subr.c revision 1.22 1 1.22 rillig /* $NetBSD: subr.c,v 1.22 2023/08/26 12:43:28 rillig Exp $ */
2 1.3 jtc
3 1.1 cgd /*
4 1.3 jtc * Copyright (c) 1980, 1993
5 1.3 jtc * The Regents of the University of California. All rights reserved.
6 1.1 cgd *
7 1.1 cgd * Redistribution and use in source and binary forms, with or without
8 1.1 cgd * modification, are permitted provided that the following conditions
9 1.1 cgd * are met:
10 1.1 cgd * 1. Redistributions of source code must retain the above copyright
11 1.1 cgd * notice, this list of conditions and the following disclaimer.
12 1.1 cgd * 2. Redistributions in binary form must reproduce the above copyright
13 1.1 cgd * notice, this list of conditions and the following disclaimer in the
14 1.1 cgd * documentation and/or other materials provided with the distribution.
15 1.14 agc * 3. Neither the name of the University nor the names of its contributors
16 1.1 cgd * may be used to endorse or promote products derived from this software
17 1.1 cgd * without specific prior written permission.
18 1.1 cgd *
19 1.1 cgd * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
20 1.1 cgd * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
21 1.1 cgd * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
22 1.1 cgd * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
23 1.1 cgd * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
24 1.1 cgd * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
25 1.1 cgd * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
26 1.1 cgd * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
27 1.1 cgd * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
28 1.1 cgd * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
29 1.1 cgd * SUCH DAMAGE.
30 1.1 cgd */
31 1.1 cgd
32 1.5 lukem #include <sys/cdefs.h>
33 1.1 cgd #ifndef lint
34 1.3 jtc #if 0
35 1.3 jtc static char sccsid[] = "@(#)subr.c 8.1 (Berkeley) 6/6/93";
36 1.3 jtc #endif
37 1.22 rillig __RCSID("$NetBSD: subr.c,v 1.22 2023/08/26 12:43:28 rillig Exp $");
38 1.1 cgd #endif /* not lint */
39 1.1 cgd
40 1.5 lukem #include <ctype.h>
41 1.5 lukem #include <err.h>
42 1.1 cgd #include <stdio.h>
43 1.1 cgd #include <stdlib.h>
44 1.1 cgd #include <string.h>
45 1.1 cgd #include "error.h"
46 1.15 dholland
47 1.1 cgd /*
48 1.16 dholland * Arrayify a list of rules
49 1.1 cgd */
50 1.5 lukem void
51 1.11 wiz arrayify(int *e_length, Eptr **e_array, Eptr header)
52 1.1 cgd {
53 1.16 dholland Eptr errorp;
54 1.16 dholland Eptr *array;
55 1.16 dholland int listlength;
56 1.16 dholland int listindex;
57 1.1 cgd
58 1.1 cgd for (errorp = header, listlength = 0;
59 1.1 cgd errorp; errorp = errorp->error_next, listlength++)
60 1.1 cgd continue;
61 1.18 dholland array = Calloc(listlength+1, sizeof (Eptr));
62 1.16 dholland for (listindex = 0, errorp = header;
63 1.1 cgd listindex < listlength;
64 1.16 dholland listindex++, errorp = errorp->error_next) {
65 1.1 cgd array[listindex] = errorp;
66 1.1 cgd errorp->error_position = listindex;
67 1.1 cgd }
68 1.8 christos array[listindex] = NULL;
69 1.1 cgd *e_length = listlength;
70 1.1 cgd *e_array = array;
71 1.1 cgd }
72 1.1 cgd
73 1.18 dholland void *
74 1.18 dholland Calloc(size_t nelements, size_t size)
75 1.1 cgd {
76 1.18 dholland void *back;
77 1.16 dholland
78 1.18 dholland back = calloc(nelements, size);
79 1.18 dholland if (back == NULL)
80 1.5 lukem errx(1, "Ran out of memory.");
81 1.22 rillig return back;
82 1.1 cgd }
83 1.1 cgd
84 1.19 dholland char *
85 1.19 dholland Strdup(const char *s)
86 1.19 dholland {
87 1.19 dholland char *ret;
88 1.19 dholland
89 1.19 dholland ret = strdup(s);
90 1.19 dholland if (ret == NULL) {
91 1.19 dholland errx(1, "Ran out of memory.");
92 1.19 dholland }
93 1.19 dholland return ret;
94 1.19 dholland }
95 1.19 dholland
96 1.1 cgd /*
97 1.16 dholland * find the position of a given character in a string
98 1.16 dholland * (one based)
99 1.1 cgd */
100 1.5 lukem int
101 1.17 dholland position(const char *string, char ch)
102 1.1 cgd {
103 1.16 dholland int i;
104 1.16 dholland
105 1.1 cgd if (string)
106 1.16 dholland for (i=1; *string; string++, i++) {
107 1.16 dholland if (*string == ch)
108 1.22 rillig return i;
109 1.16 dholland }
110 1.22 rillig return -1;
111 1.1 cgd }
112 1.5 lukem
113 1.1 cgd /*
114 1.20 andvar * clobber the first occurrence of ch in string by the new character
115 1.1 cgd */
116 1.5 lukem char *
117 1.11 wiz substitute(char *string, char chold, char chnew)
118 1.1 cgd {
119 1.16 dholland char *cp = string;
120 1.1 cgd
121 1.1 cgd if (cp)
122 1.16 dholland while (*cp) {
123 1.16 dholland if (*cp == chold) {
124 1.16 dholland *cp = chnew;
125 1.16 dholland break;
126 1.16 dholland }
127 1.16 dholland cp++;
128 1.1 cgd }
129 1.22 rillig return string;
130 1.1 cgd }
131 1.1 cgd
132 1.5 lukem char
133 1.17 dholland lastchar(const char *string)
134 1.1 cgd {
135 1.21 rillig size_t length;
136 1.16 dholland
137 1.16 dholland if (string == NULL)
138 1.22 rillig return '\0';
139 1.1 cgd length = strlen(string);
140 1.1 cgd if (length >= 1)
141 1.22 rillig return string[length-1];
142 1.1 cgd else
143 1.22 rillig return '\0';
144 1.1 cgd }
145 1.1 cgd
146 1.5 lukem char
147 1.17 dholland firstchar(const char *string)
148 1.1 cgd {
149 1.1 cgd if (string)
150 1.22 rillig return string[0];
151 1.1 cgd else
152 1.22 rillig return '\0';
153 1.1 cgd }
154 1.1 cgd
155 1.5 lukem char
156 1.17 dholland next_lastchar(const char *string)
157 1.1 cgd {
158 1.21 rillig size_t length;
159 1.16 dholland
160 1.16 dholland if (string == NULL)
161 1.22 rillig return '\0';
162 1.1 cgd length = strlen(string);
163 1.1 cgd if (length >= 2)
164 1.22 rillig return string[length - 2];
165 1.1 cgd else
166 1.22 rillig return '\0';
167 1.1 cgd }
168 1.1 cgd
169 1.5 lukem void
170 1.11 wiz clob_last(char *string, char newstuff)
171 1.1 cgd {
172 1.21 rillig if (string != NULL && string[0] != '\0')
173 1.21 rillig string[strlen(string) - 1] = newstuff;
174 1.1 cgd }
175 1.1 cgd
176 1.1 cgd /*
177 1.16 dholland * parse a string that is the result of a format %s(%d)
178 1.16 dholland * return TRUE if this is of the proper format
179 1.1 cgd */
180 1.18 dholland bool
181 1.11 wiz persperdexplode(char *string, char **r_perd, char **r_pers)
182 1.1 cgd {
183 1.16 dholland char *cp;
184 1.21 rillig size_t length = string != NULL ? strlen(string) : 0;
185 1.1 cgd
186 1.18 dholland if (length >= 4 && string[length - 1] == ')') {
187 1.1 cgd for (cp = &string[length - 2];
188 1.18 dholland isdigit((unsigned char)*cp) && *cp != '(';
189 1.1 cgd --cp)
190 1.1 cgd continue;
191 1.16 dholland if (*cp == '(') {
192 1.1 cgd string[length - 1] = '\0'; /* clobber the ) */
193 1.13 itojun *r_perd = strdup(cp+1);
194 1.1 cgd string[length - 1] = ')';
195 1.1 cgd *cp = '\0'; /* clobber the ( */
196 1.13 itojun *r_pers = strdup(string);
197 1.1 cgd *cp = '(';
198 1.18 dholland return true;
199 1.16 dholland }
200 1.1 cgd }
201 1.18 dholland return false;
202 1.1 cgd }
203 1.5 lukem
204 1.15 dholland #if 0 /* unused */
205 1.1 cgd /*
206 1.16 dholland * parse a quoted string that is the result of a format \"%s\"(%d)
207 1.16 dholland * return TRUE if this is of the proper format
208 1.1 cgd */
209 1.15 dholland static boolean
210 1.11 wiz qpersperdexplode(char *string, char **r_perd, char **r_pers)
211 1.1 cgd {
212 1.16 dholland char *cp;
213 1.16 dholland int length = 0;
214 1.1 cgd
215 1.1 cgd if (string)
216 1.1 cgd length = strlen(string);
217 1.18 dholland if (length >= 4 && string[length - 1] == ')') {
218 1.1 cgd for (cp = &string[length - 2];
219 1.18 dholland isdigit((unsigned char)*cp) && *cp != '(';
220 1.1 cgd --cp)
221 1.1 cgd continue;
222 1.16 dholland if (*cp == '(' && *(cp - 1) == '"') {
223 1.1 cgd string[length - 1] = '\0';
224 1.13 itojun *r_perd = strdup(cp+1);
225 1.1 cgd string[length - 1] = ')';
226 1.1 cgd *(cp - 1) = '\0'; /* clobber the " */
227 1.13 itojun *r_pers = strdup(string + 1);
228 1.1 cgd *(cp - 1) = '"';
229 1.18 dholland return true;
230 1.16 dholland }
231 1.1 cgd }
232 1.18 dholland return false;
233 1.1 cgd }
234 1.15 dholland #endif /* 0 - unused */
235 1.1 cgd
236 1.16 dholland static char cincomment[] = CINCOMMENT;
237 1.16 dholland static char coutcomment[] = COUTCOMMENT;
238 1.16 dholland static char fincomment[] = FINCOMMENT;
239 1.16 dholland static char foutcomment[] = FOUTCOMMENT;
240 1.16 dholland static char newline[] = NEWLINE;
241 1.16 dholland static char piincomment[] = PIINCOMMENT;
242 1.16 dholland static char pioutcomment[] = PIOUTCOMMENT;
243 1.16 dholland static char lispincomment[] = LISPINCOMMENT;
244 1.16 dholland static char riincomment[] = RIINCOMMENT;
245 1.16 dholland static char rioutcomment[] = RIOUTCOMMENT;
246 1.16 dholland static char troffincomment[] = TROFFINCOMMENT;
247 1.16 dholland static char troffoutcomment[] = TROFFOUTCOMMENT;
248 1.16 dholland static char mod2incomment[] = MOD2INCOMMENT;
249 1.16 dholland static char mod2outcomment[] = MOD2OUTCOMMENT;
250 1.16 dholland
251 1.16 dholland struct lang_desc lang_table[] = {
252 1.16 dholland { /*INUNKNOWN 0*/ "unknown", cincomment, coutcomment },
253 1.16 dholland { /*INCPP 1*/ "cpp", cincomment, coutcomment },
254 1.16 dholland { /*INCC 2*/ "cc", cincomment, coutcomment },
255 1.16 dholland { /*INAS 3*/ "as", ASINCOMMENT, newline },
256 1.16 dholland { /*INLD 4*/ "ld", cincomment, coutcomment },
257 1.16 dholland { /*INLINT 5*/ "lint", cincomment, coutcomment },
258 1.16 dholland { /*INF77 6*/ "f77", fincomment, foutcomment },
259 1.16 dholland { /*INPI 7*/ "pi", piincomment, pioutcomment },
260 1.16 dholland { /*INPC 8*/ "pc", piincomment, pioutcomment },
261 1.16 dholland { /*INFRANZ 9*/ "franz", lispincomment, newline },
262 1.16 dholland { /*INLISP 10*/ "lisp", lispincomment, newline },
263 1.16 dholland { /*INVAXIMA 11*/ "vaxima", lispincomment, newline },
264 1.16 dholland { /*INRATFOR 12*/ "ratfor", fincomment, foutcomment },
265 1.16 dholland { /*INLEX 13*/ "lex", cincomment, coutcomment },
266 1.16 dholland { /*INYACC 14*/ "yacc", cincomment, coutcomment },
267 1.16 dholland { /*INAPL 15*/ "apl", ".lm", newline },
268 1.16 dholland { /*INMAKE 16*/ "make", ASINCOMMENT, newline },
269 1.16 dholland { /*INRI 17*/ "ri", riincomment, rioutcomment },
270 1.16 dholland { /*INTROFF 18*/ "troff", troffincomment, troffoutcomment },
271 1.16 dholland { /*INMOD2 19*/ "mod2", mod2incomment, mod2outcomment },
272 1.16 dholland { 0, 0, 0 }
273 1.1 cgd };
274 1.1 cgd
275 1.5 lukem void
276 1.18 dholland printerrors(bool look_at_subclass, int errorc, Eptr errorv[])
277 1.1 cgd {
278 1.16 dholland int i;
279 1.16 dholland Eptr errorp;
280 1.1 cgd
281 1.16 dholland for (errorp = errorv[i = 0]; i < errorc; errorp = errorv[++i]) {
282 1.1 cgd if (errorp->error_e_class == C_IGNORE)
283 1.1 cgd continue;
284 1.1 cgd if (look_at_subclass && errorp->error_s_class == C_DUPL)
285 1.1 cgd continue;
286 1.1 cgd printf("Error %d, (%s error) [%s], text = \"",
287 1.1 cgd i,
288 1.1 cgd class_table[errorp->error_e_class],
289 1.1 cgd lang_table[errorp->error_language].lang_name);
290 1.1 cgd wordvprint(stdout,errorp->error_lgtext,errorp->error_text);
291 1.1 cgd printf("\"\n");
292 1.1 cgd }
293 1.1 cgd }
294 1.1 cgd
295 1.5 lukem void
296 1.11 wiz wordvprint(FILE *fyle, int wordc, char **wordv)
297 1.1 cgd {
298 1.16 dholland int i;
299 1.17 dholland const char *sep = "";
300 1.1 cgd
301 1.16 dholland for (i = 0; i < wordc; i++)
302 1.1 cgd if (wordv[i]) {
303 1.1 cgd fprintf(fyle, "%s%s",sep,wordv[i]);
304 1.1 cgd sep = " ";
305 1.1 cgd }
306 1.1 cgd }
307 1.1 cgd
308 1.1 cgd /*
309 1.16 dholland * Given a string, parse it into a number of words, and build
310 1.16 dholland * a wordc wordv combination pointing into it.
311 1.1 cgd */
312 1.5 lukem void
313 1.11 wiz wordvbuild(char *string, int *r_wordc, char ***r_wordv)
314 1.1 cgd {
315 1.16 dholland char *cp;
316 1.16 dholland char **wordv;
317 1.16 dholland int wordcount;
318 1.16 dholland int wordindex;
319 1.1 cgd
320 1.16 dholland for (wordcount = 0, cp = string; *cp; wordcount++) {
321 1.16 dholland while (*cp && isspace((unsigned char)*cp))
322 1.1 cgd cp++;
323 1.8 christos if (*cp == '\0')
324 1.1 cgd break;
325 1.7 sommerfe while (*cp && !isspace((unsigned char)*cp))
326 1.1 cgd cp++;
327 1.1 cgd }
328 1.18 dholland wordv = Calloc(wordcount + 1, sizeof (char *));
329 1.16 dholland for (cp=string, wordindex=0; wordcount; wordindex++, --wordcount) {
330 1.6 christos while (*cp && isspace((unsigned char)*cp))
331 1.1 cgd cp++;
332 1.8 christos if (*cp == '\0')
333 1.1 cgd break;
334 1.1 cgd wordv[wordindex] = cp;
335 1.16 dholland while (*cp && !isspace((unsigned char)*cp))
336 1.1 cgd cp++;
337 1.1 cgd *cp++ = '\0';
338 1.1 cgd }
339 1.1 cgd if (wordcount != 0)
340 1.5 lukem errx(6, "Initial miscount of the number of words in a line");
341 1.8 christos wordv[wordindex] = NULL;
342 1.1 cgd #ifdef FULLDEBUG
343 1.1 cgd for (wordcount = 0; wordcount < wordindex; wordcount++)
344 1.1 cgd printf("Word %d = \"%s\"\n", wordcount, wordv[wordcount]);
345 1.1 cgd printf("\n");
346 1.1 cgd #endif
347 1.1 cgd *r_wordc = wordindex;
348 1.1 cgd *r_wordv = wordv;
349 1.1 cgd }
350 1.5 lukem
351 1.1 cgd /*
352 1.16 dholland * Compare two 0 based wordvectors
353 1.1 cgd */
354 1.5 lukem int
355 1.11 wiz wordvcmp(char **wordv1, int wordc, char **wordv2)
356 1.1 cgd {
357 1.5 lukem int i;
358 1.16 dholland int back;
359 1.5 lukem
360 1.16 dholland for (i = 0; i < wordc; i++) {
361 1.8 christos if (wordv1[i] == NULL || wordv2[i] == NULL)
362 1.22 rillig return -1;
363 1.10 christos if ((back = strcmp(wordv1[i], wordv2[i])) != 0)
364 1.22 rillig return back;
365 1.1 cgd }
366 1.22 rillig return 0; /* they are equal */
367 1.1 cgd }
368 1.16 dholland
369 1.1 cgd /*
370 1.16 dholland * splice a 0 basedword vector onto the tail of a
371 1.16 dholland * new wordv, allowing the first emptyhead slots to be empty
372 1.1 cgd */
373 1.16 dholland char **
374 1.11 wiz wordvsplice(int emptyhead, int wordc, char **wordv)
375 1.1 cgd {
376 1.16 dholland char **nwordv;
377 1.16 dholland int nwordc = emptyhead + wordc;
378 1.16 dholland int i;
379 1.1 cgd
380 1.18 dholland nwordv = Calloc(nwordc, sizeof (char *));
381 1.1 cgd for (i = 0; i < emptyhead; i++)
382 1.8 christos nwordv[i] = NULL;
383 1.16 dholland for (i = emptyhead; i < nwordc; i++) {
384 1.1 cgd nwordv[i] = wordv[i-emptyhead];
385 1.1 cgd }
386 1.22 rillig return nwordv;
387 1.1 cgd }
388 1.5 lukem
389 1.1 cgd /*
390 1.18 dholland * plural and verb forms
391 1.1 cgd */
392 1.17 dholland static const char *S = "s";
393 1.17 dholland static const char *N = "";
394 1.5 lukem
395 1.17 dholland const char *
396 1.11 wiz plural(int n)
397 1.1 cgd {
398 1.22 rillig return n > 1 ? S : N;
399 1.1 cgd }
400 1.5 lukem
401 1.17 dholland const char *
402 1.11 wiz verbform(int n)
403 1.1 cgd {
404 1.22 rillig return n > 1 ? N : S;
405 1.1 cgd }
406