makecasefoldhashtable.pl
1
2
3
4
5
#!/usr/bin/perl -w
use warnings;
use strict;
6
7
8
9
10
my $HASHBUCKETS1_16 = 256;
my $HASHBUCKETS1_32 = 16;
my $HASHBUCKETS2_16 = 16;
my $HASHBUCKETS3_16 = 4;
11
12
print <<__EOF__;
/*
13
* This file is part of PhysicsFS (https://icculus.org/physfs/)
14
15
16
17
18
19
20
*
* This data generated by physfs/extras/makecasefoldhashtable.pl ...
* Do not manually edit this file!
*
* Please see the file LICENSE.txt in the source's root directory.
*/
21
22
23
#ifndef _INCLUDE_PHYSFS_CASEFOLDING_H_
#define _INCLUDE_PHYSFS_CASEFOLDING_H_
24
25
26
27
#ifndef __PHYSICSFS_INTERNAL__
#error Do not include this header from your applications.
#endif
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
/* We build three simple hashmaps here: one that maps Unicode codepoints to
a one, two, or three lowercase codepoints. To retrieve this info: look at
case_fold_hashX, where X is 1, 2, or 3. Most foldable codepoints fold to one,
a few dozen fold to two, and a handful fold to three. If the codepoint isn't
in any of these hashes, it doesn't fold (no separate upper and lowercase).
Almost all these codepoints fit into 16 bits, so we hash them as such to save
memory. If a codepoint is > 0xFFFF, we have separate hashes for them,
since there are (currently) only about 120 of them and (currently) all of them
map to a single lowercase codepoint. */
typedef struct CaseFoldMapping1_32
{
PHYSFS_uint32 from;
PHYSFS_uint32 to0;
} CaseFoldMapping1_32;
typedef struct CaseFoldMapping1_16
{
PHYSFS_uint16 from;
PHYSFS_uint16 to0;
} CaseFoldMapping1_16;
typedef struct CaseFoldMapping2_16
{
PHYSFS_uint16 from;
PHYSFS_uint16 to0;
PHYSFS_uint16 to1;
} CaseFoldMapping2_16;
typedef struct CaseFoldMapping3_16
{
PHYSFS_uint16 from;
PHYSFS_uint16 to0;
PHYSFS_uint16 to1;
PHYSFS_uint16 to2;
} CaseFoldMapping3_16;
typedef struct CaseFoldHashBucket1_16
{
const CaseFoldMapping1_16 *list;
const PHYSFS_uint8 count;
} CaseFoldHashBucket1_16;
typedef struct CaseFoldHashBucket1_32
{
const CaseFoldMapping1_32 *list;
const PHYSFS_uint8 count;
} CaseFoldHashBucket1_32;
typedef struct CaseFoldHashBucket2_16
{
const CaseFoldMapping2_16 *list;
const PHYSFS_uint8 count;
} CaseFoldHashBucket2_16;
typedef struct CaseFoldHashBucket3_16
{
const CaseFoldMapping3_16 *list;
const PHYSFS_uint8 count;
} CaseFoldHashBucket3_16;
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
my @foldPairs1_16;
my @foldPairs2_16;
my @foldPairs3_16;
my @foldPairs1_32;
for (my $i = 0; $i < $HASHBUCKETS1_16; $i++) {
$foldPairs1_16[$i] = '';
}
for (my $i = 0; $i < $HASHBUCKETS1_32; $i++) {
$foldPairs1_32[$i] = '';
}
for (my $i = 0; $i < $HASHBUCKETS2_16; $i++) {
$foldPairs2_16[$i] = '';
}
110
111
for (my $i = 0; $i < $HASHBUCKETS3_16; $i++) {
$foldPairs3_16[$i] = '';
112
113
114
115
116
117
118
119
120
121
122
123
124
125
}
open(FH,'<','casefolding.txt') or die("failed to open casefolding.txt: $!\n");
while (<FH>) {
chomp;
# strip comments from textfile...
s/\#.*\Z//;
# strip whitespace...
s/\A\s+//;
s/\s+\Z//;
next if not /\A([a-fA-F0-9]+)\;\s*(.)\;\s*(.+)\;/;
my ($code, $status, $mapping) = ($1, $2, $3);
127
128
129
130
my $hexxed = hex($code);
#print("// code '$code' status '$status' mapping '$mapping'\n");
if (($status eq 'C') or ($status eq 'F')) {
131
my ($map1, $map2, $map3) = (undef, undef, undef);
132
133
134
135
$map1 = $1 if $mapping =~ s/\A([a-fA-F0-9]+)(\s*|\Z)//;
$map2 = $1 if $mapping =~ s/\A([a-fA-F0-9]+)(\s*|\Z)//;
$map3 = $1 if $mapping =~ s/\A([a-fA-F0-9]+)(\s*|\Z)//;
die("mapping space too small for '$code'\n") if ($mapping ne '');
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
die("problem parsing mapping for '$code'\n") if (not defined($map1));
if ($hexxed < 128) {
# Just ignore these, we'll handle the low-ASCII ones ourselves.
} elsif ($hexxed > 0xFFFF) {
# We just need to add the 32-bit 2 and/or 3 codepoint maps if this die()'s here.
die("Uhoh, a codepoint > 0xFFFF that folds to multiple codepoints! Fixme.") if defined($map2);
my $hashed = (($hexxed ^ ($hexxed >> 8)) & ($HASHBUCKETS1_32-1));
#print("// hexxed '$hexxed' hashed1 '$hashed'\n");
$foldPairs1_32[$hashed] .= " { 0x$code, 0x$map1 },\n";
} elsif (not defined($map2)) {
my $hashed = (($hexxed ^ ($hexxed >> 8)) & ($HASHBUCKETS1_16-1));
#print("// hexxed '$hexxed' hashed1 '$hashed'\n");
$foldPairs1_16[$hashed] .= " { 0x$code, 0x$map1 },\n";
} elsif (not defined($map3)) {
my $hashed = (($hexxed ^ ($hexxed >> 8)) & ($HASHBUCKETS2_16-1));
#print("// hexxed '$hexxed' hashed2 '$hashed'\n");
$foldPairs2_16[$hashed] .= " { 0x$code, 0x$map1, 0x$map2 },\n";
} else {
my $hashed = (($hexxed ^ ($hexxed >> 8)) & ($HASHBUCKETS3_16-1));
#print("// hexxed '$hexxed' hashed3 '$hashed'\n");
$foldPairs3_16[$hashed] .= " { 0x$code, 0x$map1, 0x$map2, 0x$map3 },\n";
}
159
160
161
162
}
}
close(FH);
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
for (my $i = 0; $i < $HASHBUCKETS1_16; $i++) {
$foldPairs1_16[$i] =~ s/,\n\Z//;
my $str = $foldPairs1_16[$i];
next if $str eq '';
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
my $sym = "case_fold1_16_${num}";
print("static const CaseFoldMapping1_16 ${sym}[] = {\n$str\n};\n\n");
}
for (my $i = 0; $i < $HASHBUCKETS1_32; $i++) {
$foldPairs1_32[$i] =~ s/,\n\Z//;
my $str = $foldPairs1_32[$i];
next if $str eq '';
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
my $sym = "case_fold1_32_${num}";
print("static const CaseFoldMapping1_32 ${sym}[] = {\n$str\n};\n\n");
}
for (my $i = 0; $i < $HASHBUCKETS2_16; $i++) {
$foldPairs2_16[$i] =~ s/,\n\Z//;
my $str = $foldPairs2_16[$i];
next if $str eq '';
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
my $sym = "case_fold2_16_${num}";
print("static const CaseFoldMapping2_16 ${sym}[] = {\n$str\n};\n\n");
}
for (my $i = 0; $i < $HASHBUCKETS3_16; $i++) {
$foldPairs3_16[$i] =~ s/,\n\Z//;
my $str = $foldPairs3_16[$i];
196
197
198
next if $str eq '';
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
199
200
my $sym = "case_fold3_16_${num}";
print("static const CaseFoldMapping3_16 ${sym}[] = {\n$str\n};\n\n");
203
print("static const CaseFoldHashBucket1_16 case_fold_hash1_16[] = {\n");
205
206
for (my $i = 0; $i < $HASHBUCKETS1_16; $i++) {
my $str = $foldPairs1_16[$i];
207
if ($str eq '') {
208
print(" { NULL, 0 },\n");
209
210
211
} else {
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
212
213
my $sym = "case_fold1_16_${num}";
print(" { $sym, __PHYSFS_ARRAYLEN($sym) },\n");
214
215
216
217
}
}
print("};\n\n");
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
print("static const CaseFoldHashBucket1_32 case_fold_hash1_32[] = {\n");
for (my $i = 0; $i < $HASHBUCKETS1_32; $i++) {
my $str = $foldPairs1_32[$i];
if ($str eq '') {
print(" { NULL, 0 },\n");
} else {
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
my $sym = "case_fold1_32_${num}";
print(" { $sym, __PHYSFS_ARRAYLEN($sym) },\n");
}
}
print("};\n\n");
print("static const CaseFoldHashBucket2_16 case_fold_hash2_16[] = {\n");
for (my $i = 0; $i < $HASHBUCKETS2_16; $i++) {
my $str = $foldPairs2_16[$i];
if ($str eq '') {
print(" { NULL, 0 },\n");
} else {
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
my $sym = "case_fold2_16_${num}";
print(" { $sym, __PHYSFS_ARRAYLEN($sym) },\n");
}
}
print("};\n\n");
print("static const CaseFoldHashBucket3_16 case_fold_hash3_16[] = {\n");
for (my $i = 0; $i < $HASHBUCKETS3_16; $i++) {
my $str = $foldPairs3_16[$i];
if ($str eq '') {
print(" { NULL, 0 },\n");
} else {
my $num = '000' . $i;
$num =~ s/\A.*?(\d\d\d)\Z/$1/;
my $sym = "case_fold3_16_${num}";
print(" { $sym, __PHYSFS_ARRAYLEN($sym) },\n");
}
}
print("};\n\n");
print <<__EOF__;
#endif /* _INCLUDE_PHYSFS_CASEFOLDING_H_ */
/* end of physfs_casefolding.h ... */
__EOF__
273
274
275
exit 0;
# end of makecashfoldhashtable.pl ...