?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
shell32/tests: Fixed some leaks (Valgrind).
[wine.git] / tools / make_unicode
blob92b0b64a94c9593ae1829abd345242a3f4be51f0
1 #!/usr/bin/perl -w
2 #
3 # Generate code page .c files from ftp.unicode.org descriptions
4 #
5 # Copyright 2000 Alexandre Julliard
6 #
7 # This library is free software; you can redistribute it and/or
8 # modify it under the terms of the GNU Lesser General Public
9 # License as published by the Free Software Foundation; either
10 # version 2.1 of the License, or (at your option) any later version.
11 #
12 # This library is distributed in the hope that it will be useful,
13 # but WITHOUT ANY WARRANTY; without even the implied warranty of
14 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15 # Lesser General Public License for more details.
16 #
17 # You should have received a copy of the GNU Lesser General Public
18 # License along with this library; if not, write to the Free Software
19 # Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA
20 #
21
22 use strict;
23
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/10.0.0/ucd";
27 my $REPORTS = "http://www.unicode.org/reports";
28 my $RFCS = "http://www.rfc-editor.org/rfc";
29
30 # Sort keys file
31 my $SORTKEYS = "tr10/allkeys.txt";
32
33 # RFC3454 (stringprep data)
34 my $STRINGPREP = "rfc3454.txt";
35
36 # Defaults mapping
37 my $DEFAULTS = "tools/unicode-defaults";
38
39 # Default char for undefined mappings
40 my $DEF_CHAR = ord '?';
41
42 my @allfiles =
43 (
44 [ 37, "VENDORS/MICSFT/EBCDIC/CP037.TXT", 0, "IBM EBCDIC US Canada", 0, 0x6f ],
45 [ 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)" ]
116 );
117
118
119 my %ctype =
120 (
121 "upper" => 0x0001,
122 "lower" => 0x0002,
123 "digit" => 0x0004,
124 "space" => 0x0008,
125 "punct" => 0x0010,
126 "cntrl" => 0x0020,
127 "blank" => 0x0040,
128 "xdigit" => 0x0080,
129 "alpha" => 0x0100,
130 "defin" => 0x0200
131 );
132
133 my %bracket_types =
134 (
135 "o" => 0x0000,
136 "c" => 0x0001,
137 );
138
139 my %indic_types =
140 (
141 "Other" => 0x0000,
142 "Bindu" => 0x0001,
143 "Visarga" => 0x0002,
144 "Avagraha" => 0x0003,
145 "Nukta" => 0x0004,
146 "Virama" => 0x0005,
147 "Vowel_Independent" => 0x0006,
148 "Vowel_Dependent" => 0x0007,
149 "Vowel" => 0x0008,
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,
168 "Joiner" => 0x001b,
169 "Number_Joiner" => 0x001c,
170 "Number" => 0x001d,
171 "Brahmi_Joining_Number" => 0x001e,
172 "Consonant_With_Stacker" => 0x001f,
173 "Consonant_Prefixed" => 0x0020,
174 "Syllable_Modifier" => 0x0021,
175 "Consonant_Killer" => 0x0022
176 );
177
178 my %matra_types =
179 (
180 "Right" => 0x01,
181 "Left" => 0x02,
182 "Visual_Order_Left" => 0x03,
183 "Left_And_Right" => 0x04,
184 "Top" => 0x05,
185 "Bottom" => 0x06,
186 "Top_And_Bottom" => 0x07,
187 "Top_And_Right" => 0x08,
188 "Top_And_Left" => 0x09,
189 "Top_And_Left_And_Right" => 0x0a,
190 "Bottom_And_Right" => 0x0b,
191 "Top_And_Bottom_And_Right" => 0x0c,
192 "Overstruck" => 0x0d,
193 "Invisible" => 0x0e,
194 "Bottom_And_Left" => 0x0f
195 );
196
197 my %nameprep_flags =
198 (
199 "unassigned" => 0x01,
200 "prohibited" => 0x02,
201 "bidi_ral" => 0x04,
202 "bidi_l" => 0x08
203 );
204
205 my %break_types =
206 (
207 "BK" => 0x0001,
208 "CR" => 0x0002,
209 "LF" => 0x0003,
210 "CM" => 0x0004,
211 "SG" => 0x0005,
212 "GL" => 0x0006,
213 "CB" => 0x0007,
214 "SP" => 0x0008,
215 "ZW" => 0x0009,
216 "NL" => 0x000a,
217 "WJ" => 0x000b,
218 "JL" => 0x000c,
219 "JV" => 0x000d,
220 "JT" => 0x000e,
221 "H2" => 0x000f,
222 "H3" => 0x0010,
223 "XX" => 0x0011,
224 "OP" => 0x0012,
225 "CL" => 0x0013,
226 "CP" => 0x0014,
227 "QU" => 0x0015,
228 "NS" => 0x0016,
229 "EX" => 0x0017,
230 "SY" => 0x0018,
231 "IS" => 0x0019,
232 "PR" => 0x001a,
233 "PO" => 0x001b,
234 "NU" => 0x001c,
235 "AL" => 0x001d,
236 "ID" => 0x001e,
237 "IN" => 0x001f,
238 "HY" => 0x0020,
239 "BB" => 0x0021,
240 "BA" => 0x0022,
241 "SA" => 0x0023,
242 "AI" => 0x0024,
243 "B2" => 0x0025,
244 "HL" => 0x0026,
245 "CJ" => 0x0027,
246 "RI" => 0x0028,
247 "EB" => 0x0029,
248 "EM" => 0x002a,
249 "ZWJ" => 0x002b,
250 );
251
252 my %vertical_types =
253 (
254 "R" => 0x0000,
255 "U" => 0x0001,
256 "Tr" => 0x0002,
257 "Tu" => 0x0003,
258 );
259
260 my %categories =
261 (
262 "Lu" => $ctype{"defin"}|$ctype{"alpha"}|$ctype{"upper"}, # Letter, Uppercase
263 "Ll" => $ctype{"defin"}|$ctype{"alpha"}|$ctype{"lower"}, # Letter, Lowercase
264 "Lt" => $ctype{"defin"}|$ctype{"alpha"}|$ctype{"upper"}|$ctype{"lower"}, # Letter, Titlecase
265 "Mn" => $ctype{"defin"}, # Mark, Non-Spacing
266 "Mc" => $ctype{"defin"}, # Mark, Spacing Combining
267 "Me" => $ctype{"defin"}, # Mark, Enclosing
268 "Nd" => $ctype{"defin"}|$ctype{"digit"}, # Number, Decimal Digit
269 "Nl" => $ctype{"defin"}|$ctype{"alpha"}, # Number, Letter
270 "No" => $ctype{"defin"}, # Number, Other
271 "Zs" => $ctype{"defin"}|$ctype{"space"}, # Separator, Space
272 "Zl" => $ctype{"defin"}|$ctype{"space"}, # Separator, Line
273 "Zp" => $ctype{"defin"}|$ctype{"space"}, # Separator, Paragraph
274 "Cc" => $ctype{"defin"}|$ctype{"cntrl"}, # Other, Control
275 "Cf" => $ctype{"defin"}|$ctype{"cntrl"}, # Other, Format
276 "Cs" => $ctype{"defin"}, # Other, Surrogate
277 "Co" => $ctype{"defin"}, # Other, Private Use
278 "Cn" => $ctype{"defin"}, # Other, Not Assigned
279 "Lm" => $ctype{"defin"}|$ctype{"alpha"}, # Letter, Modifier
280 "Lo" => $ctype{"defin"}|$ctype{"alpha"}, # Letter, Other
281 "Pc" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Connector
282 "Pd" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Dash
283 "Ps" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Open
284 "Pe" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Close
285 "Pi" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Initial quote
286 "Pf" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Final quote
287 "Po" => $ctype{"defin"}|$ctype{"punct"}, # Punctuation, Other
288 "Sm" => $ctype{"defin"}, # Symbol, Math
289 "Sc" => $ctype{"defin"}, # Symbol, Currency
290 "Sk" => $ctype{"defin"}, # Symbol, Modifier
291 "So" => $ctype{"defin"} # Symbol, Other
292 );
293
294 # a few characters need additional categories that cannot be determined automatically
295 my %special_categories =
296 (
297 "xdigit" => [ ord('0')..ord('9'),ord('A')..ord('F'),ord('a')..ord('f'),
298 0xff10..0xff19, 0xff21..0xff26, 0xff41..0xff46 ],
299 "space" => [ 0x09..0x0d, 0x85 ],
300 "blank" => [ 0x09, 0x20, 0xa0, 0x3000, 0xfeff ],
301 "cntrl" => [ 0x070f, 0x200c, 0x200d,
302 0x200e, 0x200f, 0x202a, 0x202b, 0x202c, 0x202d, 0x202e,
303 0x206a, 0x206b, 0x206c, 0x206d, 0x206e, 0x206f, 0xfeff,
304 0xfff9, 0xfffa, 0xfffb ],
305 "punct" => [ 0x24, 0x2b, 0x3c..0x3e, 0x5e, 0x60, 0x7c, 0x7e, 0xa2..0xbe,
306 0xd7, 0xf7 ],
307 "digit" => [ 0xb2, 0xb3, 0xb9 ],
308 "lower" => [ 0x2071, 0x207f ]
309 );
310
311 my %directions =
312 (
313 "L" => 1, # Left-to-Right
314 "R" => 2, # Right-to-Left
315 "AL" => 12, # Right-to-Left Arabic
316 "EN" => 3, # European Number
317 "ES" => 4, # European Number Separator
318 "ET" => 5, # European Number Terminator
319 "AN" => 6, # Arabic Number
320 "CS" => 7, # Common Number Separator
321 "NSM" => 13, # Non-Spacing Mark
322 "BN" => 14, # Boundary Neutral
323 "B" => 8, # Paragraph Separator
324 "S" => 9, # Segment Separator
325 "WS" => 10, # Whitespace
326 "ON" => 11, # Other Neutrals
327 "LRE" => 15, # Left-to-Right Embedding
328 "LRO" => 15, # Left-to-Right Override
329 "RLE" => 15, # Right-to-Left Embedding
330 "RLO" => 15, # Right-to-Left Override
331 "PDF" => 15, # Pop Directional Format
332 "LRI" => 15, # Left-to-Right Isolate
333 "RLI" => 15, # Right-to-Left Isolate
334 "FSI" => 15, # First Strong Isolate
335 "PDI" => 15 # Pop Directional Isolate
336 );
337
338 my %joining_types =
339 (
340 "U" => 0, # Non_Joining
341 "T" => 1, # Transparent
342 "R" => 2, # Right_Joining
343 "L" => 3, # Left_Joining
344 "D" => 4, # Dual_Joining
345 "C" => 5, # Join_Causing
346 );
347
348 my @cp2uni = ();
349 my @lead_bytes = ();
350 my @uni2cp = ();
351 my @unicode_defaults = ();
352 my @unicode_aliases = ();
353 my @tolower_table = ();
354 my @toupper_table = ();
355 my @digitmap_table = ();
356 my @compatmap_table = ();
357 my @category_table = ();
358 my @joining_table = ();
359 my @direction_table = ();
360 my @decomp_table = ();
361 my @compose_table = ();
362 my $default_char;
363 my $default_wchar;
364
365 my %joining_forms =
366 (
367 "isolated" => [],
368 "final" => [],
369 "initial" => [],
370 "medial" => []
371 );
372
373 ################################################################
374 # fetch a unicode.org file and open it
375 sub open_data_file($$)
376 {
377 my ($base, $name) = @_;
378 (my $dir = "data/$name") =~ s/\/[^\/]+$//;
379 local *FILE;
380 unless (-f "data/$name")
381 {
382 print "Fetching $base/$name...\n";
383 system "mkdir", "-p", $dir;
384 !system "wget", "-q", "-O", "data/$name", "$base/$name" or die "cannot fetch $base/$name";
385 }
386 open FILE, "<data/$name" or die "cannot open data/$name";
387 return *FILE;
388 }
389
390 ################################################################
391 # read in the defaults file
392 sub READ_DEFAULTS($)
393 {
394 my $filename = shift;
395 my $start;
396
397 # first setup a few default mappings
398
399 open DEFAULTS, "$filename" or die "Cannot open $filename";
400 print "Loading $filename\n";
401 while (<DEFAULTS>)
402 {
403 next if /^\#/; # skip comments
404 next if /^$/; # skip empty lines
405 if (/^(([0-9a-fA-F]+)(,[0-9a-fA-F]+)*)\s+([0-9a-fA-F]+|'.'|none)\s+(\#.*)?/)
406 {
407 my @src = map hex, split /,/,$1;
408 my $dst = $4;
409 my $comment = $5;
410 if ($#src > 0) { push @unicode_aliases, \@src; }
411 next if ($dst eq "none");
412 $dst = ($dst =~ /\'.\'/) ? ord substr($dst,1,1) : hex $dst;
413 foreach my $src (@src)
414 {
415 die "Duplicate value" if defined($unicode_defaults[$src]);
416 $unicode_defaults[$src] = $dst;
417 }
418 next;
419 }
420 die "Unrecognized line $_\n";
421 }
422 close DEFAULTS;
423
424 # now build mappings from the decomposition field of the Unicode database
425
426 my $UNICODE_DATA = open_data_file( $UNIDATA, "UnicodeData.txt" );
427 while (<$UNICODE_DATA>)
428 {
429 # Decode the fields ...
430 my ($code, $name, $cat, $comb, $bidi,
431 $decomp, $dec, $dig, $num, $mirror,
432 $oldname, $comment, $upper, $lower, $title) = split /;/;
433 my $dst;
434 my $src = hex $code;
435
436 die "unknown category $cat" unless defined $categories{$cat};
437 die "unknown directionality $bidi" unless defined $directions{$bidi};
438
439 $category_table[$src] = $categories{$cat};
440 $direction_table[$src] = $directions{$bidi};
441 $joining_table[$src] = $joining_types{"T"} if $cat eq "Mn" || $cat eq "Me" || $cat eq "Cf";
442
443 if ($lower ne "")
444 {
445 $tolower_table[$src] = hex $lower;
446 }
447 if ($upper ne "")
448 {
449 $toupper_table[$src] = hex $upper;
450 }
451 if ($dec ne "")
452 {
453 $category_table[$src] |= $ctype{"digit"};
454 }
455 if ($dig ne "")
456 {
457 $digitmap_table[$src] = ord $dig;
458 }
459
460 # copy the category and direction for everything between First/Last pairs
461 if ($name =~ /, First>/) { $start = $src; }
462 if ($name =~ /, Last>/)
463 {
464 while ($start < $src)
465 {
466 $category_table[$start] = $category_table[$src];
467 $direction_table[$start] = $direction_table[$src];
468 $start++;
469 }
470 }
471
472 next if $decomp eq ""; # no decomposition, skip it
473
474 if ($decomp =~ /^<([a-zA-Z]+)>\s+([0-9a-fA-F]+)$/)
475 {
476 # decomposition of the form "<foo> 1234" -> use char if type is known
477 if (($src >= 0xf900 && $src < 0xfb00) || ($src >= 0xfe30 && $src < 0xfffd))
478 {
479 # Single char decomposition in the compatibility range
480 $compatmap_table[$src] = hex $2;
481 }
482 if ($1 eq "isolated" || $1 eq "final" || $1 eq "initial" || $1 eq "medial")
483 {
484 ${joining_forms{$1}}[hex $2] = $src;
485 next;
486 }
487 next unless ($1 eq "font" ||
488 $1 eq "noBreak" ||
489 $1 eq "circle" ||
490 $1 eq "super" ||
491 $1 eq "sub" ||
492 $1 eq "wide" ||
493 $1 eq "narrow" ||
494 $1 eq "compat" ||
495 $1 eq "small");
496 $dst = hex $2;
497 }
498 elsif ($decomp =~ /^<compat>\s+0020\s+([0-9a-fA-F]+)/)
499 {
500 # decomposition "<compat> 0020 1234" -> combining accent
501 $dst = hex $1;
502 }
503 elsif ($decomp =~ /^([0-9a-fA-F]+)/)
504 {
505 # decomposition contains only char values without prefix -> use first char
506 $dst = hex $1;
507 $category_table[$src] |= $category_table[$dst] if defined $category_table[$dst];
508 # store decomposition if it contains two chars
509 if ($decomp =~ /^([0-9a-fA-F]+)\s+([0-9a-fA-F]+)$/)
510 {
511 $decomp_table[$src] = [ hex $1, hex $2 ];
512 push @compose_table, [ hex $1, hex $2, $src ];
513 }
514 elsif ($decomp =~ /^(<[a-z]+>\s)*([0-9a-fA-F]+)$/ &&
515 (($src >= 0xf900 && $src < 0xfb00) || ($src >= 0xfe30 && $src < 0xfffd)))
516 {
517 # Single char decomposition in the compatibility range
518 $compatmap_table[$src] = hex $2;
519 }
520 }
521 else
522 {
523 next;
524 }
525
526 next if defined($unicode_defaults[$src]); # may have been set in the defaults file
527
528 # check for loops
529 for (my $i = $dst; ; $i = $unicode_defaults[$i])
530 {
531 die sprintf("loop detected for %04x -> %04x",$src,$dst) if $i == $src;
532 last unless defined($unicode_defaults[$i]);
533 }
534 $unicode_defaults[$src] = $dst;
535 }
536 close $UNICODE_DATA;
537
538 # patch the category of some special characters
539
540 foreach my $cat (keys %special_categories)
541 {
542 my $flag = $ctype{$cat};
543 foreach my $i (@{$special_categories{$cat}}) { $category_table[$i] |= $flag; }
544 }
545 }
546
547
548 ################################################################
549 # define a new lead byte
550 sub add_lead_byte($)
551 {
552 my $ch = shift;
553 push @lead_bytes, $ch;
554 $cp2uni[$ch] = 0;
555 }
556
557 ################################################################
558 # define a new char mapping
559 sub add_mapping($$)
560 {
561 my ($cp, $uni) = @_;
562 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
563 $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
564 if ($cp > 0xff && !defined($cp2uni[$cp >> 8]))
565 {
566 add_lead_byte( $cp >> 8 );
567 }
568 }
569
570 ################################################################
571 # parse a standard codepage file
572 sub read_codepage_file($)
573 {
574 my $name = shift;
575 my $INPUT = open_data_file( $MAPPINGS, $name );
576
577 while (<$INPUT>)
578 {
579 next if /^\#/; # skip comments
580 next if /^$/; # skip empty lines
581 next if /\x1a/; # skip ^Z
582 next if (/^0x([0-9a-fA-F]+)\s+\#UNDEFINED/); # undefined char
583
584 if (/^0x([0-9a-fA-F]+)\s+\#DBCS LEAD BYTE/)
585 {
586 add_lead_byte( hex $1 );
587 next;
588 }
589 # 0x12 0x3456
590 if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\s+(\#.*)?/)
591 {
592 add_mapping( hex $1, hex $2 );
593 next;
594 }
595 # 0x12 0x3456+0x7890
596 if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\+.*\s+(\#.*)?/)
597 {
598 add_mapping( hex $1, hex $2 );
599 next;
600 }
601 # 0x12 <LR>+0x3456
602 if (/^0x([0-9a-fA-F]+)\s+<LR>\+0x([0-9a-fA-F]+)\s+(\#.*)?/)
603 {
604 add_mapping( hex $1, hex $2 );
605 next;
606 }
607 # 0x12 <RL>+0x3456
608 if (/^0x([0-9a-fA-F]+)\s+<RL>\+0x([0-9a-fA-F]+)\s+(\#.*)?/)
609 {
610 my $uni = hex $2;
611 add_mapping( hex $1, $uni ) if $uni >= 0x80;
612 next;
613 }
614 die "$name: Unrecognized line $_\n";
615 }
616 close $INPUT;
617 }
618
619
620 ################################################################
621 # fill input data for the 20127 (us-ascii) codepage
622 sub fill_20127_codepage()
623 {
624 for (my $i = 0; $i < 128; $i++) { $cp2uni[$i] = $uni2cp[$i] = $i; }
625 for (my $i = 128; $i < 256; $i++) { $cp2uni[$i] = $i & 0x7f; }
626 }
627
628 ################################################################
629 # get a mapping including glyph chars for MB_USEGLYPHCHARS
630
631 sub get_glyphs_mapping(@)
632 {
633 $_[0x01] = 0x263a; # (WHITE SMILING FACE)
634 $_[0x02] = 0x263b; # (BLACK SMILING FACE)
635 $_[0x03] = 0x2665; # (BLACK HEART SUIT)
636 $_[0x04] = 0x2666; # (BLACK DIAMOND SUIT)
637 $_[0x05] = 0x2663; # (BLACK CLUB SUIT)
638 $_[0x06] = 0x2660; # (BLACK SPADE SUIT)
639 $_[0x07] = 0x2022; # (BULLET)
640 $_[0x08] = 0x25d8; # (INVERSE BULLET)
641 $_[0x09] = 0x25cb; # (WHITE CIRCLE)
642 $_[0x0a] = 0x25d9; # (INVERSE WHITE CIRCLE)
643 $_[0x0b] = 0x2642; # (MALE SIGN)
644 $_[0x0c] = 0x2640; # (FEMALE SIGN)
645 $_[0x0d] = 0x266a; # (EIGHTH NOTE)
646 $_[0x0e] = 0x266b; # (BEAMED EIGHTH NOTES)
647 $_[0x0f] = 0x263c; # (WHITE SUN WITH RAYS)
648 $_[0x10] = 0x25ba; # (BLACK RIGHT-POINTING POINTER)
649 $_[0x11] = 0x25c4; # (BLACK LEFT-POINTING POINTER)
650 $_[0x12] = 0x2195; # (UP DOWN ARROW)
651 $_[0x13] = 0x203c; # (DOUBLE EXCLAMATION MARK)
652 $_[0x14] = 0x00b6; # (PILCROW SIGN)
653 $_[0x15] = 0x00a7; # (SECTION SIGN)
654 $_[0x16] = 0x25ac; # (BLACK RECTANGLE)
655 $_[0x17] = 0x21a8; # (UP DOWN ARROW WITH BASE)
656 $_[0x18] = 0x2191; # (UPWARDS ARROW)
657 $_[0x19] = 0x2193; # (DOWNWARDS ARROW)
658 $_[0x1a] = 0x2192; # (RIGHTWARDS ARROW)
659 $_[0x1b] = 0x2190; # (LEFTWARDS ARROW)
660 $_[0x1c] = 0x221f; # (RIGHT ANGLE)
661 $_[0x1d] = 0x2194; # (LEFT RIGHT ARROW)
662 $_[0x1e] = 0x25b2; # (BLACK UP-POINTING TRIANGLE)
663 $_[0x1f] = 0x25bc; # (BLACK DOWN-POINTING TRIANGLE)
664 $_[0x7f] = 0x2302; # (HOUSE)
665 return @_;
666 }
667
668 ################################################################
669 # build EUC-JP table from the JIS 0208 file
670 # FIXME: for proper EUC-JP we should probably read JIS 0212 too
671 # but this would require 3-byte DBCS characters
672 sub READ_JIS0208_FILE($)
673 {
674 my $name = shift;
675
676 # ASCII chars
677 for (my $i = 0x00; $i <= 0x7f; $i++)
678 {
679 $cp2uni[$i] = $i;
680 $uni2cp[$i] = $i;
681 }
682
683 # JIS X 0201 right plane
684 for (my $i = 0xa1; $i <= 0xdf; $i++)
685 {
686 $cp2uni[0x8e00 + $i] = 0xfec0 + $i;
687 $uni2cp[0xfec0 + $i] = 0x8e00 + $i;
688 }
689
690 # lead bytes
691 foreach my $i (0x8e, 0x8f, 0xa1 .. 0xfe)
692 {
693 add_lead_byte($i);
694 }
695
696 # undefined chars
697 foreach my $i (0x80 .. 0x8d, 0x90 .. 0xa0, 0xff)
698 {
699 $cp2uni[$i] = $DEF_CHAR;
700 }
701
702 # Shift-JIS compatibility
703 $uni2cp[0x00a5] = 0x5c;
704 $uni2cp[0x203e] = 0x7e;
705
706 # Fix backslash conversion
707 $cp2uni[0xa1c0] = 0xff3c;
708 $uni2cp[0xff3c] = 0xa1c0;
709
710 my $INPUT = open_data_file( $MAPPINGS, $name );
711 while (<$INPUT>)
712 {
713 next if /^\#/; # skip comments
714 next if /^$/; # skip empty lines
715 next if /\x1a/; # skip ^Z
716 if (/^0x[0-9a-fA-F]+\s+0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\s+(\#.*)?/)
717 {
718 my $cp = 0x8080 + hex $1;
719 my $uni = hex $2;
720 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
721 $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
722 next;
723 }
724 die "$name: Unrecognized line $_\n";
725 }
726 close $INPUT;
727 }
728
729
730 ################################################################
731 # build the sort keys table
732 sub READ_SORTKEYS_FILE()
733 {
734 my @sortkeys = ();
735 for (my $i = 0; $i < 65536; $i++) { $sortkeys[$i] = [ -1, 0, 0, 0, 0 ] };
736
737 my $INPUT = open_data_file( $REPORTS, $SORTKEYS );
738 while (<$INPUT>)
739 {
740 next if /^\#/; # skip comments
741 next if /^$/; # skip empty lines
742 next if /\x1a/; # skip ^Z
743 next if /^\@version/; # skip @version header
744 if (/^([0-9a-fA-F]+)\s+;\s+\[([*.])([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]+)\]/)
745 {
746 my ($uni,$variable) = (hex $1, $2);
747 next if $uni > 65535;
748 $sortkeys[$uni] = [ $uni, hex $3, hex $4, hex $5, hex $6 ];
749 next;
750 }
751 if (/^([0-9a-fA-F]+\s+)+;\s+\[[*.]([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]+)\]/)
752 {
753 # multiple character sequence, ignored for now
754 next;
755 }
756 die "$SORTKEYS: Unrecognized line $_\n";
757 }
758 close $INPUT;
759
760 # compress the keys to 32 bit:
761 # key 1 to 16 bits, key 2 to 8 bits, key 3 to 4 bits, key 4 to 1 bit
762
763 @sortkeys = sort { ${$a}[1] <=> ${$b}[1] or
764 ${$a}[2] <=> ${$b}[2] or
765 ${$a}[3] <=> ${$b}[3] or
766 ${$a}[4] <=> ${$b}[4] or
767 $a cmp $b; } @sortkeys;
768
769 my ($n2, $n3) = (1, 1);
770 my @keys = (-1, -1, -1, -1, -1 );
771 my @flatkeys = ();
772
773 for (my $i = 0; $i < 65536; $i++)
774 {
775 my @current = @{$sortkeys[$i]};
776 next if $current[0] == -1;
777 if ($current[1] == $keys[1])
778 {
779 if ($current[2] == $keys[2])
780 {
781 if ($current[3] == $keys[3])
782 {
783 # nothing
784 }
785 else
786 {
787 $keys[3] = $current[3];
788 $n3++;
789 die if ($n3 >= 16);
790 }
791 }
792 else
793 {
794 $keys[2] = $current[2];
795 $keys[3] = $current[3];
796 $n2++;
797 $n3 = 1;
798 die if ($n2 >= 256);
799 }
800 }
801 else
802 {
803 $keys[1] = $current[1];
804 $keys[2] = $current[2];
805 $keys[3] = $current[3];
806 $n2 = 1;
807 $n3 = 1;
808 }
809
810 if ($current[2]) { $current[2] = $n2; }
811 if ($current[3]) { $current[3] = $n3; }
812 if ($current[4]) { $current[4] = 1; }
813
814 $flatkeys[$current[0]] = ($current[1] << 16) | ($current[2] << 8) | ($current[3] << 4) | $current[4];
815 }
816 return @flatkeys;
817 }
818
819
820 ################################################################
821 # build the sort keys table
822 sub DUMP_SORTKEYS($@)
823 {
824 my ($filename, @keys) = @_;
825
826 # count the number of 256-key ranges that contain something
827
828 my @offsets = ();
829 my $ranges = 2;
830 for (my $i = 0; $i < 256; $i++) { $offsets[$i] = 256; }
831 for (my $i = 0; $i < 65536; $i++)
832 {
833 next unless defined $keys[$i];
834 $offsets[$i >> 8] = $ranges * 256;
835 $ranges++;
836 $i |= 255;
837 }
838
839 # output the range offsets
840
841 open OUTPUT,">$filename.new" or die "Cannot create $filename";
842 printf "Building $filename\n";
843 printf OUTPUT "/* Unicode collation element table */\n";
844 printf OUTPUT "/* generated from %s */\n", "$REPORTS/$SORTKEYS";
845 printf OUTPUT "/* DO NOT EDIT!! */\n\n";
846
847 printf OUTPUT "const unsigned int collation_table[%d] =\n{\n", $ranges*256;
848 printf OUTPUT " /* index */\n";
849 printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%08x", 0, @offsets );
850
851 # output the default values
852
853 printf OUTPUT " /* defaults */\n";
854 printf OUTPUT "%s", DUMP_ARRAY( "0x%08x", 0, (0xffffffff) x 256 );
855
856 # output all the key ranges
857
858 for (my $i = 0; $i < 256; $i++)
859 {
860 next if $offsets[$i] == 256;
861 printf OUTPUT ",\n /* 0x%02x00 .. 0x%02xff */\n", $i, $i;
862 printf OUTPUT "%s", DUMP_ARRAY( "0x%08x", 0xffffffff, @keys[($i<<8) .. ($i<<8)+255] );
863 }
864 printf OUTPUT "\n};\n";
865 close OUTPUT;
866 save_file($filename);
867 }
868
869
870 ################################################################
871 # add default mappings once the file had been read
872 sub add_default_mappings($)
873 {
874 my $first_private = shift;
875
876 # Apply aliases
877
878 foreach my $alias (@unicode_aliases)
879 {
880 my $target = undef;
881 foreach my $src (@$alias)
882 {
883 if (defined($uni2cp[$src]))
884 {
885 $target = $uni2cp[$src];
886 last;
887 }
888 }
889 next unless defined($target);
890
891 # At least one char of the alias set is defined, set the others to the same value
892 foreach my $src (@$alias)
893 {
894 $uni2cp[$src] = $target unless defined($uni2cp[$src]);
895 }
896 }
897
898 # For every src -> target mapping in the defaults table,
899 # make uni2cp[src] = uni2cp[target] if uni2cp[target] is defined
900
901 for (my $src = 0; $src < 65536; $src++)
902 {
903 next if defined($uni2cp[$src]); # source has a definition already
904 next unless defined($unicode_defaults[$src]); # no default for this char
905 my $target = $unicode_defaults[$src];
906
907 # do a recursive mapping until we find a target char that is defined
908 while (!defined($uni2cp[$target]) &&
909 defined($unicode_defaults[$target])) { $target = $unicode_defaults[$target]; }
910
911 if (defined($uni2cp[$target])) { $uni2cp[$src] = $uni2cp[$target]; }
912 }
913
914 # Add a private char mapping for undefined chars
915
916 if ($first_private)
917 {
918 for (my $i = 32; $i < 256; $i++)
919 {
920 next if defined($cp2uni[$i]);
921 $cp2uni[$i] = $first_private;
922 $uni2cp[$first_private] = $i;
923 $first_private++;
924 }
925 }
926
927 # Add an identity mapping for all undefined chars
928
929 for (my $i = 0; $i < 256; $i++)
930 {
931 next if defined($cp2uni[$i]);
932 next if defined($uni2cp[$i]);
933 $cp2uni[$i] = $uni2cp[$i] = $i;
934 }
935 }
936
937 ################################################################
938 # dump an array of integers
939 sub DUMP_ARRAY($$@)
940 {
941 my ($format,$default,@array) = @_;
942 my $i;
943 my $ret = " ";
944 for ($i = 0; $i < $#array; $i++)
945 {
946 $ret .= sprintf($format, defined $array[$i] ? $array[$i] : $default);
947 $ret .= (($i % 8) != 7) ? ", " : ",\n ";
948 }
949 $ret .= sprintf($format, defined $array[$i] ? $array[$i] : $default);
950 return $ret;
951 }
952
953
954 ################################################################
955 # dump a unicode->ascii mapping table
956 sub dump_uni2cp_table($$)
957 {
958 my ($width, $def) = @_;
959
960 # count the number of unicode->ascii subtables that contain something
961
962 my @lowerbounds;
963 my @upperbounds;
964 my $index = 0;
965 my @filled = (-1) x 256;
966 for (my $i = 0; $i < 65536; $i++)
967 {
968 next unless defined $uni2cp[$i];
969 if ($filled[$i >> 8] == -1)
970 {
971 $lowerbounds[$index] = $i & 0xff;
972 $upperbounds[$index] = 0xff - $lowerbounds[$index];
973 $filled[$i >> 8] = $index * 256;
974 $index++;
975 }
976 else
977 {
978 $upperbounds[$index-1] = 0xff - ($i & 0xff);
979 }
980 }
981
982 # add defaults mapping at the end
983 $filled[256] = $index * 256;
984 $lowerbounds[$index] = 255;
985 $upperbounds[$index] = 0;
986 $index++;
987
988 # collapse blocks upwards if possible
989 my $removed = 0;
990 $index = 0;
991 for (my $i = 0; $i < 257; $i++)
992 {
993 next if $filled[$i] == -1;
994 if ($upperbounds[$index - 1] > $lowerbounds[$index])
995 {
996 $removed += $lowerbounds[$index];
997 }
998 else
999 {
1000 $removed += $upperbounds[$index - 1];
1001 $lowerbounds[$index] = $upperbounds[$index - 1];
1002 }
1003 $filled[$i] -= $removed;
1004 $index++;
1005 }
1006
1007 # output all the subtables into a single array
1008
1009 printf OUTPUT "static const unsigned %s uni2cp_low[%d] =\n{\n",
1010 $width == 8 ? "char" : "short", ($index + 1) * 256 - $removed;
1011
1012 my $format = $width == 8 ? "0x%02x" : "0x%04x";
1013 for (my $i = $index = 0; $i < 257; $i++)
1014 {
1015 next if $filled[$i] == -1;
1016 my $start = ($i << 8) + $lowerbounds[$index];
1017 my $end = ($i << 8) + 255;
1018 if ($i == 256)
1019 {
1020 print OUTPUT " /* defaults */\n";
1021 printf OUTPUT "%s\n};\n\n", DUMP_ARRAY( $format, 0, ($def) x ($end - $start + 1) );
1022 }
1023 else
1024 {
1025 printf OUTPUT " /* 0x%04x .. 0x%04x */\n", $start, $end;
1026 printf OUTPUT "%s,\n", DUMP_ARRAY( $format, $def, @uni2cp[$start .. $end] );
1027 }
1028 $index++;
1029 }
1030
1031 # output a table of the offsets of the subtables in the previous array
1032
1033 for (my $i = 0; $i < 256; $i++)
1034 {
1035 if ($filled[$i] == -1) { $filled[$i] = $filled[256]; }
1036 }
1037 printf OUTPUT "static const unsigned short uni2cp_high[256] =\n";
1038 printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%04x", 0, @filled[0..255] );
1039 }
1040
1041 ################################################################
1042 # dump an SBCS mapping table
1043 sub dump_sbcs_table($$$$$)
1044 {
1045 my ($codepage, $has_glyphs, $name, $def, $defw) = @_;
1046 my $i;
1047
1048 # output the ascii->unicode table
1049
1050 if ($has_glyphs)
1051 {
1052 printf OUTPUT "static const WCHAR cp2uni[512] =\n";
1053 printf OUTPUT "{\n%s", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[0 .. 255] );
1054 printf OUTPUT ",\n /* glyphs */\n%s\n};\n\n",
1055 DUMP_ARRAY( "0x%04x", $defw, get_glyphs_mapping(@cp2uni[0 .. 255]) );
1056 }
1057 else
1058 {
1059 printf OUTPUT "static const WCHAR cp2uni[256] =\n";
1060 printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[0 .. 255] );
1061 }
1062
1063 dump_uni2cp_table( 8, $def );
1064
1065 # output the code page descriptor
1066
1067 printf OUTPUT "const struct sbcs_table DECLSPEC_HIDDEN cptable_%03d =\n{\n", $codepage;
1068 printf OUTPUT " { %d, 1, 0x%04x, 0x%04x, \"%s\" },\n",
1069 $codepage, $def, $defw, $name;
1070 printf OUTPUT " cp2uni,\n";
1071 if ($has_glyphs) { printf OUTPUT " cp2uni + 256,\n"; }
1072 else { printf OUTPUT " cp2uni,\n"; }
1073 printf OUTPUT " uni2cp_low,\n";
1074 printf OUTPUT " uni2cp_high\n};\n";
1075 }
1076
1077
1078 ################################################################
1079 # dump a DBCS mapping table
1080 sub dump_dbcs_table($$$$@)
1081 {
1082 my ($codepage, $name, $def, $defw, @lb_ranges) = @_;
1083
1084 # build a list of lead bytes that are actually used
1085
1086 my @lblist = ();
1087 LBLOOP: for (my $y = 0; $y <= $#lead_bytes; $y++)
1088 {
1089 $cp2uni[$lead_bytes[$y]] = $defw;
1090 my $base = $lead_bytes[$y] << 8;
1091 for (my $x = 0; $x < 256; $x++)
1092 {
1093 if (defined $cp2uni[$base+$x])
1094 {
1095 push @lblist,$lead_bytes[$y];
1096 next LBLOOP;
1097 }
1098 }
1099 }
1100 my $unused = ($#lead_bytes > $#lblist);
1101
1102 # output the ascii->unicode table for the single byte chars
1103
1104 printf OUTPUT "static const WCHAR cp2uni[%d] =\n", 256 * ($#lblist + 2 + $unused);
1105 printf OUTPUT "{\n%s,\n", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[0 .. 255] );
1106
1107 # output the default table for unused lead bytes
1108
1109 if ($unused)
1110 {
1111 printf OUTPUT " /* unused lead bytes */\n";
1112 printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%04x", 0, ($defw) x 256 );
1113 }
1114
1115 # output the ascii->unicode table for each DBCS lead byte
1116
1117 for (my $y = 0; $y <= $#lblist; $y++)
1118 {
1119 my $base = $lblist[$y] << 8;
1120 printf OUTPUT " /* lead byte %02x */\n", $lblist[$y];
1121 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[$base .. $base+255] );
1122 printf OUTPUT ($y < $#lblist) ? ",\n" : "\n};\n\n";
1123 }
1124
1125 # output the lead byte subtables offsets
1126
1127 my @offsets = (0) x 256;
1128 for (my $x = 0; $x <= $#lblist; $x++) { $offsets[$lblist[$x]] = $x + 1; }
1129 if ($unused)
1130 {
1131 # increment all lead bytes offset to take into account the unused table
1132 for (my $x = 0; $x <= $#lead_bytes; $x++) { $offsets[$lead_bytes[$x]]++; }
1133 }
1134 printf OUTPUT "static const unsigned char cp2uni_leadbytes[256] =\n";
1135 printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%02x", 0, @offsets );
1136
1137 dump_uni2cp_table( 16, $def );
1138
1139 # output the code page descriptor
1140
1141 printf OUTPUT "const struct dbcs_table DECLSPEC_HIDDEN cptable_%03d =\n{\n", $codepage;
1142 printf OUTPUT " { %d, 2, 0x%04x, 0x%04x, \"%s\" },\n",
1143 $codepage, $def, $defw, $name;
1144 printf OUTPUT " cp2uni,\n";
1145 printf OUTPUT " cp2uni_leadbytes,\n";
1146 printf OUTPUT " uni2cp_low,\n";
1147 printf OUTPUT " uni2cp_high,\n";
1148 printf OUTPUT " {\n %s\n }\n", DUMP_ARRAY( "0x%02x", 0, @lb_ranges, 0, 0 );
1149 printf OUTPUT "};\n";
1150 }
1151
1152
1153 ################################################################
1154 # get the list of defined lead byte ranges
1155 sub get_lb_ranges()
1156 {
1157 my @list = ();
1158 my @ranges = ();
1159 my $i = 0;
1160 foreach $i (@lead_bytes) { $list[$i] = 1; }
1161 my $on = 0;
1162 for (my $i = 0; $i < 256; $i++)
1163 {
1164 if ($on)
1165 {
1166 if (!defined $list[$i]) { push @ranges, $i-1; $on = 0; }
1167 }
1168 else
1169 {
1170 if ($list[$i]) { push @ranges, $i; $on = 1; }
1171 }
1172 }
1173 if ($on) { push @ranges, 0xff; }
1174 return @ranges;
1175 }
1176
1177 ################################################################
1178 # dump the Indic Syllabic Category table
1179 sub dump_indic($)
1180 {
1181 my $filename = shift;
1182 my @indic_table;
1183
1184 my $INPUT = open_data_file( $UNIDATA, "IndicSyllabicCategory.txt" );
1185 while (<$INPUT>)
1186 {
1187 next if /^\#/; # skip comments
1188 next if /^\s*$/; # skip empty lines
1189 next if /\x1a/; # skip ^Z
1190 if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*#/)
1191 {
1192 my $type = $2;
1193 die "unknown indic $type" unless defined $indic_types{$type};
1194 if (hex $1 < 65536)
1195 {
1196 $indic_table[hex $1] = $indic_types{$type};
1197 }
1198 next;
1199 }
1200 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([A-Za-z_]+)\s*#/)
1201 {
1202 my $type = $3;
1203 die "unknown indic $type" unless defined $indic_types{$type};
1204 if (hex $1 < 65536 and hex $2 < 65536)
1205 {
1206 foreach my $i (hex $1 .. hex $2)
1207 {
1208 $indic_table[$i] = $indic_types{$type};
1209 }
1210 }
1211 next;
1212 }
1213 die "malformed line $_";
1214 }
1215 close $INPUT;
1216
1217 $INPUT = open_data_file( $UNIDATA, "IndicPositionalCategory.txt" );
1218 while (<$INPUT>)
1219 {
1220 next if /^\#/; # skip comments
1221 next if /^\s*$/; # skip empty lines
1222 next if /\x1a/; # skip ^Z
1223 if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z]+)\s*#/)
1224 {
1225 my $type = $2;
1226 die "unknown matra $type" unless defined $matra_types{$type};
1227 $indic_table[hex $1] += $matra_types{$type} << 8;
1228 next;
1229 }
1230 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([A-Za-z_]+)\s*#/)
1231 {
1232 my $type = $3;
1233 die "unknown matra $type" unless defined $matra_types{$type};
1234 foreach my $i (hex $1 .. hex $2)
1235 {
1236 $indic_table[$i] += $matra_types{$type} << 8;
1237 }
1238 next;
1239 }
1240 die "malformed line $_";
1241 }
1242 close $INPUT;
1243
1244 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1245 print "Building $filename\n";
1246 print OUTPUT "/* Unicode Indic Syllabic Category */\n";
1247 print OUTPUT "/* generated from $UNIDATA/IndicSyllabicCategory.txt */\n";
1248 print OUTPUT "/* and from $UNIDATA/IndicPositionalCategory.txt */\n";
1249 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1250 print OUTPUT "#include \"wine/unicode.h\"\n\n";
1251
1252 dump_two_level_mapping( "indic_syllabic_table", $indic_types{'Other'}, @indic_table );
1253
1254 close OUTPUT;
1255 save_file($filename);
1256 }
1257
1258 ################################################################
1259 # dump the Line Break Properties table
1260 sub dump_linebreak($)
1261 {
1262 my $filename = shift;
1263 my @break_table;
1264 my $next_group = 0;
1265
1266 my $INPUT = open_data_file( $UNIDATA, "LineBreak.txt" );
1267 while (<$INPUT>)
1268 {
1269 next if /^\#/; # skip comments
1270 next if /^\s*$/; # skip empty lines
1271 next if /\x1a/; # skip ^Z
1272 if (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z][0-9A-Z])+\s*/)
1273 {
1274 my $type = $2;
1275 die "unknown breaktype $type" unless defined $break_types{$type};
1276 $break_table[hex $1] = $break_types{$type};
1277 next;
1278 }
1279 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z][0-9A-Z])+\s*/)
1280 {
1281 my $type = $3;
1282 die "unknown breaktype $type" unless defined $break_types{$type};
1283 foreach my $i (hex $1 .. hex $2)
1284 {
1285 $break_table[$i] = $break_types{$type};
1286 }
1287 next;
1288 }
1289 elsif (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z])+\s*/)
1290 {
1291 my $type = $2;
1292 die "unknown breaktype $type" unless defined $break_types{$type};
1293 $break_table[hex $1] = $break_types{$type};
1294 next;
1295 }
1296 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z])+\s*/)
1297 {
1298 my $type = $3;
1299 die "unknown breaktype $type" unless defined $break_types{$type};
1300 foreach my $i (hex $1 .. hex $2)
1301 {
1302 $break_table[$i] = $break_types{$type};
1303 }
1304 next;
1305 }
1306 die "malformed line $_";
1307 }
1308 close $INPUT;
1309
1310 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1311 print "Building $filename\n";
1312 print OUTPUT "/* Unicode Line Break Properties */\n";
1313 print OUTPUT "/* generated from $UNIDATA/LineBreak.txt */\n";
1314 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1315 print OUTPUT "#include \"wine/unicode.h\"\n\n";
1316
1317 dump_two_level_mapping( "wine_linebreak_table", $break_types{'XX'}, @break_table );
1318
1319 close OUTPUT;
1320 save_file($filename);
1321 }
1322
1323 my %scripts =
1324 (
1325 "Unknown" => 0,
1326 "Common" => 1,
1327 "Inherited" => 2,
1328 "Arabic" => 3,
1329 "Armenian" => 4,
1330 "Avestan" => 5,
1331 "Balinese" => 6,
1332 "Bamum" => 7,
1333 "Batak" => 8,
1334 "Bengali" => 9,
1335 "Bopomofo" => 10,
1336 "Brahmi" => 11,
1337 "Braille" => 12,
1338 "Buginese" => 13,
1339 "Buhid" => 14,
1340 "Canadian_Aboriginal" => 15,
1341 "Carian" => 16,
1342 "Cham" => 17,
1343 "Cherokee" => 18,
1344 "Coptic" => 19,
1345 "Cuneiform" => 20,
1346 "Cypriot" => 21,
1347 "Cyrillic" => 22,
1348 "Deseret" => 23,
1349 "Devanagari" => 24,
1350 "Egyptian_Hieroglyphs" => 25,
1351 "Ethiopic" => 26,
1352 "Georgian" => 27,
1353 "Glagolitic" => 28,
1354 "Gothic" => 29,
1355 "Greek" => 30,
1356 "Gujarati" => 31,
1357 "Gurmukhi" => 32,
1358 "Han" => 33,
1359 "Hangul" => 34,
1360 "Hanunoo" => 35,
1361 "Hebrew" => 36,
1362 "Hiragana" => 37,
1363 "Imperial_Aramaic" => 38,
1364 "Inscriptional_Pahlavi" => 39,
1365 "Inscriptional_Parthian" => 40,
1366 "Javanese" => 41,
1367 "Kaithi" => 42,
1368 "Kannada" => 43,
1369 "Katakana" => 44,
1370 "Kayah_Li" => 45,
1371 "Kharoshthi" => 46,
1372 "Khmer" => 47,
1373 "Lao" => 48,
1374 "Latin" => 49,
1375 "Lepcha" => 50,
1376 "Limbu" => 51,
1377 "Linear_B" => 52,
1378 "Lisu" => 53,
1379 "Lycian" => 54,
1380 "Lydian" => 55,
1381 "Malayalam" => 56,
1382 "Mandaic" => 57,
1383 "Meetei_Mayek" => 58,
1384 "Mongolian" => 59,
1385 "Myanmar" => 60,
1386 "New_Tai_Lue" => 61,
1387 "Nko" => 62,
1388 "Ogham" => 63,
1389 "Ol_Chiki" => 64,
1390 "Old_Italic" => 65,
1391 "Old_Persian" => 66,
1392 "Old_South_Arabian" => 67,
1393 "Old_Turkic" => 68,
1394 "Oriya" => 69,
1395 "Osmanya" => 70,
1396 "Phags_Pa" => 71,
1397 "Phoenician" => 72,
1398 "Rejang" => 73,
1399 "Runic" => 74,
1400 "Samaritan" => 75,
1401 "Saurashtra" => 76,
1402 "Shavian" => 77,
1403 "Sinhala" => 78,
1404 "Sundanese" => 79,
1405 "Syloti_Nagri" => 80,
1406 "Syriac" => 81,
1407 "Tagalog" => 82,
1408 "Tagbanwa" => 83,
1409 "Tai_Le" => 84,
1410 "Tai_Tham" => 85,
1411 "Tai_Viet" => 86,
1412 "Tamil" => 87,
1413 "Telugu" => 88,
1414 "Thaana" => 89,
1415 "Thai" => 90,
1416 "Tibetan" => 91,
1417 "Tifinagh" => 92,
1418 "Ugaritic" => 93,
1419 "Vai" => 94,
1420 "Yi" => 95,
1421 # Win8/Win8.1
1422 "Chakma" => 96,
1423 "Meroitic_Cursive" => 97,
1424 "Meroitic_Hieroglyphs" => 98,
1425 "Miao" => 99,
1426 "Sharada" => 100,
1427 "Sora_Sompeng" => 101,
1428 "Takri" => 102,
1429 # Win10
1430 "Bassa_Vah" => 103,
1431 "Caucasian_Albanian" => 104,
1432 "Duployan" => 105,
1433 "Elbasan" => 106,
1434 "Grantha" => 107,
1435 "Khojki" => 108,
1436 "Khudawadi" => 109,
1437 "Linear_A" => 110,
1438 "Mahajani" => 111,
1439 "Manichaean" => 112,
1440 "Mende_Kikakui" => 113,
1441 "Modi" => 114,
1442 "Mro" => 115,
1443 "Nabataean" => 116,
1444 "Old_North_Arabian" => 117,
1445 "Old_Permic" => 118,
1446 "Pahawh_Hmong" => 119,
1447 "Palmyrene" => 120,
1448 "Pau_Cin_Hau" => 121,
1449 "Psalter_Pahlavi" => 122,
1450 "Siddham" => 123,
1451 "Tirhuta" => 124,
1452 "Warang_Citi" => 125,
1453 # Win10 RS1
1454 "Adlam" => 126,
1455 "Ahom" => 127,
1456 "Anatolian_Hieroglyphs" => 128,
1457 "Bhaiksuki" => 129,
1458 "Hatran" => 130,
1459 "Marchen" => 131,
1460 "Multani" => 132,
1461 "Newa" => 133,
1462 "Old_Hungarian" => 134,
1463 "Osage" => 135,
1464 "SignWriting" => 136,
1465 "Tangut" => 137
1466 );
1467
1468 ################################################################
1469 # dump Script IDs table
1470 sub dump_scripts($)
1471 {
1472 my $filename = shift;
1473 my $header = $filename;
1474 my @scripts_table;
1475 my $script_index;
1476 my $i;
1477
1478 my $INPUT = open_data_file( $UNIDATA, "Scripts.txt" );
1479 # Fill the table
1480 # Unknown script id is always 0, so undefined scripts are automatically treated as such
1481 while (<$INPUT>)
1482 {
1483 my $type = "";
1484
1485 next if /^\#/; # skip comments
1486 next if /^\s*$/; # skip empty lines
1487 next if /\x1a/; # skip ^Z
1488 if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
1489 {
1490 $type = $2;
1491 if (defined $scripts{$type})
1492 {
1493 $scripts_table[hex $1] = $scripts{$type};
1494 }
1495 next;
1496 }
1497 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
1498 {
1499 $type = $3;
1500 if (defined $scripts{$type})
1501 {
1502 foreach my $i (hex $1 .. hex $2)
1503 {
1504 $scripts_table[$i] = $scripts{$type};
1505 }
1506 }
1507 next;
1508 }
1509 }
1510
1511 close $INPUT;
1512
1513 $header = "$filename.h";
1514 open OUTPUT,">$header.new" or die "Cannot create $header";
1515 print "Building $header\n";
1516 print OUTPUT "/* Unicode Script IDs */\n";
1517 print OUTPUT "/* generated from $UNIDATA/Scripts.txt */\n";
1518 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1519
1520 print OUTPUT "enum unicode_script_id {\n";
1521 foreach my $script (sort { $scripts{$a} <=> $scripts{$b} } keys %scripts)
1522 {
1523 print OUTPUT " Script_$script = $scripts{$script},\n";
1524 }
1525 print OUTPUT " Script_LastId = ", (scalar keys %scripts) - 1, "\n";
1526 print OUTPUT "};\n";
1527
1528 close OUTPUT;
1529 save_file($header);
1530
1531 $filename = "$filename.c";
1532 open OUTPUT,">$filename.new" or die "Cannot create $header";
1533 print "Building $filename\n";
1534 print OUTPUT "/* Unicode Script IDs */\n";
1535 print OUTPUT "/* generated from $UNIDATA/Scripts.txt */\n";
1536 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1537 print OUTPUT "#include \"wine/unicode.h\"\n\n";
1538
1539 dump_two_level_mapping( "wine_scripts_table", 0, @scripts_table );
1540 close OUTPUT;
1541 save_file($filename);
1542 }
1543
1544 ################################################################
1545 # dump the BiDi mirroring table
1546 sub dump_mirroring($)
1547 {
1548 my $filename = shift;
1549 my @mirror_table = ();
1550
1551 my $INPUT = open_data_file( $UNIDATA, "BidiMirroring.txt" );
1552 while (<$INPUT>)
1553 {
1554 next if /^\#/; # skip comments
1555 next if /^$/; # skip empty lines
1556 next if /\x1a/; # skip ^Z
1557 if (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9a-fA-F]+)/)
1558 {
1559 $mirror_table[hex $1] = hex $2;
1560 next;
1561 }
1562 die "malformed line $_";
1563 }
1564 close $INPUT;
1565
1566 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1567 print "Building $filename\n";
1568 print OUTPUT "/* Unicode BiDi mirroring */\n";
1569 print OUTPUT "/* generated from $UNIDATA/BidiMirroring.txt */\n";
1570 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1571 print OUTPUT "#include \"wine/unicode.h\"\n\n";
1572 DUMP_CASE_TABLE( "DECLSPEC_HIDDEN wine_mirror_map", @mirror_table );
1573 close OUTPUT;
1574 save_file($filename);
1575 }
1576
1577 ################################################################
1578 # dump the Bidi Brackets
1579 sub dump_bracket($)
1580 {
1581 my $filename = shift;
1582 my @bracket_table;
1583
1584 my $INPUT = open_data_file( $UNIDATA, "BidiBrackets.txt" );
1585 while (<$INPUT>)
1586 {
1587 next if /^\#/; # skip comments
1588 next if /^\s*$/; # skip empty lines
1589 next if /\x1a/; # skip ^Z
1590 if (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9a-fA-F]+);\s*([con])/)
1591 {
1592 my $type = $3;
1593 die "unknown bracket $type" unless defined $bracket_types{$type};
1594 die "characters too distant $1 and $2" if abs(hex($2) - hex($1)) >= 128;
1595 $bracket_table[hex $1] = (hex($2) - hex($1)) % 255;
1596 $bracket_table[hex $1] += $bracket_types{$type} << 8;
1597 next;
1598 }
1599 die "malformed line $_";
1600 }
1601 close $INPUT;
1602
1603 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1604 print "Building $filename\n";
1605 print OUTPUT "/* Unicode Bidirectional Bracket table */\n";
1606 print OUTPUT "/* generated from $UNIDATA/BidiBrackets.txt */\n";
1607 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1608 print OUTPUT "#include \"wine/unicode.h\"\n\n";
1609
1610 dump_two_level_mapping( "bidi_bracket_table", 0, @bracket_table );
1611
1612 close OUTPUT;
1613 save_file($filename);
1614 }
1615
1616 ################################################################
1617 # dump the Arabic shaping table
1618 sub dump_shaping($)
1619 {
1620 my $filename = shift;
1621 my %groups;
1622 my $next_group = 0;
1623
1624 $groups{"No_Joining_Group"} = $next_group++;
1625
1626 my $INPUT = open_data_file( $UNIDATA, "ArabicShaping.txt" );
1627 while (<$INPUT>)
1628 {
1629 next if /^\#/; # skip comments
1630 next if /^\s*$/; # skip empty lines
1631 next if /\x1a/; # skip ^Z
1632 if (/^\s*([0-9a-fA-F]+)\s*;.*;\s*([RLDCUT])\s*;\s*(\w+)/)
1633 {
1634 my $type = $2;
1635 my $group = $3;
1636 $groups{$group} = $next_group++ unless defined $groups{$group};
1637 $joining_table[hex $1] = $joining_types{$type} | ($groups{$group} << 8);
1638 next;
1639 }
1640 die "malformed line $_";
1641 }
1642 close $INPUT;
1643
1644 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1645 print "Building $filename\n";
1646 print OUTPUT "/* Unicode Arabic shaping */\n";
1647 print OUTPUT "/* generated from $UNIDATA/ArabicShaping.txt */\n";
1648 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1649 print OUTPUT "#include \"wine/unicode.h\"\n\n";
1650
1651 dump_two_level_mapping( "wine_shaping_table", 0, @joining_table );
1652
1653 print OUTPUT "\nconst unsigned short DECLSPEC_HIDDEN wine_shaping_forms[256][4] =\n{\n";
1654 for (my $i = 0x600; $i <= 0x6ff; $i++)
1655 {
1656 printf OUTPUT " { 0x%04x, 0x%04x, 0x%04x, 0x%04x },\n",
1657 ${joining_forms{"isolated"}}[$i] || $i,
1658 ${joining_forms{"final"}}[$i] || $i,
1659 ${joining_forms{"initial"}}[$i] || $i,
1660 ${joining_forms{"medial"}}[$i] || $i;
1661 }
1662 print OUTPUT "};\n";
1663
1664 close OUTPUT;
1665 save_file($filename);
1666 }
1667
1668 ################################################################
1669 # dump the Vertical Orientation table
1670 sub dump_vertical($)
1671 {
1672 my $filename = shift;
1673 my @vertical_table;
1674
1675 my $INPUT = open_data_file( $UNIDATA, "VerticalOrientation.txt" );
1676 while (<$INPUT>)
1677 {
1678 next if /^\#/; # skip comments
1679 next if /^\s*$/; # skip empty lines
1680 next if /\x1a/; # skip ^Z
1681 if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
1682 {
1683 my $type = $2;
1684 die "unknown vertical $type" unless defined $vertical_types{$type};
1685 if (hex $1 < 65536)
1686 {
1687 $vertical_table[hex $1] = $vertical_types{$type};
1688 }
1689 next;
1690 }
1691 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([A-Za-z_]+)\s*/)
1692 {
1693 my $type = $3;
1694 die "unknown vertical $type" unless defined $vertical_types{$type};
1695 foreach my $i (hex $1 .. hex $2)
1696 {
1697 $vertical_table[$i] = $vertical_types{$type};
1698 }
1699 next;
1700 }
1701 die "malformed line $_";
1702 }
1703 close $INPUT;
1704
1705 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1706 print "Building $filename\n";
1707 print OUTPUT "/* Unicode Vertical Orientation */\n";
1708 print OUTPUT "/* generated from $UNIDATA/VerticalOrientation.txt */\n";
1709 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1710 print OUTPUT "#include \"wine/unicode.h\"\n\n";
1711
1712 dump_two_level_mapping( "vertical_orientation_table", $vertical_types{'R'}, @vertical_table );
1713
1714 close OUTPUT;
1715 save_file($filename);
1716 }
1717
1718 ################################################################
1719 # dump the digit folding tables
1720 sub dump_digit_folding($)
1721 {
1722 my $filename = shift;
1723 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1724 print "Building $filename\n";
1725 print OUTPUT "/* Unicode digit folding mappings */\n";
1726 print OUTPUT "/* generated from $UNIDATA/UnicodeData.txt */\n";
1727 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1728 print OUTPUT "#include \"wine/unicode.h\"\n\n";
1729
1730 DUMP_CASE_TABLE( "DECLSPEC_HIDDEN wine_digitmap", @digitmap_table );
1731 print OUTPUT "\n";
1732 DUMP_CASE_TABLE( "DECLSPEC_HIDDEN wine_compatmap", @compatmap_table );
1733 close OUTPUT;
1734 save_file($filename);
1735 }
1736
1737
1738 ################################################################
1739 # dump the case mapping tables
1740 sub DUMP_CASE_MAPPINGS($)
1741 {
1742 my $filename = shift;
1743 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1744 print "Building $filename\n";
1745 print OUTPUT "/* Unicode case mappings */\n";
1746 print OUTPUT "/* generated from $UNIDATA/UnicodeData.txt */\n";
1747 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1748 print OUTPUT "#include \"wine/unicode.h\"\n\n";
1749
1750 DUMP_CASE_TABLE( "wine_casemap_lower", @tolower_table );
1751 print OUTPUT "\n";
1752 DUMP_CASE_TABLE( "wine_casemap_upper", @toupper_table );
1753 close OUTPUT;
1754 save_file($filename);
1755 }
1756
1757
1758 ################################################################
1759 # dump a case mapping table
1760 sub DUMP_CASE_TABLE($@)
1761 {
1762 my ($name,@table) = @_;
1763
1764 # count the number of sub tables that contain something
1765 # also compute the low and upper populated bounds
1766
1767 my @lowerbounds = ( 0, 0 );
1768 my @upperbounds = ( 0, 255 );
1769 my $index = 0;
1770 my @filled = ();
1771 for (my $i = 0; $i < 65536; $i++)
1772 {
1773 next unless defined $table[$i];
1774 if (!defined $filled[$i >> 8])
1775 {
1776 $lowerbounds[$index] = $i & 0xff;
1777 $upperbounds[$index] = 0xff - $lowerbounds[$index];
1778 $filled[$i >> 8] = $index * 256 + 512;
1779 $index++;
1780 }
1781 else
1782 {
1783 $upperbounds[$index-1] = 0xff - ($i & 0xff);
1784 }
1785 $table[$i] = ($table[$i] - $i) & 0xffff;
1786 }
1787
1788 # Collapse blocks upwards if possible
1789 my $removed = 0;
1790 $index = 0;
1791 for (my $i = 0; $i < 256; $i++)
1792 {
1793 next unless defined $filled[$i];
1794 if ($upperbounds[$index - 1] > $lowerbounds[$index])
1795 {
1796 $removed = $removed + $lowerbounds[$index];
1797 }
1798 else
1799 {
1800 $removed = $removed + $upperbounds[$index - 1];
1801 $lowerbounds[$index] = $upperbounds[$index - 1];
1802 }
1803 $filled[$i] = $filled[$i] - $removed;
1804 $index++;
1805 }
1806
1807 # dump the table
1808
1809 printf OUTPUT "const WCHAR %s[%d] =\n", $name, $index * 256 + 512 - $removed;
1810 printf OUTPUT "{\n /* index */\n";
1811 printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%04x", 256, @filled );
1812 printf OUTPUT " /* defaults */\n";
1813 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, (0) x 256 );
1814 $index = 0;
1815 for (my $i = 0; $i < 256; $i++)
1816 {
1817 next unless $filled[$i];
1818 printf OUTPUT ",\n /* 0x%02x%02x .. 0x%02xff */\n", $i, $lowerbounds[$index], $i;
1819 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0,
1820 @table[($i<<8) + $lowerbounds[$index] .. ($i<<8)+255] );
1821 $index++;
1822 }
1823 printf OUTPUT "\n};\n";
1824 }
1825
1826 ################################################################
1827 # compress a mapping table by removing identical rows
1828 sub compress_array($$@)
1829 {
1830 my $rows = shift;
1831 my $def = shift;
1832 my @table = @_;
1833 my $len = @table / $rows;
1834 my @array = (0) x $rows;
1835 my %sequences;
1836
1837 # try to merge table rows
1838 for (my $row = 0; $row < $rows; $row++)
1839 {
1840 my @table_row = map { defined($_) ? $_ : $def; } @table[($row * $len)..($row * $len + $len - 1)];
1841 my $rowtxt = pack "S*", @table_row;
1842 if (defined($sequences{$rowtxt}))
1843 {
1844 # reuse an existing row
1845 $array[$row] = $sequences{$rowtxt};
1846 }
1847 else
1848 {
1849 # create a new row
1850 $sequences{$rowtxt} = $array[$row] = $#array + 1;
1851 push @array, @table_row;
1852 }
1853 }
1854 return @array;
1855 }
1856
1857 ################################################################
1858 # dump a simple char -> 16-bit value mapping table
1859 sub dump_simple_mapping($$@)
1860 {
1861 my $name = shift;
1862 my $def = shift;
1863 my @array = compress_array( 256, $def, @_[0..65535] );
1864
1865 printf OUTPUT "const unsigned short %s[%d] =\n{\n", $name, $#array+1;
1866 printf OUTPUT " /* offsets */\n%s,\n", DUMP_ARRAY( "0x%04x", 0, @array[0..255] );
1867 printf OUTPUT " /* values */\n%s\n};\n", DUMP_ARRAY( "0x%04x", 0, @array[256..$#array] );
1868 }
1869
1870 ################################################################
1871 # dump a char -> 16-bit value mapping table using two-level tables
1872 sub dump_two_level_mapping($$@)
1873 {
1874 my $name = shift;
1875 my $def = shift;
1876 my @row_array = compress_array( 4096, $def, @_[0..65535] );
1877 my @array = compress_array( 256, 0, @row_array[0..4095] );
1878
1879 for (my $i = 256; $i < @array; $i++) { $array[$i] += @array - 4096; }
1880
1881 printf OUTPUT "const unsigned short DECLSPEC_HIDDEN %s[%d] =\n{\n", $name, @array + @row_array - 4096;
1882 printf OUTPUT " /* level 1 offsets */\n%s,\n", DUMP_ARRAY( "0x%04x", 0, @array[0..255] );
1883 printf OUTPUT " /* level 2 offsets */\n%s,\n", DUMP_ARRAY( "0x%04x", 0, @array[256..$#array] );
1884 printf OUTPUT " /* values */\n%s\n};\n", DUMP_ARRAY( "0x%04x", 0, @row_array[4096..$#row_array] );
1885 }
1886
1887 ################################################################
1888 # dump a binary case mapping table in l_intl.nls format
1889 sub dump_binary_case_table(@)
1890 {
1891 my (@table) = @_;
1892
1893 my %difftables_hash = ();
1894 my @difftables;
1895 my %offtables2_hash = ();
1896 my @offtables2 = ();
1897
1898 my @offtable = ();
1899 for (my $i = 0; $i < 256; $i++)
1900 {
1901 my @offtable2 = ();
1902 for(my $j = 0; $j < 16; $j++) # offset table for xx00-xxFF characters
1903 {
1904 my @difftable;
1905 for (my $k = 0; $k < 16; $k++) # case map table for xxx0-xxxF characters
1906 {
1907 my $char = ($i<<8) + ($j<<4) + $k;
1908 $difftable[$k] = (defined $table[$char]) ? (($table[$char]-$char) & 0xffff) : 0;
1909 }
1910
1911 my $diff_key = pack "S*", @difftable;
1912 my $offset3 = $difftables_hash{$diff_key};
1913 if (!defined $offset3)
1914 {
1915 $offset3 = scalar @difftables;
1916 $difftables_hash{$diff_key} = $offset3;
1917 push @difftables, @difftable;
1918 }
1919 $offtable2[$j] = $offset3;
1920 }
1921
1922 my $offtable2_key = pack "S*", @offtable2;
1923 my $offset2 = $offtables2_hash{$offtable2_key};
1924 if (!defined $offset2)
1925 {
1926 $offset2 = scalar @offtables2;
1927 $offtables2_hash{$offtable2_key} = $offset2;
1928 push @offtables2, \@offtable2;
1929 }
1930 $offtable[$i] = $offset2;
1931 }
1932
1933 my @output;
1934 my $offset = 0x100; # offset of first subtable in words
1935 foreach (@offtable)
1936 {
1937 push @output, 0x10 * $_ + $offset; # offset of subtable in words
1938 }
1939
1940 $offset = 0x100 + 0x10 * scalar @offtables2; # offset of first difftable in words
1941 foreach(@offtables2)
1942 {
1943 my $table = $_;
1944 foreach(@$table)
1945 {
1946 push @output, $_ + $offset; # offset of difftable in words
1947 }
1948 }
1949
1950 my $len = 1 + scalar @output + scalar @difftables;
1951 return pack "S<*", $len, @output, @difftables;
1952 }
1953
1954
1955 ################################################################
1956 # dump case mappings for l_intl.nls
1957 sub dump_intl_nls($)
1958 {
1959 my $filename = shift;
1960 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1961 printf "Building $filename\n";
1962
1963 binmode OUTPUT;
1964 print OUTPUT pack "S<", 1; # version
1965 print OUTPUT dump_binary_case_table( @toupper_table );
1966 print OUTPUT dump_binary_case_table( @tolower_table );
1967 close OUTPUT;
1968 save_file($filename);
1969 }
1970
1971
1972 sub load_nameprep_range_table($$$)
1973 {
1974 my ($INPUT, $val, $table_ref) = @_;
1975
1976 while (<$INPUT>)
1977 {
1978 if (/^\s*([0-9a-fA-F]+)-([0-9a-fA-F]+)/)
1979 {
1980 my $last = hex $2;
1981 $last = 65535 if($last >= 65536);
1982 foreach my $i (hex $1 .. $last)
1983 {
1984 $table_ref->[$i] |= $val;
1985 }
1986 next;
1987 }
1988 elsif (/^\s*([0-9a-fA-F]+)/)
1989 {
1990 if (hex $1 < 65536)
1991 {
1992 $table_ref->[hex $1] |= $val;
1993 }
1994 next;
1995 }
1996
1997 return if (/End\sTable/);
1998 }
1999 }
2000
2001 sub load_nameprep_map_table($$)
2002 {
2003 my ($INPUT, $table_ref) = @_;
2004
2005 while (<$INPUT>)
2006 {
2007 if (/^\s*([0-9a-fA-F]+);\s;/)
2008 {
2009 # special value for map to nothing
2010 $table_ref->[hex $1] = [0xffff, 0xffff, 0xffff];
2011 next;
2012 }
2013 elsif (/^\s*([0-9a-fA-F]+);\s([0-9a-fA-F]+);/)
2014 {
2015 $table_ref->[hex $1] = [hex $2, 0, 0];
2016 next;
2017 }
2018 elsif (/^\s*([0-9a-fA-F]+);\s([0-9a-fA-F]+)\s([0-9a-fA-F]+);/)
2019 {
2020 $table_ref->[hex $1] = [hex $2, hex $3, 0];
2021 next;
2022 }
2023 elsif (/^\s*([0-9a-fA-F]+);\s([0-9a-fA-F]+)\s([0-9a-fA-F]+)\s([0-9a-fA-F]+);/)
2024 {
2025 $table_ref->[hex $1] = [hex $2, hex $3, hex $4];
2026 next;
2027 }
2028
2029 return if (/End\sTable/);
2030 }
2031 }
2032
2033 ################################################################
2034 # dump mapping table, prohibited characters set, unassigned
2035 # characters, bidirectional rules used by nameprep algorithm
2036 sub dump_nameprep($)
2037 {
2038 my $filename = shift;
2039 my @mapping_table = ();
2040 my @flags_table;
2041
2042 my $INPUT = open_data_file( $RFCS, $STRINGPREP );
2043 while (<$INPUT>)
2044 {
2045 next unless /Start\sTable/;
2046
2047 load_nameprep_range_table($INPUT, $nameprep_flags{"unassigned"}, \@flags_table) if (/A.1/);
2048 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.1.2/);
2049 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.2.2/);
2050 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.3/);
2051 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.4/);
2052 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.5/);
2053 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.6/);
2054 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.7/);
2055 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.8/);
2056 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.9/);
2057 load_nameprep_range_table($INPUT, $nameprep_flags{"bidi_ral"}, \@flags_table) if (/D.1/);
2058 load_nameprep_range_table($INPUT, $nameprep_flags{"bidi_l"}, \@flags_table) if (/D.2/);
2059
2060 load_nameprep_map_table($INPUT, \@mapping_table) if (/B.1/);
2061 load_nameprep_map_table($INPUT, \@mapping_table) if (/B.2/);
2062 }
2063 close $INPUT;
2064
2065 open OUTPUT,">$filename.new" or die "Cannot create $filename";
2066 print "Building $filename\n";
2067 print OUTPUT "/* Nameprep algorithm related data */\n";
2068 print OUTPUT "/* generated from $RFCS/$STRINGPREP */\n";
2069 print OUTPUT "/* DO NOT EDIT!! */\n\n";
2070 print OUTPUT "#include \"wine/unicode.h\"\n\n";
2071
2072 dump_two_level_mapping( "nameprep_char_type", 0, @flags_table );
2073
2074 ######### mapping table
2075 # first determine all the 16-char subsets that contain something
2076 my @filled = ();
2077 my $pos = 16*3; # for the null subset
2078 for (my $i = 0; $i < 65536; $i++)
2079 {
2080 next unless defined $mapping_table[$i];
2081 $filled[$i >> 4] = $pos;
2082 $pos += 16*3;
2083 $i |= 15;
2084 }
2085 my $total = $pos;
2086
2087 # now count the 256-char subsets that contain something
2088 my @filled_idx = (256) x 256;
2089 $pos = 256 + 16;
2090 for (my $i = 0; $i < 4096; $i++)
2091 {
2092 next unless $filled[$i];
2093 $filled_idx[$i >> 4] = $pos;
2094 $pos += 16;
2095 $i |= 15;
2096 }
2097 my $null_offset = $pos;
2098 $total += $pos;
2099
2100 # add the index offsets to the subsets positions
2101 for (my $i = 0; $i < 4096; $i++)
2102 {
2103 next unless $filled[$i];
2104 $filled[$i] += $null_offset;
2105 }
2106
2107 # dump the main index
2108 printf OUTPUT "const unsigned short DECLSPEC_HIDDEN nameprep_mapping[%d] =\n", $total;
2109 printf OUTPUT "{\n /* index */\n";
2110 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @filled_idx );
2111 printf OUTPUT ",\n /* null sub-index */\n%s", DUMP_ARRAY( "0x%04x", 0, ($null_offset) x 16 );
2112
2113 # dump the second-level indexes
2114 for (my $i = 0; $i < 256; $i++)
2115 {
2116 next unless ($filled_idx[$i] > 256);
2117 my @table = @filled[($i<<4)..($i<<4)+15];
2118 for (my $j = 0; $j < 16; $j++) { $table[$j] ||= $null_offset; }
2119 printf OUTPUT ",\n /* sub-index %02x */\n", $i;
2120 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
2121 }
2122
2123 # dump the 16-char subsets
2124 printf OUTPUT ",\n /* null mapping */\n";
2125 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, (0) x 48 );
2126
2127 for (my $i = 0; $i < 4096; $i++)
2128 {
2129 next unless $filled[$i];
2130 my @table = (0) x 48;
2131 for (my $j = 0; $j < 16; $j++)
2132 {
2133 if (defined $mapping_table[($i<<4) + $j])
2134 {
2135 $table[3 * $j] = ${$mapping_table[($i << 4) + $j]}[0];
2136 $table[3 * $j + 1] = ${$mapping_table[($i << 4) + $j]}[1];
2137 $table[3 * $j + 2] = ${$mapping_table[($i << 4) + $j]}[2];
2138 }
2139 }
2140 printf OUTPUT ",\n /* 0x%03x0 .. 0x%03xf */\n", $i, $i;
2141 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
2142 }
2143
2144 printf OUTPUT "\n};\n";
2145
2146 close OUTPUT;
2147 save_file($filename);
2148 }
2149
2150 ################################################################
2151 # dump the ctype tables
2152 sub DUMP_CTYPE_TABLES($)
2153 {
2154 my $filename = shift;
2155 open OUTPUT,">$filename.new" or die "Cannot create $filename";
2156 printf "Building $filename\n";
2157 printf OUTPUT "/* Unicode ctype tables */\n";
2158 printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
2159 printf OUTPUT "#include \"wine/unicode.h\"\n\n";
2160
2161 # add the direction in the high 4 bits of the category
2162 for (my $i = 0; $i < 65536; $i++)
2163 {
2164 $category_table[$i] |= $direction_table[$i] << 12 if defined $direction_table[$i];
2165 }
2166
2167 dump_simple_mapping( "wine_wctype_table", 0, @category_table );
2168
2169 close OUTPUT;
2170 save_file($filename);
2171 }
2172
2173
2174 ################################################################
2175 # dump the char composition table
2176 sub dump_compose_table($)
2177 {
2178 my $filename = shift;
2179
2180 open OUTPUT,">$filename.new" or die "Cannot create $filename";
2181 print "Building $filename\n";
2182 print OUTPUT "/* Unicode char composition */\n";
2183 print OUTPUT "/* generated from $UNIDATA/UnicodeData.txt */\n";
2184 print OUTPUT "/* DO NOT EDIT!! */\n\n";
2185 print OUTPUT "#include \"wine/unicode.h\"\n\n";
2186
2187 my @filled = ();
2188 foreach my $i (@compose_table)
2189 {
2190 my @comp = @$i;
2191 push @{$filled[$comp[1]]}, [ $comp[0], $comp[2] ];
2192 }
2193
2194 # count how many different second chars we have
2195
2196 my $count = 0;
2197 for (my $i = 0; $i < 65536; $i++)
2198 {
2199 next unless defined $filled[$i];
2200 $count++;
2201 }
2202
2203 # build the table of second chars and offsets
2204
2205 my $pos = $count + 1;
2206 my @table = ();
2207 for (my $i = 0; $i < 65536; $i++)
2208 {
2209 next unless defined $filled[$i];
2210 push @table, $i, $pos;
2211 $pos += @{$filled[$i]};
2212 }
2213 # terminator with last position
2214 push @table, 0, $pos;
2215 printf OUTPUT "static const WCHAR table[0x%x] =\n{\n", 2*$pos;
2216 printf OUTPUT " /* second chars + offsets */\n%s", DUMP_ARRAY( "0x%04x", 0, @table );
2217
2218 # build the table of first chars and mappings
2219
2220 for (my $i = 0; $i < 65536; $i++)
2221 {
2222 next unless defined $filled[$i];
2223 my @table = ();
2224 my @list = sort { $a->[0] <=> $b->[0] } @{$filled[$i]};
2225 for (my $j = 0; $j <= $#list; $j++)
2226 {
2227 push @table, $list[$j][0], $list[$j][1];
2228 }
2229 printf OUTPUT ",\n /* 0x%04x */\n%s", $i, DUMP_ARRAY( "0x%04x", 0, @table );
2230 }
2231 print OUTPUT "\n};\n\n";
2232 print OUTPUT <<"EOF";
2233 static inline int binary_search( WCHAR ch, int low, int high )
2234 {
2235 while (low <= high)
2236 {
2237 int pos = (low + high) / 2;
2238 if (table[2 * pos] < ch) low = pos + 1;
2239 else if (table[2 * pos] > ch) high = pos - 1;
2240 else return pos;
2241 }
2242 return -1;
2243 }
2244
2245 WCHAR DECLSPEC_HIDDEN wine_compose( const WCHAR *str )
2246 {
2247 int pos, idx = 1, start = 0, count = $count;
2248 for (;;)
2249 {
2250 if ((pos = binary_search( str[idx], start, count - 1 )) == -1) return 0;
2251 if (!idx--) return table[2 * pos + 1];
2252 start = table[2 * pos + 1];
2253 count = table[2 * pos + 3];
2254 }
2255 }
2256 EOF
2257 close OUTPUT;
2258 save_file($filename);
2259 }
2260
2261 ################################################################
2262 # dump the char decomposition table
2263 sub dump_decompose_table($)
2264 {
2265 my $filename = shift;
2266
2267 open OUTPUT,">$filename.new" or die "Cannot create $filename";
2268 print "Building $filename\n";
2269 print OUTPUT "/* Unicode char composition */\n";
2270 print OUTPUT "/* generated from $UNIDATA/UnicodeData.txt */\n";
2271 print OUTPUT "/* DO NOT EDIT!! */\n\n";
2272 print OUTPUT "#include \"wine/unicode.h\"\n\n";
2273
2274 # first determine all the 16-char subsets that contain something
2275
2276 my @filled = (0) x 4096;
2277 my $pos = 16*2; # for the null subset
2278 for (my $i = 0; $i < 65536; $i++)
2279 {
2280 next unless defined $decomp_table[$i];
2281 $filled[$i >> 4] = $pos;
2282 $pos += 16*2;
2283 $i |= 15;
2284 }
2285 my $total = $pos;
2286
2287 # now count the 256-char subsets that contain something
2288
2289 my @filled_idx = (256) x 256;
2290 $pos = 256 + 16;
2291 for (my $i = 0; $i < 4096; $i++)
2292 {
2293 next unless $filled[$i];
2294 $filled_idx[$i >> 4] = $pos;
2295 $pos += 16;
2296 $i |= 15;
2297 }
2298 my $null_offset = $pos; # null mapping
2299 $total += $pos;
2300
2301 # add the index offsets to the subsets positions
2302
2303 for (my $i = 0; $i < 4096; $i++)
2304 {
2305 next unless $filled[$i];
2306 $filled[$i] += $null_offset;
2307 }
2308
2309 # dump the main index
2310
2311 printf OUTPUT "static const WCHAR table[%d] =\n", $total;
2312 printf OUTPUT "{\n /* index */\n";
2313 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @filled_idx );
2314 printf OUTPUT ",\n /* null sub-index */\n%s", DUMP_ARRAY( "0x%04x", 0, ($null_offset) x 16 );
2315
2316 # dump the second-level indexes
2317
2318 for (my $i = 0; $i < 256; $i++)
2319 {
2320 next unless ($filled_idx[$i] > 256);
2321 my @table = @filled[($i<<4)..($i<<4)+15];
2322 for (my $j = 0; $j < 16; $j++) { $table[$j] ||= $null_offset; }
2323 printf OUTPUT ",\n /* sub-index %02x */\n", $i;
2324 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
2325 }
2326
2327 # dump the 16-char subsets
2328
2329 printf OUTPUT ",\n /* null mapping */\n";
2330 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, (0) x 32 );
2331
2332 for (my $i = 0; $i < 4096; $i++)
2333 {
2334 next unless $filled[$i];
2335 my @table = (0) x 32;
2336 for (my $j = 0; $j < 16; $j++)
2337 {
2338 if (defined $decomp_table[($i<<4) + $j])
2339 {
2340 $table[2 * $j] = ${$decomp_table[($i << 4) + $j]}[0];
2341 $table[2 * $j + 1] = ${$decomp_table[($i << 4) + $j]}[1];
2342 }
2343 }
2344 printf OUTPUT ",\n /* 0x%03x0 .. 0x%03xf */\n", $i, $i;
2345 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
2346 }
2347
2348 printf OUTPUT "\n};\n\n";
2349 print OUTPUT <<"EOF";
2350 unsigned int DECLSPEC_HIDDEN wine_decompose( WCHAR ch, WCHAR *dst, unsigned int dstlen )
2351 {
2352 const WCHAR *ptr = table + table[table[ch >> 8] + ((ch >> 4) & 0x0f)] + 2 * (ch & 0xf);
2353 unsigned int res;
2354
2355 *dst = ch;
2356 if (!*ptr) return 1;
2357 if (dstlen <= 1) return 0;
2358 /* apply the decomposition recursively to the first char */
2359 if ((res = wine_decompose( *ptr, dst, dstlen-1 ))) dst[res++] = ptr[1];
2360 return res;
2361 }
2362 EOF
2363 close OUTPUT;
2364 save_file($filename);
2365 }
2366
2367 ################################################################
2368 # output a codepage definition file from the global tables
2369 sub output_codepage_file($$$$)
2370 {
2371 my ($codepage, $filename, $comment, $has_glyphs) = @_;
2372
2373 my $output = sprintf "libs/port/c_%03d.c", $codepage;
2374 open OUTPUT,">$output.new" or die "Cannot create $output";
2375
2376 printf "Building %s from %s (%s)\n", $output, $filename || "hardcoded data", $comment;
2377
2378 # dump all tables
2379
2380 printf OUTPUT "/* code page %03d (%s) */\n", $codepage, $comment;
2381 if ($filename)
2382 {
2383 print OUTPUT "/* generated from $MAPPINGS/$filename */\n";
2384 print OUTPUT "/* DO NOT EDIT!! */\n\n";
2385 }
2386 else
2387 {
2388 printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
2389 }
2390 printf OUTPUT "#include \"wine/unicode.h\"\n\n";
2391
2392 if (!@lead_bytes) { dump_sbcs_table( $codepage, $has_glyphs, $comment, $default_char, $default_wchar ); }
2393 else { dump_dbcs_table( $codepage, $comment, $default_char, $default_wchar, get_lb_ranges() ); }
2394 close OUTPUT;
2395 save_file($output);
2396 }
2397
2398 ################################################################
2399 # read a "bestfit" Windows mapping file
2400 sub read_bestfit_file($)
2401 {
2402 my ($filename) = @_;
2403 my $state = "";
2404 my ($codepage, $width, $count);
2405 my ($lb_cur, $lb_end);
2406
2407 my $INPUT = open_data_file( $MAPPINGS, $filename ) or die "Cannot open $filename";
2408
2409 while (<$INPUT>)
2410 {
2411 next if /^;/; # skip comments
2412 next if /^\s*$/; # skip empty lines
2413 next if /\x1a/; # skip ^Z
2414 last if /^ENDCODEPAGE/;
2415
2416 if (/^CODEPAGE\s+(\d+)/)
2417 {
2418 $codepage = $1;
2419 next;
2420 }
2421 if (/^CPINFO\s+(\d+)\s+0x([0-9a-fA-f]+)\s+0x([0-9a-fA-F]+)/)
2422 {
2423 $width = $1;
2424 $default_char = hex $2;
2425 $default_wchar = hex $3;
2426 next;
2427 }
2428 if (/^(MBTABLE|WCTABLE|DBCSRANGE|DBCSTABLE)\s+(\d+)/)
2429 {
2430 $state = $1;
2431 $count = $2;
2432 next;
2433 }
2434 if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)/)
2435 {
2436 if ($state eq "MBTABLE")
2437 {
2438 my $cp = hex $1;
2439 my $uni = hex $2;
2440 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
2441 next;
2442 }
2443 if ($state eq "WCTABLE")
2444 {
2445 my $uni = hex $1;
2446 my $cp = hex $2;
2447 $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
2448 next;
2449 }
2450 if ($state eq "DBCSRANGE")
2451 {
2452 my $start = hex $1;
2453 my $end = hex $2;
2454 for (my $i = $start; $i <= $end; $i++) { add_lead_byte( $i ); }
2455 $lb_cur = $start;
2456 $lb_end = $end;
2457 next;
2458 }
2459 if ($state eq "DBCSTABLE")
2460 {
2461 my $mb = hex $1;
2462 my $uni = hex $2;
2463 my $cp = ($lb_cur << 8) | $mb;
2464 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
2465 if (!--$count)
2466 {
2467 if (++$lb_cur > $lb_end) { $state = "DBCSRANGE"; }
2468 }
2469 next;
2470 }
2471 }
2472 die "$filename: Unrecognized line $_\n";
2473 }
2474 close $INPUT;
2475 }
2476
2477
2478 ################################################################
2479 # read an input file and generate the corresponding .c file
2480 sub HANDLE_FILE(@)
2481 {
2482 my ($codepage,$filename,$has_glyphs,$comment,$first_private,$def,$defw) = @_;
2483
2484 @cp2uni = ();
2485 @lead_bytes = ();
2486 @uni2cp = ();
2487 $default_char = $def || $DEF_CHAR;
2488 $default_wchar = $defw || $DEF_CHAR;
2489
2490 # some codepage files are special
2491 if ($codepage == 20932)
2492 {
2493 READ_JIS0208_FILE( $filename );
2494 add_default_mappings( $first_private );
2495 }
2496 elsif ($codepage == 20127)
2497 {
2498 fill_20127_codepage();
2499 add_default_mappings( $first_private );
2500 }
2501 elsif ($filename =~ /\/bestfit/)
2502 {
2503 read_bestfit_file( $filename );
2504 }
2505 else
2506 {
2507 read_codepage_file( $filename );
2508 if ($codepage == 10001)
2509 {
2510 # add Shift-JIS mappings
2511 read_bestfit_file( "VENDORS/MICSFT/WindowsBestFit/bestfit932.txt" );
2512 }
2513 else
2514 {
2515 add_default_mappings( $first_private );
2516 }
2517 }
2518
2519 output_codepage_file( $codepage, $filename, $comment, $has_glyphs );
2520 }
2521
2522
2523 ################################################################
2524 # save a file if modified
2525 sub save_file($)
2526 {
2527 my $file = shift;
2528 if (-f $file && !system "cmp $file $file.new >/dev/null")
2529 {
2530 unlink "$file.new";
2531 }
2532 else
2533 {
2534 rename "$file.new", "$file";
2535 }
2536 }
2537
2538
2539 ################################################################
2540 # output the list of codepage tables into the cptable.c file
2541 sub output_cptable($)
2542 {
2543 my $output = shift;
2544 my @tables_decl = ();
2545
2546 printf "Building %s\n", $output;
2547
2548 foreach my $file (@allfiles)
2549 {
2550 my ($codepage,$filename) = @$file;
2551 push @tables_decl, sprintf("extern union cptable cptable_%03d;\n",$codepage);
2552 }
2553
2554 push @tables_decl, sprintf("\nstatic const union cptable * const cptables[%d] =\n{\n",$#allfiles+1);
2555 foreach my $file (@allfiles)
2556 {
2557 my ($codepage,$filename) = @$file;
2558 push @tables_decl, sprintf(" &cptable_%03d,\n", $codepage);
2559 }
2560 push @tables_decl, "};";
2561 REPLACE_IN_FILE( $output, @tables_decl );
2562 }
2563
2564 ################################################################
2565 # replace the contents of a file between ### cpmap ### marks
2566
2567 sub REPLACE_IN_FILE($@)
2568 {
2569 my $name = shift;
2570 my @data = @_;
2571 my @lines = ();
2572 open(FILE,$name) or die "Can't open $name";
2573 while (<FILE>)
2574 {
2575 push @lines, $_;
2576 last if /\#\#\# cpmap begin \#\#\#/;
2577 }
2578 push @lines, @data;
2579 while (<FILE>)
2580 {
2581 if (/\#\#\# cpmap end \#\#\#/) { push @lines, "\n", $_; last; }
2582 }
2583 push @lines, <FILE>;
2584 open(FILE,">$name.new") or die "Can't modify $name";
2585 print FILE @lines;
2586 close(FILE);
2587 save_file($name);
2588 }
2589
2590 ################################################################
2591 # main routine
2592
2593 chdir ".." if -f "./make_unicode";
2594 READ_DEFAULTS( $DEFAULTS );
2595 DUMP_CASE_MAPPINGS( "libs/wine/casemap.c" );
2596 DUMP_SORTKEYS( "libs/wine/collation.c", READ_SORTKEYS_FILE() );
2597 dump_compose_table( "libs/port/compose.c" );
2598 dump_decompose_table( "libs/port/decompose.c" );
2599 DUMP_CTYPE_TABLES( "libs/wine/wctype.c" );
2600 dump_digit_folding( "libs/port/digitmap.c" );
2601 dump_mirroring( "dlls/usp10/mirror.c" );
2602 dump_mirroring( "dlls/dwrite/mirror.c" );
2603 dump_bracket( "dlls/usp10/bracket.c" );
2604 dump_bracket( "dlls/dwrite/bracket.c" );
2605 dump_shaping( "dlls/usp10/shaping.c" );
2606 dump_linebreak( "dlls/usp10/linebreak.c" );
2607 dump_linebreak( "dlls/dwrite/linebreak.c" );
2608 dump_scripts( "dlls/dwrite/scripts" );
2609 dump_indic( "dlls/usp10/indicsyllable.c" );
2610 dump_intl_nls("loader/l_intl.nls");
2611 dump_vertical( "dlls/gdi32/vertical.c" );
2612 dump_vertical( "dlls/wineps.drv/vertical.c" );
2613 dump_nameprep( "dlls/kernel32/nameprep.c" );
2614
2615 foreach my $file (@allfiles) { HANDLE_FILE( @{$file} ); }
2616
2617 output_cptable("libs/port/cptable.c");
2618
2619 exit 0;
2620
2621 # Local Variables:
2622 # compile-command: "./make_unicode"
2623 # End:
Windows API implementation