| 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 | #
|
|---|
| 23 | # Sun LGPL Disclaimer: For the avoidance of doubt, except that if any license choice
|
|---|
| 24 | # other than GPL or LGPL is available it will apply instead, Sun elects to use only
|
|---|
| 25 | # the Lesser General Public License version 2.1 (LGPLv2) at this time for any software where
|
|---|
| 26 | # a choice of LGPL license versions is made available with the language indicating
|
|---|
| 27 | # that LGPLv2 or any later version may be used, or where a choice of which version
|
|---|
| 28 | # of the LGPL is applied is otherwise unspecified.
|
|---|
| 29 | #
|
|---|
| 30 |
|
|---|
| 31 | #
|
|---|
| 32 | # Sun LGPL Disclaimer: For the avoidance of doubt, except that if any license choice
|
|---|
| 33 | # other than GPL or LGPL is available it will apply instead, Sun elects to use only
|
|---|
| 34 | # the Lesser General Public License version 2.1 (LGPLv2) at this time for any software where
|
|---|
| 35 | # a choice of LGPL license versions is made available with the language indicating
|
|---|
| 36 | # that LGPLv2 or any later version may be used, or where a choice of which version
|
|---|
| 37 | # of the LGPL is applied is otherwise unspecified.
|
|---|
| 38 | #
|
|---|
| 39 |
|
|---|
| 40 | use strict;
|
|---|
| 41 |
|
|---|
| 42 | # base directory for ftp.unicode.org files
|
|---|
| 43 | my $BASEDIR = "ftp.unicode.org/Public/";
|
|---|
| 44 | my $MAPPREFIX = $BASEDIR . "MAPPINGS/";
|
|---|
| 45 |
|
|---|
| 46 | # UnicodeData file
|
|---|
| 47 | my $UNICODEDATA = $BASEDIR . "UNIDATA/UnicodeData.txt";
|
|---|
| 48 |
|
|---|
| 49 | # Sort keys file
|
|---|
| 50 | my $SORTKEYS = "www.unicode.org/reports/tr10/allkeys.txt";
|
|---|
| 51 |
|
|---|
| 52 | # Defaults mapping
|
|---|
| 53 | my $DEFAULTS = "./defaults";
|
|---|
| 54 |
|
|---|
| 55 | # Default char for undefined mappings
|
|---|
| 56 | my $DEF_CHAR = ord '?';
|
|---|
| 57 |
|
|---|
| 58 | my @allfiles =
|
|---|
| 59 | (
|
|---|
| 60 | [ 37, "VENDORS/MICSFT/EBCDIC/CP037.TXT", 0, "IBM EBCDIC US Canada" ],
|
|---|
| 61 | [ 424, "VENDORS/MISC/CP424.TXT", 0, "IBM EBCDIC Hebrew" ],
|
|---|
| 62 | [ 437, "VENDORS/MICSFT/PC/CP437.TXT", 1, "OEM United States" ],
|
|---|
| 63 | [ 500, "VENDORS/MICSFT/EBCDIC/CP500.TXT", 0, "IBM EBCDIC International" ],
|
|---|
| 64 | [ 737, "VENDORS/MICSFT/PC/CP737.TXT", 1, "OEM Greek 437G" ],
|
|---|
| 65 | [ 775, "VENDORS/MICSFT/PC/CP775.TXT", 1, "OEM Baltic" ],
|
|---|
| 66 | [ 850, "VENDORS/MICSFT/PC/CP850.TXT", 1, "OEM Multilingual Latin 1" ],
|
|---|
| 67 | [ 852, "VENDORS/MICSFT/PC/CP852.TXT", 1, "OEM Slovak Latin 2" ],
|
|---|
| 68 | [ 855, "VENDORS/MICSFT/PC/CP855.TXT", 1, "OEM Cyrillic" ],
|
|---|
| 69 | [ 856, "VENDORS/MISC/CP856.TXT", 0, "Hebrew PC" ],
|
|---|
| 70 | [ 857, "VENDORS/MICSFT/PC/CP857.TXT", 1, "OEM Turkish" ],
|
|---|
| 71 | [ 860, "VENDORS/MICSFT/PC/CP860.TXT", 1, "OEM Portuguese" ],
|
|---|
| 72 | [ 861, "VENDORS/MICSFT/PC/CP861.TXT", 1, "OEM Icelandic" ],
|
|---|
| 73 | [ 862, "VENDORS/MICSFT/PC/CP862.TXT", 1, "OEM Hebrew" ],
|
|---|
| 74 | [ 863, "VENDORS/MICSFT/PC/CP863.TXT", 1, "OEM Canadian French" ],
|
|---|
| 75 | [ 864, "VENDORS/MICSFT/PC/CP864.TXT", 0, "OEM Arabic" ],
|
|---|
| 76 | [ 865, "VENDORS/MICSFT/PC/CP865.TXT", 1, "OEM Nordic" ],
|
|---|
| 77 | [ 866, "VENDORS/MICSFT/PC/CP866.TXT", 1, "OEM Russian" ],
|
|---|
| 78 | [ 869, "VENDORS/MICSFT/PC/CP869.TXT", 1, "OEM Greek" ],
|
|---|
| 79 | [ 874, "VENDORS/MICSFT/WindowsBestFit/bestfit874.txt", 1, "ANSI/OEM Thai" ],
|
|---|
| 80 | [ 875, "VENDORS/MICSFT/EBCDIC/CP875.TXT", 0, "IBM EBCDIC Greek" ],
|
|---|
| 81 | [ 878, "VENDORS/MISC/KOI8-R.TXT", 0, "Russian KOI8" ],
|
|---|
| 82 | [ 932, "VENDORS/MICSFT/WindowsBestFit/bestfit932.txt", 0, "ANSI/OEM Japanese Shift-JIS" ],
|
|---|
| 83 | [ 936, "VENDORS/MICSFT/WindowsBestFit/bestfit936.txt", 0, "ANSI/OEM Simplified Chinese GBK" ],
|
|---|
| 84 | [ 949, "VENDORS/MICSFT/WindowsBestFit/bestfit949.txt", 0, "ANSI/OEM Korean Unified Hangul" ],
|
|---|
| 85 | [ 950, "VENDORS/MICSFT/WindowsBestFit/bestfit950.txt", 0, "ANSI/OEM Traditional Chinese Big5" ],
|
|---|
| 86 | [ 1006, "VENDORS/MISC/CP1006.TXT", 0, "IBM Arabic" ],
|
|---|
| 87 | [ 1026, "VENDORS/MICSFT/EBCDIC/CP1026.TXT", 0, "IBM EBCDIC Latin 5 Turkish" ],
|
|---|
| 88 | [ 1250, "VENDORS/MICSFT/WindowsBestFit/bestfit1250.txt", 0, "ANSI Eastern Europe" ],
|
|---|
| 89 | [ 1251, "VENDORS/MICSFT/WindowsBestFit/bestfit1251.txt", 0, "ANSI Cyrillic" ],
|
|---|
| 90 | [ 1252, "VENDORS/MICSFT/WindowsBestFit/bestfit1252.txt", 0, "ANSI Latin 1" ],
|
|---|
| 91 | [ 1253, "VENDORS/MICSFT/WindowsBestFit/bestfit1253.txt", 0, "ANSI Greek" ],
|
|---|
| 92 | [ 1254, "VENDORS/MICSFT/WindowsBestFit/bestfit1254.txt", 0, "ANSI Turkish" ],
|
|---|
| 93 | [ 1255, "VENDORS/MICSFT/WindowsBestFit/bestfit1255.txt", 0, "ANSI Hebrew" ],
|
|---|
| 94 | [ 1256, "VENDORS/MICSFT/WindowsBestFit/bestfit1256.txt", 0, "ANSI Arabic" ],
|
|---|
| 95 | [ 1257, "VENDORS/MICSFT/WindowsBestFit/bestfit1257.txt", 0, "ANSI Baltic" ],
|
|---|
| 96 | [ 1258, "VENDORS/MICSFT/WindowsBestFit/bestfit1258.txt", 0, "ANSI/OEM Viet Nam" ],
|
|---|
| 97 | [ 1361, "OBSOLETE/EASTASIA/KSC/JOHAB.TXT", 0, "Korean Johab" ],
|
|---|
| 98 | [ 10000, "VENDORS/MICSFT/MAC/ROMAN.TXT", 0, "Mac Roman" ],
|
|---|
| 99 | [ 10006, "VENDORS/MICSFT/MAC/GREEK.TXT", 0, "Mac Greek" ],
|
|---|
| 100 | [ 10007, "VENDORS/MICSFT/MAC/CYRILLIC.TXT", 0, "Mac Cyrillic" ],
|
|---|
| 101 | [ 10029, "VENDORS/MICSFT/MAC/LATIN2.TXT", 0, "Mac Latin 2" ],
|
|---|
| 102 | [ 10079, "VENDORS/MICSFT/MAC/ICELAND.TXT", 0, "Mac Icelandic" ],
|
|---|
| 103 | [ 10081, "VENDORS/MICSFT/MAC/TURKISH.TXT", 0, "Mac Turkish" ],
|
|---|
| 104 | [ 20127, undef, 0, "US-ASCII (7bit)" ],
|
|---|
| 105 | [ 20866, "VENDORS/MISC/KOI8-R.TXT", 0, "Russian KOI8" ],
|
|---|
| 106 | [ 20932, "OBSOLETE/EASTASIA/JIS/JIS0208.TXT", 0, "EUC-JP" ],
|
|---|
| 107 | [ 21866, "VENDORS/MISC/KOI8-U.TXT", 0, "Ukrainian KOI8" ],
|
|---|
| 108 | [ 28591, "ISO8859/8859-1.TXT", 0, "ISO 8859-1 Latin 1" ],
|
|---|
| 109 | [ 28592, "ISO8859/8859-2.TXT", 0, "ISO 8859-2 Latin 2 (East European)" ],
|
|---|
| 110 | [ 28593, "ISO8859/8859-3.TXT", 0, "ISO 8859-3 Latin 3 (South European)" ],
|
|---|
| 111 | [ 28594, "ISO8859/8859-4.TXT", 0, "ISO 8859-4 Latin 4 (Baltic old)" ],
|
|---|
| 112 | [ 28595, "ISO8859/8859-5.TXT", 0, "ISO 8859-5 Cyrillic" ],
|
|---|
| 113 | [ 28596, "ISO8859/8859-6.TXT", 0, "ISO 8859-6 Arabic" ],
|
|---|
| 114 | [ 28597, "ISO8859/8859-7.TXT", 0, "ISO 8859-7 Greek" ],
|
|---|
| 115 | [ 28598, "ISO8859/8859-8.TXT", 0, "ISO 8859-8 Hebrew" ],
|
|---|
| 116 | [ 28599, "ISO8859/8859-9.TXT", 0, "ISO 8859-9 Latin 5 (Turkish)" ],
|
|---|
| 117 | [ 28600, "ISO8859/8859-10.TXT", 0, "ISO 8859-10 Latin 6 (Nordic)" ],
|
|---|
| 118 | [ 28603, "ISO8859/8859-13.TXT", 0, "ISO 8859-13 Latin 7 (Baltic)" ],
|
|---|
| 119 | [ 28604, "ISO8859/8859-14.TXT", 0, "ISO 8859-14 Latin 8 (Celtic)" ],
|
|---|
| 120 | [ 28605, "ISO8859/8859-15.TXT", 0, "ISO 8859-15 Latin 9 (Euro)" ],
|
|---|
| 121 | [ 28606, "ISO8859/8859-16.TXT", 0, "ISO 8859-16 Latin 10 (Balkan)" ]
|
|---|
| 122 | );
|
|---|
| 123 |
|
|---|
| 124 |
|
|---|
| 125 | my %ctype =
|
|---|
| 126 | (
|
|---|
| 127 | "upper" => 0x0001,
|
|---|
| 128 | "lower" => 0x0002,
|
|---|
| 129 | "digit" => 0x0004,
|
|---|
| 130 | "space" => 0x0008,
|
|---|
| 131 | "punct" => 0x0010,
|
|---|
| 132 | "cntrl" => 0x0020,
|
|---|
| 133 | "blank" => 0x0040,
|
|---|
| 134 | "xdigit" => 0x0080,
|
|---|
| 135 | "alpha" => 0x0100
|
|---|
| 136 | );
|
|---|
| 137 |
|
|---|
| 138 | my %categories =
|
|---|
| 139 | (
|
|---|
| 140 | "Lu" => $ctype{"alpha"}|$ctype{"upper"}, # Letter, Uppercase
|
|---|
| 141 | "Ll" => $ctype{"alpha"}|$ctype{"lower"}, # Letter, Lowercase
|
|---|
| 142 | "Lt" => $ctype{"alpha"}, # Letter, Titlecase
|
|---|
| 143 | "Mn" => $ctype{"punct"}, # Mark, Non-Spacing
|
|---|
| 144 | "Mc" => $ctype{"punct"}, # Mark, Spacing Combining
|
|---|
| 145 | "Me" => $ctype{"punct"}, # Mark, Enclosing
|
|---|
| 146 | "Nd" => $ctype{"digit"}, # Number, Decimal Digit
|
|---|
| 147 | "Nl" => $ctype{"punct"}, # Number, Letter
|
|---|
| 148 | "No" => $ctype{"punct"}, # Number, Other
|
|---|
| 149 | "Zs" => $ctype{"space"}, # Separator, Space
|
|---|
| 150 | "Zl" => $ctype{"space"}, # Separator, Line
|
|---|
| 151 | "Zp" => $ctype{"space"}, # Separator, Paragraph
|
|---|
| 152 | "Cc" => $ctype{"cntrl"}, # Other, Control
|
|---|
| 153 | "Cf" => 0, # Other, Format
|
|---|
| 154 | "Cs" => 0, # Other, Surrogate
|
|---|
| 155 | "Co" => 0, # Other, Private Use
|
|---|
| 156 | "Cn" => 0, # Other, Not Assigned
|
|---|
| 157 | "Lm" => $ctype{"punct"}, # Letter, Modifier
|
|---|
| 158 | "Lo" => $ctype{"alpha"}, # Letter, Other
|
|---|
| 159 | "Pc" => $ctype{"punct"}, # Punctuation, Connector
|
|---|
| 160 | "Pd" => $ctype{"punct"}, # Punctuation, Dash
|
|---|
| 161 | "Ps" => $ctype{"punct"}, # Punctuation, Open
|
|---|
| 162 | "Pe" => $ctype{"punct"}, # Punctuation, Close
|
|---|
| 163 | "Pi" => $ctype{"punct"}, # Punctuation, Initial quote
|
|---|
| 164 | "Pf" => $ctype{"punct"}, # Punctuation, Final quote
|
|---|
| 165 | "Po" => $ctype{"punct"}, # Punctuation, Other
|
|---|
| 166 | "Sm" => $ctype{"punct"}, # Symbol, Math
|
|---|
| 167 | "Sc" => $ctype{"punct"}, # Symbol, Currency
|
|---|
| 168 | "Sk" => $ctype{"punct"}, # Symbol, Modifier
|
|---|
| 169 | "So" => $ctype{"punct"} # Symbol, Other
|
|---|
| 170 | );
|
|---|
| 171 |
|
|---|
| 172 | # a few characters need additional categories that cannot be determined automatically
|
|---|
| 173 | my %special_categories =
|
|---|
| 174 | (
|
|---|
| 175 | "xdigit" => [ ord('0')..ord('9'),ord('A')..ord('F'),ord('a')..ord('f'),
|
|---|
| 176 | 0xff10..0xff19, 0xff21..0xff26, 0xff41..0xff46 ],
|
|---|
| 177 | "space" => [ 0x09..0x0d, 0x85 ],
|
|---|
| 178 | "blank" => [ 0x09, 0x20, 0xa0, 0x3000, 0xfeff ],
|
|---|
| 179 | "cntrl" => [ 0x070f, 0x180b, 0x180c, 0x180d, 0x180e, 0x200c, 0x200d,
|
|---|
| 180 | 0x200e, 0x200f, 0x202a, 0x202b, 0x202c, 0x202d, 0x202e,
|
|---|
| 181 | 0x206a, 0x206b, 0x206c, 0x206d, 0x206e, 0x206f, 0xfeff,
|
|---|
| 182 | 0xfff9, 0xfffa, 0xfffb ]
|
|---|
| 183 | );
|
|---|
| 184 |
|
|---|
| 185 | my %directions =
|
|---|
| 186 | (
|
|---|
| 187 | "L" => 1, # Left-to-Right
|
|---|
| 188 | "LRE" => 11, # Left-to-Right Embedding
|
|---|
| 189 | "LRO" => 11, # Left-to-Right Override
|
|---|
| 190 | "R" => 2, # Right-to-Left
|
|---|
| 191 | "AL" => 2, # Right-to-Left Arabic
|
|---|
| 192 | "RLE" => 11, # Right-to-Left Embedding
|
|---|
| 193 | "RLO" => 11, # Right-to-Left Override
|
|---|
| 194 | "PDF" => 11, # Pop Directional Format
|
|---|
| 195 | "EN" => 3, # European Number
|
|---|
| 196 | "ES" => 4, # European Number Separator
|
|---|
| 197 | "ET" => 5, # European Number Terminator
|
|---|
| 198 | "AN" => 6, # Arabic Number
|
|---|
| 199 | "CS" => 7, # Common Number Separator
|
|---|
| 200 | "NSM" => 0, # Non-Spacing Mark
|
|---|
| 201 | "BN" => 0, # Boundary Neutral
|
|---|
| 202 | "B" => 8, # Paragraph Separator
|
|---|
| 203 | "S" => 9, # Segment Separator
|
|---|
| 204 | "WS" => 10, # Whitespace
|
|---|
| 205 | "ON" => 11 # Other Neutrals
|
|---|
| 206 | );
|
|---|
| 207 |
|
|---|
| 208 | my @cp2uni = ();
|
|---|
| 209 | my @lead_bytes = ();
|
|---|
| 210 | my @uni2cp = ();
|
|---|
| 211 | my @unicode_defaults = ();
|
|---|
| 212 | my @unicode_aliases = ();
|
|---|
| 213 | my @tolower_table = ();
|
|---|
| 214 | my @toupper_table = ();
|
|---|
| 215 | my @digitmap_table = ();
|
|---|
| 216 | my @compatmap_table = ();
|
|---|
| 217 | my @category_table = (0) x 65536;
|
|---|
| 218 | my @direction_table = ();
|
|---|
| 219 | my @decomp_table = ();
|
|---|
| 220 | my @compose_table = ();
|
|---|
| 221 |
|
|---|
| 222 |
|
|---|
| 223 | ################################################################
|
|---|
| 224 | # read in the defaults file
|
|---|
| 225 | sub READ_DEFAULTS($)
|
|---|
| 226 | {
|
|---|
| 227 | my $filename = shift;
|
|---|
| 228 | my $start;
|
|---|
| 229 |
|
|---|
| 230 | # first setup a few default mappings
|
|---|
| 231 |
|
|---|
| 232 | open DEFAULTS, "$filename" or die "Cannot open $filename";
|
|---|
| 233 | print "Loading $filename\n";
|
|---|
| 234 | while (<DEFAULTS>)
|
|---|
| 235 | {
|
|---|
| 236 | next if /^\#/; # skip comments
|
|---|
| 237 | next if /^$/; # skip empty lines
|
|---|
| 238 | if (/^(([0-9a-fA-F]+)(,[0-9a-fA-F]+)*)\s+([0-9a-fA-F]+|'.'|none)\s+(\#.*)?/)
|
|---|
| 239 | {
|
|---|
| 240 | my @src = map hex, split /,/,$1;
|
|---|
| 241 | my $dst = $4;
|
|---|
| 242 | my $comment = $5;
|
|---|
| 243 | if ($#src > 0) { push @unicode_aliases, \@src; }
|
|---|
| 244 | next if ($dst eq "none");
|
|---|
| 245 | $dst = ($dst =~ /\'.\'/) ? ord substr($dst,1,1) : hex $dst;
|
|---|
| 246 | foreach my $src (@src)
|
|---|
| 247 | {
|
|---|
| 248 | die "Duplicate value" if defined($unicode_defaults[$src]);
|
|---|
| 249 | $unicode_defaults[$src] = $dst;
|
|---|
| 250 | }
|
|---|
| 251 | next;
|
|---|
| 252 | }
|
|---|
| 253 | die "Unrecognized line $_\n";
|
|---|
| 254 | }
|
|---|
| 255 |
|
|---|
| 256 | # now build mappings from the decomposition field of the Unicode database
|
|---|
| 257 |
|
|---|
| 258 | open UNICODEDATA, "$UNICODEDATA" or die "Cannot open $UNICODEDATA";
|
|---|
| 259 | print "Loading $UNICODEDATA\n";
|
|---|
| 260 | while (<UNICODEDATA>)
|
|---|
| 261 | {
|
|---|
| 262 | # Decode the fields ...
|
|---|
| 263 | my ($code, $name, $cat, $comb, $bidi,
|
|---|
| 264 | $decomp, $dec, $dig, $num, $mirror,
|
|---|
| 265 | $oldname, $comment, $upper, $lower, $title) = split /;/;
|
|---|
| 266 | my $dst;
|
|---|
| 267 | my $src = hex $code;
|
|---|
| 268 |
|
|---|
| 269 | die "unknown category $cat" unless defined $categories{$cat};
|
|---|
| 270 | die "unknown directionality $bidi" unless defined $directions{$bidi};
|
|---|
| 271 |
|
|---|
| 272 | $category_table[$src] = $categories{$cat};
|
|---|
| 273 | $direction_table[$src] = $directions{$bidi};
|
|---|
| 274 |
|
|---|
| 275 | if ($lower ne "")
|
|---|
| 276 | {
|
|---|
| 277 | $tolower_table[$src] = hex $lower;
|
|---|
| 278 | $category_table[$src] |= $ctype{"upper"}|$ctype{"alpha"};
|
|---|
| 279 | }
|
|---|
| 280 | if ($upper ne "")
|
|---|
| 281 | {
|
|---|
| 282 | $toupper_table[$src] = hex $upper;
|
|---|
| 283 | $category_table[$src] |= $ctype{"lower"}|$ctype{"alpha"};
|
|---|
| 284 | }
|
|---|
| 285 | if ($dec ne "")
|
|---|
| 286 | {
|
|---|
| 287 | $category_table[$src] |= $ctype{"digit"};
|
|---|
| 288 | }
|
|---|
| 289 | if ($dig ne "")
|
|---|
| 290 | {
|
|---|
| 291 | $digitmap_table[$src] = ord $dig;
|
|---|
| 292 | }
|
|---|
| 293 |
|
|---|
| 294 | # copy the category and direction for everything between First/Last pairs
|
|---|
| 295 | if ($name =~ /, First>/) { $start = $src; }
|
|---|
| 296 | if ($name =~ /, Last>/)
|
|---|
| 297 | {
|
|---|
| 298 | while ($start < $src)
|
|---|
| 299 | {
|
|---|
| 300 | $category_table[$start] = $category_table[$src];
|
|---|
| 301 | $direction_table[$start] = $direction_table[$src];
|
|---|
| 302 | $start++;
|
|---|
| 303 | }
|
|---|
| 304 | }
|
|---|
| 305 |
|
|---|
| 306 | next if $decomp eq ""; # no decomposition, skip it
|
|---|
| 307 |
|
|---|
| 308 | if ($decomp =~ /^<([a-zA-Z]+)>\s+([0-9a-fA-F]+)$/)
|
|---|
| 309 | {
|
|---|
| 310 | # decomposition of the form "<foo> 1234" -> use char if type is known
|
|---|
| 311 | if (($src >= 0xf900 && $src < 0xfb00) || ($src >= 0xfe30 && $src < 0xfffd))
|
|---|
| 312 | {
|
|---|
| 313 | # Single char decomposition in the compatibility range
|
|---|
| 314 | $compatmap_table[$src] = hex $2;
|
|---|
| 315 | }
|
|---|
| 316 | next unless ($1 eq "font" ||
|
|---|
| 317 | $1 eq "noBreak" ||
|
|---|
| 318 | $1 eq "circle" ||
|
|---|
| 319 | $1 eq "super" ||
|
|---|
| 320 | $1 eq "sub" ||
|
|---|
| 321 | $1 eq "wide" ||
|
|---|
| 322 | $1 eq "narrow" ||
|
|---|
| 323 | $1 eq "compat" ||
|
|---|
| 324 | $1 eq "small");
|
|---|
| 325 | $dst = hex $2;
|
|---|
| 326 | }
|
|---|
| 327 | elsif ($decomp =~ /^<compat>\s+0020\s+([0-9a-fA-F]+)/)
|
|---|
| 328 | {
|
|---|
| 329 | # decomposition "<compat> 0020 1234" -> combining accent
|
|---|
| 330 | $dst = hex $1;
|
|---|
| 331 | }
|
|---|
| 332 | elsif ($decomp =~ /^([0-9a-fA-F]+)/)
|
|---|
| 333 | {
|
|---|
| 334 | # decomposition contains only char values without prefix -> use first char
|
|---|
| 335 | $dst = hex $1;
|
|---|
| 336 | $category_table[$src] |= $category_table[$dst] if defined $category_table[$dst];
|
|---|
| 337 | # store decomposition if it contains two chars
|
|---|
| 338 | if ($decomp =~ /^([0-9a-fA-F]+)\s+([0-9a-fA-F]+)$/)
|
|---|
| 339 | {
|
|---|
| 340 | $decomp_table[$src] = [ hex $1, hex $2 ];
|
|---|
| 341 | push @compose_table, [ hex $1, hex $2, $src ];
|
|---|
| 342 | }
|
|---|
| 343 | elsif ($decomp =~ /^(<[a-z]+>\s)*([0-9a-fA-F]+)$/ &&
|
|---|
| 344 | (($src >= 0xf900 && $src < 0xfb00) || ($src >= 0xfe30 && $src < 0xfffd)))
|
|---|
| 345 | {
|
|---|
| 346 | # Single char decomposition in the compatibility range
|
|---|
| 347 | $compatmap_table[$src] = hex $2;
|
|---|
| 348 | }
|
|---|
| 349 | }
|
|---|
| 350 | else
|
|---|
| 351 | {
|
|---|
| 352 | next;
|
|---|
| 353 | }
|
|---|
| 354 |
|
|---|
| 355 | next if defined($unicode_defaults[$src]); # may have been set in the defaults file
|
|---|
| 356 |
|
|---|
| 357 | # check for loops
|
|---|
| 358 | for (my $i = $dst; ; $i = $unicode_defaults[$i])
|
|---|
| 359 | {
|
|---|
| 360 | die sprintf("loop detected for %04x -> %04x",$src,$dst) if $i == $src;
|
|---|
| 361 | last unless defined($unicode_defaults[$i]);
|
|---|
| 362 | }
|
|---|
| 363 | $unicode_defaults[$src] = $dst;
|
|---|
| 364 | }
|
|---|
| 365 |
|
|---|
| 366 | # patch the category of some special characters
|
|---|
| 367 |
|
|---|
| 368 | foreach my $cat (keys %special_categories)
|
|---|
| 369 | {
|
|---|
| 370 | my $flag = $ctype{$cat};
|
|---|
| 371 | foreach my $i (@{$special_categories{$cat}}) { $category_table[$i] |= $flag; }
|
|---|
| 372 | }
|
|---|
| 373 | }
|
|---|
| 374 |
|
|---|
| 375 |
|
|---|
| 376 | ################################################################
|
|---|
| 377 | # parse the input file
|
|---|
| 378 | sub READ_FILE($)
|
|---|
| 379 | {
|
|---|
| 380 | my $name = shift;
|
|---|
| 381 | open INPUT,$name or die "Cannot open $name";
|
|---|
| 382 |
|
|---|
| 383 | while (<INPUT>)
|
|---|
| 384 | {
|
|---|
| 385 | next if /^\#/; # skip comments
|
|---|
| 386 | next if /^$/; # skip empty lines
|
|---|
| 387 | next if /\x1a/; # skip ^Z
|
|---|
| 388 | next if (/^0x([0-9a-fA-F]+)\s+\#UNDEFINED/); # undefined char
|
|---|
| 389 |
|
|---|
| 390 | if (/^0x([0-9a-fA-F]+)\s+\#DBCS LEAD BYTE/)
|
|---|
| 391 | {
|
|---|
| 392 | my $cp = hex $1;
|
|---|
| 393 | push @lead_bytes,$cp;
|
|---|
| 394 | $cp2uni[$cp] = 0;
|
|---|
| 395 | next;
|
|---|
| 396 | }
|
|---|
| 397 | if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\s+(\#.*)?/)
|
|---|
| 398 | {
|
|---|
| 399 | my $cp = hex $1;
|
|---|
| 400 | my $uni = hex $2;
|
|---|
| 401 | $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
|
|---|
| 402 | $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
|
|---|
| 403 | if ($cp > 0xff && !defined($cp2uni[$cp >> 8]))
|
|---|
| 404 | {
|
|---|
| 405 | push @lead_bytes,$cp >> 8;
|
|---|
| 406 | $cp2uni[$cp >> 8] = 0;
|
|---|
| 407 | }
|
|---|
| 408 | next;
|
|---|
| 409 | }
|
|---|
| 410 | die "$name: Unrecognized line $_\n";
|
|---|
| 411 | }
|
|---|
| 412 | }
|
|---|
| 413 |
|
|---|
| 414 |
|
|---|
| 415 | ################################################################
|
|---|
| 416 | # fill input data for the 20127 (us-ascii) codepage
|
|---|
| 417 | sub fill_20127_codepage()
|
|---|
| 418 | {
|
|---|
| 419 | for (my $i = 0; $i < 128; $i++) { $cp2uni[$i] = $uni2cp[$i] = $i; }
|
|---|
| 420 | for (my $i = 128; $i < 256; $i++) { $cp2uni[$i] = $i & 0x7f; }
|
|---|
| 421 | }
|
|---|
| 422 |
|
|---|
| 423 | ################################################################
|
|---|
| 424 | # get a mapping including glyph chars for MB_USEGLYPHCHARS
|
|---|
| 425 |
|
|---|
| 426 | sub get_glyphs_mapping(@)
|
|---|
| 427 | {
|
|---|
| 428 | $_[0x01] = 0x263a; # (WHITE SMILING FACE)
|
|---|
| 429 | $_[0x02] = 0x263b; # (BLACK SMILING FACE)
|
|---|
| 430 | $_[0x03] = 0x2665; # (BLACK HEART SUIT)
|
|---|
| 431 | $_[0x04] = 0x2666; # (BLACK DIAMOND SUIT)
|
|---|
| 432 | $_[0x05] = 0x2663; # (BLACK CLUB SUIT)
|
|---|
| 433 | $_[0x06] = 0x2660; # (BLACK SPADE SUIT)
|
|---|
| 434 | $_[0x07] = 0x2022; # (BULLET)
|
|---|
| 435 | $_[0x08] = 0x25d8; # (INVERSE BULLET)
|
|---|
| 436 | $_[0x09] = 0x25cb; # (WHITE CIRCLE)
|
|---|
| 437 | $_[0x0a] = 0x25d9; # (INVERSE WHITE CIRCLE)
|
|---|
| 438 | $_[0x0b] = 0x2642; # (MALE SIGN)
|
|---|
| 439 | $_[0x0c] = 0x2640; # (FEMALE SIGN)
|
|---|
| 440 | $_[0x0d] = 0x266a; # (EIGHTH NOTE)
|
|---|
| 441 | $_[0x0e] = 0x266b; # (BEAMED EIGHTH NOTES)
|
|---|
| 442 | $_[0x0f] = 0x263c; # (WHITE SUN WITH RAYS)
|
|---|
| 443 | $_[0x10] = 0x25ba; # (BLACK RIGHT-POINTING POINTER)
|
|---|
| 444 | $_[0x11] = 0x25c4; # (BLACK LEFT-POINTING POINTER)
|
|---|
| 445 | $_[0x12] = 0x2195; # (UP DOWN ARROW)
|
|---|
| 446 | $_[0x13] = 0x203c; # (DOUBLE EXCLAMATION MARK)
|
|---|
| 447 | $_[0x14] = 0x00b6; # (PILCROW SIGN)
|
|---|
| 448 | $_[0x15] = 0x00a7; # (SECTION SIGN)
|
|---|
| 449 | $_[0x16] = 0x25ac; # (BLACK RECTANGLE)
|
|---|
| 450 | $_[0x17] = 0x21a8; # (UP DOWN ARROW WITH BASE)
|
|---|
| 451 | $_[0x18] = 0x2191; # (UPWARDS ARROW)
|
|---|
| 452 | $_[0x19] = 0x2193; # (DOWNWARDS ARROW)
|
|---|
| 453 | $_[0x1a] = 0x2192; # (RIGHTWARDS ARROW)
|
|---|
| 454 | $_[0x1b] = 0x2190; # (LEFTWARDS ARROW)
|
|---|
| 455 | $_[0x1c] = 0x221f; # (RIGHT ANGLE)
|
|---|
| 456 | $_[0x1d] = 0x2194; # (LEFT RIGHT ARROW)
|
|---|
| 457 | $_[0x1e] = 0x25b2; # (BLACK UP-POINTING TRIANGLE)
|
|---|
| 458 | $_[0x1f] = 0x25bc; # (BLACK DOWN-POINTING TRIANGLE)
|
|---|
| 459 | $_[0x7f] = 0x2302; # (HOUSE)
|
|---|
| 460 | return @_;
|
|---|
| 461 | }
|
|---|
| 462 |
|
|---|
| 463 | ################################################################
|
|---|
| 464 | # build EUC-JP table from the JIS 0208 file
|
|---|
| 465 | # FIXME: for proper EUC-JP we should probably read JIS 0212 too
|
|---|
| 466 | # but this would require 3-byte DBCS characters
|
|---|
| 467 | sub READ_JIS0208_FILE($)
|
|---|
| 468 | {
|
|---|
| 469 | my $name = shift;
|
|---|
| 470 |
|
|---|
| 471 | # ASCII chars
|
|---|
| 472 | for (my $i = 0x00; $i <= 0x7f; $i++)
|
|---|
| 473 | {
|
|---|
| 474 | $cp2uni[$i] = $i;
|
|---|
| 475 | $uni2cp[$i] = $i;
|
|---|
| 476 | }
|
|---|
| 477 |
|
|---|
| 478 | # JIS X 0201 right plane
|
|---|
| 479 | for (my $i = 0xa1; $i <= 0xdf; $i++)
|
|---|
| 480 | {
|
|---|
| 481 | $cp2uni[0x8e00 + $i] = 0xfec0 + $i;
|
|---|
| 482 | $uni2cp[0xfec0 + $i] = 0x8e00 + $i;
|
|---|
| 483 | }
|
|---|
| 484 |
|
|---|
| 485 | # lead bytes
|
|---|
| 486 | foreach my $i (0x8e, 0x8f, 0xa1 .. 0xfe)
|
|---|
| 487 | {
|
|---|
| 488 | push @lead_bytes,$i;
|
|---|
| 489 | $cp2uni[$i] = 0;
|
|---|
| 490 | }
|
|---|
| 491 |
|
|---|
| 492 | # undefined chars
|
|---|
| 493 | foreach my $i (0x80 .. 0x8d, 0x90 .. 0xa0, 0xff)
|
|---|
| 494 | {
|
|---|
| 495 | $cp2uni[$i] = $DEF_CHAR;
|
|---|
| 496 | }
|
|---|
| 497 |
|
|---|
| 498 | # Shift-JIS compatibility
|
|---|
| 499 | $uni2cp[0x00a5] = 0x5c;
|
|---|
| 500 | $uni2cp[0x203e] = 0x7e;
|
|---|
| 501 |
|
|---|
| 502 | # Fix backslash conversion
|
|---|
| 503 | $cp2uni[0xa1c0] = 0xff3c;
|
|---|
| 504 | $uni2cp[0xff3c] = 0xa1c0;
|
|---|
| 505 |
|
|---|
| 506 | open INPUT, "$name" or die "Cannot open $name";
|
|---|
| 507 | while (<INPUT>)
|
|---|
| 508 | {
|
|---|
| 509 | next if /^\#/; # skip comments
|
|---|
| 510 | next if /^$/; # skip empty lines
|
|---|
| 511 | next if /\x1a/; # skip ^Z
|
|---|
| 512 | if (/^0x[0-9a-fA-F]+\s+0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\s+(\#.*)?/)
|
|---|
| 513 | {
|
|---|
| 514 | my $cp = 0x8080 + hex $1;
|
|---|
| 515 | my $uni = hex $2;
|
|---|
| 516 | $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
|
|---|
| 517 | $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
|
|---|
| 518 | next;
|
|---|
| 519 | }
|
|---|
| 520 | die "$name: Unrecognized line $_\n";
|
|---|
| 521 | }
|
|---|
| 522 | }
|
|---|
| 523 |
|
|---|
| 524 |
|
|---|
| 525 | ################################################################
|
|---|
| 526 | # build the sort keys table
|
|---|
| 527 | sub READ_SORTKEYS_FILE()
|
|---|
| 528 | {
|
|---|
| 529 | my @sortkeys = ();
|
|---|
| 530 | for (my $i = 0; $i < 65536; $i++) { $sortkeys[$i] = [ -1, 0, 0, 0, 0 ] };
|
|---|
| 531 |
|
|---|
| 532 | open INPUT, "$SORTKEYS" or die "Cannot open $SORTKEYS";
|
|---|
| 533 | print "Loading $SORTKEYS\n";
|
|---|
| 534 | while (<INPUT>)
|
|---|
| 535 | {
|
|---|
| 536 | next if /^\#/; # skip comments
|
|---|
| 537 | next if /^$/; # skip empty lines
|
|---|
| 538 | next if /\x1a/; # skip ^Z
|
|---|
| 539 | next if /^\@version/; # skip @version header
|
|---|
| 540 | if (/^([0-9a-fA-F]+)\s+;\s+\[([*.])([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]+)\]/)
|
|---|
| 541 | {
|
|---|
| 542 | my ($uni,$variable) = (hex $1, $2);
|
|---|
| 543 | next if $uni > 65535;
|
|---|
| 544 | $sortkeys[$uni] = [ $uni, hex $3, hex $4, hex $5, hex $6 ];
|
|---|
| 545 | next;
|
|---|
| 546 | }
|
|---|
| 547 | if (/^([0-9a-fA-F]+\s+)+;\s+\[[*.]([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]+)\]/)
|
|---|
| 548 | {
|
|---|
| 549 | # multiple character sequence, ignored for now
|
|---|
| 550 | next;
|
|---|
| 551 | }
|
|---|
| 552 | die "$SORTKEYS: Unrecognized line $_\n";
|
|---|
| 553 | }
|
|---|
| 554 | close INPUT;
|
|---|
| 555 |
|
|---|
| 556 | # compress the keys to 32 bit:
|
|---|
| 557 | # key 1 to 16 bits, key 2 to 8 bits, key 3 to 4 bits, key 4 to 1 bit
|
|---|
| 558 |
|
|---|
| 559 | @sortkeys = sort { ${$a}[1] <=> ${$b}[1] or
|
|---|
| 560 | ${$a}[2] <=> ${$b}[2] or
|
|---|
| 561 | ${$a}[3] <=> ${$b}[3] or
|
|---|
| 562 | ${$a}[4] <=> ${$b}[4] or
|
|---|
| 563 | $a cmp $b; } @sortkeys;
|
|---|
| 564 |
|
|---|
| 565 | my ($n2, $n3) = (1, 1);
|
|---|
| 566 | my @keys = (-1, -1, -1, -1, -1 );
|
|---|
| 567 | my @flatkeys = ();
|
|---|
| 568 |
|
|---|
| 569 | for (my $i = 0; $i < 65536; $i++)
|
|---|
| 570 | {
|
|---|
| 571 | my @current = @{$sortkeys[$i]};
|
|---|
| 572 | next if $current[0] == -1;
|
|---|
| 573 | if ($current[1] == $keys[1])
|
|---|
| 574 | {
|
|---|
| 575 | if ($current[2] == $keys[2])
|
|---|
| 576 | {
|
|---|
| 577 | if ($current[3] == $keys[3])
|
|---|
| 578 | {
|
|---|
| 579 | # nothing
|
|---|
| 580 | }
|
|---|
| 581 | else
|
|---|
| 582 | {
|
|---|
| 583 | $keys[3] = $current[3];
|
|---|
| 584 | $n3++;
|
|---|
| 585 | die if ($n3 >= 16);
|
|---|
| 586 | }
|
|---|
| 587 | }
|
|---|
| 588 | else
|
|---|
| 589 | {
|
|---|
| 590 | $keys[2] = $current[2];
|
|---|
| 591 | $keys[3] = $current[3];
|
|---|
| 592 | $n2++;
|
|---|
| 593 | $n3 = 1;
|
|---|
| 594 | die if ($n2 >= 256);
|
|---|
| 595 | }
|
|---|
| 596 | }
|
|---|
| 597 | else
|
|---|
| 598 | {
|
|---|
| 599 | $keys[1] = $current[1];
|
|---|
| 600 | $keys[2] = $current[2];
|
|---|
| 601 | $keys[3] = $current[3];
|
|---|
| 602 | $n2 = 1;
|
|---|
| 603 | $n3 = 1;
|
|---|
| 604 | }
|
|---|
| 605 |
|
|---|
| 606 | if ($current[2]) { $current[2] = $n2; }
|
|---|
| 607 | if ($current[3]) { $current[3] = $n3; }
|
|---|
| 608 | if ($current[4]) { $current[4] = 1; }
|
|---|
| 609 |
|
|---|
| 610 | $flatkeys[$current[0]] = ($current[1] << 16) | ($current[2] << 8) | ($current[3] << 4) | $current[4];
|
|---|
| 611 | }
|
|---|
| 612 | return @flatkeys;
|
|---|
| 613 | }
|
|---|
| 614 |
|
|---|
| 615 |
|
|---|
| 616 | ################################################################
|
|---|
| 617 | # build the sort keys table
|
|---|
| 618 | sub DUMP_SORTKEYS($@)
|
|---|
| 619 | {
|
|---|
| 620 | my ($filename, @keys) = @_;
|
|---|
| 621 |
|
|---|
| 622 | # count the number of 256-key ranges that contain something
|
|---|
| 623 |
|
|---|
| 624 | my @offsets = ();
|
|---|
| 625 | my $ranges = 2;
|
|---|
| 626 | for (my $i = 0; $i < 256; $i++) { $offsets[$i] = 256; }
|
|---|
| 627 | for (my $i = 0; $i < 65536; $i++)
|
|---|
| 628 | {
|
|---|
| 629 | next unless defined $keys[$i];
|
|---|
| 630 | $offsets[$i >> 8] = $ranges * 256;
|
|---|
| 631 | $ranges++;
|
|---|
| 632 | $i |= 255;
|
|---|
| 633 | }
|
|---|
| 634 |
|
|---|
| 635 | # output the range offsets
|
|---|
| 636 |
|
|---|
| 637 | open OUTPUT,">$filename.new" or die "Cannot create $filename";
|
|---|
| 638 | printf "Building $filename\n";
|
|---|
| 639 | printf OUTPUT "/* Unicode collation element table */\n";
|
|---|
| 640 | printf OUTPUT "/* generated from %s */\n", $SORTKEYS;
|
|---|
| 641 | printf OUTPUT "/* DO NOT EDIT!! */\n\n";
|
|---|
| 642 |
|
|---|
| 643 | printf OUTPUT "const unsigned int collation_table[%d] =\n{\n", $ranges*256;
|
|---|
| 644 | printf OUTPUT " /* index */\n";
|
|---|
| 645 | printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%08x", 0, @offsets );
|
|---|
| 646 |
|
|---|
| 647 | # output the default values
|
|---|
| 648 |
|
|---|
| 649 | printf OUTPUT " /* defaults */\n";
|
|---|
| 650 | printf OUTPUT "%s", DUMP_ARRAY( "0x%08x", 0, (0xffffffff) x 256 );
|
|---|
| 651 |
|
|---|
| 652 | # output all the key ranges
|
|---|
| 653 |
|
|---|
| 654 | for (my $i = 0; $i < 256; $i++)
|
|---|
| 655 | {
|
|---|
| 656 | next if $offsets[$i] == 256;
|
|---|
| 657 | printf OUTPUT ",\n /* 0x%02x00 .. 0x%02xff */\n", $i, $i;
|
|---|
| 658 | printf OUTPUT "%s", DUMP_ARRAY( "0x%08x", 0xffffffff, @keys[($i<<8) .. ($i<<8)+255] );
|
|---|
| 659 | }
|
|---|
| 660 | printf OUTPUT "\n};\n";
|
|---|
| 661 | close OUTPUT;
|
|---|
| 662 | save_file($filename);
|
|---|
| 663 | }
|
|---|
| 664 |
|
|---|
| 665 |
|
|---|
| 666 | ################################################################
|
|---|
| 667 | # add default mappings once the file had been read
|
|---|
| 668 | sub ADD_DEFAULT_MAPPINGS()
|
|---|
| 669 | {
|
|---|
| 670 | # Apply aliases
|
|---|
| 671 |
|
|---|
| 672 | foreach my $alias (@unicode_aliases)
|
|---|
| 673 | {
|
|---|
| 674 | my $target = undef;
|
|---|
| 675 | foreach my $src (@$alias)
|
|---|
| 676 | {
|
|---|
| 677 | if (defined($uni2cp[$src]))
|
|---|
| 678 | {
|
|---|
| 679 | $target = $uni2cp[$src];
|
|---|
| 680 | last;
|
|---|
| 681 | }
|
|---|
| 682 | }
|
|---|
| 683 | next unless defined($target);
|
|---|
| 684 |
|
|---|
| 685 | # At least one char of the alias set is defined, set the others to the same value
|
|---|
| 686 | foreach my $src (@$alias)
|
|---|
| 687 | {
|
|---|
| 688 | $uni2cp[$src] = $target unless defined($uni2cp[$src]);
|
|---|
| 689 | }
|
|---|
| 690 | }
|
|---|
| 691 |
|
|---|
| 692 | # For every src -> target mapping in the defaults table,
|
|---|
| 693 | # make uni2cp[src] = uni2cp[target] if uni2cp[target] is defined
|
|---|
| 694 |
|
|---|
| 695 | for (my $src = 0; $src < 65536; $src++)
|
|---|
| 696 | {
|
|---|
| 697 | next if defined($uni2cp[$src]); # source has a definition already
|
|---|
| 698 | next unless defined($unicode_defaults[$src]); # no default for this char
|
|---|
| 699 | my $target = $unicode_defaults[$src];
|
|---|
| 700 |
|
|---|
| 701 | # do a recursive mapping until we find a target char that is defined
|
|---|
| 702 | while (!defined($uni2cp[$target]) &&
|
|---|
| 703 | defined($unicode_defaults[$target])) { $target = $unicode_defaults[$target]; }
|
|---|
| 704 |
|
|---|
| 705 | if (defined($uni2cp[$target])) { $uni2cp[$src] = $uni2cp[$target]; }
|
|---|
| 706 | }
|
|---|
| 707 |
|
|---|
| 708 | # Add an identity mapping for all undefined chars
|
|---|
| 709 |
|
|---|
| 710 | for (my $i = 0; $i < 256; $i++)
|
|---|
| 711 | {
|
|---|
| 712 | next if defined($cp2uni[$i]);
|
|---|
| 713 | next if defined($uni2cp[$i]);
|
|---|
| 714 | $cp2uni[$i] = $uni2cp[$i] = $i;
|
|---|
| 715 | }
|
|---|
| 716 | }
|
|---|
| 717 |
|
|---|
| 718 | ################################################################
|
|---|
| 719 | # dump an array of integers
|
|---|
| 720 | sub DUMP_ARRAY($$@)
|
|---|
| 721 | {
|
|---|
| 722 | my ($format,$default,@array) = @_;
|
|---|
| 723 | my $i;
|
|---|
| 724 | my $ret = " ";
|
|---|
| 725 | for ($i = 0; $i < $#array; $i++)
|
|---|
| 726 | {
|
|---|
| 727 | $ret .= sprintf($format, defined $array[$i] ? $array[$i] : $default);
|
|---|
| 728 | $ret .= (($i % 8) != 7) ? ", " : ",\n ";
|
|---|
| 729 | }
|
|---|
| 730 | $ret .= sprintf($format, defined $array[$i] ? $array[$i] : $default);
|
|---|
| 731 | return $ret;
|
|---|
| 732 | }
|
|---|
| 733 |
|
|---|
| 734 | ################################################################
|
|---|
| 735 | # dump an SBCS mapping table
|
|---|
| 736 | sub dump_sbcs_table($$$$$)
|
|---|
| 737 | {
|
|---|
| 738 | my ($codepage, $has_glyphs, $name, $def, $defw) = @_;
|
|---|
| 739 | my $i;
|
|---|
| 740 |
|
|---|
| 741 | # output the ascii->unicode table
|
|---|
| 742 |
|
|---|
| 743 | if ($has_glyphs)
|
|---|
| 744 | {
|
|---|
| 745 | printf OUTPUT "static const WCHAR cp2uni[512] =\n";
|
|---|
| 746 | printf OUTPUT "{\n%s", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[0 .. 255] );
|
|---|
| 747 | printf OUTPUT ",\n /* glyphs */\n%s\n};\n\n",
|
|---|
| 748 | DUMP_ARRAY( "0x%04x", $defw, get_glyphs_mapping(@cp2uni[0 .. 255]) );
|
|---|
| 749 | }
|
|---|
| 750 | else
|
|---|
| 751 | {
|
|---|
| 752 | printf OUTPUT "static const WCHAR cp2uni[256] =\n";
|
|---|
| 753 | printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[0 .. 255] );
|
|---|
| 754 | }
|
|---|
| 755 |
|
|---|
| 756 | # count the number of unicode->ascii subtables that contain something
|
|---|
| 757 |
|
|---|
| 758 | my @filled = ();
|
|---|
| 759 | my $subtables = 1;
|
|---|
| 760 | for (my $i = 0; $i < 65536; $i++)
|
|---|
| 761 | {
|
|---|
| 762 | next unless defined $uni2cp[$i];
|
|---|
| 763 | $filled[$i >> 8] = 1;
|
|---|
| 764 | $subtables++;
|
|---|
| 765 | $i |= 255;
|
|---|
| 766 | }
|
|---|
| 767 |
|
|---|
| 768 | # output all the subtables into a single array
|
|---|
| 769 |
|
|---|
| 770 | printf OUTPUT "static const unsigned char uni2cp_low[%d] =\n{\n", $subtables*256;
|
|---|
| 771 | for (my $i = 0; $i < 256; $i++)
|
|---|
| 772 | {
|
|---|
| 773 | next unless $filled[$i];
|
|---|
| 774 | printf OUTPUT " /* 0x%02x00 .. 0x%02xff */\n", $i, $i;
|
|---|
| 775 | printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%02x", $def, @uni2cp[($i<<8) .. ($i<<8)+255] );
|
|---|
| 776 | }
|
|---|
| 777 | printf OUTPUT " /* defaults */\n";
|
|---|
| 778 | printf OUTPUT "%s\n};\n\n", DUMP_ARRAY( "0x%02x", 0, ($def) x 256 );
|
|---|
| 779 |
|
|---|
| 780 | # output a table of the offsets of the subtables in the previous array
|
|---|
| 781 |
|
|---|
| 782 | my $pos = 0;
|
|---|
| 783 | my @offsets = ();
|
|---|
| 784 | for (my $i = 0; $i < 256; $i++)
|
|---|
| 785 | {
|
|---|
| 786 | if ($filled[$i]) { push @offsets, $pos; $pos += 256; }
|
|---|
| 787 | else { push @offsets, ($subtables-1) * 256; }
|
|---|
| 788 | }
|
|---|
| 789 | printf OUTPUT "static const unsigned short uni2cp_high[256] =\n";
|
|---|
| 790 | printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%04x", 0, @offsets );
|
|---|
| 791 |
|
|---|
| 792 | # output the code page descriptor
|
|---|
| 793 |
|
|---|
| 794 | printf OUTPUT "const struct sbcs_table cptable_%03d =\n{\n", $codepage;
|
|---|
| 795 | printf OUTPUT " { %d, 1, 0x%04x, 0x%04x, \"%s\" },\n",
|
|---|
| 796 | $codepage, $def, $defw, $name;
|
|---|
| 797 | printf OUTPUT " cp2uni,\n";
|
|---|
| 798 | if ($has_glyphs) { printf OUTPUT " cp2uni + 256,\n"; }
|
|---|
| 799 | else { printf OUTPUT " cp2uni,\n"; }
|
|---|
| 800 | printf OUTPUT " uni2cp_low,\n";
|
|---|
| 801 | printf OUTPUT " uni2cp_high\n};\n";
|
|---|
| 802 | }
|
|---|
| 803 |
|
|---|
| 804 |
|
|---|
| 805 | ################################################################
|
|---|
| 806 | # dump a DBCS mapping table
|
|---|
| 807 | sub dump_dbcs_table($$$$@)
|
|---|
| 808 | {
|
|---|
| 809 | my ($codepage, $name, $def, $defw, @lb_ranges) = @_;
|
|---|
| 810 |
|
|---|
| 811 | # build a list of lead bytes that are actually used
|
|---|
| 812 |
|
|---|
| 813 | my @lblist = ();
|
|---|
| 814 | LBLOOP: for (my $y = 0; $y <= $#lead_bytes; $y++)
|
|---|
| 815 | {
|
|---|
| 816 | my $base = $lead_bytes[$y] << 8;
|
|---|
| 817 | for (my $x = 0; $x < 256; $x++)
|
|---|
| 818 | {
|
|---|
| 819 | if (defined $cp2uni[$base+$x])
|
|---|
| 820 | {
|
|---|
| 821 | push @lblist,$lead_bytes[$y];
|
|---|
| 822 | next LBLOOP;
|
|---|
| 823 | }
|
|---|
| 824 | }
|
|---|
| 825 | }
|
|---|
| 826 | my $unused = ($#lead_bytes > $#lblist);
|
|---|
| 827 |
|
|---|
| 828 | # output the ascii->unicode table for the single byte chars
|
|---|
| 829 |
|
|---|
| 830 | printf OUTPUT "static const WCHAR cp2uni[%d] =\n", 256 * ($#lblist + 2 + $unused);
|
|---|
| 831 | printf OUTPUT "{\n%s,\n", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[0 .. 255] );
|
|---|
| 832 |
|
|---|
| 833 | # output the default table for unused lead bytes
|
|---|
| 834 |
|
|---|
| 835 | if ($unused)
|
|---|
| 836 | {
|
|---|
| 837 | printf OUTPUT " /* unused lead bytes */\n";
|
|---|
| 838 | printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%04x", 0, ($defw) x 256 );
|
|---|
| 839 | }
|
|---|
| 840 |
|
|---|
| 841 | # output the ascii->unicode table for each DBCS lead byte
|
|---|
| 842 |
|
|---|
| 843 | for (my $y = 0; $y <= $#lblist; $y++)
|
|---|
| 844 | {
|
|---|
| 845 | my $base = $lblist[$y] << 8;
|
|---|
| 846 | printf OUTPUT " /* lead byte %02x */\n", $lblist[$y];
|
|---|
| 847 | printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[$base .. $base+255] );
|
|---|
| 848 | printf OUTPUT ($y < $#lblist) ? ",\n" : "\n};\n\n";
|
|---|
| 849 | }
|
|---|
| 850 |
|
|---|
| 851 | # output the lead byte subtables offsets
|
|---|
| 852 |
|
|---|
| 853 | my @offsets = ();
|
|---|
| 854 | for (my $x = 0; $x < 256; $x++) { $offsets[$x] = 0; }
|
|---|
| 855 | for (my $x = 0; $x <= $#lblist; $x++) { $offsets[$lblist[$x]] = $x + 1; }
|
|---|
| 856 | if ($unused)
|
|---|
| 857 | {
|
|---|
| 858 | # increment all lead bytes offset to take into account the unused table
|
|---|
| 859 | for (my $x = 0; $x <= $#lead_bytes; $x++) { $offsets[$lead_bytes[$x]]++; }
|
|---|
| 860 | }
|
|---|
| 861 | printf OUTPUT "static const unsigned char cp2uni_leadbytes[256] =\n";
|
|---|
| 862 | printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%02x", 0, @offsets );
|
|---|
| 863 |
|
|---|
| 864 | # count the number of unicode->ascii subtables that contain something
|
|---|
| 865 |
|
|---|
| 866 | my @filled = ();
|
|---|
| 867 | my $subtables = 1;
|
|---|
| 868 | for (my $i = 0; $i < 65536; $i++)
|
|---|
| 869 | {
|
|---|
| 870 | next unless defined $uni2cp[$i];
|
|---|
| 871 | $filled[$i >> 8] = 1;
|
|---|
| 872 | $subtables++;
|
|---|
| 873 | $i |= 255;
|
|---|
| 874 | }
|
|---|
| 875 |
|
|---|
| 876 | # output all the subtables into a single array
|
|---|
| 877 |
|
|---|
| 878 | printf OUTPUT "static const unsigned short uni2cp_low[%d] =\n{\n", $subtables*256;
|
|---|
| 879 | for (my $y = 0; $y < 256; $y++)
|
|---|
| 880 | {
|
|---|
| 881 | next unless $filled[$y];
|
|---|
| 882 | printf OUTPUT " /* 0x%02x00 .. 0x%02xff */\n", $y, $y;
|
|---|
| 883 | printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%04x", $def, @uni2cp[($y<<8) .. ($y<<8)+255] );
|
|---|
| 884 | }
|
|---|
| 885 | printf OUTPUT " /* defaults */\n";
|
|---|
| 886 | printf OUTPUT "%s\n};\n\n", DUMP_ARRAY( "0x%04x", 0, ($def) x 256 );
|
|---|
| 887 |
|
|---|
| 888 | # output a table of the offsets of the subtables in the previous array
|
|---|
| 889 |
|
|---|
| 890 | my $pos = 0;
|
|---|
| 891 | @offsets = ();
|
|---|
| 892 | for (my $y = 0; $y < 256; $y++)
|
|---|
| 893 | {
|
|---|
| 894 | if ($filled[$y]) { push @offsets, $pos; $pos += 256; }
|
|---|
| 895 | else { push @offsets, ($subtables-1) * 256; }
|
|---|
| 896 | }
|
|---|
| 897 | printf OUTPUT "static const unsigned short uni2cp_high[256] =\n";
|
|---|
| 898 | printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%04x", 0, @offsets );
|
|---|
| 899 |
|
|---|
| 900 | # output the code page descriptor
|
|---|
| 901 |
|
|---|
| 902 | printf OUTPUT "const struct dbcs_table cptable_%03d =\n{\n", $codepage;
|
|---|
| 903 | printf OUTPUT " { %d, 2, 0x%04x, 0x%04x, \"%s\" },\n",
|
|---|
| 904 | $codepage, $def, $defw, $name;
|
|---|
| 905 | printf OUTPUT " cp2uni,\n";
|
|---|
| 906 | printf OUTPUT " cp2uni_leadbytes,\n";
|
|---|
| 907 | printf OUTPUT " uni2cp_low,\n";
|
|---|
| 908 | printf OUTPUT " uni2cp_high,\n";
|
|---|
| 909 | printf OUTPUT " {\n %s\n }\n", DUMP_ARRAY( "0x%02x", 0, @lb_ranges, 0, 0 );
|
|---|
| 910 | printf OUTPUT "};\n";
|
|---|
| 911 | }
|
|---|
| 912 |
|
|---|
| 913 |
|
|---|
| 914 | ################################################################
|
|---|
| 915 | # get the list of defined lead byte ranges
|
|---|
| 916 | sub get_lb_ranges()
|
|---|
| 917 | {
|
|---|
| 918 | my @list = ();
|
|---|
| 919 | my @ranges = ();
|
|---|
| 920 | my $i = 0;
|
|---|
| 921 | foreach $i (@lead_bytes) { $list[$i] = 1; }
|
|---|
| 922 | my $on = 0;
|
|---|
| 923 | for (my $i = 0; $i < 256; $i++)
|
|---|
| 924 | {
|
|---|
| 925 | if ($on)
|
|---|
| 926 | {
|
|---|
| 927 | if (!defined $list[$i]) { push @ranges, $i-1; $on = 0; }
|
|---|
| 928 | }
|
|---|
| 929 | else
|
|---|
| 930 | {
|
|---|
| 931 | if ($list[$i]) { push @ranges, $i; $on = 1; }
|
|---|
| 932 | }
|
|---|
| 933 | }
|
|---|
| 934 | if ($on) { push @ranges, 0xff; }
|
|---|
| 935 | return @ranges;
|
|---|
| 936 | }
|
|---|
| 937 |
|
|---|
| 938 |
|
|---|
| 939 | ################################################################
|
|---|
| 940 | # dump the case mapping tables
|
|---|
| 941 | sub DUMP_CASE_MAPPINGS($)
|
|---|
| 942 | {
|
|---|
| 943 | my $filename = shift;
|
|---|
| 944 | open OUTPUT,">$filename.new" or die "Cannot create $filename";
|
|---|
| 945 | printf "Building $filename\n";
|
|---|
| 946 | printf OUTPUT "/* Unicode case mappings */\n";
|
|---|
| 947 | printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
|
|---|
| 948 | printf OUTPUT "#include \"wine/unicode.h\"\n\n";
|
|---|
| 949 |
|
|---|
| 950 | DUMP_CASE_TABLE( "wine_casemap_lower", @tolower_table );
|
|---|
| 951 | DUMP_CASE_TABLE( "wine_casemap_upper", @toupper_table );
|
|---|
| 952 | DUMP_CASE_TABLE( "wine_digitmap", @digitmap_table );
|
|---|
| 953 | DUMP_CASE_TABLE( "wine_compatmap", @compatmap_table );
|
|---|
| 954 | close OUTPUT;
|
|---|
| 955 | save_file($filename);
|
|---|
| 956 | }
|
|---|
| 957 |
|
|---|
| 958 |
|
|---|
| 959 | ################################################################
|
|---|
| 960 | # dump a case mapping table
|
|---|
| 961 | sub DUMP_CASE_TABLE($@)
|
|---|
| 962 | {
|
|---|
| 963 | my ($name,@table) = @_;
|
|---|
| 964 |
|
|---|
| 965 | # count the number of sub tables that contain something
|
|---|
| 966 | # also compute the low and upper populated bounds
|
|---|
| 967 |
|
|---|
| 968 | my @lowerbounds = ( 0, 0 );
|
|---|
| 969 | my @upperbounds = ( 0, 255 );
|
|---|
| 970 | my $index = 0;
|
|---|
| 971 | my @filled = ();
|
|---|
| 972 | for (my $i = 0; $i < 65536; $i++)
|
|---|
| 973 | {
|
|---|
| 974 | next unless defined $table[$i];
|
|---|
| 975 | if (!defined $filled[$i >> 8])
|
|---|
| 976 | {
|
|---|
| 977 | $lowerbounds[$index] = $i & 0xff;
|
|---|
| 978 | $upperbounds[$index] = 0xff - $lowerbounds[$index];
|
|---|
| 979 | $filled[$i >> 8] = $index * 256 + 512;
|
|---|
| 980 | $index++;
|
|---|
| 981 | }
|
|---|
| 982 | else
|
|---|
| 983 | {
|
|---|
| 984 | $upperbounds[$index-1] = 0xff - ($i & 0xff);
|
|---|
| 985 | }
|
|---|
| 986 | $table[$i] = ($table[$i] - $i) & 0xffff;
|
|---|
| 987 | }
|
|---|
| 988 |
|
|---|
| 989 | # Collapse blocks upwards if possible
|
|---|
| 990 | my $removed = 0;
|
|---|
| 991 | $index = 0;
|
|---|
| 992 | for (my $i = 0; $i < 256; $i++)
|
|---|
| 993 | {
|
|---|
| 994 | next unless defined $filled[$i];
|
|---|
| 995 | if ($upperbounds[$index - 1] > $lowerbounds[$index])
|
|---|
| 996 | {
|
|---|
| 997 | $removed = $removed + $lowerbounds[$index];
|
|---|
| 998 | }
|
|---|
| 999 | else
|
|---|
| 1000 | {
|
|---|
| 1001 | $removed = $removed + $upperbounds[$index - 1];
|
|---|
| 1002 | $lowerbounds[$index] = $upperbounds[$index - 1];
|
|---|
| 1003 | }
|
|---|
| 1004 | $filled[$i] = $filled[$i] - $removed;
|
|---|
| 1005 | $index++;
|
|---|
| 1006 | }
|
|---|
| 1007 |
|
|---|
| 1008 | # dump the table
|
|---|
| 1009 |
|
|---|
| 1010 | printf OUTPUT "const WCHAR %s[%d] =\n", $name, $index * 256 + 512 - $removed;
|
|---|
| 1011 | printf OUTPUT "{\n /* index */\n";
|
|---|
| 1012 | printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%04x", 256, @filled );
|
|---|
| 1013 | printf OUTPUT " /* defaults */\n";
|
|---|
| 1014 | printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, (0) x 256 );
|
|---|
| 1015 | $index = 0;
|
|---|
| 1016 | for (my $i = 0; $i < 256; $i++)
|
|---|
| 1017 | {
|
|---|
| 1018 | next unless $filled[$i];
|
|---|
| 1019 | printf OUTPUT ",\n /* 0x%02x%02x .. 0x%02xff */\n", $i, $lowerbounds[$index], $i;
|
|---|
| 1020 | printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0,
|
|---|
| 1021 | @table[($i<<8) + $lowerbounds[$index] .. ($i<<8)+255] );
|
|---|
| 1022 | $index++;
|
|---|
| 1023 | }
|
|---|
| 1024 | printf OUTPUT "\n};\n";
|
|---|
| 1025 | }
|
|---|
| 1026 |
|
|---|
| 1027 |
|
|---|
| 1028 | ################################################################
|
|---|
| 1029 | # dump the ctype tables
|
|---|
| 1030 | sub DUMP_CTYPE_TABLES($)
|
|---|
| 1031 | {
|
|---|
| 1032 | my $filename = shift;
|
|---|
| 1033 | open OUTPUT,">$filename.new" or die "Cannot create $filename";
|
|---|
| 1034 | printf "Building $filename\n";
|
|---|
| 1035 | printf OUTPUT "/* Unicode ctype tables */\n";
|
|---|
| 1036 | printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
|
|---|
| 1037 | printf OUTPUT "#include \"wine/unicode.h\"\n\n";
|
|---|
| 1038 |
|
|---|
| 1039 | my @array = (0) x 256;
|
|---|
| 1040 | my %sequences;
|
|---|
| 1041 |
|
|---|
| 1042 | # add the direction in the high 4 bits of the category
|
|---|
| 1043 | for (my $i = 0; $i < 65536; $i++)
|
|---|
| 1044 | {
|
|---|
| 1045 | $category_table[$i] |= $direction_table[$i] << 12 if defined $direction_table[$i];
|
|---|
| 1046 | }
|
|---|
| 1047 |
|
|---|
| 1048 | # try to merge table rows
|
|---|
| 1049 | for (my $row = 0; $row < 256; $row++)
|
|---|
| 1050 | {
|
|---|
| 1051 | my $rowtxt = sprintf "%04x" x 256, @category_table[($row<<8)..($row<<8)+255];
|
|---|
| 1052 | if (defined($sequences{$rowtxt}))
|
|---|
| 1053 | {
|
|---|
| 1054 | # reuse an existing row
|
|---|
| 1055 | $array[$row] = $sequences{$rowtxt};
|
|---|
| 1056 | }
|
|---|
| 1057 | else
|
|---|
| 1058 | {
|
|---|
| 1059 | # create a new row
|
|---|
| 1060 | $sequences{$rowtxt} = $array[$row] = $#array + 1;
|
|---|
| 1061 | push @array, @category_table[($row<<8)..($row<<8)+255];
|
|---|
| 1062 | }
|
|---|
| 1063 | }
|
|---|
| 1064 |
|
|---|
| 1065 | printf OUTPUT "const unsigned short wine_wctype_table[%d] =\n{\n", $#array+1;
|
|---|
| 1066 | printf OUTPUT " /* offsets */\n%s,\n", DUMP_ARRAY( "0x%04x", 0, @array[0..255] );
|
|---|
| 1067 | printf OUTPUT " /* values */\n%s\n};\n", DUMP_ARRAY( "0x%04x", 0, @array[256..$#array] );
|
|---|
| 1068 |
|
|---|
| 1069 | close OUTPUT;
|
|---|
| 1070 | save_file($filename);
|
|---|
| 1071 | }
|
|---|
| 1072 |
|
|---|
| 1073 |
|
|---|
| 1074 | ################################################################
|
|---|
| 1075 | # dump the char composition tables
|
|---|
| 1076 | sub DUMP_COMPOSE_TABLES($)
|
|---|
| 1077 | {
|
|---|
| 1078 | my $filename = shift;
|
|---|
| 1079 |
|
|---|
| 1080 | open OUTPUT,">$filename.new" or die "Cannot create $filename";
|
|---|
| 1081 | printf "Building $filename\n";
|
|---|
| 1082 | printf OUTPUT "/* Unicode char composition */\n";
|
|---|
| 1083 | printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
|
|---|
| 1084 | printf OUTPUT "#include \"wine/unicode.h\"\n\n";
|
|---|
| 1085 |
|
|---|
| 1086 | ######### composition table
|
|---|
| 1087 |
|
|---|
| 1088 | my @filled = ();
|
|---|
| 1089 | foreach my $i (@compose_table)
|
|---|
| 1090 | {
|
|---|
| 1091 | my @comp = @$i;
|
|---|
| 1092 | push @{$filled[$comp[1]]}, [ $comp[0], $comp[2] ];
|
|---|
| 1093 | }
|
|---|
| 1094 |
|
|---|
| 1095 | # count how many different second chars we have
|
|---|
| 1096 |
|
|---|
| 1097 | my $count = 0;
|
|---|
| 1098 | for (my $i = 0; $i < 65536; $i++)
|
|---|
| 1099 | {
|
|---|
| 1100 | next unless defined $filled[$i];
|
|---|
| 1101 | $count++;
|
|---|
| 1102 | }
|
|---|
| 1103 |
|
|---|
| 1104 | # build the table of second chars and offsets
|
|---|
| 1105 |
|
|---|
| 1106 | my $pos = $count + 1;
|
|---|
| 1107 | my @table = ();
|
|---|
| 1108 | for (my $i = 0; $i < 65536; $i++)
|
|---|
| 1109 | {
|
|---|
| 1110 | next unless defined $filled[$i];
|
|---|
| 1111 | push @table, $i, $pos;
|
|---|
| 1112 | $pos += @{$filled[$i]};
|
|---|
| 1113 | }
|
|---|
| 1114 | # terminator with last position
|
|---|
| 1115 | push @table, 0, $pos;
|
|---|
| 1116 | printf OUTPUT "const WCHAR unicode_compose_table[0x%x] =\n{\n", 2*$pos;
|
|---|
| 1117 | printf OUTPUT " /* second chars + offsets */\n%s", DUMP_ARRAY( "0x%04x", 0, @table );
|
|---|
| 1118 |
|
|---|
| 1119 | # build the table of first chars and mappings
|
|---|
| 1120 |
|
|---|
| 1121 | for (my $i = 0; $i < 65536; $i++)
|
|---|
| 1122 | {
|
|---|
| 1123 | next unless defined $filled[$i];
|
|---|
| 1124 | my @table = ();
|
|---|
| 1125 | my @list = sort { $a->[0] <=> $b->[0] } @{$filled[$i]};
|
|---|
| 1126 | for (my $j = 0; $j <= $#list; $j++)
|
|---|
| 1127 | {
|
|---|
| 1128 | push @table, $list[$j][0], $list[$j][1];
|
|---|
| 1129 | }
|
|---|
| 1130 | printf OUTPUT ",\n /* 0x%04x */\n%s", $i, DUMP_ARRAY( "0x%04x", 0, @table );
|
|---|
| 1131 | }
|
|---|
| 1132 | printf OUTPUT "\n};\n\nconst unsigned int unicode_compose_table_size = %d;\n\n", $count;
|
|---|
| 1133 |
|
|---|
| 1134 | ######### decomposition table
|
|---|
| 1135 |
|
|---|
| 1136 | # first determine all the 16-char subsets that contain something
|
|---|
| 1137 |
|
|---|
| 1138 | @filled = (0) x 4096;
|
|---|
| 1139 | $pos = 16*2; # for the null subset
|
|---|
| 1140 | for (my $i = 0; $i < 65536; $i++)
|
|---|
| 1141 | {
|
|---|
| 1142 | next unless defined $decomp_table[$i];
|
|---|
| 1143 | $filled[$i >> 4] = $pos;
|
|---|
| 1144 | $pos += 16*2;
|
|---|
| 1145 | $i |= 15;
|
|---|
| 1146 | }
|
|---|
| 1147 | my $total = $pos;
|
|---|
| 1148 |
|
|---|
| 1149 | # now count the 256-char subsets that contain something
|
|---|
| 1150 |
|
|---|
| 1151 | my @filled_idx = (256) x 256;
|
|---|
| 1152 | $pos = 256 + 16;
|
|---|
| 1153 | for (my $i = 0; $i < 4096; $i++)
|
|---|
| 1154 | {
|
|---|
| 1155 | next unless $filled[$i];
|
|---|
| 1156 | $filled_idx[$i >> 4] = $pos;
|
|---|
| 1157 | $pos += 16;
|
|---|
| 1158 | $i |= 15;
|
|---|
| 1159 | }
|
|---|
| 1160 | my $null_offset = $pos; # null mapping
|
|---|
| 1161 | $total += $pos;
|
|---|
| 1162 |
|
|---|
| 1163 | # add the index offsets to the subsets positions
|
|---|
| 1164 |
|
|---|
| 1165 | for (my $i = 0; $i < 4096; $i++)
|
|---|
| 1166 | {
|
|---|
| 1167 | next unless $filled[$i];
|
|---|
| 1168 | $filled[$i] += $null_offset;
|
|---|
| 1169 | }
|
|---|
| 1170 |
|
|---|
| 1171 | # dump the main index
|
|---|
| 1172 |
|
|---|
| 1173 | printf OUTPUT "const WCHAR unicode_decompose_table[%d] =\n", $total;
|
|---|
| 1174 | printf OUTPUT "{\n /* index */\n";
|
|---|
| 1175 | printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @filled_idx );
|
|---|
| 1176 | printf OUTPUT ",\n /* null sub-index */\n%s", DUMP_ARRAY( "0x%04x", 0, ($null_offset) x 16 );
|
|---|
| 1177 |
|
|---|
| 1178 | # dump the second-level indexes
|
|---|
| 1179 |
|
|---|
| 1180 | for (my $i = 0; $i < 256; $i++)
|
|---|
| 1181 | {
|
|---|
| 1182 | next unless ($filled_idx[$i] > 256);
|
|---|
| 1183 | my @table = @filled[($i<<4)..($i<<4)+15];
|
|---|
| 1184 | for (my $j = 0; $j < 16; $j++) { $table[$j] ||= $null_offset; }
|
|---|
| 1185 | printf OUTPUT ",\n /* sub-index %02x */\n", $i;
|
|---|
| 1186 | printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
|
|---|
| 1187 | }
|
|---|
| 1188 |
|
|---|
| 1189 | # dump the 16-char subsets
|
|---|
| 1190 |
|
|---|
| 1191 | printf OUTPUT ",\n /* null mapping */\n";
|
|---|
| 1192 | printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, (0) x 32 );
|
|---|
| 1193 |
|
|---|
| 1194 | for (my $i = 0; $i < 4096; $i++)
|
|---|
| 1195 | {
|
|---|
| 1196 | next unless $filled[$i];
|
|---|
| 1197 | my @table = (0) x 32;
|
|---|
| 1198 | for (my $j = 0; $j < 16; $j++)
|
|---|
| 1199 | {
|
|---|
| 1200 | if (defined $decomp_table[($i<<4) + $j])
|
|---|
| 1201 | {
|
|---|
| 1202 | $table[2 * $j] = ${$decomp_table[($i << 4) + $j]}[0];
|
|---|
| 1203 | $table[2 * $j + 1] = ${$decomp_table[($i << 4) + $j]}[1];
|
|---|
| 1204 | }
|
|---|
| 1205 | }
|
|---|
| 1206 | printf OUTPUT ",\n /* 0x%03x0 .. 0x%03xf */\n", $i, $i;
|
|---|
| 1207 | printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
|
|---|
| 1208 | }
|
|---|
| 1209 |
|
|---|
| 1210 | printf OUTPUT "\n};\n";
|
|---|
| 1211 | close OUTPUT;
|
|---|
| 1212 | save_file($filename);
|
|---|
| 1213 | }
|
|---|
| 1214 |
|
|---|
| 1215 |
|
|---|
| 1216 | ################################################################
|
|---|
| 1217 | # handle a "bestfit" Windows mapping file
|
|---|
| 1218 |
|
|---|
| 1219 | sub handle_bestfit_file($$$)
|
|---|
| 1220 | {
|
|---|
| 1221 | my ($filename, $has_glyphs, $comment) = @_;
|
|---|
| 1222 | my $state = "";
|
|---|
| 1223 | my ($codepage, $width, $def, $defw, $count);
|
|---|
| 1224 | my ($lb_cur, $lb_end);
|
|---|
| 1225 | my @lb_ranges = ();
|
|---|
| 1226 |
|
|---|
| 1227 | open INPUT,$MAPPREFIX . $filename or die "Cannot open $filename";
|
|---|
| 1228 |
|
|---|
| 1229 | while (<INPUT>)
|
|---|
| 1230 | {
|
|---|
| 1231 | next if /^;/; # skip comments
|
|---|
| 1232 | next if /^\s*$/; # skip empty lines
|
|---|
| 1233 | next if /\x1a/; # skip ^Z
|
|---|
| 1234 | last if /^ENDCODEPAGE/;
|
|---|
| 1235 |
|
|---|
| 1236 | if (/^CODEPAGE\s+(\d+)/)
|
|---|
| 1237 | {
|
|---|
| 1238 | $codepage = $1;
|
|---|
| 1239 | next;
|
|---|
| 1240 | }
|
|---|
| 1241 | if (/^CPINFO\s+(\d+)\s+0x([0-9a-fA-f]+)\s+0x([0-9a-fA-F]+)/)
|
|---|
| 1242 | {
|
|---|
| 1243 | $width = $1;
|
|---|
| 1244 | $def = hex $2;
|
|---|
| 1245 | $defw = hex $3;
|
|---|
| 1246 | next;
|
|---|
| 1247 | }
|
|---|
| 1248 | if (/^(MBTABLE|WCTABLE|DBCSRANGE|DBCSTABLE)\s+(\d+)/)
|
|---|
| 1249 | {
|
|---|
| 1250 | $state = $1;
|
|---|
| 1251 | $count = $2;
|
|---|
| 1252 | next;
|
|---|
| 1253 | }
|
|---|
| 1254 | if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)/)
|
|---|
| 1255 | {
|
|---|
| 1256 | if ($state eq "MBTABLE")
|
|---|
| 1257 | {
|
|---|
| 1258 | my $cp = hex $1;
|
|---|
| 1259 | my $uni = hex $2;
|
|---|
| 1260 | $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
|
|---|
| 1261 | next;
|
|---|
| 1262 | }
|
|---|
| 1263 | if ($state eq "WCTABLE")
|
|---|
| 1264 | {
|
|---|
| 1265 | my $uni = hex $1;
|
|---|
| 1266 | my $cp = hex $2;
|
|---|
| 1267 | $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
|
|---|
| 1268 | next;
|
|---|
| 1269 | }
|
|---|
| 1270 | if ($state eq "DBCSRANGE")
|
|---|
| 1271 | {
|
|---|
| 1272 | my $start = hex $1;
|
|---|
| 1273 | my $end = hex $2;
|
|---|
| 1274 | push @lb_ranges, $start, $end;
|
|---|
| 1275 | for (my $i = $start; $i <= $end; $i++)
|
|---|
| 1276 | {
|
|---|
| 1277 | push @lead_bytes, $i;
|
|---|
| 1278 | $cp2uni[$i] = 0;
|
|---|
| 1279 | }
|
|---|
| 1280 | $lb_cur = $start;
|
|---|
| 1281 | $lb_end = $end;
|
|---|
| 1282 | next;
|
|---|
| 1283 | }
|
|---|
| 1284 | if ($state eq "DBCSTABLE")
|
|---|
| 1285 | {
|
|---|
| 1286 | my $mb = hex $1;
|
|---|
| 1287 | my $uni = hex $2;
|
|---|
| 1288 | my $cp = ($lb_cur << 8) | $mb;
|
|---|
| 1289 | $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
|
|---|
| 1290 | if (!--$count)
|
|---|
| 1291 | {
|
|---|
| 1292 | if (++$lb_cur > $lb_end) { $state = "DBCSRANGE"; }
|
|---|
| 1293 | }
|
|---|
| 1294 | next;
|
|---|
| 1295 | }
|
|---|
| 1296 | }
|
|---|
| 1297 | die "$filename: Unrecognized line $_\n";
|
|---|
| 1298 | }
|
|---|
| 1299 | close INPUT;
|
|---|
| 1300 |
|
|---|
| 1301 | my $output = sprintf "c_%03d.c", $codepage;
|
|---|
| 1302 | open OUTPUT,">$output.new" or die "Cannot create $output";
|
|---|
| 1303 |
|
|---|
| 1304 | printf "Building %s from %s (%s)\n", $output, $filename, $comment;
|
|---|
| 1305 |
|
|---|
| 1306 | # dump all tables
|
|---|
| 1307 |
|
|---|
| 1308 | printf OUTPUT "/* code page %03d (%s) */\n", $codepage, $comment;
|
|---|
| 1309 | printf OUTPUT "/* generated from %s */\n", $MAPPREFIX . $filename;
|
|---|
| 1310 | printf OUTPUT "/* DO NOT EDIT!! */\n\n";
|
|---|
| 1311 | printf OUTPUT "#include \"wine/unicode.h\"\n\n";
|
|---|
| 1312 |
|
|---|
| 1313 | if ($width == 1) { dump_sbcs_table( $codepage, $has_glyphs, $comment, $def, $defw ); }
|
|---|
| 1314 | else { dump_dbcs_table( $codepage, $comment, $def, $defw, @lb_ranges ); }
|
|---|
| 1315 | close OUTPUT;
|
|---|
| 1316 | save_file($output);
|
|---|
| 1317 | }
|
|---|
| 1318 |
|
|---|
| 1319 |
|
|---|
| 1320 | ################################################################
|
|---|
| 1321 | # read an input file and generate the corresponding .c file
|
|---|
| 1322 | sub HANDLE_FILE(@)
|
|---|
| 1323 | {
|
|---|
| 1324 | my ($codepage,$filename,$has_glyphs,$comment) = @_;
|
|---|
| 1325 |
|
|---|
| 1326 | @cp2uni = ();
|
|---|
| 1327 | @lead_bytes = ();
|
|---|
| 1328 | @uni2cp = ();
|
|---|
| 1329 |
|
|---|
| 1330 | # symbol codepage file is special
|
|---|
| 1331 | if ($codepage == 20932) { READ_JIS0208_FILE($MAPPREFIX . $filename); }
|
|---|
| 1332 | elsif ($codepage == 20127) { fill_20127_codepage(); }
|
|---|
| 1333 | elsif ($filename =~ /\/bestfit/)
|
|---|
| 1334 | {
|
|---|
| 1335 | handle_bestfit_file( $filename, $has_glyphs, $comment );
|
|---|
| 1336 | return;
|
|---|
| 1337 | }
|
|---|
| 1338 | else { READ_FILE($MAPPREFIX . $filename); }
|
|---|
| 1339 |
|
|---|
| 1340 | ADD_DEFAULT_MAPPINGS();
|
|---|
| 1341 |
|
|---|
| 1342 | my $output = sprintf "c_%03d.c", $codepage;
|
|---|
| 1343 | open OUTPUT,">$output.new" or die "Cannot create $output";
|
|---|
| 1344 |
|
|---|
| 1345 | printf "Building %s from %s (%s)\n", $output, $filename || "hardcoded data", $comment;
|
|---|
| 1346 |
|
|---|
| 1347 | # dump all tables
|
|---|
| 1348 |
|
|---|
| 1349 | printf OUTPUT "/* code page %03d (%s) */\n", $codepage, $comment;
|
|---|
| 1350 | if ($filename)
|
|---|
| 1351 | {
|
|---|
| 1352 | printf OUTPUT "/* generated from %s */\n", $MAPPREFIX . $filename;
|
|---|
| 1353 | printf OUTPUT "/* DO NOT EDIT!! */\n\n";
|
|---|
| 1354 | }
|
|---|
| 1355 | else
|
|---|
| 1356 | {
|
|---|
| 1357 | printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
|
|---|
| 1358 | }
|
|---|
| 1359 | printf OUTPUT "#include \"wine/unicode.h\"\n\n";
|
|---|
| 1360 |
|
|---|
| 1361 | if (!@lead_bytes) { dump_sbcs_table( $codepage, $has_glyphs, $comment, $DEF_CHAR, $DEF_CHAR ); }
|
|---|
| 1362 | else { dump_dbcs_table( $codepage, $comment, $DEF_CHAR, $DEF_CHAR, get_lb_ranges() ); }
|
|---|
| 1363 | close OUTPUT;
|
|---|
| 1364 | save_file($output);
|
|---|
| 1365 | }
|
|---|
| 1366 |
|
|---|
| 1367 |
|
|---|
| 1368 | ################################################################
|
|---|
| 1369 | # save a file if modified
|
|---|
| 1370 | sub save_file($)
|
|---|
| 1371 | {
|
|---|
| 1372 | my $file = shift;
|
|---|
| 1373 | if (-f $file && !system "cmp $file $file.new >/dev/null")
|
|---|
| 1374 | {
|
|---|
| 1375 | unlink "$file.new";
|
|---|
| 1376 | }
|
|---|
| 1377 | else
|
|---|
| 1378 | {
|
|---|
| 1379 | rename "$file.new", "$file";
|
|---|
| 1380 | }
|
|---|
| 1381 | }
|
|---|
| 1382 |
|
|---|
| 1383 |
|
|---|
| 1384 | ################################################################
|
|---|
| 1385 | # output the list of codepage tables into the cptable.c file
|
|---|
| 1386 | sub OUTPUT_CPTABLE()
|
|---|
| 1387 | {
|
|---|
| 1388 | my @tables_decl = ();
|
|---|
| 1389 |
|
|---|
| 1390 | foreach my $file (@allfiles)
|
|---|
| 1391 | {
|
|---|
| 1392 | my ($codepage,$filename,$comment) = @$file;
|
|---|
| 1393 | push @tables_decl, sprintf("extern union cptable cptable_%03d;\n",$codepage);
|
|---|
| 1394 | }
|
|---|
| 1395 |
|
|---|
| 1396 | push @tables_decl, sprintf("\nstatic const union cptable * const cptables[%d] =\n{\n",$#allfiles+1);
|
|---|
| 1397 | foreach my $file (@allfiles)
|
|---|
| 1398 | {
|
|---|
| 1399 | my ($codepage,$filename,$comment) = @$file;
|
|---|
| 1400 | push @tables_decl, sprintf(" &cptable_%03d,\n", $codepage);
|
|---|
| 1401 | }
|
|---|
| 1402 | push @tables_decl, "};";
|
|---|
| 1403 | REPLACE_IN_FILE( "cptable.c", @tables_decl );
|
|---|
| 1404 | }
|
|---|
| 1405 |
|
|---|
| 1406 | ################################################################
|
|---|
| 1407 | # replace the contents of a file between ### cpmap ### marks
|
|---|
| 1408 |
|
|---|
| 1409 | sub REPLACE_IN_FILE($@)
|
|---|
| 1410 | {
|
|---|
| 1411 | my $name = shift;
|
|---|
| 1412 | my @data = @_;
|
|---|
| 1413 | my @lines = ();
|
|---|
| 1414 | open(FILE,$name) or die "Can't open $name";
|
|---|
| 1415 | while (<FILE>)
|
|---|
| 1416 | {
|
|---|
| 1417 | push @lines, $_;
|
|---|
| 1418 | last if /\#\#\# cpmap begin \#\#\#/;
|
|---|
| 1419 | }
|
|---|
| 1420 | push @lines, @data;
|
|---|
| 1421 | while (<FILE>)
|
|---|
| 1422 | {
|
|---|
| 1423 | if (/\#\#\# cpmap end \#\#\#/) { push @lines, "\n", $_; last; }
|
|---|
| 1424 | }
|
|---|
| 1425 | push @lines, <FILE>;
|
|---|
| 1426 | open(FILE,">$name.new") or die "Can't modify $name";
|
|---|
| 1427 | print FILE @lines;
|
|---|
| 1428 | close(FILE);
|
|---|
| 1429 | save_file($name);
|
|---|
| 1430 | }
|
|---|
| 1431 |
|
|---|
| 1432 | ################################################################
|
|---|
| 1433 | # main routine
|
|---|
| 1434 |
|
|---|
| 1435 | READ_DEFAULTS( $DEFAULTS );
|
|---|
| 1436 | DUMP_CASE_MAPPINGS( "casemap.c" );
|
|---|
| 1437 | DUMP_SORTKEYS( "collation.c", READ_SORTKEYS_FILE() );
|
|---|
| 1438 | DUMP_COMPOSE_TABLES( "compose.c" );
|
|---|
| 1439 | DUMP_CTYPE_TABLES( "wctype.c" );
|
|---|
| 1440 |
|
|---|
| 1441 | foreach my $file (@allfiles) { HANDLE_FILE( @{$file} ); }
|
|---|
| 1442 |
|
|---|
| 1443 | OUTPUT_CPTABLE();
|
|---|
| 1444 |
|
|---|
| 1445 | exit 0;
|
|---|
| 1446 |
|
|---|
| 1447 | # Local Variables:
|
|---|
| 1448 | # compile-command: "./cpmap.pl && make -k"
|
|---|
| 1449 | # End:
|
|---|
