make_unicode 83 KB
Newer Older
1
#!/usr/bin/perl -w
2 3 4 5 6
#
# Generate code page .c files from ftp.unicode.org descriptions
#
# Copyright 2000 Alexandre Julliard
#
7 8 9 10 11 12 13 14 15 16 17 18
# 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
19
# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA
20
#
21

22 23
use strict;

24 25
# base URLs for www.unicode.org files
my $MAPPINGS = "http://www.unicode.org/Public/MAPPINGS";
26
my $UNIDATA = "http://www.unicode.org/Public/10.0.0/ucd";
27 28
my $REPORTS = "http://www.unicode.org/reports";
my $RFCS = "http://www.rfc-editor.org/rfc";
29

30
# Sort keys file
31
my $SORTKEYS = "tr10/allkeys.txt";
32

33
# RFC3454 (stringprep data)
34
my $STRINGPREP = "rfc3454.txt";
35

36
# Defaults mapping
37
my $DEFAULTS = "tools/unicode-defaults";
38 39

# Default char for undefined mappings
40
my $DEF_CHAR = ord '?';
41

42
my @allfiles =
43
(
44
    [ 37,    "VENDORS/MICSFT/EBCDIC/CP037.TXT",   0, "IBM EBCDIC US Canada", 0, 0x6f ],
45 46
    [ 424,   "VENDORS/MISC/CP424.TXT",            0, "IBM EBCDIC Hebrew" ],
    [ 437,   "VENDORS/MICSFT/PC/CP437.TXT",       1, "OEM United States" ],
47
    [ 500,   "VENDORS/MICSFT/EBCDIC/CP500.TXT",   0, "IBM EBCDIC International", 0, 0x6f ],
48 49 50 51 52 53
    [ 737,   "VENDORS/MICSFT/PC/CP737.TXT",       1, "OEM Greek 437G" ],
    [ 775,   "VENDORS/MICSFT/PC/CP775.TXT",       1, "OEM Baltic" ],
    [ 850,   "VENDORS/MICSFT/PC/CP850.TXT",       1, "OEM Multilingual Latin 1" ],
    [ 852,   "VENDORS/MICSFT/PC/CP852.TXT",       1, "OEM Slovak Latin 2" ],
    [ 855,   "VENDORS/MICSFT/PC/CP855.TXT",       1, "OEM Cyrillic" ],
    [ 856,   "VENDORS/MISC/CP856.TXT",            0, "Hebrew PC" ],
54
    [ 857,   "VENDORS/MICSFT/PC/CP857.TXT",       1, "OEM Turkish", 0xf8bb ],
55 56 57 58
    [ 860,   "VENDORS/MICSFT/PC/CP860.TXT",       1, "OEM Portuguese" ],
    [ 861,   "VENDORS/MICSFT/PC/CP861.TXT",       1, "OEM Icelandic" ],
    [ 862,   "VENDORS/MICSFT/PC/CP862.TXT",       1, "OEM Hebrew" ],
    [ 863,   "VENDORS/MICSFT/PC/CP863.TXT",       1, "OEM Canadian French" ],
59
    [ 864,   "VENDORS/MICSFT/PC/CP864.TXT",       0, "OEM Arabic", 0xf8bb ],
60 61 62
    [ 865,   "VENDORS/MICSFT/PC/CP865.TXT",       1, "OEM Nordic" ],
    [ 866,   "VENDORS/MICSFT/PC/CP866.TXT",       1, "OEM Russian" ],
    [ 869,   "VENDORS/MICSFT/PC/CP869.TXT",       1, "OEM Greek" ],
63
    [ 874,   "VENDORS/MICSFT/WindowsBestFit/bestfit874.txt",  1, "ANSI/OEM Thai" ],
64
    [ 875,   "VENDORS/MICSFT/EBCDIC/CP875.TXT",               0, "IBM EBCDIC Greek", 0, 0x6f ],
65 66 67 68 69 70
    [ 878,   "VENDORS/MISC/KOI8-R.TXT",                       0, "Russian KOI8" ],
    [ 932,   "VENDORS/MICSFT/WindowsBestFit/bestfit932.txt",  0, "ANSI/OEM Japanese Shift-JIS" ],
    [ 936,   "VENDORS/MICSFT/WindowsBestFit/bestfit936.txt",  0, "ANSI/OEM Simplified Chinese GBK" ],
    [ 949,   "VENDORS/MICSFT/WindowsBestFit/bestfit949.txt",  0, "ANSI/OEM Korean Unified Hangul" ],
    [ 950,   "VENDORS/MICSFT/WindowsBestFit/bestfit950.txt",  0, "ANSI/OEM Traditional Chinese Big5" ],
    [ 1006,  "VENDORS/MISC/CP1006.TXT",                       0, "IBM Arabic" ],
71
    [ 1026,  "VENDORS/MICSFT/EBCDIC/CP1026.TXT",              0, "IBM EBCDIC Latin 5 Turkish", 0, 0x6f ],
72 73 74 75 76 77 78 79 80
    [ 1250,  "VENDORS/MICSFT/WindowsBestFit/bestfit1250.txt", 0, "ANSI Eastern Europe" ],
    [ 1251,  "VENDORS/MICSFT/WindowsBestFit/bestfit1251.txt", 0, "ANSI Cyrillic" ],
    [ 1252,  "VENDORS/MICSFT/WindowsBestFit/bestfit1252.txt", 0, "ANSI Latin 1" ],
    [ 1253,  "VENDORS/MICSFT/WindowsBestFit/bestfit1253.txt", 0, "ANSI Greek" ],
    [ 1254,  "VENDORS/MICSFT/WindowsBestFit/bestfit1254.txt", 0, "ANSI Turkish" ],
    [ 1255,  "VENDORS/MICSFT/WindowsBestFit/bestfit1255.txt", 0, "ANSI Hebrew" ],
    [ 1256,  "VENDORS/MICSFT/WindowsBestFit/bestfit1256.txt", 0, "ANSI Arabic" ],
    [ 1257,  "VENDORS/MICSFT/WindowsBestFit/bestfit1257.txt", 0, "ANSI Baltic" ],
    [ 1258,  "VENDORS/MICSFT/WindowsBestFit/bestfit1258.txt", 0, "ANSI/OEM Viet Nam" ],
81
    [ 1361,  "VENDORS/MICSFT/WindowsBestFit/bestfit1361.txt", 0, "Korean Johab" ],
82
    [ 10000, "VENDORS/MICSFT/MAC/ROMAN.TXT",      0, "Mac Roman" ],
83
    [ 10001, "VENDORS/APPLE/JAPANESE.TXT",        0, "Mac Japanese" ],
84
    [ 10002, "VENDORS/APPLE/CHINTRAD.TXT",        0, "Mac Traditional Chinese" ],
85
    [ 10003, "VENDORS/APPLE/KOREAN.TXT",          0, "Mac Korean", 0xf8e5 ],
86
    [ 10004, "VENDORS/APPLE/ARABIC.TXT",          0, "Mac Arabic", 0xf826 ],
87
    [ 10005, "VENDORS/APPLE/HEBREW.TXT",          0, "Mac Hebrew", 0xf7fa ],
88 89
    [ 10006, "VENDORS/MICSFT/MAC/GREEK.TXT",      0, "Mac Greek" ],
    [ 10007, "VENDORS/MICSFT/MAC/CYRILLIC.TXT",   0, "Mac Cyrillic" ],
90
    [ 10008, "VENDORS/APPLE/CHINSIMP.TXT",        0, "Mac Simplified Chinese" ],
91
    [ 10010, "VENDORS/APPLE/ROMANIAN.TXT",        0, "Mac Romanian" ],
92
    [ 10017, "VENDORS/APPLE/CYRILLIC.TXT",        0, "Mac Ukrainian" ],
93
    [ 10021, "VENDORS/APPLE/THAI.TXT",            0, "Mac Thai" ],
94 95
    [ 10029, "VENDORS/MICSFT/MAC/LATIN2.TXT",     0, "Mac Latin 2" ],
    [ 10079, "VENDORS/MICSFT/MAC/ICELAND.TXT",    0, "Mac Icelandic" ],
96
    [ 10081, "VENDORS/MICSFT/MAC/TURKISH.TXT",    0, "Mac Turkish", 0xf89f ],
97
    [ 10082, "VENDORS/APPLE/CROATIAN.TXT",        0, "Mac Croatian" ],
98 99
    [ 20127, undef,                               0, "US-ASCII (7bit)" ],
    [ 20866, "VENDORS/MISC/KOI8-R.TXT",           0, "Russian KOI8" ],
100
    [ 20932, "OBSOLETE/EASTASIA/JIS/JIS0208.TXT", 0, "EUC-JP", 0, 0x3f, 0x30fb  ],
101 102 103
    [ 21866, "VENDORS/MISC/KOI8-U.TXT",           0, "Ukrainian KOI8" ],
    [ 28591, "ISO8859/8859-1.TXT",                0, "ISO 8859-1 Latin 1" ],
    [ 28592, "ISO8859/8859-2.TXT",                0, "ISO 8859-2 Latin 2 (East European)" ],
104
    [ 28593, "ISO8859/8859-3.TXT",                0, "ISO 8859-3 Latin 3 (South European)", 0xf7f5 ],
105 106
    [ 28594, "ISO8859/8859-4.TXT",                0, "ISO 8859-4 Latin 4 (Baltic old)" ],
    [ 28595, "ISO8859/8859-5.TXT",                0, "ISO 8859-5 Cyrillic" ],
107 108 109
    [ 28596, "ISO8859/8859-6.TXT",                0, "ISO 8859-6 Arabic", 0xf7c8 ],
    [ 28597, "ISO8859/8859-7.TXT",                0, "ISO 8859-7 Greek", 0xf7c5 ],
    [ 28598, "ISO8859/8859-8.TXT",                0, "ISO 8859-8 Hebrew", 0xf79c ],
110 111 112 113 114 115
    [ 28599, "ISO8859/8859-9.TXT",                0, "ISO 8859-9 Latin 5 (Turkish)" ],
    [ 28600, "ISO8859/8859-10.TXT",               0, "ISO 8859-10 Latin 6 (Nordic)" ],
    [ 28603, "ISO8859/8859-13.TXT",               0, "ISO 8859-13 Latin 7 (Baltic)" ],
    [ 28604, "ISO8859/8859-14.TXT",               0, "ISO 8859-14 Latin 8 (Celtic)" ],
    [ 28605, "ISO8859/8859-15.TXT",               0, "ISO 8859-15 Latin 9 (Euro)" ],
    [ 28606, "ISO8859/8859-16.TXT",               0, "ISO 8859-16 Latin 10 (Balkan)" ]
116 117
);

118

119
my %ctype =
120 121 122 123 124 125 126 127 128
(
    "upper"  => 0x0001,
    "lower"  => 0x0002,
    "digit"  => 0x0004,
    "space"  => 0x0008,
    "punct"  => 0x0010,
    "cntrl"  => 0x0020,
    "blank"  => 0x0040,
    "xdigit" => 0x0080,
129 130
    "alpha"  => 0x0100,
    "defin"  => 0x0200
131 132
);

133 134 135 136 137 138
my %bracket_types =
(
    "o" => 0x0000,
    "c" => 0x0001,
);

139 140 141 142 143 144 145 146 147 148 149 150 151 152
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,
153
    "Consonant_Succeeding_Repha" => 0x000c,
154 155 156 157 158 159 160
    "Consonant_Subjoined"  => 0x000d,
    "Consonant_Medial"  => 0x000e,
    "Consonant_Final"  => 0x000f,
    "Consonant_Head_Letter"  => 0x0010,
    "Modifying_Letter"  => 0x0011,
    "Tone_Letter"  => 0x0012,
    "Tone_Mark"  => 0x0013,
161 162 163 164 165 166 167 168 169 170
    "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,
171 172 173 174 175
    "Brahmi_Joining_Number" => 0x001e,
    "Consonant_With_Stacker" => 0x001f,
    "Consonant_Prefixed" => 0x0020,
    "Syllable_Modifier" => 0x0021,
    "Consonant_Killer" => 0x0022
176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192
);

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,
193 194
    "Invisible"  => 0x0e,
    "Bottom_And_Left"  => 0x0f
195 196
);

197 198 199 200 201 202 203 204
my %nameprep_flags =
(
    "unassigned" => 0x01,
    "prohibited" => 0x02,
    "bidi_ral"   => 0x04,
    "bidi_l"     => 0x08
);

205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242
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,
243 244 245 246
    "B2"  => 0x0025,
    "HL"  => 0x0026,
    "CJ"  => 0x0027,
    "RI"  => 0x0028,
247 248 249
    "EB"  => 0x0029,
    "EM"  => 0x002a,
    "ZWJ" => 0x002b,
250 251
);

252 253 254 255 256 257 258 259
my %vertical_types =
(
    "R"  => 0x0000,
    "U"  => 0x0001,
    "Tr" => 0x0002,
    "Tu" => 0x0003,
);

260
my %categories =
261
(
262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291
    "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
292 293 294
);

# a few characters need additional categories that cannot be determined automatically
295
my %special_categories =
296 297 298
(
    "xdigit" => [ ord('0')..ord('9'),ord('A')..ord('F'),ord('a')..ord('f'),
                  0xff10..0xff19, 0xff21..0xff26, 0xff41..0xff46 ],
299 300
    "space"  => [ 0x09..0x0d, 0x85 ],
    "blank"  => [ 0x09, 0x20, 0xa0, 0x3000, 0xfeff ],
301
    "cntrl"  => [ 0x070f, 0x200c, 0x200d,
302 303
                  0x200e, 0x200f, 0x202a, 0x202b, 0x202c, 0x202d, 0x202e,
                  0x206a, 0x206b, 0x206c, 0x206d, 0x206e, 0x206f, 0xfeff,
304 305
                  0xfff9, 0xfffa, 0xfffb ],
    "punct"  => [ 0x24, 0x2b, 0x3c..0x3e, 0x5e, 0x60, 0x7c, 0x7e, 0xa2..0xbe,
306 307 308
                  0xd7, 0xf7 ],
    "digit"  => [ 0xb2, 0xb3, 0xb9 ],
    "lower"  => [ 0x2071, 0x207f ]
309 310
);

311
my %directions =
312 313 314
(
    "L"   => 1,    # Left-to-Right
    "R"   => 2,    # Right-to-Left
315
    "AL"  => 12,   # Right-to-Left Arabic
316 317 318 319 320
    "EN"  => 3,    # European Number
    "ES"  => 4,    # European Number Separator
    "ET"  => 5,    # European Number Terminator
    "AN"  => 6,    # Arabic Number
    "CS"  => 7,    # Common Number Separator
321 322
    "NSM" => 13,   # Non-Spacing Mark
    "BN"  => 14,   # Boundary Neutral
323 324 325
    "B"   => 8,    # Paragraph Separator
    "S"   => 9,    # Segment Separator
    "WS"  => 10,   # Whitespace
326 327 328 329 330 331 332 333 334 335
    "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
336 337
);

338 339 340 341 342 343 344 345 346 347
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
);

348 349 350 351 352 353 354 355 356
my @cp2uni = ();
my @lead_bytes = ();
my @uni2cp = ();
my @unicode_defaults = ();
my @unicode_aliases = ();
my @tolower_table = ();
my @toupper_table = ();
my @digitmap_table = ();
my @compatmap_table = ();
357 358
my @category_table = ();
my @joining_table = ();
359 360 361
my @direction_table = ();
my @decomp_table = ();
my @compose_table = ();
362 363
my $default_char;
my $default_wchar;
364

365 366 367 368 369 370 371
my %joining_forms =
(
   "isolated" => [],
   "final" => [],
   "initial" => [],
   "medial" => []
);
372

373 374
################################################################
# fetch a unicode.org file and open it
375
sub open_data_file($$)
376
{
377 378
    my ($base, $name) = @_;
    (my $dir = "data/$name") =~ s/\/[^\/]+$//;
379 380 381
    local *FILE;
    unless (-f "data/$name")
    {
382 383 384
        print "Fetching $base/$name...\n";
        system "mkdir", "-p", $dir;
        !system "wget", "-q", "-O", "data/$name", "$base/$name" or die "cannot fetch $base/$name";
385 386 387 388 389
    }
    open FILE, "<data/$name" or die "cannot open data/$name";
    return *FILE;
}

390 391
################################################################
# read in the defaults file
392
sub READ_DEFAULTS($)
393
{
394 395
    my $filename = shift;
    my $start;
396 397 398

    # first setup a few default mappings

399 400
    open DEFAULTS, "$filename" or die "Cannot open $filename";
    print "Loading $filename\n";
401 402 403 404 405 406 407 408 409 410 411 412
    while (<DEFAULTS>)
    {
        next if /^\#/;  # skip comments
        next if /^$/;  # skip empty lines
        if (/^(([0-9a-fA-F]+)(,[0-9a-fA-F]+)*)\s+([0-9a-fA-F]+|'.'|none)\s+(\#.*)?/)
        {
            my @src = map hex, split /,/,$1;
            my $dst = $4;
            my $comment = $5;
            if ($#src > 0) { push @unicode_aliases, \@src; }
            next if ($dst eq "none");
            $dst = ($dst =~ /\'.\'/) ? ord substr($dst,1,1) : hex $dst;
413
            foreach my $src (@src)
414 415 416 417 418 419 420 421
            {
                die "Duplicate value" if defined($unicode_defaults[$src]);
                $unicode_defaults[$src] = $dst;
            }
            next;
        }
        die "Unrecognized line $_\n";
    }
422
    close DEFAULTS;
423 424 425

    # now build mappings from the decomposition field of the Unicode database

426
    my $UNICODE_DATA = open_data_file( $UNIDATA, "UnicodeData.txt" );
427
    while (<$UNICODE_DATA>)
428 429
    {
	# Decode the fields ...
430 431 432 433
	my ($code, $name, $cat, $comb, $bidi,
            $decomp, $dec, $dig, $num, $mirror,
            $oldname, $comment, $upper, $lower, $title) = split /;/;
        my $dst;
434
        my $src = hex $code;
435

436 437 438 439 440
        die "unknown category $cat" unless defined $categories{$cat};
        die "unknown directionality $bidi" unless defined $directions{$bidi};

        $category_table[$src] = $categories{$cat};
        $direction_table[$src] = $directions{$bidi};
441
        $joining_table[$src] = $joining_types{"T"} if $cat eq "Mn" || $cat eq "Me" || $cat eq "Cf";
442 443 444 445 446 447 448 449 450 451 452 453 454

        if ($lower ne "")
        {
            $tolower_table[$src] = hex $lower;
        }
        if ($upper ne "")
        {
            $toupper_table[$src] = hex $upper;
        }
        if ($dec ne "")
        {
            $category_table[$src] |= $ctype{"digit"};
        }
455 456 457 458
        if ($dig ne "")
        {
            $digitmap_table[$src] = ord $dig;
        }
459

460 461 462 463 464 465 466 467 468 469 470
        # 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];
                $start++;
            }
        }
471

472
        next if $decomp eq "";  # no decomposition, skip it
473 474 475 476

        if ($decomp =~ /^<([a-zA-Z]+)>\s+([0-9a-fA-F]+)$/)
        {
            # decomposition of the form "<foo> 1234" -> use char if type is known
477 478
            if (($src >= 0xf900 && $src < 0xfb00) || ($src >= 0xfe30 && $src < 0xfffd))
            {
Francois Gouget's avatar
Francois Gouget committed
479
                # Single char decomposition in the compatibility range
480 481
                $compatmap_table[$src] = hex $2;
            }
482 483 484 485 486
            if ($1 eq "isolated" || $1 eq "final" || $1 eq "initial" || $1 eq "medial")
            {
                ${joining_forms{$1}}[hex $2] = $src;
                next;
            }
487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506
            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;
507
            $category_table[$src] |= $category_table[$dst] if defined $category_table[$dst];
508 509 510 511 512 513
            # store decomposition if it contains two chars
            if ($decomp =~ /^([0-9a-fA-F]+)\s+([0-9a-fA-F]+)$/)
            {
                $decomp_table[$src] = [ hex $1, hex $2 ];
                push @compose_table, [ hex $1, hex $2, $src ];
            }
514 515 516
            elsif ($decomp =~ /^(<[a-z]+>\s)*([0-9a-fA-F]+)$/ &&
                   (($src >= 0xf900 && $src < 0xfb00) || ($src >= 0xfe30 && $src < 0xfffd)))
            {
Francois Gouget's avatar
Francois Gouget committed
517
                # Single char decomposition in the compatibility range
518 519
                $compatmap_table[$src] = hex $2;
            }
520 521 522 523 524 525 526 527 528
        }
        else
        {
            next;
        }

        next if defined($unicode_defaults[$src]);  # may have been set in the defaults file

        # check for loops
529
        for (my $i = $dst; ; $i = $unicode_defaults[$i])
530 531 532 533 534 535
        {
            die sprintf("loop detected for %04x -> %04x",$src,$dst) if $i == $src;
            last unless defined($unicode_defaults[$i]);
        }
        $unicode_defaults[$src] = $dst;
    }
536
    close $UNICODE_DATA;
537 538 539

    # patch the category of some special characters

540
    foreach my $cat (keys %special_categories)
541 542
    {
        my $flag = $ctype{$cat};
543
        foreach my $i (@{$special_categories{$cat}}) { $category_table[$i] |= $flag; }
544
    }
545 546 547 548
}


################################################################
549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572
# define a new lead byte
sub add_lead_byte($)
{
    my $ch = shift;
    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 && !defined($cp2uni[$cp >> 8]))
    {
        add_lead_byte( $cp >> 8 );
    }
}

################################################################
# parse a standard codepage file
sub read_codepage_file($)
573 574
{
    my $name = shift;
575
    my $INPUT = open_data_file( $MAPPINGS, $name );
576

577
    while (<$INPUT>)
578 579 580 581
    {
        next if /^\#/;  # skip comments
        next if /^$/;  # skip empty lines
        next if /\x1a/;  # skip ^Z
582 583
        next if (/^0x([0-9a-fA-F]+)\s+\#UNDEFINED/);  # undefined char

584 585
        if (/^0x([0-9a-fA-F]+)\s+\#DBCS LEAD BYTE/)
        {
586
            add_lead_byte( hex $1 );
587 588
            next;
        }
589
        # 0x12 0x3456
590 591
        if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\s+(\#.*)?/)
        {
592
            add_mapping( hex $1, hex $2 );
593 594
            next;
        }
595 596 597 598 599 600 601
        # 0x12 0x3456+0x7890
        if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\+.*\s+(\#.*)?/)
        {
            add_mapping( hex $1, hex $2 );
            next;
        }
        # 0x12 <LR>+0x3456
602
        if (/^0x([0-9a-fA-F]+)\s+<LR>\+0x([0-9a-fA-F]+)\s+(\#.*)?/)
603 604 605 606
        {
            add_mapping( hex $1, hex $2 );
            next;
        }
607 608 609 610 611 612 613
        # 0x12 <RL>+0x3456
        if (/^0x([0-9a-fA-F]+)\s+<RL>\+0x([0-9a-fA-F]+)\s+(\#.*)?/)
        {
            my $uni = hex $2;
            add_mapping( hex $1, $uni ) if $uni >= 0x80;
            next;
        }
614 615
        die "$name: Unrecognized line $_\n";
    }
616
    close $INPUT;
617 618 619
}


620 621 622 623 624 625 626 627
################################################################
# fill input data for the 20127 (us-ascii) codepage
sub fill_20127_codepage()
{
    for (my $i = 0; $i < 128; $i++) { $cp2uni[$i] = $uni2cp[$i] = $i; }
    for (my $i = 128; $i < 256; $i++) { $cp2uni[$i] = $i & 0x7f; }
}

628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666
################################################################
# get a mapping including glyph chars for MB_USEGLYPHCHARS

sub get_glyphs_mapping(@)
{
    $_[0x01] = 0x263a;  # (WHITE SMILING FACE)
    $_[0x02] = 0x263b;  # (BLACK SMILING FACE)
    $_[0x03] = 0x2665;  # (BLACK HEART SUIT)
    $_[0x04] = 0x2666;  # (BLACK DIAMOND SUIT)
    $_[0x05] = 0x2663;  # (BLACK CLUB SUIT)
    $_[0x06] = 0x2660;  # (BLACK SPADE SUIT)
    $_[0x07] = 0x2022;  # (BULLET)
    $_[0x08] = 0x25d8;  # (INVERSE BULLET)
    $_[0x09] = 0x25cb;  # (WHITE CIRCLE)
    $_[0x0a] = 0x25d9;  # (INVERSE WHITE CIRCLE)
    $_[0x0b] = 0x2642;  # (MALE SIGN)
    $_[0x0c] = 0x2640;  # (FEMALE SIGN)
    $_[0x0d] = 0x266a;  # (EIGHTH NOTE)
    $_[0x0e] = 0x266b;  # (BEAMED EIGHTH NOTES)
    $_[0x0f] = 0x263c;  # (WHITE SUN WITH RAYS)
    $_[0x10] = 0x25ba;  # (BLACK RIGHT-POINTING POINTER)
    $_[0x11] = 0x25c4;  # (BLACK LEFT-POINTING POINTER)
    $_[0x12] = 0x2195;  # (UP DOWN ARROW)
    $_[0x13] = 0x203c;  # (DOUBLE EXCLAMATION MARK)
    $_[0x14] = 0x00b6;  # (PILCROW SIGN)
    $_[0x15] = 0x00a7;  # (SECTION SIGN)
    $_[0x16] = 0x25ac;  # (BLACK RECTANGLE)
    $_[0x17] = 0x21a8;  # (UP DOWN ARROW WITH BASE)
    $_[0x18] = 0x2191;  # (UPWARDS ARROW)
    $_[0x19] = 0x2193;  # (DOWNWARDS ARROW)
    $_[0x1a] = 0x2192;  # (RIGHTWARDS ARROW)
    $_[0x1b] = 0x2190;  # (LEFTWARDS ARROW)
    $_[0x1c] = 0x221f;  # (RIGHT ANGLE)
    $_[0x1d] = 0x2194;  # (LEFT RIGHT ARROW)
    $_[0x1e] = 0x25b2;  # (BLACK UP-POINTING TRIANGLE)
    $_[0x1f] = 0x25bc;  # (BLACK DOWN-POINTING TRIANGLE)
    $_[0x7f] = 0x2302;  # (HOUSE)
    return @_;
}
667

668 669 670 671
################################################################
# build EUC-JP table from the JIS 0208 file
# FIXME: for proper EUC-JP we should probably read JIS 0212 too
# but this would require 3-byte DBCS characters
672
sub READ_JIS0208_FILE($)
673 674 675 676
{
    my $name = shift;

    # ASCII chars
677
    for (my $i = 0x00; $i <= 0x7f; $i++)
678 679 680 681 682 683
    {
        $cp2uni[$i] = $i;
        $uni2cp[$i] = $i;
    }

    # JIS X 0201 right plane
684
    for (my $i = 0xa1; $i <= 0xdf; $i++)
685 686 687 688 689 690
    {
        $cp2uni[0x8e00 + $i] = 0xfec0 + $i;
        $uni2cp[0xfec0 + $i] = 0x8e00 + $i;
    }

    # lead bytes
691
    foreach my $i (0x8e, 0x8f, 0xa1 .. 0xfe)
692
    {
693
        add_lead_byte($i);
694 695 696
    }

    # undefined chars
697
    foreach my $i (0x80 .. 0x8d, 0x90 .. 0xa0, 0xff)
698 699 700 701 702 703 704 705 706 707 708 709
    {
        $cp2uni[$i] = $DEF_CHAR;
    }

    # Shift-JIS compatibility
    $uni2cp[0x00a5] = 0x5c;
    $uni2cp[0x203e] = 0x7e;

    # Fix backslash conversion
    $cp2uni[0xa1c0] = 0xff3c;
    $uni2cp[0xff3c] = 0xa1c0;

710
    my $INPUT = open_data_file( $MAPPINGS, $name );
711
    while (<$INPUT>)
712 713 714 715 716 717
    {
        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+(\#.*)?/)
        {
718 719
            my $cp = 0x8080 + hex $1;
            my $uni = hex $2;
720 721 722 723 724 725
            $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
            $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
            next;
        }
        die "$name: Unrecognized line $_\n";
    }
726
    close $INPUT;
727 728 729
}


730 731
################################################################
# build the sort keys table
732
sub READ_SORTKEYS_FILE()
733 734 735 736
{
    my @sortkeys = ();
    for (my $i = 0; $i < 65536; $i++) { $sortkeys[$i] = [ -1, 0, 0, 0, 0 ] };

737
    my $INPUT = open_data_file( $REPORTS, $SORTKEYS );
738
    while (<$INPUT>)
739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757
    {
        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";
    }
758
    close $INPUT;
759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821

    # 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 < 65536; $i++)
    {
        my @current = @{$sortkeys[$i]};
        next if $current[0] == -1;
        if ($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];
    }
    return @flatkeys;
}


################################################################
# build the sort keys table
822
sub DUMP_SORTKEYS($@)
823
{
824
    my ($filename, @keys) = @_;
825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840

    # count the number of 256-key ranges that contain something

    my @offsets = ();
    my $ranges = 2;
    for (my $i = 0; $i < 256; $i++) { $offsets[$i] = 256; }
    for (my $i = 0; $i < 65536; $i++)
    {
        next unless defined $keys[$i];
        $offsets[$i >> 8] = $ranges * 256;
        $ranges++;
        $i |= 255;
    }

    # output the range offsets

841 842
    open OUTPUT,">$filename.new" or die "Cannot create $filename";
    printf "Building $filename\n";
843
    printf OUTPUT "/* Unicode collation element table */\n";
844
    printf OUTPUT "/* generated from %s */\n", "$REPORTS/$SORTKEYS";
845 846 847 848 849 850 851 852 853
    printf OUTPUT "/* DO NOT EDIT!! */\n\n";

    printf OUTPUT "const unsigned int collation_table[%d] =\n{\n", $ranges*256;
    printf OUTPUT "    /* index */\n";
    printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%08x", 0, @offsets );

    # output the default values

    printf OUTPUT "    /* defaults */\n";
854
    printf OUTPUT "%s", DUMP_ARRAY( "0x%08x", 0, (0xffffffff) x 256 );
855 856 857 858 859 860 861

    # output all the key ranges

    for (my $i = 0; $i < 256; $i++)
    {
        next if $offsets[$i] == 256;
        printf OUTPUT ",\n    /* 0x%02x00 .. 0x%02xff */\n", $i, $i;
862
        printf OUTPUT "%s", DUMP_ARRAY( "0x%08x", 0xffffffff, @keys[($i<<8) .. ($i<<8)+255] );
863 864 865
    }
    printf OUTPUT "\n};\n";
    close OUTPUT;
866
    save_file($filename);
867 868 869
}


870 871
################################################################
# add default mappings once the file had been read
872
sub add_default_mappings($)
873
{
874 875
    my $first_private = shift;

876 877
    # Apply aliases

878
    foreach my $alias (@unicode_aliases)
879 880
    {
        my $target = undef;
881
        foreach my $src (@$alias)
882 883 884 885 886 887 888 889 890 891
        {
            if (defined($uni2cp[$src]))
            {
                $target = $uni2cp[$src];
                last;
            }
        }
        next unless defined($target);

        # At least one char of the alias set is defined, set the others to the same value
892
        foreach my $src (@$alias)
893 894 895 896 897 898 899 900
        {
            $uni2cp[$src] = $target unless defined($uni2cp[$src]);
        }
    }

    # For every src -> target mapping in the defaults table,
    # make uni2cp[src] = uni2cp[target] if uni2cp[target] is defined

901
    for (my $src = 0; $src < 65536; $src++)
902 903 904 905 906 907 908 909 910 911 912
    {
        next if defined($uni2cp[$src]);  # source has a definition already
        next unless defined($unicode_defaults[$src]);  # no default for this char
        my $target = $unicode_defaults[$src];

        # do a recursive mapping until we find a target char that is defined
        while (!defined($uni2cp[$target]) &&
               defined($unicode_defaults[$target])) { $target = $unicode_defaults[$target]; }

        if (defined($uni2cp[$target])) { $uni2cp[$src] = $uni2cp[$target]; }
    }
913

914 915 916 917 918 919 920 921 922 923 924 925 926
    # Add a private char mapping for undefined chars

    if ($first_private)
    {
        for (my $i = 32; $i < 256; $i++)
        {
            next if defined($cp2uni[$i]);
            $cp2uni[$i] = $first_private;
            $uni2cp[$first_private] = $i;
            $first_private++;
        }
    }

927 928
    # Add an identity mapping for all undefined chars

929
    for (my $i = 0; $i < 256; $i++)
930 931 932 933 934
    {
        next if defined($cp2uni[$i]);
        next if defined($uni2cp[$i]);
        $cp2uni[$i] = $uni2cp[$i] = $i;
    }
935 936
}

937 938
################################################################
# dump an array of integers
939
sub DUMP_ARRAY($$@)
940 941
{
    my ($format,$default,@array) = @_;
942 943
    my $i;
    my $ret = "    ";
944 945 946 947 948 949 950 951
    for ($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;
}
952

953

954
################################################################
955 956
# dump a unicode->ascii mapping table
sub dump_uni2cp_table($$)
957
{
958
    my ($width, $def) = @_;
959

960
    # count the number of unicode->ascii subtables that contain something
961

962 963 964 965 966
    my @lowerbounds;
    my @upperbounds;
    my $index = 0;
    my @filled = (-1) x 256;
    for (my $i = 0; $i < 65536; $i++)
967
    {
968 969 970 971 972 973 974 975 976 977 978 979
        next unless defined $uni2cp[$i];
        if ($filled[$i >> 8] == -1)
        {
          $lowerbounds[$index] = $i & 0xff;
          $upperbounds[$index] = 0xff - $lowerbounds[$index];
          $filled[$i >> 8] = $index * 256;
          $index++;
        }
        else
        {
          $upperbounds[$index-1] = 0xff - ($i & 0xff);
        }
980
    }
981

982 983 984 985 986
    # add defaults mapping at the end
    $filled[256] = $index * 256;
    $lowerbounds[$index] = 255;
    $upperbounds[$index] = 0;
    $index++;
987

988 989 990 991
    # collapse blocks upwards if possible
    my $removed = 0;
    $index = 0;
    for (my $i = 0; $i < 257; $i++)
992
    {
993 994 995 996 997 998 999 1000 1001 1002 1003 1004
        next if $filled[$i] == -1;
        if ($upperbounds[$index - 1] > $lowerbounds[$index])
        {
           $removed += $lowerbounds[$index];
        }
        else
        {
           $removed += $upperbounds[$index - 1];
           $lowerbounds[$index] = $upperbounds[$index - 1];
        }
        $filled[$i] -= $removed;
        $index++;
1005 1006 1007 1008
    }

    # output all the subtables into a single array

1009 1010 1011 1012 1013
    printf OUTPUT "static const unsigned %s uni2cp_low[%d] =\n{\n",
                  $width == 8 ? "char" : "short", ($index + 1) * 256 - $removed;

    my $format = $width == 8 ? "0x%02x" : "0x%04x";
    for (my $i = $index = 0; $i < 257; $i++)
1014
    {
1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028
        next if $filled[$i] == -1;
        my $start = ($i << 8) + $lowerbounds[$index];
        my $end = ($i << 8) + 255;
        if ($i == 256)
        {
            print OUTPUT "    /* defaults */\n";
            printf OUTPUT "%s\n};\n\n", DUMP_ARRAY( $format, 0, ($def) x ($end - $start + 1) );
        }
        else
        {
            printf OUTPUT "    /* 0x%04x .. 0x%04x */\n", $start, $end;
            printf OUTPUT "%s,\n", DUMP_ARRAY( $format, $def, @uni2cp[$start .. $end] );
        }
        $index++;
1029 1030 1031 1032
    }

    # output a table of the offsets of the subtables in the previous array

1033
    for (my $i = 0; $i < 256; $i++)
1034
    {
1035
        if ($filled[$i] == -1) { $filled[$i] = $filled[256]; }
1036
    }
1037
    printf OUTPUT "static const unsigned short uni2cp_high[256] =\n";
1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063
    printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%04x", 0, @filled[0..255] );
}

################################################################
# dump an SBCS mapping table
sub dump_sbcs_table($$$$$)
{
    my ($codepage, $has_glyphs, $name, $def, $defw) = @_;
    my $i;

    # output the ascii->unicode table

    if ($has_glyphs)
    {
        printf OUTPUT "static const WCHAR cp2uni[512] =\n";
        printf OUTPUT "{\n%s", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[0 .. 255] );
        printf OUTPUT ",\n    /* glyphs */\n%s\n};\n\n",
                      DUMP_ARRAY( "0x%04x", $defw, get_glyphs_mapping(@cp2uni[0 .. 255]) );
    }
    else
    {
        printf OUTPUT "static const WCHAR cp2uni[256] =\n";
        printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[0 .. 255] );
    }

    dump_uni2cp_table( 8, $def );
1064 1065 1066

    # output the code page descriptor

1067
    printf OUTPUT "const struct sbcs_table DECLSPEC_HIDDEN cptable_%03d =\n{\n", $codepage;
1068
    printf OUTPUT "    { %d, 1, 0x%04x, 0x%04x, \"%s\" },\n",
1069
                  $codepage, $def, $defw, $name;
1070
    printf OUTPUT "    cp2uni,\n";
1071 1072
    if ($has_glyphs) { printf OUTPUT "    cp2uni + 256,\n"; }
    else { printf OUTPUT "    cp2uni,\n"; }
1073 1074 1075 1076 1077 1078 1079
    printf OUTPUT "    uni2cp_low,\n";
    printf OUTPUT "    uni2cp_high\n};\n";
}


################################################################
# dump a DBCS mapping table
1080
sub dump_dbcs_table($$$$@)
1081
{
1082
    my ($codepage, $name, $def, $defw, @lb_ranges) = @_;
1083 1084 1085 1086

    # build a list of lead bytes that are actually used

    my @lblist = ();
1087
    LBLOOP: for (my $y = 0; $y <= $#lead_bytes; $y++)
1088
    {
1089
        $cp2uni[$lead_bytes[$y]] = $defw;
1090
        my $base = $lead_bytes[$y] << 8;
1091
        for (my $x = 0; $x < 256; $x++)
1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103
        {
            if (defined $cp2uni[$base+$x])
            {
                push @lblist,$lead_bytes[$y];
                next LBLOOP;
            }
        }
    }
    my $unused = ($#lead_bytes > $#lblist);

    # output the ascii->unicode table for the single byte chars

1104
    printf OUTPUT "static const WCHAR cp2uni[%d] =\n", 256 * ($#lblist + 2 + $unused);
1105
    printf OUTPUT "{\n%s,\n", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[0 .. 255] );
1106 1107 1108 1109 1110

    # output the default table for unused lead bytes

    if ($unused)
    {
1111
        printf OUTPUT "    /* unused lead bytes */\n";
1112
        printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%04x", 0, ($defw) x 256 );
1113 1114 1115 1116
    }

    # output the ascii->unicode table for each DBCS lead byte

1117
    for (my $y = 0; $y <= $#lblist; $y++)
1118 1119
    {
        my $base = $lblist[$y] << 8;
1120
        printf OUTPUT "    /* lead byte %02x */\n", $lblist[$y];
1121
        printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[$base .. $base+255] );
1122
        printf OUTPUT ($y < $#lblist) ? ",\n" : "\n};\n\n";
1123 1124 1125 1126
    }

    # output the lead byte subtables offsets

1127
    my @offsets = (0) x 256;
1128
    for (my $x = 0; $x <= $#lblist; $x++) { $offsets[$lblist[$x]] = $x + 1; }
1129 1130 1131
    if ($unused)
    {
        # increment all lead bytes offset to take into account the unused table
1132
        for (my $x = 0; $x <= $#lead_bytes; $x++) { $offsets[$lead_bytes[$x]]++; }
1133
    }
1134 1135
    printf OUTPUT "static const unsigned char cp2uni_leadbytes[256] =\n";
    printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%02x", 0, @offsets );
1136

1137
    dump_uni2cp_table( 16, $def );
1138 1139 1140

    # output the code page descriptor

1141
    printf OUTPUT "const struct dbcs_table DECLSPEC_HIDDEN cptable_%03d =\n{\n", $codepage;
1142
    printf OUTPUT "    { %d, 2, 0x%04x, 0x%04x, \"%s\" },\n",
1143
                  $codepage, $def, $defw, $name;
1144 1145 1146 1147
    printf OUTPUT "    cp2uni,\n";
    printf OUTPUT "    cp2uni_leadbytes,\n";
    printf OUTPUT "    uni2cp_low,\n";
    printf OUTPUT "    uni2cp_high,\n";
1148
    printf OUTPUT "    {\n    %s\n    }\n", DUMP_ARRAY( "0x%02x", 0, @lb_ranges, 0, 0 );
1149 1150 1151 1152 1153
    printf OUTPUT "};\n";
}


################################################################
1154 1155
# get the list of defined lead byte ranges
sub get_lb_ranges()
1156 1157
{
    my @list = ();
1158
    my @ranges = ();
1159 1160 1161
    my $i = 0;
    foreach $i (@lead_bytes) { $list[$i] = 1; }
    my $on = 0;
1162
    for (my $i = 0; $i < 256; $i++)
1163 1164 1165
    {
        if ($on)
        {
1166
            if (!defined $list[$i]) { push @ranges, $i-1; $on = 0; }
1167 1168 1169
        }
        else
        {
1170
            if ($list[$i]) { push @ranges, $i; $on = 1; }
1171 1172
        }
    }
1173 1174
    if ($on) { push @ranges, 0xff; }
    return @ranges;
1175 1176
}

1177 1178 1179 1180 1181
################################################################
# dump the Indic Syllabic Category table
sub dump_indic($)
{
    my $filename = shift;
1182
    my @indic_table;
1183

1184
    my $INPUT = open_data_file( $UNIDATA, "IndicSyllabicCategory.txt" );
1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203
    while (<$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};
1204
            if (hex $1 < 65536 and hex $2 < 65536)
1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216
            {
                foreach my $i (hex $1 .. hex $2)
                {
                    $indic_table[$i] = $indic_types{$type};
                }
            }
            next;
        }
        die "malformed line $_";
    }
    close $INPUT;

1217
    $INPUT = open_data_file( $UNIDATA, "IndicPositionalCategory.txt" );
1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247
    while (<$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";
1248
    print OUTPUT "/*       and from $UNIDATA/IndicPositionalCategory.txt */\n";
1249
    print OUTPUT "/* DO NOT EDIT!! */\n\n";
1250
    print OUTPUT "#include \"wine/unicode.h\"\n\n";
1251

1252
    dump_two_level_mapping( "indic_syllabic_table", $indic_types{'Other'}, @indic_table );
1253 1254 1255 1256 1257

    close OUTPUT;
    save_file($filename);
}

1258 1259 1260 1261 1262
################################################################
# dump the Line Break Properties table
sub dump_linebreak($)
{
    my $filename = shift;
1263
    my @break_table;
1264 1265
    my $next_group = 0;

1266
    my $INPUT = open_data_file( $UNIDATA, "LineBreak.txt" );
1267 1268 1269 1270 1271
    while (<$INPUT>)
    {
        next if /^\#/;  # skip comments
        next if /^\s*$/;  # skip empty lines
        next if /\x1a/;  # skip ^Z
1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289
        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*/)
1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314
        {
            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";
1315
    print OUTPUT "#include \"wine/unicode.h\"\n\n";
1316

1317
    dump_two_level_mapping( "wine_linebreak_table", $break_types{'XX'}, @break_table );
1318 1319 1320 1321 1322

    close OUTPUT;
    save_file($filename);
}

1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452
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,
1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465
    # 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
1466 1467
);

1468 1469 1470 1471 1472 1473
################################################################
# dump Script IDs table
sub dump_scripts($)
{
    my $filename = shift;
    my $header = $filename;
1474
    my @scripts_table;
1475
    my $script_index;
1476 1477 1478
    my $i;

    my $INPUT = open_data_file( $UNIDATA, "Scripts.txt" );
1479 1480
    # Fill the table
    # Unknown script id is always 0, so undefined scripts are automatically treated as such
1481 1482 1483 1484 1485 1486 1487
    while (<$INPUT>)
    {
        my $type = "";

        next if /^\#/;  # skip comments
        next if /^\s*$/;  # skip empty lines
        next if /\x1a/;  # skip ^Z
1488
        if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
1489 1490 1491 1492 1493 1494 1495 1496
        {
            $type = $2;
            if (defined $scripts{$type})
            {
                $scripts_table[hex $1] = $scripts{$type};
            }
            next;
        }
1497
        elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520
        {
            $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";
1521
    foreach my $script (sort { $scripts{$a} <=> $scripts{$b} } keys %scripts)
1522 1523 1524
    {
        print OUTPUT "    Script_$script = $scripts{$script},\n";
    }
1525
    print OUTPUT "    Script_LastId = ", (scalar keys %scripts) - 1, "\n";
1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536
    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";
1537
    print OUTPUT "#include \"wine/unicode.h\"\n\n";
1538

1539
    dump_two_level_mapping( "wine_scripts_table", 0, @scripts_table );
1540 1541 1542
    close OUTPUT;
    save_file($filename);
}
1543

1544 1545 1546 1547 1548 1549 1550
################################################################
# dump the BiDi mirroring table
sub dump_mirroring($)
{
    my $filename = shift;
    my @mirror_table = ();

1551
    my $INPUT = open_data_file( $UNIDATA, "BidiMirroring.txt" );
1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571
    while (<$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 $UNIDATA/BidiMirroring.txt */\n";
    print OUTPUT "/* DO NOT EDIT!! */\n\n";
    print OUTPUT "#include \"wine/unicode.h\"\n\n";
1572
    DUMP_CASE_TABLE( "DECLSPEC_HIDDEN wine_mirror_map", @mirror_table );
1573 1574 1575 1576
    close OUTPUT;
    save_file($filename);
}

1577 1578 1579 1580 1581
################################################################
# dump the Bidi Brackets
sub dump_bracket($)
{
    my $filename = shift;
1582
    my @bracket_table;
1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593

    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};
1594
            die "characters too distant $1 and $2" if abs(hex($2) - hex($1)) >= 128;
1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607
            $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";
1608
    print OUTPUT "#include \"wine/unicode.h\"\n\n";
1609

1610
    dump_two_level_mapping( "bidi_bracket_table", 0, @bracket_table );
1611 1612 1613 1614

    close OUTPUT;
    save_file($filename);
}
1615

1616 1617 1618 1619 1620 1621 1622 1623 1624 1625
################################################################
# dump the Arabic shaping table
sub dump_shaping($)
{
    my $filename = shift;
    my %groups;
    my $next_group = 0;

    $groups{"No_Joining_Group"} = $next_group++;

1626
    my $INPUT = open_data_file( $UNIDATA, "ArabicShaping.txt" );
1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648
    while (<$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";
1649
    print OUTPUT "#include \"wine/unicode.h\"\n\n";
1650

1651
    dump_two_level_mapping( "wine_shaping_table", 0, @joining_table );
1652

1653
    print OUTPUT "\nconst unsigned short DECLSPEC_HIDDEN wine_shaping_forms[256][4] =\n{\n";
1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667
    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);
}

1668 1669 1670 1671 1672
################################################################
# dump the Vertical Orientation table
sub dump_vertical($)
{
    my $filename = shift;
1673
    my @vertical_table;
1674

1675
    my $INPUT = open_data_file( $UNIDATA, "VerticalOrientation.txt" );
1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707
    while (<$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";
1708
    print OUTPUT "/* generated from $UNIDATA/VerticalOrientation.txt */\n";
1709
    print OUTPUT "/* DO NOT EDIT!! */\n\n";
1710
    print OUTPUT "#include \"wine/unicode.h\"\n\n";
1711

1712
    dump_two_level_mapping( "vertical_orientation_table", $vertical_types{'R'}, @vertical_table );
1713 1714 1715 1716

    close OUTPUT;
    save_file($filename);
}
1717

1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729
################################################################
# 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 \"wine/unicode.h\"\n\n";

1730
    DUMP_CASE_TABLE( "DECLSPEC_HIDDEN wine_digitmap",  @digitmap_table );
1731
    print OUTPUT "\n";
1732
    DUMP_CASE_TABLE( "DECLSPEC_HIDDEN wine_compatmap", @compatmap_table );
1733 1734 1735 1736 1737
    close OUTPUT;
    save_file($filename);
}


1738 1739
################################################################
# dump the case mapping tables
1740
sub DUMP_CASE_MAPPINGS($)
1741
{
1742 1743
    my $filename = shift;
    open OUTPUT,">$filename.new" or die "Cannot create $filename";
1744 1745 1746 1747 1748
    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 \"wine/unicode.h\"\n\n";
1749

1750
    DUMP_CASE_TABLE( "wine_casemap_lower", @tolower_table );
1751
    print OUTPUT "\n";
1752
    DUMP_CASE_TABLE( "wine_casemap_upper", @toupper_table );
1753
    close OUTPUT;
1754
    save_file($filename);
1755 1756 1757 1758 1759
}


################################################################
# dump a case mapping table
1760
sub DUMP_CASE_TABLE($@)
1761 1762 1763 1764
{
    my ($name,@table) = @_;

    # count the number of sub tables that contain something
1765
    # also compute the low and upper populated bounds
1766

1767 1768 1769
    my @lowerbounds = ( 0, 0 );
    my @upperbounds = ( 0, 255 );
    my $index = 0;
1770
    my @filled = ();
1771
    for (my $i = 0; $i < 65536; $i++)
1772 1773
    {
        next unless defined $table[$i];
1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785
        if (!defined $filled[$i >> 8])
        {
          $lowerbounds[$index] = $i & 0xff;
          $upperbounds[$index] = 0xff - $lowerbounds[$index];
          $filled[$i >> 8] = $index * 256 + 512;
          $index++;
        }
        else
        {
          $upperbounds[$index-1] = 0xff - ($i & 0xff);
        }
        $table[$i] = ($table[$i] - $i) & 0xffff;
1786
    }
1787 1788 1789 1790

    # Collapse blocks upwards if possible
    my $removed = 0;
    $index = 0;
1791
    for (my $i = 0; $i < 256; $i++)
1792
    {
1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804
        next unless defined $filled[$i];
        if ($upperbounds[$index - 1] > $lowerbounds[$index])
        {
           $removed = $removed + $lowerbounds[$index];
        }
        else
        {
           $removed = $removed + $upperbounds[$index - 1];
           $lowerbounds[$index] = $upperbounds[$index - 1];
        }
        $filled[$i] = $filled[$i] - $removed;
        $index++;
1805 1806 1807 1808
    }

    # dump the table

1809
    printf OUTPUT "const WCHAR %s[%d] =\n", $name, $index * 256 + 512 - $removed;
1810 1811 1812 1813
    printf OUTPUT "{\n    /* index */\n";
    printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%04x", 256, @filled );
    printf OUTPUT "    /* defaults */\n";
    printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, (0) x 256 );
1814
    $index = 0;
1815
    for (my $i = 0; $i < 256; $i++)
1816 1817
    {
        next unless $filled[$i];
1818 1819 1820 1821
        printf OUTPUT ",\n    /* 0x%02x%02x .. 0x%02xff */\n", $i, $lowerbounds[$index], $i;
        printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0,
                      @table[($i<<8) + $lowerbounds[$index] .. ($i<<8)+255] );
        $index++;
1822 1823 1824 1825
    }
    printf OUTPUT "\n};\n";
}

1826
################################################################
1827
# compress a mapping table by removing identical rows
1828
sub compress_array($$@)
1829
{
1830
    my $rows = shift;
1831
    my $def = shift;
1832
    my @table = @_;
1833 1834
    my $len = @table / $rows;
    my @array = (0) x $rows;
1835 1836 1837
    my %sequences;

    # try to merge table rows
1838
    for (my $row = 0; $row < $rows; $row++)
1839
    {
1840 1841
        my @table_row = map { defined($_) ? $_ : $def; } @table[($row * $len)..($row * $len + $len - 1)];
        my $rowtxt = pack "S*", @table_row;
1842 1843 1844 1845 1846 1847 1848 1849 1850
        if (defined($sequences{$rowtxt}))
        {
            # reuse an existing row
            $array[$row] = $sequences{$rowtxt};
        }
        else
        {
            # create a new row
            $sequences{$rowtxt} = $array[$row] = $#array + 1;
1851
            push @array, @table_row;
1852 1853
        }
    }
1854 1855 1856 1857 1858
    return @array;
}

################################################################
# dump a simple char -> 16-bit value mapping table
1859
sub dump_simple_mapping($$@)
1860 1861
{
    my $name = shift;
1862 1863
    my $def = shift;
    my @array = compress_array( 256, $def, @_[0..65535] );
1864 1865 1866 1867 1868

    printf OUTPUT "const unsigned short %s[%d] =\n{\n", $name, $#array+1;
    printf OUTPUT "    /* offsets */\n%s,\n", DUMP_ARRAY( "0x%04x", 0, @array[0..255] );
    printf OUTPUT "    /* values */\n%s\n};\n", DUMP_ARRAY( "0x%04x", 0, @array[256..$#array] );
}
1869

1870
################################################################
1871
# dump a char -> 16-bit value mapping table using two-level tables
1872
sub dump_two_level_mapping($$@)
1873 1874
{
    my $name = shift;
1875 1876 1877
    my $def = shift;
    my @row_array = compress_array( 4096, $def, @_[0..65535] );
    my @array = compress_array( 256, 0, @row_array[0..4095] );
1878 1879 1880

    for (my $i = 256; $i < @array; $i++) { $array[$i] += @array - 4096; }

1881
    printf OUTPUT "const unsigned short DECLSPEC_HIDDEN %s[%d] =\n{\n", $name, @array + @row_array - 4096;
1882 1883 1884 1885 1886 1887
    printf OUTPUT "    /* level 1 offsets */\n%s,\n", DUMP_ARRAY( "0x%04x", 0, @array[0..255] );
    printf OUTPUT "    /* level 2 offsets */\n%s,\n", DUMP_ARRAY( "0x%04x", 0, @array[256..$#array] );
    printf OUTPUT "    /* values */\n%s\n};\n", DUMP_ARRAY( "0x%04x", 0, @row_array[4096..$#row_array] );
}

################################################################
1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971
# dump a binary case mapping table in l_intl.nls format
sub dump_binary_case_table(@)
{
    my (@table) = @_;

    my %difftables_hash = ();
    my @difftables;
    my %offtables2_hash = ();
    my @offtables2 = ();
    
    my @offtable = ();
    for (my $i = 0; $i < 256; $i++)
    {
        my @offtable2 = ();
        for(my $j = 0; $j < 16; $j++) # offset table for xx00-xxFF characters
        {
            my @difftable;
            for (my $k = 0; $k < 16; $k++) # case map table for xxx0-xxxF characters
            {
                my $char = ($i<<8) + ($j<<4) + $k;
                $difftable[$k] = (defined $table[$char]) ? (($table[$char]-$char) & 0xffff) : 0;
            }

            my $diff_key = pack "S*", @difftable;
            my $offset3 = $difftables_hash{$diff_key};
            if (!defined $offset3)
            {
                $offset3 = scalar @difftables;
                $difftables_hash{$diff_key} = $offset3;
                push @difftables, @difftable;
            }
            $offtable2[$j] = $offset3;
        }

        my $offtable2_key = pack "S*", @offtable2;
        my $offset2 = $offtables2_hash{$offtable2_key};
        if (!defined $offset2)
        {
            $offset2 = scalar @offtables2;
            $offtables2_hash{$offtable2_key} = $offset2;
            push @offtables2, \@offtable2;
        }
        $offtable[$i] = $offset2;
    }

    my @output;
    my $offset = 0x100; # offset of first subtable in words
    foreach (@offtable)
    {
        push @output, 0x10 * $_ + $offset; # offset of subtable in words
    }

    $offset = 0x100 + 0x10 * scalar @offtables2; # offset of first difftable in words
    foreach(@offtables2)
    {
        my $table = $_;
        foreach(@$table)
        {
            push @output, $_ + $offset; # offset of difftable in words
        }
    }

    my $len = 1 + scalar @output + scalar @difftables;
    return pack "S<*", $len, @output, @difftables;
}


################################################################
# dump case mappings for l_intl.nls
sub dump_intl_nls($)
{
    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 dump_binary_case_table( @toupper_table );
    print OUTPUT dump_binary_case_table( @tolower_table );
    close OUTPUT;
    save_file($filename);
}


1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039
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 = ();
2040
    my @flags_table;
2041

2042
    my $INPUT = open_data_file( $RFCS, $STRINGPREP );
2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067
    while (<$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";
2068
    print OUTPUT "/* generated from $RFCS/$STRINGPREP */\n";
2069
    print OUTPUT "/* DO NOT EDIT!! */\n\n";
2070
    print OUTPUT "#include \"wine/unicode.h\"\n\n";
2071

2072
    dump_two_level_mapping( "nameprep_char_type", 0, @flags_table );
2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107

    ######### 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
2108
    printf OUTPUT "const unsigned short DECLSPEC_HIDDEN nameprep_mapping[%d] =\n", $total;
2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149
    printf OUTPUT "{\n    /* index */\n";
    printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @filled_idx );
    printf OUTPUT ",\n    /* null sub-index */\n%s", DUMP_ARRAY( "0x%04x", 0, ($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( "0x%04x", 0, @table );
    }

    # dump the 16-char subsets
    printf OUTPUT ",\n    /* null mapping */\n";
        printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 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( "0x%04x", 0, @table );
    }

    printf OUTPUT "\n};\n";

    close OUTPUT;
    save_file($filename);
}

2150 2151
################################################################
# dump the ctype tables
2152
sub DUMP_CTYPE_TABLES($)
2153
{
2154 2155 2156
    my $filename = shift;
    open OUTPUT,">$filename.new" or die "Cannot create $filename";
    printf "Building $filename\n";
2157 2158 2159 2160 2161
    printf OUTPUT "/* Unicode ctype tables */\n";
    printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
    printf OUTPUT "#include \"wine/unicode.h\"\n\n";

    # add the direction in the high 4 bits of the category
2162
    for (my $i = 0; $i < 65536; $i++)
2163
    {
2164
        $category_table[$i] |= $direction_table[$i] << 12 if defined $direction_table[$i];
2165 2166
    }

2167
    dump_simple_mapping( "wine_wctype_table", 0, @category_table );
2168 2169

    close OUTPUT;
2170
    save_file($filename);
2171 2172
}

2173 2174

################################################################
2175 2176
# dump the char composition table
sub dump_compose_table($)
2177
{
2178 2179 2180
    my $filename = shift;

    open OUTPUT,">$filename.new" or die "Cannot create $filename";
2181 2182 2183 2184 2185
    print "Building $filename\n";
    print OUTPUT "/* Unicode char composition */\n";
    print OUTPUT "/* generated from $UNIDATA/UnicodeData.txt */\n";
    print OUTPUT "/* DO NOT EDIT!! */\n\n";
    print OUTPUT "#include \"wine/unicode.h\"\n\n";
2186 2187

    my @filled = ();
2188
    foreach my $i (@compose_table)
2189 2190 2191 2192 2193 2194 2195
    {
        my @comp = @$i;
        push @{$filled[$comp[1]]}, [ $comp[0], $comp[2] ];
    }

    # count how many different second chars we have

2196 2197
    my $count = 0;
    for (my $i = 0; $i < 65536; $i++)
2198 2199 2200 2201 2202 2203 2204 2205
    {
        next unless defined $filled[$i];
        $count++;
    }

    # build the table of second chars and offsets

    my $pos = $count + 1;
2206 2207
    my @table = ();
    for (my $i = 0; $i < 65536; $i++)
2208 2209 2210 2211 2212 2213 2214
    {
        next unless defined $filled[$i];
        push @table, $i, $pos;
        $pos += @{$filled[$i]};
    }
    # terminator with last position
    push @table, 0, $pos;
2215
    printf OUTPUT "static const WCHAR table[0x%x] =\n{\n", 2*$pos;
2216 2217 2218 2219
    printf OUTPUT "    /* second chars + offsets */\n%s", DUMP_ARRAY( "0x%04x", 0, @table );

    # build the table of first chars and mappings

2220
    for (my $i = 0; $i < 65536; $i++)
2221 2222 2223 2224
    {
        next unless defined $filled[$i];
        my @table = ();
        my @list = sort { $a->[0] <=> $b->[0] } @{$filled[$i]};
2225
        for (my $j = 0; $j <= $#list; $j++)
2226 2227 2228 2229 2230
        {
            push @table, $list[$j][0], $list[$j][1];
        }
        printf OUTPUT ",\n    /* 0x%04x */\n%s", $i, DUMP_ARRAY( "0x%04x", 0, @table );
    }
2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244
    print OUTPUT "\n};\n\n";
    print OUTPUT <<"EOF";
static inline int binary_search( WCHAR ch, int low, int high )
{
    while (low <= high)
    {
        int pos = (low + high) / 2;
        if (table[2 * pos] < ch) low = pos + 1;
        else if (table[2 * pos] > ch) high = pos - 1;
        else return pos;
    }
    return -1;
}

2245
WCHAR DECLSPEC_HIDDEN wine_compose( const WCHAR *str )
2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256
{
    int pos, idx = 1, start = 0, count = $count;
    for (;;)
    {
        if ((pos = binary_search( str[idx], start, count - 1 )) == -1) return 0;
        if (!idx--) return table[2 * pos + 1];
        start = table[2 * pos + 1];
        count = table[2 * pos + 3];
    }
}
EOF
2257 2258 2259 2260 2261 2262 2263 2264 2265
    close OUTPUT;
    save_file($filename);
}

################################################################
# dump the char decomposition table
sub dump_decompose_table($)
{
    my $filename = shift;
2266

2267 2268 2269 2270 2271 2272
    open OUTPUT,">$filename.new" or die "Cannot create $filename";
    print "Building $filename\n";
    print OUTPUT "/* Unicode char composition */\n";
    print OUTPUT "/* generated from $UNIDATA/UnicodeData.txt */\n";
    print OUTPUT "/* DO NOT EDIT!! */\n\n";
    print OUTPUT "#include \"wine/unicode.h\"\n\n";
2273 2274 2275

    # first determine all the 16-char subsets that contain something

2276 2277
    my @filled = (0) x 4096;
    my $pos = 16*2;  # for the null subset
2278
    for (my $i = 0; $i < 65536; $i++)
2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290
    {
        next unless defined $decomp_table[$i];
        $filled[$i >> 4] = $pos;
        $pos += 16*2;
        $i |= 15;
    }
    my $total = $pos;

    # now count the 256-char subsets that contain something

    my @filled_idx = (256) x 256;
    $pos = 256 + 16;
2291
    for (my $i = 0; $i < 4096; $i++)
2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302
    {
        next unless $filled[$i];
        $filled_idx[$i >> 4] = $pos;
        $pos += 16;
        $i |= 15;
    }
    my $null_offset = $pos;  # null mapping
    $total += $pos;

    # add the index offsets to the subsets positions

2303
    for (my $i = 0; $i < 4096; $i++)
2304 2305 2306 2307 2308 2309 2310
    {
        next unless $filled[$i];
        $filled[$i] += $null_offset;
    }

    # dump the main index

2311
    printf OUTPUT "static const WCHAR table[%d] =\n", $total;
2312 2313 2314 2315 2316 2317
    printf OUTPUT "{\n    /* index */\n";
    printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @filled_idx );
    printf OUTPUT ",\n    /* null sub-index */\n%s", DUMP_ARRAY( "0x%04x", 0, ($null_offset) x 16 );

    # dump the second-level indexes

2318
    for (my $i = 0; $i < 256; $i++)
2319 2320 2321
    {
        next unless ($filled_idx[$i] > 256);
        my @table = @filled[($i<<4)..($i<<4)+15];
2322
        for (my $j = 0; $j < 16; $j++) { $table[$j] ||= $null_offset; }
2323 2324 2325 2326 2327 2328 2329 2330 2331
        printf OUTPUT ",\n    /* sub-index %02x */\n", $i;
        printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
    }

    # dump the 16-char subsets

    printf OUTPUT ",\n    /* null mapping */\n";
    printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, (0) x 32 );

2332
    for (my $i = 0; $i < 4096; $i++)
2333 2334 2335
    {
        next unless $filled[$i];
        my @table = (0) x 32;
2336
        for (my $j = 0; $j < 16; $j++)
2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347
        {
            if (defined $decomp_table[($i<<4) + $j])
            {
                $table[2 * $j] = ${$decomp_table[($i << 4) + $j]}[0];
                $table[2 * $j + 1] = ${$decomp_table[($i << 4) + $j]}[1];
            }
        }
        printf OUTPUT ",\n    /* 0x%03x0 .. 0x%03xf */\n", $i, $i;
        printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
    }

2348 2349
    printf OUTPUT "\n};\n\n";
    print OUTPUT <<"EOF";
2350
unsigned int DECLSPEC_HIDDEN wine_decompose( WCHAR ch, WCHAR *dst, unsigned int dstlen )
2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362
{
    const WCHAR *ptr = table + table[table[ch >> 8] + ((ch >> 4) & 0x0f)] + 2 * (ch & 0xf);
    unsigned int res;

    *dst = ch;
    if (!*ptr) return 1;
    if (dstlen <= 1) return 0;
    /* apply the decomposition recursively to the first char */
    if ((res = wine_decompose( *ptr, dst, dstlen-1 ))) dst[res++] = ptr[1];
    return res;
}
EOF
2363
    close OUTPUT;
2364
    save_file($filename);
2365 2366
}

2367
################################################################
2368 2369 2370 2371 2372
# output a codepage definition file from the global tables
sub output_codepage_file($$$$)
{
    my ($codepage, $filename, $comment, $has_glyphs) = @_;

2373
    my $output = sprintf "libs/port/c_%03d.c", $codepage;
2374
    open OUTPUT,">$output.new" or die "Cannot create $output";
2375

2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400
    printf "Building %s from %s (%s)\n", $output, $filename || "hardcoded data", $comment;

    # dump all tables

    printf OUTPUT "/* code page %03d (%s) */\n", $codepage, $comment;
    if ($filename)
    {
        print OUTPUT "/* generated from $MAPPINGS/$filename */\n";
        print OUTPUT "/* DO NOT EDIT!! */\n\n";
    }
    else
    {
        printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
    }
    printf OUTPUT "#include \"wine/unicode.h\"\n\n";

    if (!@lead_bytes) { dump_sbcs_table( $codepage, $has_glyphs, $comment, $default_char, $default_wchar ); }
    else { dump_dbcs_table( $codepage, $comment, $default_char, $default_wchar, get_lb_ranges() ); }
    close OUTPUT;
    save_file($output);
}

################################################################
# read a "bestfit" Windows mapping file
sub read_bestfit_file($)
2401
{
2402
    my ($filename) = @_;
2403
    my $state = "";
2404
    my ($codepage, $width, $count);
2405 2406
    my ($lb_cur, $lb_end);

2407
    my $INPUT = open_data_file( $MAPPINGS, $filename ) or die "Cannot open $filename";
2408

2409
    while (<$INPUT>)
2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423
    {
        next if /^;/;  # skip comments
        next if /^\s*$/;  # 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;
2424 2425
            $default_char = hex $2;
            $default_wchar = hex $3;
2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437
            next;
        }
        if (/^(MBTABLE|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")
            {
2438 2439
                my $cp = hex $1;
                my $uni = hex $2;
2440 2441 2442 2443 2444
                $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
                next;
            }
            if ($state eq "WCTABLE")
            {
2445 2446
                my $uni = hex $1;
                my $cp = hex $2;
2447 2448 2449 2450 2451
                $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
                next;
            }
            if ($state eq "DBCSRANGE")
            {
2452 2453
                my $start = hex $1;
                my $end = hex $2;
2454
                for (my $i = $start; $i <= $end; $i++) { add_lead_byte( $i ); }
2455 2456 2457 2458 2459 2460
                $lb_cur = $start;
                $lb_end = $end;
                next;
            }
            if ($state eq "DBCSTABLE")
            {
2461 2462 2463
                my $mb = hex $1;
                my $uni = hex $2;
                my $cp = ($lb_cur << 8) | $mb;
2464 2465 2466 2467 2468 2469 2470 2471
                $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
                if (!--$count)
                {
                    if (++$lb_cur > $lb_end) { $state = "DBCSRANGE"; }
                }
                next;
            }
        }
2472
        die "$filename: Unrecognized line $_\n";
2473
    }
2474
    close $INPUT;
2475 2476 2477
}


2478 2479
################################################################
# read an input file and generate the corresponding .c file
2480
sub HANDLE_FILE(@)
2481
{
2482
    my ($codepage,$filename,$has_glyphs,$comment,$first_private,$def,$defw) = @_;
2483

2484 2485 2486
    @cp2uni = ();
    @lead_bytes = ();
    @uni2cp = ();
2487 2488
    $default_char = $def || $DEF_CHAR;
    $default_wchar = $defw || $DEF_CHAR;
2489

2490 2491
    # some codepage files are special
    if ($codepage == 20932)
2492
    {
2493
        READ_JIS0208_FILE( $filename );
2494
        add_default_mappings( $first_private );
2495
    }
2496
    elsif ($codepage == 20127)
2497
    {
2498
        fill_20127_codepage();
2499
        add_default_mappings( $first_private );
2500 2501 2502 2503
    }
    elsif ($filename =~ /\/bestfit/)
    {
        read_bestfit_file( $filename );
2504 2505 2506
    }
    else
    {
2507
        read_codepage_file( $filename );
2508 2509 2510 2511 2512 2513 2514 2515 2516
        if ($codepage == 10001)
        {
            # add Shift-JIS mappings
            read_bestfit_file( "VENDORS/MICSFT/WindowsBestFit/bestfit932.txt" );
        }
        else
        {
            add_default_mappings( $first_private );
        }
2517
    }
2518

2519
    output_codepage_file( $codepage, $filename, $comment, $has_glyphs );
2520 2521 2522 2523 2524 2525 2526 2527
}


################################################################
# save a file if modified
sub save_file($)
{
    my $file = shift;
2528
    if (-f $file && !system "cmp $file $file.new >/dev/null")
2529 2530 2531 2532 2533 2534 2535
    {
        unlink "$file.new";
    }
    else
    {
        rename "$file.new", "$file";
    }
2536 2537 2538 2539 2540
}


################################################################
# output the list of codepage tables into the cptable.c file
2541
sub output_cptable($)
2542
{
2543
    my $output = shift;
2544
    my @tables_decl = ();
2545

2546 2547
    printf "Building %s\n", $output;

2548
    foreach my $file (@allfiles)
2549
    {
2550
        my ($codepage,$filename) = @$file;
2551 2552 2553 2554
        push @tables_decl, sprintf("extern union cptable cptable_%03d;\n",$codepage);
    }

    push @tables_decl, sprintf("\nstatic const union cptable * const cptables[%d] =\n{\n",$#allfiles+1);
2555
    foreach my $file (@allfiles)
2556
    {
2557
        my ($codepage,$filename) = @$file;
2558 2559 2560
        push @tables_decl, sprintf("    &cptable_%03d,\n", $codepage);
    }
    push @tables_decl, "};";
2561
    REPLACE_IN_FILE( $output, @tables_decl );
2562 2563 2564 2565 2566
}

################################################################
# replace the contents of a file between ### cpmap ### marks

2567
sub REPLACE_IN_FILE($@)
2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583
{
    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", $_; last; }
    }
    push @lines, <FILE>;
2584
    open(FILE,">$name.new") or die "Can't modify $name";
2585 2586
    print FILE @lines;
    close(FILE);
2587
    save_file($name);
2588
}
2589 2590 2591 2592

################################################################
# main routine

2593
chdir ".." if -f "./make_unicode";
2594
READ_DEFAULTS( $DEFAULTS );
2595 2596
DUMP_CASE_MAPPINGS( "libs/wine/casemap.c" );
DUMP_SORTKEYS( "libs/wine/collation.c", READ_SORTKEYS_FILE() );
2597
dump_compose_table( "libs/port/compose.c" );
2598
dump_decompose_table( "libs/port/decompose.c" );
2599
DUMP_CTYPE_TABLES( "libs/wine/wctype.c" );
2600
dump_digit_folding( "libs/port/digitmap.c" );
2601
dump_mirroring( "dlls/usp10/mirror.c" );
2602
dump_mirroring( "dlls/dwrite/mirror.c" );
2603
dump_bracket( "dlls/usp10/bracket.c" );
2604
dump_bracket( "dlls/dwrite/bracket.c" );
2605 2606
dump_shaping( "dlls/usp10/shaping.c" );
dump_linebreak( "dlls/usp10/linebreak.c" );
2607
dump_linebreak( "dlls/dwrite/linebreak.c" );
2608
dump_scripts( "dlls/dwrite/scripts" );
2609
dump_indic( "dlls/usp10/indicsyllable.c" );
2610
dump_intl_nls("loader/l_intl.nls");
2611 2612
dump_vertical( "dlls/gdi32/vertical.c" );
dump_vertical( "dlls/wineps.drv/vertical.c" );
2613
dump_nameprep( "dlls/kernel32/nameprep.c" );
2614 2615 2616

foreach my $file (@allfiles) { HANDLE_FILE( @{$file} ); }

2617
output_cptable("libs/port/cptable.c");
2618 2619 2620 2621

exit 0;

# Local Variables:
2622
# compile-command: "./make_unicode"
2623
# End: