];
my $ch = $tolower_table[$i];
$tolower_table[$i] = undef unless defined $toupper_table[$ch] && $toupper_table[$ch] == $i;
}
# load the composition exclusions
my $EXCL = open_data_file( $UNIDATA, "CompositionExclusions.txt" );
while (<$EXCL>)
{
s/\#.*//; # remove comments
if (/^([0-9a-fA-F]+)\.\.([0-9a-fA-F]+)\s*$/)
{
foreach my $i (hex $1 .. hex $2) { $comp_exclusions[$i] = 1; }
}
elsif (/^([0-9a-fA-F]+)\s*$/)
{
$comp_exclusions[hex $1] = 1;
}
}
close $EXCL;
# load the IDNA mappings
@idna_decomp_table = @decomp_compat_table;
my $IDNA = open_data_file( $IDNADATA, "IdnaMappingTable.txt" );
while (<$IDNA>)
{
s/\#.*//; # remove comments
next if /^\s*$/;
my ($char, $type, $mapping) = split /;/;
my ($ch1, $ch2);
if ($char =~ /([0-9a-fA-F]+)\.\.([0-9a-fA-F]+)/)
{
$ch1 = hex $1;
$ch2 = hex $2;
}
elsif ($char =~ /([0-9a-fA-F]+)/)
{
$ch1 = $ch2 = hex $1;
}
if ($type =~ /mapped/ || $type =~ /deviation/)
{
$mapping =~ s/^\s*(([0-9a-fA-F]+\s+)+)\s*$/$1/;
my @seq = map { hex $_; } split /\s+/, $mapping;
foreach my $i ($ch1 .. $ch2) { $idna_decomp_table[$i] = @seq ? \@seq : [ 0 ]; }
}
elsif ($type =~ /valid/)
{
}
elsif ($type =~ /ignored/)
{
foreach my $i ($ch1 .. $ch2) { $idna_decomp_table[$i] = [ 0 ]; }
}
elsif ($type =~ /disallowed/)
{
foreach my $i ($ch1 .. $ch2)
{
$idna_decomp_table[$i] = undef;
$idna_disallowed[$i] = 1;
}
}
}
close $IDNA;
}
################################################################
# define a new lead byte
sub add_lead_byte($)
{
my $ch = shift;
return if defined $cp2uni[$ch];
push @lead_bytes, $ch;
$cp2uni[$ch] = 0;
}
################################################################
# define a new char mapping
sub add_mapping($$)
{
my ($cp, $uni) = @_;
$cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
$uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
if ($cp > 0xff) { add_lead_byte( $cp >> 8 ); }
}
################################################################
# get a mapping including glyph chars for MB_USEGLYPHCHARS
sub get_glyphs_mapping(@)
{
my @table = @_;
for (my $i = 0; $i < @glyph2uni; $i++)
{
$table[$i] = $glyph2uni[$i] if defined $glyph2uni[$i];
}
return @table;
}
################################################################
# build EUC-JP table from the JIS 0208/0212 files
sub dump_eucjp_codepage()
{
@cp2uni = ();
@glyph2uni = ();
@lead_bytes = ();
@uni2cp = ();
$default_char = $DEF_CHAR;
$default_wchar = 0x30fb;
# ASCII chars
foreach my $i (0x00 .. 0x7f) { add_mapping( $i, $i ); }
# lead bytes
foreach my $i (0x8e, 0xa1 .. 0xfe) { add_lead_byte($i); }
# JIS X 0201 right plane
foreach my $i (0xa1 .. 0xdf) { add_mapping( 0x8e00 + $i, 0xfec0 + $i ); }
# undefined chars
foreach my $i (0x80 .. 0x8d, 0x8f .. 0x9f) { $cp2uni[$i] = $i; }
$cp2uni[0xa0] = 0xf8f0;
$cp2uni[0xff] = 0xf8f3;
# Fix backslash conversion
add_mapping( 0xa1c0, 0xff3c );
# Add private mappings for rows undefined in JIS 0208/0212
my $private = 0xe000;
foreach my $hi (0xf5 .. 0xfe)
{
foreach my $lo (0xa1 .. 0xfe)
{
add_mapping( ($hi << 8) + $lo, $private++ );
}
}
foreach my $hi (0xf5 .. 0xfe)
{
foreach my $lo (0x21 .. 0x7e)
{
add_mapping( ($hi << 8) + $lo, $private++ );
}
}
my $INPUT = open_data_file( $MAPPINGS, "OBSOLETE/EASTASIA/JIS/JIS0208.TXT" );
while (<$INPUT>)
{
next if /^\#/; # skip comments
next if /^$/; # skip empty lines
next if /\x1a/; # skip ^Z
if (/^0x[0-9a-fA-F]+\s+0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\s+(\#.*)?/)
{
add_mapping( 0x8080 + hex $1, hex $2 );
next;
}
die "Unrecognized line $_\n";
}
close $INPUT;
$INPUT = open_data_file( $MAPPINGS, "OBSOLETE/EASTASIA/JIS/JIS0212.TXT" );
while (<$INPUT>)
{
next if /^\#/; # skip comments
next if /^$/; # skip empty lines
next if /\x1a/; # skip ^Z
if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\s+(\#.*)?/)
{
add_mapping( 0x8000 + hex $1, hex $2 );
next;
}
die "Unrecognized line $_\n";
}
close $INPUT;
output_codepage_file( 20932 );
}
################################################################
# build the sort keys table
sub dump_sortkeys($)
{
my $filename = shift;
my @sortkeys = ();
my $INPUT = open_data_file( $REPORTS, $SORTKEYS );
while (<$INPUT>)
{
next if /^\#/; # skip comments
next if /^$/; # skip empty lines
next if /\x1a/; # skip ^Z
next if /^\@version/; # skip @version header
if (/^([0-9a-fA-F]+)\s+;\s+\[([*.])([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]+)\]/)
{
my ($uni,$variable) = (hex $1, $2);
next if $uni > 65535;
$sortkeys[$uni] = [ $uni, hex $3, hex $4, hex $5, hex $6 ];
next;
}
if (/^([0-9a-fA-F]+\s+)+;\s+\[[*.]([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]+)\]/)
{
# multiple character sequence, ignored for now
next;
}
die "$SORTKEYS: Unrecognized line $_\n";
}
close $INPUT;
# compress the keys to 32 bit:
# key 1 to 16 bits, key 2 to 8 bits, key 3 to 4 bits, key 4 to 1 bit
@sortkeys = sort { ${$a}[1] <=> ${$b}[1] or
${$a}[2] <=> ${$b}[2] or
${$a}[3] <=> ${$b}[3] or
${$a}[4] <=> ${$b}[4] or
$a cmp $b; } @sortkeys;
my ($n2, $n3) = (1, 1);
my @keys = (-1, -1, -1, -1, -1 );
my @flatkeys = ();
for (my $i = 0; $i < @sortkeys; $i++)
{
next unless defined $sortkeys[$i];
my @current = @{$sortkeys[$i]};
if ($current[1] == $keys[1])
{
if ($current[2] == $keys[2])
{
if ($current[3] == $keys[3])
{
# nothing
}
else
{
$keys[3] = $current[3];
$n3++;
die if ($n3 >= 16);
}
}
else
{
$keys[2] = $current[2];
$keys[3] = $current[3];
$n2++;
$n3 = 1;
die if ($n2 >= 256);
}
}
else
{
$keys[1] = $current[1];
$keys[2] = $current[2];
$keys[3] = $current[3];
$n2 = 1;
$n3 = 1;
}
if ($current[2]) { $current[2] = $n2; }
if ($current[3]) { $current[3] = $n3; }
if ($current[4]) { $current[4] = 1; }
$flatkeys[$current[0]] = ($current[1] << 16) | ($current[2] << 8) | ($current[3] << 4) | $current[4];
}
open OUTPUT,">$filename.new" or die "Cannot create $filename";
printf "Building $filename\n";
printf OUTPUT "/* Unicode collation element table */\n";
printf OUTPUT "/* generated from %s */\n", "$REPORTS/$SORTKEYS";
printf OUTPUT "/* DO NOT EDIT!! */\n\n";
print OUTPUT "#include \"windef.h\"\n\n";
dump_two_level_mapping( "collation_table", 0xffffffff, 32, @flatkeys );
close OUTPUT;
save_file($filename);
}
################################################################
# dump an array of integers
sub dump_array($$@)
{
my ($bit_width, $default, @array) = @_;
my $format = sprintf "0x%%0%ux", $bit_width / 4;
my $i;
my $ret = " ";
for ($i = 0; $i < $#array; $i++)
{
$ret .= sprintf($format, defined $array[$i] ? $array[$i] : $default);
$ret .= (($i % 8) != 7) ? ", " : ",\n ";
}
$ret .= sprintf($format, defined $array[$i] ? $array[$i] : $default);
return $ret;
}
################################################################
# dump an SBCS mapping table in binary format
sub dump_binary_sbcs_table($)
{
my $codepage = shift;
my @header = ( 13, $codepage, 1, $default_char, $default_wchar, $cp2uni[$default_char], $uni2cp[$default_wchar] );
my $wc_offset = 256 + 3 + (@glyph2uni ? 256 : 0);
print OUTPUT pack "S<*", @header;
print OUTPUT pack "C12", (0) x 12;
print OUTPUT pack "S<*", $wc_offset, map { $_ || 0; } @cp2uni[0 .. 255];
if (@glyph2uni)
{
print OUTPUT pack "S<*", 256, get_glyphs_mapping(@cp2uni[0 .. 255]);
}
else
{
print OUTPUT pack "S<*", 0;
}
print OUTPUT pack "S<*", 0, 0;
print OUTPUT pack "C*", map { defined $_ ? $_ : $default_char; } @uni2cp[0 .. 65535];
}
################################################################
# dump a DBCS mapping table in binary format
sub dump_binary_dbcs_table($)
{
my $codepage = shift;
my @lb_ranges = get_lb_ranges();
my @header = ( 13, $codepage, 2, $default_char, $default_wchar, $cp2uni[$default_char], $uni2cp[$default_wchar] );
my @offsets = (0) x 256;
my $pos = 0;
foreach my $i (@lead_bytes)
{
$offsets[$i] = ($pos += 256);
$cp2uni[$i] = 0;
}
my $wc_offset = 256 + 3 + 256 * (1 + scalar @lead_bytes);
print OUTPUT pack "S<*", @header;
print OUTPUT pack "C12", @lb_ranges, 0 x 12;
print OUTPUT pack "S<*", $wc_offset, map { $_ || 0; } @cp2uni[0 .. 255];
print OUTPUT pack "S<*", 0, scalar @lb_ranges / 2, @offsets;
foreach my $i (@lead_bytes)
{
my $base = $i << 8;
print OUTPUT pack "S<*", map { defined $_ ? $_ : $default_wchar; } @cp2uni[$base .. $base + 255];
}
print OUTPUT pack "S<", 4;
print OUTPUT pack "S<*", map { defined $_ ? $_ : $default_char; } @uni2cp[0 .. 65535];
}
################################################################
# get the list of defined lead byte ranges
sub get_lb_ranges()
{
my @list = ();
my @ranges = ();
foreach my $i (@lead_bytes) { $list[$i] = 1; }
my $on = 0;
for (my $i = 0; $i < 256; $i++)
{
if ($on)
{
if (!defined $list[$i]) { push @ranges, $i-1; $on = 0; }
}
else
{
if ($list[$i]) { push @ranges, $i; $on = 1; }
}
}
if ($on) { push @ranges, 0xff; }
return @ranges;
}
################################################################
# dump the Indic Syllabic Category table
sub dump_indic($)
{
my $filename = shift;
my @indic_table;
my $INPUT = open_data_file( $UNIDATA, "IndicSyllabicCategory.txt" );
while (<$INPUT>)
{
next if /^\#/; # skip comments
next if /^\s*$/; # skip empty lines
next if /\x1a/; # skip ^Z
if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*#/)
{
my $type = $2;
die "unknown indic $type" unless defined $indic_types{$type};
if (hex $1 < 65536)
{
$indic_table[hex $1] = $indic_types{$type};
}
next;
}
elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([A-Za-z_]+)\s*#/)
{
my $type = $3;
die "unknown indic $type" unless defined $indic_types{$type};
if (hex $1 < 65536 and hex $2 < 65536)
{
foreach my $i (hex $1 .. hex $2)
{
$indic_table[$i] = $indic_types{$type};
}
}
next;
}
die "malformed line $_";
}
close $INPUT;
2015-10-01 20:55:56 +02:00
$INPUT = open_data_file( $UNIDATA, "IndicPositionalCategory.txt" );
while (<$INPUT>)
{
next if /^\#/; # skip comments
next if /^\s*$/; # skip empty lines
next if /\x1a/; # skip ^Z
if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z]+)\s*#/)
{
my $type = $2;
die "unknown matra $type" unless defined $matra_types{$type};
$indic_table[hex $1] += $matra_types{$type} << 8;
next;
}
elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([A-Za-z_]+)\s*#/)
{
my $type = $3;
die "unknown matra $type" unless defined $matra_types{$type};
foreach my $i (hex $1 .. hex $2)
{
$indic_table[$i] += $matra_types{$type} << 8;
}
next;
}
die "malformed line $_";
}
close $INPUT;
open OUTPUT,">$filename.new" or die "Cannot create $filename";
print "Building $filename\n";
print OUTPUT "/* Unicode Indic Syllabic Category */\n";
print OUTPUT "/* generated from $UNIDATA:IndicSyllabicCategory.txt */\n";
print OUTPUT "/* and from $UNIDATA:IndicPositionalCategory.txt */\n";
print OUTPUT "/* DO NOT EDIT!! */\n\n";
print OUTPUT "#include \"windef.h\"\n\n";
dump_two_level_mapping( "indic_syllabic_table", $indic_types{'Other'}, 16, @indic_table );
close OUTPUT;
save_file($filename);
}
################################################################
# dump the Line Break Properties table
sub dump_linebreak($)
{
my $filename = shift;
my @break_table;
my $next_group = 0;
my $INPUT = open_data_file( $UNIDATA, "LineBreak.txt" );
while (<$INPUT>)
{
next if /^\#/; # skip comments
next if /^\s*$/; # skip empty lines
next if /\x1a/; # skip ^Z
if (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z][0-9A-Z])+\s*/)
{
my $type = $2;
die "unknown breaktype $type" unless defined $break_types{$type};
$break_table[hex $1] = $break_types{$type};
next;
}
elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z][0-9A-Z])+\s*/)
{
my $type = $3;
die "unknown breaktype $type" unless defined $break_types{$type};
foreach my $i (hex $1 .. hex $2)
{
$break_table[$i] = $break_types{$type};
}
next;
}
elsif (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z])+\s*/)
{
my $type = $2;
die "unknown breaktype $type" unless defined $break_types{$type};
$break_table[hex $1] = $break_types{$type};
next;
}
elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z])+\s*/)
{
my $type = $3;
die "unknown breaktype $type" unless defined $break_types{$type};
foreach my $i (hex $1 .. hex $2)
{
$break_table[$i] = $break_types{$type};
}
next;
}
die "malformed line $_";
}
close $INPUT;
open OUTPUT,">$filename.new" or die "Cannot create $filename";
print "Building $filename\n";
print OUTPUT "/* Unicode Line Break Properties */\n";
print OUTPUT "/* generated from $UNIDATA:LineBreak.txt */\n";
print OUTPUT "/* DO NOT EDIT!! */\n\n";
print OUTPUT "#include \"windef.h\"\n\n";
dump_two_level_mapping( "wine_linebreak_table", $break_types{'XX'}, 16, @break_table );
close OUTPUT;
save_file($filename);
}
my %scripts =
(
"Unknown" => 0,
"Common" => 1,
"Inherited" => 2,
"Arabic" => 3,
"Armenian" => 4,
"Avestan" => 5,
"Balinese" => 6,
"Bamum" => 7,
"Batak" => 8,
"Bengali" => 9,
"Bopomofo" => 10,
"Brahmi" => 11,
"Braille" => 12,
"Buginese" => 13,
"Buhid" => 14,
"Canadian_Aboriginal" => 15,
"Carian" => 16,
"Cham" => 17,
"Cherokee" => 18,
"Coptic" => 19,
"Cuneiform" => 20,
"Cypriot" => 21,
"Cyrillic" => 22,
"Deseret" => 23,
"Devanagari" => 24,
"Egyptian_Hieroglyphs" => 25,
"Ethiopic" => 26,
"Georgian" => 27,
"Glagolitic" => 28,
"Gothic" => 29,
"Greek" => 30,
"Gujarati" => 31,
"Gurmukhi" => 32,
"Han" => 33,
"Hangul" => 34,
"Hanunoo" => 35,
"Hebrew" => 36,
"Hiragana" => 37,
"Imperial_Aramaic" => 38,
"Inscriptional_Pahlavi" => 39,
"Inscriptional_Parthian" => 40,
"Javanese" => 41,
"Kaithi" => 42,
"Kannada" => 43,
"Katakana" => 44,
"Kayah_Li" => 45,
"Kharoshthi" => 46,
"Khmer" => 47,
"Lao" => 48,
"Latin" => 49,
"Lepcha" => 50,
"Limbu" => 51,
"Linear_B" => 52,
"Lisu" => 53,
"Lycian" => 54,
"Lydian" => 55,
"Malayalam" => 56,
"Mandaic" => 57,
"Meetei_Mayek" => 58,
"Mongolian" => 59,
"Myanmar" => 60,
"New_Tai_Lue" => 61,
"Nko" => 62,
"Ogham" => 63,
"Ol_Chiki" => 64,
"Old_Italic" => 65,
"Old_Persian" => 66,
"Old_South_Arabian" => 67,
"Old_Turkic" => 68,
"Oriya" => 69,
"Osmanya" => 70,
"Phags_Pa" => 71,
"Phoenician" => 72,
"Rejang" => 73,
"Runic" => 74,
"Samaritan" => 75,
"Saurashtra" => 76,
"Shavian" => 77,
"Sinhala" => 78,
"Sundanese" => 79,
"Syloti_Nagri" => 80,
"Syriac" => 81,
"Tagalog" => 82,
"Tagbanwa" => 83,
"Tai_Le" => 84,
"Tai_Tham" => 85,
"Tai_Viet" => 86,
"Tamil" => 87,
"Telugu" => 88,
"Thaana" => 89,
"Thai" => 90,
"Tibetan" => 91,
"Tifinagh" => 92,
"Ugaritic" => 93,
"Vai" => 94,
"Yi" => 95,
# Win8/Win8.1
"Chakma" => 96,
"Meroitic_Cursive" => 97,
"Meroitic_Hieroglyphs" => 98,
"Miao" => 99,
"Sharada" => 100,
"Sora_Sompeng" => 101,
"Takri" => 102,
# Win10
"Bassa_Vah" => 103,
"Caucasian_Albanian" => 104,
"Duployan" => 105,
"Elbasan" => 106,
"Grantha" => 107,
"Khojki" => 108,
"Khudawadi" => 109,
"Linear_A" => 110,
"Mahajani" => 111,
"Manichaean" => 112,
"Mende_Kikakui" => 113,
"Modi" => 114,
"Mro" => 115,
"Nabataean" => 116,
"Old_North_Arabian" => 117,
"Old_Permic" => 118,
"Pahawh_Hmong" => 119,
"Palmyrene" => 120,
"Pau_Cin_Hau" => 121,
"Psalter_Pahlavi" => 122,
"Siddham" => 123,
"Tirhuta" => 124,
"Warang_Citi" => 125,
# Win10 RS1
"Adlam" => 126,
"Ahom" => 127,
"Anatolian_Hieroglyphs" => 128,
"Bhaiksuki" => 129,
"Hatran" => 130,
"Marchen" => 131,
"Multani" => 132,
"Newa" => 133,
"Old_Hungarian" => 134,
"Osage" => 135,
"SignWriting" => 136,
"Tangut" => 137,
# Win10 RS4
"Masaram_Gondi" => 138,
"Nushu" => 139,
"Soyombo" => 140,
"Zanabazar_Square" => 141,
# Win10 1903
"Dogra" => 142,
"Gunjala_Gondi" => 143,
"Hanifi_Rohingya" => 144,
"Makasar" => 145,
"Medefaidrin" => 146,
"Old_Sogdian" => 147,
"Sogdian" => 148,
);
################################################################
# dump Script IDs table
sub dump_scripts($)
{
my $filename = shift;
my $header = $filename;
my @scripts_table;
my $script_index;
my $i;
my $INPUT = open_data_file( $UNIDATA, "Scripts.txt" );
# Fill the table
# Unknown script id is always 0, so undefined scripts are automatically treated as such
while (<$INPUT>)
{
my $type = "";
next if /^\#/; # skip comments
next if /^\s*$/; # skip empty lines
next if /\x1a/; # skip ^Z
if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
{
$type = $2;
if (defined $scripts{$type})
{
$scripts_table[hex $1] = $scripts{$type};
}
next;
}
elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
{
$type = $3;
if (defined $scripts{$type})
{
foreach my $i (hex $1 .. hex $2)
{
$scripts_table[$i] = $scripts{$type};
}
}
next;
}
}
close $INPUT;
$header = "$filename.h";
open OUTPUT,">$header.new" or die "Cannot create $header";
print "Building $header\n";
print OUTPUT "/* Unicode Script IDs */\n";
print OUTPUT "/* generated from $UNIDATA:Scripts.txt */\n";
print OUTPUT "/* DO NOT EDIT!! */\n\n";
print OUTPUT "enum unicode_script_id {\n";
foreach my $script (sort { $scripts{$a} <=> $scripts{$b} } keys %scripts)
{
print OUTPUT " Script_$script = $scripts{$script},\n";
}
print OUTPUT " Script_LastId = ", (scalar keys %scripts) - 1, "\n";
print OUTPUT "};\n";
close OUTPUT;
save_file($header);
$filename = "$filename.c";
open OUTPUT,">$filename.new" or die "Cannot create $header";
print "Building $filename\n";
print OUTPUT "/* Unicode Script IDs */\n";
print OUTPUT "/* generated from $UNIDATA:Scripts.txt */\n";
print OUTPUT "/* DO NOT EDIT!! */\n\n";
print OUTPUT "#include \"windef.h\"\n\n";
dump_two_level_mapping( "wine_scripts_table", 0, 16, @scripts_table );
close OUTPUT;
save_file($filename);
}
################################################################
# dump the BiDi mirroring table
sub dump_mirroring($)
{
my $filename = shift;
my @mirror_table = ();
my $INPUT = open_data_file( $UNIDATA, "BidiMirroring.txt" );
while (<$INPUT>)
{
next if /^\#/; # skip comments
next if /^$/; # skip empty lines
next if /\x1a/; # skip ^Z
if (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9a-fA-F]+)/)
{
$mirror_table[hex $1] = hex $2;
next;
}
die "malformed line $_";
}
close $INPUT;
open OUTPUT,">$filename.new" or die "Cannot create $filename";
print "Building $filename\n";
print OUTPUT "/* Unicode BiDi mirroring */\n";
print OUTPUT "/* generated from $UNIDATA:BidiMirroring.txt */\n";
print OUTPUT "/* DO NOT EDIT!! */\n\n";
print OUTPUT "#include \"windef.h\"\n\n";
dump_two_level_mapping( "wine_mirror_map", 0, 16, @mirror_table );
close OUTPUT;
save_file($filename);
}
################################################################
# dump the Bidi Brackets
sub dump_bracket($)
{
my $filename = shift;
my @bracket_table;
my $INPUT = open_data_file( $UNIDATA, "BidiBrackets.txt" );
while (<$INPUT>)
{
next if /^\#/; # skip comments
next if /^\s*$/; # skip empty lines
next if /\x1a/; # skip ^Z
if (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9a-fA-F]+);\s*([con])/)
{
my $type = $3;
die "unknown bracket $type" unless defined $bracket_types{$type};
die "characters too distant $1 and $2" if abs(hex($2) - hex($1)) >= 128;
$bracket_table[hex $1] = (hex($2) - hex($1)) % 255;
$bracket_table[hex $1] += $bracket_types{$type} << 8;
next;
}
die "malformed line $_";
}
close $INPUT;
open OUTPUT,">$filename.new" or die "Cannot create $filename";
print "Building $filename\n";
print OUTPUT "/* Unicode Bidirectional Bracket table */\n";
print OUTPUT "/* generated from $UNIDATA:BidiBrackets.txt */\n";
print OUTPUT "/* DO NOT EDIT!! */\n\n";
print OUTPUT "#include \"windef.h\"\n\n";
dump_two_level_mapping( "bidi_bracket_table", 0, 16, @bracket_table );
close OUTPUT;
save_file($filename);
}
################################################################
# dump the Arabic shaping table
sub dump_shaping($)
{
my $filename = shift;
my %groups;
my $next_group = 0;
$groups{"No_Joining_Group"} = $next_group++;
my $INPUT = open_data_file( $UNIDATA, "ArabicShaping.txt" );
while (<$INPUT>)
{
next if /^\#/; # skip comments
next if /^\s*$/; # skip empty lines
next if /\x1a/; # skip ^Z
if (/^\s*([0-9a-fA-F]+)\s*;.*;\s*([RLDCUT])\s*;\s*(\w+)/)
{
my $type = $2;
my $group = $3;
$groups{$group} = $next_group++ unless defined $groups{$group};
$joining_table[hex $1] = $joining_types{$type} | ($groups{$group} << 8);
next;
}
die "malformed line $_";
}
close $INPUT;
open OUTPUT,">$filename.new" or die "Cannot create $filename";
print "Building $filename\n";
print OUTPUT "/* Unicode Arabic shaping */\n";
print OUTPUT "/* generated from $UNIDATA:ArabicShaping.txt */\n";
print OUTPUT "/* DO NOT EDIT!! */\n\n";
print OUTPUT "#include \"windef.h\"\n\n";
dump_two_level_mapping( "wine_shaping_table", 0, 16, @joining_table );
print OUTPUT "\nconst unsigned short DECLSPEC_HIDDEN wine_shaping_forms[256][4] =\n{\n";
for (my $i = 0x600; $i <= 0x6ff; $i++)
{
printf OUTPUT " { 0x%04x, 0x%04x, 0x%04x, 0x%04x },\n",
${joining_forms{"isolated"}}[$i] || $i,
${joining_forms{"final"}}[$i] || $i,
${joining_forms{"initial"}}[$i] || $i,
${joining_forms{"medial"}}[$i] || $i;
}
print OUTPUT "};\n";
close OUTPUT;
save_file($filename);
}
################################################################
# dump the Vertical Orientation table
sub dump_vertical($)
{
my $filename = shift;
my @vertical_table;
my $INPUT = open_data_file( $UNIDATA, "VerticalOrientation.txt" );
while (<$INPUT>)
{
next if /^\#/; # skip comments
next if /^\s*$/; # skip empty lines
next if /\x1a/; # skip ^Z
if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
{
my $type = $2;
die "unknown vertical $type" unless defined $vertical_types{$type};
if (hex $1 < 65536)
{
$vertical_table[hex $1] = $vertical_types{$type};
}
next;
}
elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([A-Za-z_]+)\s*/)
{
my $type = $3;
die "unknown vertical $type" unless defined $vertical_types{$type};
foreach my $i (hex $1 .. hex $2)
{
$vertical_table[$i] = $vertical_types{$type};
}
next;
}
die "malformed line $_";
}
close $INPUT;
open OUTPUT,">$filename.new" or die "Cannot create $filename";
print "Building $filename\n";
print OUTPUT "/* Unicode Vertical Orientation */\n";
print OUTPUT "/* generated from $UNIDATA:VerticalOrientation.txt */\n";
print OUTPUT "/* DO NOT EDIT!! */\n\n";
print OUTPUT "#include \"windef.h\"\n\n";
dump_two_level_mapping( "vertical_orientation_table", $vertical_types{'R'}, 16, @vertical_table );
close OUTPUT;
save_file($filename);
}
################################################################
# dump the digit folding tables
sub dump_digit_folding($)
{
my ($filename) = shift;
open OUTPUT,">$filename.new" or die "Cannot create $filename";
print "Building $filename\n";
print OUTPUT "/* Unicode digit folding mappings */\n";
print OUTPUT "/* generated from $UNIDATA:UnicodeData.txt */\n";
print OUTPUT "/* DO NOT EDIT!! */\n\n";
print OUTPUT "#include \"windef.h\"\n\n";
dump_two_level_mapping( "wine_digitmap", 0, 16, @digitmap_table );
close OUTPUT;
save_file($filename);
}
################################################################
# dump the case mapping tables
sub dump_case_mappings($)
{
my $filename = shift;
open OUTPUT,">$filename.new" or die "Cannot create $filename";
print "Building $filename\n";
print OUTPUT "/* Unicode case mappings */\n";
print OUTPUT "/* generated from $UNIDATA:UnicodeData.txt */\n";
print OUTPUT "/* DO NOT EDIT!! */\n\n";
print OUTPUT "#include \"windef.h\"\n\n";
dump_case_table( "wine_casemap_lower", @tolower_table );
print OUTPUT "\n";
dump_case_table( "wine_casemap_upper", @toupper_table );
close OUTPUT;
save_file($filename);
}
################################################################
# dump a case mapping table
sub dump_case_table($@)
{
my ($name,@table) = @_;
for (my $i = 0; $i < 65536; $i++)
{
next unless defined $table[$i];
$table[$i] = ($table[$i] - $i) & 0xffff;
}
my @array = compress_array( 256, 0, @table[0..65535] );
printf OUTPUT "const WCHAR %s[%d] =\n", $name, scalar @array;
printf OUTPUT "{\n /* index */\n";
printf OUTPUT "%s,\n", dump_array( 16, 0, @array[0..255] );
printf OUTPUT " /* data */\n";
printf OUTPUT "%s", dump_array( 16, 0, @array[256..$#array] );
printf OUTPUT "\n};\n";
}
################################################################
# compress a mapping table by removing identical rows
sub compress_array($$@)
{
my $rows = shift;
my $def = shift;
my @table = @_;
my $len = @table / $rows;
my @array;
my $data = "";
# try to merge table rows
for (my $row = 0; $row < $rows; $row++)
{
my $rowtxt = pack "U*", map { defined($_) ? $_ : $def; } @table[($row * $len)..(($row + 1) * $len - 1)];
my $pos = index $data, $rowtxt;
if ($pos == -1)
{
# check if the tail of the data can match the start of the new row
my $first = substr( $rowtxt, 0, 1 );
for (my $i = length($data) - 1; $i > 0; $i--)
{
$pos = index( substr( $data, -$i ), $first );
last if $pos == -1;
$i -= $pos;
next unless substr( $data, -$i ) eq substr( $rowtxt, 0, $i );
substr( $data, -$i ) = "";
last;
}
$pos = length $data;
$data .= $rowtxt;
}
$array[$row] = $rows + $pos;
}
return @array, unpack "U*", $data;
}
################################################################
# dump a simple char -> 16-bit value mapping table
sub dump_simple_mapping($$@)
{
my $name = shift;
my $def = shift;
my @array = compress_array( 256, $def, @_[0..65535] );
printf OUTPUT "const unsigned short %s[%d] =\n{\n", $name, $#array+1;
printf OUTPUT " /* offsets */\n%s,\n", dump_array( 16, 0, @array[0..255] );
printf OUTPUT " /* values */\n%s\n};\n", dump_array( 16, 0, @array[256..$#array] );
}
################################################################
# dump a char -> 16-bit value mapping table using two-level tables
sub dump_two_level_mapping($$@)
{
my $name = shift;
my $def = shift;
my $size = shift;
my $type = $size == 16 ? "unsigned short" : "unsigned int";
my @row_array = compress_array( 4096, $def, @_[0..65535] );
my @array = compress_array( 256, 0, @row_array[0..4095] );
for (my $i = 256; $i < @array; $i++) { $array[$i] += @array - 4096; }
printf OUTPUT "const %s DECLSPEC_HIDDEN %s[%d] =\n{\n", $type, $name, @array + @row_array - 4096;
printf OUTPUT " /* level 1 offsets */\n%s,\n", dump_array( $size, 0, @array[0..255] );
printf OUTPUT " /* level 2 offsets */\n%s,\n", dump_array( $size, 0, @array[256..$#array] );
printf OUTPUT " /* values */\n%s\n};\n", dump_array( $size, 0, @row_array[4096..$#row_array] );
}
################################################################
# dump a char -> value mapping table using three-level tables
sub dump_three_level_mapping($$@)
{
my $name = shift;
my $def = shift;
my $size = shift;
my $type = $size == 16 ? "unsigned short" : "unsigned int";
my $level3 = ($MAX_CHAR + 1) / 16;
my $level2 = $level3 / 16;
my $level1 = $level2 / 16;
my @array3 = compress_array( $level3, $def, @_[0..$MAX_CHAR] );
my @array2 = compress_array( $level2, 0, @array3[0..$level3-1] );
my @array1 = compress_array( $level1, 0, @array2[0..$level2-1] );
for (my $i = $level2; $i < @array2; $i++) { $array2[$i] += @array1 + @array2 - $level2 - $level3; }
for (my $i = $level1; $i < @array1; $i++) { $array1[$i] += @array1 - $level2; }
printf OUTPUT "const %s DECLSPEC_HIDDEN %s[%u] =\n{\n", $type, $name, @array1 + (@array2 - $level2) + (@array3 - $level3);
printf OUTPUT " /* level 1 offsets */\n%s,\n", dump_array( $size, 0, @array1[0..$level1-1] );
printf OUTPUT " /* level 2 offsets */\n%s,\n", dump_array( $size, 0, @array1[$level1..$#array1] );
printf OUTPUT " /* level 3 offsets */\n%s,\n", dump_array( $size, 0, @array2[$level2..$#array2] );
printf OUTPUT " /* values */\n%s\n};\n", dump_array( $size, 0, @array3[$level3..$#array3] );
}
################################################################
# dump a binary case mapping table in l_intl.nls format
sub dump_binary_case_table(@)
{
my (@table) = @_;
my $max_char = 0x10000;
my $level1 = $max_char / 16;
my $level2 = $level1 / 16;
my @difftable;
for (my $i = 0; $i < @table; $i++)
{
next unless defined $table[$i];
$difftable[$i] = ($table[$i] - $i) & 0xffff;
}
my @row_array = compress_array( $level1, 0, @difftable[0..$max_char-1] );
my @array = compress_array( $level2, 0, @row_array[0..$level1-1] );
for (my $i = $level2; $i < @array; $i++) { $array[$i] += @array - $level1; }
return @array, @row_array[$level1..$#row_array];
}
################################################################
# dump case mappings for l_intl.nls
sub dump_intl_nls($)
{
my @upper = dump_binary_case_table( @toupper_table );
my @lower = dump_binary_case_table( @tolower_table );
my $filename = shift;
open OUTPUT,">$filename.new" or die "Cannot create $filename";
printf "Building $filename\n";
binmode OUTPUT;
print OUTPUT pack "S<", 1; # version
print OUTPUT pack "S<*", 1 + scalar @upper, @upper;
print OUTPUT pack "S<*", 1 + scalar @lower, @lower;
close OUTPUT;
save_file($filename);
}
sub load_nameprep_range_table($$$)
{
my ($INPUT, $val, $table_ref) = @_;
while (<$INPUT>)
{
if (/^\s*([0-9a-fA-F]+)-([0-9a-fA-F]+)/)
{
my $last = hex $2;
$last = 65535 if($last >= 65536);
foreach my $i (hex $1 .. $last)
{
$table_ref->[$i] |= $val;
}
next;
}
elsif (/^\s*([0-9a-fA-F]+)/)
{
if (hex $1 < 65536)
{
$table_ref->[hex $1] |= $val;
}
next;
}
return if (/End\sTable/);
}
}
sub load_nameprep_map_table($$)
{
my ($INPUT, $table_ref) = @_;
while (<$INPUT>)
{
if (/^\s*([0-9a-fA-F]+);\s;/)
{
# special value for map to nothing
$table_ref->[hex $1] = [0xffff, 0xffff, 0xffff];
next;
}
elsif (/^\s*([0-9a-fA-F]+);\s([0-9a-fA-F]+);/)
{
$table_ref->[hex $1] = [hex $2, 0, 0];
next;
}
elsif (/^\s*([0-9a-fA-F]+);\s([0-9a-fA-F]+)\s([0-9a-fA-F]+);/)
{
$table_ref->[hex $1] = [hex $2, hex $3, 0];
next;
}
elsif (/^\s*([0-9a-fA-F]+);\s([0-9a-fA-F]+)\s([0-9a-fA-F]+)\s([0-9a-fA-F]+);/)
{
$table_ref->[hex $1] = [hex $2, hex $3, hex $4];
next;
}
return if (/End\sTable/);
}
}
################################################################
# dump mapping table, prohibited characters set, unassigned
# characters, bidirectional rules used by nameprep algorithm
sub dump_nameprep($)
{
my $filename = shift;
my @mapping_table = ();
my @flags_table;
my $INPUT = open_data_file( $RFCS, $STRINGPREP );
while (<$INPUT>)
{
next unless /Start\sTable/;
load_nameprep_range_table($INPUT, $nameprep_flags{"unassigned"}, \@flags_table) if (/A.1/);
load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.1.2/);
load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.2.2/);
load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.3/);
load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.4/);
load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.5/);
load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.6/);
load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.7/);
load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.8/);
load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.9/);
load_nameprep_range_table($INPUT, $nameprep_flags{"bidi_ral"}, \@flags_table) if (/D.1/);
load_nameprep_range_table($INPUT, $nameprep_flags{"bidi_l"}, \@flags_table) if (/D.2/);
load_nameprep_map_table($INPUT, \@mapping_table) if (/B.1/);
load_nameprep_map_table($INPUT, \@mapping_table) if (/B.2/);
}
close $INPUT;
open OUTPUT,">$filename.new" or die "Cannot create $filename";
print "Building $filename\n";
print OUTPUT "/* Nameprep algorithm related data */\n";
print OUTPUT "/* generated from $RFCS/$STRINGPREP */\n";
print OUTPUT "/* DO NOT EDIT!! */\n\n";
print OUTPUT "#include \"windef.h\"\n\n";
dump_two_level_mapping( "nameprep_char_type", 0, 16, @flags_table );
######### mapping table
# first determine all the 16-char subsets that contain something
my @filled = ();
my $pos = 16*3; # for the null subset
for (my $i = 0; $i < 65536; $i++)
{
next unless defined $mapping_table[$i];
$filled[$i >> 4] = $pos;
$pos += 16*3;
$i |= 15;
}
my $total = $pos;
# now count the 256-char subsets that contain something
my @filled_idx = (256) x 256;
$pos = 256 + 16;
for (my $i = 0; $i < 4096; $i++)
{
next unless $filled[$i];
$filled_idx[$i >> 4] = $pos;
$pos += 16;
$i |= 15;
}
my $null_offset = $pos;
$total += $pos;
# add the index offsets to the subsets positions
for (my $i = 0; $i < 4096; $i++)
{
next unless $filled[$i];
$filled[$i] += $null_offset;
}
# dump the main index
printf OUTPUT "const unsigned short DECLSPEC_HIDDEN nameprep_mapping[%d] =\n", $total;
printf OUTPUT "{\n /* index */\n";
printf OUTPUT "%s", dump_array( 16, 0, @filled_idx );
printf OUTPUT ",\n /* null sub-index */\n%s", dump_array( 16, 0, ($null_offset) x 16 );
# dump the second-level indexes
for (my $i = 0; $i < 256; $i++)
{
next unless ($filled_idx[$i] > 256);
my @table = @filled[($i<<4)..($i<<4)+15];
for (my $j = 0; $j < 16; $j++) { $table[$j] ||= $null_offset; }
printf OUTPUT ",\n /* sub-index %02x */\n", $i;
printf OUTPUT "%s", dump_array( 16, 0, @table );
}
# dump the 16-char subsets
printf OUTPUT ",\n /* null mapping */\n";
printf OUTPUT "%s", dump_array( 16, 0, (0) x 48 );
for (my $i = 0; $i < 4096; $i++)
{
next unless $filled[$i];
my @table = (0) x 48;
for (my $j = 0; $j < 16; $j++)
{
if (defined $mapping_table[($i<<4) + $j])
{
$table[3 * $j] = ${$mapping_table[($i << 4) + $j]}[0];
$table[3 * $j + 1] = ${$mapping_table[($i << 4) + $j]}[1];
$table[3 * $j + 2] = ${$mapping_table[($i << 4) + $j]}[2];
}
}
printf OUTPUT ",\n /* 0x%03x0 .. 0x%03xf */\n", $i, $i;
printf OUTPUT "%s", dump_array( 16, 0, @table );
}
printf OUTPUT "\n};\n";
close OUTPUT;
save_file($filename);
}
################################################################
# dump the GetStringTypeW table
sub dump_string_type_table($)
{
my $filename = shift;
open OUTPUT,">$filename.new" or die "Cannot create $filename";
printf "Building $filename\n";
printf OUTPUT "/* Unicode wctype table */\n";
printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
printf OUTPUT "#include \"windef.h\"\n\n";
my @table = @category_table;
# add the direction in the high 4 bits of the category
for (my $i = 0; $i < 65536; $i++)
{
$table[$i] |= $c2_types{$direction_table[$i]} << 12 if defined $direction_table[$i];
}
dump_two_level_mapping( "wctype_table", 0, 16, @table );
close OUTPUT;
save_file($filename);
}
################################################################
# dump the bidi direction table
sub dump_bidi_dir_table($)
{
my $filename = shift;
open OUTPUT,">$filename.new" or die "Cannot create $filename";
printf "Building $filename\n";
printf OUTPUT "/* Unicode BiDi direction table */\n";
printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
printf OUTPUT "#include \"windef.h\"\n\n";
my @table;
for (my $i = 0; $i < 65536; $i++)
{
$table[$i] = $bidi_types{$direction_table[$i]} if defined $direction_table[$i];
}
dump_two_level_mapping( "bidi_direction_table", $bidi_types{"L"}, 16, @table );
close OUTPUT;
save_file($filename);
}
2000-08-11 22:53:40 +02:00
################################################################
# dump the ctype tables
sub dump_ctype_tables($)
2000-08-11 22:53:40 +02:00
{
my $filename = shift;
open OUTPUT,">$filename.new" or die "Cannot create $filename";
printf "Building $filename\n";
2000-08-11 22:53:40 +02:00
printf OUTPUT "/* Unicode ctype tables */\n";
printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
printf OUTPUT "#include \"windef.h\"\n\n";
2000-08-11 22:53:40 +02:00
my @table = @category_table;
2000-08-11 22:53:40 +02:00
# add the direction in the high 4 bits of the category
for (my $i = 0; $i < 65536; $i++)
2000-08-11 22:53:40 +02:00
{
$table[$i] |= $directions{$direction_table[$i]} << 12 if defined $direction_table[$i];
2000-08-11 22:53:40 +02:00
}
dump_simple_mapping( "wine_wctype_table", 0, @table );
2000-08-11 22:53:40 +02:00
close OUTPUT;
save_file($filename);
2000-08-11 22:53:40 +02:00
}
sub rol($$)
{
my ($byte, $count) = @_;
return (($byte << $count) | ($byte >> (8 - $count))) & 0xff;
}
################################################################
# compress the character properties table
sub compress_char_props_table($@)
{
my $rows = shift;
my @table = @_;
my $len = @table / $rows;
my $pos = 0;
my @array = (0) x $rows;
my %sequences;
# add some predefined sequences
foreach my $i (0, 0xfb .. 0xff) { $sequences{pack "L*", (rol($i,5)) x $len} = $i; }
# try to merge table rows
for (my $row = 0; $row < $rows; $row++)
{
my @table_row = map { defined $_ ? $_ : 0x7f; } @table[($row * $len)..(($row + 1) * $len - 1)];
my $rowtxt = pack "L*", @table_row;
if (defined($sequences{$rowtxt}))
{
# reuse an existing row
$array[$row] = $sequences{$rowtxt};
}
else
{
# create a new row
$sequences{$rowtxt} = $array[$row] = ++$pos;
push @array, @table_row;
}
}
return @array;
}
################################################################
# dump a normalization table in binary format
sub dump_norm_table($)
{
my $filename = shift;
my %forms = ( "nfc" => 1, "nfd" => 2, "nfkc" => 5, "nfkd" => 6, "idna" => 13 );
my %decomp = ( "nfc" => \@decomp_table,
"nfd" => \@decomp_table,
"nfkc" => \@decomp_compat_table,
"nfkd" => \@decomp_compat_table ,
"idna" => \@idna_decomp_table );
open OUTPUT,">$filename.new" or die "Cannot create $filename";
print "Building $filename\n";
my $type = $filename;
$type =~ s!.*/norm(\w+)\.nls!$1!;
my $compose = $forms{$type} & 1;
my $compat = !!($forms{$type} & 4) + ($type eq "idna");
my @version = split /\./, $UNIVERSION;
# combining classes
my @classes;
my @class_values;
foreach my $c (grep defined, @combining_class_table)
{
$classes[$c] = 1 if $c < 0x100;
}
for (my $i = 0; $i < @classes; $i++)
{
next unless defined $classes[$i];
$classes[$i] = @class_values;
push @class_values, $i;
}
push @class_values, 0 if (@class_values % 2);
die "too many classes" if @class_values >= 0x40;
# character properties
my @char_props;
my @decomposed;
my @comp_hash_table;
my $comp_hash_size = $compose ? 254 : 0;
for (my $i = 0; $i <= $MAX_CHAR; $i++)
{
next unless defined $combining_class_table[$i];
if (defined $decomp{$type}->[$i])
{
my @dec = get_decomposition( $i, $decomp{$type} );
if ($compose && (my @comp = get_composition( $i, $compat )))
{
my $hash = ($comp[0] + 95 * $comp[1]) % $comp_hash_size;
push @{$comp_hash_table[$hash]}, to_utf16( @comp, $i );
my $val = 0;
foreach my $d (@dec)
{
$val = $combining_class_table[$d];
last if $val;
}
$char_props[$i] = $classes[$val];
}
else
{
$char_props[$i] = 0xbf;
}
@dec = compose_hangul( @dec ) if $compose;
@dec = to_utf16( @dec );
push @dec, 0 if @dec >= 7;
$decomposed[$i] = \@dec;
}
else
{
if ($combining_class_table[$i] == 0x100)
{
$char_props[$i] = 0x7f;
}
elsif ($combining_class_table[$i])
{
$char_props[$i] = $classes[$combining_class_table[$i]] | 0x80;
}
elsif ($type eq "idna" && defined $idna_disallowed[$i])
{
$char_props[$i] = 0xff;
}
else
{
$char_props[$i] = 0;
}
}
}
if ($compose)
{
for (my $i = 0; $i <= $MAX_CHAR; $i++)
{
my @comp = get_composition( $i, $compat );
next unless @comp;
if ($combining_class_table[$comp[1]])
{
$char_props[$comp[0]] |= 0x40 unless $char_props[$comp[0]] & 0x80;
$char_props[$comp[1]] |= 0x40;
}
else
{
$char_props[$comp[0]] = ($char_props[$comp[0]] & ~0x40) | 0x80;
$char_props[$comp[1]] |= 0xc0;
}
}
}
# surrogates
foreach my $i (0xd800..0xdbff) { $char_props[$i] = 0xdf; }
foreach my $i (0xdc00..0xdfff) { $char_props[$i] = 0x9f; }
# Hangul
if ($type eq "nfc") { foreach my $i (0x1100..0x117f) { $char_props[$i] = 0xff; } }
elsif ($compose) { foreach my $i (0x1100..0x11ff) { $char_props[$i] = 0xff; } }
foreach my $i (0xac00..0xd7ff) { $char_props[$i] = 0xff; }
# invalid chars
if ($type eq "idna") { foreach my $i (0x00..0x1f, 0x7f) { $char_props[$i] = 0xff; } }
foreach my $i (0xfdd0..0xfdef) { $char_props[$i] = 0xff; }
foreach my $i (0x00..0x10)
{
$char_props[($i << 16) | 0xfffe] = 0xff;
$char_props[($i << 16) | 0xffff] = 0xff;
}
# decomposition hash table
my @decomp_hash_table;
my @decomp_hash_index;
my @decomp_hash_data;
my $decomp_hash_size = 944;
# build string of character data, reusing substrings when possible
my $decomp_char_data = "";
foreach my $i (sort { @{$b} <=> @{$a} } grep defined, @decomposed)
{
my $str = pack "U*", @{$i};
$decomp_char_data .= $str if index( $decomp_char_data, $str) == -1;
}
for (my $i = 0; $i < @decomposed; $i++)
{
next unless defined $decomposed[$i];
my $pos = index( $decomp_char_data, pack( "U*", @{$decomposed[$i]} ));
die "sequence not found" if $pos == -1;
my $len = @{$decomposed[$i]};
$len = 7 if $len > 7;
my $hash = $i % $decomp_hash_size;
push @{$decomp_hash_table[$hash]}, [ $i, ($len << 13) | $pos ];
}
for (my $i = 0; $i < $decomp_hash_size; $i++)
{
$decomp_hash_index[$i] = @decomp_hash_data / 2;
next unless defined $decomp_hash_table[$i];
if (@{$decomp_hash_table[$i]} == 1)
{
my $entry = $decomp_hash_table[$i]->[0];
if ($char_props[$entry->[0]] == 0xbf)
{
$decomp_hash_index[$i] = $entry->[1];
next;
}
}
foreach my $entry (@{$decomp_hash_table[$i]})
{
push @decomp_hash_data, $entry->[0] & 0xffff, $entry->[1];
}
}
push @decomp_hash_data, 0, 0;
# composition hash table
my @comp_hash_index;
my @comp_hash_data;
if (@comp_hash_table)
{
for (my $i = 0; $i < $comp_hash_size; $i++)
{
$comp_hash_index[$i] = @comp_hash_data;
push @comp_hash_data, @{$comp_hash_table[$i]} if defined $comp_hash_table[$i];
}
$comp_hash_index[$comp_hash_size] = @comp_hash_data;
push @comp_hash_data, 0, 0, 0;
}
my $level1 = ($MAX_CHAR + 1) / 128;
my @rows = compress_char_props_table( $level1, @char_props[0..$MAX_CHAR] );
my @header = ( $version[0], $version[1], $version[2], 0, $forms{$type}, $compat ? 18 : 3,
0, $decomp_hash_size, $comp_hash_size, 0 );
my @tables = (0) x 8;
$tables[0] = 16 + @header + @tables;
$tables[1] = $tables[0] + @class_values / 2;
$tables[2] = $tables[1] + $level1 / 2;
$tables[3] = $tables[2] + (@rows - $level1) / 2;
$tables[4] = $tables[3] + @decomp_hash_index;
$tables[5] = $tables[4] + @decomp_hash_data;
$tables[6] = $tables[5] + length $decomp_char_data;
$tables[7] = $tables[6] + @comp_hash_index;
print OUTPUT pack "S<16", unpack "U*", "norm$type.nlp";
print OUTPUT pack "S<*", @header;
print OUTPUT pack "S<*", @tables;
print OUTPUT pack "C*", @class_values;
print OUTPUT pack "C*", @rows[0..$level1-1];
print OUTPUT pack "C*", @rows[$level1..$#rows];
print OUTPUT pack "S<*", @decomp_hash_index;
print OUTPUT pack "S<*", @decomp_hash_data;
print OUTPUT pack "S<*", unpack "U*", $decomp_char_data;
print OUTPUT pack "S<*", @comp_hash_index;
print OUTPUT pack "S<*", @comp_hash_data;
close OUTPUT;
save_file($filename);
}
################################################################
# output a codepage definition file from the global tables
sub output_codepage_file($)
{
my $codepage = shift;
my $output = sprintf "nls/c_%03d.nls", $codepage;
open OUTPUT,">$output.new" or die "Cannot create $output";
printf "Building %s\n", $output;
if (!@lead_bytes) { dump_binary_sbcs_table( $codepage ); }
else { dump_binary_dbcs_table( $codepage ); }
close OUTPUT;
save_file($output);
}
################################################################
# output a codepage table from a Microsoft-style mapping file
sub dump_msdata_codepage($)
{
my $filename = shift;
my $state = "";
my ($codepage, $width, $count);
my ($lb_cur, $lb_end);
@cp2uni = ();
@glyph2uni = ();
@lead_bytes = ();
@uni2cp = ();
$default_char = $DEF_CHAR;
$default_wchar = $DEF_CHAR;
my $INPUT = open_data_file( $MSCODEPAGES, $filename ) or die "Cannot open $filename";
while (<$INPUT>)
{
next if /^;/; # skip comments
next if /^\s*$/; # skip empty lines
next if /\x1a/; # skip ^Z
last if /^ENDCODEPAGE/;
if (/^CODEPAGE\s+(\d+)/)
{
$codepage = $1;
next;
}
if (/^CPINFO\s+(\d+)\s+0x([0-9a-fA-f]+)\s+0x([0-9a-fA-F]+)/)
{
$width = $1;
$default_char = hex $2;
$default_wchar = hex $3;
next;
}
if (/^(MBTABLE|GLYPHTABLE|WCTABLE|DBCSRANGE|DBCSTABLE)\s+(\d+)/)
{
$state = $1;
$count = $2;
next;
}
if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)/)
{
if ($state eq "MBTABLE")
{
my $cp = hex $1;
my $uni = hex $2;
$cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
next;
}
if ($state eq "GLYPHTABLE")
{
my $cp = hex $1;
my $uni = hex $2;
$glyph2uni[$cp] = $uni unless defined($glyph2uni[$cp]);
next;
}
if ($state eq "WCTABLE")
{
my $uni = hex $1;
my $cp = hex $2;
$uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
next;
}
if ($state eq "DBCSRANGE")
{
my $start = hex $1;
my $end = hex $2;
for (my $i = $start; $i <= $end; $i++) { add_lead_byte( $i ); }
$lb_cur = $start;
$lb_end = $end;
next;
}
if ($state eq "DBCSTABLE")
{
my $mb = hex $1;
my $uni = hex $2;
my $cp = ($lb_cur << 8) | $mb;
$cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
if (!--$count)
{
if (++$lb_cur > $lb_end) { $state = "DBCSRANGE"; }
}
next;
}
}
die "$filename: Unrecognized line $_\n";
}
close $INPUT;
output_codepage_file( $codepage );
2005-11-17 12:43:10 +01:00
}
################################################################
# save a file if modified
sub save_file($)
{
my $file = shift;
if (-f $file && !system "cmp $file $file.new >/dev/null")
2005-11-17 12:43:10 +01:00
{
unlink "$file.new";
}
else
{
rename "$file.new", "$file";
}
}
################################################################
# replace the contents of a file between ### cpmap ### marks
sub REPLACE_IN_FILE($@)
{
my $name = shift;
my @data = @_;
my @lines = ();
open(FILE,$name) or die "Can't open $name";
while (<FILE>)
{
push @lines, $_;
last if /\#\#\# cpmap begin \#\#\#/;
}
push @lines, @data;
while (<FILE>)
{
if (/\#\#\# cpmap end \#\#\#/) { push @lines, "\n", $_; last; }
}
push @lines, <FILE>;
2005-11-17 12:43:10 +01:00
open(FILE,">$name.new") or die "Can't modify $name";
print FILE @lines;
close(FILE);
2005-11-17 12:43:10 +01:00
save_file($name);
}
################################################################
# main routine
chdir ".." if -f "./make_unicode";
load_data();
dump_case_mappings( "libs/port/casemap.c" );
dump_sortkeys( "dlls/kernelbase/collation.c" );
dump_ctype_tables( "libs/port/wctype.c" );
dump_bidi_dir_table( "dlls/gdi32/direction.c" );
dump_bidi_dir_table( "dlls/usp10/direction.c" );
dump_bidi_dir_table( "dlls/dwrite/direction.c" );
dump_string_type_table( "dlls/kernelbase/wctype.c" );
dump_digit_folding( "dlls/kernelbase/digitmap.c" );
dump_mirroring( "dlls/usp10/mirror.c" );
dump_mirroring( "dlls/dwrite/mirror.c" );
dump_bracket( "dlls/usp10/bracket.c" );
2014-09-30 07:56:39 +02:00
dump_bracket( "dlls/dwrite/bracket.c" );
dump_shaping( "dlls/usp10/shaping.c" );
dump_linebreak( "dlls/usp10/linebreak.c" );
dump_linebreak( "dlls/dwrite/linebreak.c" );
dump_scripts( "dlls/dwrite/scripts" );
dump_indic( "dlls/usp10/indicsyllable.c" );
dump_vertical( "dlls/gdi32/vertical.c" );
dump_vertical( "dlls/wineps.drv/vertical.c" );
dump_nameprep( "dlls/kernel32/nameprep.c" );
dump_intl_nls("nls/l_intl.nls");
dump_norm_table( "nls/normnfc.nls" );
dump_norm_table( "nls/normnfd.nls" );
dump_norm_table( "nls/normnfkc.nls" );
dump_norm_table( "nls/normnfkd.nls" );
dump_norm_table( "nls/normidna.nls" );
foreach my $file (@allfiles) { dump_msdata_codepage( $file ); }
dump_eucjp_codepage();
exit 0;
# Local Variables:
# compile-command: "./make_unicode"
# End: