1 /* FriBidi
2 * gen-brackets-tab.c - generate brackets.tab.i
3 *
4 * Author:
5 * Behdad Esfahbod, 2001, 2002, 2004
6 * Dov Grobgeld 2017
7 *
8 * Copyright (C) 2004 Sharif FarsiWeb, Inc
9 * Copyright (C) 2001,2002,2004 Behdad Esfahbod
10 * Copyright (C) 2017 Dov Grobgeld
11 *
12 * This library is free software; you can redistribute it and/or
13 * modify it under the terms of the GNU Lesser General Public
14 * License as published by the Free Software Foundation; either
15 * version 2.1 of the License, or (at your option) any later version.
16 *
17 * This library is distributed in the hope that it will be useful,
18 * but WITHOUT ANY WARRANTY; without even the implied warranty of
19 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
20 * Lesser General Public License for more details.
21 *
22 * You should have received a copy of the GNU Lesser General Public License
23 * along with this library, in a file named COPYING; if not, write to the
24 * Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
25 * Boston, MA 02110-1301, USA
26 *
27 * For licensing issues, contact <fribidi.license@gmail.com>.
28 */
29
30 #include <common.h>
31 #include <ctype.h>
32 #include <fribidi-unicode.h>
33
34 #include <stdio.h>
35 #ifdef HAVE_CONFIG_H
36 # include <config.h>
37 #endif
38
39 #ifdef STDC_HEADERS
40 # include <stdlib.h>
41 # include <stddef.h>
42 #else
43 # if HAVE_STDLIB_H
44 # include <stdlib.h>
45 # endif
46 #endif
47 #ifdef HAVE_STRING_H
48 # if !STDC_HEADERS && HAVE_MEMORY_H
49 # include <memory.h>
50 # endif
51 # include <string.h>
52 #endif
53 #ifdef HAVE_STRINGS_H
54 # include <strings.h>
55 #endif
56
57 #include "packtab.h"
58
59 #define appname "gen-brackets-tab"
60 #define outputname "brackets.tab.i"
61
62 static void
63 die (
64 const char *msg
65 )
66 {
67 fprintf (stderr, appname ": %s\n", msg);
68 exit (1);
69 }
70
71 static void
72 die2 (
73 const char *fmt,
74 const char *p
75 )
76 {
77 fprintf (stderr, appname ": ");
78 fprintf (stderr, fmt, p);
79 fprintf (stderr, "\n");
80 exit (1);
81 }
82
83 static void
84 die3 (
85 const char *fmt,
86 const char *p,
87 const char *q
88 )
89 {
90 fprintf (stderr, appname ": ");
91 fprintf (stderr, fmt, p, q);
92 fprintf (stderr, "\n");
93 exit (1);
94 }
95
96 static void
97 die4 (
98 const char *fmt,
99 unsigned long l,
100 unsigned long p,
101 unsigned long q
102 )
103 {
104 fprintf (stderr, appname ": ");
105 fprintf (stderr, fmt, l, p, q);
106 fprintf (stderr, "\n");
107 exit (1);
108 }
109
110 #define table_name "Brk"
111 #define macro_name "FRIBIDI_GET_BRACKETS"
112
113 static signed int table[FRIBIDI_UNICODE_CHARS];
114 static signed int equiv_table[FRIBIDI_UNICODE_CHARS];
115 static char buf[4000];
116 static signed long max_dist;
117
118 static void
119 init (
120 void
121 )
122 {
123 max_dist = 0;
124 }
125
126 static void
127 clear_tabs (
128 void
129 )
130 {
131 register FriBidiChar c;
132
133 for (c = 0; c < FRIBIDI_UNICODE_CHARS; c++)
134 {
135 table[c] = 0;
136 equiv_table[c] = 0;
137 }
138 }
139
140 static signed int table[FRIBIDI_UNICODE_CHARS];
141 static char buf[4000];
142
143 /* Read the canonical mapping of unicode characters and store them in the
144 equiv_table array. */
145 static void
146 read_unicode_data_txt_equivalence (
147 FILE *f
148 )
149 {
150 unsigned long c, l;
151
152 l = 0;
153 while (fgets (buf, sizeof buf, f))
154 {
155 int i;
156 const char *s = buf;
157 char ce_string[256]; /* For parsing the equivalence */
158 char *p = NULL;
159 int ce, in_tag;
160
161 l++;
162
163 while (*s == ' ')
164 s++;
165
166 if (s[0] == '#' || s[0] == '\0' || s[0] == '\n')
167 continue;
168 /* Field: 0 ; 1 ; 2 ; 3 ; 4 ; 5 */
169 i = sscanf (s, "%lx;%*[^;];%*[^;];%*[^;];%*[^;];%[^;]", &c, ce_string);
170 if (c >= FRIBIDI_UNICODE_CHARS)
171 {
172 fprintf (stderr, "invalid input at line %ld: %s", l, s);
173 exit(1);
174 }
175 if (i==1)
176 continue;
177
178 /* split and parse ce */
179 p = ce_string;
180 ce = -1;
181 in_tag = 0;
182 while(*p)
183 {
184 if (*p==';')
185 break;
186 else if (*p=='<')
187 in_tag = 1;
188 else if (*p=='>')
189 in_tag = 0;
190 else if (!in_tag && isalnum(*p))
191 {
192 /* Assume we got a hexa decimal */
193 ce = strtol(p,NULL,16);
194 break;
195 }
196 p++;
197 }
198
199 /* FIXME: We don't handle First..Last parts of UnicodeData.txt,
200 * but it works, since all those are LTR. */
201 equiv_table[c] = ce;
202 }
203 }
204
205 static void
206 read_bidi_brackets_txt (
207 FILE *f
208 )
209 {
210 unsigned long l;
211
212 l = 0;
213 while (fgets (buf, sizeof buf, f))
214 {
215 unsigned long i, j;
216 signed long dist;
217 int k;
218 const char *s = buf;
219 char open_close;
220
221 l++;
222
223 while (*s == ' ')
224 s++;
225
226 if (s[0] == '#' || s[0] == '\0' || s[0] == '\n')
227 continue;
228
229 k = sscanf (s, "%lx; %lx; %c", &i, &j, &open_close);
230 if (k != 3 || i >= FRIBIDI_UNICODE_CHARS || j >= FRIBIDI_UNICODE_CHARS)
231 die4 ("invalid pair in input at line %ld: %04lX, %04lX", l, i, j);
232
233 /* Open braces map to themself */
234 if (open_close=='o')
235 j = i;
236
237 /* Turn j into the unicode equivalence if it exists */
238 if (equiv_table[j])
239 {
240 /* printf("Found match for %04x->%04x\n", j, equiv_table[j]); */
241 j = equiv_table[j];
242 }
243
244 dist = ((signed long) j - (signed long) i);
245 table[i] = dist;
246 if (dist > max_dist)
247 max_dist = dist;
248 else if (-dist > max_dist)
249 max_dist = -dist;
250 }
251 }
252
253 static void
254 read_data (
255 const char *bracket_datafile_type,
256 const char *bracket_datafile_name,
257 const char *uni_datafile_type,
258 const char *uni_datafile_name
259 )
260 {
261 FILE *f;
262
263 clear_tabs ();
264
265 if (!(f = fopen (uni_datafile_name, "rt")))
266 die2 ("error: cannot open `%s' for reading", bracket_datafile_name);
267
268 if (!strcmp (uni_datafile_type, "UnicodeData.txt"))
269 read_unicode_data_txt_equivalence (f);
270 else
271 die2 ("error: unknown data-file-type %s", uni_datafile_type);
272
273 fprintf (stderr, "Reading `%s'\n", bracket_datafile_name);
274 if (!(f = fopen (bracket_datafile_name, "rt")))
275 die2 ("error: cannot open `%s' for reading", bracket_datafile_name);
276
277 if (!strcmp (bracket_datafile_type, "BidiBrackets.txt"))
278 read_bidi_brackets_txt (f);
279 else
280 die2 ("error: unknown data-file-type %s", bracket_datafile_type);
281
282 fclose (f);
283 }
284
285 static void
286 gen_brackets_tab (
287 int max_depth,
288 const char *data_file_type
289 )
290 {
291 int key_bytes;
292 const char *key_type;
293
294 printf ("/* " outputname "\n * generated by " appname " (" FRIBIDI_NAME " "
295 FRIBIDI_VERSION ")\n" " * from the file %s of Unicode version "
296 FRIBIDI_UNICODE_VERSION ". */\n\n", data_file_type);
297
298 printf ("#define PACKTAB_UINT8 uint8_t\n"
299 "#define PACKTAB_UINT16 uint16_t\n"
300 "#define PACKTAB_UINT32 uint32_t\n\n");
301
302 key_bytes = max_dist <= 0x7f ? 1 : max_dist < 0x7fff ? 2 : 4;
303 key_type = key_bytes == 1 ? "int8_t" : key_bytes == 2 ?
304 "int16_t" : "int32_t";
305
306 if (!pack_table
307 (table, FRIBIDI_UNICODE_CHARS, key_bytes, 0, max_depth, 1, NULL,
308 key_type, table_name, macro_name "_DELTA", stdout))
309 die ("error: insufficient memory, decrease max_depth");
310
311 printf ("#undef PACKTAB_UINT8\n"
312 "#undef PACKTAB_UINT16\n" "#undef PACKTAB_UINT32\n\n");
313
314 printf ("#define " macro_name "(x) ((x) + " macro_name "_DELTA(x))\n\n");
315
316 printf ("/* End of generated " outputname " */\n");
317 }
318
319 int
320 main (
321 int argc,
322 const char **argv
323 )
324 {
325 const char *bracket_datafile_type = "BidiBrackets.txt";
326 const char *uni_datafile_type = "UnicodeData.txt";
327
328 if (argc < 4)
329 die3 ("usage:\n " appname " max-depth /path/to/%s /path/to/%s [junk...]",
330 bracket_datafile_type,
331 uni_datafile_type);
332
333 {
334 int max_depth = atoi (argv[1]);
335 const char *bracket_datafile_name = argv[2];
336 const char *uni_datafile_name = argv[3];
337
338 if (max_depth < 2)
339 die ("invalid depth");
340
341 init ();
342 read_data (bracket_datafile_type, bracket_datafile_name,
343 uni_datafile_type, uni_datafile_name);
344 gen_brackets_tab (max_depth, bracket_datafile_type);
345 }
346
347 return 0;
348 }