subr.c revision 1.23 1 1.23 rillig /* $NetBSD: subr.c,v 1.23 2023/08/26 14:50:53 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 1.23 rillig __RCSID("$NetBSD: subr.c,v 1.23 2023/08/26 14:50:53 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 1.23 rillig errorp != NULL; 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 1.23 rillig if (string != NULL)
104 1.23 rillig for (int i = 1; *string != '\0'; string++, i++)
105 1.16 dholland if (*string == ch)
106 1.22 rillig return i;
107 1.22 rillig return -1;
108 1.1 cgd }
109 1.5 lukem
110 1.1 cgd /*
111 1.20 andvar * clobber the first occurrence of ch in string by the new character
112 1.1 cgd */
113 1.5 lukem char *
114 1.11 wiz substitute(char *string, char chold, char chnew)
115 1.1 cgd {
116 1.16 dholland char *cp = string;
117 1.1 cgd
118 1.23 rillig if (cp != NULL)
119 1.23 rillig while (*cp != '\0') {
120 1.16 dholland if (*cp == chold) {
121 1.16 dholland *cp = chnew;
122 1.16 dholland break;
123 1.16 dholland }
124 1.16 dholland cp++;
125 1.1 cgd }
126 1.22 rillig return string;
127 1.1 cgd }
128 1.1 cgd
129 1.5 lukem char
130 1.17 dholland lastchar(const char *string)
131 1.1 cgd {
132 1.21 rillig size_t length;
133 1.16 dholland
134 1.16 dholland if (string == NULL)
135 1.22 rillig return '\0';
136 1.1 cgd length = strlen(string);
137 1.1 cgd if (length >= 1)
138 1.22 rillig return string[length-1];
139 1.1 cgd else
140 1.22 rillig return '\0';
141 1.1 cgd }
142 1.1 cgd
143 1.5 lukem char
144 1.17 dholland firstchar(const char *string)
145 1.1 cgd {
146 1.23 rillig if (string != NULL)
147 1.22 rillig return string[0];
148 1.1 cgd else
149 1.22 rillig return '\0';
150 1.1 cgd }
151 1.1 cgd
152 1.5 lukem char
153 1.17 dholland next_lastchar(const char *string)
154 1.1 cgd {
155 1.21 rillig size_t length;
156 1.16 dholland
157 1.16 dholland if (string == NULL)
158 1.22 rillig return '\0';
159 1.1 cgd length = strlen(string);
160 1.1 cgd if (length >= 2)
161 1.22 rillig return string[length - 2];
162 1.1 cgd else
163 1.22 rillig return '\0';
164 1.1 cgd }
165 1.1 cgd
166 1.5 lukem void
167 1.11 wiz clob_last(char *string, char newstuff)
168 1.1 cgd {
169 1.21 rillig if (string != NULL && string[0] != '\0')
170 1.21 rillig string[strlen(string) - 1] = newstuff;
171 1.1 cgd }
172 1.1 cgd
173 1.1 cgd /*
174 1.16 dholland * parse a string that is the result of a format %s(%d)
175 1.16 dholland * return TRUE if this is of the proper format
176 1.1 cgd */
177 1.18 dholland bool
178 1.11 wiz persperdexplode(char *string, char **r_perd, char **r_pers)
179 1.1 cgd {
180 1.16 dholland char *cp;
181 1.21 rillig size_t length = string != NULL ? strlen(string) : 0;
182 1.1 cgd
183 1.18 dholland if (length >= 4 && string[length - 1] == ')') {
184 1.1 cgd for (cp = &string[length - 2];
185 1.18 dholland isdigit((unsigned char)*cp) && *cp != '(';
186 1.1 cgd --cp)
187 1.1 cgd continue;
188 1.16 dholland if (*cp == '(') {
189 1.1 cgd string[length - 1] = '\0'; /* clobber the ) */
190 1.13 itojun *r_perd = strdup(cp+1);
191 1.1 cgd string[length - 1] = ')';
192 1.1 cgd *cp = '\0'; /* clobber the ( */
193 1.13 itojun *r_pers = strdup(string);
194 1.1 cgd *cp = '(';
195 1.18 dholland return true;
196 1.16 dholland }
197 1.1 cgd }
198 1.18 dholland return false;
199 1.1 cgd }
200 1.5 lukem
201 1.15 dholland #if 0 /* unused */
202 1.1 cgd /*
203 1.16 dholland * parse a quoted string that is the result of a format \"%s\"(%d)
204 1.16 dholland * return TRUE if this is of the proper format
205 1.1 cgd */
206 1.23 rillig static bool
207 1.11 wiz qpersperdexplode(char *string, char **r_perd, char **r_pers)
208 1.1 cgd {
209 1.16 dholland char *cp;
210 1.16 dholland int length = 0;
211 1.1 cgd
212 1.1 cgd if (string)
213 1.1 cgd length = strlen(string);
214 1.18 dholland if (length >= 4 && string[length - 1] == ')') {
215 1.1 cgd for (cp = &string[length - 2];
216 1.18 dholland isdigit((unsigned char)*cp) && *cp != '(';
217 1.1 cgd --cp)
218 1.1 cgd continue;
219 1.16 dholland if (*cp == '(' && *(cp - 1) == '"') {
220 1.1 cgd string[length - 1] = '\0';
221 1.13 itojun *r_perd = strdup(cp+1);
222 1.1 cgd string[length - 1] = ')';
223 1.1 cgd *(cp - 1) = '\0'; /* clobber the " */
224 1.13 itojun *r_pers = strdup(string + 1);
225 1.1 cgd *(cp - 1) = '"';
226 1.18 dholland return true;
227 1.16 dholland }
228 1.1 cgd }
229 1.18 dholland return false;
230 1.1 cgd }
231 1.15 dholland #endif /* 0 - unused */
232 1.1 cgd
233 1.16 dholland static char cincomment[] = CINCOMMENT;
234 1.16 dholland static char coutcomment[] = COUTCOMMENT;
235 1.16 dholland static char fincomment[] = FINCOMMENT;
236 1.16 dholland static char foutcomment[] = FOUTCOMMENT;
237 1.16 dholland static char newline[] = NEWLINE;
238 1.16 dholland static char piincomment[] = PIINCOMMENT;
239 1.16 dholland static char pioutcomment[] = PIOUTCOMMENT;
240 1.16 dholland static char lispincomment[] = LISPINCOMMENT;
241 1.16 dholland static char riincomment[] = RIINCOMMENT;
242 1.16 dholland static char rioutcomment[] = RIOUTCOMMENT;
243 1.16 dholland static char troffincomment[] = TROFFINCOMMENT;
244 1.16 dholland static char troffoutcomment[] = TROFFOUTCOMMENT;
245 1.16 dholland static char mod2incomment[] = MOD2INCOMMENT;
246 1.16 dholland static char mod2outcomment[] = MOD2OUTCOMMENT;
247 1.16 dholland
248 1.16 dholland struct lang_desc lang_table[] = {
249 1.16 dholland { /*INUNKNOWN 0*/ "unknown", cincomment, coutcomment },
250 1.16 dholland { /*INCPP 1*/ "cpp", cincomment, coutcomment },
251 1.16 dholland { /*INCC 2*/ "cc", cincomment, coutcomment },
252 1.16 dholland { /*INAS 3*/ "as", ASINCOMMENT, newline },
253 1.16 dholland { /*INLD 4*/ "ld", cincomment, coutcomment },
254 1.16 dholland { /*INLINT 5*/ "lint", cincomment, coutcomment },
255 1.16 dholland { /*INF77 6*/ "f77", fincomment, foutcomment },
256 1.16 dholland { /*INPI 7*/ "pi", piincomment, pioutcomment },
257 1.16 dholland { /*INPC 8*/ "pc", piincomment, pioutcomment },
258 1.16 dholland { /*INFRANZ 9*/ "franz", lispincomment, newline },
259 1.16 dholland { /*INLISP 10*/ "lisp", lispincomment, newline },
260 1.16 dholland { /*INVAXIMA 11*/ "vaxima", lispincomment, newline },
261 1.16 dholland { /*INRATFOR 12*/ "ratfor", fincomment, foutcomment },
262 1.16 dholland { /*INLEX 13*/ "lex", cincomment, coutcomment },
263 1.16 dholland { /*INYACC 14*/ "yacc", cincomment, coutcomment },
264 1.16 dholland { /*INAPL 15*/ "apl", ".lm", newline },
265 1.16 dholland { /*INMAKE 16*/ "make", ASINCOMMENT, newline },
266 1.16 dholland { /*INRI 17*/ "ri", riincomment, rioutcomment },
267 1.16 dholland { /*INTROFF 18*/ "troff", troffincomment, troffoutcomment },
268 1.16 dholland { /*INMOD2 19*/ "mod2", mod2incomment, mod2outcomment },
269 1.16 dholland { 0, 0, 0 }
270 1.1 cgd };
271 1.1 cgd
272 1.5 lukem void
273 1.18 dholland printerrors(bool look_at_subclass, int errorc, Eptr errorv[])
274 1.1 cgd {
275 1.16 dholland int i;
276 1.16 dholland Eptr errorp;
277 1.1 cgd
278 1.16 dholland for (errorp = errorv[i = 0]; i < errorc; errorp = errorv[++i]) {
279 1.1 cgd if (errorp->error_e_class == C_IGNORE)
280 1.1 cgd continue;
281 1.1 cgd if (look_at_subclass && errorp->error_s_class == C_DUPL)
282 1.1 cgd continue;
283 1.1 cgd printf("Error %d, (%s error) [%s], text = \"",
284 1.1 cgd i,
285 1.1 cgd class_table[errorp->error_e_class],
286 1.1 cgd lang_table[errorp->error_language].lang_name);
287 1.1 cgd wordvprint(stdout,errorp->error_lgtext,errorp->error_text);
288 1.1 cgd printf("\"\n");
289 1.1 cgd }
290 1.1 cgd }
291 1.1 cgd
292 1.5 lukem void
293 1.11 wiz wordvprint(FILE *fyle, int wordc, char **wordv)
294 1.1 cgd {
295 1.16 dholland int i;
296 1.17 dholland const char *sep = "";
297 1.1 cgd
298 1.16 dholland for (i = 0; i < wordc; i++)
299 1.23 rillig if (wordv[i] != NULL) {
300 1.1 cgd fprintf(fyle, "%s%s",sep,wordv[i]);
301 1.1 cgd sep = " ";
302 1.1 cgd }
303 1.1 cgd }
304 1.1 cgd
305 1.1 cgd /*
306 1.16 dholland * Given a string, parse it into a number of words, and build
307 1.16 dholland * a wordc wordv combination pointing into it.
308 1.1 cgd */
309 1.5 lukem void
310 1.11 wiz wordvbuild(char *string, int *r_wordc, char ***r_wordv)
311 1.1 cgd {
312 1.16 dholland char *cp;
313 1.16 dholland char **wordv;
314 1.16 dholland int wordcount;
315 1.16 dholland int wordindex;
316 1.1 cgd
317 1.23 rillig for (wordcount = 0, cp = string; *cp != '\0'; wordcount++) {
318 1.23 rillig while (isspace((unsigned char)*cp))
319 1.1 cgd cp++;
320 1.8 christos if (*cp == '\0')
321 1.1 cgd break;
322 1.23 rillig while (*cp != '\0' && !isspace((unsigned char)*cp))
323 1.1 cgd cp++;
324 1.1 cgd }
325 1.18 dholland wordv = Calloc(wordcount + 1, sizeof (char *));
326 1.23 rillig for (cp=string, wordindex=0; wordcount > 0; wordindex++, --wordcount) {
327 1.23 rillig while (isspace((unsigned char)*cp))
328 1.1 cgd cp++;
329 1.8 christos if (*cp == '\0')
330 1.1 cgd break;
331 1.1 cgd wordv[wordindex] = cp;
332 1.23 rillig while (*cp != '\0' && !isspace((unsigned char)*cp))
333 1.1 cgd cp++;
334 1.1 cgd *cp++ = '\0';
335 1.1 cgd }
336 1.1 cgd if (wordcount != 0)
337 1.5 lukem errx(6, "Initial miscount of the number of words in a line");
338 1.8 christos wordv[wordindex] = NULL;
339 1.1 cgd #ifdef FULLDEBUG
340 1.1 cgd for (wordcount = 0; wordcount < wordindex; wordcount++)
341 1.1 cgd printf("Word %d = \"%s\"\n", wordcount, wordv[wordcount]);
342 1.1 cgd printf("\n");
343 1.1 cgd #endif
344 1.1 cgd *r_wordc = wordindex;
345 1.1 cgd *r_wordv = wordv;
346 1.1 cgd }
347 1.5 lukem
348 1.1 cgd /*
349 1.16 dholland * Compare two 0 based wordvectors
350 1.1 cgd */
351 1.5 lukem int
352 1.11 wiz wordvcmp(char **wordv1, int wordc, char **wordv2)
353 1.1 cgd {
354 1.5 lukem int i;
355 1.16 dholland int back;
356 1.5 lukem
357 1.16 dholland for (i = 0; i < wordc; i++) {
358 1.8 christos if (wordv1[i] == NULL || wordv2[i] == NULL)
359 1.22 rillig return -1;
360 1.10 christos if ((back = strcmp(wordv1[i], wordv2[i])) != 0)
361 1.22 rillig return back;
362 1.1 cgd }
363 1.22 rillig return 0; /* they are equal */
364 1.1 cgd }
365 1.16 dholland
366 1.1 cgd /*
367 1.16 dholland * splice a 0 basedword vector onto the tail of a
368 1.16 dholland * new wordv, allowing the first emptyhead slots to be empty
369 1.1 cgd */
370 1.16 dholland char **
371 1.11 wiz wordvsplice(int emptyhead, int wordc, char **wordv)
372 1.1 cgd {
373 1.16 dholland char **nwordv;
374 1.16 dholland int nwordc = emptyhead + wordc;
375 1.16 dholland int i;
376 1.1 cgd
377 1.18 dholland nwordv = Calloc(nwordc, sizeof (char *));
378 1.1 cgd for (i = 0; i < emptyhead; i++)
379 1.8 christos nwordv[i] = NULL;
380 1.16 dholland for (i = emptyhead; i < nwordc; i++) {
381 1.1 cgd nwordv[i] = wordv[i-emptyhead];
382 1.1 cgd }
383 1.22 rillig return nwordv;
384 1.1 cgd }
385 1.5 lukem
386 1.1 cgd /*
387 1.18 dholland * plural and verb forms
388 1.1 cgd */
389 1.17 dholland static const char *S = "s";
390 1.17 dholland static const char *N = "";
391 1.5 lukem
392 1.17 dholland const char *
393 1.11 wiz plural(int n)
394 1.1 cgd {
395 1.22 rillig return n > 1 ? S : N;
396 1.1 cgd }
397 1.5 lukem
398 1.17 dholland const char *
399 1.11 wiz verbform(int n)
400 1.1 cgd {
401 1.22 rillig return n > 1 ? N : S;
402 1.1 cgd }
403