tools/make_unicode

   1 #!/usr/bin/perl -w
   2 #
   3 # Generate code page .c files from ftp.unicode.org descriptions
   4 #
   5 # Copyright 2000 Alexandre Julliard
   6 #
   7 # This library is free software; you can redistribute it and/or
   8 # modify it under the terms of the GNU Lesser General Public
   9 # License as published by the Free Software Foundation; either
  10 # version 2.1 of the License, or (at your option) any later version.
  11 #
  12 # This library is distributed in the hope that it will be useful,
  13 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  14 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  15 # Lesser General Public License for more details.
  16 #
  17 # You should have received a copy of the GNU Lesser General Public
  18 # 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 use strict;
  23 use XML::LibXML;
  24 use Digest::SHA;
  25 use Encode;
  26 use Time::Local qw(timegm_modern);
  27
  28 my $UNIVERSION = "14.0.0";
  29 my $CLDRVERSION = "41";
  30 my $ISO639VERSION = "20220120";
  31 my $TZVERSION = "2022a";
  32
  33 my %data_files =
  34 (
  35  ucd       => { url  => "https://www.unicode.org/Public/$UNIVERSION/ucd/UCD.zip", name => "UCD-$UNIVERSION.zip",
  36                 sha  => "033a5276b5d7af8844589f8e3482f3977a8385e71d107d375055465178c23600" },
  37  unihan    => { url  => "https://www.unicode.org/Public/$UNIVERSION/ucd/Unihan.zip", name => "Unihan-$UNIVERSION.zip",
  38                 sha  => "2ae4519b2b82cd4d15379c17e57bfb12c33c0f54da4977de03b2b04bcf11852d" },
  39  idna      => { url  => "https://www.unicode.org/Public/idna/$UNIVERSION/IdnaMappingTable.txt", name => "IdnaMappingTable-$UNIVERSION.txt",
  40                 sha  => "d43d9ca367af27b0e4c9dc645cadc23690bdecaf7ec2687f37f01180022d4dfa" },
  41  cldr      => { url  => "https://github.com/unicode-org/cldr/archive/refs/tags/release-$CLDRVERSION.zip",
  42                 sha  => "a2b7aee281ad2f497d47995808cf5e8f24123b0814ca47f7a824556aec8a0d91" },
  43  cldr33    => { url  => "https://www.unicode.org/Public/cldr/33/cldr-common-33.0.zip",
  44                 sha  => "fa3490082c086d21257153609642f54fcf788fcfda4966fe67f3f6daca0d58b9" },
  45  sorting   => { url  => "https://download.microsoft.com/download/C/F/7/CF713A5E-9FBC-4FD6-9246-275F65C0E498/Windows 10 Sorting Weight Table.txt",
  46                 sha  => "81fcfa1e5ed3e3a94d329959ff7d97d522ddf9d653d2c4d6ddcccc5cd4df663f" },
  47  codepages => { url  => "https://download.microsoft.com/download/C/F/7/CF713A5E-9FBC-4FD6-9246-275F65C0E498/Windows Supported Code Page Data Files.zip",
  48                 sha  => "5074e6dd253056ba61fc6c870c9a955467855129c6ad3a51761c386b301b125a" },
  49  iso639    => { url  => "https://iso639-3.sil.org/sites/iso639-3/files/downloads/iso-639-3_Code_Tables_$ISO639VERSION.zip",
  50                 sha  => "d912749d10c344835f052a9f31d13f13d5ffc99bc589e1eb88f2b4663e990881" },
  51  ksx1001   => { url  => "https://www.unicode.org/Public/MAPPINGS/OBSOLETE/EASTASIA/KSC/KSX1001.TXT",
  52                 sha  => "d8d2a35206ac0ea2865f5d801c9d6717f735bf46f263a658a64a960abe59e371" },
  53  jis0208   => { url  => "https://www.unicode.org/Public/MAPPINGS/OBSOLETE/EASTASIA/JIS/JIS0208.TXT",
  54                 sha  => "1c571870457f19c97720631fa83ee491549a96ba1436da1296786a67d8632e87" },
  55  jis0212   => { url  => "https://www.unicode.org/Public/MAPPINGS/OBSOLETE/EASTASIA/JIS/JIS0212.TXT",
  56                 sha  => "477820bb3055bbcc90880d788cd95607d221dc94457bae249231adecf13c12e6" },
  57  tzdata    => { url  => "https://data.iana.org/time-zones/releases/tzdata$TZVERSION.tar.gz",
  58                 sha  => "ef7fffd9f4f50f4f58328b35022a32a5a056b245c5cb3d6791dddb342f871664" },
  59 );
  60
  61
  62 # Default char for undefined mappings
  63 my $DEF_CHAR = ord '?';
  64
  65 # Last valid Unicode character
  66 my $MAX_CHAR = 0x10ffff;
  67
  68 my $nlskey = "-SYSTEM\\-CurrentControlSet\\-Control\\-Nls";
  69 my $zonekey = "-Software\\-Microsoft\\-Windows NT\\-CurrentVersion\\Time Zones";
  70
  71 my @allfiles =
  72 (
  73     "CodpageFiles/037.txt",
  74     "CodpageFiles/437.txt",
  75     "CodpageFiles/500.txt",
  76     "CodpageFiles/708.txt",
  77     "CodpageFiles/720.txt",
  78     "CodpageFiles/737.txt",
  79     "CodpageFiles/775.txt",
  80     "CodpageFiles/850.txt",
  81     "CodpageFiles/852.txt",
  82     "CodpageFiles/855.txt",
  83     "CodpageFiles/857.txt",
  84     "CodpageFiles/860.txt",
  85     "CodpageFiles/861.txt",
  86     "CodpageFiles/862.txt",
  87     "CodpageFiles/863.txt",
  88     "CodpageFiles/864.txt",
  89     "CodpageFiles/865.txt",
  90     "CodpageFiles/866.txt",
  91     "CodpageFiles/869.txt",
  92     "CodpageFiles/874.txt",
  93     "CodpageFiles/875.txt",
  94     "CodpageFiles/932.txt",
  95     "CodpageFiles/936.txt",
  96     "CodpageFiles/949.txt",
  97     "CodpageFiles/950.txt",
  98     "CodpageFiles/1026.txt",
  99     "CodpageFiles/1250.txt",
 100     "CodpageFiles/1251.txt",
 101     "CodpageFiles/1252.txt",
 102     "CodpageFiles/1253.txt",
 103     "CodpageFiles/1254.txt",
 104     "CodpageFiles/1255.txt",
 105     "CodpageFiles/1256.txt",
 106     "CodpageFiles/1257.txt",
 107     "CodpageFiles/1258.txt",
 108     "CodpageFiles/1361.txt",
 109     "CodpageFiles/10000.txt",
 110     "CodpageFiles/10001.txt",
 111     "CodpageFiles/10002.txt",
 112     "CodpageFiles/10003.txt",
 113     "CodpageFiles/10004.txt",
 114     "CodpageFiles/10005.txt",
 115     "CodpageFiles/10006.txt",
 116     "CodpageFiles/10007.txt",
 117     "CodpageFiles/10008.txt",
 118     "CodpageFiles/10010.txt",
 119     "CodpageFiles/10017.txt",
 120     "CodpageFiles/10021.txt",
 121     "CodpageFiles/10029.txt",
 122     "CodpageFiles/10079.txt",
 123     "CodpageFiles/10081.txt",
 124     "CodpageFiles/10082.txt",
 125     "CodpageFiles/20127.txt",
 126     "CodpageFiles/20866.txt",
 127     "CodpageFiles/21866.txt",
 128     "CodpageFiles/28591.txt",
 129     "CodpageFiles/28592.txt",
 130     "CodpageFiles/28593.txt",
 131     "CodpageFiles/28594.txt",
 132     "CodpageFiles/28595.txt",
 133     "CodpageFiles/28596.txt",
 134     "CodpageFiles/28597.txt",
 135     "CodpageFiles/28598.txt",
 136     "CodpageFiles/28599.txt",
 137     "CodpageFiles/28603.txt",
 138     "CodpageFiles/28605.txt",
 139 );
 140
 141 my @timezone_files = qw(africa antarctica asia australasia europe northamerica southamerica etcetera backward);
 142
 143 my %ctype =
 144 (
 145      # CT_CTYPE1
 146     "upper"  => 0x0001,
 147     "lower"  => 0x0002,
 148     "digit"  => 0x0004,
 149     "space"  => 0x0008,
 150     "punct"  => 0x0010,
 151     "cntrl"  => 0x0020,
 152     "blank"  => 0x0040,
 153     "xdigit" => 0x0080,
 154     "alpha"  => 0x0100 | 0x80000000,
 155     "defin"  => 0x0200,
 156      # CT_CTYPE3 in high 16 bits
 157     "nonspacing"    => 0x00010000,
 158     "diacritic"     => 0x00020000,
 159     "vowelmark"     => 0x00040000,
 160     "symbol"        => 0x00080000,
 161     "katakana"      => 0x00100000,
 162     "hiragana"      => 0x00200000,
 163     "halfwidth"     => 0x00400000,
 164     "fullwidth"     => 0x00800000,
 165     "ideograph"     => 0x01000000,
 166     "kashida"       => 0x02000000,
 167     "lexical"       => 0x04000000,
 168     "highsurrogate" => 0x08000000,
 169     "lowsurrogate"  => 0x10000000,
 170 );
 171
 172 my %bracket_types =
 173 (
 174     "o" => 0x0000,
 175     "c" => 0x0001,
 176 );
 177
 178 my %indic_types =
 179 (
 180     "Other"    => 0x0000,
 181     "Bindu"    => 0x0001,
 182     "Visarga"  => 0x0002,
 183     "Avagraha" => 0x0003,
 184     "Nukta"    => 0x0004,
 185     "Virama"   => 0x0005,
 186     "Vowel_Independent"  => 0x0006,
 187     "Vowel_Dependent"  => 0x0007,
 188     "Vowel"  => 0x0008,
 189     "Consonant_Placeholder"  => 0x0009,
 190     "Consonant"  => 0x000a,
 191     "Consonant_Dead"  => 0x000b,
 192     "Consonant_Succeeding_Repha" => 0x000c,
 193     "Consonant_Subjoined"  => 0x000d,
 194     "Consonant_Medial"  => 0x000e,
 195     "Consonant_Final"  => 0x000f,
 196     "Consonant_Head_Letter"  => 0x0010,
 197     "Modifying_Letter"  => 0x0011,
 198     "Tone_Letter"  => 0x0012,
 199     "Tone_Mark"  => 0x0013,
 200     "Register_Shifter"  => 0x0014,
 201     "Consonant_Preceding_Repha" => 0x0015,
 202     "Pure_Killer" => 0x0016,
 203     "Invisible_Stacker" => 0x0017,
 204     "Gemination_Mark" => 0x0018,
 205     "Cantillation_Mark" => 0x0019,
 206     "Non_Joiner" => 0x001a,
 207     "Joiner" => 0x001b,
 208     "Number_Joiner" => 0x001c,
 209     "Number" => 0x001d,
 210     "Brahmi_Joining_Number" => 0x001e,
 211     "Consonant_With_Stacker" => 0x001f,
 212     "Consonant_Prefixed" => 0x0020,
 213     "Syllable_Modifier" => 0x0021,
 214     "Consonant_Killer" => 0x0022,
 215     "Consonant_Initial_Postfixed" => 0x0023,
 216 );
 217
 218 my %matra_types =
 219 (
 220     "Right"    => 0x01,
 221     "Left"  => 0x02,
 222     "Visual_Order_Left" => 0x03,
 223     "Left_And_Right"    => 0x04,
 224     "Top"   => 0x05,
 225     "Bottom"  => 0x06,
 226     "Top_And_Bottom"  => 0x07,
 227     "Top_And_Right"  => 0x08,
 228     "Top_And_Left"  => 0x09,
 229     "Top_And_Left_And_Right"  => 0x0a,
 230     "Bottom_And_Right"  => 0x0b,
 231     "Top_And_Bottom_And_Right"  => 0x0c,
 232     "Overstruck"  => 0x0d,
 233     "Invisible"  => 0x0e,
 234     "Bottom_And_Left"  => 0x0f,
 235     "Top_And_Bottom_And_Left"  => 0x10,
 236 );
 237
 238 my %break_types =
 239 (
 240     "BK"  => 0x0001,
 241     "CR"  => 0x0002,
 242     "LF"  => 0x0003,
 243     "CM"  => 0x0004,
 244     "SG"  => 0x0005,
 245     "GL"  => 0x0006,
 246     "CB"  => 0x0007,
 247     "SP"  => 0x0008,
 248     "ZW"  => 0x0009,
 249     "NL"  => 0x000a,
 250     "WJ"  => 0x000b,
 251     "JL"  => 0x000c,
 252     "JV"  => 0x000d,
 253     "JT"  => 0x000e,
 254     "H2"  => 0x000f,
 255     "H3"  => 0x0010,
 256     "XX"  => 0x0011,
 257     "OP"  => 0x0012,
 258     "CL"  => 0x0013,
 259     "CP"  => 0x0014,
 260     "QU"  => 0x0015,
 261     "NS"  => 0x0016,
 262     "EX"  => 0x0017,
 263     "SY"  => 0x0018,
 264     "IS"  => 0x0019,
 265     "PR"  => 0x001a,
 266     "PO"  => 0x001b,
 267     "NU"  => 0x001c,
 268     "AL"  => 0x001d,
 269     "ID"  => 0x001e,
 270     "IN"  => 0x001f,
 271     "HY"  => 0x0020,
 272     "BB"  => 0x0021,
 273     "BA"  => 0x0022,
 274     "SA"  => 0x0023,
 275     "AI"  => 0x0024,
 276     "B2"  => 0x0025,
 277     "HL"  => 0x0026,
 278     "CJ"  => 0x0027,
 279     "RI"  => 0x0028,
 280     "EB"  => 0x0029,
 281     "EM"  => 0x002a,
 282     "ZWJ" => 0x002b,
 283 );
 284
 285 my %vertical_types =
 286 (
 287     "R"  => 0x0000,
 288     "U"  => 0x0001,
 289     "Tr" => 0x0002,
 290     "Tu" => 0x0003,
 291 );
 292
 293 my %categories =
 294 (
 295     "Lu" => $ctype{"defin"}|$ctype{"alpha"}|$ctype{"upper"}, # Letter, Uppercase
 296     "Ll" => $ctype{"defin"}|$ctype{"alpha"}|$ctype{"lower"}, # Letter, Lowercase
 297     "Lt" => $ctype{"defin"}|$ctype{"alpha"}|$ctype{"upper"}|$ctype{"lower"},    # Letter, Titlecase
 298     "Mn" => $ctype{"defin"}|$ctype{"nonspacing"}, # Mark, Non-Spacing
 299     "Mc" => $ctype{"defin"},                    # Mark, Spacing Combining
 300     "Me" => $ctype{"defin"},                    # Mark, Enclosing
 301     "Nd" => $ctype{"defin"}|$ctype{"digit"},    # Number, Decimal Digit
 302     "Nl" => $ctype{"defin"}|$ctype{"alpha"},    # Number, Letter
 303     "No" => $ctype{"defin"},                    # Number, Other
 304     "Zs" => $ctype{"defin"}|$ctype{"space"},    # Separator, Space
 305     "Zl" => $ctype{"defin"}|$ctype{"space"},    # Separator, Line
 306     "Zp" => $ctype{"defin"}|$ctype{"space"},    # Separator, Paragraph
 307     "Cc" => $ctype{"defin"}|$ctype{"cntrl"},    # Other, Control
 308     "Cf" => $ctype{"defin"}|$ctype{"cntrl"},    # Other, Format
 309     "Cs" => $ctype{"defin"},                    # Other, Surrogate
 310     "Co" => $ctype{"defin"},                    # Other, Private Use
 311     "Cn" => $ctype{"defin"},                    # Other, Not Assigned
 312     "Lm" => $ctype{"defin"}|$ctype{"alpha"},    # Letter, Modifier
 313     "Lo" => $ctype{"defin"}|$ctype{"alpha"},    # Letter, Other
 314     "Pc" => $ctype{"defin"}|$ctype{"punct"},    # Punctuation, Connector
 315     "Pd" => $ctype{"defin"}|$ctype{"punct"},    # Punctuation, Dash
 316     "Ps" => $ctype{"defin"}|$ctype{"punct"},    # Punctuation, Open
 317     "Pe" => $ctype{"defin"}|$ctype{"punct"},    # Punctuation, Close
 318     "Pi" => $ctype{"defin"}|$ctype{"punct"},    # Punctuation, Initial quote
 319     "Pf" => $ctype{"defin"}|$ctype{"punct"},    # Punctuation, Final quote
 320     "Po" => $ctype{"defin"}|$ctype{"punct"},    # Punctuation, Other
 321     "Sm" => $ctype{"defin"}|$ctype{"symbol"},   # Symbol, Math
 322     "Sc" => $ctype{"defin"}|$ctype{"symbol"},   # Symbol, Currency
 323     "Sk" => $ctype{"defin"}|$ctype{"symbol"},   # Symbol, Modifier
 324     "So" => $ctype{"defin"}|$ctype{"symbol"}    # Symbol, Other
 325 );
 326
 327 # a few characters need additional categories that cannot be determined automatically
 328 my %special_categories =
 329 (
 330     "xdigit" => [ ord('0')..ord('9'),ord('A')..ord('F'),ord('a')..ord('f'),
 331                   0xff10..0xff19, 0xff21..0xff26, 0xff41..0xff46 ],
 332     "space"  => [ 0x09..0x0d, 0x85 ],
 333     "blank"  => [ 0x09, 0x20, 0xa0, 0x3000, 0xfeff ],
 334     "cntrl"  => [ 0x070f, 0x200c, 0x200d,
 335                   0x200e, 0x200f, 0x202a, 0x202b, 0x202c, 0x202d, 0x202e,
 336                   0x206a, 0x206b, 0x206c, 0x206d, 0x206e, 0x206f, 0xfeff,
 337                   0xfff9, 0xfffa, 0xfffb ],
 338     "punct"  => [ 0x24, 0x2b, 0x3c..0x3e, 0x5e, 0x60, 0x7c, 0x7e, 0xa2..0xbe,
 339                   0xd7, 0xf7 ],
 340     "digit"  => [ 0xb2, 0xb3, 0xb9 ],
 341     "lower"  => [ 0xaa, 0xba, 0x2071, 0x207f ],
 342     "nonspacing" => [ 0xc0..0xc5, 0xc7..0xcf, 0xd1..0xd6, 0xd8..0xdd, 0xe0..0xe5, 0xe7..0xef,
 343                       0xf1..0xf6, 0xf8..0xfd, 0xff, 0x6de, 0x1929..0x192b, 0x302e..0x302f ],
 344     "diacritic" => [ 0x5e, 0x60, 0xb7, 0xd8, 0xf8 ],
 345     "symbol" => [ 0x09..0x0d, 0x20..0x23, 0x25, 0x26, 0x28..0x2a, 0x2c, 0x2e..0x2f, 0x3a..0x40,
 346                   0x5b..0x60, 0x7b..0x7e, 0xa0..0xa9, 0xab..0xb1, 0xb4..0xb8, 0xbb, 0xbf,
 347                   0x02b9..0x02ba, 0x02c6..0x02cf ],
 348     "halfwidth" => [ 0x20..0x7e, 0xa2..0xa3, 0xa5..0xa6, 0xac, 0xaf, 0x20a9 ],
 349     "fullwidth" => [ 0x2018..0x2019, 0x201c..0x201d, 0x3000..0x3002, 0x300c..0x300d, 0x309b..0x309c,
 350                      0x30a1..0x30ab, 0x30ad, 0x30ad, 0x30af, 0x30b1, 0x30b3, 0x30b5, 0x30b7, 0x30b9,
 351                      0x30bb, 0x30bd, 0x30bf, 0x30c1, 0x30c3, 0x30c4, 0x30c6, 0x30c8, 0x30ca..0x30cf,
 352                      0x30d2, 0x30d5, 0x30d8, 0x30db, 0x30de..0x30ed, 0x30ef, 0x30f2..0x30f3, 0x30fb,
 353                      0x3131..0x3164 ],
 354     "ideograph" => [ 0x3006..0x3007 ],
 355     "lexical" => [ 0x22, 0x24, 0x27, 0x2d, 0x2f, 0x3d, 0x40, 0x5c, 0x5e..0x60, 0x7e,
 356                    0xa8, 0xaa, 0xad, 0xaf, 0xb4, 0xb8, 0xba,
 357                    0x02b0..0x02b8, 0x02bc, 0x02c7, 0x02ca..0x02cb, 0x02cf, 0x02d8..0x02dd, 0x02e0..0x02e3,
 358                    0x037a, 0x0384..0x0385, 0x0387, 0x0559..0x055a, 0x0640, 0x1fbd..0x1fc1,
 359                    0x1fcd..0x1fcf, 0x1fdd..0x1fdf, 0x1fed..0x1fef, 0x1ffd..0x1ffe, 0x2010..0x2015,
 360                    0x2032..0x2034, 0x2038, 0x2043..0x2044, 0x207b..0x207c, 0x207f, 0x208b..0x208c,
 361                    0x2212, 0x2215..0x2216, 0x2500, 0x2504..0x2505, 0x2508..0x2509, 0x254c..0x254d,
 362                    0x3003, 0x301c, 0x3030..0x3035, 0x309b..0x309e, 0x30fd..0x30fe, 0xfe31..0xfe32,
 363                    0xfe58, 0xfe63, 0xfe66, 0xfe68..0xfe69, 0xfe6b, 0xff04, 0xff07, 0xff0d, 0xff0f,
 364                    0xff1d, 0xff20, 0xff3c, 0xff3e, 0xff40, 0xff5e ],
 365     "kashida" => [ 0x0640 ],
 366 );
 367
 368 my %directions =
 369 (
 370     "L"   => 1,    # Left-to-Right
 371     "R"   => 2,    # Right-to-Left
 372     "AL"  => 12,   # Right-to-Left Arabic
 373     "EN"  => 3,    # European Number
 374     "ES"  => 4,    # European Number Separator
 375     "ET"  => 5,    # European Number Terminator
 376     "AN"  => 6,    # Arabic Number
 377     "CS"  => 7,    # Common Number Separator
 378     "NSM" => 13,   # Non-Spacing Mark
 379     "BN"  => 14,   # Boundary Neutral
 380     "B"   => 8,    # Paragraph Separator
 381     "S"   => 9,    # Segment Separator
 382     "WS"  => 10,   # Whitespace
 383     "ON"  => 11,   # Other Neutrals
 384     "LRE" => 15,   # Left-to-Right Embedding
 385     "LRO" => 15,   # Left-to-Right Override
 386     "RLE" => 15,   # Right-to-Left Embedding
 387     "RLO" => 15,   # Right-to-Left Override
 388     "PDF" => 15,   # Pop Directional Format
 389     "LRI" => 15,   # Left-to-Right Isolate
 390     "RLI" => 15,   # Right-to-Left Isolate
 391     "FSI" => 15,   # First Strong Isolate
 392     "PDI" => 15    # Pop Directional Isolate
 393 );
 394
 395 my %c2_types =
 396 (
 397     "L"   => 1,    # C2_LEFTTORIGHT
 398     "R"   => 2,    # C2_RIGHTTOLEFT
 399     "AL"  => 2,    # C2_RIGHTTOLEFT
 400     "EN"  => 3,    # C2_EUROPENUMBER
 401     "ES"  => 4,    # C2_EUROPESEPARATOR
 402     "ET"  => 5,    # C2_EUROPETERMINATOR
 403     "AN"  => 6,    # C2_ARABICNUMBER
 404     "CS"  => 7,    # C2_COMMONSEPARATOR
 405     "NSM" => 11,   # C2_OTHERNEUTRAL
 406     "BN"  => 0,    # C2_NOTAPPLICABLE
 407     "B"   => 8,    # C2_BLOCKSEPARATOR
 408     "S"   => 9,    # C2_SEGMENTSEPARATOR
 409     "WS"  => 10,   # C2_WHITESPACE
 410     "ON"  => 11,   # C2_OTHERNEUTRAL
 411     "LRE" => 11,   # C2_OTHERNEUTRAL
 412     "LRO" => 11,   # C2_OTHERNEUTRAL
 413     "RLE" => 11,   # C2_OTHERNEUTRAL
 414     "RLO" => 11,   # C2_OTHERNEUTRAL
 415     "PDF" => 11,   # C2_OTHERNEUTRAL
 416     "LRI" => 11,   # C2_OTHERNEUTRAL
 417     "RLI" => 11,   # C2_OTHERNEUTRAL
 418     "FSI" => 11,   # C2_OTHERNEUTRAL
 419     "PDI" => 11    # C2_OTHERNEUTRAL
 420 );
 421
 422 my %bidi_types =
 423 (
 424     "ON"  => 0,    # Other Neutrals
 425     "L"   => 1,    # Left-to-Right
 426     "R"   => 2,    # Right-to-Left
 427     "AN"  => 3,    # Arabic Number
 428     "EN"  => 4,    # European Number
 429     "AL"  => 5,    # Right-to-Left Arabic
 430     "NSM" => 6,    # Non-Spacing Mark
 431     "CS"  => 7,    # Common Number Separator
 432     "ES"  => 8,    # European Number Separator
 433     "ET"  => 9,    # European Number Terminator
 434     "BN"  => 10,   # Boundary Neutral
 435     "S"   => 11,   # Segment Separator
 436     "WS"  => 12,   # Whitespace
 437     "B"   => 13,   # Paragraph Separator
 438     "RLO" => 14,   # Right-to-Left Override
 439     "RLE" => 15,   # Right-to-Left Embedding
 440     "LRO" => 16,   # Left-to-Right Override
 441     "LRE" => 17,   # Left-to-Right Embedding
 442     "PDF" => 18,   # Pop Directional Format
 443     "LRI" => 19,   # Left-to-Right Isolate
 444     "RLI" => 20,   # Right-to-Left Isolate
 445     "FSI" => 21,   # First Strong Isolate
 446     "PDI" => 22    # Pop Directional Isolate
 447 );
 448
 449 my %joining_types =
 450 (
 451    "U" => 0,           # Non_Joining
 452    "L" => 1,           # Left_Joining
 453    "R" => 2,           # Right_Joining
 454    "D" => 3,           # Dual_Joining
 455    "C" => 3,           # Join_Causing
 456    "ALAPH" => 4,       # Syriac ALAPH
 457    "DALATH RISH" => 5, # Syriac DALATH RISH group
 458    "T" => 6,           # Transparent
 459 );
 460
 461 my @locales =
 462 (
 463  { name => "", lcid => 0x0000007f, file => "root", territory => "IV", sabbrevlangname => "IVL", sopentypelang =>"dflt" },
 464  { name => "aa", dir => "seed", sopentypelang => "AFR" },
 465  { name => "aa-DJ", dir => "seed" },
 466  { name => "aa-ER", dir => "seed" },
 467  { name => "aa-ET", dir => "seed" },
 468  { name => "af", lcid => 0x00000036, oemcp => 850, sabbrevlangname => "AFK", sopentypelang => "AFK" },
 469  { name => "af-NA" },
 470  { name => "af-ZA", lcid => 0x00000436 },
 471  { name => "agq" },
 472  { name => "agq-CM" },
 473  { name => "ak", sopentypelang => "TWI" },
 474  { name => "ak-GH" },
 475  { name => "am", lcid => 0x0000005e, sabbrevlangname => "AMH" },
 476  { name => "am-ET", lcid => 0x0000045e },
 477  { name => "ar", lcid => 0x00000001, territory => "SA", oemcp => 720, group => 13 },
 478  { name => "ar-001" },
 479  { name => "ar-AE", lcid => 0x00003801, sabbrevlangname => "ARU" },
 480  { name => "ar-BH", lcid => 0x00003c01, sabbrevlangname => "ARH" },
 481  { name => "ar-DJ" },
 482  { name => "ar-DZ", lcid => 0x00001401, sabbrevlangname => "ARG" },
 483  { name => "ar-EG", lcid => 0x00000c01, sabbrevlangname => "ARE" },
 484  { name => "ar-EH" },
 485  { name => "ar-ER" },
 486  { name => "ar-IL" },
 487  { name => "ar-IQ", lcid => 0x00000801, sabbrevlangname => "ARI" },
 488  { name => "ar-JO", lcid => 0x00002c01, sabbrevlangname => "ARJ" },
 489  { name => "ar-KM" },
 490  { name => "ar-KW", lcid => 0x00003401, sabbrevlangname => "ARK" },
 491  { name => "ar-LB", lcid => 0x00003001, sabbrevlangname => "ARB" },
 492  { name => "ar-LY", lcid => 0x00001001, sabbrevlangname => "ARL" },
 493  { name => "ar-MA", lcid => 0x00001801, sabbrevlangname => "ARM" },
 494  { name => "ar-MR" },
 495  { name => "ar-OM", lcid => 0x00002001, sabbrevlangname => "ARO" },
 496  { name => "ar-PS" },
 497  { name => "ar-QA", lcid => 0x00004001, sabbrevlangname => "ARQ" },
 498  { name => "ar-SA", lcid => 0x00000401, sabbrevlangname => "ARA" },
 499  { name => "ar-SD" },
 500  { name => "ar-SO" },
 501  { name => "ar-SS" },
 502  { name => "ar-SY", lcid => 0x00002801, sabbrevlangname => "ARS" },
 503  { name => "ar-TD" },
 504  { name => "ar-TN", lcid => 0x00001c01, sabbrevlangname => "ART" },
 505  { name => "ar-YE", lcid => 0x00002401, sabbrevlangname => "ARY" },
 506  { name => "arn", lcid => 0x0000007a, oemcp => 850, ebcdiccp => 20284, slist => ",", dir => "seed", sabbrevlangname => "MPD", sopentypelang => "MAP" },
 507  { name => "arn-CL", lcid => 0x0000047a, dir => "seed" },
 508  { name => "arn-Latn", alias => "arn" },
 509  { name => "arn-Latn-CL", alias => "arn-CL" },
 510  { name => "as", lcid => 0x0000004d, slist => ",", group => 15 },
 511  { name => "as-IN", lcid => 0x0000044d },
 512  { name => "asa" },
 513  { name => "asa-TZ" },
 514  { name => "ast" },
 515  { name => "ast-ES" },
 516  { name => "az", lcid => 0x0000002c, oemcp => 857, ebcdiccp => 20905, group => 2 },
 517  { name => "az-Cyrl", lcid => 0x0000742c, oemcp => 866, ebcdiccp => 20880, group => 5, sabbrevlangname => "AZC" },
 518  { name => "az-Cyrl-AZ", lcid => 0x0000082c },
 519  { name => "az-Latn", lcid => 0x0000782c },
 520  { name => "az-Latn-AZ", lcid => 0x0000042c },
 521  { name => "ba", lcid => 0x0000006d, oemcp => 866, group => 5, dir => "seed", sabbrevlangname => "BAS", sopentypelang => "BSH" },
 522  { name => "ba-Cyrl", alias => "ba" },
 523  { name => "ba-Cyrl-RU", alias => "ba-RU" },
 524  { name => "ba-RU", lcid => 0x0000046d, dir => "seed" },
 525  { name => "bas" },
 526  { name => "bas-CM" },
 527  { name => "be", lcid => 0x00000023, oemcp => 866, ebcdiccp => 500, group => 5 },
 528  { name => "be-BY", lcid => 0x00000423 },
 529  { name => "bem" },
 530  { name => "bem-ZM" },
 531  { name => "bez" },
 532  { name => "bez-TZ" },
 533  { name => "bg", lcid => 0x00000002, oemcp => 866, ebcdiccp => 21025, group => 5, sabbrevlangname => "BGR", sopentypelang => "BGR" },
 534  { name => "bg-BG", lcid => 0x00000402 },
 535  { name => "bin", lcid => 0x00000066, oemcp => 850, dir => "exemplars", sabbrevlangname => "ZZZ", sopentypelang => "EDO" },
 536  { name => "bin-NG", lcid => 0x00000466, file => "bin", dir => "exemplars" },
 537  { name => "bm", sopentypelang => "BMB" },
 538  { name => "bm-Latn", file => "bm" },
 539  { name => "bm-Latn-ML", file => "bm_ML" },
 540  { name => "bm-ML", alias => "bm-Latn-ML" },
 541  { name => "bn", lcid => 0x00000045, slist => ",", group => 15, sabbrevlangname => "BNB" },
 542  { name => "bn-BD", lcid => 0x00000845 },
 543  { name => "bn-IN", lcid => 0x00000445, sabbrevlangname => "BNG" },
 544  { name => "bo", lcid => 0x00000051, slist => ",", group => 15, sabbrevlangname => "BOB", sopentypelang => "TIB" },
 545  { name => "bo-CN", lcid => 0x00000451 },
 546  { name => "bo-IN", slist => "," },
 547  { name => "bo-Tibt", alias => "bo" },
 548  { name => "bo-Tibt-CN", alias => "bo-CN" },
 549  { name => "bo-Tibt-IN", alias => "bo-IN" },
 550  { name => "br", lcid => 0x0000007e, oemcp => 850, ebcdiccp => 20297 },
 551  { name => "br-FR", lcid => 0x0000047e },
 552  { name => "br-Latn", alias => "br" },
 553  { name => "br-Latn-FR", alias => "br-FR" },
 554  { name => "brx" },
 555  { name => "brx-IN" },
 556  { name => "bs", lcid => 0x0000781a, oemcp => 852, maccp => 10082, ebcdiccp => 870, group => 2, sabbrevlangname => "BSB" },
 557  { name => "bs-Cyrl", lcid => 0x0000641a, oemcp => 855, group => 5, sabbrevlangname => "BSC" },
 558  { name => "bs-Cyrl-BA", lcid => 0x0000201a },
 559  { name => "bs-Latn", lcid => 0x0000681a },
 560  { name => "bs-Latn-BA", lcid => 0x0000141a },
 561  { name => "byn", dir => "seed", sopentypelang => "BIL" },
 562  { name => "byn-ER", dir => "seed" },
 563  { name => "ca", lcid => 0x00000003, oemcp => 850 },
 564  { name => "ca-AD", maccp => 65001 },
 565  { name => "ca-ES", lcid => 0x00000403 },
 566  { name => "ca-ES-valencia", lcid => 0x00000803, file => "ca_ES_VALENCIA", sabbrevlangname => "VAL" },
 567  { name => "ca-FR", maccp => 65001 },
 568  { name => "ca-IT", maccp => 65001 },
 569  { name => "ccp" },
 570  { name => "ccp-BD", alias => "ccp-Cakm-BD" },
 571  { name => "ccp-Cakm", file => "ccp" },
 572  { name => "ccp-Cakm-BD", file => "ccp_BD" },
 573  { name => "ccp-Cakm-IN", file => "ccp_IN" },
 574  { name => "ccp-IN", alias => "ccp-Cakm-IN" },
 575  { name => "ce" },
 576  { name => "ce-RU" },
 577  { name => "ceb" },
 578  { name => "ceb-Latn", file => "ceb" },
 579  { name => "ceb-Latn-PH", file => "ceb_PH" },
 580  { name => "ceb-PH", alias => "ceb-Latn-PH" },
 581  { name => "cgg" },
 582  { name => "cgg-UG" },
 583  { name => "chr", lcid => 0x0000005c, slist => ",", sabbrevlangname => "CRE" },
 584  { name => "chr-Cher", lcid => 0x00007c5c, file => "chr" },
 585  { name => "chr-Cher-US", lcid => 0x0000045c, file => "chr_US" },
 586  { name => "chr-US", alias => "chr-Cher-US" },
 587  { name => "ckb", alias => "ku" },
 588  { name => "ckb-IQ", alias => "ku-Arab-IQ" },
 589  { name => "ckb-IR", alias => "ku-Arab-IR" },
 590  { name => "co", lcid => 0x00000083, oemcp => 850, ebcdiccp => 20297, dir => "seed" },
 591  { name => "co-FR", lcid => 0x00000483, dir => "seed" },
 592  { name => "co-Latn", alias => "co" },
 593  { name => "co-Latn-FR", alias => "co-FR" },
 594  { name => "cs", lcid => 0x00000005, oemcp => 852, group => 2, sabbrevlangname => "CSY", sopentypelang => "CSY" },
 595  { name => "cs-CZ", lcid => 0x00000405 },
 596  { name => "cu", dir => "seed", sopentypelang => "CSL" },
 597  { name => "cu-RU", dir => "seed" },
 598  { name => "cy", lcid => 0x00000052, oemcp => 850, ebcdiccp => 20285, sabbrevlangname => "CYM", sopentypelang => "WEL" },
 599  { name => "cy-GB", lcid => 0x00000452 },
 600  { name => "da", lcid => 0x00000006, oemcp => 850, ebcdiccp => 20277 },
 601  { name => "da-DK", lcid => 0x00000406 },
 602  { name => "da-GL", maccp => 65001 },
 603  { name => "dav" },
 604  { name => "dav-KE" },
 605  { name => "de", lcid => 0x00000007, oemcp => 850, ebcdiccp => 20273 },
 606  { name => "de-AT", lcid => 0x00000c07, sabbrevlangname => "DEA" },
 607  { name => "de-BE" },
 608  { name => "de-CH", lcid => 0x00000807, sabbrevlangname => "DES" },
 609  { name => "de-DE", lcid => 0x00000407 },
 610  { name => "de-DE_phoneb", lcid => 0x00010407, alias => "de-DE" },
 611  { name => "de-DE-u-co-phonebk", alias => "de-DE_phoneb" },
 612  { name => "de-IT", oemcp => 65001 },
 613  { name => "de-LI", lcid => 0x00001407, sabbrevlangname => "DEC" },
 614  { name => "de-LU", lcid => 0x00001007, sabbrevlangname => "DEL" },
 615  { name => "dje", sopentypelang => "DJR" },
 616  { name => "dje-NE" },
 617  { name => "doi" },
 618  { name => "doi-IN" },
 619  { name => "dsb", lcid => 0x00007c2e, sparent => "hsb", oemcp => 850, ebcdiccp => 870, sabbrevlangname => "DSB", sopentypelang => "LSB" },
 620  { name => "dsb-DE", lcid => 0x0000082e },
 621  { name => "dua" },
 622  { name => "dua-CM" },
 623  { name => "dv", lcid => 0x00000065, slist => "\x{060c}", group => 13, dir => "seed" },
 624  { name => "dv-MV", lcid => 0x00000465, dir => "seed" },
 625  { name => "dyo" },
 626  { name => "dyo-SN" },
 627  { name => "dz", sopentypelang => "DZN" },
 628  { name => "dz-BT", lcid => 0x00000c51, sabbrevlangname => "ZZZ" },
 629  { name => "ebu" },
 630  { name => "ebu-KE" },
 631  { name => "ee" },
 632  { name => "ee-GH" },
 633  { name => "ee-TG" },
 634  { name => "el", lcid => 0x00000008, oemcp => 737, group => 4 },
 635  { name => "el-CY" },
 636  { name => "el-GR", lcid => 0x00000408 },
 637  { name => "en", lcid => 0x00000009, oemcp => 437, slist => ",", sabbrevlangname => "ENU" },
 638  { name => "en-001", oemcp => 850 },
 639  { name => "en-029", lcid => 0x00002409, file => "en", oemcp => 850, sabbrevlangname => "ENB" },
 640  { name => "en-150", oemcp => 65001 },
 641  { name => "en-AE", lcid => 0x00004c09, oemcp => 65001, sabbrevlangname => "ZZZ" },
 642  { name => "en-AG", oemcp => 850 },
 643  { name => "en-AI", oemcp => 850 },
 644  { name => "en-AS", oemcp => 850 },
 645  { name => "en-AT", oemcp => 65001 },
 646  { name => "en-AU", lcid => 0x00000c09, oemcp => 850, sabbrevlangname => "ENA" },
 647  { name => "en-BB", oemcp => 850 },
 648  { name => "en-BE", oemcp => 850 },
 649  { name => "en-BI", oemcp => 65001 },
 650  { name => "en-BM", oemcp => 850 },
 651  { name => "en-BS", oemcp => 850 },
 652  { name => "en-BW", oemcp => 850 },
 653  { name => "en-BZ", lcid => 0x00002809, oemcp => 850, sabbrevlangname => "ENL" },
 654  { name => "en-CA", lcid => 0x00001009, oemcp => 850, ebcdiccp => 37, sabbrevlangname => "ENC" },
 655  { name => "en-CC", oemcp => 850 },
 656  { name => "en-CH", oemcp => 65001 },
 657  { name => "en-CK", oemcp => 850 },
 658  { name => "en-CM", oemcp => 850 },
 659  { name => "en-CX", oemcp => 850 },
 660  { name => "en-CY", oemcp => 65001 },
 661  { name => "en-DE", oemcp => 65001 },
 662  { name => "en-DG", oemcp => 850 },
 663  { name => "en-DK", oemcp => 65001 },
 664  { name => "en-DM", oemcp => 850 },
 665  { name => "en-ER", oemcp => 850 },
 666  { name => "en-FI", oemcp => 65001 },
 667  { name => "en-FJ", oemcp => 850 },
 668  { name => "en-FK", oemcp => 850 },
 669  { name => "en-FM", oemcp => 850 },
 670  { name => "en-GB", lcid => 0x00000809, oemcp => 850, ebcdiccp => 20285, sabbrevlangname => "ENG" },
 671  { name => "en-GD", oemcp => 850 },
 672  { name => "en-GG", oemcp => 850 },
 673  { name => "en-GH", oemcp => 850 },
 674  { name => "en-GI", oemcp => 850 },
 675  { name => "en-GM", oemcp => 850 },
 676  { name => "en-GU", oemcp => 850 },
 677  { name => "en-GY", oemcp => 850 },
 678  { name => "en-HK", lcid => 0x00003c09, oemcp => 850, sabbrevlangname => "ENH" },
 679  { name => "en-ID", lcid => 0x00003809, file => "en", oemcp => 850, sabbrevlangname => "ZZZ" },
 680  { name => "en-IE", lcid => 0x00001809, oemcp => 850, sabbrevlangname => "ENI" },
 681  { name => "en-IL", oemcp => 65001 },
 682  { name => "en-IM", oemcp => 850 },
 683  { name => "en-IN", lcid => 0x00004009, sabbrevlangname => "ENN" },
 684  { name => "en-IO", oemcp => 850 },
 685  { name => "en-JE", oemcp => 850 },
 686  { name => "en-JM", lcid => 0x00002009, oemcp => 850, sabbrevlangname => "ENJ" },
 687  { name => "en-KE", oemcp => 850 },
 688  { name => "en-KI", oemcp => 850 },
 689  { name => "en-KN", oemcp => 850 },
 690  { name => "en-KY", oemcp => 850 },
 691  { name => "en-LC", oemcp => 850 },
 692  { name => "en-LR", oemcp => 850 },
 693  { name => "en-LS", oemcp => 850 },
 694  { name => "en-MG", oemcp => 850 },
 695  { name => "en-MH", oemcp => 850 },
 696  { name => "en-MO", oemcp => 850 },
 697  { name => "en-MP", oemcp => 850 },
 698  { name => "en-MS", oemcp => 850 },
 699  { name => "en-MT", oemcp => 850 },
 700  { name => "en-MU", oemcp => 850 },
 701  { name => "en-MW", oemcp => 850 },
 702  { name => "en-MY", lcid => 0x00004409, sabbrevlangname => "ENM" },
 703  { name => "en-NA", oemcp => 850 },
 704  { name => "en-NF", oemcp => 850 },
 705  { name => "en-NG", oemcp => 850 },
 706  { name => "en-NL", oemcp => 65001 },
 707  { name => "en-NR", oemcp => 850 },
 708  { name => "en-NU", oemcp => 850 },
 709  { name => "en-NZ", lcid => 0x00001409, oemcp => 850, sabbrevlangname => "ENZ" },
 710  { name => "en-PG", oemcp => 850 },
 711  { name => "en-PH", lcid => 0x00003409, ebcdiccp => 500, sabbrevlangname => "ENP" },
 712  { name => "en-PK", oemcp => 850 },
 713  { name => "en-PN", oemcp => 850 },
 714  { name => "en-PR", oemcp => 850 },
 715  { name => "en-PW", oemcp => 850 },
 716  { name => "en-RW", oemcp => 850 },
 717  { name => "en-SB", oemcp => 850 },
 718  { name => "en-SC", oemcp => 850 },
 719  { name => "en-SD", oemcp => 850 },
 720  { name => "en-SE", oemcp => 65001 },
 721  { name => "en-SG", lcid => 0x00004809, sabbrevlangname => "ENE" },
 722  { name => "en-SH", oemcp => 850 },
 723  { name => "en-SI", oemcp => 65001 },
 724  { name => "en-SL", oemcp => 850 },
 725  { name => "en-SS", oemcp => 850 },
 726  { name => "en-SX", oemcp => 850 },
 727  { name => "en-SZ", oemcp => 850 },
 728  { name => "en-TC", oemcp => 850 },
 729  { name => "en-TK", oemcp => 850 },
 730  { name => "en-TO", oemcp => 850 },
 731  { name => "en-TT", lcid => 0x00002c09, oemcp => 850, sabbrevlangname => "ENT" },
 732  { name => "en-TV", oemcp => 850 },
 733  { name => "en-TZ", oemcp => 850 },
 734  { name => "en-UG", oemcp => 850 },
 735  { name => "en-UM", oemcp => 850 },
 736  { name => "en-US", lcid => 0x00000409 },
 737  { name => "en-VC", oemcp => 850 },
 738  { name => "en-VG", oemcp => 850 },
 739  { name => "en-VI", oemcp => 850 },
 740  { name => "en-VU", oemcp => 850 },
 741  { name => "en-WS", oemcp => 850 },
 742  { name => "en-ZA", lcid => 0x00001c09, ebcdiccp => 500, sabbrevlangname => "ENS" },
 743  { name => "en-ZM", oemcp => 850 },
 744  { name => "en-ZW", lcid => 0x00003009, ebcdiccp => 500, sabbrevlangname => "ENW" },
 745  { name => "eo", sopentypelang => "NTO" },
 746  { name => "eo-001" },
 747  { name => "es", lcid => 0x0000000a, oemcp => 850, ebcdiccp => 20284, sabbrevlangname => "ESP", sopentypelang => "ESP" },
 748  { name => "es-419", lcid => 0x0000580a, sabbrevlangname => "ESJ" },
 749  { name => "es-AR", lcid => 0x00002c0a, sabbrevlangname => "ESS" },
 750  { name => "es-BO", lcid => 0x0000400a, sabbrevlangname => "ESB" },
 751  { name => "es-BR", oemcp => 65001 },
 752  { name => "es-BZ", oemcp => 65001 },
 753  { name => "es-CL", lcid => 0x0000340a, sabbrevlangname => "ESL" },
 754  { name => "es-CO", lcid => 0x0000240a, sabbrevlangname => "ESO" },
 755  { name => "es-CR", lcid => 0x0000140a, sabbrevlangname => "ESC" },
 756  { name => "es-CU", lcid => 0x00005c0a, sabbrevlangname => "ESK" },
 757  { name => "es-DO", lcid => 0x00001c0a, sabbrevlangname => "ESD" },
 758  { name => "es-EA" },
 759  { name => "es-EC", lcid => 0x0000300a, sabbrevlangname => "ESF" },
 760  { name => "es-ES", lcid => 0x00000c0a, sabbrevlangname => "ESN" },
 761  { name => "es-ES_tradnl", lcid => 0x0000040a, file => "es_ES" },
 762  { name => "es-ES-u-co-trad", alias => "es-ES_tradnl" },
 763  { name => "es-GQ" },
 764  { name => "es-GT", lcid => 0x0000100a, sabbrevlangname => "ESG" },
 765  { name => "es-HN", lcid => 0x0000480a, sabbrevlangname => "ESH" },
 766  { name => "es-IC" },
 767  { name => "es-MX", lcid => 0x0000080a, sabbrevlangname => "ESM" },
 768  { name => "es-NI", lcid => 0x00004c0a, sabbrevlangname => "ESI" },
 769  { name => "es-PA", lcid => 0x0000180a, sabbrevlangname => "ESA" },
 770  { name => "es-PE", lcid => 0x0000280a, sabbrevlangname => "ESR" },
 771  { name => "es-PH" },
 772  { name => "es-PR", lcid => 0x0000500a, sabbrevlangname => "ESU" },
 773  { name => "es-PY", lcid => 0x00003c0a, sabbrevlangname => "ESZ" },
 774  { name => "es-SV", lcid => 0x0000440a, sabbrevlangname => "ESE" },
 775  { name => "es-US", lcid => 0x0000540a, sabbrevlangname => "EST" },
 776  { name => "es-UY", lcid => 0x0000380a, sabbrevlangname => "ESY" },
 777  { name => "es-VE", lcid => 0x0000200a, sabbrevlangname => "ESV" },
 778  { name => "et", lcid => 0x00000025, oemcp => 775, group => 3, sabbrevlangname => "ETI", sopentypelang => "ETI" },
 779  { name => "et-EE", lcid => 0x00000425 },
 780  { name => "eu", lcid => 0x0000002d, oemcp => 850, maccp => 65001, sabbrevlangname => "EUQ", sopentypelang => "EUQ" },
 781  { name => "eu-ES", lcid => 0x0000042d },
 782  { name => "ewo" },
 783  { name => "ewo-CM" },
 784  { name => "fa", lcid => 0x00000029, inegnumber => 3, oemcp => 720, slist => "\x{061b}", group => 13, sabbrevlangname => "FAR", sopentypelang => "FAR" },
 785  { name => "fa-AF", alias => "prs-AF" },
 786  { name => "fa-IR", lcid => 0x00000429 },
 787  { name => "ff", lcid => 0x00000067, oemcp => 850, ebcdiccp => 20297 },
 788  { name => "ff-CM", alias => "ff-Latn-CM" },
 789  { name => "ff-GN", alias => "ff-Latn-GN" },
 790  { name => "ff-MR", alias => "ff-Latn-MR" },
 791  { name => "ff-NG", alias => "ff-Latn-NG" },
 792  { name => "ff-SN", alias => "ff-Latn-SN" },
 793  { name => "ff-Adlm" },
 794  { name => "ff-Adlm-BF" },
 795  { name => "ff-Adlm-CM" },
 796  { name => "ff-Adlm-GH" },
 797  { name => "ff-Adlm-GM" },
 798  { name => "ff-Adlm-GN" },
 799  { name => "ff-Adlm-GW" },
 800  { name => "ff-Adlm-LR" },
 801  { name => "ff-Adlm-MR" },
 802  { name => "ff-Adlm-NE" },
 803  { name => "ff-Adlm-NG" },
 804  { name => "ff-Adlm-SL" },
 805  { name => "ff-Adlm-SN" },
 806  { name => "ff-Latn", lcid => 0x00007c67 },
 807  { name => "ff-Latn-BF", oemcp => 65001 },
 808  { name => "ff-Latn-CM" },
 809  { name => "ff-Latn-GH", oemcp => 65001 },
 810  { name => "ff-Latn-GM", oemcp => 65001 },
 811  { name => "ff-Latn-GN" },
 812  { name => "ff-Latn-GW", oemcp => 65001 },
 813  { name => "ff-Latn-LR", oemcp => 65001 },
 814  { name => "ff-Latn-MR" },
 815  { name => "ff-Latn-NE", oemcp => 65001 },
 816  { name => "ff-Latn-NG", lcid => 0x00000467, sabbrevlangname => "ZZZ" },
 817  { name => "ff-Latn-SL", oemcp => 65001 },
 818  { name => "ff-Latn-SN", lcid => 0x00000867 },
 819  { name => "fi", lcid => 0x0000000b, oemcp => 850, ebcdiccp => 20278 },
 820  { name => "fi-FI", lcid => 0x0000040b },
 821  { name => "fil", lcid => 0x00000064, oemcp => 437, ebcdiccp => 500, sabbrevlangname => "FPO", sopentypelang => "PIL" },
 822  { name => "fil-PH", lcid => 0x00000464 },
 823  { name => "fil-Latn", alias => "fil" },
 824  { name => "fil-Latn-PH", alias => "fil-PH" },
 825  { name => "fo", lcid => 0x00000038, oemcp => 850, maccp => 10079, ebcdiccp => 20277, sabbrevlangname => "FOS", sopentypelang => "FOS" },
 826  { name => "fo-DK", oemcp => 65001, maccp => 65001 },
 827  { name => "fo-FO", lcid => 0x00000438 },
 828  { name => "fr", lcid => 0x0000000c, oemcp => 850, ebcdiccp => 20297 },
 829  { name => "fr-029", lcid => 0x00001c0c, file => "fr", sabbrevlangname => "ZZZ" },
 830  { name => "fr-BE", lcid => 0x0000080c, sabbrevlangname => "FRB" },
 831  { name => "fr-BF" },
 832  { name => "fr-BI" },
 833  { name => "fr-BJ" },
 834  { name => "fr-BL" },
 835  { name => "fr-CA", lcid => 0x00000c0c, sabbrevlangname => "FRC" },
 836  { name => "fr-CD", lcid => 0x0000240c, sabbrevlangname => "FRD" },
 837  { name => "fr-CF" },
 838  { name => "fr-CG" },
 839  { name => "fr-CH", lcid => 0x0000100c, sabbrevlangname => "FRS" },
 840  { name => "fr-CI", lcid => 0x0000300c, sabbrevlangname => "FRI" },
 841  { name => "fr-CM", lcid => 0x00002c0c, sabbrevlangname => "FRE" },
 842  { name => "fr-DJ" },
 843  { name => "fr-DZ" },
 844  { name => "fr-FR", lcid => 0x0000040c },
 845  { name => "fr-GA" },
 846  { name => "fr-GF" },
 847  { name => "fr-GN" },
 848  { name => "fr-GP" },
 849  { name => "fr-GQ" },
 850  { name => "fr-HT", lcid => 0x00003c0c, sabbrevlangname => "FRH" },
 851  { name => "fr-KM" },
 852  { name => "fr-LU", lcid => 0x0000140c, sabbrevlangname => "FRL" },
 853  { name => "fr-MA", lcid => 0x0000380c, sabbrevlangname => "FRO" },
 854  { name => "fr-MC", lcid => 0x0000180c, sabbrevlangname => "FRM" },
 855  { name => "fr-MF" },
 856  { name => "fr-MG" },
 857  { name => "fr-ML", lcid => 0x0000340c, sabbrevlangname => "FRF" },
 858  { name => "fr-MQ" },
 859  { name => "fr-MR" },
 860  { name => "fr-MU" },
 861  { name => "fr-NC" },
 862  { name => "fr-NE" },
 863  { name => "fr-PF" },
 864  { name => "fr-PM" },
 865  { name => "fr-RE", lcid => 0x0000200c, sabbrevlangname => "FRR" },
 866  { name => "fr-RW" },
 867  { name => "fr-SC" },
 868  { name => "fr-SN", lcid => 0x0000280c, sabbrevlangname => "FRN" },
 869  { name => "fr-SY" },
 870  { name => "fr-TD" },
 871  { name => "fr-TG" },
 872  { name => "fr-TN" },
 873  { name => "fr-VU" },
 874  { name => "fr-WF" },
 875  { name => "fr-YT" },
 876  { name => "fur", sopentypelang => "FRL" },
 877  { name => "fur-IT" },
 878  { name => "fy", lcid => 0x00000062, oemcp => 850, sabbrevlangname => "FYN", sopentypelang => "FRI" },
 879  { name => "fy-NL", lcid => 0x00000462 },
 880  { name => "ga", lcid => 0x0000003c, oemcp => 850, sabbrevlangname => "IRE", sopentypelang => "IRI" },
 881  { name => "ga-GB" },
 882  { name => "ga-IE", lcid => 0x0000083c },
 883  { name => "gd", lcid => 0x00000091, oemcp => 850, ebcdiccp => 20285, sopentypelang => "GAE" },
 884  { name => "gd-GB", lcid => 0x00000491 },
 885  { name => "gd-Latn", alias => "gd" },
 886  { name => "gl", lcid => 0x00000056, oemcp => 850, sabbrevlangname => "GLC", sopentypelang => "GAL" },
 887  { name => "gl-ES", lcid => 0x00000456 },
 888  { name => "gn", lcid => 0x00000074, oemcp => 850, ebcdiccp => 20284, slist => ",", dir => "seed", sopentypelang => "GUA" },
 889  { name => "gn-PY", lcid => 0x00000474, dir => "seed" },
 890  { name => "gsw", lcid => 0x00000084, oemcp => 850, ebcdiccp => 20297, sabbrevlangname => "ZZZ", sopentypelang => "ALS" },
 891  { name => "gsw-CH" },
 892  { name => "gsw-FR", lcid => 0x00000484, sabbrevlangname => "GSW" },
 893  { name => "gsw-LI" },
 894  { name => "gu", lcid => 0x00000047, slist => ",", group => 15 },
 895  { name => "gu-IN", lcid => 0x00000447 },
 896  { name => "guz" },
 897  { name => "guz-KE" },
 898  { name => "gv", sopentypelang => "MNX" },
 899  { name => "gv-IM" },
 900  { name => "ha", lcid => 0x00000068, oemcp => 437 },
 901  { name => "ha-GH", alias => "ha-Latn-GH" },
 902  { name => "ha-Latn", lcid => 0x00007c68, file => "ha" },
 903  { name => "ha-Latn-GH", file => "ha_GH", ebcdiccp => 500 },
 904  { name => "ha-Latn-NE", file => "ha_NE", ebcdiccp => 500 },
 905  { name => "ha-Latn-NG", lcid => 0x00000468, file => "ha_NG" },
 906  { name => "ha-NE", alias => "ha-Latn-NE" },
 907  { name => "ha-NG", alias => "ha-Latn-NG" },
 908  { name => "haw", lcid => 0x00000075, oemcp => 437 },
 909  { name => "haw-Latn", alias => "haw" },
 910  { name => "haw-Latn-US", alias => "haw-US" },
 911  { name => "haw-US", lcid => 0x00000475 },
 912  { name => "he", lcid => 0x0000000d, oemcp => 862, slist => ",", group => 12, sopentypelang => "IWR" },
 913  { name => "he-IL", lcid => 0x0000040d },
 914  { name => "hi", lcid => 0x00000039, slist => ",", group => 15 },
 915  { name => "hi-IN", lcid => 0x00000439 },
 916  { name => "hr", lcid => 0x0000001a, inegnumber => 2, oemcp => 852, maccp => 10082, group => 2 },
 917  { name => "hr-BA", lcid => 0x0000101a, ebcdiccp => 870, inegnumber => 1, sabbrevlangname => "HRB" },
 918  { name => "hr-HR", lcid => 0x0000041a },
 919  { name => "hsb", lcid => 0x0000002e, oemcp => 850, ebcdiccp => 870, sopentypelang => "USB" },
 920  { name => "hsb-DE", lcid => 0x0000042e },
 921  { name => "hu", lcid => 0x0000000e, oemcp => 852, group => 2 },
 922  { name => "hu-HU", lcid => 0x0000040e },
 923  { name => "hu-HU_technl", lcid => 0x0001040e, alias => "hu-HU" },
 924  { name => "hy", lcid => 0x0000002b, slist => ",", group => 17 },
 925  { name => "hy-AM", lcid => 0x0000042b },
 926  { name => "ia" },
 927  { name => "ia-001" },
 928 ## name => "ibb", lcid => 0x00000069 },
 929 ## name => "ibb-NG", lcid => 0x00000469 },
 930  { name => "id", lcid => 0x00000021, oemcp => 850 },
 931  { name => "id-ID", lcid => 0x00000421 },
 932  { name => "ig", lcid => 0x00000070, oemcp => 437 },
 933  { name => "ig-Latn", alias => "ig" },
 934  { name => "ig-Latn-NG", alias => "ig-NG" },
 935  { name => "ig-NG", lcid => 0x00000470 },
 936  { name => "ii", lcid => 0x00000078, group => 9, sopentypelang => "YIM" },
 937  { name => "ii-CN", lcid => 0x00000478 },
 938  { name => "ii-Yiii", alias => "ii" },
 939  { name => "ii-Yiii-CN", alias => "ii-CN" },
 940  { name => "is", lcid => 0x0000000f, oemcp => 850, maccp => 10079, ebcdiccp => 20871 },
 941  { name => "is-IS", lcid => 0x0000040f },
 942  { name => "it", lcid => 0x00000010, oemcp => 850, ebcdiccp => 20280 },
 943  { name => "it-CH", lcid => 0x00000810, ebcdiccp => 500, sabbrevlangname => "ITS" },
 944  { name => "it-IT", lcid => 0x00000410 },
 945  { name => "it-SM" },
 946  { name => "it-VA", oemcp => 65001 },
 947  { name => "iu", lcid => 0x0000005d, oemcp => 437, slist => ",", sortlocale => "iu-Latn-CA", dir => "seed", sabbrevlangname => "IUK", sopentypelang => "INU" },
 948  { name => "iu-Cans", lcid => 0x0000785d, file => "iu", oemcp => 65001, dir => "seed", sabbrevlangname => "IUS" },
 949  { name => "iu-Cans-CA", lcid => 0x0000045d, file => "iu_CA", dir => "seed" },
 950  { name => "iu-Latn", lcid => 0x00007c5d, dir => "seed" },
 951  { name => "iu-Latn-CA", lcid => 0x0000085d, dir => "seed" },
 952  { name => "ja", lcid => 0x00000011, ireadinglayout => 2, oemcp => 932, slist => ",",  sscripts => "Hani Hira Jpan Kana", group => 7, sopentypelang => "JAN" },
 953  { name => "ja-JP", lcid => 0x00000411 },
 954  { name => "ja-JP_radstr", lcid => 0x00040411, alias => "ja-JP" },
 955  { name => "ja-JP-u-co-unihan", alias => "ja-JP_radstr" },
 956  { name => "jgo" },
 957  { name => "jgo-CM" },
 958  { name => "jmc" },
 959  { name => "jmc-TZ" },
 960  { name => "jv", oemcp => 850 },
 961  { name => "jv-ID", alias => "jv-Latn-ID" },
 962 ## name => "jv-Java" },
 963 ## name => "jv-Java-ID" },
 964  { name => "jv-Latn", file => "jv" },
 965  { name => "jv-Latn-ID", file => "jv_ID" },
 966  { name => "ka", lcid => 0x00000037, group => 16 },
 967  { name => "ka-GE", lcid => 0x00000437 },
 968  { name => "ka-GE_modern", lcid => 0x00010437, alias => "ka-GE" },
 969  { name => "kab", sopentypelang => "KAB0" },
 970  { name => "kab-DZ" },
 971  { name => "kam", sopentypelang => "KMB" },
 972  { name => "kam-KE" },
 973  { name => "kde" },
 974  { name => "kde-TZ" },
 975  { name => "kea" },
 976  { name => "kea-CV" },
 977  { name => "kgp" },
 978  { name => "kgp-BR" },
 979  { name => "khq" },
 980  { name => "khq-ML" },
 981  { name => "ki" },
 982  { name => "ki-KE" },
 983  { name => "kk", lcid => 0x0000003f, group => 5, sabbrevlangname => "KKZ" },
 984  { name => "kk-Cyrl", alias => "kk" },
 985  { name => "kk-Cyrl-KZ", alias => "kk-KZ" },
 986  { name => "kk-KZ", lcid => 0x0000043f },
 987  { name => "kkj" },
 988  { name => "kkj-CM" },
 989  { name => "kl", lcid => 0x0000006f, oemcp => 850, ebcdiccp => 20277, sopentypelang => "GRN" },
 990  { name => "kl-GL", lcid => 0x0000046f },
 991  { name => "kln", sopentypelang => "KAL" },
 992  { name => "kln-KE" },
 993  { name => "km", lcid => 0x00000053, inegnumber => 2, slist => ",", group => 15 },
 994  { name => "km-KH", lcid => 0x00000453 },
 995  { name => "kn", lcid => 0x0000004b, slist => ",", group => 15, sabbrevlangname => "KDI" },
 996  { name => "kn-IN", lcid => 0x0000044b },
 997  { name => "ko", lcid => 0x00000012, ireadinglayout => 2, slist => ",", oemcp => 949, ebcdiccp => 20833, sscripts => "Hang Hani Kore", group => 8 },
 998  { name => "ko-KP", oemcp => 65001 },
 999  { name => "ko-KR", lcid => 0x00000412 },
1000  { name => "kok", lcid => 0x00000057, slist => ",", group => 15, sabbrevlangname => "KNK" },
1001  { name => "kok-IN", lcid => 0x00000457 },
1002  { name => "kr", lcid => 0x00000071, sortlocale => "kr-Latn-NG", oemcp => 850, dir => "exemplars", sabbrevlangname => "ZZZ", sopentypelang => "KNR" },
1003  { name => "kr-Latn", file => "kr", dir => "exemplars" },
1004  { name => "kr-Latn-NG", lcid => 0x00000471, file => "kr", dir => "exemplars" },
1005  { name => "kr-NG", alias => "kr-Latn-NG" },
1006  { name => "ks", lcid => 0x00000060, group => 15, sabbrevlangname => "ZZZ", sopentypelang => "KSH" },
1007  { name => "ks-Arab", lcid => 0x00000460 },
1008  { name => "ks-Arab-IN" },
1009  { name => "ks-Deva", slist => "," },
1010  { name => "ks-Deva-IN", lcid => 0x00000860 },
1011  { name => "ks-IN", alias => "ks-Arab-IN" },
1012  { name => "ksb" },
1013  { name => "ksb-TZ" },
1014  { name => "ksf" },
1015  { name => "ksf-CM" },
1016  { name => "ksh", sopentypelang => "KSH0" },
1017  { name => "ksh-DE" },
1018  { name => "ku", lcid => 0x00000092, file => "ckb", slist => "\x{061b}", sortlocale => "ku-Arab-IQ", oemcp => 720 },
1019  { name => "ku-Arab", lcid => 0x00007c92, file => "ckb", group => 13 },
1020  { name => "ku-Arab-IQ", lcid => 0x00000492, file => "ckb_IQ" },
1021  { name => "ku-Arab-IR", file => "ckb_IR", oemcp => 65001 },
1022  { name => "kw" },
1023  { name => "kw-GB" },
1024  { name => "ky", lcid => 0x00000040, oemcp => 866, group => 5, sabbrevlangname => "KYR" },
1025  { name => "ky-Cyrl", alias => "ky" },
1026  { name => "ky-Cyrl-KG", alias => "ky-KG" },
1027  { name => "ky-KG", lcid => 0x00000440 },
1028  { name => "la", lcid => 0x00000076, oemcp => 437, slist => ",", dir => "seed", sabbrevlangname => "ZZZ" },
1029  { name => "la-001", lcid => 0x00000476, file => "la", dir => "seed" },
1030  { name => "lag" },
1031  { name => "lag-TZ" },
1032  { name => "lb", lcid => 0x0000006e, oemcp => 850, ebcdiccp => 20297, sabbrevlangname => "LBX" },
1033  { name => "lb-LU", lcid => 0x0000046e },
1034  { name => "lg" },
1035  { name => "lg-UG" },
1036  { name => "lkt" },
1037  { name => "lkt-US" },
1038  { name => "ln" },
1039  { name => "ln-AO" },
1040  { name => "ln-CD" },
1041  { name => "ln-CF" },
1042  { name => "ln-CG" },
1043  { name => "lo", lcid => 0x00000054, group => 15 },
1044  { name => "lo-LA", lcid => 0x00000454 },
1045  { name => "lrc" },
1046  { name => "lrc-IQ" },
1047  { name => "lrc-IR" },
1048  { name => "lt", lcid => 0x00000027, oemcp => 775, group => 3, sabbrevlangname => "LTH", sopentypelang => "LTH" },
1049  { name => "lt-LT", lcid => 0x00000427 },
1050  { name => "lu" },
1051  { name => "lu-CD" },
1052  { name => "luo" },
1053  { name => "luo-KE" },
1054  { name => "luy", sopentypelang => "LUH" },
1055  { name => "luy-KE" },
1056  { name => "lv", lcid => 0x00000026, oemcp => 775, group => 3, sabbrevlangname => "LVI", sopentypelang => "LVI" },
1057  { name => "lv-LV", lcid => 0x00000426 },
1058  { name => "mai" },
1059  { name => "mai-IN" },
1060  { name => "mas" },
1061  { name => "mas-KE" },
1062  { name => "mas-TZ" },
1063  { name => "mer" },
1064  { name => "mer-KE" },
1065  { name => "mfe" },
1066  { name => "mfe-MU" },
1067  { name => "mg" },
1068  { name => "mg-MG" },
1069  { name => "mgh" },
1070  { name => "mgh-MZ" },
1071  { name => "mgo" },
1072  { name => "mgo-CM" },
1073  { name => "mi", lcid => 0x00000081, slist => "," },
1074  { name => "mi-Latn", alias => "mi" },
1075  { name => "mi-Latn-NZ", alias => "mi-NZ" },
1076  { name => "mi-NZ", lcid => 0x00000481 },
1077  { name => "mk", lcid => 0x0000002f, oemcp => 866, ebcdiccp => 500, group => 5, sabbrevlangname => "MKI" },
1078  { name => "mk-MK", lcid => 0x0000042f },
1079  { name => "ml", lcid => 0x0000004c, group => 15, sabbrevlangname => "MYM", sopentypelang => "MLR" },
1080  { name => "ml-IN", lcid => 0x0000044c },
1081  { name => "mn", lcid => 0x00000050, oemcp => 866, sopentypelang => "MNG" },
1082  { name => "mn-Cyrl", lcid => 0x00007850, file => "mn", sabbrevlangname => "MNN" },
1083  { name => "mn-Cyrl-MN", alias => "mn-MN" },
1084  { name => "mn-MN", lcid => 0x00000450, sparent => "mn-Cyrl", group => 5 },
1085  { name => "mn-Mong", lcid => 0x00007c50, oemcp => 65001, slist => ",", group => 15, dir => "seed", sabbrevlangname => "MNG" },
1086  { name => "mn-Mong-CN", lcid => 0x00000850, dir => "seed" },
1087  { name => "mn-Mong-MN", lcid => 0x00000c50, dir => "seed", sabbrevlangname => "MNM" },
1088  { name => "mni", lcid => 0x00000058, slist => ",", sabbrevlangname => "ZZZ" },
1089  { name => "mni-IN", lcid => 0x00000458, file => "mni_Beng_IN" },
1090  { name => "moh", lcid => 0x0000007c, oemcp => 850, ebcdiccp => 37, slist => ",", dir => "seed", sabbrevlangname => "MWK" },
1091  { name => "moh-CA", lcid => 0x0000047c, dir => "seed" },
1092  { name => "moh-Latn", alias => "moh" },
1093  { name => "moh-Latn-CA", alias => "moh-CA" },
1094  { name => "mr", lcid => 0x0000004e, slist => ",", group => 15 },
1095  { name => "mr-IN", lcid => 0x0000044e },
1096  { name => "ms", lcid => 0x0000003e, oemcp => 850, sabbrevlangname => "MSL", sopentypelang => "MLY" },
1097  { name => "ms-BN", lcid => 0x0000083e, sabbrevlangname => "MSB" },
1098  { name => "ms-ID" },
1099  { name => "ms-Latn", alias => "ms" },
1100  { name => "ms-Latn-BN", alias => "ms-BN" },
1101  { name => "ms-Latn-MY", alias => "ms-MY" },
1102  { name => "ms-Latn-SG", alias => "ms-SG" },
1103  { name => "ms-MY", lcid => 0x0000043e },
1104  { name => "ms-SG" },
1105  { name => "mt", lcid => 0x0000003a, sopentypelang => "MTS" },
1106  { name => "mt-MT", lcid => 0x0000043a },
1107  { name => "mua" },
1108  { name => "mua-CM" },
1109  { name => "my", lcid => 0x00000055, sopentypelang => "BRM" },
1110  { name => "my-MM", lcid => 0x00000455 },
1111  { name => "mzn" },
1112  { name => "mzn-IR" },
1113  { name => "naq" },
1114  { name => "naq-NA" },
1115  { name => "nb", lcid => 0x00007c14, oemcp => 850, ebcdiccp => 20277, sabbrevlangname => "NOR", sopentypelang => "NOR" },
1116  { name => "nb-NO", lcid => 0x00000414 },
1117  { name => "nb-SJ" },
1118  { name => "nd", sopentypelang => "NDB" },
1119  { name => "nd-ZW" },
1120  { name => "nds" },
1121  { name => "nds-DE" },
1122  { name => "nds-NL" },
1123  { name => "ne", lcid => 0x00000061, slist => "," },
1124  { name => "ne-IN", lcid => 0x00000861, sabbrevlangname => "NEI" },
1125  { name => "ne-NP", lcid => 0x00000461, group => 15 },
1126  { name => "nl", lcid => 0x00000013, oemcp => 850 },
1127  { name => "nl-AW" },
1128  { name => "nl-BE", lcid => 0x00000813, sabbrevlangname => "NLB" },
1129  { name => "nl-BQ" },
1130  { name => "nl-CW" },
1131  { name => "nl-NL", lcid => 0x00000413 },
1132  { name => "nl-SR" },
1133  { name => "nl-SX" },
1134  { name => "nmg" },
1135  { name => "nmg-CM" },
1136  { name => "nn", lcid => 0x00007814, oemcp => 850, ebcdiccp => 20277, sabbrevlangname => "NON", sopentypelang => "NYN" },
1137  { name => "nn-NO", lcid => 0x00000814 },
1138  { name => "nnh" },
1139  { name => "nnh-CM" },
1140  { name => "no", lcid => 0x00000014, oemcp => 850, ebcdiccp => 20277, sortlocale => "nb-NO" },
1141  { name => "nqo", idigits => 3, inegnumber => 3, slist => "\x{060c}", dir => "seed", sopentypelang => "NKO" },
1142  { name => "nqo-GN", dir => "seed" },
1143  { name => "nr", dir => "seed", sopentypelang => "NDB" },
1144  { name => "nr-ZA", dir => "seed" },
1145  { name => "nso", lcid => 0x0000006c, oemcp => 850, dir => "seed", sopentypelang => "SOT" },
1146  { name => "nso-ZA", lcid => 0x0000046c, dir => "seed" },
1147  { name => "nus" },
1148  { name => "nus-SD", alias => "nus-SS" },
1149  { name => "nus-SS" },
1150  { name => "nyn", sopentypelang => "NKL" },
1151  { name => "nyn-UG" },
1152  { name => "oc", lcid => 0x00000082, oemcp => 850, ebcdiccp => 20297, dir => "seed" },
1153  { name => "oc-FR", lcid => 0x00000482, dir => "seed" },
1154  { name => "oc-Latn", alias => "oc" },
1155  { name => "oc-Latn-FR", alias => "oc-FR" },
1156  { name => "om", lcid => 0x00000072, sopentypelang => "ORO" },
1157  { name => "om-ET", lcid => 0x00000472 },
1158  { name => "om-KE" },
1159  { name => "or", lcid => 0x00000048, slist => ",", group => 15 },
1160  { name => "or-IN", lcid => 0x00000448 },
1161  { name => "os" },
1162  { name => "os-GE" },
1163  { name => "os-RU" },
1164  { name => "pa", lcid => 0x00000046, slist => "," },
1165  { name => "pa-Arab", lcid => 0x00007c46, slist => ";", inegnumber => 2, oemcp => 720, group => 13, sabbrevlangname => "PAP" },
1166  { name => "pa-Arab-PK", lcid => 0x00000846 },
1167  { name => "pa-Guru" },
1168  { name => "pa-Guru-IN", alias => "pa-IN" },
1169  { name => "pa-IN", lcid => 0x00000446, sparent => "pa-Guru", file => "pa_Guru_IN", group => 15 },
1170 ## name => "pap", lcid => 0x00000079 },
1171 ## name => "pap-029", lcid => 0x00000479 },
1172  { name => "pcm" },
1173  { name => "pcm-NG" },
1174  { name => "pl", lcid => 0x00000015, oemcp => 852, ebcdiccp => 20880, group => 2, sabbrevlangname => "PLK", sopentypelang => "PLK" },
1175  { name => "pl-PL", lcid => 0x00000415 },
1176 ## name => "prg" },
1177 ## name => "prg-001" },
1178  { name => "prs", lcid => 0x0000008c, file => "fa", inegnumber => 3, oemcp => 720, group => 13, sopentypelang => "DRI" },
1179  { name => "prs-AF", lcid => 0x0000048c, file => "fa_AF" },
1180  { name => "prs-Arab", alias => "prs" },
1181  { name => "prs-Arab-AF", alias => "prs-AF" },
1182  { name => "ps", lcid => 0x00000063, group => 13, sabbrevlangname => "PAS", sopentypelang => "PAS" },
1183  { name => "ps-AF", lcid => 0x00000463 },
1184  { name => "ps-PK" },
1185  { name => "pt", lcid => 0x00000016, oemcp => 850, sabbrevlangname => "PTB", sopentypelang => "PTG" },
1186  { name => "pt-AO" },
1187  { name => "pt-BR", lcid => 0x00000416 },
1188  { name => "pt-CH", oemcp => 65001 },
1189  { name => "pt-CV" },
1190  { name => "pt-GQ", oemcp => 65001 },
1191  { name => "pt-GW" },
1192  { name => "pt-LU", oemcp => 65001 },
1193  { name => "pt-MO" },
1194  { name => "pt-MZ" },
1195  { name => "pt-PT", lcid => 0x00000816, sabbrevlangname => "PTG" },
1196  { name => "pt-ST" },
1197  { name => "pt-TL" },
1198 ## name => qps-Latn-x-sh", lcid => 0x80000901 },
1199 ## name => qps-ploc", lcid => 0x80000501 },
1200 ## name => qps-ploca", lcid => 0x800005fe },
1201 ## name => qps-plocm", lcid => 0x800009ff },
1202  { name => "qu", alias => "quz" },
1203  { name => "qu-BO", alias => "quz-BO" },
1204  { name => "qu-EC", alias => "quz-EC" },
1205  { name => "qu-PE", alias => "quz-PE" },
1206  { name => "quc", lcid => 0x00000086, oemcp => 850, ebcdiccp => 20284, slist => ",", dir => "seed" },
1207  { name => "quc-Latn", lcid => 0x00007c86, file => "quc", dir => "seed" },
1208  { name => "quc-Latn-GT", lcid => 0x00000486, file => "quc_GT", dir => "seed" },
1209  { name => "qut", alias => "quc" },
1210  { name => "qut-GT", alias => "quc-Latn-GT" },
1211  { name => "quz", lcid => 0x0000006b, file => "qu", territory => "BO", oemcp => 850, ebcdiccp => 20284, slist => "," },
1212  { name => "quz-BO", lcid => 0x0000046b, file => "qu_BO" },
1213  { name => "quz-EC", lcid => 0x0000086b, file => "qu_EC" },
1214  { name => "quz-Latn", alias => "quz" },
1215  { name => "quz-Latn-BO", alias => "quz-BO" },
1216  { name => "quz-Latn-EC", alias => "quz-EC" },
1217  { name => "quz-Latn-PE", alias => "quz-PE" },
1218  { name => "quz-PE", lcid => 0x00000c6b, file => "qu_PE" },
1219  { name => "rm", lcid => 0x00000017, oemcp => 850, ebcdiccp => 20273, sabbrevlangname => "RMC", sopentypelang => "RMS" },
1220  { name => "rm-CH", lcid => 0x00000417 },
1221  { name => "rn" },
1222  { name => "rn-BI" },
1223  { name => "ro", lcid => 0x00000018, oemcp => 852, ebcdiccp => 20880, sabbrevlangname => "ROM", sopentypelang => "ROM" },
1224  { name => "ro-MD", lcid => 0x00000818, maccp => 65001, sabbrevlangname => "ROD" },
1225  { name => "ro-RO", lcid => 0x00000418, group => 2 },
1226  { name => "rof" },
1227  { name => "rof-TZ" },
1228  { name => "ru", lcid => 0x00000019, oemcp => 866 },
1229  { name => "ru-BY", maccp => 65001 },
1230  { name => "ru-KG", maccp => 65001 },
1231  { name => "ru-KZ", maccp => 65001 },
1232  { name => "ru-MD", lcid => 0x00000819, maccp => 65001, sabbrevlangname => "RUM" },
1233  { name => "ru-RU", lcid => 0x00000419, group => 5 },
1234  { name => "ru-UA", maccp => 65001 },
1235  { name => "rw", lcid => 0x00000087, oemcp => 437, sopentypelang => "RUA" },
1236  { name => "rw-RW", lcid => 0x00000487 },
1237  { name => "rwk" },
1238  { name => "rwk-TZ" },
1239  { name => "sa", lcid => 0x0000004f, slist => ",", group => 15 },
1240  { name => "sa-Deva", alias => "sa" },
1241  { name => "sa-Deva-IN", alias => "sa-IN" },
1242  { name => "sa-IN", lcid => 0x0000044f },
1243  { name => "sah", lcid => 0x00000085, oemcp => 866, group => 5, sopentypelang => "YAK" },
1244  { name => "sah-Cyrl", alias => "sah" },
1245  { name => "sah-Cyrl-RU", alias => "sah-RU" },
1246  { name => "sah-RU", lcid => 0x00000485 },
1247  { name => "saq" },
1248  { name => "saq-KE" },
1249  { name => "sat" },
1250  { name => "sat-Olck" },
1251  { name => "sat-Olck-IN" },
1252  { name => "sbp" },
1253  { name => "sbp-TZ" },
1254  { name => "sc" },
1255  { name => "sc-IT" },
1256  { name => "sd", lcid => 0x00000059, inegnumber => 3, oemcp => 720, sabbrevlangname => "SIP" },
1257  { name => "sd-Arab", lcid => 0x00007c59, group => 13 },
1258  { name => "sd-Arab-PK", lcid => 0x00000859 },
1259  { name => "sd-Deva", inegnumber => 1, slist => ",", oemcp => 65001, group => 15 },
1260  { name => "sd-Deva-IN", lcid => 0x00000459, sabbrevlangname => "ZZZ" },
1261  { name => "sd-PK", alias => "sd-Arab-PK" },
1262  { name => "se", lcid => 0x0000003b, oemcp => 850, ebcdiccp => 20277, sopentypelang => "NSM" },
1263  { name => "se-FI", lcid => 0x00000c3b, ebcdiccp => 20278, sabbrevlangname => "SMG" },
1264  { name => "se-NO", lcid => 0x0000043b },
1265  { name => "se-SE", lcid => 0x0000083b, ebcdiccp => 20278, sabbrevlangname => "SMF" },
1266  { name => "se-Latn", alias => "se" },
1267  { name => "se-Latn-FI", alias => "se-FI" },
1268  { name => "se-Latn-NO", alias => "se-NO" },
1269  { name => "se-Latn-SE", alias => "se-SE" },
1270  { name => "seh" },
1271  { name => "seh-MZ" },
1272  { name => "ses" },
1273  { name => "ses-ML" },
1274  { name => "sg", sopentypelang => "SGO" },
1275  { name => "sg-CF" },
1276  { name => "shi" },
1277  { name => "shi-Latn" },
1278  { name => "shi-Latn-MA" },
1279  { name => "shi-Tfng" },
1280  { name => "shi-Tfng-MA" },
1281  { name => "si", lcid => 0x0000005b, group => 15, sopentypelang => "SNH" },
1282  { name => "si-LK", lcid => 0x0000045b },
1283  { name => "sk", lcid => 0x0000001b, oemcp => 852, ebcdiccp => 20880, group => 2, sabbrevlangname => "SKY", sopentypelang => "SKY" },
1284  { name => "sk-SK", lcid => 0x0000041b },
1285  { name => "sl", lcid => 0x00000024, oemcp => 852, ebcdiccp => 20880, group => 2 },
1286  { name => "sl-SI", lcid => 0x00000424 },
1287  { name => "sma", lcid => 0x0000783b, sparent => "se", ebcdiccp => 20278, dir => "seed", sabbrevlangname => "SMB", sopentypelang => "SSM" },
1288  { name => "sma-Latn", alias => "sma" },
1289  { name => "sma-Latn-NO", alias => "sma-NO" },
1290  { name => "sma-Latn-SE", alias => "sma-SE" },
1291  { name => "sma-NO", lcid => 0x0000183b, ebcdiccp => 20277, dir => "seed", sabbrevlangname => "SMA" },
1292  { name => "sma-SE", lcid => 0x00001c3b, dir => "seed" },
1293  { name => "smj", lcid => 0x00007c3b, sparent => "se", ebcdiccp => 20278, dir => "seed", sabbrevlangname => "SMK", sopentypelang => "LSM" },
1294  { name => "smj-Latn", alias => "smj" },
1295  { name => "smj-Latn-NO", alias => "smj-NO" },
1296  { name => "smj-Latn-SE", alias => "smj-SE" },
1297  { name => "smj-NO", lcid => 0x0000103b, ebcdiccp => 20277, dir => "seed", sabbrevlangname => "SMJ" },
1298  { name => "smj-SE", lcid => 0x0000143b, dir => "seed" },
1299  { name => "smn", lcid => 0x0000703b, sparent => "se", ebcdiccp => 20278, sopentypelang => "ISM" },
1300  { name => "smn-FI", lcid => 0x0000243b },
1301  { name => "smn-Latn", alias => "smn" },
1302  { name => "smn-Latn-FI", alias => "smn-FI" },
1303  { name => "sms", lcid => 0x0000743b, sparent => "se", ebcdiccp => 20278, dir => "seed", sopentypelang => "SKS" },
1304  { name => "sms-FI", lcid => 0x0000203b, dir => "seed" },
1305  { name => "sms-Latn", alias => "sms" },
1306  { name => "sms-Latn-FI", alias => "sms-FI" },
1307  { name => "sn", sopentypelang => "SNA0" },
1308  { name => "sn-Latn", file => "sn" },
1309  { name => "sn-Latn-ZW", file => "sn_ZW" },
1310  { name => "sn-ZW", alias => "sn-Latn-ZW" },
1311  { name => "so", lcid => 0x00000077, sopentypelang => "SML" },
1312  { name => "so-DJ" },
1313  { name => "so-ET" },
1314  { name => "so-KE" },
1315  { name => "so-SO", lcid => 0x00000477 },
1316  { name => "sq", lcid => 0x0000001c, oemcp => 852, ebcdiccp => 20880, group => 2 },
1317  { name => "sq-AL", lcid => 0x0000041c },
1318  { name => "sq-MK" },
1319  { name => "sq-XK" },
1320  { name => "sr", lcid => 0x00007c1a, sortlocale => "sr-Latn-RS", oemcp => 852, group => 2, sabbrevlangname => "SRB", sopentypelang => "SRB" },
1321  { name => "sr-Cyrl", lcid => 0x00006c1a, oemcp => 855, ebcdiccp => 21025, group => 5, sabbrevlangname => "SRO" },
1322  { name => "sr-Cyrl-BA", lcid => 0x00001c1a, sabbrevlangname => "SRN" },
1323  { name => "sr-Cyrl-ME", lcid => 0x0000301a, sabbrevlangname => "SRQ" },
1324  { name => "sr-Cyrl-RS", lcid => 0x0000281a },
1325  { name => "sr-Cyrl-XK" },
1326  { name => "sr-Latn", lcid => 0x0000701a, sabbrevlangname => "SRM" },
1327  { name => "sr-Latn-BA", lcid => 0x0000181a, maccp => 10082, ebcdiccp => 870, sabbrevlangname => "SRS" },
1328  { name => "sr-Latn-ME", lcid => 0x00002c1a, sabbrevlangname => "SRP" },
1329  { name => "sr-Latn-RS", lcid => 0x0000241a, sabbrevlangname => "SRM" },
1330  { name => "sr-Latn-XK" },
1331 ## name => "sr-Cyrl-CS", lcid => 0x00000c1a },
1332 ## name => "sr-Latn-CS", lcid => 0x0000081a },
1333  { name => "ss", dir => "seed", sopentypelang => "SWZ" },
1334  { name => "ss-SZ", dir => "seed" },
1335  { name => "ss-ZA", dir => "seed" },
1336  { name => "ssy", dir => "seed" },
1337  { name => "ssy-ER", dir => "seed" },
1338  { name => "st", lcid => 0x00000030, dir => "seed" },
1339  { name => "st-LS", dir => "seed" },
1340  { name => "st-ZA", lcid => 0x00000430, dir => "seed" },
1341  { name => "su" },
1342  { name => "su-Latn" },
1343  { name => "su-Latn-ID" },
1344  { name => "sv", lcid => 0x0000001d, oemcp => 850, ebcdiccp => 20278, sabbrevlangname => "SVE", sopentypelang => "SVE" },
1345  { name => "sv-AX" },
1346  { name => "sv-FI", lcid => 0x0000081d, sabbrevlangname => "SVF" },
1347  { name => "sv-SE", lcid => 0x0000041d, sabbrevlangname => "SVE" },
1348  { name => "sw", lcid => 0x00000041, territory => "KE", oemcp => 437, ebcdiccp => 500, sabbrevlangname => "SWK", sopentypelang => "SWK" },
1349  { name => "sw-CD" },
1350  { name => "sw-KE", lcid => 0x00000441 },
1351  { name => "sw-TZ" },
1352  { name => "sw-UG" },
1353  { name => "swc-CD", alias => "sw-CD" },
1354  { name => "syr", lcid => 0x0000005a, slist => ",", group => 13, dir => "seed" },
1355  { name => "syr-SY", lcid => 0x0000045a, dir => "seed" },
1356  { name => "syr-Syrc", alias => "syr" },
1357  { name => "syr-Syrc-SY", alias => "syr-SY" },
1358  { name => "ta", lcid => 0x00000049, slist => ",", group => 15, sabbrevlangname => "TAI" },
1359  { name => "ta-IN", lcid => 0x00000449 },
1360  { name => "ta-LK", lcid => 0x00000849, sabbrevlangname => "TAM" },
1361  { name => "ta-MY" },
1362  { name => "ta-SG" },
1363  { name => "te", lcid => 0x0000004a, group => 15 },
1364  { name => "te-IN", lcid => 0x0000044a },
1365  { name => "teo" },
1366  { name => "teo-KE" },
1367  { name => "teo-UG" },
1368  { name => "tg", lcid => 0x00000028, oemcp => 866, group => 5, sabbrevlangname => "TAJ", sopentypelang => "TAJ" },
1369  { name => "tg-Cyrl", lcid => 0x00007c28, file => "tg" },
1370  { name => "tg-Cyrl-TJ", lcid => 0x00000428, file => "tg_TJ" },
1371  { name => "tg-TJ", alias => "tg-Cyrl-TJ" },
1372  { name => "th", lcid => 0x0000001e, oemcp => 874, ebcdiccp => 20838, slist => ",", group => 11 },
1373  { name => "th-TH", lcid => 0x0000041e },
1374  { name => "ti", lcid => 0x00000073, territory => "ER", sopentypelang => "TGY" },
1375  { name => "ti-ER", lcid => 0x00000873 },
1376  { name => "ti-ET", lcid => 0x00000473, sabbrevlangname => "TIE" },
1377  { name => "tig", dir => "seed", sopentypelang => "TGR" },
1378  { name => "tig-ER", dir => "seed" },
1379  { name => "tk", lcid => 0x00000042, oemcp => 852, ebcdiccp => 20880, group => 2, sopentypelang => "TKM" },
1380  { name => "tk-Latn", alias => "tk" },
1381  { name => "tk-Latn-TM", alias => "tk-TM" },
1382  { name => "tk-TM", lcid => 0x00000442 },
1383  { name => "tn", lcid => 0x00000032, oemcp => 850, dir => "seed", sopentypelang => "TNA" },
1384  { name => "tn-BW", lcid => 0x00000832, dir => "seed", sabbrevlangname => "TSB" },
1385  { name => "tn-ZA", lcid => 0x00000432, dir => "seed" },
1386  { name => "to", sopentypelang => "TGN" },
1387  { name => "to-TO" },
1388  { name => "tr", lcid => 0x0000001f, oemcp => 857, ebcdiccp => 20905, group => 6, sabbrevlangname => "TRK", sopentypelang => "TRK" },
1389  { name => "tr-CY" },
1390  { name => "tr-TR", lcid => 0x0000041f },
1391  { name => "ts", lcid => 0x00000031, dir => "seed", sopentypelang => "TSG" },
1392  { name => "ts-ZA", lcid => 0x00000431, dir => "seed" },
1393  { name => "tt", lcid => 0x00000044, oemcp => 866, group => 5, sabbrevlangname => "TTT" },
1394  { name => "tt-Cyrl", alias => "tt" },
1395  { name => "tt-Cyrl-RU", alias => "tt-RU" },
1396  { name => "tt-RU", lcid => 0x00000444 },
1397  { name => "twq" },
1398  { name => "twq-NE" },
1399  { name => "tzm", lcid => 0x0000005f, sortlocale => "tzm-Latn-DZ", oemcp => 850, ebcdiccp => 20297, sabbrevlangname => "TZA" },
1400  { name => "tzm-Latn", lcid => 0x00007c5f, territory => "DZ", file => "tzm" },
1401  { name => "tzm-Latn-MA", file => "tzm_MA", oemcp => 65001 },
1402  { name => "tzm-Latn-DZ", lcid => 0x0000085f, file => "tzm" },
1403  { name => "tzm-MA", alias => "tzm-Latn-MA" },
1404 ## name => "tzm-Arab", group => 13 },
1405 ## name => "tzm-Arab-MA", lcid => 0x0000045f },
1406 ## name => "tzm-Tfng", lcid => 0x0000785f },
1407 ## name => "tzm-Tfng-MA", lcid => 0x0000105f },
1408  { name => "ug", lcid => 0x00000080, oemcp => 720, slist => ",", group => 13, sopentypelang => "UYG" },
1409  { name => "ug-Arab", alias => "ug" },
1410  { name => "ug-Arab-CN", alias => "ug-CN" },
1411  { name => "ug-CN", lcid => 0x00000480 },
1412  { name => "uk", lcid => 0x00000022, oemcp => 866, maccp => 10017, ebcdiccp => 500, group => 5 },
1413  { name => "uk-UA", lcid => 0x00000422 },
1414  { name => "ur", lcid => 0x00000020, oemcp => 720 },
1415  { name => "ur-IN", lcid => 0x00000820, maccp => 65001, sabbrevlangname => "URI" },
1416  { name => "ur-PK", lcid => 0x00000420, group => 13 },
1417  { name => "uz", lcid => 0x00000043, oemcp => 857, maccp => 10029, group => 2 },
1418  { name => "uz-Arab", oemcp => 65001, maccp => 65001 },
1419  { name => "uz-Arab-AF" },
1420  { name => "uz-Cyrl", lcid => 0x00007843, oemcp => 866, maccp => 10007, group => 5, sabbrevlangname => "UZC" },
1421  { name => "uz-Cyrl-UZ", lcid => 0x00000843 },
1422  { name => "uz-Latn", lcid => 0x00007c43 },
1423  { name => "uz-Latn-UZ", lcid => 0x00000443 },
1424  { name => "vai" },
1425  { name => "vai-Latn" },
1426  { name => "vai-Latn-LR" },
1427  { name => "vai-Vaii" },
1428  { name => "vai-Vaii-LR" },
1429  { name => "ve", lcid => 0x00000033, dir => "seed", sabbrevlangname => "ZZZ" },
1430  { name => "ve-ZA", lcid => 0x00000433, dir => "seed" },
1431  { name => "vi", lcid => 0x0000002a, oemcp => 1258, slist => ",", group => 14, sabbrevlangname => "VIT", sopentypelang => "VIT" },
1432  { name => "vi-VN", lcid => 0x0000042a },
1433  { name => "vo", dir => "seed" },
1434  { name => "vo-001", dir => "seed" },
1435  { name => "vun" },
1436  { name => "vun-TZ" },
1437  { name => "wae" },
1438  { name => "wae-CH" },
1439  { name => "wal", dir => "seed" },
1440  { name => "wal-ET", dir => "seed" },
1441  { name => "wo", lcid => 0x00000088, oemcp => 850, ebcdiccp => 20297, sopentypelang => "WLF" },
1442  { name => "wo-Latn", alias => "wo" },
1443  { name => "wo-Latn-SN", alias => "wo-SN" },
1444  { name => "wo-SN", lcid => 0x00000488 },
1445  { name => "x-IV_mathan", lcid => 0x0001007f, alias => "" },
1446  { name => "xh", lcid => 0x00000034, oemcp => 850, sopentypelang => "XHS" },
1447  { name => "xh-ZA", lcid => 0x00000434 },
1448  { name => "xog" },
1449  { name => "xog-UG" },
1450  { name => "yav" },
1451  { name => "yav-CM" },
1452  { name => "yi", lcid => 0x0000003d, sabbrevlangname => "ZZZ", sopentypelang => "JII" },
1453  { name => "yi-001", lcid => 0x0000043d },
1454  { name => "yo", lcid => 0x0000006a, oemcp => 437, sopentypelang => "YBA" },
1455  { name => "yo-BJ", ebcdiccp => 500 },
1456  { name => "yo-Latn", alias => "yo" },
1457  { name => "yo-Latn-NG", alias => "yo-NG" },
1458  { name => "yo-NG", lcid => 0x0000046a },
1459  { name => "yrl" },
1460  { name => "yrl-BR" },
1461  { name => "yrl-CO" },
1462  { name => "yrl-VE" },
1463  { name => "yue" },
1464  { name => "yue-Hans" },
1465  { name => "yue-Hans-CN" },
1466  { name => "yue-Hant" },
1467  { name => "yue-Hant-HK" },
1468  { name => "zgh" },
1469  { name => "zgh-MA", alias => "zgh-Tfng-MA" },
1470  { name => "zgh-Tfng", file => "zgh" },
1471  { name => "zgh-Tfng-MA", file => "zgh_MA" },
1472  { name => "zh", lcid => 0x00007804, ireadinglayout => 2, oemcp => 936, slist => ",", sscripts => "Hani Hans", sabbrevlangname => "CHS", sopentypelang => "ZHS" },
1473  { name => "zh-CN", lcid => 0x00000804, file => "zh_Hans_CN", sparent => "zh-Hans" },
1474  { name => "zh-CN_phoneb", lcid => 0x00050804, alias => "zh-CN" },
1475  { name => "zh-CN_stroke", lcid => 0x00020804, alias => "zh-CN" },
1476  { name => "zh-Hans", lcid => 0x00000004, group => 10 },
1477  { name => "zh-Hans-CN", alias => "zh-CN" },
1478  { name => "zh-Hans-CN-u-co-phonebk", alias => "zh-CN_phoneb" },
1479  { name => "zh-Hans-CN-u-co-stroke", alias => "zh-CN_stroke" },
1480  { name => "zh-Hans-HK", slist => ";" },
1481  { name => "zh-Hans-MO", slist => ";" },
1482  { name => "zh-Hans-SG", alias => "zh-SG" },
1483  { name => "zh-Hans-SG-u-co-phonebk", alias => "zh-SG_phoneb" },
1484  { name => "zh-Hans-SG-u-co-stroke", alias => "zh-SG_stroke" },
1485  { name => "zh-Hant", lcid => 0x00007c04, sortlocale => "zh-HK", ireadinglayout => 2, oemcp => 950, slist => ",", sscripts => "Hani Hant", group => 9, sabbrevlangname => "CHT", sopentypelang => "ZHH" },
1486  { name => "zh-Hant-HK", alias => "zh-HK" },
1487  { name => "zh-Hant-HK-u-co-unihan", alias => "zh-HK_radstr" },
1488  { name => "zh-Hant-MO", alias => "zh-MO" },
1489  { name => "zh-Hant-MO-u-co-stroke", alias => "zh-MO_stroke" },
1490  { name => "zh-Hant-MO-u-co-unihan", alias => "zh-MO_radstr" },
1491  { name => "zh-Hant-TW", alias => "zh-TW" },
1492  { name => "zh-Hant-TW-u-co-phonetic", alias => "zh-TW_pronun" },
1493  { name => "zh-Hant-TW-u-co-unihan", alias => "zh-TW_radstr" },
1494  { name => "zh-HK", lcid => 0x00000c04, file => "zh_Hant_HK", sparent => "zh-Hant", sabbrevlangname => "ZHH" },
1495  { name => "zh-HK_radstr", lcid => 0x00040c04, alias => "zh-HK" },
1496  { name => "zh-MO", lcid => 0x00001404, file => "zh_Hant_MO", sparent => "zh-Hant", sabbrevlangname => "ZHM", sopentypelang => "ZHT" },
1497  { name => "zh-MO_radstr", lcid => 0x00041404, alias => "zh-MO" },
1498  { name => "zh-MO_stroke", lcid => 0x00021404, alias => "zh-MO" },
1499  { name => "zh-SG", lcid => 0x00001004, file => "zh_Hans_SG", sparent => "zh-Hans", sabbrevlangname => "ZHI" },
1500  { name => "zh-SG_phoneb", lcid => 0x00051004, alias => "zh-SG" },
1501  { name => "zh-SG_stroke", lcid => 0x00021004, alias => "zh-SG" },
1502  { name => "zh-TW", lcid => 0x00000404, file => "zh_Hant_TW", sparent => "zh-Hant", sopentypelang => "ZHT" },
1503  { name => "zh-TW_pronun", lcid => 0x00030404, alias => "zh-TW" },
1504  { name => "zh-TW_radstr", lcid => 0x00040404, alias => "zh-TW" },
1505  { name => "zu", lcid => 0x00000035, oemcp => 850 },
1506  { name => "zu-ZA", lcid => 0x00000435 },
1507 );
1508
1509 my @calendars =
1510 (
1511  { id => 1, name => "Gregorian", itwodigityearmax => 2049 },
1512  { id => 2, type => "gregorian", locale => "en-US", itwodigityearmax => 2049 },
1513  { id => 3, type => "japanese", locale => "ja-JP", eras => [ 232..236 ] },
1514  { id => 4, type => "roc", locale => "zh-TW", eras => [ 1 ] },
1515  { id => 5, type => "dangi", locale => "ko-KR", eras => [ 0 ] },
1516  { id => 6, type => "islamic", locale => "ar-SA", itwodigityearmax => 1451 },
1517  { id => 7, type => "buddhist", locale => "th-TH", eras => [ 0 ] },
1518  { id => 8, type => "hebrew", locale => "he-IL", itwodigityearmax => 5810 },
1519  { id => 9, type => "gregorian", locale => "fr-FR", itwodigityearmax => 2049 },
1520  { id => 10, type => "gregorian", locale => "ar-SA", itwodigityearmax => 2049 },
1521  { id => 11, type => "gregorian", locale => "ar-SA", itwodigityearmax => 2049 },
1522  { id => 12, type => "gregorian", locale => "ar-SA", itwodigityearmax => 2049 },
1523  { id => 13, name => "Julian", locale => "en-US", itwodigityearmax => 2049 },
1524  { id => 14, name => "Japanese Lunisolar" },
1525  { id => 15, name => "Chinese Lunisolar" },
1526  { id => 16, name => "Saka" },
1527  { id => 17, name => "Lunar ETO Chinese" },
1528  { id => 18, name => "Lunar ETO Korean" },
1529  { id => 19, name => "Lunar ETO Rokuyou" },
1530  { id => 20, name => "Korean Lunisolar" },
1531  { id => 21, name => "Taiwan Lunisolar" },
1532  { id => 22, type => "persian", locale => "prs-AF", itwodigityearmax => 1429 },
1533  { id => 23, type => "islamic-umalqura", locale => "ar-SA", itwodigityearmax => 1451 },
1534 );
1535
1536 my @geoids =
1537 (
1538  { id => 2,   name => "AG" },     # Antigua and Barbuda
1539  { id => 3,   name => "AF" },     # Afghanistan
1540  { id => 4,   name => "DZ" },     # Algeria
1541  { id => 5,   name => "AZ" },     # Azerbaijan
1542  { id => 6,   name => "AL" },     # Albania
1543  { id => 7,   name => "AM" },     # Armenia
1544  { id => 8,   name => "AD" },     # Andorra
1545  { id => 9,   name => "AO" },     # Angola
1546  { id => 10,  name => "AS" },     # American Samoa
1547  { id => 11,  name => "AR" },     # Argentina
1548  { id => 12,  name => "AU" },     # Australia
1549  { id => 14,  name => "AT" },     # Austria
1550  { id => 17,  name => "BH" },     # Bahrain
1551  { id => 18,  name => "BB" },     # Barbados
1552  { id => 19,  name => "BW" },     # Botswana
1553  { id => 20,  name => "BM" },     # Bermuda
1554  { id => 21,  name => "BE" },     # Belgium
1555  { id => 22,  name => "BS" },     # Bahamas, The
1556  { id => 23,  name => "BD" },     # Bangladesh
1557  { id => 24,  name => "BZ" },     # Belize
1558  { id => 25,  name => "BA" },     # Bosnia and Herzegovina
1559  { id => 26,  name => "BO" },     # Bolivia
1560  { id => 27,  name => "MM" },     # Myanmar
1561  { id => 28,  name => "BJ" },     # Benin
1562  { id => 29,  name => "BY" },     # Belarus
1563  { id => 30,  name => "SB" },     # Solomon Islands
1564  { id => 32,  name => "BR" },     # Brazil
1565  { id => 34,  name => "BT" },     # Bhutan
1566  { id => 35,  name => "BG" },     # Bulgaria
1567  { id => 37,  name => "BN" },     # Brunei
1568  { id => 38,  name => "BI" },     # Burundi
1569  { id => 39,  name => "CA" },     # Canada
1570  { id => 40,  name => "KH" },     # Cambodia
1571  { id => 41,  name => "TD" },     # Chad
1572  { id => 42,  name => "LK" },     # Sri Lanka
1573  { id => 43,  name => "CG" },     # Congo
1574  { id => 44,  name => "CD" },     # Congo (DRC)
1575  { id => 45,  name => "CN" },     # China
1576  { id => 46,  name => "CL" },     # Chile
1577  { id => 49,  name => "CM" },     # Cameroon
1578  { id => 50,  name => "KM" },     # Comoros
1579  { id => 51,  name => "CO" },     # Colombia
1580  { id => 54,  name => "CR" },     # Costa Rica
1581  { id => 55,  name => "CF" },     # Central African Republic
1582  { id => 56,  name => "CU" },     # Cuba
1583  { id => 57,  name => "CV" },     # Cape Verde
1584  { id => 59,  name => "CY" },     # Cyprus
1585  { id => 61,  name => "DK" },     # Denmark
1586  { id => 62,  name => "DJ" },     # Djibouti
1587  { id => 63,  name => "DM" },     # Dominica
1588  { id => 65,  name => "DO" },     # Dominican Republic
1589  { id => 66,  name => "EC" },     # Ecuador
1590  { id => 67,  name => "EG" },     # Egypt
1591  { id => 68,  name => "IE" },     # Ireland
1592  { id => 69,  name => "GQ" },     # Equatorial Guinea
1593  { id => 70,  name => "EE" },     # Estonia
1594  { id => 71,  name => "ER" },     # Eritrea
1595  { id => 72,  name => "SV" },     # El Salvador
1596  { id => 73,  name => "ET" },     # Ethiopia
1597  { id => 75,  name => "CZ" },     # Czech Republic
1598  { id => 77,  name => "FI" },     # Finland
1599  { id => 78,  name => "FJ" },     # Fiji Islands
1600  { id => 80,  name => "FM" },     # Micronesia
1601  { id => 81,  name => "FO" },     # Faroe Islands
1602  { id => 84,  name => "FR" },     # France
1603  { id => 86,  name => "GM" },     # Gambia, The
1604  { id => 87,  name => "GA" },     # Gabon
1605  { id => 88,  name => "GE" },     # Georgia
1606  { id => 89,  name => "GH" },     # Ghana
1607  { id => 90,  name => "GI" },     # Gibraltar
1608  { id => 91,  name => "GD" },     # Grenada
1609  { id => 93,  name => "GL" },     # Greenland
1610  { id => 94,  name => "DE" },     # Germany
1611  { id => 98,  name => "GR" },     # Greece
1612  { id => 99,  name => "GT" },     # Guatemala
1613  { id => 100, name => "GN" },     # Guinea
1614  { id => 101, name => "GY" },     # Guyana
1615  { id => 103, name => "HT" },     # Haiti
1616  { id => 104, name => "HK" },     # Hong Kong S.A.R.
1617  { id => 106, name => "HN" },     # Honduras
1618  { id => 108, name => "HR" },     # Croatia
1619  { id => 109, name => "HU" },     # Hungary
1620  { id => 110, name => "IS" },     # Iceland
1621  { id => 111, name => "ID" },     # Indonesia
1622  { id => 113, name => "IN" },     # India
1623  { id => 114, name => "IO" },     # British Indian Ocean Territory
1624  { id => 116, name => "IR" },     # Iran
1625  { id => 117, name => "IL" },     # Israel
1626  { id => 118, name => "IT" },     # Italy
1627  { id => 119, name => "CI" },     # Côte d'Ivoire
1628  { id => 121, name => "IQ" },     # Iraq
1629  { id => 122, name => "JP" },     # Japan
1630  { id => 124, name => "JM" },     # Jamaica
1631  { id => 125, name => "SJ" },     # Jan Mayen
1632  { id => 126, name => "JO" },     # Jordan
1633  { id => 127, parent => "UM" },   # Johnston Atoll
1634  { id => 129, name => "KE" },     # Kenya
1635  { id => 130, name => "KG" },     # Kyrgyzstan
1636  { id => 131, name => "KP" },     # North Korea
1637  { id => 133, name => "KI" },     # Kiribati
1638  { id => 134, name => "KR" },     # Korea
1639  { id => 136, name => "KW" },     # Kuwait
1640  { id => 137, name => "KZ" },     # Kazakhstan
1641  { id => 138, name => "LA" },     # Laos
1642  { id => 139, name => "LB" },     # Lebanon
1643  { id => 140, name => "LV" },     # Latvia
1644  { id => 141, name => "LT" },     # Lithuania
1645  { id => 142, name => "LR" },     # Liberia
1646  { id => 143, name => "SK" },     # Slovakia
1647  { id => 145, name => "LI" },     # Liechtenstein
1648  { id => 146, name => "LS" },     # Lesotho
1649  { id => 147, name => "LU" },     # Luxembourg
1650  { id => 148, name => "LY" },     # Libya
1651  { id => 149, name => "MG" },     # Madagascar
1652  { id => 151, name => "MO" },     # Macao S.A.R.
1653  { id => 152, name => "MD" },     # Moldova
1654  { id => 154, name => "MN" },     # Mongolia
1655  { id => 156, name => "MW" },     # Malawi
1656  { id => 157, name => "ML" },     # Mali
1657  { id => 158, name => "MC" },     # Monaco
1658  { id => 159, name => "MA" },     # Morocco
1659  { id => 160, name => "MU" },     # Mauritius
1660  { id => 162, name => "MR" },     # Mauritania
1661  { id => 163, name => "MT" },     # Malta
1662  { id => 164, name => "OM" },     # Oman
1663  { id => 165, name => "MV" },     # Maldives
1664  { id => 166, name => "MX" },     # Mexico
1665  { id => 167, name => "MY" },     # Malaysia
1666  { id => 168, name => "MZ" },     # Mozambique
1667  { id => 173, name => "NE" },     # Niger
1668  { id => 174, name => "VU" },     # Vanuatu
1669  { id => 175, name => "NG" },     # Nigeria
1670  { id => 176, name => "NL" },     # Netherlands
1671  { id => 177, name => "NO" },     # Norway
1672  { id => 178, name => "NP" },     # Nepal
1673  { id => 180, name => "NR" },     # Nauru
1674  { id => 181, name => "SR" },     # Suriname
1675  { id => 182, name => "NI" },     # Nicaragua
1676  { id => 183, name => "NZ" },     # New Zealand
1677  { id => 184, name => "PS" },     # Palestinian Authority
1678  { id => 185, name => "PY" },     # Paraguay
1679  { id => 187, name => "PE" },     # Peru
1680  { id => 190, name => "PK" },     # Pakistan
1681  { id => 191, name => "PL" },     # Poland
1682  { id => 192, name => "PA" },     # Panama
1683  { id => 193, name => "PT" },     # Portugal
1684  { id => 194, name => "PG" },     # Papua New Guinea
1685  { id => 195, name => "PW" },     # Palau
1686  { id => 196, name => "GW" },     # Guinea-Bissau
1687  { id => 197, name => "QA" },     # Qatar
1688  { id => 198, name => "RE" },     # Reunion
1689  { id => 199, name => "MH" },     # Marshall Islands
1690  { id => 200, name => "RO" },     # Romania
1691  { id => 201, name => "PH" },     # Philippines
1692  { id => 202, name => "PR" },     # Puerto Rico
1693  { id => 203, name => "RU" },     # Russia
1694  { id => 204, name => "RW" },     # Rwanda
1695  { id => 205, name => "SA" },     # Saudi Arabia
1696  { id => 206, name => "PM" },     # St. Pierre and Miquelon
1697  { id => 207, name => "KN" },     # St. Kitts and Nevis
1698  { id => 208, name => "SC" },     # Seychelles
1699  { id => 209, name => "ZA" },     # South Africa
1700  { id => 210, name => "SN" },     # Senegal
1701  { id => 212, name => "SI" },     # Slovenia
1702  { id => 213, name => "SL" },     # Sierra Leone
1703  { id => 214, name => "SM" },     # San Marino
1704  { id => 215, name => "SG" },     # Singapore
1705  { id => 216, name => "SO" },     # Somalia
1706  { id => 217, name => "ES" },     # Spain
1707  { id => 218, name => "LC" },     # St. Lucia
1708  { id => 219, name => "SD" },     # Sudan
1709  { id => 220, name => "SJ" },     # Svalbard
1710  { id => 221, name => "SE" },     # Sweden
1711  { id => 222, name => "SY" },     # Syria
1712  { id => 223, name => "CH" },     # Switzerland
1713  { id => 224, name => "AE" },     # United Arab Emirates
1714  { id => 225, name => "TT" },     # Trinidad and Tobago
1715  { id => 227, name => "TH" },     # Thailand
1716  { id => 228, name => "TJ" },     # Tajikistan
1717  { id => 231, name => "TO" },     # Tonga
1718  { id => 232, name => "TG" },     # Togo
1719  { id => 233, name => "ST" },     # São Tomé and Príncipe
1720  { id => 234, name => "TN" },     # Tunisia
1721  { id => 235, name => "TR" },     # Turkey
1722  { id => 236, name => "TV" },     # Tuvalu
1723  { id => 237, name => "TW" },     # Taiwan
1724  { id => 238, name => "TM" },     # Turkmenistan
1725  { id => 239, name => "TZ" },     # Tanzania
1726  { id => 240, name => "UG" },     # Uganda
1727  { id => 241, name => "UA" },     # Ukraine
1728  { id => 242, name => "GB" },     # United Kingdom
1729  { id => 244, name => "US" },     # United States
1730  { id => 245, name => "BF" },     # Burkina Faso
1731  { id => 246, name => "UY" },     # Uruguay
1732  { id => 247, name => "UZ" },     # Uzbekistan
1733  { id => 248, name => "VC" },     # St. Vincent and the Grenadines
1734  { id => 249, name => "VE" },     # Bolivarian Republic of Venezuela
1735  { id => 251, name => "VN" },     # Vietnam
1736  { id => 252, name => "VI" },     # Virgin Islands
1737  { id => 253, name => "VA" },     # Vatican City
1738  { id => 254, name => "NA" },     # Namibia
1739  { id => 257, name => "EH" },     # Western Sahara (disputed)
1740  { id => 258, parent => "UM" },   # Wake Island
1741  { id => 259, name => "WS" },     # Samoa
1742  { id => 260, name => "SZ" },     # Swaziland
1743  { id => 261, name => "YE" },     # Yemen
1744  { id => 263, name => "ZM" },     # Zambia
1745  { id => 264, name => "ZW" },     # Zimbabwe
1746  { id => 269, name => "CS" },     # Serbia and Montenegro (Former)
1747  { id => 270, name => "ME" },     # Montenegro
1748  { id => 271, name => "RS" },     # Serbia
1749  { id => 273, name => "CW" },     # Curaçao
1750  { id => 276, name => "SS" },     # South Sudan
1751  { id => 300, name => "AI" },     # Anguilla
1752  { id => 301, name => "AQ" },     # Antarctica
1753  { id => 302, name => "AW" },     # Aruba
1754  { id => 303, parent => "SH" },   # Ascension Island
1755  { id => 304, parent => "053" },  # Ashmore and Cartier Islands
1756  { id => 305, parent => "UM" },   # Baker Island
1757  { id => 306, name => "BV" },     # Bouvet Island
1758  { id => 307, name => "KY" },     # Cayman Islands
1759  { id => 308, name => "830", parent => "155" }, # Channel Islands
1760  { id => 309, name => "CX" },     # Christmas Island
1761  { id => 310, parent => "009" },  # Clipperton Island
1762  { id => 311, name => "CC" },     # Cocos (Keeling) Islands
1763  { id => 312, name => "CK" },     # Cook Islands
1764  { id => 313, parent => "053" },  # Coral Sea Islands
1765  { id => 314, parent => "IO" },   # Diego Garcia
1766  { id => 315, name => "FK" },     # Falkland Islands (Islas Malvinas)
1767  { id => 317, name => "GF" },     # French Guiana
1768  { id => 318, name => "PF" },     # French Polynesia
1769  { id => 319, name => "TF" },     # French Southern and Antarctic Lands
1770  { id => 321, name => "GP" },     # Guadeloupe
1771  { id => 322, name => "GU" },     # Guam
1772  { id => 323 },                   # Guantanamo Bay
1773  { id => 324, name => "GG" },     # Guernsey
1774  { id => 325, name => "HM" },     # Heard Island and McDonald Islands
1775  { id => 326, parent => "UM" },   # Howland Island
1776  { id => 327, parent => "UM" },   # Jarvis Island
1777  { id => 328, name => "JE" },     # Jersey
1778  { id => 329, parent => "UM" },   # Kingman Reef
1779  { id => 330, name => "MQ" },     # Martinique
1780  { id => 331, name => "YT" },     # Mayotte
1781  { id => 332, name => "MS" },     # Montserrat
1782  { id => 333, name => "AN", region => 1 }, # Netherlands Antilles (Former)
1783  { id => 334, name => "NC" },     # New Caledonia
1784  { id => 335, name => "NU" },     # Niue
1785  { id => 336, name => "NF" },     # Norfolk Island
1786  { id => 337, name => "MP" },     # Northern Mariana Islands
1787  { id => 338, parent => "UM" },   # Palmyra Atoll
1788  { id => 339, name => "PN" },     # Pitcairn Islands
1789  { id => 340, parent => "MP" },   # Rota Island
1790  { id => 341, parent => "MP" },   # Saipan
1791  { id => 342, name => "GS" },     # South Georgia and the South Sandwich Islands
1792  { id => 343, name => "SH" },     # St. Helena
1793  { id => 346, parent => "MP" },   # Tinian Island
1794  { id => 347, name => "TK" },     # Tokelau
1795  { id => 348, parent => "SH" },   # Tristan da Cunha
1796  { id => 349, name => "TC" },     # Turks and Caicos Islands
1797  { id => 351, name => "VG" },     # Virgin Islands, British
1798  { id => 352, name => "WF" },     # Wallis and Futuna
1799  { id => 742, name => "002" },    # Africa
1800  { id => 2129, name => "142" },   # Asia
1801  { id => 10541, name => "150" },  # Europe
1802  { id => 15126, name => "IM" },   # Man, Isle of
1803  { id => 19618, name => "MK" },   # Macedonia, Former Yugoslav Republic of
1804  { id => 20900, name => "054" },  # Melanesia
1805  { id => 21206, name => "057" },  # Micronesia
1806  { id => 21242, parent => "UM" }, # Midway Islands
1807  { id => 23581, name => "021" },  # Northern America
1808  { id => 26286, name => "061" },  # Polynesia
1809  { id => 27082, name => "013" },  # Central America
1810  { id => 27114, name => "009" },  # Oceania
1811  { id => 30967, name => "SX" },   # Sint Maarten (Dutch part)
1812  { id => 31396, name => "005" },  # South America
1813  { id => 31706, name => "MF" },   # Saint Martin (French part)
1814  { id => 39070, name => "001" },  # World
1815  { id => 42483, name => "011" },  # Western Africa
1816  { id => 42484, name => "017" },  # Middle Africa
1817  { id => 42487, name => "015" },  # Northern Africa
1818  { id => 47590, name => "143" },  # Central Asia
1819  { id => 47599, name => "035" },  # South-Eastern Asia
1820  { id => 47600, name => "030" },  # Eastern Asia
1821  { id => 47603, name => "014" },  # Eastern Africa
1822  { id => 47609, name => "151" },  # Eastern Europe
1823  { id => 47610, name => "039" },  # Southern Europe
1824  { id => 47611, name => "145" },  # Middle East
1825  { id => 47614, name => "034" },  # Southern Asia
1826  { id => 7299303, name => "TL" }, # Democratic Republic of Timor-Leste
1827  { id => 9914689, name => "XK" }, # Kosovo
1828  { id => 10026358, name => "019" }, # Americas
1829  { id => 10028789, name => "AX" },  # Åland Islands
1830  { id => 10039880, name => "029", sintlsymbol => "XCD" }, # Caribbean
1831  { id => 10039882, name => "154" }, # Northern Europe
1832  { id => 10039883, name => "018" }, # Southern Africa
1833  { id => 10210824, name => "155" }, # Western Europe
1834  { id => 10210825, name => "053" }, # Australia and New Zealand
1835  { id => 161832015, name => "BL" }, # Saint Barthélemy
1836  { id => 161832256, name => "UM" }, # U.S. Minor Outlying Islands
1837  { id => 161832257, name => "419", parent => "019" }, # Latin America and the Caribbean
1838  { id => 161832258, name => "BQ" },   # Bonaire, Sint Eustatius and Saba
1839 );
1840
1841 my @cp2uni = ();
1842 my @glyph2uni = ();
1843 my @lead_bytes = ();
1844 my @uni2cp = ();
1845 my @tolower_table = ();
1846 my @toupper_table = ();
1847 my @digitmap_table = ();
1848 my @halfwidth_table = ();
1849 my @fullwidth_table = ();
1850 my @cjk_compat_table = ();
1851 my @chinese_traditional_table = ();
1852 my @chinese_simplified_table = ();
1853 my @category_table = ();
1854 my @initial_joining_table = ();
1855 my @direction_table = ();
1856 my @decomp_table = ();
1857 my @combining_class_table = ();
1858 my @decomp_compat_table = ();
1859 my @comp_exclusions = ();
1860 my @idna_decomp_table = ();
1861 my @idna_disallowed = ();
1862 my %registry_keys;
1863 my $default_char;
1864 my $default_wchar;
1865
1866 my %joining_forms =
1867 (
1868    "isolated" => [],
1869    "final" => [],
1870    "initial" => [],
1871    "medial" => []
1872 );
1873
1874 my $current_data_file;
1875
1876 sub to_utf16(@)
1877 {
1878     my @ret;
1879     foreach my $ch (@_)
1880     {
1881         if ($ch < 0x10000)
1882         {
1883             push @ret, $ch;
1884         }
1885         else
1886         {
1887             my $val = $ch - 0x10000;
1888             push @ret, 0xd800 | ($val >> 10), 0xdc00 | ($val & 0x3ff);
1889         }
1890     }
1891     return @ret;
1892 }
1893
1894 ################################################################
1895 # fetch a unicode.org file and open it
1896 sub open_data_file($@)
1897 {
1898     my ($id, $name) = @_;
1899     my $data = $data_files{$id};
1900     my $cache = ($ENV{XDG_CACHE_HOME} || "$ENV{HOME}/.cache") . "/wine";
1901     local *FILE;
1902
1903     my $url = $data->{url};
1904     my $filename = "$cache/" . ($data->{name} || ($url =~ s/.*\/([^\/]+)$/$1/r));
1905     unless (-f $filename)
1906     {
1907         print "Fetching $url...\n";
1908         system "mkdir", "-p", $cache;
1909         !system "wget", "-q", "-O", $filename, $url or die "cannot fetch $url";
1910     }
1911
1912     my $sha = Digest::SHA->new( "sha256" )->addfile( $filename )->hexdigest;
1913     die "invalid checksum $sha for $filename" unless $sha eq $data->{sha};
1914
1915     if ($filename =~ /\.zip$/)
1916     {
1917         open FILE, "-|", "unzip", "-p", $filename, $name or die "cannot extract $name from $filename";
1918     }
1919     elsif ($filename =~ /\.tar\.gz$/)
1920     {
1921         open FILE, "-|", "tar", "-x", "-f", $filename, "-O", $name or die "cannot extract $name from $filename";
1922     }
1923     else
1924     {
1925         open FILE, "<$filename" or die "cannot open $filename";
1926     }
1927     $current_data_file = $name ? "$url:$name" : $url;
1928     return *FILE;
1929 }
1930
1931 ################################################################
1932 # load a unicode.org file as XML data
1933 sub load_xml_data_file($@)
1934 {
1935     my ($id, $name) = @_;
1936     my $FILE = open_data_file( $id, $name );
1937     my $xml = XML::LibXML->load_xml( IO => $FILE );
1938     close FILE;
1939     return $xml;
1940 }
1941
1942 ################################################################
1943 # recursively get the decomposition for a character
1944 sub get_decomposition($$);
1945 sub get_decomposition($$)
1946 {
1947     my ($char, $table) = @_;
1948     my @ret;
1949
1950     return $char unless defined $table->[$char];
1951     foreach my $ch (@{$table->[$char]})
1952     {
1953         push @ret, get_decomposition( $ch, $table );
1954     }
1955     return @ret;
1956 }
1957
1958 ################################################################
1959 # get the composition that results in a given character
1960 sub get_composition($$)
1961 {
1962     my ($ch, $compat) = @_;
1963     return () unless defined $decomp_table[$ch];  # no decomposition
1964     my @ret = @{$decomp_table[$ch]};
1965     return () if @ret < 2;                        # singleton decomposition
1966     return () if $comp_exclusions[$ch];           # composition exclusion
1967     return () if $combining_class_table[$ch];     # non-starter
1968     return () if $combining_class_table[$ret[0]]; # first char is non-starter
1969     return () if $compat == 1 && !defined $decomp_table[$ret[0]] &&
1970         defined $decomp_compat_table[$ret[0]];    # first char has compat decomposition
1971     return () if $compat == 2 && !defined $decomp_table[$ret[0]] &&
1972         defined $idna_decomp_table[$ret[0]];      # first char has IDNA decomposition
1973     return () if $compat == 2 && defined $idna_decomp_table[$ret[0]] &&
1974         defined $idna_decomp_table[$idna_decomp_table[$ret[0]]->[0]];  # first char's decomposition has IDNA decomposition
1975     return () if $compat == 2 && defined $idna_decomp_table[$ret[1]];  # second char has IDNA decomposition
1976     return @ret;
1977 }
1978
1979 ################################################################
1980 # recursively build decompositions
1981 sub build_decompositions(@)
1982 {
1983     my @src = @_;
1984     my @dst;
1985
1986     for (my $i = 0; $i < @src; $i++)
1987     {
1988         next unless defined $src[$i];
1989         my @decomp = to_utf16( get_decomposition( $i, \@src ));
1990         $dst[$i] = \@decomp;
1991     }
1992     return @dst;
1993 }
1994
1995 ################################################################
1996 # compose Hangul sequences
1997 sub compose_hangul(@)
1998 {
1999     my $SBASE  = 0xac00;
2000     my $LBASE  = 0x1100;
2001     my $VBASE  = 0x1161;
2002     my $TBASE  = 0x11a7;
2003     my $LCOUNT = 19;
2004     my $VCOUNT = 21;
2005     my $TCOUNT = 28;
2006     my $NCOUNT = $VCOUNT * $TCOUNT;
2007     my $SCOUNT = $LCOUNT * $NCOUNT;
2008
2009     my @seq = @_;
2010     my @ret;
2011     my $i;
2012
2013     for ($i = 0; $i < @seq; $i++)
2014     {
2015         my $ch = $seq[$i];
2016         if ($ch >= $LBASE && $ch < $LBASE + $LCOUNT && $i < @seq - 1 &&
2017             $seq[$i+1] >= $VBASE && $seq[$i+1] < $VBASE + $VCOUNT)
2018         {
2019             $ch = $SBASE + (($seq[$i] - $LBASE) * $VCOUNT + ($seq[$i+1] - $VBASE)) * $TCOUNT;
2020             $i++;
2021         }
2022         if ($ch >= $SBASE && $ch < $SBASE + $SCOUNT && !(($ch - $SBASE) % $TCOUNT) && $i < @seq - 1 &&
2023             $seq[$i+1] > $TBASE && $seq[$i+1] < $TBASE + $TCOUNT)
2024         {
2025             $ch += $seq[$i+1] - $TBASE;
2026             $i++;
2027         }
2028         push @ret, $ch;
2029     }
2030     return @ret;
2031 }
2032
2033 ################################################################
2034 # remove linguistic-only mappings from the case table
2035 sub remove_linguistic_mappings($$)
2036 {
2037     my ($upper, $lower) = @_;
2038
2039     # remove case mappings that don't round-trip
2040
2041     for (my $i = 0; $i < @{$upper}; $i++)
2042     {
2043         next unless defined ${$upper}[$i];
2044         my $ch = ${$upper}[$i];
2045         ${$upper}[$i] = undef unless defined ${$lower}[$ch] && ${$lower}[$ch] == $i;
2046     }
2047     for (my $i = 0; $i < @{$lower}; $i++)
2048     {
2049         next unless defined ${$lower}[$i];
2050         my $ch = ${$lower}[$i];
2051         ${$lower}[$i] = undef unless defined ${$upper}[$ch] && ${$upper}[$ch] == $i;
2052     }
2053 }
2054
2055 ################################################################
2056 # read in the Unicode database files
2057 sub load_data()
2058 {
2059     my $start;
2060
2061     # now build mappings from the decomposition field of the Unicode database
2062
2063     my $UNICODE_DATA = open_data_file( "ucd", "UnicodeData.txt" );
2064     while (<$UNICODE_DATA>)
2065     {
2066         # Decode the fields ...
2067         my ($code, $name, $cat, $comb, $bidi,
2068             $decomp, $dec, $dig, $num, $mirror,
2069             $oldname, $comment, $upper, $lower, $title) = split /;/;
2070         my $src = hex $code;
2071
2072         die "unknown category $cat" unless defined $categories{$cat};
2073         die "unknown directionality $bidi" unless defined $directions{$bidi};
2074
2075         $category_table[$src] = $categories{$cat};
2076         $direction_table[$src] = $bidi;
2077         if ($cat eq "Mn" || $cat eq "Me" || $cat eq "Cf")
2078         {
2079             $initial_joining_table[$src] = $joining_types{"T"};
2080         }
2081         else
2082         {
2083             $initial_joining_table[$src] = $joining_types{"U"};
2084         }
2085
2086         if ($lower ne "")
2087         {
2088             $tolower_table[$src] = hex $lower;
2089         }
2090         if ($upper ne "")
2091         {
2092             $toupper_table[$src] = hex $upper;
2093         }
2094         if ($dec ne "")
2095         {
2096             $category_table[$src] |= $ctype{"digit"};
2097         }
2098         if ($dig ne "")
2099         {
2100             $digitmap_table[$src] = ord $dig;
2101         }
2102         $combining_class_table[$src] = ($cat ne "Co") ? $comb : 0x100; # Private Use
2103
2104         $category_table[$src] |= $ctype{"nonspacing"}    if $bidi eq "NSM";
2105         $category_table[$src] |= $ctype{"diacritic"}     if $name =~ /^(COMBINING)|(MODIFIER LETTER)\W/;
2106         $category_table[$src] |= $ctype{"vowelmark"}     if $name =~ /\sVOWEL/ || $oldname =~ /\sVOWEL/;
2107         $category_table[$src] |= $ctype{"halfwidth"}     if $name =~ /^HALFWIDTH\s/;
2108         $category_table[$src] |= $ctype{"fullwidth"}     if $name =~ /^FULLWIDTH\s/;
2109         $category_table[$src] |= $ctype{"hiragana"}      if $name =~ /(HIRAGANA)|(\WKANA\W)/;
2110         $category_table[$src] |= $ctype{"katakana"}      if $name =~ /(KATAKANA)|(\WKANA\W)/;
2111         $category_table[$src] |= $ctype{"ideograph"}     if $name =~ /^<CJK Ideograph/;
2112         $category_table[$src] |= $ctype{"ideograph"}     if $name =~ /^CJK COMPATIBILITY IDEOGRAPH/;
2113         $category_table[$src] |= $ctype{"ideograph"}     if $name =~ /^HANGZHOU/;
2114         $category_table[$src] |= $ctype{"highsurrogate"} if $name =~ /High Surrogate/;
2115         $category_table[$src] |= $ctype{"lowsurrogate"}  if $name =~ /Low Surrogate/;
2116
2117         # copy the category and direction for everything between First/Last pairs
2118         if ($name =~ /, First>/) { $start = $src; }
2119         if ($name =~ /, Last>/)
2120         {
2121             while ($start < $src)
2122             {
2123                 $category_table[$start] = $category_table[$src];
2124                 $direction_table[$start] = $direction_table[$src];
2125                 $combining_class_table[$start] = $combining_class_table[$src];
2126                 $start++;
2127             }
2128         }
2129
2130         next if $decomp eq "";  # no decomposition, skip it
2131
2132         if ($decomp =~ /^<([a-zA-Z]+)>\s+([0-9a-fA-F]+)/)
2133         {
2134             my @seq = map { hex $_; } (split /\s+/, (split /\s+/, $decomp, 2)[1]);
2135             $decomp_compat_table[$src] = \@seq;
2136         }
2137
2138         if ($decomp =~ /^<([a-zA-Z]+)>\s+([0-9a-fA-F]+)$/)
2139         {
2140             # decomposition of the form "<foo> 1234" -> use char if type is known
2141             my $dst = hex $2;
2142             if ($1 eq "narrow")
2143             {
2144                 $halfwidth_table[$dst] = $src;
2145                 $fullwidth_table[$src] = $dst;
2146             }
2147             elsif ($1 eq "wide")
2148             {
2149                 next if $dst == 0x5c; # don't remap backslash
2150                 $fullwidth_table[$dst] = $src;
2151                 $halfwidth_table[$src] = $dst;
2152             }
2153             elsif ($1 eq "font" || $1 eq "square" || $1 eq "circle")
2154             {
2155                 $fullwidth_table[$src] = $dst if $src >= 0x10000;
2156             }
2157             elsif ($1 eq "isolated" || $1 eq "final" || $1 eq "initial" || $1 eq "medial")
2158             {
2159                 ${joining_forms{$1}}[$dst] = $src;
2160             }
2161         }
2162         elsif ($decomp =~ /^<compat>\s+0020\s+([0-9a-fA-F]+)/)
2163         {
2164             # decomposition "<compat> 0020 1234" -> combining accent
2165         }
2166         elsif ($decomp =~ /^([0-9a-fA-F]+)/)
2167         {
2168             # store decomposition
2169             if ($decomp =~ /^([0-9a-fA-F]+)\s+([0-9a-fA-F]+)$/)
2170             {
2171                 $decomp_table[$src] = $decomp_compat_table[$src] = [ hex $1, hex $2 ];
2172             }
2173             elsif ($decomp =~ /^([0-9a-fA-F]+)$/)
2174             {
2175                 my $dst = hex $1;
2176                 # Single char decomposition
2177                 $decomp_table[$src] = $decomp_compat_table[$src] = [ $dst ];
2178                 if ($name =~ /^CJK COMPATIBILITY IDEOGRAPH/)
2179                 {
2180                     $cjk_compat_table[$src] = $dst;
2181                     $fullwidth_table[$src] = $dst if $src >= 0x10000;
2182                 }
2183             }
2184         }
2185     }
2186     close $UNICODE_DATA;
2187
2188     # patch the category of some special characters
2189
2190     for (my $i = 0; $i < @decomp_table; $i++)
2191     {
2192         next unless defined $decomp_table[$i];
2193         $category_table[$i] |= $category_table[$decomp_table[$i]->[0]];
2194     }
2195     foreach my $cat (keys %special_categories)
2196     {
2197         my $flag = $ctype{$cat};
2198         foreach my $i (@{$special_categories{$cat}}) { $category_table[$i] |= $flag; }
2199     }
2200     for (my $i = 0; $i < @decomp_compat_table; $i++)
2201     {
2202         next unless defined $decomp_compat_table[$i];
2203         next unless @{$decomp_compat_table[$i]} == 2;
2204         $category_table[$i] |= $category_table[$decomp_compat_table[$i]->[1]] & $ctype{"diacritic"};
2205     }
2206
2207     # load the composition exclusions
2208
2209     my $EXCL = open_data_file( "ucd", "CompositionExclusions.txt" );
2210     while (<$EXCL>)
2211     {
2212         s/\#.*//;  # remove comments
2213         if (/^([0-9a-fA-F]+)\.\.([0-9a-fA-F]+)\s*$/)
2214         {
2215             foreach my $i (hex $1 .. hex $2) { $comp_exclusions[$i] = 1; }
2216         }
2217         elsif (/^([0-9a-fA-F]+)\s*$/)
2218         {
2219             $comp_exclusions[hex $1] = 1;
2220         }
2221     }
2222     close $EXCL;
2223
2224     # load the IDNA mappings
2225
2226     @idna_decomp_table = @decomp_compat_table;
2227     my $IDNA = open_data_file( "idna", "IdnaMappingTable.txt" );
2228     while (<$IDNA>)
2229     {
2230         s/\#.*//;  # remove comments
2231         next if /^\s*$/;
2232         my ($char, $type, $mapping) = split /;/;
2233         my ($ch1, $ch2);
2234         if ($char =~ /([0-9a-fA-F]+)\.\.([0-9a-fA-F]+)/)
2235         {
2236             $ch1 = hex $1;
2237             $ch2 = hex $2;
2238         }
2239         elsif ($char =~ /([0-9a-fA-F]+)/)
2240         {
2241             $ch1 = $ch2 = hex $1;
2242         }
2243
2244         if ($type =~ /mapped/ || $type =~ /deviation/)
2245         {
2246             $mapping =~ s/^\s*(([0-9a-fA-F]+\s+)+)\s*$/$1/;
2247             my @seq = map { hex $_; } split /\s+/, $mapping;
2248             foreach my $i ($ch1 .. $ch2) { $idna_decomp_table[$i] = @seq ? \@seq : [ 0 ]; }
2249         }
2250         elsif ($type =~ /valid/)
2251         {
2252         }
2253         elsif ($type =~ /ignored/)
2254         {
2255             foreach my $i ($ch1 .. $ch2) { $idna_decomp_table[$i] = [ 0 ]; }
2256         }
2257         elsif ($type =~ /disallowed/)
2258         {
2259             foreach my $i ($ch1 .. $ch2)
2260             {
2261                 $idna_decomp_table[$i] = undef;
2262                 $idna_disallowed[$i] = 1;
2263             }
2264         }
2265     }
2266     close $IDNA;
2267
2268     # load the Unihan mappings
2269
2270     my $UNIHAN = open_data_file( "unihan", "Unihan_Variants.txt" );
2271     while (<$UNIHAN>)
2272     {
2273         s/\#.*//;  # remove comments
2274         next if /^\s*$/;
2275         if (/^U\+([0-9a-fA-F]{4})\s+kTraditionalVariant\s+U\+([0-9a-fA-F]{4})$/)
2276         {
2277             next if hex $1 < 0x4dc0;  # skip extension A
2278             $chinese_traditional_table[hex $1] = hex $2;
2279         }
2280         elsif (/^U\+([0-9a-fA-F]{4})\s+kSimplifiedVariant\s+U\+([0-9a-fA-F]{4})$/)
2281         {
2282             next if hex $1 < 0x4dc0;  # skip extension A
2283             $chinese_simplified_table[hex $1] = hex $2;
2284         }
2285     }
2286     close $UNIHAN;
2287     foreach my $i (0xf900..0xfaff)
2288     {
2289         next unless defined $cjk_compat_table[$i];
2290         next if defined $chinese_simplified_table[$cjk_compat_table[$i]];
2291         $chinese_simplified_table[$i] = $cjk_compat_table[$i];
2292     }
2293 }
2294
2295
2296 ################################################################
2297 # add a new registry key
2298 sub add_registry_key($$$)
2299 {
2300     my ($base, $key, $defval) = @_;
2301     $registry_keys{"$base\\$key"} = [ $defval ] unless defined $registry_keys{"$base\\$key"};
2302 }
2303
2304 ################################################################
2305 # add a new registry value with explicit type
2306 sub add_registry_value($$$$)
2307 {
2308     my ($base, $key, $name, $value) = @_;
2309     add_registry_key( $base, $key, undef );
2310     push @{$registry_keys{"$base\\$key"}}, "'$name' = $value";
2311 }
2312
2313 ################################################################
2314 # add a new registry string value
2315 sub add_registry_string_value($$$$)
2316 {
2317     my ($base, $key, $name, $value) = @_;
2318     $value =~ s/\'/\'\'/g;
2319     add_registry_value( $base, $key, $name, "s '$value'" );
2320 }
2321
2322 ################################################################
2323 # add a new registry dword value
2324 sub add_registry_dword_value($$$$)
2325 {
2326     my ($base, $key, $name, $value) = @_;
2327     add_registry_value( $base, $key, $name, "d $value" );
2328 }
2329
2330 ################################################################
2331 # add a new registry binary value
2332 sub add_registry_binary_value($$$$)
2333 {
2334     my ($base, $key, $name, $value) = @_;
2335     add_registry_value( $base, $key, $name, "b " . join "", map { sprintf "%02x", $_; } unpack( "C*", $value ));
2336 }
2337
2338 ################################################################
2339 # define a new lead byte
2340 sub add_lead_byte($)
2341 {
2342     my $ch = shift;
2343     return if defined $cp2uni[$ch];
2344     push @lead_bytes, $ch;
2345     $cp2uni[$ch] = 0;
2346 }
2347
2348 ################################################################
2349 # define a new char mapping
2350 sub add_mapping($$)
2351 {
2352     my ($cp, $uni) = @_;
2353     $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
2354     $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
2355     if ($cp > 0xff) { add_lead_byte( $cp >> 8 ); }
2356 }
2357
2358 ################################################################
2359 # get a mapping including glyph chars for MB_USEGLYPHCHARS
2360 sub get_glyphs_mapping(@)
2361 {
2362     my @table = @_;
2363
2364     for (my $i = 0; $i < @glyph2uni; $i++)
2365     {
2366         $table[$i] = $glyph2uni[$i] if defined $glyph2uni[$i];
2367     }
2368     return @table;
2369 }
2370
2371 ################################################################
2372 # build EUC-JP table from the JIS 0208/0212 files
2373 sub dump_eucjp_codepage()
2374 {
2375     @cp2uni = ();
2376     @glyph2uni = ();
2377     @lead_bytes = ();
2378     @uni2cp = ();
2379     $default_char = $DEF_CHAR;
2380     $default_wchar = 0x30fb;
2381
2382     # ASCII chars
2383     foreach my $i (0x00 .. 0x7f) { add_mapping( $i, $i ); }
2384
2385     # lead bytes
2386     foreach my $i (0x8e, 0xa1 .. 0xfe) { add_lead_byte($i); }
2387
2388     # JIS X 0201 right plane
2389     foreach my $i (0xa1 .. 0xdf) { add_mapping( 0x8e00 + $i, 0xfec0 + $i ); }
2390
2391     # undefined chars
2392     foreach my $i (0x80 .. 0x8d, 0x8f .. 0x9f) { $cp2uni[$i] = $i; }
2393     $cp2uni[0xa0] = 0xf8f0;
2394     $cp2uni[0xff] = 0xf8f3;
2395
2396     # Fix backslash conversion
2397     add_mapping( 0xa1c0, 0xff3c );
2398
2399     # Add private mappings for rows undefined in JIS 0208/0212
2400     my $private = 0xe000;
2401     foreach my $hi (0xf5 .. 0xfe)
2402     {
2403         foreach my $lo (0xa1 .. 0xfe)
2404         {
2405             add_mapping( ($hi << 8) + $lo, $private++ );
2406         }
2407     }
2408     foreach my $hi (0xf5 .. 0xfe)
2409     {
2410         foreach my $lo (0x21 .. 0x7e)
2411         {
2412             add_mapping( ($hi << 8) + $lo, $private++ );
2413         }
2414     }
2415
2416     my $INPUT = open_data_file( "jis0208" );
2417     while (<$INPUT>)
2418     {
2419         next if /^\#/;  # skip comments
2420         next if /^$/;  # skip empty lines
2421         next if /\x1a/;  # skip ^Z
2422         if (/^0x[0-9a-fA-F]+\s+0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\s+(\#.*)?/)
2423         {
2424             add_mapping( 0x8080 + hex $1, hex $2 );
2425             next;
2426         }
2427         die "Unrecognized line $_\n";
2428     }
2429     close $INPUT;
2430
2431     $INPUT = open_data_file( "jis0212" );
2432     while (<$INPUT>)
2433     {
2434         next if /^\#/;  # skip comments
2435         next if /^$/;  # skip empty lines
2436         next if /\x1a/;  # skip ^Z
2437         if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\s+(\#.*)?/)
2438         {
2439             add_mapping( 0x8000 + hex $1, hex $2 );
2440             next;
2441         }
2442         die "Unrecognized line $_\n";
2443     }
2444     close $INPUT;
2445
2446     output_codepage_file( 20932 );
2447 }
2448
2449 ################################################################
2450 # build Korean Wansung table from the KSX1001 file
2451 sub dump_krwansung_codepage(@)
2452 {
2453     my @cp949 = @_;
2454     @cp2uni = ();
2455     @glyph2uni = ();
2456     @lead_bytes = ();
2457     @uni2cp = ();
2458     $default_char = 0x3f;
2459     $default_wchar = 0x003f;
2460
2461     # ASCII and undefined chars
2462     foreach my $i (0x00 .. 0x9f) { add_mapping( $i, $i ); }
2463     add_mapping( 0xa0, 0xf8e6 );
2464     add_mapping( 0xad, 0xf8e7 );
2465     add_mapping( 0xae, 0xf8e8 );
2466     add_mapping( 0xaf, 0xf8e9 );
2467     add_mapping( 0xfe, 0xf8ea );
2468     add_mapping( 0xff, 0xf8eb );
2469
2470     my $INPUT = open_data_file( "ksx1001" );
2471     while (<$INPUT>)
2472     {
2473         next if /^\#/;  # skip comments
2474         next if /^$/;  # skip empty lines
2475         next if /\x1a/;  # skip ^Z
2476         if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\s+(\#.*)?/)
2477         {
2478             add_mapping( 0x8080 + hex $1, hex $2 );
2479             next;
2480         }
2481         die "Unrecognized line $_\n";
2482     }
2483     close $INPUT;
2484
2485     # get some extra mappings from cp 949
2486     my @defined_lb;
2487     map { $defined_lb[$_] = 1; } @lead_bytes;
2488     foreach my $i (0x0000 .. 0xffff)
2489     {
2490         next if ($i >= 0x1100 && $i <= 0x11ff);  # range not used in 20949
2491         next unless defined $cp949[$i];
2492         if ($cp949[$i] >= 0xff)
2493         {
2494             # only add chars for lead bytes that exist in 20949
2495             my $hi = $cp949[$i] >> 8;
2496             my $lo = $cp949[$i] & 0xff;
2497             next unless $defined_lb[$hi];
2498             next unless $lo >= 0xa1 && $lo <= 0xfe;
2499         }
2500         add_mapping( $cp949[$i], $i );
2501     }
2502
2503     output_codepage_file( 20949 );
2504 }
2505
2506
2507 ################################################################
2508 # dump an array of integers
2509 sub dump_array($$@)
2510 {
2511     my ($bit_width, $default, @array) = @_;
2512     my $format = sprintf "0x%%0%ux", $bit_width / 4;
2513     my $i;
2514     my $ret = "    ";
2515     for ($i = 0; $i < $#array; $i++)
2516     {
2517         $ret .= sprintf($format, defined $array[$i] ? $array[$i] : $default);
2518         $ret .= (($i % 8) != 7) ? ", " : ",\n    ";
2519     }
2520     $ret .= sprintf($format, defined $array[$i] ? $array[$i] : $default);
2521     return $ret;
2522 }
2523
2524
2525 ################################################################
2526 # dump an SBCS mapping table in binary format
2527 sub dump_binary_sbcs_table($)
2528 {
2529     my $codepage = shift;
2530
2531     my @header = ( 13, $codepage, 1, $default_char, $default_wchar, $cp2uni[$default_char], $uni2cp[$default_wchar] );
2532     my $wc_offset = 256 + 3 + (@glyph2uni ? 256 : 0);
2533
2534     print OUTPUT pack "S<*", @header;
2535     print OUTPUT pack "C12", (0) x 12;
2536     print OUTPUT pack "S<*", $wc_offset, map { $_ || 0; } @cp2uni[0 .. 255];
2537
2538     if (@glyph2uni)
2539     {
2540         print OUTPUT pack "S<*", 256, get_glyphs_mapping(@cp2uni[0 .. 255]);
2541     }
2542     else
2543     {
2544         print OUTPUT pack "S<*", 0;
2545     }
2546
2547     print OUTPUT pack "S<*", 0, 0;
2548
2549     print OUTPUT pack "C*", map { defined $_ ? $_ : $default_char; } @uni2cp[0 .. 65535];
2550 }
2551
2552
2553 ################################################################
2554 # dump a DBCS mapping table in binary format
2555 sub dump_binary_dbcs_table($)
2556 {
2557     my $codepage = shift;
2558     my @lb_ranges = get_lb_ranges();
2559     my @header = ( 13, $codepage, 2, $default_char, $default_wchar, $cp2uni[$default_char], $uni2cp[$default_wchar] );
2560
2561     my @offsets = (0) x 256;
2562     my $pos = 0;
2563     foreach my $i (@lead_bytes)
2564     {
2565         $offsets[$i] = ($pos += 256);
2566         $cp2uni[$i] = 0;
2567     }
2568
2569     my $wc_offset = 256 + 3 + 256 * (1 + scalar @lead_bytes);
2570
2571     print OUTPUT pack "S<*", @header;
2572     print OUTPUT pack "C12", @lb_ranges, 0 x 12;
2573     print OUTPUT pack "S<*", $wc_offset, map { $_ || 0; } @cp2uni[0 .. 255];
2574     print OUTPUT pack "S<*", 0, scalar @lb_ranges / 2, @offsets;
2575
2576     foreach my $i (@lead_bytes)
2577     {
2578         my $base = $i << 8;
2579         print OUTPUT pack "S<*", map { defined $_ ? $_ : $default_wchar; } @cp2uni[$base .. $base + 255];
2580     }
2581
2582     print OUTPUT pack "S<", 4;
2583     print OUTPUT pack "S<*", map { defined $_ ? $_ : $default_char; } @uni2cp[0 .. 65535];
2584 }
2585
2586
2587 ################################################################
2588 # get the list of defined lead byte ranges
2589 sub get_lb_ranges()
2590 {
2591     my @list = ();
2592     my @ranges = ();
2593
2594     foreach my $i (@lead_bytes) { $list[$i] = 1; }
2595     my $on = 0;
2596     for (my $i = 0; $i < 256; $i++)
2597     {
2598         if ($on)
2599         {
2600             if (!defined $list[$i]) { push @ranges, $i-1; $on = 0; }
2601         }
2602         else
2603         {
2604             if ($list[$i]) { push @ranges, $i; $on = 1; }
2605         }
2606     }
2607     if ($on) { push @ranges, 0xff; }
2608     return @ranges;
2609 }
2610
2611 ################################################################
2612 # dump the Indic Syllabic Category table
2613 sub dump_indic($)
2614 {
2615     my $filename = shift;
2616     my @indic_table;
2617
2618     my $INPUT = open_data_file( "ucd", "IndicSyllabicCategory.txt" );
2619     while (<$INPUT>)
2620     {
2621         next if /^\#/;  # skip comments
2622         next if /^\s*$/;  # skip empty lines
2623         next if /\x1a/;  # skip ^Z
2624         if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*#/)
2625         {
2626             my $type = $2;
2627             die "unknown indic $type" unless defined $indic_types{$type};
2628             if (hex $1 < 65536)
2629             {
2630                 $indic_table[hex $1] = $indic_types{$type};
2631             }
2632             next;
2633         }
2634         elsif (/^\s*([0-9a-fA-F]+)\.\.\s*([0-9a-fA-F]+)\s*;\s*([A-Za-z_]+)\s*#/)
2635         {
2636             my $type = $3;
2637             die "unknown indic $type" unless defined $indic_types{$type};
2638             if (hex $1 < 65536 and hex $2 < 65536)
2639             {
2640                 foreach my $i (hex $1 .. hex $2)
2641                 {
2642                     $indic_table[$i] = $indic_types{$type};
2643                 }
2644             }
2645             next;
2646         }
2647         die "malformed line $_";
2648     }
2649     close $INPUT;
2650
2651     my $prev_data_file = $current_data_file;
2652     $INPUT = open_data_file( "ucd", "IndicPositionalCategory.txt" );
2653     while (<$INPUT>)
2654     {
2655         next if /^\#/;  # skip comments
2656         next if /^\s*$/;  # skip empty lines
2657         next if /\x1a/;  # skip ^Z
2658         if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*#/)
2659         {
2660             my $type = $2;
2661             die "unknown matra $type" unless defined $matra_types{$type};
2662             $indic_table[hex $1] |= $matra_types{$type} << 8;
2663             next;
2664         }
2665         elsif (/^\s*([0-9a-fA-F]+)\.\.\s*([0-9a-fA-F]+)\s*;\s*([A-Za-z_]+)\s*#/)
2666         {
2667             my $type = $3;
2668             die "unknown matra $type" unless defined $matra_types{$type};
2669             foreach my $i (hex $1 .. hex $2)
2670             {
2671                 $indic_table[$i] |= $matra_types{$type} << 8;
2672             }
2673             next;
2674         }
2675         die "malformed line $_";
2676     }
2677     close $INPUT;
2678
2679     open OUTPUT,">$filename.new" or die "Cannot create $filename";
2680     print "Building $filename\n";
2681     print OUTPUT "/* Unicode Indic Syllabic Category */\n";
2682     print OUTPUT "/* generated from $prev_data_file */\n";
2683     print OUTPUT "/*       and from $current_data_file */\n";
2684     print OUTPUT "/* DO NOT EDIT!! */\n\n";
2685     print OUTPUT "#include \"windef.h\"\n\n";
2686
2687     dump_two_level_mapping( "indic_syllabic_table", $indic_types{'Other'}, 16, @indic_table );
2688
2689     close OUTPUT;
2690     save_file($filename);
2691 }
2692
2693 ################################################################
2694 # dump the Line Break Properties table
2695 sub dump_linebreak($)
2696 {
2697     my $filename = shift;
2698     my @break_table;
2699
2700     my $INPUT = open_data_file( "ucd", "LineBreak.txt" );
2701     while (<$INPUT>)
2702     {
2703         next if /^\#/;  # skip comments
2704         next if /^\s*$/;  # skip empty lines
2705         next if /\x1a/;  # skip ^Z
2706         if (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z][0-9A-Z])+\s*/)
2707         {
2708             my $type = $2;
2709             die "unknown breaktype $type" unless defined $break_types{$type};
2710             $break_table[hex $1] = $break_types{$type};
2711             next;
2712         }
2713         elsif (/^\s*([0-9a-fA-F]+)\.\.\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z][0-9A-Z])+\s*/)
2714         {
2715             my $type = $3;
2716             die "unknown breaktype $type" unless defined $break_types{$type};
2717             foreach my $i (hex $1 .. hex $2)
2718             {
2719                 $break_table[$i] = $break_types{$type};
2720             }
2721             next;
2722         }
2723         elsif (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z])+\s*/)
2724         {
2725             my $type = $2;
2726             die "unknown breaktype $type" unless defined $break_types{$type};
2727             $break_table[hex $1] = $break_types{$type};
2728             next;
2729         }
2730         elsif (/^\s*([0-9a-fA-F]+)\.\.\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z])+\s*/)
2731         {
2732             my $type = $3;
2733             die "unknown breaktype $type" unless defined $break_types{$type};
2734             foreach my $i (hex $1 .. hex $2)
2735             {
2736                 $break_table[$i] = $break_types{$type};
2737             }
2738             next;
2739         }
2740         die "malformed line $_";
2741     }
2742     close $INPUT;
2743
2744     open OUTPUT,">$filename.new" or die "Cannot create $filename";
2745     print "Building $filename\n";
2746     print OUTPUT "/* Unicode Line Break Properties */\n";
2747     print OUTPUT "/* generated from $current_data_file */\n";
2748     print OUTPUT "/* DO NOT EDIT!! */\n\n";
2749     print OUTPUT "#include \"windef.h\"\n\n";
2750
2751     dump_two_level_mapping( "wine_linebreak_table", $break_types{'XX'}, 16, @break_table );
2752
2753     close OUTPUT;
2754     save_file($filename);
2755 }
2756
2757 my %scripts =
2758 (
2759     "Unknown"                => 0,
2760     "Common"                 => 1,
2761     "Inherited"              => 2,
2762     "Arabic"                 => 3,
2763     "Armenian"               => 4,
2764     "Avestan"                => 5,
2765     "Balinese"               => 6,
2766     "Bamum"                  => 7,
2767     "Batak"                  => 8,
2768     "Bengali"                => 9,
2769     "Bopomofo"               => 10,
2770     "Brahmi"                 => 11,
2771     "Braille"                => 12,
2772     "Buginese"               => 13,
2773     "Buhid"                  => 14,
2774     "Canadian_Aboriginal"    => 15,
2775     "Carian"                 => 16,
2776     "Cham"                   => 17,
2777     "Cherokee"               => 18,
2778     "Coptic"                 => 19,
2779     "Cuneiform"              => 20,
2780     "Cypriot"                => 21,
2781     "Cyrillic"               => 22,
2782     "Deseret"                => 23,
2783     "Devanagari"             => 24,
2784     "Egyptian_Hieroglyphs"   => 25,
2785     "Ethiopic"               => 26,
2786     "Georgian"               => 27,
2787     "Glagolitic"             => 28,
2788     "Gothic"                 => 29,
2789     "Greek"                  => 30,
2790     "Gujarati"               => 31,
2791     "Gurmukhi"               => 32,
2792     "Han"                    => 33,
2793     "Hangul"                 => 34,
2794     "Hanunoo"                => 35,
2795     "Hebrew"                 => 36,
2796     "Hiragana"               => 37,
2797     "Imperial_Aramaic"       => 38,
2798     "Inscriptional_Pahlavi"  => 39,
2799     "Inscriptional_Parthian" => 40,
2800     "Javanese"               => 41,
2801     "Kaithi"                 => 42,
2802     "Kannada"                => 43,
2803     "Katakana"               => 44,
2804     "Kayah_Li"               => 45,
2805     "Kharoshthi"             => 46,
2806     "Khmer"                  => 47,
2807     "Lao"                    => 48,
2808     "Latin"                  => 49,
2809     "Lepcha"                 => 50,
2810     "Limbu"                  => 51,
2811     "Linear_B"               => 52,
2812     "Lisu"                   => 53,
2813     "Lycian"                 => 54,
2814     "Lydian"                 => 55,
2815     "Malayalam"              => 56,
2816     "Mandaic"                => 57,
2817     "Meetei_Mayek"           => 58,
2818     "Mongolian"              => 59,
2819     "Myanmar"                => 60,
2820     "New_Tai_Lue"            => 61,
2821     "Nko"                    => 62,
2822     "Ogham"                  => 63,
2823     "Ol_Chiki"               => 64,
2824     "Old_Italic"             => 65,
2825     "Old_Persian"            => 66,
2826     "Old_South_Arabian"      => 67,
2827     "Old_Turkic"             => 68,
2828     "Oriya"                  => 69,
2829     "Osmanya"                => 70,
2830     "Phags_Pa"               => 71,
2831     "Phoenician"             => 72,
2832     "Rejang"                 => 73,
2833     "Runic"                  => 74,
2834     "Samaritan"              => 75,
2835     "Saurashtra"             => 76,
2836     "Shavian"                => 77,
2837     "Sinhala"                => 78,
2838     "Sundanese"              => 79,
2839     "Syloti_Nagri"           => 80,
2840     "Syriac"                 => 81,
2841     "Tagalog"                => 82,
2842     "Tagbanwa"               => 83,
2843     "Tai_Le"                 => 84,
2844     "Tai_Tham"               => 85,
2845     "Tai_Viet"               => 86,
2846     "Tamil"                  => 87,
2847     "Telugu"                 => 88,
2848     "Thaana"                 => 89,
2849     "Thai"                   => 90,
2850     "Tibetan"                => 91,
2851     "Tifinagh"               => 92,
2852     "Ugaritic"               => 93,
2853     "Vai"                    => 94,
2854     "Yi"                     => 95,
2855     # Win8/Win8.1
2856     "Chakma"                 => 96,
2857     "Meroitic_Cursive"       => 97,
2858     "Meroitic_Hieroglyphs"   => 98,
2859     "Miao"                   => 99,
2860     "Sharada"                => 100,
2861     "Sora_Sompeng"           => 101,
2862     "Takri"                  => 102,
2863     # Win10
2864     "Bassa_Vah"              => 103,
2865     "Caucasian_Albanian"     => 104,
2866     "Duployan"               => 105,
2867     "Elbasan"                => 106,
2868     "Grantha"                => 107,
2869     "Khojki"                 => 108,
2870     "Khudawadi"              => 109,
2871     "Linear_A"               => 110,
2872     "Mahajani"               => 111,
2873     "Manichaean"             => 112,
2874     "Mende_Kikakui"          => 113,
2875     "Modi"                   => 114,
2876     "Mro"                    => 115,
2877     "Nabataean"              => 116,
2878     "Old_North_Arabian"      => 117,
2879     "Old_Permic"             => 118,
2880     "Pahawh_Hmong"           => 119,
2881     "Palmyrene"              => 120,
2882     "Pau_Cin_Hau"            => 121,
2883     "Psalter_Pahlavi"        => 122,
2884     "Siddham"                => 123,
2885     "Tirhuta"                => 124,
2886     "Warang_Citi"            => 125,
2887     # Win10 RS1
2888     "Adlam"                  => 126,
2889     "Ahom"                   => 127,
2890     "Anatolian_Hieroglyphs"  => 128,
2891     "Bhaiksuki"              => 129,
2892     "Hatran"                 => 130,
2893     "Marchen"                => 131,
2894     "Multani"                => 132,
2895     "Newa"                   => 133,
2896     "Old_Hungarian"          => 134,
2897     "Osage"                  => 135,
2898     "SignWriting"            => 136,
2899     "Tangut"                 => 137,
2900     # Win10 RS4
2901     "Masaram_Gondi"          => 138,
2902     "Nushu"                  => 139,
2903     "Soyombo"                => 140,
2904     "Zanabazar_Square"       => 141,
2905     # Win10 1903
2906     "Dogra"                  => 142,
2907     "Gunjala_Gondi"          => 143,
2908     "Hanifi_Rohingya"        => 144,
2909     "Makasar"                => 145,
2910     "Medefaidrin"            => 146,
2911     "Old_Sogdian"            => 147,
2912     "Sogdian"                => 148,
2913     # Win10 2004
2914     "Elymaic"                => 149,
2915     "Nyiakeng_Puachue_Hmong" => 150,
2916     "Nandinagari"            => 151,
2917     "Wancho"                 => 152,
2918     # Win11
2919     "Chorasmian"             => 153,
2920     "Dives_Akuru"            => 154,
2921     "Khitan_Small_Script"    => 155,
2922     "Yezidi"                 => 156,
2923 );
2924
2925 ################################################################
2926 # dump Script IDs table
2927 sub dump_scripts($)
2928 {
2929     my $filename = shift;
2930     my $header = $filename;
2931     my @scripts_table;
2932     my $script_index;
2933     my $i;
2934
2935     my $INPUT = open_data_file( "ucd", "Scripts.txt" );
2936     # Fill the table
2937     # Unknown script id is always 0, so undefined scripts are automatically treated as such
2938     while (<$INPUT>)
2939     {
2940         my $type = "";
2941
2942         next if /^\#/;  # skip comments
2943         next if /^\s*$/;  # skip empty lines
2944         next if /\x1a/;  # skip ^Z
2945         if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
2946         {
2947             $type = $2;
2948             if (defined $scripts{$type})
2949             {
2950                 $scripts_table[hex $1] = $scripts{$type};
2951             }
2952             next;
2953         }
2954         elsif (/^\s*([0-9a-fA-F]+)\.\.\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
2955         {
2956             $type = $3;
2957             if (defined $scripts{$type})
2958             {
2959                 foreach my $i (hex $1 .. hex $2)
2960                 {
2961                     $scripts_table[$i] = $scripts{$type};
2962                 }
2963             }
2964             next;
2965         }
2966     }
2967
2968     close $INPUT;
2969
2970     $header = "$filename.h";
2971     open OUTPUT,">$header.new" or die "Cannot create $header";
2972     print "Building $header\n";
2973     print OUTPUT "/* Unicode Script IDs */\n";
2974     print OUTPUT "/* generated from $current_data_file */\n";
2975     print OUTPUT "/* DO NOT EDIT!! */\n\n";
2976
2977     print OUTPUT "enum unicode_script_id {\n";
2978     foreach my $script (sort { $scripts{$a} <=> $scripts{$b} } keys %scripts)
2979     {
2980         print OUTPUT "    Script_$script = $scripts{$script},\n";
2981     }
2982     print OUTPUT "    Script_LastId = ", (scalar keys %scripts) - 1, "\n";
2983     print OUTPUT "};\n";
2984
2985     close OUTPUT;
2986     save_file($header);
2987
2988     $filename = "$filename.c";
2989     open OUTPUT,">$filename.new" or die "Cannot create $header";
2990     print "Building $filename\n";
2991     print OUTPUT "/* Unicode Script IDs */\n";
2992     print OUTPUT "/* generated from $current_data_file */\n";
2993     print OUTPUT "/* DO NOT EDIT!! */\n\n";
2994     print OUTPUT "#include \"windef.h\"\n\n";
2995
2996     dump_two_level_mapping( "wine_scripts_table", 0, 16, @scripts_table );
2997     close OUTPUT;
2998     save_file($filename);
2999 }
3000
3001 ################################################################
3002 # dump the BiDi mirroring table
3003 sub dump_mirroring($)
3004 {
3005     my $filename = shift;
3006     my @mirror_table = ();
3007
3008     my $INPUT = open_data_file( "ucd", "BidiMirroring.txt" );
3009     while (<$INPUT>)
3010     {
3011         next if /^\#/;  # skip comments
3012         next if /^$/;  # skip empty lines
3013         next if /\x1a/;  # skip ^Z
3014         if (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9a-fA-F]+)/)
3015         {
3016             $mirror_table[hex $1] = hex $2;
3017             next;
3018         }
3019         die "malformed line $_";
3020     }
3021     close $INPUT;
3022
3023     open OUTPUT,">$filename.new" or die "Cannot create $filename";
3024     print "Building $filename\n";
3025     print OUTPUT "/* Unicode BiDi mirroring */\n";
3026     print OUTPUT "/* generated from $current_data_file */\n";
3027     print OUTPUT "/* DO NOT EDIT!! */\n\n";
3028     print OUTPUT "#include \"windef.h\"\n\n";
3029     dump_two_level_mapping( "wine_mirror_map", 0, 16, @mirror_table );
3030     close OUTPUT;
3031     save_file($filename);
3032 }
3033
3034 ################################################################
3035 # dump the Bidi Brackets
3036 sub dump_bracket($)
3037 {
3038     my $filename = shift;
3039     my @bracket_table;
3040
3041     my $INPUT = open_data_file( "ucd", "BidiBrackets.txt" );
3042     while (<$INPUT>)
3043     {
3044         next if /^\#/;  # skip comments
3045         next if /^\s*$/;  # skip empty lines
3046         next if /\x1a/;  # skip ^Z
3047         if (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9a-fA-F]+);\s*([con])/)
3048         {
3049             my $type = $3;
3050             die "unknown bracket $type" unless defined $bracket_types{$type};
3051             die "characters too distant $1 and $2" if abs(hex($2) - hex($1)) >= 128;
3052             $bracket_table[hex $1] = (hex($2) - hex($1)) % 255;
3053             $bracket_table[hex $1] += $bracket_types{$type} << 8;
3054             next;
3055         }
3056         die "malformed line $_";
3057     }
3058     close $INPUT;
3059
3060     open OUTPUT,">$filename.new" or die "Cannot create $filename";
3061     print "Building $filename\n";
3062     print OUTPUT "/* Unicode Bidirectional Bracket table */\n";
3063     print OUTPUT "/* generated from $current_data_file */\n";
3064     print OUTPUT "/* DO NOT EDIT!! */\n\n";
3065     print OUTPUT "#include \"windef.h\"\n\n";
3066
3067     dump_two_level_mapping( "bidi_bracket_table", 0, 16, @bracket_table );
3068
3069     close OUTPUT;
3070     save_file($filename);
3071 }
3072
3073 ################################################################
3074 # dump the Arabic shaping table
3075 sub dump_shaping($)
3076 {
3077     my $filename = shift;
3078     my @joining_table = @initial_joining_table;
3079
3080     my $INPUT = open_data_file( "ucd", "ArabicShaping.txt" );
3081     while (<$INPUT>)
3082     {
3083         next if /^\#/;  # skip comments
3084         next if /^\s*$/;  # skip empty lines
3085         next if /\x1a/;  # skip ^Z
3086         if (/^\s*([0-9a-fA-F]+)\s*;.*;\s*([RLDCUT])\s*;\s*(\w+)/)
3087         {
3088             my $type = $2;
3089             $joining_table[hex $1] = $joining_types{$type};
3090             next;
3091         }
3092         die "malformed line $_";
3093     }
3094     close $INPUT;
3095
3096     open OUTPUT,">$filename.new" or die "Cannot create $filename";
3097     print "Building $filename\n";
3098     print OUTPUT "/* Unicode Arabic shaping */\n";
3099     print OUTPUT "/* generated from $current_data_file */\n";
3100     print OUTPUT "/* DO NOT EDIT!! */\n\n";
3101     print OUTPUT "#include \"windef.h\"\n\n";
3102
3103     dump_two_level_mapping( "wine_shaping_table", 0, 16, @joining_table );
3104
3105     print OUTPUT "\nconst unsigned short DECLSPEC_HIDDEN wine_shaping_forms[256][4] =\n{\n";
3106     for (my $i = 0x600; $i <= 0x6ff; $i++)
3107     {
3108         printf OUTPUT "    { 0x%04x, 0x%04x, 0x%04x, 0x%04x },\n",
3109             ${joining_forms{"isolated"}}[$i] || $i,
3110             ${joining_forms{"final"}}[$i] || $i,
3111             ${joining_forms{"initial"}}[$i] || $i,
3112             ${joining_forms{"medial"}}[$i] || $i;
3113     }
3114     print OUTPUT "};\n";
3115
3116     close OUTPUT;
3117     save_file($filename);
3118 }
3119
3120 ################################################################
3121 # dump the Arabic shaping table
3122 sub dump_arabic_shaping($)
3123 {
3124     my $filename = shift;
3125     my @joining_table = @initial_joining_table;
3126
3127     my $INPUT = open_data_file( "ucd", "ArabicShaping.txt" );
3128     while (<$INPUT>)
3129     {
3130         next if /^\#/;  # skip comments
3131         next if /^\s*$/;  # skip empty lines
3132         next if /\x1a/;  # skip ^Z
3133         if (/^\s*([0-9a-fA-F]+)\s*;.*;\s*([RLDCUT])\s*;\s*(\w+)/)
3134         {
3135             my $type = $2;
3136             my $group = $3;
3137
3138             if ($group eq "ALAPH" || $group eq "DALATH RISH")
3139             {
3140                 $joining_table[hex $1] = $joining_types{$group};
3141             }
3142             else
3143             {
3144                 $joining_table[hex $1] = $joining_types{$type};
3145             }
3146
3147             next;
3148         }
3149         die "malformed line $_";
3150     }
3151     close $INPUT;
3152
3153     open OUTPUT,">$filename.new" or die "Cannot create $filename";
3154     print "Building $filename\n";
3155     print OUTPUT "/* Unicode Arabic shaping */\n";
3156     print OUTPUT "/* generated from $current_data_file */\n";
3157     print OUTPUT "/* DO NOT EDIT!! */\n\n";
3158     print OUTPUT "#include \"windef.h\"\n\n";
3159
3160     dump_two_level_mapping( "arabic_shaping_table", 0, 16, @joining_table );
3161
3162     close OUTPUT;
3163     save_file($filename);
3164 }
3165
3166 ################################################################
3167 # dump the Vertical Orientation table
3168 sub dump_vertical($$)
3169 {
3170     my ($filename, $unix) = @_;
3171     my @vertical_table;
3172
3173     my $INPUT = open_data_file( "ucd", "VerticalOrientation.txt" );
3174     while (<$INPUT>)
3175     {
3176         next if /^\#/;  # skip comments
3177         next if /^\s*$/;  # skip empty lines
3178         next if /\x1a/;  # skip ^Z
3179         if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
3180         {
3181             my $type = $2;
3182             die "unknown vertical $type" unless defined $vertical_types{$type};
3183             if (hex $1 < 65536)
3184             {
3185                 $vertical_table[hex $1] = $vertical_types{$type};
3186             }
3187             next;
3188         }
3189         elsif (/^\s*([0-9a-fA-F]+)\.\.\s*([0-9a-fA-F]+)\s*;\s*([A-Za-z_]+)\s*/)
3190         {
3191             my $type = $3;
3192             die "unknown vertical $type" unless defined $vertical_types{$type};
3193             foreach my $i (hex $1 .. hex $2)
3194             {
3195                 $vertical_table[$i] = $vertical_types{$type};
3196             }
3197             next;
3198         }
3199         die "malformed line $_";
3200     }
3201     close $INPUT;
3202
3203     open OUTPUT,">$filename.new" or die "Cannot create $filename";
3204     print "Building $filename\n";
3205     print OUTPUT "/* Unicode Vertical Orientation */\n";
3206     print OUTPUT "/* generated from $current_data_file */\n";
3207     print OUTPUT "/* DO NOT EDIT!! */\n\n";
3208     if ($unix)
3209     {
3210         print OUTPUT "#if 0\n";
3211         print OUTPUT "#pragma makedep unix\n";
3212         print OUTPUT "#endif\n\n";
3213     }
3214     print OUTPUT "#include \"windef.h\"\n\n";
3215
3216     dump_two_level_mapping( "vertical_orientation_table", $vertical_types{'R'}, 16, @vertical_table );
3217
3218     close OUTPUT;
3219     save_file($filename);
3220 }
3221
3222 ################################################################
3223 # compress a mapping table by removing identical rows
3224 sub compress_array($$@)
3225 {
3226     my $rows = shift;
3227     my $def = shift;
3228     my @table = @_;
3229     my $len = @table / $rows;
3230     my @array;
3231     my $data = "";
3232
3233     # try to merge table rows
3234     for (my $row = 0; $row < $rows; $row++)
3235     {
3236         my $rowtxt = pack "U*", map { defined($_) ? $_ : $def; } @table[($row * $len)..(($row + 1) * $len - 1)];
3237         my $pos = index $data, $rowtxt;
3238         if ($pos == -1)
3239         {
3240             # check if the tail of the data can match the start of the new row
3241             my $first = substr( $rowtxt, 0, 1 );
3242             for (my $i = length($data) - 1; $i > 0; $i--)
3243             {
3244                 $pos = index( substr( $data, -$i ), $first );
3245                 last if $pos == -1;
3246                 $i -= $pos;
3247                 next unless substr( $data, -$i ) eq substr( $rowtxt, 0, $i );
3248                 substr( $data, -$i ) = "";
3249                 last;
3250             }
3251             $pos = length $data;
3252             $data .= $rowtxt;
3253         }
3254         $array[$row] = $rows + $pos;
3255     }
3256     return @array, unpack "U*", $data;
3257 }
3258
3259 ################################################################
3260 # dump a char -> 16-bit value mapping table using two-level tables
3261 sub dump_two_level_mapping($$$@)
3262 {
3263     my $name = shift;
3264     my $def = shift;
3265     my $size = shift;
3266     my $type = $size == 16 ? "unsigned short" : "unsigned int";
3267     my (@array, @row_array, @data, @row_data);
3268     (@row_array[0..4095], @data) = compress_array( 4096, $def, @_[0..65535] );
3269     (@array[0..255], @row_data) = compress_array( 256, 0, @row_array );
3270
3271     for (my $i = 0; $i < @row_data; $i++) { $row_data[$i] += @row_data + 256 - 4096; }
3272
3273     printf OUTPUT "const %s DECLSPEC_HIDDEN %s[%d] =\n{\n", $type, $name, @array + @row_data + @data;
3274     printf OUTPUT "    /* level 1 offsets */\n%s,\n", dump_array( $size, 0, @array );
3275     printf OUTPUT "    /* level 2 offsets */\n%s,\n", dump_array( $size, 0, @row_data );
3276     printf OUTPUT "    /* values */\n%s\n};\n", dump_array( $size, 0, @data );
3277 }
3278
3279 ################################################################
3280 # dump a char -> value mapping table using three-level tables
3281 sub dump_three_level_mapping($$@)
3282 {
3283     my $name = shift;
3284     my $def = shift;
3285     my $size = shift;
3286     my $type = $size == 16 ? "unsigned short" : "unsigned int";
3287     my $level3 = ($MAX_CHAR + 1) / 16;
3288     my $level2 = $level3 / 16;
3289     my $level1 = $level2 / 16;
3290     my @array3 = compress_array( $level3, $def, @_[0..$MAX_CHAR] );
3291     my @array2 = compress_array( $level2, 0, @array3[0..$level3-1] );
3292     my @array1 = compress_array( $level1, 0, @array2[0..$level2-1] );
3293
3294     for (my $i = $level2; $i < @array2; $i++) { $array2[$i] += @array1 + @array2 - $level2 - $level3; }
3295     for (my $i = $level1; $i < @array1; $i++) { $array1[$i] += @array1 - $level2; }
3296
3297     printf OUTPUT "const %s DECLSPEC_HIDDEN %s[%u] =\n{\n", $type, $name, @array1 + (@array2 - $level2) + (@array3 - $level3);
3298     printf OUTPUT "    /* level 1 offsets */\n%s,\n", dump_array( $size, 0, @array1[0..$level1-1] );
3299     printf OUTPUT "    /* level 2 offsets */\n%s,\n", dump_array( $size, 0, @array1[$level1..$#array1] );
3300     printf OUTPUT "    /* level 3 offsets */\n%s,\n", dump_array( $size, 0, @array2[$level2..$#array2] );
3301     printf OUTPUT "    /* values */\n%s\n};\n", dump_array( $size, 0, @array3[$level3..$#array3] );
3302 }
3303
3304 ################################################################
3305 # dump a binary case mapping table in l_intl.nls format
3306 sub dump_binary_case_table(@)
3307 {
3308     my (@table) = @_;
3309     my @difftable;
3310     my @res;
3311
3312     for (my $i = 0; $i < @table; $i++)
3313     {
3314         next unless defined $table[$i];
3315         $difftable[$i] = ($table[$i] - $i) & 0xffffffff;
3316     }
3317
3318     my (@low_array1, @low_array2, @low_data, @low_row_data);
3319     (@low_array2[0..4095], @low_data) = compress_array( 4096, 0, @difftable[0..65535] );
3320     (@low_array1[0..255], @low_row_data) = compress_array( 256, 0, @low_array2 );
3321
3322     if (scalar @table > 0x10000)
3323     {
3324         my (@high_array1, @high_array2, @high_data, @high_row_data);
3325         (@high_array2[0..32767], @high_data) = compress_array( 32768, 0, @difftable[65536..$MAX_CHAR] );
3326         (@high_array1[0..1023], @high_row_data) = compress_array( 1024, 0, @high_array2 );
3327
3328         push @res, map { $_ + 1024; } @low_array1;
3329         push @res, map { $_ + @res + @low_row_data + @low_data; } @high_array1;
3330         push @res, map { $_ + @res + @low_row_data - 4096; } @low_row_data;
3331         push @res, @low_data;
3332         push @res, map { 2 * ($_ - 32768) + @res + @high_row_data; } @high_row_data;
3333         return pack( "S<*", 1 + scalar @res  + 2 * scalar @high_data, @res ) . pack( "L<*", @high_data );
3334     }
3335     else
3336     {
3337         push @res, @low_array1;
3338         push @res, map { $_ + @res + @low_row_data - 4096; } @low_row_data;
3339         push @res, @low_data;
3340         return pack "S<*", 1 + scalar @res, @res;
3341     }
3342 }
3343
3344 ################################################################
3345 # dump case mappings for l_intl.nls
3346 sub dump_intl_nls($)
3347 {
3348     my @upper_table = @toupper_table;
3349     my @lower_table = @tolower_table;
3350     remove_linguistic_mappings( \@upper_table, \@lower_table );
3351
3352     my $upper = dump_binary_case_table( @upper_table[0..65535] );
3353     my $lower = dump_binary_case_table( @lower_table[0..65535] );
3354
3355     my $filename = shift;
3356     open OUTPUT,">$filename.new" or die "Cannot create $filename";
3357     printf "Building $filename\n";
3358
3359     binmode OUTPUT;
3360     print OUTPUT pack "S<", 1;  # version
3361     print OUTPUT $upper;
3362     print OUTPUT $lower;
3363     close OUTPUT;
3364     save_file($filename);
3365 }
3366
3367
3368 ################################################################
3369 # dump the bidi direction table
3370 sub dump_bidi_dir_table($)
3371 {
3372     my $filename = shift;
3373     open OUTPUT,">$filename.new" or die "Cannot create $filename";
3374     printf "Building $filename\n";
3375     printf OUTPUT "/* Unicode BiDi direction table */\n";
3376     printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
3377     printf OUTPUT "#include \"windef.h\"\n\n";
3378
3379     my @table;
3380
3381     for (my $i = 0; $i < 65536; $i++)
3382     {
3383         $table[$i] = $bidi_types{$direction_table[$i]} if defined $direction_table[$i];
3384     }
3385
3386     dump_two_level_mapping( "bidi_direction_table", $bidi_types{"L"}, 16, @table );
3387
3388     close OUTPUT;
3389     save_file($filename);
3390 }
3391
3392
3393 sub rol($$)
3394 {
3395     my ($byte, $count) = @_;
3396     return (($byte << $count) | ($byte >> (8 - $count))) & 0xff;
3397 }
3398
3399 ################################################################
3400 # compress the character properties table
3401 sub compress_char_props_table($@)
3402 {
3403     my $rows = shift;
3404     my @table = @_;
3405     my $len = @table / $rows;
3406     my $pos = 0;
3407     my @array = (0) x $rows;
3408     my %sequences;
3409
3410     # add some predefined sequences
3411     foreach my $i (0, 0xfb .. 0xff) { $sequences{pack "L*", (rol($i,5)) x $len} = $i; }
3412
3413     # try to merge table rows
3414     for (my $row = 0; $row < $rows; $row++)
3415     {
3416         my @table_row = map { defined $_ ? $_ : 0x7f; } @table[($row * $len)..(($row + 1) * $len - 1)];
3417         my $rowtxt = pack "L*", @table_row;
3418         if (defined($sequences{$rowtxt}))
3419         {
3420             # reuse an existing row
3421             $array[$row] = $sequences{$rowtxt};
3422         }
3423         else
3424         {
3425             # create a new row
3426             $sequences{$rowtxt} = $array[$row] = ++$pos;
3427             push @array, @table_row;
3428         }
3429     }
3430     return @array;
3431 }
3432
3433 ################################################################
3434 # dump a normalization table in binary format
3435 sub dump_norm_table($)
3436 {
3437     my $filename = shift;
3438
3439     my %forms  = ( "nfc" => 1, "nfd" => 2, "nfkc" => 5, "nfkd" => 6, "idna" => 13 );
3440     my %decomp = ( "nfc" => \@decomp_table,
3441                    "nfd" => \@decomp_table,
3442                    "nfkc" => \@decomp_compat_table,
3443                    "nfkd" => \@decomp_compat_table ,
3444                    "idna" => \@idna_decomp_table );
3445
3446     open OUTPUT,">$filename.new" or die "Cannot create $filename";
3447     print "Building $filename\n";
3448
3449     my $type = $filename;
3450     $type =~ s!.*/norm(\w+)\.nls!$1!;
3451
3452     my $compose = $forms{$type} & 1;
3453     my $compat = !!($forms{$type} & 4) + ($type eq "idna");
3454
3455     my @version = split /\./, $UNIVERSION;
3456
3457     # combining classes
3458
3459     my @classes;
3460     my @class_values;
3461
3462     foreach my $c (grep defined, @combining_class_table)
3463     {
3464         $classes[$c] = 1 if $c < 0x100;
3465     }
3466     for (my $i = 0; $i < @classes; $i++)
3467     {
3468         next unless defined $classes[$i];
3469         $classes[$i] = @class_values;
3470         push @class_values, $i;
3471     }
3472     push @class_values, 0 if (@class_values % 2);
3473     die "too many classes" if @class_values >= 0x40;
3474
3475     # character properties
3476
3477     my @char_props;
3478     my @decomposed;
3479     my @comp_hash_table;
3480     my $comp_hash_size = $compose ? 254 : 0;
3481
3482     for (my $i = 0; $i <= $MAX_CHAR; $i++)
3483     {
3484         next unless defined $combining_class_table[$i];
3485         if (defined $decomp{$type}->[$i])
3486         {
3487             my @dec = get_decomposition( $i, $decomp{$type} );
3488             if ($compose && (my @comp = get_composition( $i, $compat )))
3489             {
3490                 my $hash = ($comp[0] + 95 * $comp[1]) % $comp_hash_size;
3491                 push @{$comp_hash_table[$hash]}, to_utf16( @comp, $i );
3492
3493                 my $val = 0;
3494                 foreach my $d (@dec)
3495                 {
3496                     $val = $combining_class_table[$d];
3497                     last if $val;
3498                 }
3499                 $char_props[$i] = $classes[$val];
3500             }
3501             else
3502             {
3503                 $char_props[$i] = 0xbf;
3504             }
3505             @dec = compose_hangul( @dec ) if $compose;
3506             @dec = to_utf16( @dec );
3507             push @dec, 0 if @dec >= 7;
3508             $decomposed[$i] = \@dec;
3509         }
3510         else
3511         {
3512             if ($combining_class_table[$i] == 0x100)
3513             {
3514                 $char_props[$i] = 0x7f;
3515             }
3516             elsif ($combining_class_table[$i])
3517             {
3518                 $char_props[$i] = $classes[$combining_class_table[$i]] | 0x80;
3519             }
3520             elsif ($type eq "idna" && defined $idna_disallowed[$i])
3521             {
3522                 $char_props[$i] = 0xff;
3523             }
3524             else
3525             {
3526                 $char_props[$i] = 0;
3527             }
3528         }
3529     }
3530
3531     if ($compose)
3532     {
3533         for (my $i = 0; $i <= $MAX_CHAR; $i++)
3534         {
3535             my @comp = get_composition( $i, $compat );
3536             next unless @comp;
3537             if ($combining_class_table[$comp[1]])
3538             {
3539                 $char_props[$comp[0]] |= 0x40 unless $char_props[$comp[0]] & 0x80;
3540                 $char_props[$comp[1]] |= 0x40;
3541             }
3542             else
3543             {
3544                 $char_props[$comp[0]] = ($char_props[$comp[0]] & ~0x40) | 0x80;
3545                 $char_props[$comp[1]] |= 0xc0;
3546             }
3547         }
3548     }
3549
3550     # surrogates
3551     foreach my $i (0xd800..0xdbff) { $char_props[$i] = 0xdf; }
3552     foreach my $i (0xdc00..0xdfff) { $char_props[$i] = 0x9f; }
3553
3554     # Hangul
3555     if ($type eq "nfc") { foreach my $i (0x1100..0x117f) { $char_props[$i] = 0xff; } }
3556     elsif ($compose) { foreach my $i (0x1100..0x11ff) { $char_props[$i] = 0xff; } }
3557     foreach my $i (0xac00..0xd7ff) { $char_props[$i] = 0xff; }
3558
3559     # invalid chars
3560     if ($type eq "idna") { foreach my $i (0x00..0x1f, 0x7f) { $char_props[$i] = 0xff; } }
3561     foreach my $i (0xfdd0..0xfdef) { $char_props[$i] = 0xff; }
3562     foreach my $i (0x00..0x10)
3563     {
3564         $char_props[($i << 16) | 0xfffe] = 0xff;
3565         $char_props[($i << 16) | 0xffff] = 0xff;
3566     }
3567
3568     # decomposition hash table
3569
3570     my @decomp_hash_table;
3571     my @decomp_hash_index;
3572     my @decomp_hash_data;
3573     my $decomp_hash_size = 944;
3574
3575     # build string of character data, reusing substrings when possible
3576     my $decomp_char_data = "";
3577     foreach my $i (sort { @{$b} <=> @{$a} } grep defined, @decomposed)
3578     {
3579         my $str = pack "U*", @{$i};
3580         $decomp_char_data .= $str if index( $decomp_char_data, $str) == -1;
3581     }
3582     for (my $i = 0; $i < @decomposed; $i++)
3583     {
3584         next unless defined $decomposed[$i];
3585         my $pos = index( $decomp_char_data, pack( "U*", @{$decomposed[$i]} ));
3586         die "sequence not found" if $pos == -1;
3587         my $len = @{$decomposed[$i]};
3588         $len = 7 if $len > 7;
3589         my $hash = $i % $decomp_hash_size;
3590         push @{$decomp_hash_table[$hash]}, [ $i, ($len << 13) | $pos ];
3591     }
3592     for (my $i = 0; $i < $decomp_hash_size; $i++)
3593     {
3594         $decomp_hash_index[$i] = @decomp_hash_data / 2;
3595         next unless defined $decomp_hash_table[$i];
3596         if (@{$decomp_hash_table[$i]} == 1)
3597         {
3598             my $entry = $decomp_hash_table[$i]->[0];
3599             if ($char_props[$entry->[0]] == 0xbf)
3600             {
3601                 $decomp_hash_index[$i] = $entry->[1];
3602                 next;
3603             }
3604         }
3605         foreach my $entry (@{$decomp_hash_table[$i]})
3606         {
3607             push @decomp_hash_data, $entry->[0] & 0xffff, $entry->[1];
3608         }
3609     }
3610     push @decomp_hash_data, 0, 0;
3611
3612     # composition hash table
3613
3614     my @comp_hash_index;
3615     my @comp_hash_data;
3616     if (@comp_hash_table)
3617     {
3618         for (my $i = 0; $i < $comp_hash_size; $i++)
3619         {
3620             $comp_hash_index[$i] = @comp_hash_data;
3621             push @comp_hash_data, @{$comp_hash_table[$i]} if defined $comp_hash_table[$i];
3622         }
3623         $comp_hash_index[$comp_hash_size] = @comp_hash_data;
3624         push @comp_hash_data, 0, 0, 0;
3625     }
3626
3627     my $level1 = ($MAX_CHAR + 1) / 128;
3628     my @rows = compress_char_props_table( $level1, @char_props[0..$MAX_CHAR] );
3629
3630     my @header = ( $version[0], $version[1], $version[2], 0, $forms{$type}, $compat ? 18 : 3,
3631                    0, $decomp_hash_size, $comp_hash_size, 0 );
3632     my @tables = (0) x 8;
3633
3634     $tables[0] = 16 + @header + @tables;
3635     $tables[1] = $tables[0] + @class_values / 2;
3636     $tables[2] = $tables[1] + $level1 / 2;
3637     $tables[3] = $tables[2] + (@rows - $level1) / 2;
3638     $tables[4] = $tables[3] + @decomp_hash_index;
3639     $tables[5] = $tables[4] + @decomp_hash_data;
3640     $tables[6] = $tables[5] + length $decomp_char_data;
3641     $tables[7] = $tables[6] + @comp_hash_index;
3642
3643     print OUTPUT pack "S<16", unpack "U*", "norm$type.nlp";
3644     print OUTPUT pack "S<*", @header;
3645     print OUTPUT pack "S<*", @tables;
3646     print OUTPUT pack "C*", @class_values;
3647
3648     print OUTPUT pack "C*", @rows[0..$level1-1];
3649     print OUTPUT pack "C*", @rows[$level1..$#rows];
3650     print OUTPUT pack "S<*", @decomp_hash_index;
3651     print OUTPUT pack "S<*", @decomp_hash_data;
3652     print OUTPUT pack "S<*", unpack "U*", $decomp_char_data;
3653     print OUTPUT pack "S<*", @comp_hash_index;
3654     print OUTPUT pack "S<*", @comp_hash_data;
3655
3656     close OUTPUT;
3657     save_file($filename);
3658
3659     add_registry_string_value( $nlskey, "Normalization", sprintf( "%x", $forms{$type} ), "norm$type.nls" );
3660 }
3661
3662
3663 ################################################################
3664 # output a codepage definition file from the global tables
3665 sub output_codepage_file($)
3666 {
3667     my $codepage = shift;
3668
3669     my $output = sprintf "nls/c_%03d.nls", $codepage;
3670     open OUTPUT,">$output.new" or die "Cannot create $output";
3671
3672     printf "Building %s\n", $output;
3673     if (!@lead_bytes) { dump_binary_sbcs_table( $codepage ); }
3674     else { dump_binary_dbcs_table( $codepage ); }
3675
3676     close OUTPUT;
3677     save_file($output);
3678
3679     add_registry_string_value( $nlskey, "Codepage", sprintf( "%d", $codepage ), sprintf( "c_%03d.nls", $codepage ));
3680 }
3681
3682 ################################################################
3683 # output a codepage table from a Microsoft-style mapping file
3684 sub dump_msdata_codepage($)
3685 {
3686     my $filename = shift;
3687
3688     my $state = "";
3689     my ($codepage, $width, $count);
3690     my ($lb_cur, $lb_end);
3691
3692     @cp2uni = ();
3693     @glyph2uni = ();
3694     @lead_bytes = ();
3695     @uni2cp = ();
3696     $default_char = $DEF_CHAR;
3697     $default_wchar = $DEF_CHAR;
3698
3699     my $INPUT = open_data_file( "codepages", $filename );
3700
3701     while (<$INPUT>)
3702     {
3703         next if /^;/;  # skip comments
3704         next if /^\s*$/;  # skip empty lines
3705         next if /\x1a/;  # skip ^Z
3706         last if /^ENDCODEPAGE/;
3707
3708         if (/^CODEPAGE\s+(\d+)/)
3709         {
3710             $codepage = $1;
3711             next;
3712         }
3713         if (/^CPINFO\s+(\d+)\s+0x([0-9a-fA-f]+)\s+0x([0-9a-fA-F]+)/)
3714         {
3715             $width = $1;
3716             $default_char = hex $2;
3717             $default_wchar = hex $3;
3718             next;
3719         }
3720         if (/^(MBTABLE|GLYPHTABLE|WCTABLE|DBCSRANGE|DBCSTABLE)\s+(\d+)/)
3721         {
3722             $state = $1;
3723             $count = $2;
3724             next;
3725         }
3726         if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)/)
3727         {
3728             if ($state eq "MBTABLE")
3729             {
3730                 my $cp = hex $1;
3731                 my $uni = hex $2;
3732                 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
3733                 next;
3734             }
3735             if ($state eq "GLYPHTABLE")
3736             {
3737                 my $cp = hex $1;
3738                 my $uni = hex $2;
3739                 $glyph2uni[$cp] = $uni unless defined($glyph2uni[$cp]);
3740                 next;
3741             }
3742             if ($state eq "WCTABLE")
3743             {
3744                 my $uni = hex $1;
3745                 my $cp = hex $2;
3746                 $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
3747                 next;
3748             }
3749             if ($state eq "DBCSRANGE")
3750             {
3751                 my $start = hex $1;
3752                 my $end = hex $2;
3753                 for (my $i = $start; $i <= $end; $i++) { add_lead_byte( $i ); }
3754                 $lb_cur = $start;
3755                 $lb_end = $end;
3756                 next;
3757             }
3758             if ($state eq "DBCSTABLE")
3759             {
3760                 my $mb = hex $1;
3761                 my $uni = hex $2;
3762                 my $cp = ($lb_cur << 8) | $mb;
3763                 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
3764                 if (!--$count)
3765                 {
3766                     if (++$lb_cur > $lb_end) { $state = "DBCSRANGE"; }
3767                 }
3768                 next;
3769             }
3770         }
3771         die "$filename: Unrecognized line $_\n";
3772     }
3773     close $INPUT;
3774
3775     output_codepage_file( $codepage );
3776
3777     if ($codepage == 949) { dump_krwansung_codepage( @uni2cp ); }
3778 }
3779
3780 ################################################################
3781 # align a string length
3782 sub align_string($$)
3783 {
3784     my ($align, $str) = @_;
3785     $str .= pack "C*", (0) x ($align - length($str) % $align) if length($str) % $align;
3786     return $str;
3787 }
3788
3789 ################################################################
3790 # pad a string with zeros
3791 sub pad_string($$)
3792 {
3793     my ($pad, $str) = @_;
3794     $str .= pack "C*", (0) x ($pad - length($str)) if length($str) < $pad;
3795     return $str;
3796 }
3797
3798 ################################################################
3799 # pack a GUID string
3800 sub pack_guid($)
3801 {
3802     $_ = shift;
3803     /([0-9A-Fa-f]{8})-([0-9A-Fa-f]{4})-([0-9A-Fa-f]{4})-([0-9A-Fa-f]{2})([0-9A-Fa-f]{2})-([0-9A-Fa-f]{2})([0-9A-Fa-f]{2})([0-9A-Fa-f]{2})([0-9A-Fa-f]{2})([0-9A-Fa-f]{2})([0-9A-Fa-f]{2})/;
3804     return pack "L<S<2C8", hex $1, hex $2, hex $3, hex $4, hex $5, hex $6, hex $7, hex $8, hex $9, hex $10, hex $11;
3805 }
3806
3807 ################################################################
3808 # comparison function for compression sort
3809 sub cmp_compression
3810 {
3811     return scalar @{$a} <=> scalar @{$b} ||
3812         $a->[4] <=> $b->[4] ||
3813         $a->[5] <=> $b->[5] ||
3814         $a->[6] <=> $b->[6] ||
3815         $a->[7] <=> $b->[7] ||
3816         $a->[8] <=> $b->[8] ||
3817         $a->[9] <=> $b->[9] ||
3818         $a->[10] <=> $b->[10] ||
3819         $a->[11] <=> $b->[11] ||
3820         $a->[12] <=> $b->[12];
3821 }
3822
3823 ################################################################
3824 # build a binary sort keys table
3825 sub dump_sortkey_table($)
3826 {
3827     my $filename = shift;
3828     my @keys;
3829     my ($part, $section, $subsection, $guid, $version, $ling_flag);
3830     my @multiple_weights;
3831     my @expansions;
3832     my @compressions;
3833     my %exceptions;
3834     my %guids;
3835     my %compr_flags;
3836     my %locales;
3837     my $default_guid = "00000001-57ee-1e5c-00b4-d0000bb1e11e";
3838     my $jamostr = "";
3839
3840     my $re_hex = '0x[0-9A-Fa-f]+';
3841     my $re_key = '(\d+\s+\d+\s+\d+\s+\d+)';
3842     $guids{$default_guid} = { };
3843
3844     my %flags = ( "HAS_3_BYTE_WEIGHTS" => 0x01, "REVERSEDIACRITICS" => 0x10, "DOUBLECOMPRESSION" => 0x20, "INVERSECASING" => 0x40 );
3845
3846     my $KEYS = open_data_file( "sorting" );
3847
3848     printf "Building $filename\n";
3849
3850     while (<$KEYS>)
3851     {
3852         s/\s*;.*$//;
3853         next if /^\s*$/;  # skip empty lines
3854         if (/^\s*(SORTKEY|SORTTABLES)/)
3855         {
3856             $part = $1;
3857             next;
3858         }
3859         if (/^\s*(ENDSORTKEY|ENDSORTTABLES)/)
3860         {
3861             $part = $section = "";
3862             next;
3863         }
3864         if (/^\s*(DEFAULT|RELEASE|REVERSEDIACRITICS|DOUBLECOMPRESSION|INVERSECASING|MULTIPLEWEIGHTS|EXPANSION|COMPATIBILITY|COMPRESSION|EXCEPTION|JAMOSORT)\s+/)
3865         {
3866             $section = $1;
3867             $guid = undef;
3868             next;
3869         }
3870         next unless $part;
3871         if ("$part.$section" eq "SORTKEY.DEFAULT")
3872         {
3873             if (/^\s*($re_hex)\s+$re_key/)
3874             {
3875                 $keys[hex $1] = [ split(/\s+/,$2) ];
3876                 next;
3877             }
3878         }
3879         elsif ("$part.$section" eq "SORTTABLES.RELEASE")
3880         {
3881             if (/^\s*NLSVERSION\s+0x([0-9A-Fa-f]+)/)
3882             {
3883                 $version = hex $1;
3884                 next;
3885             }
3886             if (/^\s*DEFINEDVERSION\s+0x([0-9A-Fa-f]+)/)
3887             {
3888                 # ignore for now
3889                 next;
3890             }
3891         }
3892         elsif ("$part.$section" eq "SORTTABLES.REVERSEDIACRITICS" ||
3893                "$part.$section" eq "SORTTABLES.DOUBLECOMPRESSION" ||
3894                "$part.$section" eq "SORTTABLES.INVERSECASING")
3895         {
3896             if (/^\s*SORTGUID\s+([-0-9A-Fa-f]+)/)
3897             {
3898                 $guid = lc $1;
3899                 $guids{$guid} = { } unless defined $guids{$guid};
3900                 $guids{$guid}->{flags} |= $flags{$section};
3901                 next;
3902             }
3903             if (/^\s*LOCALENAME\s+([A-Za-z0-9-_]+)/)
3904             {
3905                 $locales{$1} = $guid;
3906                 next;
3907             }
3908         }
3909         elsif ("$part.$section" eq "SORTTABLES.MULTIPLEWEIGHTS")
3910         {
3911             if (/^\s*(\d+)\s+(\d+)/)
3912             {
3913                 push @multiple_weights, $1, $2;
3914                 next;
3915             }
3916         }
3917         elsif ("$part.$section" eq "SORTTABLES.EXPANSION")
3918         {
3919             if (/^\s*0x([0-9A-Fa-f]+)\s+0x([0-9A-Fa-f]+)\s+0x([0-9A-Fa-f]+)/)
3920             {
3921                 my $pos = scalar @expansions / 2;
3922                 $keys[hex $1] = [ 2, 0, $pos & 0xff, $pos >> 8 ] unless defined $keys[hex $1];
3923                 push @expansions, hex $2, hex $3;
3924                 next;
3925             }
3926         }
3927         elsif ("$part.$section" eq "SORTTABLES.COMPATIBILITY")
3928         {
3929             if (/^\s*0x([0-9A-Fa-f]+)\s+0x([0-9A-Fa-f]+)/)
3930             {
3931                 $keys[hex $1] = $keys[hex $2];
3932                 next;
3933             }
3934         }
3935         elsif ("$part.$section" eq "SORTTABLES.COMPRESSION")
3936         {
3937             if (/^\s*SORTGUID\s+([-0-9A-Fa-f]+)\s+\d*\s*([A-Z0-9_]+)?/)
3938             {
3939                 if ($subsection || !$guid)  # start a new one
3940                 {
3941                     $guid = lc $1;
3942                     $subsection = "";
3943                     $guids{$guid} = { } unless defined $guids{$guid};
3944                     $guids{$guid}->{flags} |= $flags{$2} if $2;
3945                     $guids{$guid}->{compr} = @compressions;
3946                     $exceptions{"$guid-"} = [ ] unless defined $exceptions{"$guid-"};
3947                     $compr_flags{$guid} = [ ] unless defined $compr_flags{$guid};
3948                     push @compressions, [ ];
3949                 }
3950                 else  # merge with current one
3951                 {
3952                     $guids{lc $1} = { } unless defined $guids{lc $1};
3953                     $guids{lc $1}->{flags} |= $flags{$2} if $2;
3954                     $guids{lc $1}->{compr} = $guids{$guid}->{compr};
3955                     $compr_flags{lc $1} = $compr_flags{$guid};
3956                 }
3957                 next;
3958             }
3959             if (/^\s*LOCALENAME\s+([A-Za-z0-9-_]+)/)
3960             {
3961                 $locales{$1} = $guid;
3962                 next;
3963             }
3964             if (/^\s*(TWO|THREE|FOUR|FIVE|SIX|SEVEN|EIGHT)/)
3965             {
3966                 $subsection = $1;
3967                 next;
3968             }
3969             if ($subsection && /^\s*(($re_hex\s+){2,8})$re_key/)
3970             {
3971                 my @comp = map { hex $_; } split(/\s+/,$1);
3972                 push @{$compressions[$#compressions]}, [ split(/\s+/,$3), @comp ];
3973                 # add compression flags
3974                 $compr_flags{$guid}->[$comp[0]] |= @comp >= 6 ? 0xc0 : @comp >= 4 ? 0x80 : 0x40;
3975                 next;
3976             }
3977         }
3978         elsif ("$part.$section" eq "SORTTABLES.EXCEPTION")
3979         {
3980             if (/^\s*SORTGUID\s+([-0-9A-Fa-f]+)\s+\d*\s*(LINGUISTIC_CASING)?/)
3981             {
3982                 $guid = lc $1;
3983                 $guids{$guid} = { } unless defined $guids{lc $1};
3984                 $ling_flag = ($2 ? "+" : "-");
3985                 $exceptions{"$guid$ling_flag"} = [ ] unless defined $exceptions{"$guid$ling_flag"};
3986                 next;
3987             }
3988             if (/^\s*LOCALENAME\s+([A-Za-z0-9-_]+)/)
3989             {
3990                 $locales{$1} = $guid;
3991                 next;
3992             }
3993             if (/^\s*($re_hex)\s+$re_key/)
3994             {
3995                 $exceptions{"$guid$ling_flag"}->[hex $1] = [ split(/\s+/,$2) ];
3996                 next;
3997             }
3998         }
3999         elsif ("$part.$section" eq "SORTTABLES.JAMOSORT")
4000         {
4001             if (/^\s*$re_hex\s+(($re_hex\s*){5})/)
4002             {
4003                 $jamostr .= pack "C8", map { hex $_; } split /\s+/, $1;
4004                 next;
4005             }
4006         }
4007         die "$current_data_file: $part.$section: unrecognized line $_\n";
4008     }
4009     close $KEYS;
4010
4011     # Sortkey table
4012
4013     my $table;
4014     for (my $i = 0; $i < 0x10000; $i++)
4015     {
4016         my @k = defined $keys[$i] ? @{$keys[$i]} : (0) x 4;
4017         $table .= pack "C4", $k[1], $k[0], $k[2], $k[3];
4018     }
4019
4020     foreach my $id (sort keys %exceptions)
4021     {
4022         my $pos = length($table) / 4;
4023         my @exc = @{$exceptions{$id}};
4024         my @filled;
4025         my $key = (substr( $id, -1 ) eq "+" ? "ling_except" : "except");
4026         my $guid = substr( $id, 0, -1 );
4027         $guids{$guid}->{$key} = $pos;
4028         $pos += 0x100;
4029         my @flags = @{$compr_flags{$guid}} if defined $compr_flags{$guid};
4030         for (my $j = 0; $j < 0x10000; $j++)
4031         {
4032             next unless defined $exc[$j] || defined $flags[$j];
4033             $filled[$j >> 8] = 1;
4034             $j |= 0xff;
4035         }
4036         for (my $j = 0; $j < 0x100; $j++)
4037         {
4038             $table .= pack "L<", $filled[$j] ? $pos : $j * 0x100;
4039             $pos += 0x100 if $filled[$j];
4040         }
4041         for (my $j = 0; $j < 0x10000; $j++)
4042         {
4043             next unless $filled[$j >> 8];
4044             my @k = defined $exc[$j] ? @{$exc[$j]} : defined $keys[$j] ? @{$keys[$j]} : (0) x 4;
4045             $k[3] |= $flags[$j] || 0;
4046             $table .= pack "C4", $k[1], $k[0], $k[2], $k[3];
4047         }
4048     }
4049
4050     # Case mapping tables
4051
4052     # standard table
4053     my @casemaps;
4054     my @upper = @toupper_table;
4055     my @lower = @tolower_table;
4056     remove_linguistic_mappings( \@upper, \@lower );
4057     $casemaps[0] = pack( "S<*", 1) . dump_binary_case_table( @upper ) . dump_binary_case_table( @lower );
4058
4059     # linguistic table
4060     $casemaps[1] = pack( "S<*", 1) . dump_binary_case_table( @toupper_table ) . dump_binary_case_table( @tolower_table );
4061
4062     # Turkish table
4063     @upper = @toupper_table;
4064     @lower = @tolower_table;
4065     $upper[ord 'i'] = 0x130; # LATIN CAPITAL LETTER I WITH DOT ABOVE
4066     $lower[ord 'I'] = 0x131; # LATIN SMALL LETTER DOTLESS I
4067     $casemaps[2] = pack( "S<*", 1) . dump_binary_case_table( @upper ) . dump_binary_case_table( @lower );
4068     my $casemaps = align_string( 8, $casemaps[0] . $casemaps[1] . $casemaps[2] );
4069
4070     # Char type table
4071
4072     my @table;
4073     my $types = "";
4074     my %typestr;
4075     for (my $i = 0; $i < 0x10000; $i++)
4076     {
4077         my $str = pack "S<3",
4078             ($category_table[$i] || 0) & 0xffff,
4079             defined($direction_table[$i]) ? $c2_types{$direction_table[$i]} : 0,
4080             ($category_table[$i] || 0) >> 16;
4081
4082         if (!defined($typestr{$str}))
4083         {
4084             $typestr{$str} = length($types) / 6;
4085             $types .= $str;
4086         }
4087         $table[$i] = $typestr{$str};
4088     }
4089
4090     my (@rows, @array, @data, @row_data);
4091     (@rows[0..4095], @data) = compress_array( 4096, 0, @table[0..65535] );
4092     (@array[0..255], @row_data) = compress_array( 256, 0, @rows );
4093     for (my $i = 0; $i < 256; $i++) { $array[$i] *= 2; }  # we need byte offsets
4094     for (my $i = 0; $i < @row_data; $i++) { $row_data[$i] += 2 * @row_data + 512 - 4096; }
4095
4096     my $arraystr = pack("S<*", @array, @row_data) . pack("C*", @data);
4097     my $chartypes = pack "S<2", 4 + length($types) + length($arraystr), 2 + length($types);
4098     $chartypes = align_string( 8, $chartypes . $types . $arraystr );
4099
4100     # Sort tables
4101
4102     # guids
4103     my $sorttables = pack "L<2", $version, scalar %guids;
4104     foreach my $id (sort keys %guids)
4105     {
4106         my %guid = %{$guids{$id}};
4107         my $flags = $guid{flags} || 0;
4108         my $map = length($casemaps[0]) + (defined $guid{ling_except} ? length($casemaps[1]) : 0);
4109         $sorttables .= pack_guid($id) . pack "L<5",
4110             $flags,
4111             defined($guid{compr}) ? $guid{compr} : 0xffffffff,
4112             $guid{except} || 0,
4113             $guid{ling_except} || 0,
4114             $map / 2;
4115     }
4116
4117     # expansions
4118     $sorttables .= pack "L<S<*", scalar @expansions / 2, @expansions;
4119
4120     # compressions
4121     $sorttables .= pack "L<", scalar @compressions;
4122     my $rowstr = "";
4123     foreach my $c (@compressions)
4124     {
4125         my $pos = length($rowstr) / 2;
4126         my $min = 0xffff;
4127         my $max = 0;
4128         my @lengths = (0) x 8;
4129         foreach my $r (sort cmp_compression @{$c})
4130         {
4131             my @row = @{$r};
4132             $lengths[scalar @row - 6]++;
4133             foreach my $val (@row[4..$#row])
4134             {
4135                 $min = $val if $min > $val;
4136                 $max = $val if $max < $val;
4137             }
4138             $rowstr .= align_string( 4, pack "S<*", @row[4..$#row] );
4139             $rowstr .= pack "C4", $row[1], $row[0], $row[2], $row[3];
4140         }
4141         $sorttables .= pack "L<S<10", $pos, $min, $max, @lengths;
4142     }
4143     $sorttables .= $rowstr;
4144
4145     # multiple weights
4146     $sorttables .= align_string( 4, pack "L<C*", scalar @multiple_weights / 2, @multiple_weights );
4147
4148     # jamo sort
4149     $sorttables .= pack("L<", length($jamostr) / 8) . $jamostr;
4150
4151     # Locales
4152
4153     add_registry_key( $nlskey, "Sorting\\Ids", "{$default_guid}" );
4154     foreach my $loc (sort keys %locales)
4155     {
4156         # skip specific locales that match more general ones
4157         my @parts = split /[-_]/, $loc;
4158         next if @parts > 1 && defined($locales{$parts[0]}) && $locales{$parts[0]} eq $locales{$loc};
4159         next if @parts > 2 && defined($locales{"$parts[0]-$parts[1]"}) && $locales{"$parts[0]-$parts[1]"} eq $locales{$loc};
4160         add_registry_string_value( $nlskey, "Sorting\\Ids", $loc, "\{$locales{$loc}\}" );
4161     }
4162
4163     # File header
4164
4165     my @header;
4166     $header[0] = 16;
4167     $header[1] = $header[0] + length $table;
4168     $header[2] = $header[1] + length $casemaps;
4169     $header[3] = $header[2] + length $chartypes;
4170
4171     open OUTPUT, ">$filename.new" or die "Cannot create $filename";
4172     print OUTPUT pack "L<*", @header;
4173     print OUTPUT $table, $casemaps, $chartypes, $sorttables;
4174     close OUTPUT;
4175     save_file($filename);
4176     return $chartypes;
4177 }
4178
4179
4180 my %lcnames;
4181
4182 sub locale_parent($)
4183 {
4184     my $loc = shift;
4185
4186     return undef unless $loc;
4187     return $lcnames{$loc}->{sparent} if defined $lcnames{$loc} && defined $lcnames{$loc}->{sparent};
4188     return $lcnames{$loc}->{parent} if defined $lcnames{$loc} && defined $lcnames{$loc}->{parent};
4189     if ($loc =~ /(.*)-[0-9A-Za-z]+/) { return $1; }
4190     return "";
4191 }
4192
4193 sub compare_locales
4194 {
4195     (my $n1 = $a) =~ tr/A-Z_/a-z-/;
4196     (my $n2 = $b) =~ tr/A-Z_/a-z-/;
4197     return $n1 cmp $n2;
4198 }
4199
4200 # query an xml key
4201 sub xml_query($$)
4202 {
4203     my ($xml, $query) = @_;
4204     my $ret = $xml->find( $query );
4205     return undef unless $ret;
4206     printf STDERR "multiple entries for %s\n", $query if (@{$ret} > 1);
4207     return @{$ret}[0]->textContent;
4208 }
4209
4210 # query an xml key for a locale, with fallback to the parents
4211 sub loc_query($$)
4212 {
4213     my ($loc, $query) = @_;
4214
4215     $loc = $lcnames{"en-US"} unless $loc->{name};  # fallback to "en-US" for root locale
4216
4217     for (my $cur = $loc->{name}; defined $cur; $cur = locale_parent( $cur ))
4218     {
4219         next unless defined $lcnames{$cur};
4220         my $xml = $lcnames{$cur}->{xml};
4221         my $ret = $xml->find( $query );
4222         next unless $ret;
4223         printf STDERR "%s: multiple entries for %s\n", $cur, $query if (@{$ret} > 1);
4224         next if @{$ret}[0]->textContent eq "\x{2191}\x{2191}\x{2191}"; # "↑↑↑"
4225         return @{$ret}[0]->textContent;
4226     }
4227     return undef;
4228 }
4229
4230 # retrieve a locale field entry by going up the parents tree
4231 sub locale_entry($$$)
4232 {
4233     my ($loc, $field, $def) = @_;
4234
4235     return $loc->{$field} if defined $loc->{$field};
4236
4237     unless ($loc->{name})  # fallback to "en-US" for root locale
4238     {
4239         $loc = $lcnames{"en-US"};
4240         return $loc->{$field} if defined $loc->{$field};
4241     }
4242     while (defined $loc->{alias})  # resolve aliases
4243     {
4244         $loc = $lcnames{$loc->{alias}};
4245         return $loc->{$field} if defined $loc->{$field};
4246     }
4247     my $cur = $loc->{name};
4248     while ($cur)
4249     {
4250         if (defined $lcnames{$cur} && defined $lcnames{$cur}->{sparent})
4251         {
4252             $cur = $lcnames{$cur}->{sparent};
4253         }
4254         elsif ($cur =~ /(.*)-[0-9A-Za-z]+/)
4255         {
4256             $cur = $1;
4257         }
4258         else
4259         {
4260             return $def;
4261         }
4262         return $lcnames{$cur}->{$field} if defined $lcnames{$cur} && defined $lcnames{$cur}->{$field};
4263     }
4264     return $def;
4265 }
4266
4267 my $string_data;
4268
4269 sub add_str_data($)
4270 {
4271     my $txt = shift;
4272     my $ret = index( $string_data, $txt );
4273     if ($ret == -1)
4274     {
4275         $ret = length($string_data);
4276         $string_data .= $txt
4277     }
4278     return $ret / 2;
4279 }
4280
4281 sub add_string($)
4282 {
4283     my $str = shift;
4284     return 0 unless defined($str) && $str ne "";
4285     my $utf = encode( "UTF16LE", $str );
4286     return add_str_data( (pack "S<", length($utf) / 2) . $utf . (pack "S", 0) );
4287 }
4288
4289 sub add_fontsig(@)
4290 {
4291     return add_str_data( pack "S<L<*", scalar(@_) * 2, @_ );
4292 }
4293
4294 sub add_strarray(@)
4295 {
4296     return 0 unless @_;
4297     return add_str_data( pack "S<L<*", scalar @_, map { add_string($_) } @_);
4298 }
4299
4300 sub format_to_grouping($)
4301 {
4302     my $format = shift;
4303     if ($format =~ /#,(#+),(#+0)/) { return chr(length($2)) . chr(length($1)); }
4304     if ($format =~ /#,(#+0)/) { return chr(length($1)); }
4305 #    printf STDERR "unknown format %s\n", $format;
4306     return chr(3);
4307 }
4308
4309 sub parse_currency_format($$)
4310 {
4311     my $name = shift;
4312     my ($posfmt, $negfmt) = split /;/, shift;
4313     my @pospatterns = ( "\xa4[^\xa0]*#",      # $1.1
4314                         "00[^\xa0]*\xa4",     # 1.1$
4315                         "\xa4.*\xa0.*#",      # $ 1.1
4316                         "00.*\xa0.*\xa4" );   # 1.1 $
4317     my @negpatterns = ( "\\(\xa4[^\xa0]*#",   # ($1.1)
4318                         "-\xa4[^\xa0]*#",     # -$1.1
4319                         "\xa4[^\xa0]*-#",     # $-1.1
4320                         "\xa4[^\xa0]*#.*00-", # $1.1-
4321                         "00[^\xa0]*\xa4\\)",  # (1.1$)
4322                         "-#.*00[^\xa0]*\xa4", # -1.1$
4323                         "00-[^\xa0]*\xa4",    # 1.1-$
4324                         "00[^\xa0]*\xa4-",    # 1.1$-
4325                         "-#.*00.*\xa0.*\xa4", # -1.1 $
4326                         "-\xa4.*\xa0.*#",     # -$ 1.1
4327                         "00.*\xa0.*\xa4-",    # 1.1 $-
4328                         "\xa4.*\xa0.*#.*00-", # $ 1.1-
4329                         "\xa4.*\xa0.*-#",     # $ -1.1
4330                         "00-.*\xa0.*\xa4",    # 1.1- $
4331                         "\\(\xa4.*\xa0.*#",   # ($ 1.1)
4332                         "00.*\xa0.*\xa4\\)"); # (1.1 $)
4333     my ($pos, $neg);
4334
4335     for ($pos = 0; $pos < @pospatterns; $pos++)
4336     {
4337         last if ($posfmt =~ /$pospatterns[$pos]/);
4338     }
4339     #printf STDERR "$name: unknown format '%s'\n", $posfmt if ($pos == @pospatterns);
4340     $pos = 0 if ($pos == @pospatterns);
4341
4342     if (defined $negfmt)
4343     {
4344         for ($neg = 0; $neg < @negpatterns; $neg++)
4345         {
4346             last if ($negfmt =~ /$negpatterns[$neg]/);
4347         }
4348         #printf STDERR "$name: unknown format '%s'\n", $negfmt if ($neg == @negpatterns);
4349         $neg = 0 if ($neg == @negpatterns);
4350     }
4351     elsif ($pos == 0) { $neg = 1; }
4352     elsif ($pos == 1) { $neg = 5; }
4353     elsif ($pos == 2) { $neg = 9; }
4354     elsif ($pos == 3) { $neg = 8; }
4355
4356     return ($pos, $neg);
4357 }
4358
4359 sub parse_percent_format($)
4360 {
4361     my $fmt = shift;
4362     my @patterns = ( "0.+%",    # 1 %
4363                      "0%",      # 1%
4364                      "%#",      # %1
4365                      "%.+#" );  # % 1
4366     my $pos;
4367     for ($pos = 0; $pos < @patterns; $pos++)
4368     {
4369         last if ($fmt =~ /$patterns[$pos]/);
4370     }
4371     printf STDERR "unknown format '%s'\n", $fmt if ($pos == @patterns);
4372     return ($pos, ($pos == 3) ? 7 : $pos);
4373 }
4374
4375 sub convert_date_format($)
4376 {
4377     my $fmt = shift;
4378     $fmt =~ s/G+/gg/;
4379     $fmt =~ s/LLLL/MMMM/;
4380     $fmt =~ s/LLL/MMM/;
4381     $fmt =~ s/E+/dddd/;
4382     $fmt =~ s/ccc+/dddd/;
4383     $fmt =~ s/([^gy])y([^y])/$1yyyy$2/;
4384     $fmt =~ s/^y([^y])/yyyy$1/;
4385     $fmt =~ s/([^gy])y$/$1yyyy/;
4386     return $fmt;
4387 }
4388
4389 sub convert_time_format($)
4390 {
4391     my $fmt = shift;
4392     $fmt =~ s/a+/tt/;
4393     $fmt =~ s/B+/tt/;
4394     return $fmt;
4395 }
4396
4397 sub load_iso639()
4398 {
4399     my %iso639;
4400     my $DATA = open_data_file( "iso639", "iso-639-3_Code_Tables_$ISO639VERSION/iso-639-3.tab" );
4401     while (<$DATA>)
4402     {
4403         if (/^\s*[a-z]{3}\s+[a-z]{3}\s+([a-z]{3})\s+([a-z]{2})\s/) { $iso639{$2} = $1; }
4404     }
4405     close $DATA;
4406     return %iso639;
4407 }
4408
4409
4410 ################################################################
4411 # build the locale table for locale.nls
4412 sub build_locale_data()
4413 {
4414     my $base = "cldr-release-$CLDRVERSION";
4415     my $suppl = load_xml_data_file( "cldr", "$base/common/supplemental/supplementalData.xml" );
4416     my $subtags = load_xml_data_file( "cldr", "$base/common/supplemental/likelySubtags.xml" );
4417     my $numbers = load_xml_data_file( "cldr", "$base/common/supplemental/numberingSystems.xml" );
4418     # obsolete phone data from CLDR version 33
4419     my $phone = load_xml_data_file( "cldr33", "common/supplemental/telephoneCodeData.xml" );
4420     my %iso639 = load_iso639();
4421     $string_data = pack "S2", 0, 0;  # offset 0 == empty string
4422
4423     %lcnames = map { $_->{name} => $_ } @locales;
4424
4425     my %lcids;
4426     foreach my $loc (@locales) { $lcids{$loc->{lcid}} = $loc if defined $loc->{lcid}; }
4427
4428     my %days = ( "sun" => 0, "mon" => 1, "tue" => 2, "wed" => 3, "thu" => 4, "fri" => 5, "sat" => 6 );
4429
4430     # assign locale parents
4431
4432     foreach my $loc (@locales)
4433     {
4434         next if $loc->{name} eq "";
4435         next if defined $loc->{parent};
4436         (my $unix_name = $loc->{name}) =~ s/-/_/g;
4437         my $parent = xml_query( $suppl, "/supplementalData/parentLocales/parentLocale[contains(concat(' ',\@locales,' '),' $unix_name ')]/\@parent" );
4438         if ($parent)
4439         {
4440             $parent =~ s/_/-/g;
4441             $parent = "" if $parent eq "root";
4442         }
4443         elsif ($loc->{name} =~ /(.*)-[0-9A-Za-z]+/) { $parent = $1; }
4444         $loc->{parent} = $parent || "";
4445     }
4446
4447     # load per-locale XML files
4448
4449     foreach my $loc (@locales)
4450     {
4451         next if defined $loc->{alias};
4452         (my $file = $loc->{file} || $loc->{name}) =~ s/-/_/g;
4453         $file = "$base/" . ($loc->{dir} || "common") . "/main/$file.xml";
4454         my $xml = load_xml_data_file( "cldr", $file );
4455         $loc->{xml} = $xml;
4456         $loc->{language} ||= xml_query( $xml, "/ldml/identity/language/\@type" );
4457         $loc->{territory} ||= xml_query( $xml, "/ldml/identity/territory/\@type" );
4458         $loc->{script} = xml_query( $xml, "/ldml/identity/script/\@type" );
4459         if (!defined($loc->{territory}) && $loc->{name} =~ /-([A-Z]{2}|[0-9]{3})$/) { $loc->{territory} = $1; }
4460         if (!defined($loc->{script}) && $loc->{name} =~ /-([A-Z][a-z]{3})(-[A-Z]{2})?$/) { $loc->{script} = $1; }
4461     }
4462
4463     # assign a default territory and sort locale
4464
4465     foreach my $loc (@locales)
4466     {
4467         next if defined $loc->{alias};
4468         next if defined $loc->{territory};
4469         my $id = $loc->{sortlocale};
4470         if (defined $id && ($id =~ /[-_]([A-Z0-9]+)$/))
4471         {
4472             $loc->{territory} = $1;
4473             next;
4474         }
4475         my @children = grep /^$loc->{name}-[A-Z0-9]+$/ && !defined $lcnames{$_}->{alias}, keys %lcnames;
4476         if (@children == 1)
4477         {
4478             $id = $children[0];
4479         }
4480         else
4481         {
4482             my $name = $loc->{file} || $loc->{name};
4483             $name =~ s/-(Arab|Cyrl|Deva|Guru|Hans|Hant|Latn|Tfng|Vaii)$//;
4484             $name =~ s/-/_/g;
4485             $id = xml_query( $subtags, "/supplementalData/likelySubtags/likelySubtag[\@from='$name']/\@to" );
4486             $id =~ s/_/-/g if $id;
4487         }
4488         if ($id =~ /[-_]([A-Z0-9]+)$/)
4489         {
4490             $loc->{territory} = $1;
4491             next if defined $loc->{sortlocale};
4492             next unless $id =~ /^$loc->{name}/;
4493             while (defined $lcnames{$id} && defined $lcnames{$id}->{alias}) { $id = $lcnames{$id}->{alias}; }
4494             $loc->{sortlocale} = $id if defined $lcnames{$id};
4495             next;
4496         }
4497         print STDERR "no territory found for $loc->{name}\n";
4498     }
4499
4500     # fill geoid table
4501
4502     my %geotable;
4503     foreach my $geo (@geoids)
4504     {
4505         my $name = $geo->{name};
4506         next unless defined $name;
4507         $geo->{alias} = $geotable{$name} if defined $geotable{$name};
4508         $geotable{$name} ||= $geo;
4509     }
4510     foreach my $loc (@locales)
4511     {
4512         next if defined $loc->{alias};
4513         my $territory = $loc->{territory};
4514         $geotable{$territory} ||= { name => $territory };
4515     }
4516     foreach my $name (keys %geotable)
4517     {
4518         my $geo = $geotable{$name};
4519         $geo->{dialcode} = xml_query( $phone, "(/supplementalData/telephoneCodeData/codesByTerritory[\@territory='$name']/telephoneCountryCode)[1]/\@code" );
4520         if ($name =~ /\d+/)
4521         {
4522             $geo->{uncode} = $name;
4523             next;
4524         }
4525         $geo->{iso2} = $name;
4526         $geo->{iso3} = xml_query( $suppl, "/supplementalData/codeMappings/territoryCodes[\@type='$name']/\@alpha3");
4527         $geo->{uncode} = xml_query( $suppl, "/supplementalData/codeMappings/territoryCodes[\@type='$name']/\@numeric");
4528         $geo->{sintlsymbol} ||= xml_query( $suppl, "(/supplementalData/currencyData/region[\@iso3166='$name']/currency[not(\@to)])[1]/\@iso4217") || "XXX";
4529         $geo->{sintlsymbol} =~ s/XXX/XDR/;
4530     }
4531     foreach my $geo (@geoids)
4532     {
4533         $geo->{parentid} = $geotable{$geo->{parent}}->{id} if defined $geo->{parent};
4534         next if defined $geo->{iso2};
4535         next if defined $geo->{alias};
4536         next unless defined $geo->{uncode};
4537         my @contains;
4538         my $list = xml_query( $suppl, "/supplementalData/territoryContainment/group[\@type='$geo->{uncode}' and not(\@status)]/\@contains");
4539         push @contains, split /\s+/, $list if defined $list;
4540         $list = xml_query( $suppl, "/supplementalData/territoryContainment/group[\@type='$geo->{uncode}' and \@status='deprecated']/\@contains");
4541         push @contains, split /\s+/, $list if defined $list;
4542         while (@contains)
4543         {
4544             my $territory = pop @contains;
4545             if (defined $geotable{$territory})
4546             {
4547                 $geotable{$territory}->{parentid} ||= $geo->{id};
4548             }
4549             elsif ($territory =~ /\d+/)
4550             {
4551                 # expand region recursively
4552                 $list = xml_query( $suppl, "/supplementalData/territoryContainment/group[\@type='$territory' and not(\@status)]/\@contains" );
4553                 push @contains, split /\s+/, $list if defined $list;
4554             }
4555         }
4556     }
4557
4558     # assign calendars to their locale
4559
4560     foreach my $cal (@calendars)
4561     {
4562         next unless defined $cal->{locale};
4563         my $loc = $lcnames{$cal->{locale}};
4564         $loc->{calendar} = [ ] unless defined $loc->{calendar};
4565         push @{$loc->{calendar}}, $cal;
4566     }
4567
4568     # assign default lcid to aliases
4569
4570     foreach my $loc (@locales)
4571     {
4572         next unless defined $loc->{alias};
4573         next if defined $loc->{lcid};
4574         my $alias = $loc->{alias};
4575         my $lcid = $lcnames{$alias}->{lcid} || 0x1000;
4576         $loc->{lcid} = $lcid | 0x80000000;
4577     }
4578
4579     # assign sort aliases to parent locale
4580
4581     foreach my $loc (@locales)
4582     {
4583         next unless $loc->{name} =~ /_/;
4584         next unless defined $loc->{alias};
4585         my $alias = $loc->{alias};
4586         my $parent = $lcnames{$alias};
4587         my $basename = $parent->{name};
4588         while (1)
4589         {
4590             @{$parent->{sortnames}}[($loc->{lcid} >> 16) - 1] = $loc->{name};
4591             $alias = locale_parent( $alias );
4592             last unless $alias && defined $lcnames{$alias};
4593             $parent = $lcnames{$alias};
4594             last if defined $parent->{sortbase} && $parent->{sortbase} ne $basename;
4595             $parent->{sortbase} = $basename;
4596         }
4597     }
4598
4599     # assign an array index to all locales
4600
4601     my $idx = 0;
4602     foreach my $loc (@locales)
4603     {
4604         next if defined $loc->{alias};
4605         $loc->{idx} = $idx++;
4606     }
4607     foreach my $loc (@locales)
4608     {
4609         my $alias = $loc->{alias};
4610         next unless defined $alias;
4611         while (defined $lcnames{$alias}->{alias}) { $alias = $lcnames{$alias}->{alias}; }
4612         $loc->{idx} = $lcnames{$alias}->{idx};
4613     }
4614
4615     # output lcids table
4616
4617     my $lcid_data = "";
4618     foreach my $id (sort { $a <=> $b } keys %lcids)
4619     {
4620         my $loc = $lcids{$id};
4621         $lcid_data .= pack "L<S<2", $id, $loc->{idx}, add_string($loc->{name});
4622     }
4623
4624     # output lcnames table
4625
4626     my $lcname_data = "";
4627     foreach my $name (sort compare_locales keys %lcnames)
4628     {
4629         my $loc = $lcnames{$name};
4630         $lcname_data .= pack "S<2L<", add_string($name), $loc->{idx}, $loc->{lcid} || 0x1000;
4631     }
4632
4633     # output locales array
4634
4635     my $locale_data = "";
4636     my $default_lcid = 0x8001;
4637     foreach my $loc (@locales)
4638     {
4639         next if defined $loc->{alias};
4640         my $sname = $loc->{name};
4641         my $language = $loc->{language};
4642         my $territory = $loc->{territory};
4643         my $script = $loc->{script};
4644         my $neutral = ($sname && $sname !~ /-$territory/);
4645         my $sparent = $loc->{sparent} || (($sname =~ /(.*)-[0-9A-Za-z]+/) ? $1 : $loc->{parent});
4646         my $unique_lcid = $loc->{lcid};
4647         unless (defined $unique_lcid) { $unique_lcid = $default_lcid++; }
4648         my $geo = $geotable{$territory};
4649         my $territory_match = "contains(concat(' ',normalize-space(\@territories),' '),' $territory ')";
4650
4651         # languages and scripts
4652
4653         my $ssortlocale = $loc->{sortlocale} || ($neutral ? "$sname-$territory" : $sname);
4654         my $idefaultlanguage = defined $lcnames{$ssortlocale} ? $lcnames{$ssortlocale}->{lcid} : undef;
4655         $idefaultlanguage = $lcnames{"en-US"}->{lcid} unless $ssortlocale;
4656         (my $siso639langname = $sname) =~ s/-.*$//;
4657         my $siso639langname2 = $iso639{$siso639langname} || $siso639langname;
4658         my $sopentypelang = sprintf "%-4s", locale_entry( $loc, "sopentypelang", uc $siso639langname2 );
4659         my $sabbrevlangname = defined $loc->{lcid} ? locale_entry( $loc, "sabbrevlangname", uc $siso639langname2 ) : "ZZZ";
4660         my $siso3166ctryname2 = $geo->{iso3} || $geo->{uncode};
4661         my $senglanguage = loc_query( $lcnames{en}, "/ldml/localeDisplayNames/languages/language[\@type='$language' and not(\@alt)]" ) || "";
4662         my $sengcountry = loc_query( $lcnames{en}, "/ldml/localeDisplayNames/territories/territory[\@type='$territory' and not(\@alt)]" ) || "";
4663         my $snativelangname = loc_query( $loc, "/ldml/localeDisplayNames/languages/language[\@type='$language' and not(\@alt)]" );
4664         my $snativectryname = loc_query( $loc, "/ldml/localeDisplayNames/territories/territory[\@type='$territory' and not(\@alt)]" );
4665         $sengcountry =~ s/South Korea/Korea/;
4666         $snativelangname ||= $senglanguage;
4667         $snativectryname ||= $sengcountry;
4668         if ($script)
4669         {
4670             my $engscript = loc_query( $lcnames{en}, "/ldml/localeDisplayNames/scripts/script[\@type='$script' and not(\@alt)]" );
4671             my $nativescript = loc_query( $loc, "/ldml/localeDisplayNames/scripts/script[\@type='$script' and not(\@alt)]" );
4672             $senglanguage .= " ($engscript)" if $engscript;
4673             $snativelangname .= " ($nativescript)" if $nativescript;
4674         }
4675         my $sengdisplayname = $neutral ? $senglanguage : "$senglanguage ($sengcountry)";
4676         my $snativedisplayname = $neutral ? $snativelangname : "$snativelangname ($snativectryname)";
4677         $sengdisplayname =~ s/\) \(/, /;
4678         $snativedisplayname =~ s/\) \(/, /;
4679         my $sscripts = locale_entry( $loc, "sscripts", $script ) || xml_query( $suppl, "/supplementalData/languageData/language[\@type='$language' and not(\@alt)]/\@scripts" );
4680         $sscripts = (join ";", (sort split / /, ($sscripts || "Latn"))) . ";";
4681         my $ireadinglayout = locale_entry( $loc, "ireadinglayout", 0 );
4682         my $charlayout = loc_query( $loc, "/ldml/layout/orientation/characterOrder" );
4683         if ($charlayout eq "right-to-left")
4684         {
4685             $ireadinglayout = 1;
4686         }
4687         elsif ($charlayout eq "top-to-bottom")
4688         {
4689             my $linelayout = loc_query( $loc, "/ldml/layout/orientation/lineOrder" );
4690             $ireadinglayout = $linelayout eq "right-to-left" ? 2 : 3;
4691         }
4692         my $igeoid = $geo->{id} || 0;
4693
4694         # numbers
4695
4696         my $sdecimal = loc_query( $loc, "/ldml/numbers/symbols[\@numberSystem='latn']/decimal" );
4697         my $slist = locale_entry( $loc, "slist", ";" );
4698         my $smondecimalsep = loc_query( $loc, "/ldml/numbers/symbols[\@numberSystem='latn']/currencyDecimal" ) || $sdecimal;
4699         my $sthousand = loc_query( $loc, "/ldml/numbers/symbols[\@numberSystem='latn']/group" );
4700         $sthousand =~ s/\x{202f}/\x{00a0}/;
4701         my $smonthousandsep = loc_query( $loc, "/ldml/numbers/symbols[\@numberSystem='latn']/currencyGroup" ) || $sthousand;
4702         my $spositivesign = "";
4703         my $snegativesign = "-";
4704         my $spercent = loc_query( $loc, "/ldml/numbers/symbols[\@numberSystem='latn']/percentSign" );
4705         my $snan = loc_query( $loc, "/ldml/numbers/symbols[\@numberSystem='latn']/nan" );
4706         my $sposinfinity = loc_query( $loc, "/ldml/numbers/symbols[\@numberSystem='latn']/infinity" );
4707         my $sneginfinity = $sposinfinity ? "-$sposinfinity" : "";
4708         my $sgrouping = format_to_grouping( loc_query( $loc, "/ldml/numbers/decimalFormats[\@numberSystem='latn']/decimalFormatLength[not(\@type)]/decimalFormat/pattern" ));
4709         my $percentformat = loc_query( $loc, "/ldml/numbers/percentFormats[\@numberSystem='latn']/percentFormatLength[not(\@type)]/percentFormat/pattern" );
4710         my $currencyformat = loc_query( $loc, "/ldml/numbers/currencyFormats[\@numberSystem='latn']/currencyFormatLength[not(\@type)]/currencyFormat[\@type='accounting']/pattern" ) ||
4711                              loc_query( $loc, "/ldml/numbers/currencyFormats[\@numberSystem='latn']/currencyFormatLength[not(\@type)]/currencyFormat[\@type='standard']/pattern" );
4712         my $smongrouping = format_to_grouping( $currencyformat );
4713         my ($icurrency, $inegcurr) = parse_currency_format( $sname, $currencyformat );
4714         my ($ipospercent, $inegpercent) = parse_percent_format( $percentformat );
4715         my $native_numbering = loc_query( $loc, "/ldml/numbers/otherNumberingSystems/native" );
4716         my @snativedigits = split //, xml_query( $numbers, "/supplementalData/numberingSystems/numberingSystem[\@id='$native_numbering']/\@digits" );
4717         my $digitsubstitution = !(ord($snativedigits[0]) >= 0x600 && ord($snativedigits[0]) <= 0x6ff);
4718         my $measure = defined xml_query( $suppl, "/supplementalData/measurementData/measurementSystem[\@type='US' and $territory_match]" );
4719         my $papersize = defined xml_query( $suppl, "/supplementalData/measurementData/paperSize[\@type='US-Letter' and $territory_match]" );
4720
4721         # currencies
4722
4723         my $sintlsymbol = $geo->{sintlsymbol} || "XDR";
4724         my $scurrency = $geo->{scurrency} || loc_query( $loc, "/ldml/numbers/currencies/currency[\@type='$sintlsymbol']/symbol[\@alt='narrow']" );
4725         $scurrency ||= loc_query( $loc, "/ldml/numbers/currencies/currency[\@type='$sintlsymbol']/symbol[not(\@alt)]" );
4726         $geo->{scurrency} = $scurrency if $scurrency;
4727         $scurrency ||= $sintlsymbol;
4728         my $sengcurrname = $loc->{sengcurrname} || loc_query( $lcnames{en}, "/ldml/numbers/currencies/currency[\@type='$sintlsymbol']/displayName[not(\@count)]" );
4729         my $snativecurrname = $loc->{sengcurrname} || loc_query( $loc, "/ldml/numbers/currencies/currency[\@type='$sintlsymbol']/displayName[not(\@count)]" ) || $sengcurrname;
4730         my $icurrdigits = xml_query( $suppl, "/supplementalData/currencyData/fractions/info[\@iso4217='$sintlsymbol']/\@digits" );
4731         $icurrdigits = 2 unless defined $icurrdigits;
4732
4733         # calendars
4734
4735         my $firstday = xml_query( $suppl, "/supplementalData/weekData/firstDay[not(\@alt) and $territory_match]/\@day" );
4736         my $ifirstdayofweek = $firstday ? $days{$firstday} : 1;
4737         my $firstweekofyear = (xml_query( $suppl, "/supplementalData/weekData/minDays[$territory_match]/\@count" ) || 0) == 4 ? 2 : 0;
4738         my $serastring = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/eras/eraAbbr/era[\@type='1' and not(\@alt)]" );
4739         my (@sdayname, @sabbrevdayname, @sshortestdayname);
4740         foreach my $d (sort { $days{$a} <=> $days{$b} } keys %days)
4741         {
4742             my $n = $days{$d};
4743             my %name;
4744             foreach my $type (qw(wide abbreviated short))
4745             {
4746                 $name{$type} = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/days/dayContext[\@type='format']/dayWidth[\@type='$type']/day[\@type='$d' and not(\@alt)]" );
4747             }
4748             push @sdayname, $name{wide};
4749             push @sabbrevdayname, $name{abbreviated} || $name{wide};
4750             push @sshortestdayname, $name{short} || $name{abbreviated} || $name{wide};
4751         }
4752         my (@smonthname, @sabbrevmonthname, @sgenitivemonth, @sabbrevgenitivemonth);
4753         foreach my $n (1..13)
4754         {
4755             my $name = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/months/monthContext[\@type='stand-alone']/monthWidth[\@type='wide']/month[\@type='$n']" );
4756             my $abbrev = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/months/monthContext[\@type='stand-alone']/monthWidth[\@type='abbreviated']/month[\@type='$n']" );
4757             my $genitive = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/months/monthContext[\@type='format']/monthWidth[\@type='wide']/month[\@type='$n']" );
4758             my $abbrevgen = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/months/monthContext[\@type='format']/monthWidth[\@type='abbreviated']/month[\@type='$n']" );
4759             push @smonthname, $name || $genitive || "";
4760             push @sabbrevmonthname, $abbrev || $abbrevgen || $name || $genitive || "";
4761             push @sgenitivemonth, $genitive || "";
4762             push @sabbrevgenitivemonth, $abbrevgen || $genitive || "";
4763         }
4764         @sgenitivemonth = () if join("|",@smonthname) eq join("|",@sgenitivemonth);
4765         @sabbrevgenitivemonth = () if join("|",@sabbrevmonthname) eq join("|",@sabbrevgenitivemonth);
4766         my %caltypes = ( "gregorian" => 1, "japanese" => 3, "chinese" => 4, "dangi" => 5, "islamic" => 6, "buddhist" => 7, "hebrew" => 8,
4767                          "persian" => 22, "islamic-civil" => 23, "islamic-umalqura" => 23 );
4768         my $calpref = xml_query( $suppl, "/supplementalData/calendarPreferenceData/calendarPreference[$territory_match]/\@ordering" ) || "gregorian";
4769         my $icalendartype;
4770         my @scalnames;
4771         foreach my $c (split /\s+/, $calpref)
4772         {
4773             next unless defined $caltypes{$c};
4774             $icalendartype .= chr($caltypes{$c});
4775             $scalnames[$caltypes{$c} - 1] = loc_query( $loc, "/ldml/localeDisplayNames/types/type[\@key='calendar' and \@type='$c']" );
4776         }
4777
4778         # date/time formats
4779
4780         my $s1159 = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dayPeriods/dayPeriodContext[\@type='format']/dayPeriodWidth[\@type='abbreviated']/dayPeriod[\@type='am' and not(\@alt)]" );
4781         my $s2359 = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dayPeriods/dayPeriodContext[\@type='format']/dayPeriodWidth[\@type='abbreviated']/dayPeriod[\@type='pm' and not (\@alt)]" );
4782         my $sshortestam = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dayPeriods/dayPeriodContext[\@type='format']/dayPeriodWidth[\@type='narrow']/dayPeriod[\@type='am' and not(\@alt)]" );
4783         my $sshortestpm = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dayPeriods/dayPeriodContext[\@type='format']/dayPeriodWidth[\@type='narrow']/dayPeriod[\@type='pm' and not (\@alt)]" );
4784         my @stimeformat = (loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/timeFormats/timeFormatLength[\@type='medium']/timeFormat/pattern[not(\@alt)]" ));
4785         push @stimeformat, loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dateTimeFormats/availableFormats/dateFormatItem[\@id='Hms' and not(\@alt)]" );
4786         pop @stimeformat if $stimeformat[0] eq $stimeformat[1];
4787         @stimeformat = map convert_time_format($_), @stimeformat;
4788         my @sshorttime = (loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/timeFormats/timeFormatLength[\@type='short']/timeFormat/pattern[not(\@alt)]" ));
4789         push @sshorttime, loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dateTimeFormats/availableFormats/dateFormatItem[\@id='Hm' and not(\@alt)]" );
4790         pop @sshorttime if $sshorttime[0] eq $sshorttime[1];
4791         @sshorttime = map convert_time_format($_), @sshorttime;
4792         my @sshortdate = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dateTimeFormats/availableFormats/dateFormatItem[\@id='yMd' and not(\@alt)]" );
4793         push @sshortdate, loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dateTimeFormats/availableFormats/dateFormatItem[\@id='yMMMd' and not(\@alt)]" );
4794         @sshortdate = map convert_date_format($_), @sshortdate;
4795         my @slongdate = (loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dateFormats/dateFormatLength[\@type='full']/dateFormat/pattern[not(\@alt)]" ));
4796         push @slongdate, loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dateFormats/dateFormatLength[\@type='long']/dateFormat/pattern[not(\@alt)]" );
4797         @slongdate = map convert_date_format($_), @slongdate;
4798         my @smonthday = (loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dateTimeFormats/availableFormats/dateFormatItem[\@id='MMMMd' and not(\@alt)]" ));
4799         push @smonthday, loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dateTimeFormats/availableFormats/dateFormatItem[\@id='Md' and not(\@alt)]" );
4800         push @smonthday, loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dateTimeFormats/availableFormats/dateFormatItem[\@id='MMMd' and not(\@alt)]" );
4801         @smonthday = map convert_date_format($_), @smonthday;
4802         my @syearmonth = map convert_date_format($_), loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dateTimeFormats/availableFormats/dateFormatItem[\@id='yMMMM' and not(\@alt)]" );
4803         my @sduration = map convert_time_format( lc $_ ), loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dateTimeFormats/availableFormats/dateFormatItem[\@id='Hms' and not(\@alt)]" );
4804         my $srelativelongdate = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dateTimeFormats/availableFormats/dateFormatItem[\@id='MMMMEd' and not(\@alt)]" ) ||
4805                                 loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='gregorian']/dateTimeFormats/availableFormats/dateFormatItem[\@id='MMMEd' and not(\@alt)]" );
4806         $srelativelongdate = convert_date_format( $srelativelongdate );
4807
4808         if (defined $loc->{calendar})
4809         {
4810             foreach my $cal (@{$loc->{calendar}})
4811             {
4812                 $cal->{sshortdate} = \@sshortdate;
4813                 $cal->{syearmonth} = \@syearmonth;
4814                 $cal->{slongdate} = \@slongdate;
4815                 $cal->{serastring} = [ $serastring ];
4816                 $cal->{sdayname} = \@sdayname;
4817                 $cal->{sabbrevdayname} = \@sabbrevdayname;
4818                 $cal->{smonthname} = \@smonthname;
4819                 $cal->{sabbrevmonthname} = \@sabbrevmonthname;
4820                 $cal->{scalname} = $scalnames[$cal->{id}];
4821                 $cal->{smonthday} = \@smonthday;
4822                 $cal->{sshortestdayname} = \@sshortestdayname;
4823                 $cal->{sabbreverastring} = [ $serastring ];
4824                 $cal->{sshortestdayname} = \@sshortestdayname;
4825                 $cal->{srelativelongdate} = $srelativelongdate;
4826             }
4827         }
4828
4829         # codepages
4830
4831         my %ansicpmap = ( 437 => 1252, 720 => 1256, 737 => 1253, 775 => 1257, 850 => 1252,
4832                           852 => 1250, 855 => 1251, 866 => 1251, 857 => 1254, 862 => 1255 );
4833         my %maccpmap = ( 437 => 10000, 720 => 10004, 737 => 10006, 775 => 10029, 850 => 10000,
4834                          852 => 10029, 855 => 10007, 857 => 10081, 862 => 10005, 866 => 10007,
4835                          874 => 10021, 932 => 10001, 936 => 10008, 949 => 10003, 950 => 10002,
4836                          1258 => 10000 );
4837         my %ebcdiccpmap = ( 437 => 37, 720 => 20420, 737 => 20273, 866 => 20880, 932 => 20290 );
4838         my %codepagemasks = ( 874 =>   [ 0x01000000, 0x00000000, 0x00000000, 0, 0x00010000, 0x00000000, 0x00010000, 0x00000000 ],
4839                               932 =>   [ 0x00000000, 0x28c70000, 0x00000010, 0, 0x00020000, 0x00000000, 0x00020000, 0x00000000 ],
4840                               936 =>   [ 0x00000000, 0x28010000, 0x00000002, 0, 0x00040000, 0x00000000, 0x00040000, 0x00000000 ],
4841                               949 =>   [ 0x00000000, 0x00000000, 0x00000000, 0, 0x00080000, 0x00000000, 0x00080000, 0x00000000 ],
4842                               950 =>   [ 0x00000000, 0x28c10000, 0x00000012, 0, 0x00100000, 0x00000000, 0x00100000, 0x00000000 ],
4843                               1258 =>  [ 0x2000000f, 0x00000000, 0x00000000, 0, 0x00000100, 0x00008000, 0x00000100, 0x00008000 ],
4844                               866 =>   [ 0x00000200, 0x00000000, 0x00000000, 0, 0x00000004, 0x00020000, 0x00000004, 0x02020000 ],
4845                               862 =>   [ 0x00000800, 0x40000000, 0x00000000, 0, 0x00000020, 0x00200000, 0x00000020, 0x00200000 ],
4846                               857 =>   [ 0x0000001f, 0x00000000, 0x00000000, 0, 0x00000010, 0x01000000, 0x00000010, 0x01000000 ],
4847                               855 =>   [ 0x00000200, 0x00000000, 0x00000000, 0, 0x00000004, 0x02000000, 0x00000004, 0x02000000 ],
4848                               852 =>   [ 0x00000027, 0x00000000, 0x00000000, 0, 0x00000002, 0x04000000, 0x00000002, 0x04000000 ],
4849                               775 =>   [ 0x00000007, 0x00000000, 0x00000000, 0, 0x00000080, 0x08000000, 0x00000080, 0x08000000 ],
4850                               737 =>   [ 0x00000080, 0x00000000, 0x00000000, 0, 0x00000008, 0x10000000, 0x00000008, 0x10010000 ],
4851                               720 =>   [ 0x00002000, 0x00000000, 0x00000000, 0, 0x00000040, 0x20000000, 0x00000040, 0x20080000 ],
4852                               850 =>   [ 0x00000003, 0x00000000, 0x00000000, 0, 0x00000001, 0x40000000, 0x0000019f, 0xdfd70000 ],
4853                               437 =>   [ 0x00000003, 0x00000000, 0x00000000, 0, 0x00000001, 0x80000000, 0x0000019f, 0xdfd70000 ],
4854                               65001 => [ 0x00000000, 0x00000000, 0x00000000, 0, 0x00000000, 0x00000000, 0x0000019f, 0xdfd70000 ] );
4855         my $oemcp = locale_entry( $loc, "oemcp", 65001 );
4856         my $maccp = locale_entry( $loc, "maccp", undef ) || $maccpmap{$oemcp} || 65001;
4857         my $ebcdiccp = locale_entry( $loc, "ebcdiccp", undef ) || $ebcdiccpmap{$oemcp} || 500;
4858         $ebcdiccp = 500 if (defined $loc->{oemcp} && $loc->{oemcp} == 65001) || (defined $loc->{maccp} && $loc->{maccp} == 65001);
4859         my $ansicp = $ansicpmap{$oemcp} || $oemcp;
4860         my @fontsig = (0) x 8;
4861         my $sig = locale_entry( $loc, "fontsig", [] );
4862         foreach my $i (0..7) { $fontsig[$i] |= $codepagemasks{$oemcp}->[$i]; }
4863         foreach my $i (0..$#{$sig}) { $fontsig[$i] |= $sig->[$i]; }
4864         $fontsig[3] |= 1 << 31;
4865         $fontsig[3] |= 1 << 27 if $ireadinglayout == 1;
4866         $fontsig[3] |= 1 << 28 if $ireadinglayout == 3;
4867
4868         # special cases for invariant locale
4869
4870         unless ($loc->{name})
4871         {
4872             $siso639langname = "iv";
4873             $siso639langname2 = "ivl";
4874             $senglanguage = $snativelangname = "Invariant Language";
4875             $sengcountry = $snativectryname = "Invariant Country";
4876             $sengdisplayname = "Invariant Language (Invariant Country)";
4877             $snativedisplayname = "Invariant Language (Invariant Region)";
4878             $sengcurrname = $snativecurrname = "International Monetary Fund";
4879             $scurrency = "\x{00a4}";
4880             $ifirstdayofweek = 0;
4881             $igeoid = $geotable{"US"}->{id};
4882             @stimeformat = ("HH:mm:ss");
4883             @sshortdate = ("MM/dd/yyyy", "yyyy-MM-dd");
4884             @slongdate = ("dddd, dd MMMM yyyy");
4885             @syearmonth = ("yyyy MMMM");
4886             @smonthday = ("MMMM dd", "MMMM d", "M/d", "MMM d");
4887             @sshorttime = ("HH:mm", "hh:mm tt", "H:mm", "h:mm tt");
4888             $srelativelongdate = "dddd, MMMM dd";
4889             $sposinfinity = "Infinity";
4890             $sneginfinity = "-Infinity";
4891             $spositivesign = "+";
4892             $ipospercent = $inegpercent = 0;
4893         }
4894
4895         # output data
4896
4897         $locale_data .= pack "L<2",
4898             add_string( $sname ),                  # name
4899             add_string( $sopentypelang );          # LOCALE_SOPENTYPELANGUAGETAG
4900
4901         $locale_data .= pack "S<14",
4902             $loc->{lcid} || 0x1000,                # LOCALE_ILANGUAGE
4903             $unique_lcid,                          # unique_lcid
4904             locale_entry( $loc, "idigits", 2 ),    # LOCALE_IDIGITS
4905             locale_entry( $loc, "inegnumber", 1 ), # LOCALE_INEGNUMBER
4906             $icurrdigits,                          # LOCALE_ICURRDIGITS
4907             $icurrency,                            # LOCALE_ICURRENCY
4908             $inegcurr,                             # LOCALE_INEGCURR
4909             locale_entry( $loc, "ilzero", 1 ),     # LOCALE_ILZERO
4910             !$neutral,                             # LOCALE_INEUTRAL
4911             $ifirstdayofweek,                      # LOCALE_IFIRSTDAYOFWEEK
4912             $firstweekofyear,                      # LOCALE_IFIRSTWEEKOFYEAR
4913             $geo->{dialcode} || 1 ,                # LOCALE_ICOUNTRY,
4914             $measure,                              # LOCALE_IMEASURE
4915             $digitsubstitution;                    # LOCALE_IDIGITSUBSTITUTION
4916
4917         $locale_data .= pack "L<18",
4918             add_string( $sgrouping ),              # LOCALE_SGROUPING
4919             add_string( $smongrouping ),           # LOCALE_SMONGROUPING
4920             add_string( $slist ),                  # LOCALE_SLIST
4921             add_string( $sdecimal ),               # LOCALE_SDECIMAL
4922             add_string( $sthousand ),              # LOCALE_STHOUSAND
4923             add_string( $scurrency ),              # LOCALE_SCURRENCY
4924             add_string( $smondecimalsep ),         # LOCALE_SMONDECIMALSEP
4925             add_string( $smonthousandsep ),        # LOCALE_SMONTHOUSANDSEP
4926             add_string( $spositivesign ),          # LOCALE_SPOSITIVESIGN
4927             add_string( $snegativesign ),          # LOCALE_SNEGATIVESIGN
4928             add_string( $s1159 ),                  # LOCALE_S1159
4929             add_string( $s2359 ),                  # LOCALE_S2359
4930             add_strarray( @snativedigits ),        # LOCALE_SNATIVEDIGITS
4931             add_strarray( @stimeformat ),          # LOCALE_STIMEFORMAT
4932             add_strarray( @sshortdate ),           # LOCALE_SSHORTDATE
4933             add_strarray( @slongdate ),            # LOCALE_SLONGDATE
4934             add_strarray( @syearmonth ),           # LOCALE_SYEARMONTH
4935             add_strarray( @sduration );            # LOCALE_SDURATION
4936
4937         $locale_data .= pack "S<8",
4938             $idefaultlanguage || 0x1000,           # LOCALE_IDEFAULTLANGUAGE
4939             $ansicp,                               # LOCALE_IDEFAULTANSICODEPAGE
4940             $oemcp,                                # LOCALE_IDEFAULTCODEPAGE
4941             $maccp,                                # LOCALE_IDEFAULTMACCODEPAGE
4942             $ebcdiccp,                             # LOCALE_IDEFAULTEBCDICCODEPAGE
4943             $igeoid < 65536 ? $igeoid : 39070,     # old_geoid
4944             $papersize ? 1 : 9,                    # LOCALE_IPAPERSIZE
4945             0;  # FIXME                            # islamic_cal
4946
4947         $locale_data .= pack "L<24",
4948             add_string( $icalendartype ),          # LOCALE_ICALENDARTYPE
4949             add_string( $sabbrevlangname ),        # LOCALE_SABBREVLANGNAME
4950             add_string( $siso639langname ),        # LOCALE_SISO639LANGNAME
4951             add_string( $senglanguage ),           # LOCALE_SENGLANGUAGE
4952             add_string( $snativelangname ),        # LOCALE_SNATIVELANGNAME
4953             add_string( $sengcountry ),            # LOCALE_SENGCOUNTRY
4954             add_string( $snativectryname ),        # LOCALE_SNATIVECTRYNAME
4955             add_string( $siso3166ctryname2 ),      # LOCALE_SABBREVCTRYNAME
4956             add_string( $territory ),              # LOCALE_SISO3166CTRYNAME
4957             add_string( $sintlsymbol ),            # LOCALE_SINTLSYMBOL
4958             add_string( $sengcurrname ),           # LOCALE_SENGCURRNAME
4959             add_string( $snativecurrname ),        # LOCALE_SNATIVECURRNAME
4960             add_fontsig( @fontsig ),               # LOCALE_FONTSIGNATURE
4961             add_string( $siso639langname2 ),       # LOCALE_SISO639LANGNAME2
4962             add_string( $siso3166ctryname2 ),      # LOCALE_SISO3166CTRYNAME2
4963             add_string( $sparent ),                # LOCALE_SPARENT
4964             add_strarray( @sdayname ),             # LOCALE_SDAYNAME
4965             add_strarray( @sabbrevdayname ),       # LOCALE_SABBREVDAYNAME
4966             add_strarray( @smonthname ),           # LOCALE_SMONTHNAME
4967             add_strarray( @sabbrevmonthname ),     # LOCALE_SABBREVMONTHNAME
4968             add_strarray( @sgenitivemonth ),       # LOCALE_SGENITIVEMONTH
4969             add_strarray( @sabbrevgenitivemonth ), # LOCALE_SABBREVGENITIVEMONTH
4970             add_strarray( @scalnames ),            # LOCALE_SCALNAMES
4971             add_strarray( @{$loc->{sortnames}} );  # LOCALE_SSORTNAMES
4972
4973         $locale_data .= pack "S<6",
4974             $inegpercent,                          # LOCALE_INEGATIVEPERCENT
4975             $ipospercent,                          # LOCALE_IPOSITIVEPERCENT
4976             0,                                     # unknown
4977             $ireadinglayout,                       # LOCALE_IREADINGLAYOUT
4978             0x2a,                                  # unknown
4979             0x2a;                                  # unknown
4980
4981         $locale_data .= pack "L<24",
4982             0,                                     # unknown
4983             add_string( $sengdisplayname ),        # LOCALE_SENGLISHDISPLAYNAME
4984             add_string( $snativedisplayname ),     # LOCALE_SNATIVEDISPLAYNAME
4985             add_string( $spercent ),               # LOCALE_SPERCENT
4986             add_string( $snan ),                   # LOCALE_SNAN
4987             add_string( $sposinfinity ),           # LOCALE_SPOSINFINITY
4988             add_string( $sneginfinity ),           # LOCALE_SNEGINFINITY
4989             0,                                     # unknown
4990             add_string( $serastring ),             # CAL_SERASTRING
4991             add_string( $serastring ),             # CAL_SABBREVERASTRING
4992             0,                                     # unknown
4993             add_string( $ssortlocale ),            # LOCALE_SCONSOLEFALLBACKNAME
4994             add_strarray( @sshorttime ),           # LOCALE_SSHORTTIME
4995             add_strarray( @sshortestdayname ),     # CAL_SSHORTESTDAYNAME
4996             0,                                     # unknown
4997             add_string( $ssortlocale ),            # LOCALE_SSORTLOCALE
4998             add_string( "0409:00000409" ), # FIXME # LOCALE_SKEYBOARDSTOINSTALL
4999             add_string( $sscripts ),               # LOCALE_SSCRIPTS
5000             add_string( $srelativelongdate ),      # LOCALE_SRELATIVELONGDATE
5001             $igeoid,                               # LOCALE_IGEOID
5002             add_string( $sshortestam || "a" ),     # LOCALE_SSHORTESTAM
5003             add_string( $sshortestpm || "p" ),     # LOCALE_SSHORTESTPM
5004             add_strarray( @smonthday ),            # LOCALE_SMONTHDAY
5005             add_string( "k0-windows-us" ) # FIXME  # keyboard_layout
5006     }
5007
5008     # output language groups
5009
5010     my %groups;
5011     add_registry_key( $nlskey, "Locale", "00000409" );
5012     foreach my $loc (@locales)
5013     {
5014         next unless defined $loc->{lcid};
5015         next if ($loc->{lcid} & 0x80000000);
5016         next if !defined($loc->{alias}) && $loc->{name} !~ /-$loc->{territory}/; # skip neutral locales
5017         my $group = locale_entry( $loc, "group", 1 );
5018         my $name = sprintf( "%08x", $loc->{lcid} );
5019         my $val = sprintf( "%x", $group );
5020         add_registry_string_value( $nlskey, "Locale", $name, $val ) unless ($loc->{lcid} & 0x000f0000);
5021         add_registry_string_value( $nlskey, "Locale\\Alternate Sorts", $name, $val ) if $loc->{name} =~ /_/;
5022         $groups{$val} = 1;
5023     }
5024     foreach my $group (keys %groups) { add_registry_string_value( $nlskey, "Language Groups", $group, "1" ); }
5025
5026     # output calendar data
5027
5028     my $calendar_data = "";
5029     foreach my $cal (@calendars)
5030     {
5031         my $scalname = $cal->{name};
5032         my $iyearoffsetrange = 0;
5033         my $itwodigityearmax = $cal->{itwodigityearmax};
5034         my @sshortdate;
5035         my @syearmonth;
5036         my @slongdate;
5037         my @serastring;
5038         my @sdayname;
5039         my @sabbrevdayname;
5040         my @smonthname;
5041         my @sabbrevmonthname;
5042         my @smonthday;
5043         my @sabbreverastring;
5044         my @sshortestdayname;
5045
5046         my $type = $cal->{type};
5047         if (defined $cal->{locale} && defined $type)
5048         {
5049             my $loc = $lcnames{$cal->{locale}};
5050             my $fmt = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='$type']/dateTimeFormats/availableFormats/dateFormatItem[\@id='yMd' and not(\@alt)]" );
5051             push @sshortdate, $fmt if $fmt;
5052             $fmt = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='$type']/dateTimeFormats/availableFormats/dateFormatItem[\@id='yyyyMd' and not(\@alt)]" );
5053             push @sshortdate, $fmt if $fmt;
5054             $fmt = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='$type']/dateTimeFormats/availableFormats/dateFormatItem[\@id='yMMMd' and not(\@alt)]" );
5055             push @sshortdate, $fmt if $fmt;
5056             $fmt = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='$type']/dateTimeFormats/availableFormats/dateFormatItem[\@id='yyyyMMMd' and not(\@alt)]" );
5057             push @sshortdate, $fmt if $fmt;
5058             @sshortdate = map convert_date_format($_), @sshortdate;
5059             $fmt = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='$type']/dateFormats/dateFormatLength[\@type='full']/dateFormat/pattern[not(\@alt)]" );
5060             push @slongdate, $fmt if $fmt;
5061             $fmt = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='$type']/dateFormats/dateFormatLength[\@type='long']/dateFormat/pattern[not(\@alt)]" );
5062             push @slongdate, $fmt if $fmt;
5063             @slongdate = map convert_date_format($_), @slongdate;
5064
5065             foreach my $n (1..13)
5066             {
5067                 my $name = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='$type']/months/monthContext[\@type='format']/monthWidth[\@type='wide']/month[\@type='$n' and not(\@yeartype)]" );
5068                 my $abbrev = loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='$type']/months/monthContext[\@type='format']/monthWidth[\@type='abbreviated']/month[\@type='$n' and not(\@yeartype)]" );
5069                 push @smonthname, $name || "";
5070                 push @sabbrevmonthname, $abbrev || $name || "";
5071             }
5072
5073             $scalname ||= loc_query( $loc, "/ldml/localeDisplayNames/types/type[\@key='calendar' and \@type='$type']" );
5074             if (defined $cal->{eras})
5075             {
5076                 my @eras;
5077                 my $idx = 1;
5078                 foreach my $era (@{$cal->{eras}})
5079                 {
5080                     my $start = xml_query( $suppl, "/supplementalData/calendarData/calendar[\@type='$type']/eras/era[\@type='$era']/\@start" );
5081                     next unless $start =~ /^(-?\d+)-(\d+)-(\d+)/;
5082                     my ($year, $mon, $day, $zero, $first) = ($1, $2, $3, $1 - 1, 1);
5083                     if ($zero < 0)
5084                     {
5085                         $first -= $zero;
5086                         $year = 1;
5087                         $itwodigityearmax = 2049 - $zero;
5088                     }
5089                     unshift @eras, pack( "S<8", 6, $idx++, $year, $mon, $day, $zero, $first, 0 );
5090                     push @serastring, loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='$type']/eras/eraAbbr/era[\@type='$era']" );
5091                     push @sabbreverastring, loc_query( $loc, "/ldml/dates/calendars/calendar[\@type='$type']/eras/eraNarrow/era[\@type='$era']" );
5092                 }
5093                 $iyearoffsetrange = add_str_data( pack "S<L<*", scalar @eras, map { add_str_data($_); } @eras );
5094             }
5095         }
5096
5097         @sshortdate = @{$cal->{sshortdate}} if defined $cal->{sshortdate} && !@sshortdate;
5098         @syearmonth = @{$cal->{syearmonth}} if defined $cal->{syearmonth};
5099         @slongdate = @{$cal->{slongdate}} if defined $cal->{slongdate} && !@slongdate;
5100         @serastring = @{$cal->{serastring}} if defined $cal->{serastring} && !@serastring;
5101         @sdayname = @{$cal->{sdayname}} if defined $cal->{sdayname};
5102         @sabbrevdayname = @{$cal->{sabbrevdayname}} if defined $cal->{sabbrevdayname};
5103         @smonthname = @{$cal->{smonthname}} if defined $cal->{smonthname} && !join("",@smonthname);
5104         @sabbrevmonthname = @{$cal->{sabbrevmonthname}} if defined $cal->{sabbrevmonthname} && !join("",@sabbrevmonthname);
5105         @smonthday = @{$cal->{smonthday}} if defined $cal->{smonthday};
5106         @sabbreverastring = @{$cal->{sabbreverastring}} if defined $cal->{sabbreverastring} && !@sabbreverastring;
5107         @sshortestdayname = @{$cal->{sshortestdayname}} if defined $cal->{sshortestdayname};
5108         my $srelativelongdate = $cal->{srelativelongdate};
5109
5110         @serastring = ("A.D.") unless @serastring;
5111         @sabbreverastring = ("AD") unless @sabbreverastring;
5112
5113         if ($cal->{id} != 1)  # calendar 1 is a placeholder, information is fetched from locale instead
5114         {
5115             @sshortdate = ("") unless @sshortdate;
5116             @syearmonth = ("") unless @syearmonth;
5117             @slongdate = ("") unless @slongdate;
5118             @sdayname = ("") x 7 unless @sdayname;
5119             @sabbrevdayname = ("") x 7 unless @sabbrevdayname;
5120             @sshortestdayname = ("") x 7 unless @sshortestdayname;
5121             @smonthname = ("") x 13 unless @smonthname;
5122             @sabbrevmonthname = ("") x 13 unless @sabbrevmonthname;
5123             @smonthday = ("") unless @smonthday;
5124         }
5125
5126         $calendar_data .= pack "S<2L<17",
5127             $cal->{id},                        # CAL_ICALINTVALUE
5128             $itwodigityearmax || 99,           # CAL_ITWODIGITYEARMAX
5129             add_strarray( @sshortdate ),       # CAL_SSHORTDATE
5130             add_strarray( @syearmonth ),       # CAL_SYEARMONTH
5131             add_strarray( @slongdate ),        # CAL_SLONGDATE
5132             add_strarray( @serastring ),       # CAL_SERASTRING
5133             $iyearoffsetrange,                 # CAL_IYEAROFFSETRANGE
5134             add_strarray( @sdayname ),         # CAL_SDAYNAME
5135             add_strarray( @sabbrevdayname ),   # CAL_SABBREVDAYNAME
5136             add_strarray( @smonthname ),       # CAL_SMONTHNAME
5137             add_strarray( @sabbrevmonthname ), # CAL_SABBREVMONTHNAME
5138             add_string( $scalname ),           # CAL_SCALNAME
5139             add_strarray( @smonthday ),        # CAL_SMONTHDAY
5140             add_strarray( @sabbreverastring ), # CAL_SABBREVERASTRING
5141             add_strarray( @sshortestdayname ), # CAL_SSHORTESTDAYNAME
5142             add_string( $srelativelongdate );  # CAL_SRELATIVELONGDATE
5143     }
5144
5145     # output locale header
5146
5147     my $nb_lcids = scalar keys %lcids;
5148     my $nb_locales = scalar grep { !defined $_->{alias} } @locales;
5149     my $nb_lcnames = scalar keys %lcnames;
5150     my $locale_size = length($locale_data) / $nb_locales;
5151     my $nb_calendars = scalar @calendars;
5152     my $calendar_size = length($calendar_data) / $nb_calendars;
5153     my $lcids_offset = 19 * 4;  # size of header
5154     my $lcnames_offset = $lcids_offset + length $lcid_data;
5155     my $locales_offset = $lcnames_offset + length $lcname_data;
5156     my $calendar_offset = $locales_offset + length $locale_data;
5157     my $strings_offset = $calendar_offset + length $calendar_data;
5158
5159     my $locale_header = pack "L<7S<4L<S<2L<3S<2L<4",
5160         8,  # offset
5161         0,
5162         7,  # version
5163         0x5344534e,  # magic
5164         0, 0, 0,
5165         0,
5166         $nb_lcids,
5167         $nb_locales,
5168         $locale_size,
5169         $locales_offset,
5170         $nb_lcnames,
5171         0,
5172         $lcids_offset,
5173         $lcnames_offset,
5174         0,
5175         $nb_calendars,
5176         $calendar_size,
5177         $calendar_offset,
5178         $strings_offset,
5179         0, 0;
5180
5181     return align_string( 4, $locale_header . $lcid_data . $lcname_data . $locale_data . $calendar_data . $string_data );
5182 }
5183
5184
5185 ################################################################
5186 # build the charmaps table for locale.nls
5187 sub build_charmaps_data()
5188 {
5189     my $data = "";
5190
5191     # MAP_FOLDDIGITS
5192     my @digits = (ord('0') .. ord('9'));
5193     $digitmap_table[0x3007] = $digits[0];               # Ideographic Zero
5194     @digitmap_table[0x0c78..0x0c7b] = @digits[0..3];    # Telugu Fraction Digits
5195     @digitmap_table[0x0c7c..0x0c7e] = @digits[1..3];    # Telugu Fraction Digits
5196     @digitmap_table[0x3021..0x3029] = @digits[1..9];    # Hangzhou Numerals
5197     @digitmap_table[0xa8e0..0xa8e9] = @digits;          # Combining Devanagari Digits
5198     @digitmap_table[0x10107..0x1010f] = @digits[1..9];  # Aegean Numbers
5199     $digitmap_table[0x10320] = $digits[1];              # Old Italic Numerals
5200     $digitmap_table[0x10321] = $digits[5];              # Old Italic Numerals
5201     $data .= dump_binary_case_table( @digitmap_table );
5202
5203     # CJK compatibility map
5204     $data .= dump_binary_case_table( @cjk_compat_table );
5205
5206     # LCMAP_HIRAGANA/KATAKANA
5207     my (@hiragana_table, @katakana_table);
5208     foreach my $ch (0x3041..0x3096, 0x309d..0x309e)
5209     {
5210         $hiragana_table[$ch + 0x60] = $ch;
5211         $katakana_table[$ch] = $ch + 0x60;
5212     }
5213     $data .= dump_binary_case_table( @hiragana_table ) . dump_binary_case_table( @katakana_table );
5214
5215     # LCMAP_HALFWIDTH/FULLWIDTH
5216     $halfwidth_table[0x2018] = 0x0027;
5217     $halfwidth_table[0x2019] = 0x0027;
5218     $halfwidth_table[0x201c] = 0x0022;
5219     $halfwidth_table[0x201d] = 0x0022;
5220     $halfwidth_table[0x309b] = 0xff9e;
5221     $halfwidth_table[0x309c] = 0xff9f;
5222     $fullwidth_table[0x309b] = 0x3099;
5223     $fullwidth_table[0x309c] = 0x309a;
5224     $data .= dump_binary_case_table( @halfwidth_table ) . dump_binary_case_table( @fullwidth_table );
5225
5226     # LCMAP_TRADITIONAL/SIMPLIFIED_CHINESE
5227     $data .= dump_binary_case_table( @chinese_traditional_table ) . dump_binary_case_table( @chinese_simplified_table );
5228
5229     # FIXME: some more unknown tables here
5230
5231     return $data;
5232 }
5233
5234
5235 ################################################################
5236 # build the geoids table for locale.nls
5237 sub build_geoids_data()
5238 {
5239     my $data = "";
5240     my %index;
5241     my $idx = 0;
5242     my @geo_header = (0x00650067, 0x0000006f, 0, 4 * 7, scalar @geoids, 0, 0);
5243
5244     foreach my $geo (@geoids)
5245     {
5246         my $id = $geo->{id};
5247         $geo = $geo->{alias} if defined $geo->{alias};
5248         my $lat = "0.000";
5249         my $long = "0.000";
5250         my $iso2 = $geo->{iso2} || "XX";
5251         my $iso3 = $geo->{iso3} || "XX";
5252         my $isregion = $geo->{region} || (defined $geo->{uncode} && !defined $geo->{iso2});
5253         my $sintlsymbol = $geo->{sintlsymbol} || "XDR";
5254         my $scurrency = $geo->{scurrency} || "\x{00a4}";
5255
5256         $data .= pack( "L<", $id );
5257         $data .= pad_string( 24, encode( "UTF16LE", $lat ));
5258         $data .= pad_string( 24, encode( "UTF16LE", $long ));
5259         $data .= pack( "L<2", $isregion ? 14 : 16, $geo->{parentid} || 39070 );
5260         $data .= pad_string( 8, encode( "UTF16LE", $iso2 ));
5261         $data .= pad_string( 8, encode( "UTF16LE", $iso3 ));
5262         $data .= pack( "S<2", $geo->{uncode} || 0, $geo->{dialcode} || 0 );
5263         $data .= pad_string( 8, encode( "UTF16LE", $sintlsymbol ));
5264         $data .= pad_string( 16, encode( "UTF16LE", $scurrency ));
5265         $index{$geo->{name}} = $idx if $geo->{name};
5266         $idx++;
5267     }
5268     $index{"XX"} = $index{"001"};
5269
5270     $geo_header[5] = $geo_header[3] + length $data;
5271     $geo_header[6] = scalar keys %index;
5272
5273     foreach my $name (sort keys %index)
5274     {
5275         $data .= pad_string( 8, encode( "UTF16LE", $name ));
5276         $data .= pack "L<", $index{$name};
5277     }
5278
5279     $geo_header[2] = $geo_header[3] + length $data;
5280     return pack( "L<7", @geo_header ) . $data;
5281 }
5282
5283
5284 ################################################################
5285 # build a binary locale table
5286 sub dump_locales($$)
5287 {
5288     my ($filename, $chartypes) = @_;
5289
5290     printf "Building $filename\n";
5291
5292     my $locale_data = build_locale_data();
5293     my $charmaps_data = build_charmaps_data();
5294     my $geoids_data = build_geoids_data();
5295     my $scripts_data = "";  # FIXME
5296
5297     my @header = ( 0 ) x 8;
5298     $header[0] = 4 * scalar @header;  # chartypes offset
5299     $header[4] = $header[0] + length $chartypes;  # locales offset
5300     $header[5] = $header[4] + length $locale_data;  # charmaps offset
5301     $header[6] = $header[5] + length $charmaps_data;  # geoids offset
5302     $header[7] = $header[6] + length $geoids_data;  # scripts offset
5303
5304     open OUTPUT, ">$filename.new" or die "Cannot create $filename";
5305     print OUTPUT pack "L<*", @header;
5306     print OUTPUT $chartypes, $locale_data, $charmaps_data, $geoids_data, $scripts_data;
5307     close OUTPUT;
5308     save_file($filename);
5309 }
5310
5311
5312 ################################################################
5313 # return the day of week of the first of the month
5314 sub month_first_dow($$)
5315 {
5316     my ($year, $month) = @_;
5317     my @time = gmtime( timegm_modern( 0, 0, 0, 1, $month - 1, $year ));
5318     return $time[6];
5319 }
5320
5321
5322 ################################################################
5323 # compare system time values
5324 sub compare_systime($$)
5325 {
5326     my ($a, $b) = @_;
5327     return $a->[0] <=> $b->[0] ||
5328            $a->[1] <=> $b->[1] ||
5329            $a->[2] <=> $b->[2] ||
5330            $a->[3] <=> $b->[3] ||
5331            $a->[4] <=> $b->[4] ||
5332            $a->[5] <=> $b->[5] ||
5333            $a->[6] <=> $b->[6];
5334 }
5335
5336
5337 ################################################################
5338 # compare the zone transition date with the rule date
5339 sub compare_transition_date($$$$)
5340 {
5341     my ($stdoff, $isdst, $zone, $rule) = @_;
5342
5343     if (scalar @{$zone} <= 1)
5344     {
5345         return (!defined($zone->[0]) || $zone->[0] > $rule->[0]) ? 1 : -1;
5346     }
5347
5348     my @date = parse_transition_date( $stdoff, $isdst, $zone->[0], $zone->[1], $zone->[2], $zone->[3] || 0 );
5349     return compare_systime( \@date, $rule );
5350 }
5351
5352
5353 ################################################################
5354 # get the Windows zone names from the CLDR data
5355 sub load_windows_zones()
5356 {
5357     my $current_name;
5358     my %names;
5359     my $base = "cldr-release-$CLDRVERSION";
5360     my $INPUT = open_data_file( "cldr", "$base/common/supplemental/windowsZones.xml" );
5361     while (<$INPUT>)
5362     {
5363         if (/<!-- +(\(UTC.*) -->.*/)
5364         {
5365             $current_name = $1;
5366         }
5367         if (/<mapZone other="(.*)" territory="001" type="(.*)"\/>/)
5368         {
5369             $names{$1} = [ $current_name, $2 ];
5370         }
5371     }
5372     close $INPUT;
5373     return %names;
5374 }
5375
5376
5377 ################################################################
5378 # parse a transition date specification from the tzdata files
5379 sub parse_transition_date($$@)
5380 {
5381     use integer;
5382     my ($stdoff, $isdst, $year, $in, $on, $at) = @_;
5383
5384     $on = "1" unless defined $on;
5385     $at = "0" unless defined $at;
5386
5387     my %months = ( Jan => 1, Feb => 2, Mar => 3, Apr => 4, May => 5, Jun => 6,
5388                    Jul => 7, Aug => 8, Sep => 9, Oct => 10, Nov => 11, Dec => 12 );
5389     my %days = ( Sun => 0, Mon => 1, Tue => 2, Wed => 3, Thu => 4, Fri => 5, Sat => 6 );
5390
5391     my $mon = $in ? $months{$in} : 1;
5392     my ($week, $dow, $flag, $time, $sec);
5393     my $first = month_first_dow( $year, $mon );
5394
5395     if ($on =~ /^last(.*)$/)
5396     {
5397         $week = 5;
5398         $dow = $days{$1};
5399     }
5400     elsif ($on =~ /^(.*)>=(\d+)$/)
5401     {
5402         $dow = $days{$1};
5403         my $diff = ($first + 6 - $dow) % 7;
5404         $week = $2 >= 25 ? 5 : ($2 + 6 + $diff) / 7;
5405     }
5406     elsif ($on =~ /^(.*)<=(\d+)$/)
5407     {
5408         $dow = $days{$1};
5409         my $diff = ($first + $2 + 6 - $dow) % 7;
5410         $week = ($2 + 6 - $diff) / 7;
5411         if (!$week)
5412         {
5413             $week = 5;
5414             if (!--$mon) { $mon = 12; $year--; }
5415         }
5416     }
5417     elsif ($on =~ /^\d+$/)
5418     {
5419         $dow = ($first + $on - 1) % 7;
5420         $week = $on >= 25 ? 5 : ($on + 6) / 7;
5421     }
5422     else
5423     {
5424         die "unsupported date specification $year $in $on $at";
5425     }
5426
5427     if ($at =~ /^(\d+):(\d+):(\d+)([uws]?)$/)
5428     {
5429         $time = $1 * 60 + $2;
5430         $sec = $3;
5431         $flag = $4;
5432     }
5433     elsif ($at =~ /^(\d+):(\d+)([uws]?)$/)
5434     {
5435         $time = $1 * 60 + $2;
5436         $flag = $3;
5437     }
5438     elsif ($at =~ /^(\d+)([uws]?)$/)
5439     {
5440         $time = $1 * 60;
5441         $flag = $2;
5442     }
5443     else
5444     {
5445         die "unsupported time specification $year $in $on $at";
5446     }
5447
5448     $flag ||= "w";
5449     $time -= $stdoff if $flag eq "u";
5450     $time += 60 if !$isdst && $flag ne "w";
5451
5452     if ($time < 0)  # previous day
5453     {
5454         $week-- if $week < 5 && $dow == month_first_dow( $year, $mon );
5455         $week-- if $week == 5 && $dow == month_first_dow( $year + ($mon == 12), $mon % 12 + 1 );
5456         if (!$week)
5457         {
5458             $week = 5;
5459             if (!--$mon) { $mon = 12; $year--; }
5460         }
5461         $dow = ($dow + 6) % 7;
5462         $time += 24 * 60;
5463     }
5464
5465     return ($year, $mon, $week, $dow, $time / 60, $time % 60, $sec || 0);
5466 }
5467
5468
5469 ################################################################
5470 # parse a system time value as a SYSTEMTIME structure
5471 sub pack_systime(@)
5472 {
5473     my ($year, $mon, $week, $dow, $hour, $min, $sec) = @_;
5474     return pack "S<8", 0, $mon, $dow, $week, $hour < 24 ? ($hour, $min, $sec, 0) : (23, 59, 59, 999);
5475 }
5476
5477
5478 ################################################################
5479 # parse a timezone offset from the tzdata files
5480 sub parse_tz_offset($)
5481 {
5482     my ($hour, $min) = split /:/, shift;
5483     $min ||= 0;
5484     return $hour < 0 ? -$hour * 60 + $min : -$hour * 60 - $min;  # invert sign
5485 }
5486
5487
5488 ################################################################
5489 # build the timezone data
5490 sub dump_timezones($@)
5491 {
5492     my $filename = shift;
5493     my $FIRST_YEAR = 2000;
5494     my $LAST_YEAR = 2030;
5495
5496     my %names = load_windows_zones();
5497     my %zones;
5498     my %rules;
5499     my %links;
5500     my %res_indices;
5501
5502     printf "Building $filename\n";
5503
5504     open OUTPUT, ">$filename.new" or die "Cannot create $filename";
5505     print OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
5506     print OUTPUT "#include \"winresrc.h\"\n\n";
5507     print OUTPUT "#pragma makedep po\n\n";
5508     print OUTPUT "LANGUAGE LANG_ENGLISH, SUBLANG_DEFAULT\n\n";
5509     print OUTPUT "STRINGTABLE\n{\n";
5510
5511     # load tzdata files
5512
5513     foreach my $filename (@_)
5514     {
5515         my $FILE = open_data_file( "tzdata", $filename );
5516         my $zonename;
5517         while (<$FILE>)
5518         {
5519             chomp;
5520             s/\#.*$//;
5521             next if /^\s*$/;
5522             my @fields = split /\s+/;
5523             if ($fields[0] eq "Zone" || ($zonename && $fields[0] eq ""))
5524             {
5525                 shift @fields;
5526                 $zonename = shift @fields unless $zonename;
5527                 my ($stdoff, $rules, $dummy, @date) = @fields;
5528                 $zones{$zonename} ||= [ ];
5529                 push @{$zones{$zonename}}, [ parse_tz_offset( $stdoff ), $rules, @date ];
5530                 $zonename = undef unless @date;  # last entry doesn't have an until date
5531                 next;
5532             }
5533             if ($fields[0] eq "Rule")
5534             {
5535                 shift @fields;
5536                 my ($rulename, $from, $to, $dummy, $in, $on, $at, $save) = @fields;
5537                 $to = $from if $to eq "only";
5538                 $to = $LAST_YEAR if $to eq "max";
5539                 push @{$rules{$rulename}}, [ parse_tz_offset( $save ), $from, $to, $in, $on, $at ];
5540                 next;
5541             }
5542             if ($fields[0] eq "Link")
5543             {
5544                 $links{$fields[2]} = $fields[1];
5545                 next;
5546             }
5547             die "unrecognized line $_";
5548         }
5549         close $FILE;
5550     }
5551
5552     foreach my $name (sort { uc($a) cmp uc($b) } keys %names)
5553     {
5554         my ($display, $zone) = @{$names{$name}};
5555         $zone = $links{$zone} if defined $links{$zone};
5556
5557         # build list of transitions
5558
5559         my @transitions;
5560         my @from_date = ( 1 );
5561         my $last_stdoff = 0;
5562         for (my $i = 0; $i < scalar @{$zones{$zone}}; $i++)
5563         {
5564             my ($stdoff, $rule, @until_date) = @{$zones{$zone}->[$i]};
5565             my $isdst = ($last_stdoff != $stdoff);
5566             $from_date[0] ||= $LAST_YEAR;
5567             my @systime = parse_transition_date( $stdoff, $isdst, @from_date );
5568             push @transitions, [ $stdoff, -1, \@systime ];
5569
5570             if (defined $rules{$rule})
5571             {
5572                 foreach my $r (@{$rules{$rule}})
5573                 {
5574                     my ($offset, $from, $to, $in, $on, $at) = @{$r};
5575                     foreach my $year ($from..$to)
5576                     {
5577                         next if $year < $from_date[0];
5578                         next if $until_date[0] && $year > $until_date[0];
5579                         my @systime = parse_transition_date( $stdoff, !!$offset, $year, $in, $on, $at );
5580                         next if compare_transition_date( $stdoff, $isdst, \@until_date, \@systime ) <= 0;
5581                         my $ret = compare_transition_date( $stdoff, $isdst, \@from_date, \@systime );
5582                         next if $ret > 0;
5583                         pop @transitions if !$ret;  # remove transition if there's a dst change at the same time
5584                         push @transitions, [ $stdoff, $offset, \@systime ];
5585                     }
5586                 }
5587             }
5588             @from_date = @until_date;
5589             $last_stdoff = $stdoff;
5590         }
5591         @transitions = sort { compare_systime( $a->[2], $b->[2] ) }  @transitions;
5592
5593         # build per-year dynamic info
5594
5595         my @info;
5596         my $last_dstoff = 0;
5597         my $last_dst = 0;
5598         my $year = $FIRST_YEAR;
5599         while ($year <= $LAST_YEAR)
5600         {
5601             if (@transitions && $transitions[0]->[2]->[0] < $year)
5602             {
5603                 $last_stdoff = $transitions[0]->[0];
5604                 shift @transitions;
5605                 next;
5606             }
5607             my ($std, $dst, @trans);
5608             my $cur_stdoff = $last_stdoff;
5609             my $cur_dstoff = ($name =~ /^UTC/) ? 0 : -60;
5610             while (@transitions && $transitions[0]->[2]->[0] == $year)
5611             {
5612                 my $t = shift @transitions;
5613                 my ($stdoff, $dstoff, $systime) = @{$t};
5614                 $systime = pack_systime( @{$systime} );
5615                 if (!$dstoff)  # std
5616                 {
5617                     $cur_stdoff = $stdoff unless $std;
5618                     $std = $systime;
5619                 }
5620                 elsif ($dstoff != -1)  # dst
5621                 {
5622                     $cur_dstoff = $dstoff unless $dst;
5623                     $dst ||= $systime;
5624                 }
5625                 elsif ($stdoff != $last_stdoff)  # rule transition
5626                 {
5627                     # Handle a special case: Samoa moved to the other side of
5628                     # the date line between 2011-12-03 and 2012-01-01,
5629                     # entirely skipping the day 2011-12-31. We ignore this
5630                     # change because it happens on a year boundary and more
5631                     # importantly it would generate on offset of -25 hours,
5632                     # which some programs (e.g., Mono) do not like. See
5633                     # https://bugs.winehq.org/show_bug.cgi?id=51758
5634
5635                     if ($last_stdoff - $stdoff < 24 * 60)
5636                     {
5637                         @trans = ($last_stdoff, $stdoff, $systime);
5638                         $cur_stdoff = $stdoff;
5639                     }
5640                 }
5641                 elsif ($dst)  # rule transition with no stdoff change
5642                 {
5643                     $std = $systime;
5644                 }
5645                 $last_dstoff = ($dstoff == -1) ? 0 : $dstoff;
5646             }
5647             $last_stdoff = $cur_stdoff;
5648
5649             if ($cur_dstoff > 0)  # swap std and dst to ensure that offset is negative
5650             {
5651                 ($std, $dst) = ($dst, $std);
5652                 $cur_stdoff += $cur_dstoff;
5653                 $cur_dstoff = -$cur_dstoff;
5654             }
5655
5656             if (@trans)
5657             {
5658                 # heuristic to prefer switching dst
5659                 if ($last_dst == $year - 1 || (!$last_dst && $trans[0] > $trans[1]))
5660                 {
5661                     $dst ||= $trans[2];
5662                     $cur_stdoff = $trans[0];
5663                     $cur_dstoff = $trans[1] - $trans[0];
5664                 }
5665                 else
5666                 {
5667                     $std ||= $trans[2];
5668                     $cur_stdoff = $trans[1];
5669                     $cur_dstoff = $trans[0] - $trans[1];
5670                 }
5671             }
5672
5673             if ($std || $dst)
5674             {
5675                 $std ||= pack_systime( parse_transition_date( 0, 0, $year, "Jan", 1 ));
5676                 $dst ||= pack_systime( parse_transition_date( 0, 0, $year, "Jan", 1 ));
5677                 $last_dst = $year;
5678             }
5679             else
5680             {
5681                 $std = pack "S<8", 0;
5682                 $dst = pack "S<8", 0;
5683                 $cur_stdoff += $last_dstoff;
5684             }
5685             $info[$year++] = pack( "l<3", $cur_stdoff, 0, $cur_dstoff ) . $std . $dst;
5686         }
5687
5688         # output registry keys
5689
5690         my $std_name = $name eq "UTC" ? "Coordinated Universal Time" : $name;
5691         my $dlt_name = $std_name =~ s/Standard Time/Daylight Time/r;
5692         my $res_idx = hex( substr( Digest::SHA::sha1_hex($name), -3, 3 )) << 4;
5693         $res_idx += 16 while exists $res_indices{$res_idx};
5694         $res_indices{$res_idx} = 1;
5695
5696         add_registry_string_value( $zonekey, $name, "Display", $display );
5697         add_registry_string_value( $zonekey, $name, "Std", $std_name );
5698         add_registry_string_value( $zonekey, $name, "Dlt", $dlt_name );
5699         add_registry_string_value( $zonekey, $name, "MUI_Std", sprintf( "\@tzres.dll,-%u", $res_idx ));
5700         add_registry_string_value( $zonekey, $name, "MUI_Dlt", sprintf( "\@tzres.dll,-%u", $res_idx + 1 ));
5701         add_registry_string_value( $zonekey, $name, "MUI_Display", sprintf( "\@tzres.dll,-%u", $res_idx + 2 ));
5702         add_registry_binary_value( $zonekey, $name, "TZI", $info[$LAST_YEAR] );
5703
5704         printf OUTPUT "%7d \"#msgctxt#maximum 31 characters#%s\"\n", $res_idx, $std_name;
5705         printf OUTPUT "%7d \"#msgctxt#maximum 31 characters#%s\"\n", $res_idx + 1, $dlt_name;
5706         printf OUTPUT "%7d \"%s\"\n", $res_idx + 2, $display;
5707
5708         my $first_year = $FIRST_YEAR;
5709         my $last_year = $LAST_YEAR;
5710         $last_year-- while $last_year > $FIRST_YEAR && $info[$last_year] eq $info[$last_year - 1];
5711         $first_year++ while $first_year < $last_year && $info[$first_year] eq $info[$last_year];
5712
5713         next if $last_year <= $first_year;
5714
5715         foreach my $i ($first_year..$last_year)
5716         {
5717             add_registry_binary_value( $zonekey, "$name\\Dynamic DST", $i, $info[$i] );
5718         }
5719         add_registry_dword_value( $zonekey, "$name\\Dynamic DST", "FirstEntry", $first_year );
5720         add_registry_dword_value( $zonekey, "$name\\Dynamic DST", "LastEntry", $last_year );
5721     }
5722
5723     print OUTPUT "}\n";
5724     close OUTPUT;
5725     save_file($filename);
5726 }
5727
5728
5729 ################################################################
5730 # build the script to create registry keys
5731 sub dump_registry_script($%)
5732 {
5733     my ($filename, %keys) = @_;
5734     my $indent = 1;
5735     my @prev;
5736
5737     printf "Building %s\n", $filename;
5738     open OUTPUT, ">$filename.new" or die "Cannot create $filename";
5739     print OUTPUT "HKLM\n{\n";
5740     foreach my $k (sort { ($a =~ tr/a-z\\/A-Z\001/r) cmp ($b =~ tr/a-z\\/A-Z\001/r) } keys %keys)
5741     {
5742         my @subkeys = split /\\/, $k;
5743         while (@prev && @subkeys && $prev[0] eq $subkeys[0]) { shift @prev; shift @subkeys; }
5744         while (@prev) { printf OUTPUT "%*s}\n", 4 * --$indent, ""; shift @prev; }
5745         my ($def, @vals) = @{$keys{$k}};
5746         for (my $i = 0; $i < @subkeys; $i++)
5747         {
5748             my $name = $subkeys[$i];
5749             my $prefix = "";
5750             if ($name =~ /^-/)
5751             {
5752                 $name =~ s/^-//;
5753                 $prefix = "NoRemove ";
5754             }
5755             if ($name =~ /\s/)
5756             {
5757                 $name = "'$name'";
5758             }
5759             printf OUTPUT "%*s%s%s%s\n%*s{\n", 4 * $indent, "", $prefix, $name,
5760                 $i == $#subkeys && $def ? " = s '$def'" : "", 4 * $indent, "";
5761             $indent++;
5762         }
5763         foreach my $v (sort @vals) { printf OUTPUT "%*sval $v\n", 4 * $indent, ""; }
5764         @prev = split /\\/, $k;
5765     }
5766     while (@prev) { printf OUTPUT "%*s}\n", 4 * --$indent, ""; shift @prev; }
5767     printf OUTPUT "}\n";
5768     close OUTPUT;
5769     save_file($filename);
5770 }
5771
5772
5773 ################################################################
5774 # save a file if modified
5775 sub save_file($)
5776 {
5777     my $file = shift;
5778     if (-f $file && !system "cmp $file $file.new >/dev/null")
5779     {
5780         unlink "$file.new";
5781     }
5782     else
5783     {
5784         rename "$file.new", "$file";
5785     }
5786 }
5787
5788
5789 ################################################################
5790 # main routine
5791
5792 chdir ".." if -f "./make_unicode";
5793 load_data();
5794 dump_bidi_dir_table( "dlls/gdi32/uniscribe/direction.c" );
5795 dump_bidi_dir_table( "dlls/dwrite/direction.c" );
5796 dump_mirroring( "dlls/gdi32/uniscribe/mirror.c" );
5797 dump_mirroring( "dlls/dwrite/mirror.c" );
5798 dump_bracket( "dlls/gdi32/uniscribe/bracket.c" );
5799 dump_bracket( "dlls/dwrite/bracket.c" );
5800 dump_shaping( "dlls/gdi32/uniscribe/shaping.c" );
5801 dump_arabic_shaping( "dlls/dwrite/shapers/arabic_table.c" );
5802 dump_linebreak( "dlls/gdi32/uniscribe/linebreak.c" );
5803 dump_linebreak( "dlls/dwrite/linebreak.c" );
5804 dump_scripts( "dlls/dwrite/scripts" );
5805 dump_indic( "dlls/gdi32/uniscribe/indicsyllable.c" );
5806 dump_vertical( "dlls/win32u/vertical.c", 1 );
5807 dump_vertical( "dlls/wineps.drv/vertical.c", 0 );
5808 dump_intl_nls("nls/l_intl.nls");
5809 dump_norm_table( "nls/normnfc.nls" );
5810 dump_norm_table( "nls/normnfd.nls" );
5811 dump_norm_table( "nls/normnfkc.nls" );
5812 dump_norm_table( "nls/normnfkd.nls" );
5813 dump_norm_table( "nls/normidna.nls" );
5814 my $chartypes = dump_sortkey_table( "nls/sortdefault.nls" );
5815 dump_locales( "nls/locale.nls", $chartypes );
5816 foreach my $file (@allfiles) { dump_msdata_codepage( $file ); }
5817 dump_eucjp_codepage();
5818 dump_timezones( "dlls/tzres/tzres.rc", @timezone_files );
5819 dump_registry_script( "dlls/kernelbase/kernelbase.rgs", %registry_keys );
5820
5821 exit 0;
5822
5823 # Local Variables:
5824 # compile-command: "./make_unicode"
5825 # End: