1 | /* $NetBSD: mksyntax.c,v 1.23 2000/07/18 19:13:21 cgd Exp $ */
|
---|
2 |
|
---|
3 | /*-
|
---|
4 | * Copyright (c) 1991, 1993
|
---|
5 | * The Regents of the University of California. All rights reserved.
|
---|
6 | *
|
---|
7 | * This code is derived from software contributed to Berkeley by
|
---|
8 | * Kenneth Almquist.
|
---|
9 | *
|
---|
10 | * Redistribution and use in source and binary forms, with or without
|
---|
11 | * modification, are permitted provided that the following conditions
|
---|
12 | * are met:
|
---|
13 | * 1. Redistributions of source code must retain the above copyright
|
---|
14 | * notice, this list of conditions and the following disclaimer.
|
---|
15 | * 2. Redistributions in binary form must reproduce the above copyright
|
---|
16 | * notice, this list of conditions and the following disclaimer in the
|
---|
17 | * documentation and/or other materials provided with the distribution.
|
---|
18 | * 3. All advertising materials mentioning features or use of this software
|
---|
19 | * must display the following acknowledgement:
|
---|
20 | * This product includes software developed by the University of
|
---|
21 | * California, Berkeley and its contributors.
|
---|
22 | * 4. Neither the name of the University nor the names of its contributors
|
---|
23 | * may be used to endorse or promote products derived from this software
|
---|
24 | * without specific prior written permission.
|
---|
25 | *
|
---|
26 | * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
|
---|
27 | * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
---|
28 | * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
---|
29 | * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
|
---|
30 | * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
---|
31 | * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
---|
32 | * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
---|
33 | * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
---|
34 | * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
---|
35 | * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
---|
36 | * SUCH DAMAGE.
|
---|
37 | */
|
---|
38 |
|
---|
39 | #ifndef lint
|
---|
40 | static const char copyright[] =
|
---|
41 | "@(#) Copyright (c) 1991, 1993\n\
|
---|
42 | The Regents of the University of California. All rights reserved.\n";
|
---|
43 | #endif /* not lint */
|
---|
44 |
|
---|
45 | #ifndef lint
|
---|
46 | #if 0
|
---|
47 | static char sccsid[] = "@(#)mksyntax.c 8.2 (Berkeley) 5/4/95";
|
---|
48 | #else
|
---|
49 | static const char rcsid[] =
|
---|
50 | "$NetBSD: mksyntax.c,v 1.23 2000/07/18 19:13:21 cgd Exp $";
|
---|
51 | #endif
|
---|
52 | #endif /* not lint */
|
---|
53 |
|
---|
54 | /*
|
---|
55 | * This program creates syntax.h and syntax.c.
|
---|
56 | */
|
---|
57 |
|
---|
58 | #include <stdio.h>
|
---|
59 | #include <stdlib.h>
|
---|
60 | #include <string.h>
|
---|
61 | #include <sys/types.h>
|
---|
62 | #include "../parser.h"
|
---|
63 |
|
---|
64 |
|
---|
65 | struct synclass {
|
---|
66 | char *name;
|
---|
67 | char *comment;
|
---|
68 | };
|
---|
69 |
|
---|
70 | /* Syntax classes */
|
---|
71 | struct synclass synclass[] = {
|
---|
72 | { "CWORD", "character is nothing special" },
|
---|
73 | { "CNL", "newline character" },
|
---|
74 | { "CBACK", "a backslash character" },
|
---|
75 | { "CSQUOTE", "single quote" },
|
---|
76 | { "CDQUOTE", "double quote" },
|
---|
77 | { "CENDQUOTE", "a terminating quote" },
|
---|
78 | { "CBQUOTE", "backwards single quote" },
|
---|
79 | { "CVAR", "a dollar sign" },
|
---|
80 | { "CENDVAR", "a '}' character" },
|
---|
81 | { "CLP", "a left paren in arithmetic" },
|
---|
82 | { "CRP", "a right paren in arithmetic" },
|
---|
83 | { "CEOF", "end of file" },
|
---|
84 | { "CCTL", "like CWORD, except it must be escaped" },
|
---|
85 | { "CSPCL", "these terminate a word" },
|
---|
86 | { NULL, NULL }
|
---|
87 | };
|
---|
88 |
|
---|
89 |
|
---|
90 | /*
|
---|
91 | * Syntax classes for is_ functions. Warning: if you add new classes
|
---|
92 | * you may have to change the definition of the is_in_name macro.
|
---|
93 | */
|
---|
94 | struct synclass is_entry[] = {
|
---|
95 | { "ISDIGIT", "a digit" },
|
---|
96 | { "ISUPPER", "an upper case letter" },
|
---|
97 | { "ISLOWER", "a lower case letter" },
|
---|
98 | { "ISUNDER", "an underscore" },
|
---|
99 | { "ISSPECL", "the name of a special parameter" },
|
---|
100 | { NULL, NULL }
|
---|
101 | };
|
---|
102 |
|
---|
103 | static char writer[] = "\
|
---|
104 | /*\n\
|
---|
105 | * This file was generated by the mksyntax program.\n\
|
---|
106 | */\n\
|
---|
107 | \n";
|
---|
108 |
|
---|
109 |
|
---|
110 | static FILE *cfile;
|
---|
111 | static FILE *hfile;
|
---|
112 | static char *syntax[513];
|
---|
113 | static int base;
|
---|
114 | static int size; /* number of values which a char variable can have */
|
---|
115 | static int nbits; /* number of bits in a character */
|
---|
116 | static int digit_contig;/* true if digits are contiguous */
|
---|
117 |
|
---|
118 | static void filltable(char *);
|
---|
119 | static void init(void);
|
---|
120 | static void add(char *, char *);
|
---|
121 | static void print(char *);
|
---|
122 | static void output_type_macros(void);
|
---|
123 | static void digit_convert(void);
|
---|
124 | int main(int, char **);
|
---|
125 |
|
---|
126 | int
|
---|
127 | main(argc, argv)
|
---|
128 | int argc;
|
---|
129 | char **argv;
|
---|
130 | {
|
---|
131 | #ifdef TARGET_CHAR
|
---|
132 | TARGET_CHAR c;
|
---|
133 | TARGET_CHAR d;
|
---|
134 | #else
|
---|
135 | char c;
|
---|
136 | char d;
|
---|
137 | #endif
|
---|
138 | int sign;
|
---|
139 | int i;
|
---|
140 | char buf[80];
|
---|
141 | int pos;
|
---|
142 | static char digit[] = "0123456789";
|
---|
143 |
|
---|
144 | /* Create output files */
|
---|
145 | if ((cfile = fopen("syntax.c", "w")) == NULL) {
|
---|
146 | perror("syntax.c");
|
---|
147 | exit(2);
|
---|
148 | }
|
---|
149 | if ((hfile = fopen("syntax.h", "w")) == NULL) {
|
---|
150 | perror("syntax.h");
|
---|
151 | exit(2);
|
---|
152 | }
|
---|
153 | fputs(writer, hfile);
|
---|
154 | fputs(writer, cfile);
|
---|
155 |
|
---|
156 | /* Determine the characteristics of chars. */
|
---|
157 | c = -1;
|
---|
158 | if (c <= 0)
|
---|
159 | sign = 1;
|
---|
160 | else
|
---|
161 | sign = 0;
|
---|
162 | for (nbits = 1 ; ; nbits++) {
|
---|
163 | d = (1 << nbits) - 1;
|
---|
164 | if (d == c)
|
---|
165 | break;
|
---|
166 | }
|
---|
167 | printf("%s %d bit chars\n", sign? "signed" : "unsigned", nbits);
|
---|
168 | if (nbits > 9) {
|
---|
169 | fputs("Characters can't have more than 9 bits\n", stderr);
|
---|
170 | exit(2);
|
---|
171 | }
|
---|
172 | size = (1 << nbits) + 1;
|
---|
173 | base = 1;
|
---|
174 | if (sign)
|
---|
175 | base += 1 << (nbits - 1);
|
---|
176 | digit_contig = 1;
|
---|
177 | for (i = 0 ; i < 10 ; i++) {
|
---|
178 | if (digit[i] != '0' + i)
|
---|
179 | digit_contig = 0;
|
---|
180 | }
|
---|
181 |
|
---|
182 | fputs("#include <sys/cdefs.h>\n", hfile);
|
---|
183 | fputs("#include <ctype.h>\n", hfile);
|
---|
184 |
|
---|
185 | /* Generate the #define statements in the header file */
|
---|
186 | fputs("/* Syntax classes */\n", hfile);
|
---|
187 | for (i = 0 ; synclass[i].name ; i++) {
|
---|
188 | sprintf(buf, "#define %s %d", synclass[i].name, i);
|
---|
189 | fputs(buf, hfile);
|
---|
190 | for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
|
---|
191 | putc('\t', hfile);
|
---|
192 | fprintf(hfile, "/* %s */\n", synclass[i].comment);
|
---|
193 | }
|
---|
194 | putc('\n', hfile);
|
---|
195 | fputs("/* Syntax classes for is_ functions */\n", hfile);
|
---|
196 | for (i = 0 ; is_entry[i].name ; i++) {
|
---|
197 | sprintf(buf, "#define %s %#o", is_entry[i].name, 1 << i);
|
---|
198 | fputs(buf, hfile);
|
---|
199 | for (pos = strlen(buf) ; pos < 32 ; pos = (pos + 8) & ~07)
|
---|
200 | putc('\t', hfile);
|
---|
201 | fprintf(hfile, "/* %s */\n", is_entry[i].comment);
|
---|
202 | }
|
---|
203 | putc('\n', hfile);
|
---|
204 | fprintf(hfile, "#define SYNBASE %d\n", base);
|
---|
205 | fprintf(hfile, "#define PEOF %d\n\n", -base);
|
---|
206 | if (sign)
|
---|
207 | fprintf(hfile, "#define UPEOF %d\n\n", -base);
|
---|
208 | else
|
---|
209 | fprintf(hfile, "#define UPEOF ((unsigned char) %d)\n\n", -base);
|
---|
210 | putc('\n', hfile);
|
---|
211 | fputs("#define BASESYNTAX (basesyntax + SYNBASE)\n", hfile);
|
---|
212 | fputs("#define DQSYNTAX (dqsyntax + SYNBASE)\n", hfile);
|
---|
213 | fputs("#define SQSYNTAX (sqsyntax + SYNBASE)\n", hfile);
|
---|
214 | fputs("#define ARISYNTAX (arisyntax + SYNBASE)\n", hfile);
|
---|
215 | putc('\n', hfile);
|
---|
216 | output_type_macros(); /* is_digit, etc. */
|
---|
217 | putc('\n', hfile);
|
---|
218 |
|
---|
219 | /* Generate the syntax tables. */
|
---|
220 | fputs("#include \"shell.h\"\n", cfile);
|
---|
221 | fputs("#include \"syntax.h\"\n\n", cfile);
|
---|
222 | init();
|
---|
223 | fputs("/* syntax table used when not in quotes */\n", cfile);
|
---|
224 | add("\n", "CNL");
|
---|
225 | add("\\", "CBACK");
|
---|
226 | add("'", "CSQUOTE");
|
---|
227 | add("\"", "CDQUOTE");
|
---|
228 | add("`", "CBQUOTE");
|
---|
229 | add("$", "CVAR");
|
---|
230 | add("}", "CENDVAR");
|
---|
231 | add("<>();&| \t", "CSPCL");
|
---|
232 | print("basesyntax");
|
---|
233 | init();
|
---|
234 | fputs("\n/* syntax table used when in double quotes */\n", cfile);
|
---|
235 | add("\n", "CNL");
|
---|
236 | add("\\", "CBACK");
|
---|
237 | add("\"", "CENDQUOTE");
|
---|
238 | add("`", "CBQUOTE");
|
---|
239 | add("$", "CVAR");
|
---|
240 | add("}", "CENDVAR");
|
---|
241 | /* ':/' for tilde expansion, '-' for [a\-x] pattern ranges */
|
---|
242 | add("!*?[=~:/-]", "CCTL");
|
---|
243 | print("dqsyntax");
|
---|
244 | init();
|
---|
245 | fputs("\n/* syntax table used when in single quotes */\n", cfile);
|
---|
246 | add("\n", "CNL");
|
---|
247 | add("'", "CENDQUOTE");
|
---|
248 | /* ':/' for tilde expansion, '-' for [a\-x] pattern ranges */
|
---|
249 | add("!*?[=~:/-]\\", "CCTL");
|
---|
250 | print("sqsyntax");
|
---|
251 | init();
|
---|
252 | fputs("\n/* syntax table used when in arithmetic */\n", cfile);
|
---|
253 | add("\n", "CNL");
|
---|
254 | add("\\", "CBACK");
|
---|
255 | add("`", "CBQUOTE");
|
---|
256 | add("'", "CSQUOTE");
|
---|
257 | add("\"", "CDQUOTE");
|
---|
258 | add("$", "CVAR");
|
---|
259 | add("}", "CENDVAR");
|
---|
260 | add("(", "CLP");
|
---|
261 | add(")", "CRP");
|
---|
262 | print("arisyntax");
|
---|
263 | filltable("0");
|
---|
264 | fputs("\n/* character classification table */\n", cfile);
|
---|
265 | add("0123456789", "ISDIGIT");
|
---|
266 | add("abcdefghijklmnopqrstucvwxyz", "ISLOWER");
|
---|
267 | add("ABCDEFGHIJKLMNOPQRSTUCVWXYZ", "ISUPPER");
|
---|
268 | add("_", "ISUNDER");
|
---|
269 | add("#?$!-*@", "ISSPECL");
|
---|
270 | print("is_type");
|
---|
271 | if (! digit_contig)
|
---|
272 | digit_convert();
|
---|
273 | exit(0);
|
---|
274 | /* NOTREACHED */
|
---|
275 | }
|
---|
276 |
|
---|
277 |
|
---|
278 |
|
---|
279 | /*
|
---|
280 | * Clear the syntax table.
|
---|
281 | */
|
---|
282 |
|
---|
283 | static void
|
---|
284 | filltable(dftval)
|
---|
285 | char *dftval;
|
---|
286 | {
|
---|
287 | int i;
|
---|
288 |
|
---|
289 | for (i = 0 ; i < size ; i++)
|
---|
290 | syntax[i] = dftval;
|
---|
291 | }
|
---|
292 |
|
---|
293 |
|
---|
294 | /*
|
---|
295 | * Initialize the syntax table with default values.
|
---|
296 | */
|
---|
297 |
|
---|
298 | static void
|
---|
299 | init()
|
---|
300 | {
|
---|
301 | filltable("CWORD");
|
---|
302 | syntax[0] = "CEOF";
|
---|
303 | #ifdef TARGET_CHAR
|
---|
304 | syntax[base + (TARGET_CHAR)CTLESC] = "CCTL";
|
---|
305 | syntax[base + (TARGET_CHAR)CTLVAR] = "CCTL";
|
---|
306 | syntax[base + (TARGET_CHAR)CTLENDVAR] = "CCTL";
|
---|
307 | syntax[base + (TARGET_CHAR)CTLBACKQ] = "CCTL";
|
---|
308 | syntax[base + (TARGET_CHAR)CTLBACKQ + (TARGET_CHAR)CTLQUOTE] = "CCTL";
|
---|
309 | syntax[base + (TARGET_CHAR)CTLARI] = "CCTL";
|
---|
310 | syntax[base + (TARGET_CHAR)CTLENDARI] = "CCTL";
|
---|
311 | syntax[base + (TARGET_CHAR)CTLQUOTEMARK] = "CCTL";
|
---|
312 | #else
|
---|
313 | syntax[base + CTLESC] = "CCTL";
|
---|
314 | syntax[base + CTLVAR] = "CCTL";
|
---|
315 | syntax[base + CTLENDVAR] = "CCTL";
|
---|
316 | syntax[base + CTLBACKQ] = "CCTL";
|
---|
317 | syntax[base + CTLBACKQ + CTLQUOTE] = "CCTL";
|
---|
318 | syntax[base + CTLARI] = "CCTL";
|
---|
319 | syntax[base + CTLENDARI] = "CCTL";
|
---|
320 | syntax[base + CTLQUOTEMARK] = "CCTL";
|
---|
321 | #endif /* TARGET_CHAR */
|
---|
322 | }
|
---|
323 |
|
---|
324 |
|
---|
325 | /*
|
---|
326 | * Add entries to the syntax table.
|
---|
327 | */
|
---|
328 |
|
---|
329 | static void
|
---|
330 | add(p, type)
|
---|
331 | char *p, *type;
|
---|
332 | {
|
---|
333 | while (*p)
|
---|
334 | syntax[*p++ + base] = type;
|
---|
335 | }
|
---|
336 |
|
---|
337 |
|
---|
338 |
|
---|
339 | /*
|
---|
340 | * Output the syntax table.
|
---|
341 | */
|
---|
342 |
|
---|
343 | static void
|
---|
344 | print(name)
|
---|
345 | char *name;
|
---|
346 | {
|
---|
347 | int i;
|
---|
348 | int col;
|
---|
349 |
|
---|
350 | fprintf(hfile, "extern const char %s[];\n", name);
|
---|
351 | fprintf(cfile, "const char %s[%d] = {\n", name, size);
|
---|
352 | col = 0;
|
---|
353 | for (i = 0 ; i < size ; i++) {
|
---|
354 | if (i == 0) {
|
---|
355 | fputs(" ", cfile);
|
---|
356 | } else if ((i & 03) == 0) {
|
---|
357 | fputs(",\n ", cfile);
|
---|
358 | col = 0;
|
---|
359 | } else {
|
---|
360 | putc(',', cfile);
|
---|
361 | while (++col < 9 * (i & 03))
|
---|
362 | putc(' ', cfile);
|
---|
363 | }
|
---|
364 | fputs(syntax[i], cfile);
|
---|
365 | col += strlen(syntax[i]);
|
---|
366 | }
|
---|
367 | fputs("\n};\n", cfile);
|
---|
368 | }
|
---|
369 |
|
---|
370 |
|
---|
371 |
|
---|
372 | /*
|
---|
373 | * Output character classification macros (e.g. is_digit). If digits are
|
---|
374 | * contiguous, we can test for them quickly.
|
---|
375 | */
|
---|
376 |
|
---|
377 | static char *macro[] = {
|
---|
378 | "#define is_digit(c)\t((is_type+SYNBASE)[c] & ISDIGIT)",
|
---|
379 | "#define is_alpha(c)\t((c) != UPEOF && ((c) < CTLESC || (c) > CTLENDARI) && isalpha((unsigned char) (c)))",
|
---|
380 | "#define is_name(c)\t((c) != UPEOF && ((c) < CTLESC || (c) > CTLENDARI) && ((c) == '_' || isalpha((unsigned char) (c))))",
|
---|
381 | "#define is_in_name(c)\t((c) != UPEOF && ((c) < CTLESC || (c) > CTLENDARI) && ((c) == '_' || isalnum((unsigned char) (c))))",
|
---|
382 | "#define is_special(c)\t((is_type+SYNBASE)[c] & (ISSPECL|ISDIGIT))",
|
---|
383 | NULL
|
---|
384 | };
|
---|
385 |
|
---|
386 | static void
|
---|
387 | output_type_macros()
|
---|
388 | {
|
---|
389 | char **pp;
|
---|
390 |
|
---|
391 | if (digit_contig)
|
---|
392 | macro[0] = "#define is_digit(c)\t((unsigned)((c) - '0') <= 9)";
|
---|
393 | for (pp = macro ; *pp ; pp++)
|
---|
394 | fprintf(hfile, "%s\n", *pp);
|
---|
395 | if (digit_contig)
|
---|
396 | fputs("#define digit_val(c)\t((c) - '0')\n", hfile);
|
---|
397 | else
|
---|
398 | fputs("#define digit_val(c)\t(digit_value[c])\n", hfile);
|
---|
399 | }
|
---|
400 |
|
---|
401 |
|
---|
402 |
|
---|
403 | /*
|
---|
404 | * Output digit conversion table (if digits are not contiguous).
|
---|
405 | */
|
---|
406 |
|
---|
407 | static void
|
---|
408 | digit_convert()
|
---|
409 | {
|
---|
410 | int maxdigit;
|
---|
411 | static char digit[] = "0123456789";
|
---|
412 | char *p;
|
---|
413 | int i;
|
---|
414 |
|
---|
415 | maxdigit = 0;
|
---|
416 | for (p = digit ; *p ; p++)
|
---|
417 | if (*p > maxdigit)
|
---|
418 | maxdigit = *p;
|
---|
419 | fputs("extern const char digit_value[];\n", hfile);
|
---|
420 | fputs("\n\nconst char digit_value[] = {\n", cfile);
|
---|
421 | for (i = 0 ; i <= maxdigit ; i++) {
|
---|
422 | for (p = digit ; *p && *p != i ; p++);
|
---|
423 | if (*p == '\0')
|
---|
424 | p = digit;
|
---|
425 | fprintf(cfile, " %ld,\n", (long)(p - digit));
|
---|
426 | }
|
---|
427 | fputs("};\n", cfile);
|
---|
428 | }
|
---|