1/* sub.c: This file contains the substitution routines for the ed
2   line editor */
3/*-
4 * Copyright (c) 1993 Andrew Moore, Talke Studio.
5 * All rights reserved.
6 *
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
9 * are met:
10 * 1. Redistributions of source code must retain the above copyright
11 *    notice, this list of conditions and the following disclaimer.
12 * 2. Redistributions in binary form must reproduce the above copyright
13 *    notice, this list of conditions and the following disclaimer in the
14 *    documentation and/or other materials provided with the distribution.
15 *
16 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
17 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
18 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
19 * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
20 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
21 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
22 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
23 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
24 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
25 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
26 * SUCH DAMAGE.
27 */
28
29#include <sys/cdefs.h>
30__FBSDID("$FreeBSD: src/bin/ed/sub.c,v 1.15 2002/06/30 05:13:53 obrien Exp $");
31
32#include "ed.h"
33
34
35char *rhbuf;			/* rhs substitution buffer */
36int rhbufsz;			/* rhs substitution buffer size */
37int rhbufi;			/* rhs substitution buffer index */
38
39/* extract_subst_tail: extract substitution tail from the command buffer */
40int
41extract_subst_tail(int *flagp, long *np)
42{
43	char delimiter;
44
45	*flagp = *np = 0;
46	if ((delimiter = *ibufp) == '\n') {
47		rhbufi = 0;
48		*flagp = GPR;
49		return 0;
50	} else if (extract_subst_template() == NULL)
51		return  ERR;
52	else if (*ibufp == '\n') {
53		*flagp = GPR;
54		return 0;
55	} else if (*ibufp == delimiter)
56		ibufp++;
57	if ('1' <= *ibufp && *ibufp <= '9') {
58		STRTOL(*np, ibufp);
59		return 0;
60	} else if (*ibufp == 'g') {
61		ibufp++;
62		*flagp = GSG;
63		return 0;
64	}
65	return 0;
66}
67
68
69/* extract_subst_template: return pointer to copy of substitution template
70   in the command buffer */
71char *
72extract_subst_template(void)
73{
74	int n = 0;
75	int i = 0;
76	char c;
77	char delimiter = *ibufp++;
78
79	if (*ibufp == '%' && *(ibufp + 1) == delimiter) {
80		ibufp++;
81		if (!rhbuf)
82			errmsg = "no previous substitution";
83		return rhbuf;
84	}
85	while (*ibufp != delimiter) {
86		REALLOC(rhbuf, rhbufsz, i + 2, NULL);
87		if ((c = rhbuf[i++] = *ibufp++) == '\n' && *ibufp == '\0') {
88			i--, ibufp--;
89			break;
90		} else if (c != '\\')
91			;
92		else if ((rhbuf[i++] = *ibufp++) != '\n')
93			;
94		else if (!isglobal) {
95			while ((n = get_tty_line()) == 0 ||
96			    (n > 0 && ibuf[n - 1] != '\n'))
97				clearerr(stdin);
98			if (n < 0)
99				return NULL;
100		}
101	}
102	REALLOC(rhbuf, rhbufsz, i + 1, NULL);
103	rhbuf[rhbufi = i] = '\0';
104	return  rhbuf;
105}
106
107
108char *rbuf;			/* substitute_matching_text buffer */
109int rbufsz;			/* substitute_matching_text buffer size */
110
111/* search_and_replace: for each line in a range, change text matching a pattern
112   according to a substitution template; return status  */
113int
114search_and_replace(pattern_t *pat, int gflag, int kth)
115{
116	undo_t *up;
117	const char *txt;
118	const char *eot;
119	long lc;
120	long xa = current_addr;
121	int nsubs = 0;
122	line_t *lp;
123	int len;
124
125	current_addr = first_addr - 1;
126	for (lc = 0; lc <= second_addr - first_addr; lc++) {
127		lp = get_addressed_line_node(++current_addr);
128		if ((len = substitute_matching_text(pat, lp, gflag, kth)) < 0)
129			return ERR;
130		else if (len) {
131			up = NULL;
132			if (delete_lines(current_addr, current_addr) < 0)
133				return ERR;
134			txt = rbuf;
135			eot = rbuf + len;
136			SPL1();
137			do {
138				if ((txt = put_sbuf_line(txt)) == NULL) {
139					SPL0();
140					return ERR;
141				} else if (up)
142					up->t = get_addressed_line_node(current_addr);
143				else if ((up = push_undo_stack(UADD,
144				    current_addr, current_addr)) == NULL) {
145					SPL0();
146					return ERR;
147				}
148			} while (txt != eot);
149			SPL0();
150			nsubs++;
151			/* 3751351 */
152			replace_marks(lp, get_addressed_line_node(current_addr));
153			xa = current_addr;
154		}
155	}
156	current_addr = xa;
157	if  (nsubs == 0 && !(gflag & GLB)) {
158		errmsg = "no match";
159		return ERR;
160	} else if ((gflag & (GPR | GLS | GNP)) &&
161	    display_lines(current_addr, current_addr, gflag) < 0)
162		return ERR;
163	return 0;
164}
165
166
167/* substitute_matching_text: replace text matched by a pattern according to
168   a substitution template; return pointer to the modified text */
169int
170substitute_matching_text(pattern_t *pat, line_t *lp, int gflag, int kth)
171{
172	int off = 0;
173	int changed = 0;
174	int matchno = 0;
175	int i = 0;
176	regmatch_t rm[SE_MAX];
177	char *txt;
178	char *eot;
179
180	if ((txt = get_sbuf_line(lp)) == NULL)
181		return ERR;
182	if (isbinary)
183		NUL_TO_NEWLINE(txt, lp->len);
184	eot = txt + lp->len;
185	if (!regexec(pat, txt, SE_MAX, rm, 0)) {
186		do {
187			if (!kth || kth == ++matchno) {
188				changed++;
189				i = rm[0].rm_so;
190				REALLOC(rbuf, rbufsz, off + i, ERR);
191				if (isbinary)
192					NEWLINE_TO_NUL(txt, rm[0].rm_eo);
193				memcpy(rbuf + off, txt, i);
194				off += i;
195				if ((off = apply_subst_template(txt, rm, off,
196				    pat->re_nsub)) < 0)
197					return ERR;
198			} else {
199				i = rm[0].rm_eo;
200				REALLOC(rbuf, rbufsz, off + i, ERR);
201				if (isbinary)
202					NEWLINE_TO_NUL(txt, i);
203				memcpy(rbuf + off, txt, i);
204				off += i;
205			}
206			txt += rm[0].rm_eo;
207		} while (*txt &&
208                        (!changed || ((gflag & GSG) && rm[0].rm_eo)) &&
209		        !regexec(pat, txt, SE_MAX, rm, REG_NOTBOL));
210		i = eot - txt;
211		REALLOC(rbuf, rbufsz, off + i + 2, ERR);
212		if (i > 0 && !rm[0].rm_eo && (gflag & GSG)) {
213			errmsg = "infinite substitution loop";
214			return  ERR;
215		}
216		if (isbinary)
217			NEWLINE_TO_NUL(txt, i);
218		memcpy(rbuf + off, txt, i);
219		memcpy(rbuf + off + i, "\n", 2);
220	}
221	return changed ? off + i + 1 : 0;
222}
223
224
225/* apply_subst_template: modify text according to a substitution template;
226   return offset to end of modified text */
227int
228apply_subst_template(const char *boln, regmatch_t *rm, int off, int re_nsub)
229{
230	int j = 0;
231	int k = 0;
232	int n;
233	char *sub = rhbuf;
234
235	for (; sub - rhbuf < rhbufi; sub++)
236		if (*sub == '&') {
237			j = rm[0].rm_so;
238			k = rm[0].rm_eo;
239			REALLOC(rbuf, rbufsz, off + k - j, ERR);
240			while (j < k)
241				rbuf[off++] = boln[j++];
242		} else if (*sub == '\\' && '1' <= *++sub && *sub <= '9' &&
243		    (n = *sub - '0') <= re_nsub) {
244			j = rm[n].rm_so;
245			k = rm[n].rm_eo;
246			REALLOC(rbuf, rbufsz, off + k - j, ERR);
247			while (j < k)
248				rbuf[off++] = boln[j++];
249		} else {
250			REALLOC(rbuf, rbufsz, off + 1, ERR);
251			rbuf[off++] = *sub;
252		}
253	REALLOC(rbuf, rbufsz, off + 1, ERR);
254	rbuf[off] = '\0';
255	return off;
256}
257