3 # Generate code page .c files from ftp.unicode.org descriptions
5 # Copyright 2000 Alexandre Julliard
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.
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.
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., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
22 # base directory for ftp.unicode.org files
23 $BASEDIR = "ftp.unicode.org/Public/";
24 $MAPPREFIX = $BASEDIR . "MAPPINGS/";
27 $UNICODEDATA = $BASEDIR . "UNIDATA/UnicodeData.txt";
30 $DEFAULTS = "./defaults";
32 # Default char for undefined mappings
37 [ 37, "VENDORS/MICSFT/EBCDIC/CP037.TXT", "IBM EBCDIC US Canada" ],
38 [ 42, "VENDORS/ADOBE/symbol.txt", "Symbol" ],
39 [ 424, "VENDORS/MISC/CP424.TXT", "IBM EBCDIC Hebrew" ],
40 [ 437, "VENDORS/MICSFT/PC/CP437.TXT", "OEM United States" ],
41 [ 500, "VENDORS/MICSFT/EBCDIC/CP500.TXT", "IBM EBCDIC International" ],
42 [ 737, "VENDORS/MICSFT/PC/CP737.TXT", "OEM Greek 437G" ],
43 [ 775, "VENDORS/MICSFT/PC/CP775.TXT", "OEM Baltic" ],
44 [ 850, "VENDORS/MICSFT/PC/CP850.TXT", "OEM Multilingual Latin 1" ],
45 [ 852, "VENDORS/MICSFT/PC/CP852.TXT", "OEM Slovak Latin 2" ],
46 [ 855, "VENDORS/MICSFT/PC/CP855.TXT", "OEM Cyrillic" ],
47 [ 856, "VENDORS/MISC/CP856.TXT", "Hebrew PC" ],
48 [ 857, "VENDORS/MICSFT/PC/CP857.TXT", "OEM Turkish" ],
49 [ 860, "VENDORS/MICSFT/PC/CP860.TXT", "OEM Portuguese" ],
50 [ 861, "VENDORS/MICSFT/PC/CP861.TXT", "OEM Icelandic" ],
51 [ 862, "VENDORS/MICSFT/PC/CP862.TXT", "OEM Hebrew" ],
52 [ 863, "VENDORS/MICSFT/PC/CP863.TXT", "OEM Canadian French" ],
53 [ 864, "VENDORS/MICSFT/PC/CP864.TXT", "OEM Arabic" ],
54 [ 865, "VENDORS/MICSFT/PC/CP865.TXT", "OEM Nordic" ],
55 [ 866, "VENDORS/MICSFT/PC/CP866.TXT", "OEM Russian" ],
56 [ 869, "VENDORS/MICSFT/PC/CP869.TXT", "OEM Greek" ],
57 [ 874, "VENDORS/MICSFT/PC/CP874.TXT", "ANSI/OEM Thai" ],
58 [ 875, "VENDORS/MICSFT/EBCDIC/CP875.TXT", "IBM EBCDIC Greek" ],
59 [ 878, "VENDORS/MISC/KOI8-R.TXT", "Russian KOI8" ],
60 [ 932, "VENDORS/MICSFT/WINDOWS/CP932.TXT", "ANSI/OEM Japanese Shift-JIS" ],
61 [ 936, "VENDORS/MICSFT/WINDOWS/CP936.TXT", "ANSI/OEM Simplified Chinese GBK" ],
62 [ 949, "VENDORS/MICSFT/WINDOWS/CP949.TXT", "ANSI/OEM Korean Unified Hangul" ],
63 [ 950, "VENDORS/MICSFT/WINDOWS/CP950.TXT", "ANSI/OEM Traditional Chinese Big5" ],
64 [ 1006, "VENDORS/MISC/CP1006.TXT", "IBM Arabic" ],
65 [ 1026, "VENDORS/MICSFT/EBCDIC/CP1026.TXT", "IBM EBCDIC Latin 5 Turkish" ],
66 [ 1250, "VENDORS/MICSFT/WINDOWS/CP1250.TXT", "ANSI Eastern Europe" ],
67 [ 1251, "VENDORS/MICSFT/WINDOWS/CP1251.TXT", "ANSI Cyrillic" ],
68 [ 1252, "VENDORS/MICSFT/WINDOWS/CP1252.TXT", "ANSI Latin 1" ],
69 [ 1253, "VENDORS/MICSFT/WINDOWS/CP1253.TXT", "ANSI Greek" ],
70 [ 1254, "VENDORS/MICSFT/WINDOWS/CP1254.TXT", "ANSI Turkish" ],
71 [ 1255, "VENDORS/MICSFT/WINDOWS/CP1255.TXT", "ANSI Hebrew" ],
72 [ 1256, "VENDORS/MICSFT/WINDOWS/CP1256.TXT", "ANSI Arabic" ],
73 [ 1257, "VENDORS/MICSFT/WINDOWS/CP1257.TXT", "ANSI Baltic" ],
74 [ 1258, "VENDORS/MICSFT/WINDOWS/CP1258.TXT", "ANSI/OEM Viet Nam" ],
75 [ 10000, "VENDORS/MICSFT/MAC/ROMAN.TXT", "Mac Roman" ],
76 [ 10006, "VENDORS/MICSFT/MAC/GREEK.TXT", "Mac Greek" ],
77 [ 10007, "VENDORS/MICSFT/MAC/CYRILLIC.TXT", "Mac Cyrillic" ],
78 [ 10029, "VENDORS/MICSFT/MAC/LATIN2.TXT", "Mac Latin 2" ],
79 [ 10079, "VENDORS/MICSFT/MAC/ICELAND.TXT", "Mac Icelandic" ],
80 [ 10081, "VENDORS/MICSFT/MAC/TURKISH.TXT", "Mac Turkish" ],
81 [ 20866, "VENDORS/MISC/KOI8-R.TXT", "Russian KOI8" ],
82 [ 28591, "ISO8859/8859-1.TXT", "ISO 8859-1 Latin 1" ],
83 [ 28592, "ISO8859/8859-2.TXT", "ISO 8859-2 Latin 2 (East European)" ],
84 [ 28593, "ISO8859/8859-3.TXT", "ISO 8859-3 Latin 3 (South European)" ],
85 [ 28594, "ISO8859/8859-4.TXT", "ISO 8859-4 Latin 4 (Baltic old)" ],
86 [ 28595, "ISO8859/8859-5.TXT", "ISO 8859-5 Cyrillic" ],
87 [ 28596, "ISO8859/8859-6.TXT", "ISO 8859-6 Arabic" ],
88 [ 28597, "ISO8859/8859-7.TXT", "ISO 8859-7 Greek" ],
89 [ 28598, "ISO8859/8859-8.TXT", "ISO 8859-8 Hebrew" ],
90 [ 28599, "ISO8859/8859-9.TXT", "ISO 8859-9 Latin 5 (Turkish)" ],
91 [ 28600, "ISO8859/8859-10.TXT", "ISO 8859-10 Latin 6 (Nordic)" ],
92 [ 28603, "ISO8859/8859-13.TXT", "ISO 8859-13 Latin 7 (Baltic)" ],
93 [ 28604, "ISO8859/8859-14.TXT", "ISO 8859-14 Latin 8 (Celtic)" ],
94 [ 28605, "ISO8859/8859-15.TXT", "ISO 8859-15 Latin 9 (Euro)" ],
95 [ 28606, "ISO8859/8859-16.TXT", "ISO 8859-16 Latin 10 (Balkan)" ]
114 "Lu" => $ctype{"alpha"}|$ctype{"upper"}, # Letter, Uppercase
115 "Ll" => $ctype{"alpha"}|$ctype{"lower"}, # Letter, Lowercase
116 "Lt" => $ctype{"alpha"}, # Letter, Titlecase
117 "Mn" => $ctype{"punct"}, # Mark, Non-Spacing
118 "Mc" => $ctype{"punct"}, # Mark, Spacing Combining
119 "Me" => $ctype{"punct"}, # Mark, Enclosing
120 "Nd" => $ctype{"digit"}, # Number, Decimal Digit
121 "Nl" => $ctype{"punct"}, # Number, Letter
122 "No" => $ctype{"punct"}, # Number, Other
123 "Zs" => $ctype{"space"}, # Separator, Space
124 "Zl" => 0, # Separator, Line
125 "Zp" => 0, # Separator, Paragraph
126 "Cc" => $ctype{"cntrl"}, # Other, Control
127 "Cf" => 0, # Other, Format
128 "Cs" => 0, # Other, Surrogate
129 "Co" => 0, # Other, Private Use
130 "Cn" => 0, # Other, Not Assigned
131 "Lm" => $ctype{"punct"}, # Letter, Modifier
132 "Lo" => $ctype{"alpha"}, # Letter, Other
133 "Pc" => $ctype{"punct"}, # Punctuation, Connector
134 "Pd" => $ctype{"punct"}, # Punctuation, Dash
135 "Ps" => $ctype{"punct"}, # Punctuation, Open
136 "Pe" => $ctype{"punct"}, # Punctuation, Close
137 "Pi" => $ctype{"punct"}, # Punctuation, Initial quote
138 "Pf" => $ctype{"punct"}, # Punctuation, Final quote
139 "Po" => $ctype{"punct"}, # Punctuation, Other
140 "Sm" => $ctype{"punct"}, # Symbol, Math
141 "Sc" => $ctype{"punct"}, # Symbol, Currency
142 "Sk" => $ctype{"punct"}, # Symbol, Modifier
143 "So" => $ctype{"punct"} # Symbol, Other
146 # a few characters need additional categories that cannot be determined automatically
147 %special_categories =
149 "xdigit" => [ ord('0')..ord('9'),ord('A')..ord('F'),ord('a')..ord('f'),
150 0xff10..0xff19, 0xff21..0xff26, 0xff41..0xff46 ],
151 "space" => [ 0x09..0x0d, 0xfeff ],
152 "blank" => [ 0x09, 0x20, 0xa0, 0xfeff ]
157 "L" => 1, # Left-to-Right
158 "LRE" => 11, # Left-to-Right Embedding
159 "LRO" => 11, # Left-to-Right Override
160 "R" => 2, # Right-to-Left
161 "AL" => 2, # Right-to-Left Arabic
162 "RLE" => 11, # Right-to-Left Embedding
163 "RLO" => 11, # Right-to-Left Override
164 "PDF" => 11, # Pop Directional Format
165 "EN" => 3, # European Number
166 "ES" => 4, # European Number Separator
167 "ET" => 5, # European Number Terminator
168 "AN" => 6, # Arabic Number
169 "CS" => 7, # Common Number Separator
170 "NSM" => 0, # Non-Spacing Mark
171 "BN" => 0, # Boundary Neutral
172 "B" => 8, # Paragraph Separator
173 "S" => 9, # Segment Separator
174 "WS" => 10, # Whitespace
175 "ON" => 11 # Other Neutrals
179 ################################################################
183 DUMP_CASE_MAPPINGS();
184 DUMP_COMPOSE_TABLES();
187 foreach $file (@allfiles) { HANDLE_FILE( @$file ); }
194 ################################################################
195 # read in the defaults file
198 @unicode_defaults = ();
199 @unicode_aliases = ();
202 @category_table = ();
203 @direction_table = ();
207 # first setup a few default mappings
209 open DEFAULTS or die "Cannot open $DEFAULTS";
210 print "Loading $DEFAULTS\n";
213 next if /^\#/; # skip comments
214 next if /^$/; # skip empty lines
215 if (/^(([0-9a-fA-F]+)(,[0-9a-fA-F]+)*)\s+([0-9a-fA-F]+|'.'|none)\s+(\#.*)?/)
217 my @src = map hex, split /,/,$1;
220 if ($#src > 0) { push @unicode_aliases, \@src; }
221 next if ($dst eq "none");
222 $dst = ($dst =~ /\'.\'/) ? ord substr($dst,1,1) : hex $dst;
225 die "Duplicate value" if defined($unicode_defaults[$src]);
226 $unicode_defaults[$src] = $dst;
230 die "Unrecognized line $_\n";
233 # now build mappings from the decomposition field of the Unicode database
235 open UNICODEDATA or die "Cannot open $UNICODEDATA";
236 print "Loading $UNICODEDATA\n";
237 while (<UNICODEDATA>)
239 # Decode the fields ...
240 ($code, $name, $cat, $comb, $bidi,
241 $decomp, $dec, $dig, $num, $mirror,
242 $oldname, $comment, $upper, $lower, $title) = split /;/;
246 die "unknown category $cat" unless defined $categories{$cat};
247 die "unknown directionality $bidi" unless defined $directions{$bidi};
249 $uniname[$src] = $name;
250 $category_table[$src] = $categories{$cat};
251 $direction_table[$src] = $directions{$bidi};
255 $tolower_table[$src] = hex $lower;
256 $category_table[$src] |= $ctype{"upper"}|$ctype{"alpha"};
260 $toupper_table[$src] = hex $upper;
261 $category_table[$src] |= $ctype{"lower"}|$ctype{"alpha"};
265 $category_table[$src] |= $ctype{"digit"};
268 # copy the category and direction for everything between First/Last pairs
269 if ($name =~ /, First>/) { $start = $src; }
270 if ($name =~ /, Last>/)
272 while ($start < $src)
274 $category_table[$start] = $category_table[$src];
275 $direction_table[$start] = $direction_table[$src];
280 next if $decomp eq ""; # no decomposition, skip it
282 if ($decomp =~ /^<([a-zA-Z]+)>\s+([0-9a-fA-F]+)$/)
284 # decomposition of the form "<foo> 1234" -> use char if type is known
285 next unless ($1 eq "font" ||
296 elsif ($decomp =~ /^<compat>\s+0020\s+([0-9a-fA-F]+)/)
298 # decomposition "<compat> 0020 1234" -> combining accent
301 elsif ($decomp =~ /^([0-9a-fA-F]+)/)
303 # decomposition contains only char values without prefix -> use first char
305 $category_table[$src] |= $category_table[$dst];
306 # store decomposition if it contains two chars
307 if ($decomp =~ /^([0-9a-fA-F]+)\s+([0-9a-fA-F]+)$/)
309 $decomp_table[$src] = [ hex $1, hex $2 ];
310 push @compose_table, [ hex $1, hex $2, $src ];
318 next if defined($unicode_defaults[$src]); # may have been set in the defaults file
321 for ($i = $dst; ; $i = $unicode_defaults[$i])
323 die sprintf("loop detected for %04x -> %04x",$src,$dst) if $i == $src;
324 last unless defined($unicode_defaults[$i]);
326 $unicode_defaults[$src] = $dst;
329 # patch the category of some special characters
331 foreach $cat (keys %special_categories)
333 my $flag = $ctype{$cat};
334 foreach $i (@{$special_categories{$cat}}) { $category_table[$i] |= $flag; }
339 ################################################################
340 # parse the input file
344 open INPUT,$name or die "Cannot open $name";
351 next if /^\#/; # skip comments
352 next if /^$/; # skip empty lines
353 next if /\x1a/; # skip ^Z
354 next if (/^0x([0-9a-fA-F]+)\s+\#UNDEFINED/); # undefined char
356 if (/^0x([0-9a-fA-F]+)\s+\#DBCS LEAD BYTE/)
359 push @lead_bytes,$cp;
363 if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\s+(\#.*)?/)
367 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
368 $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
371 die "$name: Unrecognized line $_\n";
376 ################################################################
377 # parse the symbol.txt file, since its syntax is different from the other ones
381 open INPUT,$name or die "Cannot open $name";
388 next if /^\#/; # skip comments
389 next if /^$/; # skip empty lines
390 next if /\x1a/; # skip ^Z
391 if (/^([0-9a-fA-F]+)\s+([0-9a-fA-F]+)\s+(\#.*)?/)
395 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
396 $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
399 die "$name: Unrecognized line $_\n";
404 ################################################################
405 # add default mappings once the file had been read
406 sub ADD_DEFAULT_MAPPINGS
410 foreach $alias (@unicode_aliases)
413 foreach $src (@$alias)
415 if (defined($uni2cp[$src]))
417 $target = $uni2cp[$src];
421 next unless defined($target);
423 # At least one char of the alias set is defined, set the others to the same value
424 foreach $src (@$alias)
426 $uni2cp[$src] = $target unless defined($uni2cp[$src]);
430 # For every src -> target mapping in the defaults table,
431 # make uni2cp[src] = uni2cp[target] if uni2cp[target] is defined
433 for ($src = 0; $src < 65536; $src++)
435 next if defined($uni2cp[$src]); # source has a definition already
436 next unless defined($unicode_defaults[$src]); # no default for this char
437 my $target = $unicode_defaults[$src];
439 # do a recursive mapping until we find a target char that is defined
440 while (!defined($uni2cp[$target]) &&
441 defined($unicode_defaults[$target])) { $target = $unicode_defaults[$target]; }
443 if (defined($uni2cp[$target])) { $uni2cp[$src] = $uni2cp[$target]; }
446 # Add an identity mapping for all undefined chars
448 for ($i = 0; $i < 256; $i++)
450 next if defined($cp2uni[$i]);
451 next if defined($uni2cp[$i]);
452 $cp2uni[$i] = $uni2cp[$i] = $i;
456 ################################################################
457 # dump an array of integers
460 my ($format,$default,@array) = @_;
462 for ($i = 0; $i < $#array; $i++)
464 $ret .= sprintf($format, defined $array[$i] ? $array[$i] : $default);
465 $ret .= (($i % 8) != 7) ? ", " : ",\n ";
467 $ret .= sprintf($format, defined $array[$i] ? $array[$i] : $default);
471 ################################################################
472 # dump an SBCS mapping table
475 my ($codepage, $name) = @_;
478 # output the ascii->unicode table
480 printf OUTPUT "static const WCHAR cp2uni[256] =\n";
481 printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%04x", $DEF_CHAR, @cp2uni[0 .. 255] );
483 # count the number of unicode->ascii subtables that contain something
487 for ($i = 0; $i < 65536; $i++)
489 next unless defined $uni2cp[$i];
490 $filled[$i >> 8] = 1;
495 # output all the subtables into a single array
497 printf OUTPUT "static const unsigned char uni2cp_low[%d] =\n{\n", $subtables*256;
498 for ($i = 0; $i < 256; $i++)
500 next unless $filled[$i];
501 printf OUTPUT " /* 0x%02x00 .. 0x%02xff */\n", $i, $i;
502 printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%02x", $DEF_CHAR, @uni2cp[($i<<8) .. ($i<<8)+255] );
504 printf OUTPUT " /* defaults */\n";
505 printf OUTPUT "%s\n};\n\n", DUMP_ARRAY( "0x%02x", 0, ($DEF_CHAR) x 256 );
507 # output a table of the offsets of the subtables in the previous array
511 for ($i = 0; $i < 256; $i++)
513 if ($filled[$i]) { push @offsets, $pos; $pos += 256; }
514 else { push @offsets, ($subtables-1) * 256; }
516 printf OUTPUT "static const unsigned short uni2cp_high[256] =\n";
517 printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%04x", 0, @offsets );
519 # output the code page descriptor
521 printf OUTPUT "const struct sbcs_table cptable_%03d =\n{\n", $codepage;
522 printf OUTPUT " { %d, 1, 0x%04x, 0x%04x, \"%s\" },\n",
523 $codepage, $DEF_CHAR, $DEF_CHAR, $name;
524 printf OUTPUT " cp2uni,\n";
525 printf OUTPUT " uni2cp_low,\n";
526 printf OUTPUT " uni2cp_high\n};\n";
530 ################################################################
531 # dump a DBCS mapping table
534 my ($codepage, $name) = @_;
537 # build a list of lead bytes that are actually used
540 LBLOOP: for ($y = 0; $y <= $#lead_bytes; $y++)
542 my $base = $lead_bytes[$y] << 8;
543 for ($x = 0; $x < 256; $x++)
545 if (defined $cp2uni[$base+$x])
547 push @lblist,$lead_bytes[$y];
552 my $unused = ($#lead_bytes > $#lblist);
554 # output the ascii->unicode table for the single byte chars
556 printf OUTPUT "static const WCHAR cp2uni[%d] =\n", 256 * ($#lblist + 2 + $unused);
557 printf OUTPUT "{\n%s,\n", DUMP_ARRAY( "0x%04x", $DEF_CHAR, @cp2uni[0 .. 255] );
559 # output the default table for unused lead bytes
563 printf OUTPUT " /* unused lead bytes */\n";
564 printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%04x", 0, ($DEF_CHAR) x 256 );
567 # output the ascii->unicode table for each DBCS lead byte
569 for ($y = 0; $y <= $#lblist; $y++)
571 my $base = $lblist[$y] << 8;
572 printf OUTPUT " /* lead byte %02x */\n", $lblist[$y];
573 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", $DEF_CHAR, @cp2uni[$base .. $base+255] );
574 printf OUTPUT ($y < $#lblist) ? ",\n" : "\n};\n\n";
577 # output the lead byte subtables offsets
580 for ($x = 0; $x < 256; $x++) { $offsets[$x] = 0; }
581 for ($x = 0; $x <= $#lblist; $x++) { $offsets[$lblist[$x]] = $x + 1; }
584 # increment all lead bytes offset to take into account the unused table
585 for ($x = 0; $x <= $#lead_bytes; $x++) { $offsets[$lead_bytes[$x]]++; }
587 printf OUTPUT "static const unsigned char cp2uni_leadbytes[256] =\n";
588 printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%02x", 0, @offsets );
590 # count the number of unicode->ascii subtables that contain something
594 for ($i = 0; $i < 65536; $i++)
596 next unless defined $uni2cp[$i];
597 $filled[$i >> 8] = 1;
602 # output all the subtables into a single array
604 printf OUTPUT "static const unsigned short uni2cp_low[%d] =\n{\n", $subtables*256;
605 for ($y = 0; $y < 256; $y++)
607 next unless $filled[$y];
608 printf OUTPUT " /* 0x%02x00 .. 0x%02xff */\n", $y, $y;
609 printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%04x", $DEF_CHAR, @uni2cp[($y<<8) .. ($y<<8)+255] );
611 printf OUTPUT " /* defaults */\n";
612 printf OUTPUT "%s\n};\n\n", DUMP_ARRAY( "0x%04x", 0, ($DEF_CHAR) x 256 );
614 # output a table of the offsets of the subtables in the previous array
618 for ($y = 0; $y < 256; $y++)
620 if ($filled[$y]) { push @offsets, $pos; $pos += 256; }
621 else { push @offsets, ($subtables-1) * 256; }
623 printf OUTPUT "static const unsigned short uni2cp_high[256] =\n";
624 printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%04x", 0, @offsets );
626 # output the code page descriptor
628 printf OUTPUT "const struct dbcs_table cptable_%03d =\n{\n", $codepage;
629 printf OUTPUT " { %d, 2, 0x%04x, 0x%04x, \"%s\" },\n",
630 $codepage, $DEF_CHAR, $DEF_CHAR, $name;
631 printf OUTPUT " cp2uni,\n";
632 printf OUTPUT " cp2uni_leadbytes,\n";
633 printf OUTPUT " uni2cp_low,\n";
634 printf OUTPUT " uni2cp_high,\n";
636 printf OUTPUT "};\n";
640 ################################################################
641 # dump the list of defined lead byte ranges
646 foreach $i (@lead_bytes) { $list[$i] = 1; }
649 for ($i = 0; $i < 256; $i++)
653 if (!defined $list[$i]) { printf OUTPUT "0x%02x, ", $i-1; $on = 0; }
657 if ($list[$i]) { printf OUTPUT "0x%02x, ", $i; $on = 1; }
660 if ($on) { printf OUTPUT "0xff, "; }
661 printf OUTPUT "0x00, 0x00 }\n";
665 ################################################################
666 # dump the case mapping tables
667 sub DUMP_CASE_MAPPINGS
669 open OUTPUT,">casemap.c" or die "Cannot create casemap.c";
670 printf "Building casemap.c\n";
671 printf OUTPUT "/* Unicode case mappings */\n";
672 printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
673 printf OUTPUT "#include \"wine/unicode.h\"\n\n";
675 DUMP_CASE_TABLE( "casemap_lower", @tolower_table );
676 DUMP_CASE_TABLE( "casemap_upper", @toupper_table );
681 ################################################################
682 # dump a case mapping table
685 my ($name,@table) = @_;
687 # count the number of sub tables that contain something
691 for ($i = 0; $i < 65536; $i++)
693 next unless defined $table[$i];
694 $filled[$i >> 8] = $pos;
698 for ($i = 0; $i < 65536; $i++)
700 next unless defined $table[$i];
701 $table[$i] = ($table[$i] - $i) & 0xffff;
706 printf OUTPUT "const WCHAR %s[%d] =\n", $name, $pos;
707 printf OUTPUT "{\n /* index */\n";
708 printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%04x", 256, @filled );
709 printf OUTPUT " /* defaults */\n";
710 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, (0) x 256 );
711 for ($i = 0; $i < 256; $i++)
713 next unless $filled[$i];
714 printf OUTPUT ",\n /* 0x%02x00 .. 0x%02xff */\n", $i, $i;
715 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table[($i<<8) .. ($i<<8)+255] );
717 printf OUTPUT "\n};\n";
721 ################################################################
722 # dump the ctype tables
723 sub DUMP_CTYPE_TABLES
725 open OUTPUT,">wctype.c" or die "Cannot create casemap.c";
726 printf "Building wctype.c\n";
727 printf OUTPUT "/* Unicode ctype tables */\n";
728 printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
729 printf OUTPUT "#include \"wine/unicode.h\"\n\n";
732 my @array = (0) x 256;
734 # add the direction in the high 4 bits of the category
735 for ($i = 0; $i < 65536; $i++)
737 $category_table[$i] |= $direction_table[$i] << 12;
740 # try to merge table rows
741 for ($row = 0; $row < 256; $row++)
743 my $rowtxt = sprintf "%04x" x 256, @category_table[($row<<8)..($row<<8)+255];
744 if (defined($sequences{$rowtxt}))
746 # reuse an existing row
747 $array[$row] = $sequences{$rowtxt};
752 $sequences{$rowtxt} = $array[$row] = $#array + 1;
753 push @array, @category_table[($row<<8)..($row<<8)+255];
757 printf OUTPUT "const unsigned short wctype_table[%d] =\n{\n", $#array+1;
758 printf OUTPUT " /* offsets */\n%s,\n", DUMP_ARRAY( "0x%04x", 0, @array[0..255] );
759 printf OUTPUT " /* values */\n%s\n};\n", DUMP_ARRAY( "0x%04x", 0, @array[256..$#array] );
765 ################################################################
766 # dump the char composition tables
767 sub DUMP_COMPOSE_TABLES
769 open OUTPUT,">compose.c" or die "Cannot create compose.c";
770 printf "Building compose.c\n";
771 printf OUTPUT "/* Unicode char composition */\n";
772 printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
773 printf OUTPUT "#include \"wine/unicode.h\"\n\n";
775 ######### composition table
778 foreach $i (@compose_table)
781 push @{$filled[$comp[1]]}, [ $comp[0], $comp[2] ];
784 # count how many different second chars we have
786 for ($i = $count = 0; $i < 65536; $i++)
788 next unless defined $filled[$i];
792 # build the table of second chars and offsets
794 my $pos = $count + 1;
795 for ($i = 0; $i < 65536; $i++)
797 next unless defined $filled[$i];
798 push @table, $i, $pos;
799 $pos += @{$filled[$i]};
801 # terminator with last position
802 push @table, 0, $pos;
803 printf OUTPUT "const WCHAR unicode_compose_table[0x%x] =\n{\n", 2*$pos;
804 printf OUTPUT " /* second chars + offsets */\n%s", DUMP_ARRAY( "0x%04x", 0, @table );
806 # build the table of first chars and mappings
808 for ($i = 0; $i < 65536; $i++)
810 next unless defined $filled[$i];
812 my @list = sort { $a->[0] <=> $b->[0] } @{$filled[$i]};
813 for ($j = 0; $j <= $#list; $j++)
815 push @table, $list[$j][0], $list[$j][1];
817 printf OUTPUT ",\n /* 0x%04x */\n%s", $i, DUMP_ARRAY( "0x%04x", 0, @table );
819 printf OUTPUT "\n};\n\nconst unsigned int unicode_compose_table_size = %d;\n\n", $count;
821 ######### decomposition table
823 # first determine all the 16-char subsets that contain something
825 my @filled = (0) x 4096;
826 my $pos = 16*2; # for the null subset
827 for ($i = 0; $i < 65536; $i++)
829 next unless defined $decomp_table[$i];
830 $filled[$i >> 4] = $pos;
836 # now count the 256-char subsets that contain something
838 my @filled_idx = (256) x 256;
840 for ($i = 0; $i < 4096; $i++)
842 next unless $filled[$i];
843 $filled_idx[$i >> 4] = $pos;
847 my $null_offset = $pos; # null mapping
850 # add the index offsets to the subsets positions
852 for ($i = 0; $i < 4096; $i++)
854 next unless $filled[$i];
855 $filled[$i] += $null_offset;
858 # dump the main index
860 printf OUTPUT "const WCHAR unicode_decompose_table[%d] =\n", $total;
861 printf OUTPUT "{\n /* index */\n";
862 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @filled_idx );
863 printf OUTPUT ",\n /* null sub-index */\n%s", DUMP_ARRAY( "0x%04x", 0, ($null_offset) x 16 );
865 # dump the second-level indexes
867 for ($i = 0; $i < 256; $i++)
869 next unless ($filled_idx[$i] > 256);
870 my @table = @filled[($i<<4)..($i<<4)+15];
871 for ($j = 0; $j < 16; $j++) { $table[$j] ||= $null_offset; }
872 printf OUTPUT ",\n /* sub-index %02x */\n", $i;
873 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
876 # dump the 16-char subsets
878 printf OUTPUT ",\n /* null mapping */\n";
879 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, (0) x 32 );
881 for ($i = 0; $i < 4096; $i++)
883 next unless $filled[$i];
884 my @table = (0) x 32;
885 for ($j = 0; $j < 16; $j++)
887 if (defined $decomp_table[($i<<4) + $j])
889 $table[2 * $j] = ${$decomp_table[($i << 4) + $j]}[0];
890 $table[2 * $j + 1] = ${$decomp_table[($i << 4) + $j]}[1];
893 printf OUTPUT ",\n /* 0x%03x0 .. 0x%03xf */\n", $i, $i;
894 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
897 printf OUTPUT "\n};\n";
902 ################################################################
903 # read an input file and generate the corresponding .c file
906 my ($codepage,$filename,$comment) = @_;
908 # symbol codepage file is special
909 if ($codepage == 42) { READ_SYMBOL_FILE($MAPPREFIX . $filename); }
910 else { READ_FILE($MAPPREFIX . $filename); }
912 ADD_DEFAULT_MAPPINGS();
914 my $output = sprintf "c_%03d.c", $codepage;
915 open OUTPUT,">$output" or die "Cannot create $output";
917 printf "Building %s from %s (%s)\n", $output, $filename, $comment;
921 printf OUTPUT "/* code page %03d (%s) */\n", $codepage, $comment;
922 printf OUTPUT "/* generated from %s */\n", $MAPPREFIX . $filename;
923 printf OUTPUT "/* DO NOT EDIT!! */\n\n";
924 printf OUTPUT "#include \"wine/unicode.h\"\n\n";
926 if ($#lead_bytes == -1) { DUMP_SBCS_TABLE( $codepage, $comment ); }
927 else { DUMP_DBCS_TABLE( $codepage, $comment ); }
932 ################################################################
933 # output the list of codepage tables into the cptable.c file
938 foreach $file (@allfiles)
940 my ($codepage,$filename,$comment) = @$file;
941 push @tables_decl, sprintf("extern union cptable cptable_%03d;\n",$codepage);
944 push @tables_decl, sprintf("\nstatic const union cptable * const cptables[%d] =\n{\n",$#allfiles+1);
945 foreach $file (@allfiles)
947 my ($codepage,$filename,$comment) = @$file;
948 push @tables_decl, sprintf(" &cptable_%03d,\n", $codepage);
950 push @tables_decl, "};";
951 REPLACE_IN_FILE( "cptable.c", @tables_decl );
954 ################################################################
955 # replace the contents of a file between ### cpmap ### marks
962 open(FILE,$name) or die "Can't open $name";
966 last if /\#\#\# cpmap begin \#\#\#/;
971 if (/\#\#\# cpmap end \#\#\#/) { push @lines, "\n", $_; last; }
974 open(FILE,">$name") or die "Can't modify $name";