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., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA
24 # base URLs for www.unicode.org files
25 my $MAPPINGS = "http://www.unicode.org/Public/MAPPINGS";
26 my $UNIDATA = "http://www.unicode.org/Public/11.0.0/ucd";
27 my $REPORTS = "http://www.unicode.org/reports";
28 my $RFCS = "http://www.rfc-editor.org/rfc";
31 my $SORTKEYS = "tr10/allkeys.txt";
33 # RFC3454 (stringprep data)
34 my $STRINGPREP = "rfc3454.txt";
37 my $DEFAULTS = "tools/unicode-defaults";
39 # Default char for undefined mappings
40 my $DEF_CHAR = ord '?';
44 [ 37, "VENDORS/MICSFT/EBCDIC/CP037.TXT", 0, "IBM EBCDIC US Canada", 0, 0x6f ],
45 [ 424, "VENDORS/MISC/CP424.TXT", 0, "IBM EBCDIC Hebrew" ],
46 [ 437, "VENDORS/MICSFT/PC/CP437.TXT", 1, "OEM United States" ],
47 [ 500, "VENDORS/MICSFT/EBCDIC/CP500.TXT", 0, "IBM EBCDIC International", 0, 0x6f ],
48 [ 737, "VENDORS/MICSFT/PC/CP737.TXT", 1, "OEM Greek 437G" ],
49 [ 775, "VENDORS/MICSFT/PC/CP775.TXT", 1, "OEM Baltic" ],
50 [ 850, "VENDORS/MICSFT/PC/CP850.TXT", 1, "OEM Multilingual Latin 1" ],
51 [ 852, "VENDORS/MICSFT/PC/CP852.TXT", 1, "OEM Slovak Latin 2" ],
52 [ 855, "VENDORS/MICSFT/PC/CP855.TXT", 1, "OEM Cyrillic" ],
53 [ 856, "VENDORS/MISC/CP856.TXT", 0, "Hebrew PC" ],
54 [ 857, "VENDORS/MICSFT/PC/CP857.TXT", 1, "OEM Turkish", 0xf8bb ],
55 [ 860, "VENDORS/MICSFT/PC/CP860.TXT", 1, "OEM Portuguese" ],
56 [ 861, "VENDORS/MICSFT/PC/CP861.TXT", 1, "OEM Icelandic" ],
57 [ 862, "VENDORS/MICSFT/PC/CP862.TXT", 1, "OEM Hebrew" ],
58 [ 863, "VENDORS/MICSFT/PC/CP863.TXT", 1, "OEM Canadian French" ],
59 [ 864, "VENDORS/MICSFT/PC/CP864.TXT", 0, "OEM Arabic", 0xf8bb ],
60 [ 865, "VENDORS/MICSFT/PC/CP865.TXT", 1, "OEM Nordic" ],
61 [ 866, "VENDORS/MICSFT/PC/CP866.TXT", 1, "OEM Russian" ],
62 [ 869, "VENDORS/MICSFT/PC/CP869.TXT", 1, "OEM Greek" ],
63 [ 874, "VENDORS/MICSFT/WindowsBestFit/bestfit874.txt", 1, "ANSI/OEM Thai" ],
64 [ 875, "VENDORS/MICSFT/EBCDIC/CP875.TXT", 0, "IBM EBCDIC Greek", 0, 0x6f ],
65 [ 878, "VENDORS/MISC/KOI8-R.TXT", 0, "Russian KOI8" ],
66 [ 932, "VENDORS/MICSFT/WindowsBestFit/bestfit932.txt", 0, "ANSI/OEM Japanese Shift-JIS" ],
67 [ 936, "VENDORS/MICSFT/WindowsBestFit/bestfit936.txt", 0, "ANSI/OEM Simplified Chinese GBK" ],
68 [ 949, "VENDORS/MICSFT/WindowsBestFit/bestfit949.txt", 0, "ANSI/OEM Korean Unified Hangul" ],
69 [ 950, "VENDORS/MICSFT/WindowsBestFit/bestfit950.txt", 0, "ANSI/OEM Traditional Chinese Big5" ],
70 [ 1006, "VENDORS/MISC/CP1006.TXT", 0, "IBM Arabic" ],
71 [ 1026, "VENDORS/MICSFT/EBCDIC/CP1026.TXT", 0, "IBM EBCDIC Latin 5 Turkish", 0, 0x6f ],
72 [ 1250, "VENDORS/MICSFT/WindowsBestFit/bestfit1250.txt", 0, "ANSI Eastern Europe" ],
73 [ 1251, "VENDORS/MICSFT/WindowsBestFit/bestfit1251.txt", 0, "ANSI Cyrillic" ],
74 [ 1252, "VENDORS/MICSFT/WindowsBestFit/bestfit1252.txt", 0, "ANSI Latin 1" ],
75 [ 1253, "VENDORS/MICSFT/WindowsBestFit/bestfit1253.txt", 0, "ANSI Greek" ],
76 [ 1254, "VENDORS/MICSFT/WindowsBestFit/bestfit1254.txt", 0, "ANSI Turkish" ],
77 [ 1255, "VENDORS/MICSFT/WindowsBestFit/bestfit1255.txt", 0, "ANSI Hebrew" ],
78 [ 1256, "VENDORS/MICSFT/WindowsBestFit/bestfit1256.txt", 0, "ANSI Arabic" ],
79 [ 1257, "VENDORS/MICSFT/WindowsBestFit/bestfit1257.txt", 0, "ANSI Baltic" ],
80 [ 1258, "VENDORS/MICSFT/WindowsBestFit/bestfit1258.txt", 0, "ANSI/OEM Viet Nam" ],
81 [ 1361, "VENDORS/MICSFT/WindowsBestFit/bestfit1361.txt", 0, "Korean Johab" ],
82 [ 10000, "VENDORS/MICSFT/MAC/ROMAN.TXT", 0, "Mac Roman" ],
83 [ 10001, "VENDORS/APPLE/JAPANESE.TXT", 0, "Mac Japanese" ],
84 [ 10002, "VENDORS/APPLE/CHINTRAD.TXT", 0, "Mac Traditional Chinese" ],
85 [ 10003, "VENDORS/APPLE/KOREAN.TXT", 0, "Mac Korean", 0xf8e5 ],
86 [ 10004, "VENDORS/APPLE/ARABIC.TXT", 0, "Mac Arabic", 0xf826 ],
87 [ 10005, "VENDORS/APPLE/HEBREW.TXT", 0, "Mac Hebrew", 0xf7fa ],
88 [ 10006, "VENDORS/MICSFT/MAC/GREEK.TXT", 0, "Mac Greek" ],
89 [ 10007, "VENDORS/MICSFT/MAC/CYRILLIC.TXT", 0, "Mac Cyrillic" ],
90 [ 10008, "VENDORS/APPLE/CHINSIMP.TXT", 0, "Mac Simplified Chinese" ],
91 [ 10010, "VENDORS/APPLE/ROMANIAN.TXT", 0, "Mac Romanian" ],
92 [ 10017, "VENDORS/APPLE/CYRILLIC.TXT", 0, "Mac Ukrainian" ],
93 [ 10021, "VENDORS/APPLE/THAI.TXT", 0, "Mac Thai" ],
94 [ 10029, "VENDORS/MICSFT/MAC/LATIN2.TXT", 0, "Mac Latin 2" ],
95 [ 10079, "VENDORS/MICSFT/MAC/ICELAND.TXT", 0, "Mac Icelandic" ],
96 [ 10081, "VENDORS/MICSFT/MAC/TURKISH.TXT", 0, "Mac Turkish", 0xf89f ],
97 [ 10082, "VENDORS/APPLE/CROATIAN.TXT", 0, "Mac Croatian" ],
98 [ 20127, undef, 0, "US-ASCII (7bit)" ],
99 [ 20866, "VENDORS/MISC/KOI8-R.TXT", 0, "Russian KOI8" ],
100 [ 20932, "OBSOLETE/EASTASIA/JIS/JIS0208.TXT", 0, "EUC-JP", 0, 0x3f, 0x30fb ],
101 [ 21866, "VENDORS/MISC/KOI8-U.TXT", 0, "Ukrainian KOI8" ],
102 [ 28591, "ISO8859/8859-1.TXT", 0, "ISO 8859-1 Latin 1" ],
103 [ 28592, "ISO8859/8859-2.TXT", 0, "ISO 8859-2 Latin 2 (East European)" ],
104 [ 28593, "ISO8859/8859-3.TXT", 0, "ISO 8859-3 Latin 3 (South European)", 0xf7f5 ],
105 [ 28594, "ISO8859/8859-4.TXT", 0, "ISO 8859-4 Latin 4 (Baltic old)" ],
106 [ 28595, "ISO8859/8859-5.TXT", 0, "ISO 8859-5 Cyrillic" ],
107 [ 28596, "ISO8859/8859-6.TXT", 0, "ISO 8859-6 Arabic", 0xf7c8 ],
108 [ 28597, "ISO8859/8859-7.TXT", 0, "ISO 8859-7 Greek", 0xf7c5 ],
109 [ 28598, "ISO8859/8859-8.TXT", 0, "ISO 8859-8 Hebrew", 0xf79c ],
110 [ 28599, "ISO8859/8859-9.TXT", 0, "ISO 8859-9 Latin 5 (Turkish)" ],
111 [ 28600, "ISO8859/8859-10.TXT", 0, "ISO 8859-10 Latin 6 (Nordic)" ],
112 [ 28603, "ISO8859/8859-13.TXT", 0, "ISO 8859-13 Latin 7 (Baltic)" ],
113 [ 28604, "ISO8859/8859-14.TXT", 0, "ISO 8859-14 Latin 8 (Celtic)" ],
114 [ 28605, "ISO8859/8859-15.TXT", 0, "ISO 8859-15 Latin 9 (Euro)" ],
115 [ 28606, "ISO8859/8859-16.TXT", 0, "ISO 8859-16 Latin 10 (Balkan)" ]
144 "Avagraha" => 0x0003,
147 "Vowel_Independent" => 0x0006,
148 "Vowel_Dependent" => 0x0007,
150 "Consonant_Placeholder" => 0x0009,
151 "Consonant" => 0x000a,
152 "Consonant_Dead" => 0x000b,
153 "Consonant_Succeeding_Repha" => 0x000c,
154 "Consonant_Subjoined" => 0x000d,
155 "Consonant_Medial" => 0x000e,
156 "Consonant_Final" => 0x000f,
157 "Consonant_Head_Letter" => 0x0010,
158 "Modifying_Letter" => 0x0011,
159 "Tone_Letter" => 0x0012,
160 "Tone_Mark" => 0x0013,
161 "Register_Shifter" => 0x0014,
162 "Consonant_Preceding_Repha" => 0x0015,
163 "Pure_Killer" => 0x0016,
164 "Invisible_Stacker" => 0x0017,
165 "Gemination_Mark" => 0x0018,
166 "Cantillation_Mark" => 0x0019,
167 "Non_Joiner" => 0x001a,
169 "Number_Joiner" => 0x001c,
171 "Brahmi_Joining_Number" => 0x001e,
172 "Consonant_With_Stacker" => 0x001f,
173 "Consonant_Prefixed" => 0x0020,
174 "Syllable_Modifier" => 0x0021,
175 "Consonant_Killer" => 0x0022,
176 "Consonant_Initial_Postfixed" => 0x0023,
183 "Visual_Order_Left" => 0x03,
184 "Left_And_Right" => 0x04,
187 "Top_And_Bottom" => 0x07,
188 "Top_And_Right" => 0x08,
189 "Top_And_Left" => 0x09,
190 "Top_And_Left_And_Right" => 0x0a,
191 "Bottom_And_Right" => 0x0b,
192 "Top_And_Bottom_And_Right" => 0x0c,
193 "Overstruck" => 0x0d,
195 "Bottom_And_Left" => 0x0f
200 "unassigned" => 0x01,
201 "prohibited" => 0x02,
263 "Lu" => $ctype{"defin"}|$ctype{"alpha"}|$ctype{"upper"}, # Letter, Uppercase
264 "Ll" => $ctype{"defin"}|$ctype{"alpha"}|$ctype{"lower"}, # Letter, Lowercase
265 "Lt" => $ctype{"defin"}|$ctype{"alpha"}|$ctype{"upper"}|$ctype{"lower"}, # Letter, Titlecase
266 "Mn" => $ctype{"defin"}, # Mark, Non-Spacing
267 "Mc" => $ctype{"defin"}, # Mark, Spacing Combining
268 "Me" => $ctype{"defin"}, # Mark, Enclosing
269 "Nd" => $ctype{"defin"}|$ctype{"digit"}, # Number, Decimal Digit
270 "Nl" => $ctype{"defin"}|$ctype{"alpha"}, # Number, Letter
271 "No" => $ctype{"defin"}, # Number, Other
272 "Zs" => $ctype{"defin"}|$ctype{"space"}, # Separator, Space
273 "Zl" => $ctype{"defin"}|$ctype{"space"}, # Separator, Line
274 "Zp" => $ctype{"defin"}|$ctype{"space"}, # Separator, Paragraph
275 "Cc" => $ctype{"defin"}|$ctype{"cntrl"}, # Other, Control
276 "Cf" => $ctype{"defin"}|$ctype{"cntrl"}, # Other, Format
277 "Cs" => $ctype{"defin"}, # Other, Surrogate
278 "Co" => $ctype{"defin"}, # Other, Private Use
279 "Cn" => $ctype{"defin"}, # Other, Not Assigned
280 "Lm" => $ctype{"defin"}|$ctype{"alpha"}, # Letter, Modifier
281 "Lo" => $ctype{"defin"}|$ctype{"alpha"}, # Letter, Other
282 "Pc" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Connector
283 "Pd" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Dash
284 "Ps" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Open
285 "Pe" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Close
286 "Pi" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Initial quote
287 "Pf" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Final quote
288 "Po" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Other
289 "Sm" => $ctype{"defin"}, # Symbol, Math
290 "Sc" => $ctype{"defin"}, # Symbol, Currency
291 "Sk" => $ctype{"defin"}, # Symbol, Modifier
292 "So" => $ctype{"defin"} # Symbol, Other
295 # a few characters need additional categories that cannot be determined automatically
296 my %special_categories =
298 "xdigit" => [ ord('0')..ord('9'),ord('A')..ord('F'),ord('a')..ord('f'),
299 0xff10..0xff19
, 0xff21..0xff26
, 0xff41..0xff46
],
300 "space" => [ 0x09..0x0d
, 0x85 ],
301 "blank" => [ 0x09, 0x20, 0xa0, 0x3000, 0xfeff ],
302 "cntrl" => [ 0x070f, 0x200c, 0x200d,
303 0x200e, 0x200f, 0x202a, 0x202b, 0x202c, 0x202d, 0x202e,
304 0x206a, 0x206b, 0x206c, 0x206d, 0x206e, 0x206f, 0xfeff,
305 0xfff9, 0xfffa, 0xfffb ],
306 "punct" => [ 0x24, 0x2b, 0x3c..0x3e
, 0x5e, 0x60, 0x7c, 0x7e, 0xa2..0xbe
,
308 "digit" => [ 0xb2, 0xb3, 0xb9 ],
309 "lower" => [ 0x2071, 0x207f ]
314 "L" => 1, # Left-to-Right
315 "R" => 2, # Right-to-Left
316 "AL" => 12, # Right-to-Left Arabic
317 "EN" => 3, # European Number
318 "ES" => 4, # European Number Separator
319 "ET" => 5, # European Number Terminator
320 "AN" => 6, # Arabic Number
321 "CS" => 7, # Common Number Separator
322 "NSM" => 13, # Non-Spacing Mark
323 "BN" => 14, # Boundary Neutral
324 "B" => 8, # Paragraph Separator
325 "S" => 9, # Segment Separator
326 "WS" => 10, # Whitespace
327 "ON" => 11, # Other Neutrals
328 "LRE" => 15, # Left-to-Right Embedding
329 "LRO" => 15, # Left-to-Right Override
330 "RLE" => 15, # Right-to-Left Embedding
331 "RLO" => 15, # Right-to-Left Override
332 "PDF" => 15, # Pop Directional Format
333 "LRI" => 15, # Left-to-Right Isolate
334 "RLI" => 15, # Right-to-Left Isolate
335 "FSI" => 15, # First Strong Isolate
336 "PDI" => 15 # Pop Directional Isolate
341 "U" => 0, # Non_Joining
342 "T" => 1, # Transparent
343 "R" => 2, # Right_Joining
344 "L" => 3, # Left_Joining
345 "D" => 4, # Dual_Joining
346 "C" => 5, # Join_Causing
352 my @unicode_defaults = ();
353 my @unicode_aliases = ();
354 my @tolower_table = ();
355 my @toupper_table = ();
356 my @digitmap_table = ();
357 my @compatmap_table = ();
358 my @category_table = ();
359 my @joining_table = ();
360 my @direction_table = ();
361 my @decomp_table = ();
362 my @compose_table = ();
374 ################################################################
375 # fetch a unicode.org file and open it
376 sub open_data_file
($$)
378 my ($base, $name) = @_;
379 (my $dir = "data/$name") =~ s
/\
/[^\/]+$//;
381 unless (-f
"data/$name")
383 print "Fetching $base/$name...\n";
384 system "mkdir", "-p", $dir;
385 !system "wget", "-q", "-O", "data/$name", "$base/$name" or die "cannot fetch $base/$name";
387 open FILE
, "<data/$name" or die "cannot open data/$name";
391 ################################################################
392 # read in the defaults file
395 my $filename = shift;
398 # first setup a few default mappings
400 open DEFAULTS
, "$filename" or die "Cannot open $filename";
401 print "Loading $filename\n";
404 next if /^\#/; # skip comments
405 next if /^$/; # skip empty lines
406 if (/^(([0-9a-fA-F]+)(,[0-9a-fA-F]+)*)\s+([0-9a-fA-F]+|'.'|none)\s+(\#.*)?/)
408 my @src = map hex, split /,/,$1;
411 if ($#src > 0) { push @unicode_aliases, \
@src; }
412 next if ($dst eq "none");
413 $dst = ($dst =~ /\'.\'/) ?
ord substr($dst,1,1) : hex $dst;
414 foreach my $src (@src)
416 die "Duplicate value" if defined($unicode_defaults[$src]);
417 $unicode_defaults[$src] = $dst;
421 die "Unrecognized line $_\n";
425 # now build mappings from the decomposition field of the Unicode database
427 my $UNICODE_DATA = open_data_file
( $UNIDATA, "UnicodeData.txt" );
428 while (<$UNICODE_DATA>)
430 # Decode the fields ...
431 my ($code, $name, $cat, $comb, $bidi,
432 $decomp, $dec, $dig, $num, $mirror,
433 $oldname, $comment, $upper, $lower, $title) = split /;/;
437 die "unknown category $cat" unless defined $categories{$cat};
438 die "unknown directionality $bidi" unless defined $directions{$bidi};
440 $category_table[$src] = $categories{$cat};
441 $direction_table[$src] = $directions{$bidi};
442 $joining_table[$src] = $joining_types{"T"} if $cat eq "Mn" || $cat eq "Me" || $cat eq "Cf";
446 $tolower_table[$src] = hex $lower;
450 $toupper_table[$src] = hex $upper;
454 $category_table[$src] |= $ctype{"digit"};
458 $digitmap_table[$src] = ord $dig;
461 # copy the category and direction for everything between First/Last pairs
462 if ($name =~ /, First>/) { $start = $src; }
463 if ($name =~ /, Last>/)
465 while ($start < $src)
467 $category_table[$start] = $category_table[$src];
468 $direction_table[$start] = $direction_table[$src];
473 next if $decomp eq ""; # no decomposition, skip it
475 if ($decomp =~ /^<([a-zA-Z]+)>\s+([0-9a-fA-F]+)$/)
477 # decomposition of the form "<foo> 1234" -> use char if type is known
478 if (($src >= 0xf900 && $src < 0xfb00) || ($src >= 0xfe30 && $src < 0xfffd))
480 # Single char decomposition in the compatibility range
481 $compatmap_table[$src] = hex $2;
483 if ($1 eq "isolated" || $1 eq "final" || $1 eq "initial" || $1 eq "medial")
485 ${joining_forms
{$1}}[hex $2] = $src;
488 next unless ($1 eq "font" ||
499 elsif ($decomp =~ /^<compat>\s+0020\s+([0-9a-fA-F]+)/)
501 # decomposition "<compat> 0020 1234" -> combining accent
504 elsif ($decomp =~ /^([0-9a-fA-F]+)/)
506 # decomposition contains only char values without prefix -> use first char
508 $category_table[$src] |= $category_table[$dst] if defined $category_table[$dst];
509 # store decomposition if it contains two chars
510 if ($decomp =~ /^([0-9a-fA-F]+)\s+([0-9a-fA-F]+)$/)
512 $decomp_table[$src] = [ hex $1, hex $2 ];
513 push @compose_table, [ hex $1, hex $2, $src ];
515 elsif ($decomp =~ /^(<[a-z]+>\s)*([0-9a-fA-F]+)$/ &&
516 (($src >= 0xf900 && $src < 0xfb00) || ($src >= 0xfe30 && $src < 0xfffd)))
518 # Single char decomposition in the compatibility range
519 $compatmap_table[$src] = hex $2;
527 next if defined($unicode_defaults[$src]); # may have been set in the defaults file
530 for (my $i = $dst; ; $i = $unicode_defaults[$i])
532 die sprintf("loop detected for %04x -> %04x",$src,$dst) if $i == $src;
533 last unless defined($unicode_defaults[$i]);
535 $unicode_defaults[$src] = $dst;
539 # patch the category of some special characters
541 foreach my $cat (keys %special_categories)
543 my $flag = $ctype{$cat};
544 foreach my $i (@
{$special_categories{$cat}}) { $category_table[$i] |= $flag; }
549 ################################################################
550 # define a new lead byte
554 push @lead_bytes, $ch;
558 ################################################################
559 # define a new char mapping
563 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
564 $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
565 if ($cp > 0xff && !defined($cp2uni[$cp >> 8]))
567 add_lead_byte
( $cp >> 8 );
571 ################################################################
572 # parse a standard codepage file
573 sub read_codepage_file
($)
576 my $INPUT = open_data_file
( $MAPPINGS, $name );
580 next if /^\#/; # skip comments
581 next if /^$/; # skip empty lines
582 next if /\x1a/; # skip ^Z
583 next if (/^0x([0-9a-fA-F]+)\s+\#UNDEFINED/); # undefined char
585 if (/^0x([0-9a-fA-F]+)\s+\#DBCS LEAD BYTE/)
587 add_lead_byte
( hex $1 );
591 if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\s+(\#.*)?/)
593 add_mapping
( hex $1, hex $2 );
597 if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\+.*\s+(\#.*)?/)
599 add_mapping
( hex $1, hex $2 );
603 if (/^0x([0-9a-fA-F]+)\s+<LR>\+0x([0-9a-fA-F]+)\s+(\#.*)?/)
605 add_mapping
( hex $1, hex $2 );
609 if (/^0x([0-9a-fA-F]+)\s+<RL>\+0x([0-9a-fA-F]+)\s+(\#.*)?/)
612 add_mapping
( hex $1, $uni ) if $uni >= 0x80;
615 die "$name: Unrecognized line $_\n";
621 ################################################################
622 # fill input data for the 20127 (us-ascii) codepage
623 sub fill_20127_codepage
()
625 for (my $i = 0; $i < 128; $i++) { $cp2uni[$i] = $uni2cp[$i] = $i; }
626 for (my $i = 128; $i < 256; $i++) { $cp2uni[$i] = $i & 0x7f; }
629 ################################################################
630 # get a mapping including glyph chars for MB_USEGLYPHCHARS
632 sub get_glyphs_mapping
(@
)
634 $_[0x01] = 0x263a; # (WHITE SMILING FACE)
635 $_[0x02] = 0x263b; # (BLACK SMILING FACE)
636 $_[0x03] = 0x2665; # (BLACK HEART SUIT)
637 $_[0x04] = 0x2666; # (BLACK DIAMOND SUIT)
638 $_[0x05] = 0x2663; # (BLACK CLUB SUIT)
639 $_[0x06] = 0x2660; # (BLACK SPADE SUIT)
640 $_[0x07] = 0x2022; # (BULLET)
641 $_[0x08] = 0x25d8; # (INVERSE BULLET)
642 $_[0x09] = 0x25cb; # (WHITE CIRCLE)
643 $_[0x0a] = 0x25d9; # (INVERSE WHITE CIRCLE)
644 $_[0x0b] = 0x2642; # (MALE SIGN)
645 $_[0x0c] = 0x2640; # (FEMALE SIGN)
646 $_[0x0d] = 0x266a; # (EIGHTH NOTE)
647 $_[0x0e] = 0x266b; # (BEAMED EIGHTH NOTES)
648 $_[0x0f] = 0x263c; # (WHITE SUN WITH RAYS)
649 $_[0x10] = 0x25ba; # (BLACK RIGHT-POINTING POINTER)
650 $_[0x11] = 0x25c4; # (BLACK LEFT-POINTING POINTER)
651 $_[0x12] = 0x2195; # (UP DOWN ARROW)
652 $_[0x13] = 0x203c; # (DOUBLE EXCLAMATION MARK)
653 $_[0x14] = 0x00b6; # (PILCROW SIGN)
654 $_[0x15] = 0x00a7; # (SECTION SIGN)
655 $_[0x16] = 0x25ac; # (BLACK RECTANGLE)
656 $_[0x17] = 0x21a8; # (UP DOWN ARROW WITH BASE)
657 $_[0x18] = 0x2191; # (UPWARDS ARROW)
658 $_[0x19] = 0x2193; # (DOWNWARDS ARROW)
659 $_[0x1a] = 0x2192; # (RIGHTWARDS ARROW)
660 $_[0x1b] = 0x2190; # (LEFTWARDS ARROW)
661 $_[0x1c] = 0x221f; # (RIGHT ANGLE)
662 $_[0x1d] = 0x2194; # (LEFT RIGHT ARROW)
663 $_[0x1e] = 0x25b2; # (BLACK UP-POINTING TRIANGLE)
664 $_[0x1f] = 0x25bc; # (BLACK DOWN-POINTING TRIANGLE)
665 $_[0x7f] = 0x2302; # (HOUSE)
669 ################################################################
670 # build EUC-JP table from the JIS 0208 file
671 # FIXME: for proper EUC-JP we should probably read JIS 0212 too
672 # but this would require 3-byte DBCS characters
673 sub READ_JIS0208_FILE
($)
678 for (my $i = 0x00; $i <= 0x7f; $i++)
684 # JIS X 0201 right plane
685 for (my $i = 0xa1; $i <= 0xdf; $i++)
687 $cp2uni[0x8e00 + $i] = 0xfec0 + $i;
688 $uni2cp[0xfec0 + $i] = 0x8e00 + $i;
692 foreach my $i (0x8e, 0x8f, 0xa1 .. 0xfe)
698 foreach my $i (0x80 .. 0x8d, 0x90 .. 0xa0, 0xff)
700 $cp2uni[$i] = $DEF_CHAR;
703 # Shift-JIS compatibility
704 $uni2cp[0x00a5] = 0x5c;
705 $uni2cp[0x203e] = 0x7e;
707 # Fix backslash conversion
708 $cp2uni[0xa1c0] = 0xff3c;
709 $uni2cp[0xff3c] = 0xa1c0;
711 my $INPUT = open_data_file
( $MAPPINGS, $name );
714 next if /^\#/; # skip comments
715 next if /^$/; # skip empty lines
716 next if /\x1a/; # skip ^Z
717 if (/^0x[0-9a-fA-F]+\s+0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\s+(\#.*)?/)
719 my $cp = 0x8080 + hex $1;
721 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
722 $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
725 die "$name: Unrecognized line $_\n";
731 ################################################################
732 # build the sort keys table
733 sub READ_SORTKEYS_FILE
()
736 for (my $i = 0; $i < 65536; $i++) { $sortkeys[$i] = [ -1, 0, 0, 0, 0 ] };
738 my $INPUT = open_data_file
( $REPORTS, $SORTKEYS );
741 next if /^\#/; # skip comments
742 next if /^$/; # skip empty lines
743 next if /\x1a/; # skip ^Z
744 next if /^\@version/; # skip @version header
745 if (/^([0-9a-fA-F]+)\s+;\s+\[([*.])([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]+)\]/)
747 my ($uni,$variable) = (hex $1, $2);
748 next if $uni > 65535;
749 $sortkeys[$uni] = [ $uni, hex $3, hex $4, hex $5, hex $6 ];
752 if (/^([0-9a-fA-F]+\s+)+;\s+\[[*.]([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]+)\]/)
754 # multiple character sequence, ignored for now
757 die "$SORTKEYS: Unrecognized line $_\n";
761 # compress the keys to 32 bit:
762 # key 1 to 16 bits, key 2 to 8 bits, key 3 to 4 bits, key 4 to 1 bit
764 @sortkeys = sort { ${$a}[1] <=> ${$b}[1] or
765 ${$a}[2] <=> ${$b}[2] or
766 ${$a}[3] <=> ${$b}[3] or
767 ${$a}[4] <=> ${$b}[4] or
768 $a cmp $b; } @sortkeys;
770 my ($n2, $n3) = (1, 1);
771 my @keys = (-1, -1, -1, -1, -1 );
774 for (my $i = 0; $i < 65536; $i++)
776 my @current = @
{$sortkeys[$i]};
777 next if $current[0] == -1;
778 if ($current[1] == $keys[1])
780 if ($current[2] == $keys[2])
782 if ($current[3] == $keys[3])
788 $keys[3] = $current[3];
795 $keys[2] = $current[2];
796 $keys[3] = $current[3];
804 $keys[1] = $current[1];
805 $keys[2] = $current[2];
806 $keys[3] = $current[3];
811 if ($current[2]) { $current[2] = $n2; }
812 if ($current[3]) { $current[3] = $n3; }
813 if ($current[4]) { $current[4] = 1; }
815 $flatkeys[$current[0]] = ($current[1] << 16) | ($current[2] << 8) | ($current[3] << 4) | $current[4];
821 ################################################################
822 # build the sort keys table
823 sub DUMP_SORTKEYS
($@
)
825 my ($filename, @keys) = @_;
827 # count the number of 256-key ranges that contain something
831 for (my $i = 0; $i < 256; $i++) { $offsets[$i] = 256; }
832 for (my $i = 0; $i < 65536; $i++)
834 next unless defined $keys[$i];
835 $offsets[$i >> 8] = $ranges * 256;
840 # output the range offsets
842 open OUTPUT
,">$filename.new" or die "Cannot create $filename";
843 printf "Building $filename\n";
844 printf OUTPUT
"/* Unicode collation element table */\n";
845 printf OUTPUT
"/* generated from %s */\n", "$REPORTS/$SORTKEYS";
846 printf OUTPUT
"/* DO NOT EDIT!! */\n\n";
848 printf OUTPUT
"const unsigned int collation_table[%d] =\n{\n", $ranges*256;
849 printf OUTPUT
" /* index */\n";
850 printf OUTPUT
"%s,\n", DUMP_ARRAY
( "0x%08x", 0, @offsets );
852 # output the default values
854 printf OUTPUT
" /* defaults */\n";
855 printf OUTPUT
"%s", DUMP_ARRAY
( "0x%08x", 0, (0xffffffff) x
256 );
857 # output all the key ranges
859 for (my $i = 0; $i < 256; $i++)
861 next if $offsets[$i] == 256;
862 printf OUTPUT
",\n /* 0x%02x00 .. 0x%02xff */\n", $i, $i;
863 printf OUTPUT
"%s", DUMP_ARRAY
( "0x%08x", 0xffffffff, @keys[($i<<8) .. ($i<<8)+255] );
865 printf OUTPUT
"\n};\n";
867 save_file
($filename);
871 ################################################################
872 # add default mappings once the file had been read
873 sub add_default_mappings
($)
875 my $first_private = shift;
879 foreach my $alias (@unicode_aliases)
882 foreach my $src (@
$alias)
884 if (defined($uni2cp[$src]))
886 $target = $uni2cp[$src];
890 next unless defined($target);
892 # At least one char of the alias set is defined, set the others to the same value
893 foreach my $src (@
$alias)
895 $uni2cp[$src] = $target unless defined($uni2cp[$src]);
899 # For every src -> target mapping in the defaults table,
900 # make uni2cp[src] = uni2cp[target] if uni2cp[target] is defined
902 for (my $src = 0; $src < 65536; $src++)
904 next if defined($uni2cp[$src]); # source has a definition already
905 next unless defined($unicode_defaults[$src]); # no default for this char
906 my $target = $unicode_defaults[$src];
908 # do a recursive mapping until we find a target char that is defined
909 while (!defined($uni2cp[$target]) &&
910 defined($unicode_defaults[$target])) { $target = $unicode_defaults[$target]; }
912 if (defined($uni2cp[$target])) { $uni2cp[$src] = $uni2cp[$target]; }
915 # Add a private char mapping for undefined chars
919 for (my $i = 32; $i < 256; $i++)
921 next if defined($cp2uni[$i]);
922 $cp2uni[$i] = $first_private;
923 $uni2cp[$first_private] = $i;
928 # Add an identity mapping for all undefined chars
930 for (my $i = 0; $i < 256; $i++)
932 next if defined($cp2uni[$i]);
933 next if defined($uni2cp[$i]);
934 $cp2uni[$i] = $uni2cp[$i] = $i;
938 ################################################################
939 # dump an array of integers
942 my ($format,$default,@array) = @_;
945 for ($i = 0; $i < $#array; $i++)
947 $ret .= sprintf($format, defined $array[$i] ?
$array[$i] : $default);
948 $ret .= (($i % 8) != 7) ?
", " : ",\n ";
950 $ret .= sprintf($format, defined $array[$i] ?
$array[$i] : $default);
955 ################################################################
956 # dump a unicode->ascii mapping table
957 sub dump_uni2cp_table
($$)
959 my ($width, $def) = @_;
961 # count the number of unicode->ascii subtables that contain something
966 my @filled = (-1) x
256;
967 for (my $i = 0; $i < 65536; $i++)
969 next unless defined $uni2cp[$i];
970 if ($filled[$i >> 8] == -1)
972 $lowerbounds[$index] = $i & 0xff;
973 $upperbounds[$index] = 0xff - $lowerbounds[$index];
974 $filled[$i >> 8] = $index * 256;
979 $upperbounds[$index-1] = 0xff - ($i & 0xff);
983 # add defaults mapping at the end
984 $filled[256] = $index * 256;
985 $lowerbounds[$index] = 255;
986 $upperbounds[$index] = 0;
989 # collapse blocks upwards if possible
992 for (my $i = 0; $i < 257; $i++)
994 next if $filled[$i] == -1;
995 if ($upperbounds[$index - 1] > $lowerbounds[$index])
997 $removed += $lowerbounds[$index];
1001 $removed += $upperbounds[$index - 1];
1002 $lowerbounds[$index] = $upperbounds[$index - 1];
1004 $filled[$i] -= $removed;
1008 # output all the subtables into a single array
1010 printf OUTPUT
"static const unsigned %s uni2cp_low[%d] =\n{\n",
1011 $width == 8 ?
"char" : "short", ($index + 1) * 256 - $removed;
1013 my $format = $width == 8 ?
"0x%02x" : "0x%04x";
1014 for (my $i = $index = 0; $i < 257; $i++)
1016 next if $filled[$i] == -1;
1017 my $start = ($i << 8) + $lowerbounds[$index];
1018 my $end = ($i << 8) + 255;
1021 print OUTPUT
" /* defaults */\n";
1022 printf OUTPUT
"%s\n};\n\n", DUMP_ARRAY
( $format, 0, ($def) x
($end - $start + 1) );
1026 printf OUTPUT
" /* 0x%04x .. 0x%04x */\n", $start, $end;
1027 printf OUTPUT
"%s,\n", DUMP_ARRAY
( $format, $def, @uni2cp[$start .. $end] );
1032 # output a table of the offsets of the subtables in the previous array
1034 for (my $i = 0; $i < 256; $i++)
1036 if ($filled[$i] == -1) { $filled[$i] = $filled[256]; }
1038 printf OUTPUT
"static const unsigned short uni2cp_high[256] =\n";
1039 printf OUTPUT
"{\n%s\n};\n\n", DUMP_ARRAY
( "0x%04x", 0, @filled[0..255] );
1042 ################################################################
1043 # dump an SBCS mapping table
1044 sub dump_sbcs_table
($$$$$)
1046 my ($codepage, $has_glyphs, $name, $def, $defw) = @_;
1049 # output the ascii->unicode table
1053 printf OUTPUT
"static const WCHAR cp2uni[512] =\n";
1054 printf OUTPUT
"{\n%s", DUMP_ARRAY
( "0x%04x", $defw, @cp2uni[0 .. 255] );
1055 printf OUTPUT
",\n /* glyphs */\n%s\n};\n\n",
1056 DUMP_ARRAY
( "0x%04x", $defw, get_glyphs_mapping
(@cp2uni[0 .. 255]) );
1060 printf OUTPUT
"static const WCHAR cp2uni[256] =\n";
1061 printf OUTPUT
"{\n%s\n};\n\n", DUMP_ARRAY
( "0x%04x", $defw, @cp2uni[0 .. 255] );
1064 dump_uni2cp_table
( 8, $def );
1066 # output the code page descriptor
1068 printf OUTPUT
"const struct sbcs_table DECLSPEC_HIDDEN cptable_%03d =\n{\n", $codepage;
1069 printf OUTPUT
" { %d, 1, 0x%04x, 0x%04x, \"%s\" },\n",
1070 $codepage, $def, $defw, $name;
1071 printf OUTPUT
" cp2uni,\n";
1072 if ($has_glyphs) { printf OUTPUT
" cp2uni + 256,\n"; }
1073 else { printf OUTPUT
" cp2uni,\n"; }
1074 printf OUTPUT
" uni2cp_low,\n";
1075 printf OUTPUT
" uni2cp_high\n};\n";
1079 ################################################################
1080 # dump a DBCS mapping table
1081 sub dump_dbcs_table
($$$$@
)
1083 my ($codepage, $name, $def, $defw, @lb_ranges) = @_;
1085 # build a list of lead bytes that are actually used
1088 LBLOOP
: for (my $y = 0; $y <= $#lead_bytes; $y++)
1090 $cp2uni[$lead_bytes[$y]] = $defw;
1091 my $base = $lead_bytes[$y] << 8;
1092 for (my $x = 0; $x < 256; $x++)
1094 if (defined $cp2uni[$base+$x])
1096 push @lblist,$lead_bytes[$y];
1101 my $unused = ($#lead_bytes > $#lblist);
1103 # output the ascii->unicode table for the single byte chars
1105 printf OUTPUT
"static const WCHAR cp2uni[%d] =\n", 256 * ($#lblist + 2 + $unused);
1106 printf OUTPUT
"{\n%s,\n", DUMP_ARRAY
( "0x%04x", $defw, @cp2uni[0 .. 255] );
1108 # output the default table for unused lead bytes
1112 printf OUTPUT
" /* unused lead bytes */\n";
1113 printf OUTPUT
"%s,\n", DUMP_ARRAY
( "0x%04x", 0, ($defw) x
256 );
1116 # output the ascii->unicode table for each DBCS lead byte
1118 for (my $y = 0; $y <= $#lblist; $y++)
1120 my $base = $lblist[$y] << 8;
1121 printf OUTPUT
" /* lead byte %02x */\n", $lblist[$y];
1122 printf OUTPUT
"%s", DUMP_ARRAY
( "0x%04x", $defw, @cp2uni[$base .. $base+255] );
1123 printf OUTPUT
($y < $#lblist) ?
",\n" : "\n};\n\n";
1126 # output the lead byte subtables offsets
1128 my @offsets = (0) x
256;
1129 for (my $x = 0; $x <= $#lblist; $x++) { $offsets[$lblist[$x]] = $x + 1; }
1132 # increment all lead bytes offset to take into account the unused table
1133 for (my $x = 0; $x <= $#lead_bytes; $x++) { $offsets[$lead_bytes[$x]]++; }
1135 printf OUTPUT
"static const unsigned char cp2uni_leadbytes[256] =\n";
1136 printf OUTPUT
"{\n%s\n};\n\n", DUMP_ARRAY
( "0x%02x", 0, @offsets );
1138 dump_uni2cp_table
( 16, $def );
1140 # output the code page descriptor
1142 printf OUTPUT
"const struct dbcs_table DECLSPEC_HIDDEN cptable_%03d =\n{\n", $codepage;
1143 printf OUTPUT
" { %d, 2, 0x%04x, 0x%04x, \"%s\" },\n",
1144 $codepage, $def, $defw, $name;
1145 printf OUTPUT
" cp2uni,\n";
1146 printf OUTPUT
" cp2uni_leadbytes,\n";
1147 printf OUTPUT
" uni2cp_low,\n";
1148 printf OUTPUT
" uni2cp_high,\n";
1149 printf OUTPUT
" {\n %s\n }\n", DUMP_ARRAY
( "0x%02x", 0, @lb_ranges, 0, 0 );
1150 printf OUTPUT
"};\n";
1154 ################################################################
1155 # get the list of defined lead byte ranges
1161 foreach $i (@lead_bytes) { $list[$i] = 1; }
1163 for (my $i = 0; $i < 256; $i++)
1167 if (!defined $list[$i]) { push @ranges, $i-1; $on = 0; }
1171 if ($list[$i]) { push @ranges, $i; $on = 1; }
1174 if ($on) { push @ranges, 0xff; }
1178 ################################################################
1179 # dump the Indic Syllabic Category table
1182 my $filename = shift;
1185 my $INPUT = open_data_file
( $UNIDATA, "IndicSyllabicCategory.txt" );
1188 next if /^\#/; # skip comments
1189 next if /^\s*$/; # skip empty lines
1190 next if /\x1a/; # skip ^Z
1191 if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*#/)
1194 die "unknown indic $type" unless defined $indic_types{$type};
1197 $indic_table[hex $1] = $indic_types{$type};
1201 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([A-Za-z_]+)\s*#/)
1204 die "unknown indic $type" unless defined $indic_types{$type};
1205 if (hex $1 < 65536 and hex $2 < 65536)
1207 foreach my $i (hex $1 .. hex $2)
1209 $indic_table[$i] = $indic_types{$type};
1214 die "malformed line $_";
1218 $INPUT = open_data_file
( $UNIDATA, "IndicPositionalCategory.txt" );
1221 next if /^\#/; # skip comments
1222 next if /^\s*$/; # skip empty lines
1223 next if /\x1a/; # skip ^Z
1224 if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z]+)\s*#/)
1227 die "unknown matra $type" unless defined $matra_types{$type};
1228 $indic_table[hex $1] += $matra_types{$type} << 8;
1231 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([A-Za-z_]+)\s*#/)
1234 die "unknown matra $type" unless defined $matra_types{$type};
1235 foreach my $i (hex $1 .. hex $2)
1237 $indic_table[$i] += $matra_types{$type} << 8;
1241 die "malformed line $_";
1245 open OUTPUT
,">$filename.new" or die "Cannot create $filename";
1246 print "Building $filename\n";
1247 print OUTPUT
"/* Unicode Indic Syllabic Category */\n";
1248 print OUTPUT
"/* generated from $UNIDATA/IndicSyllabicCategory.txt */\n";
1249 print OUTPUT
"/* and from $UNIDATA/IndicPositionalCategory.txt */\n";
1250 print OUTPUT
"/* DO NOT EDIT!! */\n\n";
1251 print OUTPUT
"#include \"wine/unicode.h\"\n\n";
1253 dump_two_level_mapping
( "indic_syllabic_table", $indic_types{'Other'}, @indic_table );
1256 save_file
($filename);
1259 ################################################################
1260 # dump the Line Break Properties table
1261 sub dump_linebreak
($)
1263 my $filename = shift;
1267 my $INPUT = open_data_file
( $UNIDATA, "LineBreak.txt" );
1270 next if /^\#/; # skip comments
1271 next if /^\s*$/; # skip empty lines
1272 next if /\x1a/; # skip ^Z
1273 if (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z][0-9A-Z])+\s*/)
1276 die "unknown breaktype $type" unless defined $break_types{$type};
1277 $break_table[hex $1] = $break_types{$type};
1280 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z][0-9A-Z])+\s*/)
1283 die "unknown breaktype $type" unless defined $break_types{$type};
1284 foreach my $i (hex $1 .. hex $2)
1286 $break_table[$i] = $break_types{$type};
1290 elsif (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z])+\s*/)
1293 die "unknown breaktype $type" unless defined $break_types{$type};
1294 $break_table[hex $1] = $break_types{$type};
1297 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z])+\s*/)
1300 die "unknown breaktype $type" unless defined $break_types{$type};
1301 foreach my $i (hex $1 .. hex $2)
1303 $break_table[$i] = $break_types{$type};
1307 die "malformed line $_";
1311 open OUTPUT
,">$filename.new" or die "Cannot create $filename";
1312 print "Building $filename\n";
1313 print OUTPUT
"/* Unicode Line Break Properties */\n";
1314 print OUTPUT
"/* generated from $UNIDATA/LineBreak.txt */\n";
1315 print OUTPUT
"/* DO NOT EDIT!! */\n\n";
1316 print OUTPUT
"#include \"wine/unicode.h\"\n\n";
1318 dump_two_level_mapping
( "wine_linebreak_table", $break_types{'XX'}, @break_table );
1321 save_file
($filename);
1341 "Canadian_Aboriginal" => 15,
1351 "Egyptian_Hieroglyphs" => 25,
1364 "Imperial_Aramaic" => 38,
1365 "Inscriptional_Pahlavi" => 39,
1366 "Inscriptional_Parthian" => 40,
1384 "Meetei_Mayek" => 58,
1387 "New_Tai_Lue" => 61,
1392 "Old_Persian" => 66,
1393 "Old_South_Arabian" => 67,
1406 "Syloti_Nagri" => 80,
1424 "Meroitic_Cursive" => 97,
1425 "Meroitic_Hieroglyphs" => 98,
1428 "Sora_Sompeng" => 101,
1432 "Caucasian_Albanian" => 104,
1440 "Manichaean" => 112,
1441 "Mende_Kikakui" => 113,
1445 "Old_North_Arabian" => 117,
1446 "Old_Permic" => 118,
1447 "Pahawh_Hmong" => 119,
1449 "Pau_Cin_Hau" => 121,
1450 "Psalter_Pahlavi" => 122,
1453 "Warang_Citi" => 125,
1457 "Anatolian_Hieroglyphs" => 128,
1463 "Old_Hungarian" => 134,
1465 "SignWriting" => 136,
1468 "Masaram_Gondi" => 138,
1471 "Zanabazar_Square" => 141,
1474 ################################################################
1475 # dump Script IDs table
1478 my $filename = shift;
1479 my $header = $filename;
1484 my $INPUT = open_data_file
( $UNIDATA, "Scripts.txt" );
1486 # Unknown script id is always 0, so undefined scripts are automatically treated as such
1491 next if /^\#/; # skip comments
1492 next if /^\s*$/; # skip empty lines
1493 next if /\x1a/; # skip ^Z
1494 if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
1497 if (defined $scripts{$type})
1499 $scripts_table[hex $1] = $scripts{$type};
1503 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
1506 if (defined $scripts{$type})
1508 foreach my $i (hex $1 .. hex $2)
1510 $scripts_table[$i] = $scripts{$type};
1519 $header = "$filename.h";
1520 open OUTPUT
,">$header.new" or die "Cannot create $header";
1521 print "Building $header\n";
1522 print OUTPUT
"/* Unicode Script IDs */\n";
1523 print OUTPUT
"/* generated from $UNIDATA/Scripts.txt */\n";
1524 print OUTPUT
"/* DO NOT EDIT!! */\n\n";
1526 print OUTPUT
"enum unicode_script_id {\n";
1527 foreach my $script (sort { $scripts{$a} <=> $scripts{$b} } keys %scripts)
1529 print OUTPUT
" Script_$script = $scripts{$script},\n";
1531 print OUTPUT
" Script_LastId = ", (scalar keys %scripts) - 1, "\n";
1532 print OUTPUT
"};\n";
1537 $filename = "$filename.c";
1538 open OUTPUT
,">$filename.new" or die "Cannot create $header";
1539 print "Building $filename\n";
1540 print OUTPUT
"/* Unicode Script IDs */\n";
1541 print OUTPUT
"/* generated from $UNIDATA/Scripts.txt */\n";
1542 print OUTPUT
"/* DO NOT EDIT!! */\n\n";
1543 print OUTPUT
"#include \"wine/unicode.h\"\n\n";
1545 dump_two_level_mapping
( "wine_scripts_table", 0, @scripts_table );
1547 save_file
($filename);
1550 ################################################################
1551 # dump the BiDi mirroring table
1552 sub dump_mirroring
($)
1554 my $filename = shift;
1555 my @mirror_table = ();
1557 my $INPUT = open_data_file
( $UNIDATA, "BidiMirroring.txt" );
1560 next if /^\#/; # skip comments
1561 next if /^$/; # skip empty lines
1562 next if /\x1a/; # skip ^Z
1563 if (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9a-fA-F]+)/)
1565 $mirror_table[hex $1] = hex $2;
1568 die "malformed line $_";
1572 open OUTPUT
,">$filename.new" or die "Cannot create $filename";
1573 print "Building $filename\n";
1574 print OUTPUT
"/* Unicode BiDi mirroring */\n";
1575 print OUTPUT
"/* generated from $UNIDATA/BidiMirroring.txt */\n";
1576 print OUTPUT
"/* DO NOT EDIT!! */\n\n";
1577 print OUTPUT
"#include \"wine/unicode.h\"\n\n";
1578 DUMP_CASE_TABLE
( "DECLSPEC_HIDDEN wine_mirror_map", @mirror_table );
1580 save_file
($filename);
1583 ################################################################
1584 # dump the Bidi Brackets
1587 my $filename = shift;
1590 my $INPUT = open_data_file
( $UNIDATA, "BidiBrackets.txt" );
1593 next if /^\#/; # skip comments
1594 next if /^\s*$/; # skip empty lines
1595 next if /\x1a/; # skip ^Z
1596 if (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9a-fA-F]+);\s*([con])/)
1599 die "unknown bracket $type" unless defined $bracket_types{$type};
1600 die "characters too distant $1 and $2" if abs(hex($2) - hex($1)) >= 128;
1601 $bracket_table[hex $1] = (hex($2) - hex($1)) % 255;
1602 $bracket_table[hex $1] += $bracket_types{$type} << 8;
1605 die "malformed line $_";
1609 open OUTPUT
,">$filename.new" or die "Cannot create $filename";
1610 print "Building $filename\n";
1611 print OUTPUT
"/* Unicode Bidirectional Bracket table */\n";
1612 print OUTPUT
"/* generated from $UNIDATA/BidiBrackets.txt */\n";
1613 print OUTPUT
"/* DO NOT EDIT!! */\n\n";
1614 print OUTPUT
"#include \"wine/unicode.h\"\n\n";
1616 dump_two_level_mapping
( "bidi_bracket_table", 0, @bracket_table );
1619 save_file
($filename);
1622 ################################################################
1623 # dump the Arabic shaping table
1626 my $filename = shift;
1630 $groups{"No_Joining_Group"} = $next_group++;
1632 my $INPUT = open_data_file
( $UNIDATA, "ArabicShaping.txt" );
1635 next if /^\#/; # skip comments
1636 next if /^\s*$/; # skip empty lines
1637 next if /\x1a/; # skip ^Z
1638 if (/^\s*([0-9a-fA-F]+)\s*;.*;\s*([RLDCUT])\s*;\s*(\w+)/)
1642 $groups{$group} = $next_group++ unless defined $groups{$group};
1643 $joining_table[hex $1] = $joining_types{$type} | ($groups{$group} << 8);
1646 die "malformed line $_";
1650 open OUTPUT
,">$filename.new" or die "Cannot create $filename";
1651 print "Building $filename\n";
1652 print OUTPUT
"/* Unicode Arabic shaping */\n";
1653 print OUTPUT
"/* generated from $UNIDATA/ArabicShaping.txt */\n";
1654 print OUTPUT
"/* DO NOT EDIT!! */\n\n";
1655 print OUTPUT
"#include \"wine/unicode.h\"\n\n";
1657 dump_two_level_mapping
( "wine_shaping_table", 0, @joining_table );
1659 print OUTPUT
"\nconst unsigned short DECLSPEC_HIDDEN wine_shaping_forms[256][4] =\n{\n";
1660 for (my $i = 0x600; $i <= 0x6ff; $i++)
1662 printf OUTPUT
" { 0x%04x, 0x%04x, 0x%04x, 0x%04x },\n",
1663 ${joining_forms
{"isolated"}}[$i] || $i,
1664 ${joining_forms
{"final"}}[$i] || $i,
1665 ${joining_forms
{"initial"}}[$i] || $i,
1666 ${joining_forms
{"medial"}}[$i] || $i;
1668 print OUTPUT
"};\n";
1671 save_file
($filename);
1674 ################################################################
1675 # dump the Vertical Orientation table
1676 sub dump_vertical
($)
1678 my $filename = shift;
1681 my $INPUT = open_data_file
( $UNIDATA, "VerticalOrientation.txt" );
1684 next if /^\#/; # skip comments
1685 next if /^\s*$/; # skip empty lines
1686 next if /\x1a/; # skip ^Z
1687 if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
1690 die "unknown vertical $type" unless defined $vertical_types{$type};
1693 $vertical_table[hex $1] = $vertical_types{$type};
1697 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([A-Za-z_]+)\s*/)
1700 die "unknown vertical $type" unless defined $vertical_types{$type};
1701 foreach my $i (hex $1 .. hex $2)
1703 $vertical_table[$i] = $vertical_types{$type};
1707 die "malformed line $_";
1711 open OUTPUT
,">$filename.new" or die "Cannot create $filename";
1712 print "Building $filename\n";
1713 print OUTPUT
"/* Unicode Vertical Orientation */\n";
1714 print OUTPUT
"/* generated from $UNIDATA/VerticalOrientation.txt */\n";
1715 print OUTPUT
"/* DO NOT EDIT!! */\n\n";
1716 print OUTPUT
"#include \"wine/unicode.h\"\n\n";
1718 dump_two_level_mapping
( "vertical_orientation_table", $vertical_types{'R'}, @vertical_table );
1721 save_file
($filename);
1724 ################################################################
1725 # dump the digit folding tables
1726 sub dump_digit_folding
($)
1728 my $filename = shift;
1729 open OUTPUT
,">$filename.new" or die "Cannot create $filename";
1730 print "Building $filename\n";
1731 print OUTPUT
"/* Unicode digit folding mappings */\n";
1732 print OUTPUT
"/* generated from $UNIDATA/UnicodeData.txt */\n";
1733 print OUTPUT
"/* DO NOT EDIT!! */\n\n";
1734 print OUTPUT
"#include \"wine/unicode.h\"\n\n";
1736 DUMP_CASE_TABLE
( "DECLSPEC_HIDDEN wine_digitmap", @digitmap_table );
1738 DUMP_CASE_TABLE
( "DECLSPEC_HIDDEN wine_compatmap", @compatmap_table );
1740 save_file
($filename);
1744 ################################################################
1745 # dump the case mapping tables
1746 sub DUMP_CASE_MAPPINGS
($)
1748 my $filename = shift;
1749 open OUTPUT
,">$filename.new" or die "Cannot create $filename";
1750 print "Building $filename\n";
1751 print OUTPUT
"/* Unicode case mappings */\n";
1752 print OUTPUT
"/* generated from $UNIDATA/UnicodeData.txt */\n";
1753 print OUTPUT
"/* DO NOT EDIT!! */\n\n";
1754 print OUTPUT
"#include \"wine/unicode.h\"\n\n";
1756 DUMP_CASE_TABLE
( "wine_casemap_lower", @tolower_table );
1758 DUMP_CASE_TABLE
( "wine_casemap_upper", @toupper_table );
1760 save_file
($filename);
1764 ################################################################
1765 # dump a case mapping table
1766 sub DUMP_CASE_TABLE
($@
)
1768 my ($name,@table) = @_;
1770 # count the number of sub tables that contain something
1771 # also compute the low and upper populated bounds
1773 my @lowerbounds = ( 0, 0 );
1774 my @upperbounds = ( 0, 255 );
1777 for (my $i = 0; $i < 65536; $i++)
1779 next unless defined $table[$i];
1780 if (!defined $filled[$i >> 8])
1782 $lowerbounds[$index] = $i & 0xff;
1783 $upperbounds[$index] = 0xff - $lowerbounds[$index];
1784 $filled[$i >> 8] = $index * 256 + 512;
1789 $upperbounds[$index-1] = 0xff - ($i & 0xff);
1791 $table[$i] = ($table[$i] - $i) & 0xffff;
1794 # Collapse blocks upwards if possible
1797 for (my $i = 0; $i < 256; $i++)
1799 next unless defined $filled[$i];
1800 if ($upperbounds[$index - 1] > $lowerbounds[$index])
1802 $removed = $removed + $lowerbounds[$index];
1806 $removed = $removed + $upperbounds[$index - 1];
1807 $lowerbounds[$index] = $upperbounds[$index - 1];
1809 $filled[$i] = $filled[$i] - $removed;
1815 printf OUTPUT
"const WCHAR %s[%d] =\n", $name, $index * 256 + 512 - $removed;
1816 printf OUTPUT
"{\n /* index */\n";
1817 printf OUTPUT
"%s,\n", DUMP_ARRAY
( "0x%04x", 256, @filled );
1818 printf OUTPUT
" /* defaults */\n";
1819 printf OUTPUT
"%s", DUMP_ARRAY
( "0x%04x", 0, (0) x
256 );
1821 for (my $i = 0; $i < 256; $i++)
1823 next unless $filled[$i];
1824 printf OUTPUT
",\n /* 0x%02x%02x .. 0x%02xff */\n", $i, $lowerbounds[$index], $i;
1825 printf OUTPUT
"%s", DUMP_ARRAY
( "0x%04x", 0,
1826 @table[($i<<8) + $lowerbounds[$index] .. ($i<<8)+255] );
1829 printf OUTPUT
"\n};\n";
1832 ################################################################
1833 # compress a mapping table by removing identical rows
1834 sub compress_array
($$@
)
1839 my $len = @table / $rows;
1840 my @array = (0) x
$rows;
1843 # try to merge table rows
1844 for (my $row = 0; $row < $rows; $row++)
1846 my @table_row = map { defined($_) ?
$_ : $def; } @table[($row * $len)..($row * $len + $len - 1)];
1847 my $rowtxt = pack "S*", @table_row;
1848 if (defined($sequences{$rowtxt}))
1850 # reuse an existing row
1851 $array[$row] = $sequences{$rowtxt};
1856 $sequences{$rowtxt} = $array[$row] = $#array + 1;
1857 push @array, @table_row;
1863 ################################################################
1864 # dump a simple char -> 16-bit value mapping table
1865 sub dump_simple_mapping
($$@
)
1869 my @array = compress_array
( 256, $def, @_[0..65535] );
1871 printf OUTPUT
"const unsigned short %s[%d] =\n{\n", $name, $#array+1;
1872 printf OUTPUT
" /* offsets */\n%s,\n", DUMP_ARRAY
( "0x%04x", 0, @array[0..255] );
1873 printf OUTPUT
" /* values */\n%s\n};\n", DUMP_ARRAY
( "0x%04x", 0, @array[256..$#array] );
1876 ################################################################
1877 # dump a char -> 16-bit value mapping table using two-level tables
1878 sub dump_two_level_mapping
($$@
)
1882 my @row_array = compress_array
( 4096, $def, @_[0..65535] );
1883 my @array = compress_array
( 256, 0, @row_array[0..4095] );
1885 for (my $i = 256; $i < @array; $i++) { $array[$i] += @array - 4096; }
1887 printf OUTPUT
"const unsigned short DECLSPEC_HIDDEN %s[%d] =\n{\n", $name, @array + @row_array - 4096;
1888 printf OUTPUT
" /* level 1 offsets */\n%s,\n", DUMP_ARRAY
( "0x%04x", 0, @array[0..255] );
1889 printf OUTPUT
" /* level 2 offsets */\n%s,\n", DUMP_ARRAY
( "0x%04x", 0, @array[256..$#array] );
1890 printf OUTPUT
" /* values */\n%s\n};\n", DUMP_ARRAY
( "0x%04x", 0, @row_array[4096..$#row_array] );
1893 ################################################################
1894 # dump a binary case mapping table in l_intl.nls format
1895 sub dump_binary_case_table
(@
)
1899 my %difftables_hash = ();
1901 my %offtables2_hash = ();
1902 my @offtables2 = ();
1905 for (my $i = 0; $i < 256; $i++)
1908 for(my $j = 0; $j < 16; $j++) # offset table for xx00-xxFF characters
1911 for (my $k = 0; $k < 16; $k++) # case map table for xxx0-xxxF characters
1913 my $char = ($i<<8) + ($j<<4) + $k;
1914 $difftable[$k] = (defined $table[$char]) ?
(($table[$char]-$char) & 0xffff) : 0;
1917 my $diff_key = pack "S*", @difftable;
1918 my $offset3 = $difftables_hash{$diff_key};
1919 if (!defined $offset3)
1921 $offset3 = scalar @difftables;
1922 $difftables_hash{$diff_key} = $offset3;
1923 push @difftables, @difftable;
1925 $offtable2[$j] = $offset3;
1928 my $offtable2_key = pack "S*", @offtable2;
1929 my $offset2 = $offtables2_hash{$offtable2_key};
1930 if (!defined $offset2)
1932 $offset2 = scalar @offtables2;
1933 $offtables2_hash{$offtable2_key} = $offset2;
1934 push @offtables2, \
@offtable2;
1936 $offtable[$i] = $offset2;
1940 my $offset = 0x100; # offset of first subtable in words
1943 push @output, 0x10 * $_ + $offset; # offset of subtable in words
1946 $offset = 0x100 + 0x10 * scalar @offtables2; # offset of first difftable in words
1947 foreach(@offtables2)
1952 push @output, $_ + $offset; # offset of difftable in words
1956 my $len = 1 + scalar @output + scalar @difftables;
1957 return pack "S<*", $len, @output, @difftables;
1961 ################################################################
1962 # dump case mappings for l_intl.nls
1963 sub dump_intl_nls
($)
1965 my $filename = shift;
1966 open OUTPUT
,">$filename.new" or die "Cannot create $filename";
1967 printf "Building $filename\n";
1970 print OUTPUT
pack "S<", 1; # version
1971 print OUTPUT dump_binary_case_table
( @toupper_table );
1972 print OUTPUT dump_binary_case_table
( @tolower_table );
1974 save_file
($filename);
1978 sub load_nameprep_range_table
($$$)
1980 my ($INPUT, $val, $table_ref) = @_;
1984 if (/^\s*([0-9a-fA-F]+)-([0-9a-fA-F]+)/)
1987 $last = 65535 if($last >= 65536);
1988 foreach my $i (hex $1 .. $last)
1990 $table_ref->[$i] |= $val;
1994 elsif (/^\s*([0-9a-fA-F]+)/)
1998 $table_ref->[hex $1] |= $val;
2003 return if (/End\sTable/);
2007 sub load_nameprep_map_table
($$)
2009 my ($INPUT, $table_ref) = @_;
2013 if (/^\s*([0-9a-fA-F]+);\s;/)
2015 # special value for map to nothing
2016 $table_ref->[hex $1] = [0xffff, 0xffff, 0xffff];
2019 elsif (/^\s*([0-9a-fA-F]+);\s([0-9a-fA-F]+);/)
2021 $table_ref->[hex $1] = [hex $2, 0, 0];
2024 elsif (/^\s*([0-9a-fA-F]+);\s([0-9a-fA-F]+)\s([0-9a-fA-F]+);/)
2026 $table_ref->[hex $1] = [hex $2, hex $3, 0];
2029 elsif (/^\s*([0-9a-fA-F]+);\s([0-9a-fA-F]+)\s([0-9a-fA-F]+)\s([0-9a-fA-F]+);/)
2031 $table_ref->[hex $1] = [hex $2, hex $3, hex $4];
2035 return if (/End\sTable/);
2039 ################################################################
2040 # dump mapping table, prohibited characters set, unassigned
2041 # characters, bidirectional rules used by nameprep algorithm
2042 sub dump_nameprep
($)
2044 my $filename = shift;
2045 my @mapping_table = ();
2048 my $INPUT = open_data_file
( $RFCS, $STRINGPREP );
2051 next unless /Start\sTable/;
2053 load_nameprep_range_table
($INPUT, $nameprep_flags{"unassigned"}, \
@flags_table) if (/A.1/);
2054 load_nameprep_range_table
($INPUT, $nameprep_flags{"prohibited"}, \
@flags_table) if (/C.1.2/);
2055 load_nameprep_range_table
($INPUT, $nameprep_flags{"prohibited"}, \
@flags_table) if (/C.2.2/);
2056 load_nameprep_range_table
($INPUT, $nameprep_flags{"prohibited"}, \
@flags_table) if (/C.3/);
2057 load_nameprep_range_table
($INPUT, $nameprep_flags{"prohibited"}, \
@flags_table) if (/C.4/);
2058 load_nameprep_range_table
($INPUT, $nameprep_flags{"prohibited"}, \
@flags_table) if (/C.5/);
2059 load_nameprep_range_table
($INPUT, $nameprep_flags{"prohibited"}, \
@flags_table) if (/C.6/);
2060 load_nameprep_range_table
($INPUT, $nameprep_flags{"prohibited"}, \
@flags_table) if (/C.7/);
2061 load_nameprep_range_table
($INPUT, $nameprep_flags{"prohibited"}, \
@flags_table) if (/C.8/);
2062 load_nameprep_range_table
($INPUT, $nameprep_flags{"prohibited"}, \
@flags_table) if (/C.9/);
2063 load_nameprep_range_table
($INPUT, $nameprep_flags{"bidi_ral"}, \
@flags_table) if (/D.1/);
2064 load_nameprep_range_table
($INPUT, $nameprep_flags{"bidi_l"}, \
@flags_table) if (/D.2/);
2066 load_nameprep_map_table
($INPUT, \
@mapping_table) if (/B.1/);
2067 load_nameprep_map_table
($INPUT, \
@mapping_table) if (/B.2/);
2071 open OUTPUT
,">$filename.new" or die "Cannot create $filename";
2072 print "Building $filename\n";
2073 print OUTPUT
"/* Nameprep algorithm related data */\n";
2074 print OUTPUT
"/* generated from $RFCS/$STRINGPREP */\n";
2075 print OUTPUT
"/* DO NOT EDIT!! */\n\n";
2076 print OUTPUT
"#include \"wine/unicode.h\"\n\n";
2078 dump_two_level_mapping
( "nameprep_char_type", 0, @flags_table );
2080 ######### mapping table
2081 # first determine all the 16-char subsets that contain something
2083 my $pos = 16*3; # for the null subset
2084 for (my $i = 0; $i < 65536; $i++)
2086 next unless defined $mapping_table[$i];
2087 $filled[$i >> 4] = $pos;
2093 # now count the 256-char subsets that contain something
2094 my @filled_idx = (256) x
256;
2096 for (my $i = 0; $i < 4096; $i++)
2098 next unless $filled[$i];
2099 $filled_idx[$i >> 4] = $pos;
2103 my $null_offset = $pos;
2106 # add the index offsets to the subsets positions
2107 for (my $i = 0; $i < 4096; $i++)
2109 next unless $filled[$i];
2110 $filled[$i] += $null_offset;
2113 # dump the main index
2114 printf OUTPUT
"const unsigned short DECLSPEC_HIDDEN nameprep_mapping[%d] =\n", $total;
2115 printf OUTPUT
"{\n /* index */\n";
2116 printf OUTPUT
"%s", DUMP_ARRAY
( "0x%04x", 0, @filled_idx );
2117 printf OUTPUT
",\n /* null sub-index */\n%s", DUMP_ARRAY
( "0x%04x", 0, ($null_offset) x
16 );
2119 # dump the second-level indexes
2120 for (my $i = 0; $i < 256; $i++)
2122 next unless ($filled_idx[$i] > 256);
2123 my @table = @filled[($i<<4)..($i<<4)+15];
2124 for (my $j = 0; $j < 16; $j++) { $table[$j] ||= $null_offset; }
2125 printf OUTPUT
",\n /* sub-index %02x */\n", $i;
2126 printf OUTPUT
"%s", DUMP_ARRAY
( "0x%04x", 0, @table );
2129 # dump the 16-char subsets
2130 printf OUTPUT
",\n /* null mapping */\n";
2131 printf OUTPUT
"%s", DUMP_ARRAY
( "0x%04x", 0, (0) x
48 );
2133 for (my $i = 0; $i < 4096; $i++)
2135 next unless $filled[$i];
2136 my @table = (0) x
48;
2137 for (my $j = 0; $j < 16; $j++)
2139 if (defined $mapping_table[($i<<4) + $j])
2141 $table[3 * $j] = ${$mapping_table[($i << 4) + $j]}[0];
2142 $table[3 * $j + 1] = ${$mapping_table[($i << 4) + $j]}[1];
2143 $table[3 * $j + 2] = ${$mapping_table[($i << 4) + $j]}[2];
2146 printf OUTPUT
",\n /* 0x%03x0 .. 0x%03xf */\n", $i, $i;
2147 printf OUTPUT
"%s", DUMP_ARRAY
( "0x%04x", 0, @table );
2150 printf OUTPUT
"\n};\n";
2153 save_file
($filename);
2156 ################################################################
2157 # dump the ctype tables
2158 sub DUMP_CTYPE_TABLES
($)
2160 my $filename = shift;
2161 open OUTPUT
,">$filename.new" or die "Cannot create $filename";
2162 printf "Building $filename\n";
2163 printf OUTPUT
"/* Unicode ctype tables */\n";
2164 printf OUTPUT
"/* Automatically generated; DO NOT EDIT!! */\n\n";
2165 printf OUTPUT
"#include \"wine/unicode.h\"\n\n";
2167 # add the direction in the high 4 bits of the category
2168 for (my $i = 0; $i < 65536; $i++)
2170 $category_table[$i] |= $direction_table[$i] << 12 if defined $direction_table[$i];
2173 dump_simple_mapping
( "wine_wctype_table", 0, @category_table );
2176 save_file
($filename);
2180 ################################################################
2181 # dump the char composition table
2182 sub dump_compose_table
($)
2184 my $filename = shift;
2186 open OUTPUT
,">$filename.new" or die "Cannot create $filename";
2187 print "Building $filename\n";
2188 print OUTPUT
"/* Unicode char composition */\n";
2189 print OUTPUT
"/* generated from $UNIDATA/UnicodeData.txt */\n";
2190 print OUTPUT
"/* DO NOT EDIT!! */\n\n";
2191 print OUTPUT
"#include \"wine/unicode.h\"\n\n";
2194 foreach my $i (@compose_table)
2197 push @
{$filled[$comp[1]]}, [ $comp[0], $comp[2] ];
2200 # count how many different second chars we have
2203 for (my $i = 0; $i < 65536; $i++)
2205 next unless defined $filled[$i];
2209 # build the table of second chars and offsets
2211 my $pos = $count + 1;
2213 for (my $i = 0; $i < 65536; $i++)
2215 next unless defined $filled[$i];
2216 push @table, $i, $pos;
2217 $pos += @
{$filled[$i]};
2219 # terminator with last position
2220 push @table, 0, $pos;
2221 printf OUTPUT
"static const WCHAR table[0x%x] =\n{\n", 2*$pos;
2222 printf OUTPUT
" /* second chars + offsets */\n%s", DUMP_ARRAY
( "0x%04x", 0, @table );
2224 # build the table of first chars and mappings
2226 for (my $i = 0; $i < 65536; $i++)
2228 next unless defined $filled[$i];
2230 my @list = sort { $a->[0] <=> $b->[0] } @
{$filled[$i]};
2231 for (my $j = 0; $j <= $#list; $j++)
2233 push @table, $list[$j][0], $list[$j][1];
2235 printf OUTPUT
",\n /* 0x%04x */\n%s", $i, DUMP_ARRAY
( "0x%04x", 0, @table );
2237 print OUTPUT
"\n};\n\n";
2238 print OUTPUT
<<"EOF";
2239 static inline int binary_search( WCHAR ch, int low, int high )
2243 int pos = (low + high) / 2;
2244 if (table[2 * pos] < ch) low = pos + 1;
2245 else if (table[2 * pos] > ch) high = pos - 1;
2251 WCHAR DECLSPEC_HIDDEN wine_compose( const WCHAR *str )
2253 int pos, idx = 1, start = 0, count = $count;
2256 if ((pos = binary_search( str[idx], start, count - 1 )) == -1) return 0;
2257 if (!idx--) return table[2 * pos + 1];
2258 start = table[2 * pos + 1];
2259 count = table[2 * pos + 3];
2264 save_file
($filename);
2267 ################################################################
2268 # dump the char decomposition table
2269 sub dump_decompose_table
($)
2271 my $filename = shift;
2273 open OUTPUT
,">$filename.new" or die "Cannot create $filename";
2274 print "Building $filename\n";
2275 print OUTPUT
"/* Unicode char composition */\n";
2276 print OUTPUT
"/* generated from $UNIDATA/UnicodeData.txt */\n";
2277 print OUTPUT
"/* DO NOT EDIT!! */\n\n";
2278 print OUTPUT
"#include \"wine/unicode.h\"\n\n";
2280 # first determine all the 16-char subsets that contain something
2282 my @filled = (0) x
4096;
2283 my $pos = 16*2; # for the null subset
2284 for (my $i = 0; $i < 65536; $i++)
2286 next unless defined $decomp_table[$i];
2287 $filled[$i >> 4] = $pos;
2293 # now count the 256-char subsets that contain something
2295 my @filled_idx = (256) x
256;
2297 for (my $i = 0; $i < 4096; $i++)
2299 next unless $filled[$i];
2300 $filled_idx[$i >> 4] = $pos;
2304 my $null_offset = $pos; # null mapping
2307 # add the index offsets to the subsets positions
2309 for (my $i = 0; $i < 4096; $i++)
2311 next unless $filled[$i];
2312 $filled[$i] += $null_offset;
2315 # dump the main index
2317 printf OUTPUT
"static const WCHAR table[%d] =\n", $total;
2318 printf OUTPUT
"{\n /* index */\n";
2319 printf OUTPUT
"%s", DUMP_ARRAY
( "0x%04x", 0, @filled_idx );
2320 printf OUTPUT
",\n /* null sub-index */\n%s", DUMP_ARRAY
( "0x%04x", 0, ($null_offset) x
16 );
2322 # dump the second-level indexes
2324 for (my $i = 0; $i < 256; $i++)
2326 next unless ($filled_idx[$i] > 256);
2327 my @table = @filled[($i<<4)..($i<<4)+15];
2328 for (my $j = 0; $j < 16; $j++) { $table[$j] ||= $null_offset; }
2329 printf OUTPUT
",\n /* sub-index %02x */\n", $i;
2330 printf OUTPUT
"%s", DUMP_ARRAY
( "0x%04x", 0, @table );
2333 # dump the 16-char subsets
2335 printf OUTPUT
",\n /* null mapping */\n";
2336 printf OUTPUT
"%s", DUMP_ARRAY
( "0x%04x", 0, (0) x
32 );
2338 for (my $i = 0; $i < 4096; $i++)
2340 next unless $filled[$i];
2341 my @table = (0) x
32;
2342 for (my $j = 0; $j < 16; $j++)
2344 if (defined $decomp_table[($i<<4) + $j])
2346 $table[2 * $j] = ${$decomp_table[($i << 4) + $j]}[0];
2347 $table[2 * $j + 1] = ${$decomp_table[($i << 4) + $j]}[1];
2350 printf OUTPUT
",\n /* 0x%03x0 .. 0x%03xf */\n", $i, $i;
2351 printf OUTPUT
"%s", DUMP_ARRAY
( "0x%04x", 0, @table );
2354 printf OUTPUT
"\n};\n\n";
2355 print OUTPUT
<<"EOF";
2356 unsigned int DECLSPEC_HIDDEN wine_decompose( WCHAR ch, WCHAR *dst, unsigned int dstlen )
2358 const WCHAR *ptr = table + table[table[ch >> 8] + ((ch >> 4) & 0x0f)] + 2 * (ch & 0xf);
2362 if (!*ptr) return 1;
2363 if (dstlen <= 1) return 0;
2364 /* apply the decomposition recursively to the first char */
2365 if ((res = wine_decompose( *ptr, dst, dstlen-1 ))) dst[res++] = ptr[1];
2370 save_file
($filename);
2373 ################################################################
2374 # output a codepage definition file from the global tables
2375 sub output_codepage_file
($$$$)
2377 my ($codepage, $filename, $comment, $has_glyphs) = @_;
2379 my $output = sprintf "libs/port/c_%03d.c", $codepage;
2380 open OUTPUT
,">$output.new" or die "Cannot create $output";
2382 printf "Building %s from %s (%s)\n", $output, $filename || "hardcoded data", $comment;
2386 printf OUTPUT
"/* code page %03d (%s) */\n", $codepage, $comment;
2389 print OUTPUT
"/* generated from $MAPPINGS/$filename */\n";
2390 print OUTPUT
"/* DO NOT EDIT!! */\n\n";
2394 printf OUTPUT
"/* Automatically generated; DO NOT EDIT!! */\n\n";
2396 printf OUTPUT
"#include \"wine/unicode.h\"\n\n";
2398 if (!@lead_bytes) { dump_sbcs_table
( $codepage, $has_glyphs, $comment, $default_char, $default_wchar ); }
2399 else { dump_dbcs_table
( $codepage, $comment, $default_char, $default_wchar, get_lb_ranges
() ); }
2404 ################################################################
2405 # read a "bestfit" Windows mapping file
2406 sub read_bestfit_file
($)
2408 my ($filename) = @_;
2410 my ($codepage, $width, $count);
2411 my ($lb_cur, $lb_end);
2413 my $INPUT = open_data_file
( $MAPPINGS, $filename ) or die "Cannot open $filename";
2417 next if /^;/; # skip comments
2418 next if /^\s*$/; # skip empty lines
2419 next if /\x1a/; # skip ^Z
2420 last if /^ENDCODEPAGE/;
2422 if (/^CODEPAGE\s+(\d+)/)
2427 if (/^CPINFO\s+(\d+)\s+0x([0-9a-fA-f]+)\s+0x([0-9a-fA-F]+)/)
2430 $default_char = hex $2;
2431 $default_wchar = hex $3;
2434 if (/^(MBTABLE|WCTABLE|DBCSRANGE|DBCSTABLE)\s+(\d+)/)
2440 if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)/)
2442 if ($state eq "MBTABLE")
2446 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
2449 if ($state eq "WCTABLE")
2453 $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
2456 if ($state eq "DBCSRANGE")
2460 for (my $i = $start; $i <= $end; $i++) { add_lead_byte
( $i ); }
2465 if ($state eq "DBCSTABLE")
2469 my $cp = ($lb_cur << 8) | $mb;
2470 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
2473 if (++$lb_cur > $lb_end) { $state = "DBCSRANGE"; }
2478 die "$filename: Unrecognized line $_\n";
2484 ################################################################
2485 # read an input file and generate the corresponding .c file
2488 my ($codepage,$filename,$has_glyphs,$comment,$first_private,$def,$defw) = @_;
2493 $default_char = $def || $DEF_CHAR;
2494 $default_wchar = $defw || $DEF_CHAR;
2496 # some codepage files are special
2497 if ($codepage == 20932)
2499 READ_JIS0208_FILE
( $filename );
2500 add_default_mappings
( $first_private );
2502 elsif ($codepage == 20127)
2504 fill_20127_codepage
();
2505 add_default_mappings
( $first_private );
2507 elsif ($filename =~ /\/bestfit
/)
2509 read_bestfit_file
( $filename );
2513 read_codepage_file
( $filename );
2514 if ($codepage == 10001)
2516 # add Shift-JIS mappings
2517 read_bestfit_file
( "VENDORS/MICSFT/WindowsBestFit/bestfit932.txt" );
2521 add_default_mappings
( $first_private );
2525 output_codepage_file
( $codepage, $filename, $comment, $has_glyphs );
2529 ################################################################
2530 # save a file if modified
2534 if (-f
$file && !system "cmp $file $file.new >/dev/null")
2540 rename "$file.new", "$file";
2545 ################################################################
2546 # output the list of codepage tables into the cptable.c file
2547 sub output_cptable
($)
2550 my @tables_decl = ();
2552 printf "Building %s\n", $output;
2554 foreach my $file (@allfiles)
2556 my ($codepage,$filename) = @
$file;
2557 push @tables_decl, sprintf("extern union cptable cptable_%03d;\n",$codepage);
2560 push @tables_decl, sprintf("\nstatic const union cptable * const cptables[%d] =\n{\n",$#allfiles+1);
2561 foreach my $file (@allfiles)
2563 my ($codepage,$filename) = @
$file;
2564 push @tables_decl, sprintf(" &cptable_%03d,\n", $codepage);
2566 push @tables_decl, "};";
2567 REPLACE_IN_FILE
( $output, @tables_decl );
2570 ################################################################
2571 # replace the contents of a file between ### cpmap ### marks
2573 sub REPLACE_IN_FILE
($@
)
2578 open(FILE
,$name) or die "Can't open $name";
2582 last if /\#\#\# cpmap begin \#\#\#/;
2587 if (/\#\#\# cpmap end \#\#\#/) { push @lines, "\n", $_; last; }
2589 push @lines, <FILE
>;
2590 open(FILE
,">$name.new") or die "Can't modify $name";
2596 ################################################################
2599 chdir ".." if -f
"./make_unicode";
2600 READ_DEFAULTS
( $DEFAULTS );
2601 DUMP_CASE_MAPPINGS
( "libs/wine/casemap.c" );
2602 DUMP_SORTKEYS
( "libs/wine/collation.c", READ_SORTKEYS_FILE
() );
2603 dump_compose_table
( "libs/port/compose.c" );
2604 dump_decompose_table
( "libs/port/decompose.c" );
2605 DUMP_CTYPE_TABLES
( "libs/wine/wctype.c" );
2606 dump_digit_folding
( "libs/port/digitmap.c" );
2607 dump_mirroring
( "dlls/usp10/mirror.c" );
2608 dump_mirroring
( "dlls/dwrite/mirror.c" );
2609 dump_bracket
( "dlls/usp10/bracket.c" );
2610 dump_bracket
( "dlls/dwrite/bracket.c" );
2611 dump_shaping
( "dlls/usp10/shaping.c" );
2612 dump_linebreak
( "dlls/usp10/linebreak.c" );
2613 dump_linebreak
( "dlls/dwrite/linebreak.c" );
2614 dump_scripts
( "dlls/dwrite/scripts" );
2615 dump_indic
( "dlls/usp10/indicsyllable.c" );
2616 dump_intl_nls
("loader/l_intl.nls");
2617 dump_vertical
( "dlls/gdi32/vertical.c" );
2618 dump_vertical
( "dlls/wineps.drv/vertical.c" );
2619 dump_nameprep
( "dlls/kernel32/nameprep.c" );
2621 foreach my $file (@allfiles) { HANDLE_FILE
( @
{$file} ); }
2623 output_cptable
("libs/port/cptable.c");
2628 # compile-command: "./make_unicode"