?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
bcrypt: Add BCryptDeriveKey stub.
[wine.git] / tools / make_unicode
blob4b79668a1a0f10fe794246652409638a1c064be5
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/12.0.0/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 "Consonant_Initial_Postfixed" => 0x0023,
177 );
178
179 my %matra_types =
180 (
181 "Right" => 0x01,
182 "Left" => 0x02,
183 "Visual_Order_Left" => 0x03,
184 "Left_And_Right" => 0x04,
185 "Top" => 0x05,
186 "Bottom" => 0x06,
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,
194 "Invisible" => 0x0e,
195 "Bottom_And_Left" => 0x0f
196 );
197
198 my %nameprep_flags =
199 (
200 "unassigned" => 0x01,
201 "prohibited" => 0x02,
202 "bidi_ral" => 0x04,
203 "bidi_l" => 0x08
204 );
205
206 my %break_types =
207 (
208 "BK" => 0x0001,
209 "CR" => 0x0002,
210 "LF" => 0x0003,
211 "CM" => 0x0004,
212 "SG" => 0x0005,
213 "GL" => 0x0006,
214 "CB" => 0x0007,
215 "SP" => 0x0008,
216 "ZW" => 0x0009,
217 "NL" => 0x000a,
218 "WJ" => 0x000b,
219 "JL" => 0x000c,
220 "JV" => 0x000d,
221 "JT" => 0x000e,
222 "H2" => 0x000f,
223 "H3" => 0x0010,
224 "XX" => 0x0011,
225 "OP" => 0x0012,
226 "CL" => 0x0013,
227 "CP" => 0x0014,
228 "QU" => 0x0015,
229 "NS" => 0x0016,
230 "EX" => 0x0017,
231 "SY" => 0x0018,
232 "IS" => 0x0019,
233 "PR" => 0x001a,
234 "PO" => 0x001b,
235 "NU" => 0x001c,
236 "AL" => 0x001d,
237 "ID" => 0x001e,
238 "IN" => 0x001f,
239 "HY" => 0x0020,
240 "BB" => 0x0021,
241 "BA" => 0x0022,
242 "SA" => 0x0023,
243 "AI" => 0x0024,
244 "B2" => 0x0025,
245 "HL" => 0x0026,
246 "CJ" => 0x0027,
247 "RI" => 0x0028,
248 "EB" => 0x0029,
249 "EM" => 0x002a,
250 "ZWJ" => 0x002b,
251 );
252
253 my %vertical_types =
254 (
255 "R" => 0x0000,
256 "U" => 0x0001,
257 "Tr" => 0x0002,
258 "Tu" => 0x0003,
259 );
260
261 my %categories =
262 (
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
293 );
294
295 # a few characters need additional categories that cannot be determined automatically
296 my %special_categories =
297 (
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,
307 0xd7, 0xf7 ],
308 "digit" => [ 0xb2, 0xb3, 0xb9 ],
309 "lower" => [ 0x2071, 0x207f ]
310 );
311
312 my %directions =
313 (
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
337 );
338
339 my %c2_types =
340 (
341 "L" => 1, # C2_LEFTTORIGHT
342 "R" => 2, # C2_RIGHTTOLEFT
343 "AL" => 2, # C2_RIGHTTOLEFT
344 "EN" => 3, # C2_EUROPENUMBER
345 "ES" => 4, # C2_EUROPESEPARATOR
346 "ET" => 5, # C2_EUROPETERMINATOR
347 "AN" => 6, # C2_ARABICNUMBER
348 "CS" => 7, # C2_COMMONSEPARATOR
349 "NSM" => 0, # C2_NOTAPPLICABLE
350 "BN" => 0, # C2_NOTAPPLICABLE
351 "B" => 8, # C2_BLOCKSEPARATOR
352 "S" => 9, # C2_SEGMENTSEPARATOR
353 "WS" => 10, # C2_WHITESPACE
354 "ON" => 11, # C2_OTHERNEUTRAL
355 "LRE" => 11, # C2_OTHERNEUTRAL
356 "LRO" => 11, # C2_OTHERNEUTRAL
357 "RLE" => 11, # C2_OTHERNEUTRAL
358 "RLO" => 11, # C2_OTHERNEUTRAL
359 "PDF" => 11, # C2_OTHERNEUTRAL
360 "LRI" => 11, # C2_OTHERNEUTRAL
361 "RLI" => 11, # C2_OTHERNEUTRAL
362 "FSI" => 11, # C2_OTHERNEUTRAL
363 "PDI" => 11 # C2_OTHERNEUTRAL
364 );
365
366 my %bidi_types =
367 (
368 "ON" => 0, # Other Neutrals
369 "L" => 1, # Left-to-Right
370 "R" => 2, # Right-to-Left
371 "AN" => 3, # Arabic Number
372 "EN" => 4, # European Number
373 "AL" => 5, # Right-to-Left Arabic
374 "NSM" => 6, # Non-Spacing Mark
375 "CS" => 7, # Common Number Separator
376 "ES" => 8, # European Number Separator
377 "ET" => 9, # European Number Terminator
378 "BN" => 10, # Boundary Neutral
379 "S" => 11, # Segment Separator
380 "WS" => 12, # Whitespace
381 "B" => 13, # Paragraph Separator
382 "RLO" => 14, # Right-to-Left Override
383 "RLE" => 15, # Right-to-Left Embedding
384 "LRO" => 16, # Left-to-Right Override
385 "LRE" => 17, # Left-to-Right Embedding
386 "PDF" => 18, # Pop Directional Format
387 "LRI" => 19, # Left-to-Right Isolate
388 "RLI" => 20, # Right-to-Left Isolate
389 "FSI" => 21, # First Strong Isolate
390 "PDI" => 22 # Pop Directional Isolate
391 );
392
393 my %joining_types =
394 (
395 "U" => 0, # Non_Joining
396 "T" => 1, # Transparent
397 "R" => 2, # Right_Joining
398 "L" => 3, # Left_Joining
399 "D" => 4, # Dual_Joining
400 "C" => 5, # Join_Causing
401 );
402
403 my @cp2uni = ();
404 my @lead_bytes = ();
405 my @uni2cp = ();
406 my @unicode_defaults = ();
407 my @unicode_aliases = ();
408 my @tolower_table = ();
409 my @toupper_table = ();
410 my @digitmap_table = ();
411 my @compatmap_table = ();
412 my @category_table = ();
413 my @joining_table = ();
414 my @direction_table = ();
415 my @decomp_table = ();
416 my @compose_table = ();
417 my @combining_class_table = ();
418 my @full_compatmap_table = ();
419 my $default_char;
420 my $default_wchar;
421
422 my %joining_forms =
423 (
424 "isolated" => [],
425 "final" => [],
426 "initial" => [],
427 "medial" => []
428 );
429
430 ################################################################
431 # fetch a unicode.org file and open it
432 sub open_data_file($$)
433 {
434 my ($base, $name) = @_;
435 (my $dir = "data/$name") =~ s/\/[^\/]+$//;
436 local *FILE;
437 unless (-f "data/$name")
438 {
439 print "Fetching $base/$name...\n";
440 system "mkdir", "-p", $dir;
441 !system "wget", "-q", "-O", "data/$name", "$base/$name" or die "cannot fetch $base/$name";
442 }
443 open FILE, "<data/$name" or die "cannot open data/$name";
444 return *FILE;
445 }
446
447 ################################################################
448 # read in the defaults file
449 sub READ_DEFAULTS($)
450 {
451 my $filename = shift;
452 my $start;
453
454 # first setup a few default mappings
455
456 open DEFAULTS, "$filename" or die "Cannot open $filename";
457 print "Loading $filename\n";
458 while (<DEFAULTS>)
459 {
460 next if /^\#/; # skip comments
461 next if /^$/; # skip empty lines
462 if (/^(([0-9a-fA-F]+)(,[0-9a-fA-F]+)*)\s+([0-9a-fA-F]+|'.'|none)\s+(\#.*)?/)
463 {
464 my @src = map hex, split /,/,$1;
465 my $dst = $4;
466 my $comment = $5;
467 if ($#src > 0) { push @unicode_aliases, \@src; }
468 next if ($dst eq "none");
469 $dst = ($dst =~ /\'.\'/) ? ord substr($dst,1,1) : hex $dst;
470 foreach my $src (@src)
471 {
472 die "Duplicate value" if defined($unicode_defaults[$src]);
473 $unicode_defaults[$src] = $dst;
474 }
475 next;
476 }
477 die "Unrecognized line $_\n";
478 }
479 close DEFAULTS;
480
481 # now build mappings from the decomposition field of the Unicode database
482
483 my $UNICODE_DATA = open_data_file( $UNIDATA, "UnicodeData.txt" );
484 while (<$UNICODE_DATA>)
485 {
486 # Decode the fields ...
487 my ($code, $name, $cat, $comb, $bidi,
488 $decomp, $dec, $dig, $num, $mirror,
489 $oldname, $comment, $upper, $lower, $title) = split /;/;
490 my $dst;
491 my $src = hex $code;
492
493 die "unknown category $cat" unless defined $categories{$cat};
494 die "unknown directionality $bidi" unless defined $directions{$bidi};
495
496 $category_table[$src] = $categories{$cat};
497 $direction_table[$src] = $bidi;
498 $joining_table[$src] = $joining_types{"T"} if $cat eq "Mn" || $cat eq "Me" || $cat eq "Cf";
499
500 if ($lower ne "")
501 {
502 $tolower_table[$src] = hex $lower;
503 }
504 if ($upper ne "")
505 {
506 $toupper_table[$src] = hex $upper;
507 }
508 if ($dec ne "")
509 {
510 $category_table[$src] |= $ctype{"digit"};
511 }
512 if ($dig ne "")
513 {
514 $digitmap_table[$src] = ord $dig;
515 }
516 if ($comb ne "")
517 {
518 $combining_class_table[$src] = $comb;
519 }
520
521 # copy the category and direction for everything between First/Last pairs
522 if ($name =~ /, First>/) { $start = $src; }
523 if ($name =~ /, Last>/)
524 {
525 while ($start < $src)
526 {
527 $category_table[$start] = $category_table[$src];
528 $direction_table[$start] = $direction_table[$src];
529 $start++;
530 }
531 }
532
533 next if $decomp eq ""; # no decomposition, skip it
534
535 if ($decomp =~ /^<([a-zA-Z]+)>\s+([0-9a-fA-F]+)/)
536 {
537 my @seq = ();
538 for my $ch (split /\s+/, (split /\s+/, $decomp, 2)[1])
539 {
540 push @seq, (hex $ch);
541 }
542 $full_compatmap_table[$src] = \@seq;
543 }
544
545 if ($decomp =~ /^<([a-zA-Z]+)>\s+([0-9a-fA-F]+)$/)
546 {
547 # decomposition of the form "<foo> 1234" -> use char if type is known
548 if (($src >= 0xf900 && $src < 0xfb00) || ($src >= 0xfe30 && $src < 0xfffd))
549 {
550 # Single char decomposition in the compatibility range
551 $compatmap_table[$src] = hex $2;
552 }
553 if ($1 eq "isolated" || $1 eq "final" || $1 eq "initial" || $1 eq "medial")
554 {
555 ${joining_forms{$1}}[hex $2] = $src;
556 next;
557 }
558 next unless ($1 eq "font" ||
559 $1 eq "noBreak" ||
560 $1 eq "circle" ||
561 $1 eq "super" ||
562 $1 eq "sub" ||
563 $1 eq "wide" ||
564 $1 eq "narrow" ||
565 $1 eq "compat" ||
566 $1 eq "small");
567 $dst = hex $2;
568 }
569 elsif ($decomp =~ /^<compat>\s+0020\s+([0-9a-fA-F]+)/)
570 {
571 # decomposition "<compat> 0020 1234" -> combining accent
572 $dst = hex $1;
573 }
574 elsif ($decomp =~ /^([0-9a-fA-F]+)/)
575 {
576 # decomposition contains only char values without prefix -> use first char
577 $dst = hex $1;
578 $category_table[$src] |= $category_table[$dst] if defined $category_table[$dst];
579 # store decomposition if it contains two chars
580 if ($decomp =~ /^([0-9a-fA-F]+)\s+([0-9a-fA-F]+)$/)
581 {
582 $decomp_table[$src] = [ hex $1, hex $2 ];
583 push @compose_table, [ hex $1, hex $2, $src ];
584 }
585 elsif ($decomp =~ /^([0-9a-fA-F]+)$/)
586 {
587 # Single char decomposition
588 if (hex $1 < 65536)
589 {
590 $decomp_table[$src] = [ hex $1, 0 ];
591 }
592
593 if (($src >= 0xf900 && $src < 0xfb00) || ($src >= 0xfe30 && $src < 0xfffd))
594 {
595 # Single char decomposition in the compatibility range
596 $compatmap_table[$src] = hex $1;
597 }
598 }
599 }
600 else
601 {
602 next;
603 }
604
605 next if defined($unicode_defaults[$src]); # may have been set in the defaults file
606
607 # check for loops
608 for (my $i = $dst; ; $i = $unicode_defaults[$i])
609 {
610 die sprintf("loop detected for %04x -> %04x",$src,$dst) if $i == $src;
611 last unless defined($unicode_defaults[$i]);
612 }
613 $unicode_defaults[$src] = $dst;
614 }
615 close $UNICODE_DATA;
616
617 # patch the category of some special characters
618
619 foreach my $cat (keys %special_categories)
620 {
621 my $flag = $ctype{$cat};
622 foreach my $i (@{$special_categories{$cat}}) { $category_table[$i] |= $flag; }
623 }
624 }
625
626
627 ################################################################
628 # define a new lead byte
629 sub add_lead_byte($)
630 {
631 my $ch = shift;
632 push @lead_bytes, $ch;
633 $cp2uni[$ch] = 0;
634 }
635
636 ################################################################
637 # define a new char mapping
638 sub add_mapping($$)
639 {
640 my ($cp, $uni) = @_;
641 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
642 $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
643 if ($cp > 0xff && !defined($cp2uni[$cp >> 8]))
644 {
645 add_lead_byte( $cp >> 8 );
646 }
647 }
648
649 ################################################################
650 # parse a standard codepage file
651 sub read_codepage_file($)
652 {
653 my $name = shift;
654 my $INPUT = open_data_file( $MAPPINGS, $name );
655
656 while (<$INPUT>)
657 {
658 next if /^\#/; # skip comments
659 next if /^$/; # skip empty lines
660 next if /\x1a/; # skip ^Z
661 next if (/^0x([0-9a-fA-F]+)\s+\#UNDEFINED/); # undefined char
662
663 if (/^0x([0-9a-fA-F]+)\s+\#DBCS LEAD BYTE/)
664 {
665 add_lead_byte( hex $1 );
666 next;
667 }
668 # 0x12 0x3456
669 if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\s+(\#.*)?/)
670 {
671 add_mapping( hex $1, hex $2 );
672 next;
673 }
674 # 0x12 0x3456+0x7890
675 if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\+.*\s+(\#.*)?/)
676 {
677 add_mapping( hex $1, hex $2 );
678 next;
679 }
680 # 0x12 <LR>+0x3456
681 if (/^0x([0-9a-fA-F]+)\s+<LR>\+0x([0-9a-fA-F]+)\s+(\#.*)?/)
682 {
683 add_mapping( hex $1, hex $2 );
684 next;
685 }
686 # 0x12 <RL>+0x3456
687 if (/^0x([0-9a-fA-F]+)\s+<RL>\+0x([0-9a-fA-F]+)\s+(\#.*)?/)
688 {
689 my $uni = hex $2;
690 add_mapping( hex $1, $uni ) if $uni >= 0x80;
691 next;
692 }
693 die "$name: Unrecognized line $_\n";
694 }
695 close $INPUT;
696 }
697
698
699 ################################################################
700 # fill input data for the 20127 (us-ascii) codepage
701 sub fill_20127_codepage()
702 {
703 for (my $i = 0; $i < 128; $i++) { $cp2uni[$i] = $uni2cp[$i] = $i; }
704 for (my $i = 128; $i < 256; $i++) { $cp2uni[$i] = $i & 0x7f; }
705 }
706
707 ################################################################
708 # get a mapping including glyph chars for MB_USEGLYPHCHARS
709
710 sub get_glyphs_mapping(@)
711 {
712 $_[0x01] = 0x263a; # (WHITE SMILING FACE)
713 $_[0x02] = 0x263b; # (BLACK SMILING FACE)
714 $_[0x03] = 0x2665; # (BLACK HEART SUIT)
715 $_[0x04] = 0x2666; # (BLACK DIAMOND SUIT)
716 $_[0x05] = 0x2663; # (BLACK CLUB SUIT)
717 $_[0x06] = 0x2660; # (BLACK SPADE SUIT)
718 $_[0x07] = 0x2022; # (BULLET)
719 $_[0x08] = 0x25d8; # (INVERSE BULLET)
720 $_[0x09] = 0x25cb; # (WHITE CIRCLE)
721 $_[0x0a] = 0x25d9; # (INVERSE WHITE CIRCLE)
722 $_[0x0b] = 0x2642; # (MALE SIGN)
723 $_[0x0c] = 0x2640; # (FEMALE SIGN)
724 $_[0x0d] = 0x266a; # (EIGHTH NOTE)
725 $_[0x0e] = 0x266b; # (BEAMED EIGHTH NOTES)
726 $_[0x0f] = 0x263c; # (WHITE SUN WITH RAYS)
727 $_[0x10] = 0x25ba; # (BLACK RIGHT-POINTING POINTER)
728 $_[0x11] = 0x25c4; # (BLACK LEFT-POINTING POINTER)
729 $_[0x12] = 0x2195; # (UP DOWN ARROW)
730 $_[0x13] = 0x203c; # (DOUBLE EXCLAMATION MARK)
731 $_[0x14] = 0x00b6; # (PILCROW SIGN)
732 $_[0x15] = 0x00a7; # (SECTION SIGN)
733 $_[0x16] = 0x25ac; # (BLACK RECTANGLE)
734 $_[0x17] = 0x21a8; # (UP DOWN ARROW WITH BASE)
735 $_[0x18] = 0x2191; # (UPWARDS ARROW)
736 $_[0x19] = 0x2193; # (DOWNWARDS ARROW)
737 $_[0x1a] = 0x2192; # (RIGHTWARDS ARROW)
738 $_[0x1b] = 0x2190; # (LEFTWARDS ARROW)
739 $_[0x1c] = 0x221f; # (RIGHT ANGLE)
740 $_[0x1d] = 0x2194; # (LEFT RIGHT ARROW)
741 $_[0x1e] = 0x25b2; # (BLACK UP-POINTING TRIANGLE)
742 $_[0x1f] = 0x25bc; # (BLACK DOWN-POINTING TRIANGLE)
743 $_[0x7f] = 0x2302; # (HOUSE)
744 return @_;
745 }
746
747 ################################################################
748 # build EUC-JP table from the JIS 0208 file
749 # FIXME: for proper EUC-JP we should probably read JIS 0212 too
750 # but this would require 3-byte DBCS characters
751 sub READ_JIS0208_FILE($)
752 {
753 my $name = shift;
754
755 # ASCII chars
756 for (my $i = 0x00; $i <= 0x7f; $i++)
757 {
758 $cp2uni[$i] = $i;
759 $uni2cp[$i] = $i;
760 }
761
762 # JIS X 0201 right plane
763 for (my $i = 0xa1; $i <= 0xdf; $i++)
764 {
765 $cp2uni[0x8e00 + $i] = 0xfec0 + $i;
766 $uni2cp[0xfec0 + $i] = 0x8e00 + $i;
767 }
768
769 # lead bytes
770 foreach my $i (0x8e, 0x8f, 0xa1 .. 0xfe)
771 {
772 add_lead_byte($i);
773 }
774
775 # undefined chars
776 foreach my $i (0x80 .. 0x8d, 0x90 .. 0xa0, 0xff)
777 {
778 $cp2uni[$i] = $DEF_CHAR;
779 }
780
781 # Shift-JIS compatibility
782 $uni2cp[0x00a5] = 0x5c;
783 $uni2cp[0x203e] = 0x7e;
784
785 # Fix backslash conversion
786 $cp2uni[0xa1c0] = 0xff3c;
787 $uni2cp[0xff3c] = 0xa1c0;
788
789 my $INPUT = open_data_file( $MAPPINGS, $name );
790 while (<$INPUT>)
791 {
792 next if /^\#/; # skip comments
793 next if /^$/; # skip empty lines
794 next if /\x1a/; # skip ^Z
795 if (/^0x[0-9a-fA-F]+\s+0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)\s+(\#.*)?/)
796 {
797 my $cp = 0x8080 + hex $1;
798 my $uni = hex $2;
799 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
800 $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
801 next;
802 }
803 die "$name: Unrecognized line $_\n";
804 }
805 close $INPUT;
806 }
807
808
809 ################################################################
810 # build the sort keys table
811 sub READ_SORTKEYS_FILE()
812 {
813 my @sortkeys = ();
814 for (my $i = 0; $i < 65536; $i++) { $sortkeys[$i] = [ -1, 0, 0, 0, 0 ] };
815
816 my $INPUT = open_data_file( $REPORTS, $SORTKEYS );
817 while (<$INPUT>)
818 {
819 next if /^\#/; # skip comments
820 next if /^$/; # skip empty lines
821 next if /\x1a/; # skip ^Z
822 next if /^\@version/; # skip @version header
823 if (/^([0-9a-fA-F]+)\s+;\s+\[([*.])([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]+)\]/)
824 {
825 my ($uni,$variable) = (hex $1, $2);
826 next if $uni > 65535;
827 $sortkeys[$uni] = [ $uni, hex $3, hex $4, hex $5, hex $6 ];
828 next;
829 }
830 if (/^([0-9a-fA-F]+\s+)+;\s+\[[*.]([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]{4})\.([0-9a-fA-F]+)\]/)
831 {
832 # multiple character sequence, ignored for now
833 next;
834 }
835 die "$SORTKEYS: Unrecognized line $_\n";
836 }
837 close $INPUT;
838
839 # compress the keys to 32 bit:
840 # key 1 to 16 bits, key 2 to 8 bits, key 3 to 4 bits, key 4 to 1 bit
841
842 @sortkeys = sort { ${$a}[1] <=> ${$b}[1] or
843 ${$a}[2] <=> ${$b}[2] or
844 ${$a}[3] <=> ${$b}[3] or
845 ${$a}[4] <=> ${$b}[4] or
846 $a cmp $b; } @sortkeys;
847
848 my ($n2, $n3) = (1, 1);
849 my @keys = (-1, -1, -1, -1, -1 );
850 my @flatkeys = ();
851
852 for (my $i = 0; $i < 65536; $i++)
853 {
854 my @current = @{$sortkeys[$i]};
855 next if $current[0] == -1;
856 if ($current[1] == $keys[1])
857 {
858 if ($current[2] == $keys[2])
859 {
860 if ($current[3] == $keys[3])
861 {
862 # nothing
863 }
864 else
865 {
866 $keys[3] = $current[3];
867 $n3++;
868 die if ($n3 >= 16);
869 }
870 }
871 else
872 {
873 $keys[2] = $current[2];
874 $keys[3] = $current[3];
875 $n2++;
876 $n3 = 1;
877 die if ($n2 >= 256);
878 }
879 }
880 else
881 {
882 $keys[1] = $current[1];
883 $keys[2] = $current[2];
884 $keys[3] = $current[3];
885 $n2 = 1;
886 $n3 = 1;
887 }
888
889 if ($current[2]) { $current[2] = $n2; }
890 if ($current[3]) { $current[3] = $n3; }
891 if ($current[4]) { $current[4] = 1; }
892
893 $flatkeys[$current[0]] = ($current[1] << 16) | ($current[2] << 8) | ($current[3] << 4) | $current[4];
894 }
895 return @flatkeys;
896 }
897
898
899 ################################################################
900 # build the sort keys table
901 sub dump_sortkeys($@)
902 {
903 my ($filename, @keys) = @_;
904
905 # count the number of 256-key ranges that contain something
906
907 my @offsets = ();
908 my $ranges = 2;
909 for (my $i = 0; $i < 256; $i++) { $offsets[$i] = 256; }
910 for (my $i = 0; $i < 65536; $i++)
911 {
912 next unless defined $keys[$i];
913 $offsets[$i >> 8] = $ranges * 256;
914 $ranges++;
915 $i |= 255;
916 }
917
918 # output the range offsets
919
920 open OUTPUT,">$filename.new" or die "Cannot create $filename";
921 printf "Building $filename\n";
922 printf OUTPUT "/* Unicode collation element table */\n";
923 printf OUTPUT "/* generated from %s */\n", "$REPORTS/$SORTKEYS";
924 printf OUTPUT "/* DO NOT EDIT!! */\n\n";
925 print OUTPUT "#include \"windef.h\"\n\n";
926
927 printf OUTPUT "const unsigned int DECLSPEC_HIDDEN collation_table[%d] =\n{\n", $ranges*256;
928 printf OUTPUT " /* index */\n";
929 printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%08x", 0, @offsets );
930
931 # output the default values
932
933 printf OUTPUT " /* defaults */\n";
934 printf OUTPUT "%s", DUMP_ARRAY( "0x%08x", 0, (0xffffffff) x 256 );
935
936 # output all the key ranges
937
938 for (my $i = 0; $i < 256; $i++)
939 {
940 next if $offsets[$i] == 256;
941 printf OUTPUT ",\n /* 0x%02x00 .. 0x%02xff */\n", $i, $i;
942 printf OUTPUT "%s", DUMP_ARRAY( "0x%08x", 0xffffffff, @keys[($i<<8) .. ($i<<8)+255] );
943 }
944 printf OUTPUT "\n};\n";
945 close OUTPUT;
946 save_file($filename);
947 }
948
949
950 ################################################################
951 # add default mappings once the file had been read
952 sub add_default_mappings($)
953 {
954 my $first_private = shift;
955
956 # Apply aliases
957
958 foreach my $alias (@unicode_aliases)
959 {
960 my $target = undef;
961 foreach my $src (@$alias)
962 {
963 if (defined($uni2cp[$src]))
964 {
965 $target = $uni2cp[$src];
966 last;
967 }
968 }
969 next unless defined($target);
970
971 # At least one char of the alias set is defined, set the others to the same value
972 foreach my $src (@$alias)
973 {
974 $uni2cp[$src] = $target unless defined($uni2cp[$src]);
975 }
976 }
977
978 # For every src -> target mapping in the defaults table,
979 # make uni2cp[src] = uni2cp[target] if uni2cp[target] is defined
980
981 for (my $src = 0; $src < 65536; $src++)
982 {
983 next if defined($uni2cp[$src]); # source has a definition already
984 next unless defined($unicode_defaults[$src]); # no default for this char
985 my $target = $unicode_defaults[$src];
986
987 # do a recursive mapping until we find a target char that is defined
988 while (!defined($uni2cp[$target]) &&
989 defined($unicode_defaults[$target])) { $target = $unicode_defaults[$target]; }
990
991 if (defined($uni2cp[$target])) { $uni2cp[$src] = $uni2cp[$target]; }
992 }
993
994 # Add a private char mapping for undefined chars
995
996 if ($first_private)
997 {
998 for (my $i = 32; $i < 256; $i++)
999 {
1000 next if defined($cp2uni[$i]);
1001 $cp2uni[$i] = $first_private;
1002 $uni2cp[$first_private] = $i;
1003 $first_private++;
1004 }
1005 }
1006
1007 # Add an identity mapping for all undefined chars
1008
1009 for (my $i = 0; $i < 256; $i++)
1010 {
1011 next if defined($cp2uni[$i]);
1012 next if defined($uni2cp[$i]);
1013 $cp2uni[$i] = $uni2cp[$i] = $i;
1014 }
1015 }
1016
1017 ################################################################
1018 # dump an array of integers
1019 sub DUMP_ARRAY($$@)
1020 {
1021 my ($format,$default,@array) = @_;
1022 my $i;
1023 my $ret = " ";
1024 for ($i = 0; $i < $#array; $i++)
1025 {
1026 $ret .= sprintf($format, defined $array[$i] ? $array[$i] : $default);
1027 $ret .= (($i % 8) != 7) ? ", " : ",\n ";
1028 }
1029 $ret .= sprintf($format, defined $array[$i] ? $array[$i] : $default);
1030 return $ret;
1031 }
1032
1033
1034 ################################################################
1035 # dump a unicode->ascii mapping table
1036 sub dump_uni2cp_table($$)
1037 {
1038 my ($width, $def) = @_;
1039
1040 # count the number of unicode->ascii subtables that contain something
1041
1042 my @lowerbounds;
1043 my @upperbounds;
1044 my $index = 0;
1045 my @filled = (-1) x 256;
1046 for (my $i = 0; $i < 65536; $i++)
1047 {
1048 next unless defined $uni2cp[$i];
1049 if ($filled[$i >> 8] == -1)
1050 {
1051 $lowerbounds[$index] = $i & 0xff;
1052 $upperbounds[$index] = 0xff - $lowerbounds[$index];
1053 $filled[$i >> 8] = $index * 256;
1054 $index++;
1055 }
1056 else
1057 {
1058 $upperbounds[$index-1] = 0xff - ($i & 0xff);
1059 }
1060 }
1061
1062 # add defaults mapping at the end
1063 $filled[256] = $index * 256;
1064 $lowerbounds[$index] = 255;
1065 $upperbounds[$index] = 0;
1066 $index++;
1067
1068 # collapse blocks upwards if possible
1069 my $removed = 0;
1070 $index = 0;
1071 for (my $i = 0; $i < 257; $i++)
1072 {
1073 next if $filled[$i] == -1;
1074 if ($upperbounds[$index - 1] > $lowerbounds[$index])
1075 {
1076 $removed += $lowerbounds[$index];
1077 }
1078 else
1079 {
1080 $removed += $upperbounds[$index - 1];
1081 $lowerbounds[$index] = $upperbounds[$index - 1];
1082 }
1083 $filled[$i] -= $removed;
1084 $index++;
1085 }
1086
1087 # output all the subtables into a single array
1088
1089 printf OUTPUT "static const unsigned %s uni2cp_low[] =\n{\n", $width == 8 ? "char" : "short";
1090
1091 my $format = $width == 8 ? "0x%02x" : "0x%04x";
1092 for (my $i = $index = 0; $i < 257; $i++)
1093 {
1094 next if $filled[$i] == -1;
1095 my $start = ($i << 8) + $lowerbounds[$index];
1096 my $end = ($i << 8) + 255;
1097 if ($i == 256)
1098 {
1099 print OUTPUT " /* defaults */\n";
1100 printf OUTPUT "%s\n};\n\n", DUMP_ARRAY( $format, 0, ($def) x ($end - $start + 1) );
1101 }
1102 else
1103 {
1104 printf OUTPUT " /* 0x%04x .. 0x%04x */\n", $start, $end;
1105 printf OUTPUT "%s,\n", DUMP_ARRAY( $format, $def, @uni2cp[$start .. $end] );
1106 }
1107 $index++;
1108 }
1109
1110 # output a table of the offsets of the subtables in the previous array
1111
1112 for (my $i = 0; $i < 256; $i++)
1113 {
1114 if ($filled[$i] == -1) { $filled[$i] = $filled[256]; }
1115 }
1116 printf OUTPUT "static const unsigned short uni2cp_high[256] =\n";
1117 printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%04x", 0, @filled[0..255] );
1118 }
1119
1120 ################################################################
1121 # dump an SBCS mapping table
1122 sub dump_sbcs_table($$$$$)
1123 {
1124 my ($codepage, $has_glyphs, $name, $def, $defw) = @_;
1125 my $i;
1126
1127 # output the ascii->unicode table
1128
1129 if ($has_glyphs)
1130 {
1131 printf OUTPUT "static const WCHAR cp2uni[512] =\n";
1132 printf OUTPUT "{\n%s", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[0 .. 255] );
1133 printf OUTPUT ",\n /* glyphs */\n%s\n};\n\n",
1134 DUMP_ARRAY( "0x%04x", $defw, get_glyphs_mapping(@cp2uni[0 .. 255]) );
1135 }
1136 else
1137 {
1138 printf OUTPUT "static const WCHAR cp2uni[256] =\n";
1139 printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[0 .. 255] );
1140 }
1141
1142 dump_uni2cp_table( 8, $def );
1143
1144 # output the code page descriptor
1145
1146 printf OUTPUT "const struct sbcs_table DECLSPEC_HIDDEN cptable_%03d =\n{\n", $codepage;
1147 printf OUTPUT " { %d, 1, 0x%04x, 0x%04x, \"%s\" },\n",
1148 $codepage, $def, $defw, $name;
1149 printf OUTPUT " cp2uni,\n";
1150 if ($has_glyphs) { printf OUTPUT " cp2uni + 256,\n"; }
1151 else { printf OUTPUT " cp2uni,\n"; }
1152 printf OUTPUT " uni2cp_low,\n";
1153 printf OUTPUT " uni2cp_high\n};\n";
1154 }
1155
1156
1157 ################################################################
1158 # dump a DBCS mapping table
1159 sub dump_dbcs_table($$$$@)
1160 {
1161 my ($codepage, $name, $def, $defw, @lb_ranges) = @_;
1162
1163 # build a list of lead bytes that are actually used
1164
1165 my @lblist = ();
1166 LBLOOP: for (my $y = 0; $y <= $#lead_bytes; $y++)
1167 {
1168 $cp2uni[$lead_bytes[$y]] = $defw;
1169 my $base = $lead_bytes[$y] << 8;
1170 for (my $x = 0; $x < 256; $x++)
1171 {
1172 if (defined $cp2uni[$base+$x])
1173 {
1174 push @lblist,$lead_bytes[$y];
1175 next LBLOOP;
1176 }
1177 }
1178 }
1179 my $unused = ($#lead_bytes > $#lblist);
1180
1181 # output the ascii->unicode table for the single byte chars
1182
1183 printf OUTPUT "static const WCHAR cp2uni[%d] =\n", 256 * ($#lblist + 2 + $unused);
1184 printf OUTPUT "{\n%s,\n", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[0 .. 255] );
1185
1186 # output the default table for unused lead bytes
1187
1188 if ($unused)
1189 {
1190 printf OUTPUT " /* unused lead bytes */\n";
1191 printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%04x", 0, ($defw) x 256 );
1192 }
1193
1194 # output the ascii->unicode table for each DBCS lead byte
1195
1196 for (my $y = 0; $y <= $#lblist; $y++)
1197 {
1198 my $base = $lblist[$y] << 8;
1199 printf OUTPUT " /* lead byte %02x */\n", $lblist[$y];
1200 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", $defw, @cp2uni[$base .. $base+255] );
1201 printf OUTPUT ($y < $#lblist) ? ",\n" : "\n};\n\n";
1202 }
1203
1204 # output the lead byte subtables offsets
1205
1206 my @offsets = (0) x 256;
1207 for (my $x = 0; $x <= $#lblist; $x++) { $offsets[$lblist[$x]] = $x + 1; }
1208 if ($unused)
1209 {
1210 # increment all lead bytes offset to take into account the unused table
1211 for (my $x = 0; $x <= $#lead_bytes; $x++) { $offsets[$lead_bytes[$x]]++; }
1212 }
1213 printf OUTPUT "static const unsigned char cp2uni_leadbytes[256] =\n";
1214 printf OUTPUT "{\n%s\n};\n\n", DUMP_ARRAY( "0x%02x", 0, @offsets );
1215
1216 dump_uni2cp_table( 16, $def );
1217
1218 # output the code page descriptor
1219
1220 printf OUTPUT "const struct dbcs_table DECLSPEC_HIDDEN cptable_%03d =\n{\n", $codepage;
1221 printf OUTPUT " { %d, 2, 0x%04x, 0x%04x, \"%s\" },\n",
1222 $codepage, $def, $defw, $name;
1223 printf OUTPUT " cp2uni,\n";
1224 printf OUTPUT " cp2uni_leadbytes,\n";
1225 printf OUTPUT " uni2cp_low,\n";
1226 printf OUTPUT " uni2cp_high,\n";
1227 printf OUTPUT " {\n %s\n }\n", DUMP_ARRAY( "0x%02x", 0, @lb_ranges, 0, 0 );
1228 printf OUTPUT "};\n";
1229 }
1230
1231
1232 ################################################################
1233 # get the list of defined lead byte ranges
1234 sub get_lb_ranges()
1235 {
1236 my @list = ();
1237 my @ranges = ();
1238 my $i = 0;
1239 foreach $i (@lead_bytes) { $list[$i] = 1; }
1240 my $on = 0;
1241 for (my $i = 0; $i < 256; $i++)
1242 {
1243 if ($on)
1244 {
1245 if (!defined $list[$i]) { push @ranges, $i-1; $on = 0; }
1246 }
1247 else
1248 {
1249 if ($list[$i]) { push @ranges, $i; $on = 1; }
1250 }
1251 }
1252 if ($on) { push @ranges, 0xff; }
1253 return @ranges;
1254 }
1255
1256 ################################################################
1257 # dump the Indic Syllabic Category table
1258 sub dump_indic($)
1259 {
1260 my $filename = shift;
1261 my @indic_table;
1262
1263 my $INPUT = open_data_file( $UNIDATA, "IndicSyllabicCategory.txt" );
1264 while (<$INPUT>)
1265 {
1266 next if /^\#/; # skip comments
1267 next if /^\s*$/; # skip empty lines
1268 next if /\x1a/; # skip ^Z
1269 if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*#/)
1270 {
1271 my $type = $2;
1272 die "unknown indic $type" unless defined $indic_types{$type};
1273 if (hex $1 < 65536)
1274 {
1275 $indic_table[hex $1] = $indic_types{$type};
1276 }
1277 next;
1278 }
1279 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([A-Za-z_]+)\s*#/)
1280 {
1281 my $type = $3;
1282 die "unknown indic $type" unless defined $indic_types{$type};
1283 if (hex $1 < 65536 and hex $2 < 65536)
1284 {
1285 foreach my $i (hex $1 .. hex $2)
1286 {
1287 $indic_table[$i] = $indic_types{$type};
1288 }
1289 }
1290 next;
1291 }
1292 die "malformed line $_";
1293 }
1294 close $INPUT;
1295
1296 $INPUT = open_data_file( $UNIDATA, "IndicPositionalCategory.txt" );
1297 while (<$INPUT>)
1298 {
1299 next if /^\#/; # skip comments
1300 next if /^\s*$/; # skip empty lines
1301 next if /\x1a/; # skip ^Z
1302 if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z]+)\s*#/)
1303 {
1304 my $type = $2;
1305 die "unknown matra $type" unless defined $matra_types{$type};
1306 $indic_table[hex $1] += $matra_types{$type} << 8;
1307 next;
1308 }
1309 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([A-Za-z_]+)\s*#/)
1310 {
1311 my $type = $3;
1312 die "unknown matra $type" unless defined $matra_types{$type};
1313 foreach my $i (hex $1 .. hex $2)
1314 {
1315 $indic_table[$i] += $matra_types{$type} << 8;
1316 }
1317 next;
1318 }
1319 die "malformed line $_";
1320 }
1321 close $INPUT;
1322
1323 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1324 print "Building $filename\n";
1325 print OUTPUT "/* Unicode Indic Syllabic Category */\n";
1326 print OUTPUT "/* generated from $UNIDATA/IndicSyllabicCategory.txt */\n";
1327 print OUTPUT "/* and from $UNIDATA/IndicPositionalCategory.txt */\n";
1328 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1329 print OUTPUT "#include \"windef.h\"\n\n";
1330
1331 dump_two_level_mapping( "indic_syllabic_table", $indic_types{'Other'}, @indic_table );
1332
1333 close OUTPUT;
1334 save_file($filename);
1335 }
1336
1337 ################################################################
1338 # dump the Line Break Properties table
1339 sub dump_linebreak($)
1340 {
1341 my $filename = shift;
1342 my @break_table;
1343 my $next_group = 0;
1344
1345 my $INPUT = open_data_file( $UNIDATA, "LineBreak.txt" );
1346 while (<$INPUT>)
1347 {
1348 next if /^\#/; # skip comments
1349 next if /^\s*$/; # skip empty lines
1350 next if /\x1a/; # skip ^Z
1351 if (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z][0-9A-Z])+\s*/)
1352 {
1353 my $type = $2;
1354 die "unknown breaktype $type" unless defined $break_types{$type};
1355 $break_table[hex $1] = $break_types{$type};
1356 next;
1357 }
1358 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z][0-9A-Z])+\s*/)
1359 {
1360 my $type = $3;
1361 die "unknown breaktype $type" unless defined $break_types{$type};
1362 foreach my $i (hex $1 .. hex $2)
1363 {
1364 $break_table[$i] = $break_types{$type};
1365 }
1366 next;
1367 }
1368 elsif (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z])+\s*/)
1369 {
1370 my $type = $2;
1371 die "unknown breaktype $type" unless defined $break_types{$type};
1372 $break_table[hex $1] = $break_types{$type};
1373 next;
1374 }
1375 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([0-9A-Z][0-9A-Z])+\s*/)
1376 {
1377 my $type = $3;
1378 die "unknown breaktype $type" unless defined $break_types{$type};
1379 foreach my $i (hex $1 .. hex $2)
1380 {
1381 $break_table[$i] = $break_types{$type};
1382 }
1383 next;
1384 }
1385 die "malformed line $_";
1386 }
1387 close $INPUT;
1388
1389 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1390 print "Building $filename\n";
1391 print OUTPUT "/* Unicode Line Break Properties */\n";
1392 print OUTPUT "/* generated from $UNIDATA/LineBreak.txt */\n";
1393 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1394 print OUTPUT "#include \"windef.h\"\n\n";
1395
1396 dump_two_level_mapping( "wine_linebreak_table", $break_types{'XX'}, @break_table );
1397
1398 close OUTPUT;
1399 save_file($filename);
1400 }
1401
1402 my %scripts =
1403 (
1404 "Unknown" => 0,
1405 "Common" => 1,
1406 "Inherited" => 2,
1407 "Arabic" => 3,
1408 "Armenian" => 4,
1409 "Avestan" => 5,
1410 "Balinese" => 6,
1411 "Bamum" => 7,
1412 "Batak" => 8,
1413 "Bengali" => 9,
1414 "Bopomofo" => 10,
1415 "Brahmi" => 11,
1416 "Braille" => 12,
1417 "Buginese" => 13,
1418 "Buhid" => 14,
1419 "Canadian_Aboriginal" => 15,
1420 "Carian" => 16,
1421 "Cham" => 17,
1422 "Cherokee" => 18,
1423 "Coptic" => 19,
1424 "Cuneiform" => 20,
1425 "Cypriot" => 21,
1426 "Cyrillic" => 22,
1427 "Deseret" => 23,
1428 "Devanagari" => 24,
1429 "Egyptian_Hieroglyphs" => 25,
1430 "Ethiopic" => 26,
1431 "Georgian" => 27,
1432 "Glagolitic" => 28,
1433 "Gothic" => 29,
1434 "Greek" => 30,
1435 "Gujarati" => 31,
1436 "Gurmukhi" => 32,
1437 "Han" => 33,
1438 "Hangul" => 34,
1439 "Hanunoo" => 35,
1440 "Hebrew" => 36,
1441 "Hiragana" => 37,
1442 "Imperial_Aramaic" => 38,
1443 "Inscriptional_Pahlavi" => 39,
1444 "Inscriptional_Parthian" => 40,
1445 "Javanese" => 41,
1446 "Kaithi" => 42,
1447 "Kannada" => 43,
1448 "Katakana" => 44,
1449 "Kayah_Li" => 45,
1450 "Kharoshthi" => 46,
1451 "Khmer" => 47,
1452 "Lao" => 48,
1453 "Latin" => 49,
1454 "Lepcha" => 50,
1455 "Limbu" => 51,
1456 "Linear_B" => 52,
1457 "Lisu" => 53,
1458 "Lycian" => 54,
1459 "Lydian" => 55,
1460 "Malayalam" => 56,
1461 "Mandaic" => 57,
1462 "Meetei_Mayek" => 58,
1463 "Mongolian" => 59,
1464 "Myanmar" => 60,
1465 "New_Tai_Lue" => 61,
1466 "Nko" => 62,
1467 "Ogham" => 63,
1468 "Ol_Chiki" => 64,
1469 "Old_Italic" => 65,
1470 "Old_Persian" => 66,
1471 "Old_South_Arabian" => 67,
1472 "Old_Turkic" => 68,
1473 "Oriya" => 69,
1474 "Osmanya" => 70,
1475 "Phags_Pa" => 71,
1476 "Phoenician" => 72,
1477 "Rejang" => 73,
1478 "Runic" => 74,
1479 "Samaritan" => 75,
1480 "Saurashtra" => 76,
1481 "Shavian" => 77,
1482 "Sinhala" => 78,
1483 "Sundanese" => 79,
1484 "Syloti_Nagri" => 80,
1485 "Syriac" => 81,
1486 "Tagalog" => 82,
1487 "Tagbanwa" => 83,
1488 "Tai_Le" => 84,
1489 "Tai_Tham" => 85,
1490 "Tai_Viet" => 86,
1491 "Tamil" => 87,
1492 "Telugu" => 88,
1493 "Thaana" => 89,
1494 "Thai" => 90,
1495 "Tibetan" => 91,
1496 "Tifinagh" => 92,
1497 "Ugaritic" => 93,
1498 "Vai" => 94,
1499 "Yi" => 95,
1500 # Win8/Win8.1
1501 "Chakma" => 96,
1502 "Meroitic_Cursive" => 97,
1503 "Meroitic_Hieroglyphs" => 98,
1504 "Miao" => 99,
1505 "Sharada" => 100,
1506 "Sora_Sompeng" => 101,
1507 "Takri" => 102,
1508 # Win10
1509 "Bassa_Vah" => 103,
1510 "Caucasian_Albanian" => 104,
1511 "Duployan" => 105,
1512 "Elbasan" => 106,
1513 "Grantha" => 107,
1514 "Khojki" => 108,
1515 "Khudawadi" => 109,
1516 "Linear_A" => 110,
1517 "Mahajani" => 111,
1518 "Manichaean" => 112,
1519 "Mende_Kikakui" => 113,
1520 "Modi" => 114,
1521 "Mro" => 115,
1522 "Nabataean" => 116,
1523 "Old_North_Arabian" => 117,
1524 "Old_Permic" => 118,
1525 "Pahawh_Hmong" => 119,
1526 "Palmyrene" => 120,
1527 "Pau_Cin_Hau" => 121,
1528 "Psalter_Pahlavi" => 122,
1529 "Siddham" => 123,
1530 "Tirhuta" => 124,
1531 "Warang_Citi" => 125,
1532 # Win10 RS1
1533 "Adlam" => 126,
1534 "Ahom" => 127,
1535 "Anatolian_Hieroglyphs" => 128,
1536 "Bhaiksuki" => 129,
1537 "Hatran" => 130,
1538 "Marchen" => 131,
1539 "Multani" => 132,
1540 "Newa" => 133,
1541 "Old_Hungarian" => 134,
1542 "Osage" => 135,
1543 "SignWriting" => 136,
1544 "Tangut" => 137,
1545 # Win10 RS4
1546 "Masaram_Gondi" => 138,
1547 "Nushu" => 139,
1548 "Soyombo" => 140,
1549 "Zanabazar_Square" => 141,
1550 );
1551
1552 ################################################################
1553 # dump Script IDs table
1554 sub dump_scripts($)
1555 {
1556 my $filename = shift;
1557 my $header = $filename;
1558 my @scripts_table;
1559 my $script_index;
1560 my $i;
1561
1562 my $INPUT = open_data_file( $UNIDATA, "Scripts.txt" );
1563 # Fill the table
1564 # Unknown script id is always 0, so undefined scripts are automatically treated as such
1565 while (<$INPUT>)
1566 {
1567 my $type = "";
1568
1569 next if /^\#/; # skip comments
1570 next if /^\s*$/; # skip empty lines
1571 next if /\x1a/; # skip ^Z
1572 if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
1573 {
1574 $type = $2;
1575 if (defined $scripts{$type})
1576 {
1577 $scripts_table[hex $1] = $scripts{$type};
1578 }
1579 next;
1580 }
1581 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
1582 {
1583 $type = $3;
1584 if (defined $scripts{$type})
1585 {
1586 foreach my $i (hex $1 .. hex $2)
1587 {
1588 $scripts_table[$i] = $scripts{$type};
1589 }
1590 }
1591 next;
1592 }
1593 }
1594
1595 close $INPUT;
1596
1597 $header = "$filename.h";
1598 open OUTPUT,">$header.new" or die "Cannot create $header";
1599 print "Building $header\n";
1600 print OUTPUT "/* Unicode Script IDs */\n";
1601 print OUTPUT "/* generated from $UNIDATA/Scripts.txt */\n";
1602 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1603
1604 print OUTPUT "enum unicode_script_id {\n";
1605 foreach my $script (sort { $scripts{$a} <=> $scripts{$b} } keys %scripts)
1606 {
1607 print OUTPUT " Script_$script = $scripts{$script},\n";
1608 }
1609 print OUTPUT " Script_LastId = ", (scalar keys %scripts) - 1, "\n";
1610 print OUTPUT "};\n";
1611
1612 close OUTPUT;
1613 save_file($header);
1614
1615 $filename = "$filename.c";
1616 open OUTPUT,">$filename.new" or die "Cannot create $header";
1617 print "Building $filename\n";
1618 print OUTPUT "/* Unicode Script IDs */\n";
1619 print OUTPUT "/* generated from $UNIDATA/Scripts.txt */\n";
1620 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1621 print OUTPUT "#include \"windef.h\"\n\n";
1622
1623 dump_two_level_mapping( "wine_scripts_table", 0, @scripts_table );
1624 close OUTPUT;
1625 save_file($filename);
1626 }
1627
1628 ################################################################
1629 # dump the BiDi mirroring table
1630 sub dump_mirroring($)
1631 {
1632 my $filename = shift;
1633 my @mirror_table = ();
1634
1635 my $INPUT = open_data_file( $UNIDATA, "BidiMirroring.txt" );
1636 while (<$INPUT>)
1637 {
1638 next if /^\#/; # skip comments
1639 next if /^$/; # skip empty lines
1640 next if /\x1a/; # skip ^Z
1641 if (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9a-fA-F]+)/)
1642 {
1643 $mirror_table[hex $1] = hex $2;
1644 next;
1645 }
1646 die "malformed line $_";
1647 }
1648 close $INPUT;
1649
1650 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1651 print "Building $filename\n";
1652 print OUTPUT "/* Unicode BiDi mirroring */\n";
1653 print OUTPUT "/* generated from $UNIDATA/BidiMirroring.txt */\n";
1654 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1655 print OUTPUT "#include \"windef.h\"\n\n";
1656 dump_case_table( "DECLSPEC_HIDDEN wine_mirror_map", @mirror_table );
1657 close OUTPUT;
1658 save_file($filename);
1659 }
1660
1661 ################################################################
1662 # dump the Bidi Brackets
1663 sub dump_bracket($)
1664 {
1665 my $filename = shift;
1666 my @bracket_table;
1667
1668 my $INPUT = open_data_file( $UNIDATA, "BidiBrackets.txt" );
1669 while (<$INPUT>)
1670 {
1671 next if /^\#/; # skip comments
1672 next if /^\s*$/; # skip empty lines
1673 next if /\x1a/; # skip ^Z
1674 if (/^\s*([0-9a-fA-F]+)\s*;\s*([0-9a-fA-F]+);\s*([con])/)
1675 {
1676 my $type = $3;
1677 die "unknown bracket $type" unless defined $bracket_types{$type};
1678 die "characters too distant $1 and $2" if abs(hex($2) - hex($1)) >= 128;
1679 $bracket_table[hex $1] = (hex($2) - hex($1)) % 255;
1680 $bracket_table[hex $1] += $bracket_types{$type} << 8;
1681 next;
1682 }
1683 die "malformed line $_";
1684 }
1685 close $INPUT;
1686
1687 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1688 print "Building $filename\n";
1689 print OUTPUT "/* Unicode Bidirectional Bracket table */\n";
1690 print OUTPUT "/* generated from $UNIDATA/BidiBrackets.txt */\n";
1691 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1692 print OUTPUT "#include \"windef.h\"\n\n";
1693
1694 dump_two_level_mapping( "bidi_bracket_table", 0, @bracket_table );
1695
1696 close OUTPUT;
1697 save_file($filename);
1698 }
1699
1700 ################################################################
1701 # dump the Arabic shaping table
1702 sub dump_shaping($)
1703 {
1704 my $filename = shift;
1705 my %groups;
1706 my $next_group = 0;
1707
1708 $groups{"No_Joining_Group"} = $next_group++;
1709
1710 my $INPUT = open_data_file( $UNIDATA, "ArabicShaping.txt" );
1711 while (<$INPUT>)
1712 {
1713 next if /^\#/; # skip comments
1714 next if /^\s*$/; # skip empty lines
1715 next if /\x1a/; # skip ^Z
1716 if (/^\s*([0-9a-fA-F]+)\s*;.*;\s*([RLDCUT])\s*;\s*(\w+)/)
1717 {
1718 my $type = $2;
1719 my $group = $3;
1720 $groups{$group} = $next_group++ unless defined $groups{$group};
1721 $joining_table[hex $1] = $joining_types{$type} | ($groups{$group} << 8);
1722 next;
1723 }
1724 die "malformed line $_";
1725 }
1726 close $INPUT;
1727
1728 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1729 print "Building $filename\n";
1730 print OUTPUT "/* Unicode Arabic shaping */\n";
1731 print OUTPUT "/* generated from $UNIDATA/ArabicShaping.txt */\n";
1732 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1733 print OUTPUT "#include \"windef.h\"\n\n";
1734
1735 dump_two_level_mapping( "wine_shaping_table", 0, @joining_table );
1736
1737 print OUTPUT "\nconst unsigned short DECLSPEC_HIDDEN wine_shaping_forms[256][4] =\n{\n";
1738 for (my $i = 0x600; $i <= 0x6ff; $i++)
1739 {
1740 printf OUTPUT " { 0x%04x, 0x%04x, 0x%04x, 0x%04x },\n",
1741 ${joining_forms{"isolated"}}[$i] || $i,
1742 ${joining_forms{"final"}}[$i] || $i,
1743 ${joining_forms{"initial"}}[$i] || $i,
1744 ${joining_forms{"medial"}}[$i] || $i;
1745 }
1746 print OUTPUT "};\n";
1747
1748 close OUTPUT;
1749 save_file($filename);
1750 }
1751
1752 ################################################################
1753 # dump the Vertical Orientation table
1754 sub dump_vertical($)
1755 {
1756 my $filename = shift;
1757 my @vertical_table;
1758
1759 my $INPUT = open_data_file( $UNIDATA, "VerticalOrientation.txt" );
1760 while (<$INPUT>)
1761 {
1762 next if /^\#/; # skip comments
1763 next if /^\s*$/; # skip empty lines
1764 next if /\x1a/; # skip ^Z
1765 if (/^\s*([0-9a-fA-F]+)\s*;\s*([a-zA-Z_]+)\s*/)
1766 {
1767 my $type = $2;
1768 die "unknown vertical $type" unless defined $vertical_types{$type};
1769 if (hex $1 < 65536)
1770 {
1771 $vertical_table[hex $1] = $vertical_types{$type};
1772 }
1773 next;
1774 }
1775 elsif (/^\s*([0-9a-fA-F]+)..\s*([0-9a-fA-F]+)\s*;\s*([A-Za-z_]+)\s*/)
1776 {
1777 my $type = $3;
1778 die "unknown vertical $type" unless defined $vertical_types{$type};
1779 foreach my $i (hex $1 .. hex $2)
1780 {
1781 $vertical_table[$i] = $vertical_types{$type};
1782 }
1783 next;
1784 }
1785 die "malformed line $_";
1786 }
1787 close $INPUT;
1788
1789 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1790 print "Building $filename\n";
1791 print OUTPUT "/* Unicode Vertical Orientation */\n";
1792 print OUTPUT "/* generated from $UNIDATA/VerticalOrientation.txt */\n";
1793 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1794 print OUTPUT "#include \"windef.h\"\n\n";
1795
1796 dump_two_level_mapping( "vertical_orientation_table", $vertical_types{'R'}, @vertical_table );
1797
1798 close OUTPUT;
1799 save_file($filename);
1800 }
1801
1802 ################################################################
1803 # dump the digit folding tables
1804 sub dump_digit_folding($)
1805 {
1806 my $filename = shift;
1807 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1808 print "Building $filename\n";
1809 print OUTPUT "/* Unicode digit folding mappings */\n";
1810 print OUTPUT "/* generated from $UNIDATA/UnicodeData.txt */\n";
1811 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1812 print OUTPUT "#include \"windef.h\"\n\n";
1813
1814 dump_case_table( "DECLSPEC_HIDDEN wine_digitmap", @digitmap_table );
1815 print OUTPUT "\n";
1816 dump_case_table( "DECLSPEC_HIDDEN wine_compatmap", @compatmap_table );
1817 close OUTPUT;
1818 save_file($filename);
1819 }
1820
1821
1822 ################################################################
1823 # dump the case mapping tables
1824 sub dump_case_mappings($)
1825 {
1826 my $filename = shift;
1827 open OUTPUT,">$filename.new" or die "Cannot create $filename";
1828 print "Building $filename\n";
1829 print OUTPUT "/* Unicode case mappings */\n";
1830 print OUTPUT "/* generated from $UNIDATA/UnicodeData.txt */\n";
1831 print OUTPUT "/* DO NOT EDIT!! */\n\n";
1832 print OUTPUT "#include \"windef.h\"\n\n";
1833
1834 dump_case_table( "wine_casemap_lower", @tolower_table );
1835 print OUTPUT "\n";
1836 dump_case_table( "wine_casemap_upper", @toupper_table );
1837 close OUTPUT;
1838 save_file($filename);
1839 }
1840
1841
1842 ################################################################
1843 # dump a case mapping table
1844 sub dump_case_table($@)
1845 {
1846 my ($name,@table) = @_;
1847
1848 # count the number of sub tables that contain something
1849 # also compute the low and upper populated bounds
1850
1851 my @lowerbounds = ( 0, 0 );
1852 my @upperbounds = ( 0, 255 );
1853 my $index = 0;
1854 my @filled = ();
1855 for (my $i = 0; $i < 65536; $i++)
1856 {
1857 next unless defined $table[$i];
1858 if (!defined $filled[$i >> 8])
1859 {
1860 $lowerbounds[$index] = $i & 0xff;
1861 $upperbounds[$index] = 0xff - $lowerbounds[$index];
1862 $filled[$i >> 8] = $index * 256 + 512;
1863 $index++;
1864 }
1865 else
1866 {
1867 $upperbounds[$index-1] = 0xff - ($i & 0xff);
1868 }
1869 $table[$i] = ($table[$i] - $i) & 0xffff;
1870 }
1871
1872 # Collapse blocks upwards if possible
1873 my $removed = 0;
1874 $index = 0;
1875 for (my $i = 0; $i < 256; $i++)
1876 {
1877 next unless defined $filled[$i];
1878 if ($upperbounds[$index - 1] > $lowerbounds[$index])
1879 {
1880 $removed = $removed + $lowerbounds[$index];
1881 }
1882 else
1883 {
1884 $removed = $removed + $upperbounds[$index - 1];
1885 $lowerbounds[$index] = $upperbounds[$index - 1];
1886 }
1887 $filled[$i] = $filled[$i] - $removed;
1888 $index++;
1889 }
1890
1891 # dump the table
1892
1893 printf OUTPUT "const WCHAR %s[%d] =\n", $name, $index * 256 + 512 - $removed;
1894 printf OUTPUT "{\n /* index */\n";
1895 printf OUTPUT "%s,\n", DUMP_ARRAY( "0x%04x", 256, @filled );
1896 printf OUTPUT " /* defaults */\n";
1897 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, (0) x 256 );
1898 $index = 0;
1899 for (my $i = 0; $i < 256; $i++)
1900 {
1901 next unless $filled[$i];
1902 printf OUTPUT ",\n /* 0x%02x%02x .. 0x%02xff */\n", $i, $lowerbounds[$index], $i;
1903 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0,
1904 @table[($i<<8) + $lowerbounds[$index] .. ($i<<8)+255] );
1905 $index++;
1906 }
1907 printf OUTPUT "\n};\n";
1908 }
1909
1910 ################################################################
1911 # compress a mapping table by removing identical rows
1912 sub compress_array($$@)
1913 {
1914 my $rows = shift;
1915 my $def = shift;
1916 my @table = @_;
1917 my $len = @table / $rows;
1918 my @array = (0) x $rows;
1919 my %sequences;
1920
1921 # try to merge table rows
1922 for (my $row = 0; $row < $rows; $row++)
1923 {
1924 my @table_row = map { defined($_) ? $_ : $def; } @table[($row * $len)..($row * $len + $len - 1)];
1925 my $rowtxt = pack "S*", @table_row;
1926 if (defined($sequences{$rowtxt}))
1927 {
1928 # reuse an existing row
1929 $array[$row] = $sequences{$rowtxt};
1930 }
1931 else
1932 {
1933 # create a new row
1934 $sequences{$rowtxt} = $array[$row] = $#array + 1;
1935 push @array, @table_row;
1936 }
1937 }
1938 return @array;
1939 }
1940
1941 ################################################################
1942 # dump a simple char -> 16-bit value mapping table
1943 sub dump_simple_mapping($$@)
1944 {
1945 my $name = shift;
1946 my $def = shift;
1947 my @array = compress_array( 256, $def, @_[0..65535] );
1948
1949 printf OUTPUT "const unsigned short %s[%d] =\n{\n", $name, $#array+1;
1950 printf OUTPUT " /* offsets */\n%s,\n", DUMP_ARRAY( "0x%04x", 0, @array[0..255] );
1951 printf OUTPUT " /* values */\n%s\n};\n", DUMP_ARRAY( "0x%04x", 0, @array[256..$#array] );
1952 }
1953
1954 ################################################################
1955 # dump a char -> 16-bit value mapping table using two-level tables
1956 sub dump_two_level_mapping($$@)
1957 {
1958 my $name = shift;
1959 my $def = shift;
1960 my @row_array = compress_array( 4096, $def, @_[0..65535] );
1961 my @array = compress_array( 256, 0, @row_array[0..4095] );
1962
1963 for (my $i = 256; $i < @array; $i++) { $array[$i] += @array - 4096; }
1964
1965 printf OUTPUT "const unsigned short DECLSPEC_HIDDEN %s[%d] =\n{\n", $name, @array + @row_array - 4096;
1966 printf OUTPUT " /* level 1 offsets */\n%s,\n", DUMP_ARRAY( "0x%04x", 0, @array[0..255] );
1967 printf OUTPUT " /* level 2 offsets */\n%s,\n", DUMP_ARRAY( "0x%04x", 0, @array[256..$#array] );
1968 printf OUTPUT " /* values */\n%s\n};\n", DUMP_ARRAY( "0x%04x", 0, @row_array[4096..$#row_array] );
1969 }
1970
1971 ################################################################
1972 # dump a binary case mapping table in l_intl.nls format
1973 sub dump_binary_case_table(@)
1974 {
1975 my (@table) = @_;
1976
1977 my %difftables_hash = ();
1978 my @difftables;
1979 my %offtables2_hash = ();
1980 my @offtables2 = ();
1981
1982 my @offtable = ();
1983 for (my $i = 0; $i < 256; $i++)
1984 {
1985 my @offtable2 = ();
1986 for(my $j = 0; $j < 16; $j++) # offset table for xx00-xxFF characters
1987 {
1988 my @difftable;
1989 for (my $k = 0; $k < 16; $k++) # case map table for xxx0-xxxF characters
1990 {
1991 my $char = ($i<<8) + ($j<<4) + $k;
1992 $difftable[$k] = (defined $table[$char]) ? (($table[$char]-$char) & 0xffff) : 0;
1993 }
1994
1995 my $diff_key = pack "S*", @difftable;
1996 my $offset3 = $difftables_hash{$diff_key};
1997 if (!defined $offset3)
1998 {
1999 $offset3 = scalar @difftables;
2000 $difftables_hash{$diff_key} = $offset3;
2001 push @difftables, @difftable;
2002 }
2003 $offtable2[$j] = $offset3;
2004 }
2005
2006 my $offtable2_key = pack "S*", @offtable2;
2007 my $offset2 = $offtables2_hash{$offtable2_key};
2008 if (!defined $offset2)
2009 {
2010 $offset2 = scalar @offtables2;
2011 $offtables2_hash{$offtable2_key} = $offset2;
2012 push @offtables2, \@offtable2;
2013 }
2014 $offtable[$i] = $offset2;
2015 }
2016
2017 my @output;
2018 my $offset = 0x100; # offset of first subtable in words
2019 foreach (@offtable)
2020 {
2021 push @output, 0x10 * $_ + $offset; # offset of subtable in words
2022 }
2023
2024 $offset = 0x100 + 0x10 * scalar @offtables2; # offset of first difftable in words
2025 foreach(@offtables2)
2026 {
2027 my $table = $_;
2028 foreach(@$table)
2029 {
2030 push @output, $_ + $offset; # offset of difftable in words
2031 }
2032 }
2033
2034 my $len = 1 + scalar @output + scalar @difftables;
2035 return pack "S<*", $len, @output, @difftables;
2036 }
2037
2038
2039 ################################################################
2040 # dump case mappings for l_intl.nls
2041 sub dump_intl_nls($)
2042 {
2043 my $filename = shift;
2044 open OUTPUT,">$filename.new" or die "Cannot create $filename";
2045 printf "Building $filename\n";
2046
2047 binmode OUTPUT;
2048 print OUTPUT pack "S<", 1; # version
2049 print OUTPUT dump_binary_case_table( @toupper_table );
2050 print OUTPUT dump_binary_case_table( @tolower_table );
2051 close OUTPUT;
2052 save_file($filename);
2053 }
2054
2055
2056 sub load_nameprep_range_table($$$)
2057 {
2058 my ($INPUT, $val, $table_ref) = @_;
2059
2060 while (<$INPUT>)
2061 {
2062 if (/^\s*([0-9a-fA-F]+)-([0-9a-fA-F]+)/)
2063 {
2064 my $last = hex $2;
2065 $last = 65535 if($last >= 65536);
2066 foreach my $i (hex $1 .. $last)
2067 {
2068 $table_ref->[$i] |= $val;
2069 }
2070 next;
2071 }
2072 elsif (/^\s*([0-9a-fA-F]+)/)
2073 {
2074 if (hex $1 < 65536)
2075 {
2076 $table_ref->[hex $1] |= $val;
2077 }
2078 next;
2079 }
2080
2081 return if (/End\sTable/);
2082 }
2083 }
2084
2085 sub load_nameprep_map_table($$)
2086 {
2087 my ($INPUT, $table_ref) = @_;
2088
2089 while (<$INPUT>)
2090 {
2091 if (/^\s*([0-9a-fA-F]+);\s;/)
2092 {
2093 # special value for map to nothing
2094 $table_ref->[hex $1] = [0xffff, 0xffff, 0xffff];
2095 next;
2096 }
2097 elsif (/^\s*([0-9a-fA-F]+);\s([0-9a-fA-F]+);/)
2098 {
2099 $table_ref->[hex $1] = [hex $2, 0, 0];
2100 next;
2101 }
2102 elsif (/^\s*([0-9a-fA-F]+);\s([0-9a-fA-F]+)\s([0-9a-fA-F]+);/)
2103 {
2104 $table_ref->[hex $1] = [hex $2, hex $3, 0];
2105 next;
2106 }
2107 elsif (/^\s*([0-9a-fA-F]+);\s([0-9a-fA-F]+)\s([0-9a-fA-F]+)\s([0-9a-fA-F]+);/)
2108 {
2109 $table_ref->[hex $1] = [hex $2, hex $3, hex $4];
2110 next;
2111 }
2112
2113 return if (/End\sTable/);
2114 }
2115 }
2116
2117 ################################################################
2118 # dump mapping table, prohibited characters set, unassigned
2119 # characters, bidirectional rules used by nameprep algorithm
2120 sub dump_nameprep($)
2121 {
2122 my $filename = shift;
2123 my @mapping_table = ();
2124 my @flags_table;
2125
2126 my $INPUT = open_data_file( $RFCS, $STRINGPREP );
2127 while (<$INPUT>)
2128 {
2129 next unless /Start\sTable/;
2130
2131 load_nameprep_range_table($INPUT, $nameprep_flags{"unassigned"}, \@flags_table) if (/A.1/);
2132 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.1.2/);
2133 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.2.2/);
2134 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.3/);
2135 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.4/);
2136 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.5/);
2137 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.6/);
2138 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.7/);
2139 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.8/);
2140 load_nameprep_range_table($INPUT, $nameprep_flags{"prohibited"}, \@flags_table) if (/C.9/);
2141 load_nameprep_range_table($INPUT, $nameprep_flags{"bidi_ral"}, \@flags_table) if (/D.1/);
2142 load_nameprep_range_table($INPUT, $nameprep_flags{"bidi_l"}, \@flags_table) if (/D.2/);
2143
2144 load_nameprep_map_table($INPUT, \@mapping_table) if (/B.1/);
2145 load_nameprep_map_table($INPUT, \@mapping_table) if (/B.2/);
2146 }
2147 close $INPUT;
2148
2149 open OUTPUT,">$filename.new" or die "Cannot create $filename";
2150 print "Building $filename\n";
2151 print OUTPUT "/* Nameprep algorithm related data */\n";
2152 print OUTPUT "/* generated from $RFCS/$STRINGPREP */\n";
2153 print OUTPUT "/* DO NOT EDIT!! */\n\n";
2154 print OUTPUT "#include \"windef.h\"\n\n";
2155
2156 dump_two_level_mapping( "nameprep_char_type", 0, @flags_table );
2157
2158 ######### mapping table
2159 # first determine all the 16-char subsets that contain something
2160 my @filled = ();
2161 my $pos = 16*3; # for the null subset
2162 for (my $i = 0; $i < 65536; $i++)
2163 {
2164 next unless defined $mapping_table[$i];
2165 $filled[$i >> 4] = $pos;
2166 $pos += 16*3;
2167 $i |= 15;
2168 }
2169 my $total = $pos;
2170
2171 # now count the 256-char subsets that contain something
2172 my @filled_idx = (256) x 256;
2173 $pos = 256 + 16;
2174 for (my $i = 0; $i < 4096; $i++)
2175 {
2176 next unless $filled[$i];
2177 $filled_idx[$i >> 4] = $pos;
2178 $pos += 16;
2179 $i |= 15;
2180 }
2181 my $null_offset = $pos;
2182 $total += $pos;
2183
2184 # add the index offsets to the subsets positions
2185 for (my $i = 0; $i < 4096; $i++)
2186 {
2187 next unless $filled[$i];
2188 $filled[$i] += $null_offset;
2189 }
2190
2191 # dump the main index
2192 printf OUTPUT "const unsigned short DECLSPEC_HIDDEN nameprep_mapping[%d] =\n", $total;
2193 printf OUTPUT "{\n /* index */\n";
2194 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @filled_idx );
2195 printf OUTPUT ",\n /* null sub-index */\n%s", DUMP_ARRAY( "0x%04x", 0, ($null_offset) x 16 );
2196
2197 # dump the second-level indexes
2198 for (my $i = 0; $i < 256; $i++)
2199 {
2200 next unless ($filled_idx[$i] > 256);
2201 my @table = @filled[($i<<4)..($i<<4)+15];
2202 for (my $j = 0; $j < 16; $j++) { $table[$j] ||= $null_offset; }
2203 printf OUTPUT ",\n /* sub-index %02x */\n", $i;
2204 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
2205 }
2206
2207 # dump the 16-char subsets
2208 printf OUTPUT ",\n /* null mapping */\n";
2209 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, (0) x 48 );
2210
2211 for (my $i = 0; $i < 4096; $i++)
2212 {
2213 next unless $filled[$i];
2214 my @table = (0) x 48;
2215 for (my $j = 0; $j < 16; $j++)
2216 {
2217 if (defined $mapping_table[($i<<4) + $j])
2218 {
2219 $table[3 * $j] = ${$mapping_table[($i << 4) + $j]}[0];
2220 $table[3 * $j + 1] = ${$mapping_table[($i << 4) + $j]}[1];
2221 $table[3 * $j + 2] = ${$mapping_table[($i << 4) + $j]}[2];
2222 }
2223 }
2224 printf OUTPUT ",\n /* 0x%03x0 .. 0x%03xf */\n", $i, $i;
2225 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
2226 }
2227
2228 printf OUTPUT "\n};\n";
2229
2230 close OUTPUT;
2231 save_file($filename);
2232 }
2233
2234
2235 ################################################################
2236 # dump the GetStringTypeW table
2237 sub dump_string_type_table($)
2238 {
2239 my $filename = shift;
2240 open OUTPUT,">$filename.new" or die "Cannot create $filename";
2241 printf "Building $filename\n";
2242 printf OUTPUT "/* Unicode wctype table */\n";
2243 printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
2244 printf OUTPUT "#include \"windef.h\"\n\n";
2245
2246 my @table = @category_table;
2247
2248 # add the direction in the high 4 bits of the category
2249 for (my $i = 0; $i < 65536; $i++)
2250 {
2251 $table[$i] |= $c2_types{$direction_table[$i]} << 12 if defined $direction_table[$i];
2252 }
2253
2254 dump_two_level_mapping( "wctype_table", 0, @table );
2255
2256 close OUTPUT;
2257 save_file($filename);
2258 }
2259
2260
2261 ################################################################
2262 # dump the bidi direction table
2263 sub dump_bidi_dir_table($)
2264 {
2265 my $filename = shift;
2266 open OUTPUT,">$filename.new" or die "Cannot create $filename";
2267 printf "Building $filename\n";
2268 printf OUTPUT "/* Unicode BiDi direction table */\n";
2269 printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
2270 printf OUTPUT "#include \"windef.h\"\n\n";
2271
2272 my @table;
2273
2274 for (my $i = 0; $i < 65536; $i++)
2275 {
2276 $table[$i] = $bidi_types{$direction_table[$i]} if defined $direction_table[$i];
2277 }
2278
2279 dump_two_level_mapping( "bidi_direction_table", $bidi_types{"L"}, @table );
2280
2281 close OUTPUT;
2282 save_file($filename);
2283 }
2284
2285
2286 ################################################################
2287 # dump the ctype tables
2288 sub dump_ctype_tables($)
2289 {
2290 my $filename = shift;
2291 open OUTPUT,">$filename.new" or die "Cannot create $filename";
2292 printf "Building $filename\n";
2293 printf OUTPUT "/* Unicode ctype tables */\n";
2294 printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
2295 printf OUTPUT "#include \"windef.h\"\n\n";
2296
2297 my @table = @category_table;
2298
2299 # add the direction in the high 4 bits of the category
2300 for (my $i = 0; $i < 65536; $i++)
2301 {
2302 $table[$i] |= $directions{$direction_table[$i]} << 12 if defined $direction_table[$i];
2303 }
2304
2305 dump_simple_mapping( "wine_wctype_table", 0, @table );
2306
2307 close OUTPUT;
2308 save_file($filename);
2309 }
2310
2311
2312 ################################################################
2313 # dump the char composition table
2314 sub dump_compose_table($)
2315 {
2316 my $filename = shift;
2317
2318 open OUTPUT,">$filename.new" or die "Cannot create $filename";
2319 print "Building $filename\n";
2320 print OUTPUT "/* Unicode char composition */\n";
2321 print OUTPUT "/* generated from $UNIDATA/UnicodeData.txt */\n";
2322 print OUTPUT "/* DO NOT EDIT!! */\n\n";
2323 print OUTPUT "#include \"windef.h\"\n\n";
2324
2325 my @filled = ();
2326 foreach my $i (@compose_table)
2327 {
2328 my @comp = @$i;
2329 push @{$filled[$comp[1]]}, [ $comp[0], $comp[2] ];
2330 }
2331
2332 # count how many different second chars we have
2333
2334 my $count = 0;
2335 for (my $i = 0; $i < 65536; $i++)
2336 {
2337 next unless defined $filled[$i];
2338 $count++;
2339 }
2340
2341 # build the table of second chars and offsets
2342
2343 my $pos = $count + 1;
2344 my @table = ();
2345 for (my $i = 0; $i < 65536; $i++)
2346 {
2347 next unless defined $filled[$i];
2348 push @table, $i, $pos;
2349 $pos += @{$filled[$i]};
2350 }
2351 # terminator with last position
2352 push @table, 0, $pos;
2353 printf OUTPUT "static const WCHAR table[0x%x] =\n{\n", 2*$pos;
2354 printf OUTPUT " /* second chars + offsets */\n%s", DUMP_ARRAY( "0x%04x", 0, @table );
2355
2356 # build the table of first chars and mappings
2357
2358 for (my $i = 0; $i < 65536; $i++)
2359 {
2360 next unless defined $filled[$i];
2361 my @table = ();
2362 my @list = sort { $a->[0] <=> $b->[0] } @{$filled[$i]};
2363 for (my $j = 0; $j <= $#list; $j++)
2364 {
2365 push @table, $list[$j][0], $list[$j][1];
2366 }
2367 printf OUTPUT ",\n /* 0x%04x */\n%s", $i, DUMP_ARRAY( "0x%04x", 0, @table );
2368 }
2369 print OUTPUT "\n};\n\n";
2370 print OUTPUT <<"EOF";
2371 static inline int binary_search( WCHAR ch, int low, int high )
2372 {
2373 while (low <= high)
2374 {
2375 int pos = (low + high) / 2;
2376 if (table[2 * pos] < ch) low = pos + 1;
2377 else if (table[2 * pos] > ch) high = pos - 1;
2378 else return pos;
2379 }
2380 return -1;
2381 }
2382
2383 WCHAR DECLSPEC_HIDDEN wine_compose( const WCHAR *str )
2384 {
2385 int pos, idx = 1, start = 0, count = $count;
2386 for (;;)
2387 {
2388 if ((pos = binary_search( str[idx], start, count - 1 )) == -1) return 0;
2389 if (!idx--) return table[2 * pos + 1];
2390 start = table[2 * pos + 1];
2391 count = table[2 * pos + 3];
2392 }
2393 }
2394 EOF
2395 close OUTPUT;
2396 save_file($filename);
2397 }
2398
2399 sub dump_full_compat_table()
2400 {
2401 # first determine all the 16-char subsets that contain something
2402
2403 my @filled = (0) x 4096;
2404 my $pos = 16; # for the null subset
2405 my $data_total = 0;
2406 for (my $i = 0; $i < 65536; $i++)
2407 {
2408 next unless defined $full_compatmap_table[$i];
2409 if ($filled[$i >> 4] == 0)
2410 {
2411 $filled[$i >> 4] = $pos;
2412 $pos += 16;
2413 }
2414 $data_total += $#{$full_compatmap_table[$i]} + 1;
2415 }
2416 my $total = $pos;
2417
2418 # now count the 256-char subsets that contain something
2419
2420 my @filled_idx = (256) x 256;
2421 $pos = 256 + 16;
2422 for (my $i = 0; $i < 4096; $i++)
2423 {
2424 next unless $filled[$i];
2425 $filled_idx[$i >> 4] = $pos;
2426 $pos += 16;
2427 $i |= 15;
2428 }
2429 my $null_offset = $pos; # null mapping
2430 $total += $pos + 1; # add the offset sentinel
2431
2432 # add the index offsets to the subsets positions
2433
2434 for (my $i = 0; $i < 4096; $i++)
2435 {
2436 next unless $filled[$i];
2437 $filled[$i] += $null_offset;
2438 }
2439
2440 # dump the main index
2441
2442 printf OUTPUT "static const WCHAR compatmap_table[%d] =\n", $total + $data_total;
2443 printf OUTPUT "{\n /* index */\n";
2444 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @filled_idx );
2445 printf OUTPUT ",\n /* null sub-index */\n%s", DUMP_ARRAY( "0x%04x", 0, ($null_offset) x 16 );
2446
2447 # dump the second-level indexes
2448
2449 for (my $i = 0; $i < 256; $i++)
2450 {
2451 next unless ($filled_idx[$i] > 256);
2452 my @table = @filled[($i<<4)..($i<<4)+15];
2453 for (my $j = 0; $j < 16; $j++) { $table[$j] ||= $null_offset; }
2454 printf OUTPUT ",\n /* sub-index %02x */\n", $i;
2455 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
2456 }
2457
2458 # dump the 16-char offsets
2459
2460 printf OUTPUT ",\n /* null offsets */\n";
2461 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, ($total) x (16) );
2462
2463 $pos = $total;
2464
2465 my @data;
2466 for (my $i = 0; $i < 4096; $i++)
2467 {
2468 next unless $filled[$i];
2469 my @table = (0) x (16);
2470 for (my $j = 0; $j < 16; $j++)
2471 {
2472 $table[$j] = $pos;
2473 if (defined $full_compatmap_table[($i<<4) + $j])
2474 {
2475 $pos += $#{$full_compatmap_table[($i<<4) + $j]} + 1;
2476 push @data, @{$full_compatmap_table[($i<<4) + $j]};
2477 }
2478 }
2479 printf OUTPUT ",\n /* offsets 0x%03x0 .. 0x%03xf */\n", $i, $i;
2480 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
2481 }
2482
2483 my @sentinel = $pos;
2484 printf OUTPUT ",\n /* offset sentinel */\n";
2485 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @sentinel );
2486
2487 printf OUTPUT ",\n /* data */\n";
2488 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @data );
2489
2490 printf OUTPUT "\n};\n\n";
2491 }
2492
2493 ################################################################
2494 # dump the char decomposition table
2495 sub dump_decompose_table($)
2496 {
2497 my $filename = shift;
2498
2499 open OUTPUT,">$filename.new" or die "Cannot create $filename";
2500 print "Building $filename\n";
2501 print OUTPUT "/* Unicode char composition */\n";
2502 print OUTPUT "/* generated from $UNIDATA/UnicodeData.txt */\n";
2503 print OUTPUT "/* DO NOT EDIT!! */\n\n";
2504 print OUTPUT "#include \"wine/unicode.h\"\n\n";
2505
2506 # first determine all the 16-char subsets that contain something
2507
2508 my @filled = (0) x 4096;
2509 my $pos = 16*2; # for the null subset
2510 for (my $i = 0; $i < 65536; $i++)
2511 {
2512 next unless defined $decomp_table[$i];
2513 $filled[$i >> 4] = $pos;
2514 $pos += 16*2;
2515 $i |= 15;
2516 }
2517 my $total = $pos;
2518
2519 # now count the 256-char subsets that contain something
2520
2521 my @filled_idx = (256) x 256;
2522 $pos = 256 + 16;
2523 for (my $i = 0; $i < 4096; $i++)
2524 {
2525 next unless $filled[$i];
2526 $filled_idx[$i >> 4] = $pos;
2527 $pos += 16;
2528 $i |= 15;
2529 }
2530 my $null_offset = $pos; # null mapping
2531 $total += $pos;
2532
2533 # add the index offsets to the subsets positions
2534
2535 for (my $i = 0; $i < 4096; $i++)
2536 {
2537 next unless $filled[$i];
2538 $filled[$i] += $null_offset;
2539 }
2540
2541 # dump the main index
2542
2543 printf OUTPUT "static const WCHAR table[%d] =\n", $total;
2544 printf OUTPUT "{\n /* index */\n";
2545 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @filled_idx );
2546 printf OUTPUT ",\n /* null sub-index */\n%s", DUMP_ARRAY( "0x%04x", 0, ($null_offset) x 16 );
2547
2548 # dump the second-level indexes
2549
2550 for (my $i = 0; $i < 256; $i++)
2551 {
2552 next unless ($filled_idx[$i] > 256);
2553 my @table = @filled[($i<<4)..($i<<4)+15];
2554 for (my $j = 0; $j < 16; $j++) { $table[$j] ||= $null_offset; }
2555 printf OUTPUT ",\n /* sub-index %02x */\n", $i;
2556 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
2557 }
2558
2559 # dump the 16-char subsets
2560
2561 printf OUTPUT ",\n /* null mapping */\n";
2562 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, (0) x 32 );
2563
2564 for (my $i = 0; $i < 4096; $i++)
2565 {
2566 next unless $filled[$i];
2567 my @table = (0) x 32;
2568 for (my $j = 0; $j < 16; $j++)
2569 {
2570 if (defined $decomp_table[($i<<4) + $j])
2571 {
2572 $table[2 * $j] = ${$decomp_table[($i << 4) + $j]}[0];
2573 $table[2 * $j + 1] = ${$decomp_table[($i << 4) + $j]}[1];
2574 }
2575 }
2576 printf OUTPUT ",\n /* 0x%03x0 .. 0x%03xf */\n", $i, $i;
2577 printf OUTPUT "%s", DUMP_ARRAY( "0x%04x", 0, @table );
2578 }
2579
2580 printf OUTPUT "\n};\n\n";
2581
2582 dump_full_compat_table();
2583
2584 print OUTPUT <<"EOF";
2585 static const WCHAR *get_compatmap_entry( WCHAR ch, unsigned int *len )
2586 {
2587 unsigned short offset_offs = compatmap_table[compatmap_table[ch >> 8] + ((ch >> 4) & 0xf)] + (ch & 0xf);
2588 unsigned short start = compatmap_table[offset_offs];
2589 unsigned short end = compatmap_table[offset_offs + 1];
2590
2591 if (end > start)
2592 {
2593 *len = end - start;
2594 return compatmap_table + start;
2595 }
2596
2597 return NULL;
2598 }
2599
2600 unsigned int DECLSPEC_HIDDEN wine_decompose( int flags, WCHAR ch, WCHAR *dst, unsigned int dstlen )
2601 {
2602 const WCHAR *ptr = NULL;
2603 unsigned int res, len, dst_pos = 0;
2604
2605 *dst = ch;
2606
2607 if (flags & WINE_DECOMPOSE_COMPAT)
2608 ptr = get_compatmap_entry( ch, &len );
2609
2610 if (!ptr)
2611 {
2612 ptr = table + table[table[ch >> 8] + ((ch >> 4) & 0x0f)] + 2 * (ch & 0xf);
2613 len = ptr[1] ? 2 : 1;
2614 }
2615
2616 if (!*ptr) return 1;
2617 if (dstlen < len) return 0;
2618 /* apply the decomposition recursively */
2619 while (len--)
2620 {
2621 if (!(res = wine_decompose( flags, *ptr++, dst + dst_pos, dstlen - dst_pos - len ))) return 0;
2622 dst_pos += res;
2623 }
2624
2625 return dst_pos;
2626 }
2627 EOF
2628 close OUTPUT;
2629 save_file($filename);
2630 }
2631
2632 ################################################################
2633 # dump the combining class table
2634 sub dump_combining_class($)
2635 {
2636 my $filename = shift;
2637
2638 open OUTPUT,">$filename.new" or die "Cannot create $filename";
2639 print "Building $filename\n";
2640 print OUTPUT "/* Unicode Combining Classes */\n";
2641 print OUTPUT "/* generated from $UNIDATA/UnicodeData.txt */\n";
2642 print OUTPUT "/* DO NOT EDIT!! */\n\n";
2643 print OUTPUT "#include \"windef.h\"\n\n";
2644
2645 dump_two_level_mapping( "combining_class_table", 0, @combining_class_table );
2646 close OUTPUT;
2647 save_file($filename);
2648 }
2649
2650 ################################################################
2651 # output a codepage definition file from the global tables
2652 sub output_codepage_file($$$$)
2653 {
2654 my ($codepage, $filename, $comment, $has_glyphs) = @_;
2655
2656 my $output = sprintf "libs/port/c_%03d.c", $codepage;
2657 open OUTPUT,">$output.new" or die "Cannot create $output";
2658
2659 printf "Building %s from %s (%s)\n", $output, $filename || "hardcoded data", $comment;
2660
2661 # dump all tables
2662
2663 printf OUTPUT "/* code page %03d (%s) */\n", $codepage, $comment;
2664 if ($filename)
2665 {
2666 print OUTPUT "/* generated from $MAPPINGS/$filename */\n";
2667 print OUTPUT "/* DO NOT EDIT!! */\n\n";
2668 }
2669 else
2670 {
2671 printf OUTPUT "/* Automatically generated; DO NOT EDIT!! */\n\n";
2672 }
2673 printf OUTPUT "#include \"wine/unicode.h\"\n\n";
2674
2675 if (!@lead_bytes) { dump_sbcs_table( $codepage, $has_glyphs, $comment, $default_char, $default_wchar ); }
2676 else { dump_dbcs_table( $codepage, $comment, $default_char, $default_wchar, get_lb_ranges() ); }
2677 close OUTPUT;
2678 save_file($output);
2679 }
2680
2681 ################################################################
2682 # read a "bestfit" Windows mapping file
2683 sub read_bestfit_file($)
2684 {
2685 my ($filename) = @_;
2686 my $state = "";
2687 my ($codepage, $width, $count);
2688 my ($lb_cur, $lb_end);
2689
2690 my $INPUT = open_data_file( $MAPPINGS, $filename ) or die "Cannot open $filename";
2691
2692 while (<$INPUT>)
2693 {
2694 next if /^;/; # skip comments
2695 next if /^\s*$/; # skip empty lines
2696 next if /\x1a/; # skip ^Z
2697 last if /^ENDCODEPAGE/;
2698
2699 if (/^CODEPAGE\s+(\d+)/)
2700 {
2701 $codepage = $1;
2702 next;
2703 }
2704 if (/^CPINFO\s+(\d+)\s+0x([0-9a-fA-f]+)\s+0x([0-9a-fA-F]+)/)
2705 {
2706 $width = $1;
2707 $default_char = hex $2;
2708 $default_wchar = hex $3;
2709 next;
2710 }
2711 if (/^(MBTABLE|WCTABLE|DBCSRANGE|DBCSTABLE)\s+(\d+)/)
2712 {
2713 $state = $1;
2714 $count = $2;
2715 next;
2716 }
2717 if (/^0x([0-9a-fA-F]+)\s+0x([0-9a-fA-F]+)/)
2718 {
2719 if ($state eq "MBTABLE")
2720 {
2721 my $cp = hex $1;
2722 my $uni = hex $2;
2723 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
2724 next;
2725 }
2726 if ($state eq "WCTABLE")
2727 {
2728 my $uni = hex $1;
2729 my $cp = hex $2;
2730 $uni2cp[$uni] = $cp unless defined($uni2cp[$uni]);
2731 next;
2732 }
2733 if ($state eq "DBCSRANGE")
2734 {
2735 my $start = hex $1;
2736 my $end = hex $2;
2737 for (my $i = $start; $i <= $end; $i++) { add_lead_byte( $i ); }
2738 $lb_cur = $start;
2739 $lb_end = $end;
2740 next;
2741 }
2742 if ($state eq "DBCSTABLE")
2743 {
2744 my $mb = hex $1;
2745 my $uni = hex $2;
2746 my $cp = ($lb_cur << 8) | $mb;
2747 $cp2uni[$cp] = $uni unless defined($cp2uni[$cp]);
2748 if (!--$count)
2749 {
2750 if (++$lb_cur > $lb_end) { $state = "DBCSRANGE"; }
2751 }
2752 next;
2753 }
2754 }
2755 die "$filename: Unrecognized line $_\n";
2756 }
2757 close $INPUT;
2758 }
2759
2760
2761 ################################################################
2762 # read an input file and generate the corresponding .c file
2763 sub HANDLE_FILE(@)
2764 {
2765 my ($codepage,$filename,$has_glyphs,$comment,$first_private,$def,$defw) = @_;
2766
2767 @cp2uni = ();
2768 @lead_bytes = ();
2769 @uni2cp = ();
2770 $default_char = $def || $DEF_CHAR;
2771 $default_wchar = $defw || $DEF_CHAR;
2772
2773 # some codepage files are special
2774 if ($codepage == 20932)
2775 {
2776 READ_JIS0208_FILE( $filename );
2777 add_default_mappings( $first_private );
2778 }
2779 elsif ($codepage == 20127)
2780 {
2781 fill_20127_codepage();
2782 add_default_mappings( $first_private );
2783 }
2784 elsif ($filename =~ /\/bestfit/)
2785 {
2786 read_bestfit_file( $filename );
2787 }
2788 else
2789 {
2790 read_codepage_file( $filename );
2791 if ($codepage == 10001)
2792 {
2793 # add Shift-JIS mappings
2794 read_bestfit_file( "VENDORS/MICSFT/WindowsBestFit/bestfit932.txt" );
2795 }
2796 else
2797 {
2798 add_default_mappings( $first_private );
2799 }
2800 }
2801
2802 output_codepage_file( $codepage, $filename, $comment, $has_glyphs );
2803 }
2804
2805
2806 ################################################################
2807 # save a file if modified
2808 sub save_file($)
2809 {
2810 my $file = shift;
2811 if (-f $file && !system "cmp $file $file.new >/dev/null")
2812 {
2813 unlink "$file.new";
2814 }
2815 else
2816 {
2817 rename "$file.new", "$file";
2818 }
2819 }
2820
2821
2822 ################################################################
2823 # output the list of codepage tables into the cptable.c file
2824 sub output_cptable($)
2825 {
2826 my $output = shift;
2827 my @tables_decl = ();
2828
2829 printf "Building %s\n", $output;
2830
2831 foreach my $file (@allfiles)
2832 {
2833 my ($codepage,$filename) = @$file;
2834 push @tables_decl, sprintf("extern union cptable cptable_%03d;\n",$codepage);
2835 }
2836
2837 push @tables_decl, sprintf("\nstatic const union cptable * const cptables[%d] =\n{\n",$#allfiles+1);
2838 foreach my $file (@allfiles)
2839 {
2840 my ($codepage,$filename) = @$file;
2841 push @tables_decl, sprintf(" &cptable_%03d,\n", $codepage);
2842 }
2843 push @tables_decl, "};";
2844 REPLACE_IN_FILE( $output, @tables_decl );
2845 }
2846
2847 ################################################################
2848 # replace the contents of a file between ### cpmap ### marks
2849
2850 sub REPLACE_IN_FILE($@)
2851 {
2852 my $name = shift;
2853 my @data = @_;
2854 my @lines = ();
2855 open(FILE,$name) or die "Can't open $name";
2856 while (<FILE>)
2857 {
2858 push @lines, $_;
2859 last if /\#\#\# cpmap begin \#\#\#/;
2860 }
2861 push @lines, @data;
2862 while (<FILE>)
2863 {
2864 if (/\#\#\# cpmap end \#\#\#/) { push @lines, "\n", $_; last; }
2865 }
2866 push @lines, <FILE>;
2867 open(FILE,">$name.new") or die "Can't modify $name";
2868 print FILE @lines;
2869 close(FILE);
2870 save_file($name);
2871 }
2872
2873 ################################################################
2874 # main routine
2875
2876 chdir ".." if -f "./make_unicode";
2877 READ_DEFAULTS( $DEFAULTS );
2878 dump_case_mappings( "libs/port/casemap.c" );
2879 dump_sortkeys( "libs/port/collation.c", READ_SORTKEYS_FILE() );
2880 dump_compose_table( "libs/port/compose.c" );
2881 dump_decompose_table( "libs/port/decompose.c" );
2882 dump_ctype_tables( "libs/port/wctype.c" );
2883 dump_bidi_dir_table( "dlls/gdi32/direction.c" );
2884 dump_bidi_dir_table( "dlls/usp10/direction.c" );
2885 dump_bidi_dir_table( "dlls/dwrite/direction.c" );
2886 dump_string_type_table( "dlls/kernel32/wctype.c" );
2887 dump_digit_folding( "libs/port/digitmap.c" );
2888 dump_combining_class( "libs/port/combclass.c" );
2889 dump_mirroring( "dlls/usp10/mirror.c" );
2890 dump_mirroring( "dlls/dwrite/mirror.c" );
2891 dump_bracket( "dlls/usp10/bracket.c" );
2892 dump_bracket( "dlls/dwrite/bracket.c" );
2893 dump_shaping( "dlls/usp10/shaping.c" );
2894 dump_linebreak( "dlls/usp10/linebreak.c" );
2895 dump_linebreak( "dlls/dwrite/linebreak.c" );
2896 dump_scripts( "dlls/dwrite/scripts" );
2897 dump_indic( "dlls/usp10/indicsyllable.c" );
2898 dump_intl_nls("loader/l_intl.nls");
2899 dump_vertical( "dlls/gdi32/vertical.c" );
2900 dump_vertical( "dlls/wineps.drv/vertical.c" );
2901 dump_nameprep( "dlls/kernel32/nameprep.c" );
2902
2903 foreach my $file (@allfiles) { HANDLE_FILE( @{$file} ); }
2904
2905 output_cptable("libs/port/cptable.c");
2906
2907 exit 0;
2908
2909 # Local Variables:
2910 # compile-command: "./make_unicode"
2911 # End:
Windows API implementation