1/*	$NetBSD: file.c,v 1.5 2011/02/16 18:35:39 joerg Exp $	*/
2/*	$FreeBSD$	*/
3/*	$OpenBSD: file.c,v 1.11 2010/07/02 20:48:48 nicm Exp $	*/
4
5/*-
6 * SPDX-License-Identifier: BSD-2-Clause-FreeBSD
7 *
8 * Copyright (c) 1999 James Howard and Dag-Erling Co��dan Sm��rgrav
9 * Copyright (C) 2008-2010 Gabor Kovesdan <gabor@FreeBSD.org>
10 * Copyright (C) 2010 Dimitry Andric <dimitry@andric.com>
11 * All rights reserved.
12 *
13 * Redistribution and use in source and binary forms, with or without
14 * modification, are permitted provided that the following conditions
15 * are met:
16 * 1. Redistributions of source code must retain the above copyright
17 *    notice, this list of conditions and the following disclaimer.
18 * 2. Redistributions in binary form must reproduce the above copyright
19 *    notice, this list of conditions and the following disclaimer in the
20 *    documentation and/or other materials provided with the distribution.
21 *
22 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
23 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
24 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
25 * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
26 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
27 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
28 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
29 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
30 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
31 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32 * SUCH DAMAGE.
33 */
34
35#include <sys/cdefs.h>
36__FBSDID("$FreeBSD$");
37
38#include <sys/param.h>
39#include <sys/mman.h>
40#include <sys/stat.h>
41#include <sys/types.h>
42
43#include <err.h>
44#include <errno.h>
45#include <fcntl.h>
46#include <stddef.h>
47#include <stdlib.h>
48#include <string.h>
49#include <unistd.h>
50#include <wchar.h>
51#include <wctype.h>
52
53#include "grep.h"
54
55#define	MAXBUFSIZ	(32 * 1024)
56#define	LNBUFBUMP	80
57
58static char *buffer;
59static char *bufpos;
60static size_t bufrem;
61static size_t fsiz;
62
63static char *lnbuf;
64static size_t lnbuflen;
65
66static inline int
67grep_refill(struct file *f)
68{
69	ssize_t nr;
70
71	if (filebehave == FILE_MMAP)
72		return (0);
73
74	bufpos = buffer;
75	bufrem = 0;
76
77	nr = read(f->fd, buffer, MAXBUFSIZ);
78	if (nr < 0)
79		return (-1);
80
81	bufrem = nr;
82	return (0);
83}
84
85static inline int
86grep_lnbufgrow(size_t newlen)
87{
88
89	if (lnbuflen < newlen) {
90		lnbuf = grep_realloc(lnbuf, newlen);
91		lnbuflen = newlen;
92	}
93
94	return (0);
95}
96
97char *
98grep_fgetln(struct file *f, struct parsec *pc)
99{
100	char *p;
101	size_t len;
102	size_t off;
103	ptrdiff_t diff;
104
105	/* Fill the buffer, if necessary */
106	if (bufrem == 0 && grep_refill(f) != 0)
107		goto error;
108
109	if (bufrem == 0) {
110		/* Return zero length to indicate EOF */
111		pc->ln.len= 0;
112		return (bufpos);
113	}
114
115	/* Look for a newline in the remaining part of the buffer */
116	if ((p = memchr(bufpos, fileeol, bufrem)) != NULL) {
117		++p; /* advance over newline */
118		len = p - bufpos;
119		if (grep_lnbufgrow(len + 1))
120			goto error;
121		memcpy(lnbuf, bufpos, len);
122		bufrem -= len;
123		bufpos = p;
124		pc->ln.len = len;
125		lnbuf[len] = '\0';
126		return (lnbuf);
127	}
128
129	/* We have to copy the current buffered data to the line buffer */
130	for (len = bufrem, off = 0; ; len += bufrem) {
131		/* Make sure there is room for more data */
132		if (grep_lnbufgrow(len + LNBUFBUMP))
133			goto error;
134		memcpy(lnbuf + off, bufpos, len - off);
135		/* With FILE_MMAP, this is EOF; there's no more to refill */
136		if (filebehave == FILE_MMAP) {
137			bufrem -= len;
138			break;
139		}
140		off = len;
141		/* Fetch more to try and find EOL/EOF */
142		if (grep_refill(f) != 0)
143			goto error;
144		if (bufrem == 0)
145			/* EOF: return partial line */
146			break;
147		if ((p = memchr(bufpos, fileeol, bufrem)) == NULL)
148			continue;
149		/* got it: finish up the line (like code above) */
150		++p;
151		diff = p - bufpos;
152		len += diff;
153		if (grep_lnbufgrow(len + 1))
154		    goto error;
155		memcpy(lnbuf + off, bufpos, diff);
156		bufrem -= diff;
157		bufpos = p;
158		break;
159	}
160	pc->ln.len = len;
161	lnbuf[len] = '\0';
162	return (lnbuf);
163
164error:
165	pc->ln.len = 0;
166	return (NULL);
167}
168
169/*
170 * Opens a file for processing.
171 */
172struct file *
173grep_open(const char *path)
174{
175	struct file *f;
176
177	f = grep_malloc(sizeof *f);
178	memset(f, 0, sizeof *f);
179	if (path == NULL) {
180		/* Processing stdin implies --line-buffered. */
181		lbflag = true;
182		f->fd = STDIN_FILENO;
183	} else if ((f->fd = open(path, O_RDONLY)) == -1)
184		goto error1;
185
186	if (filebehave == FILE_MMAP) {
187		struct stat st;
188
189		if ((fstat(f->fd, &st) == -1) || (st.st_size > OFF_MAX) ||
190		    (!S_ISREG(st.st_mode)))
191			filebehave = FILE_STDIO;
192		else {
193			int flags = MAP_PRIVATE | MAP_NOCORE | MAP_NOSYNC;
194#ifdef MAP_PREFAULT_READ
195			flags |= MAP_PREFAULT_READ;
196#endif
197			fsiz = st.st_size;
198			buffer = mmap(NULL, fsiz, PROT_READ, flags,
199			     f->fd, (off_t)0);
200			if (buffer == MAP_FAILED)
201				filebehave = FILE_STDIO;
202			else {
203				bufrem = st.st_size;
204				bufpos = buffer;
205				madvise(buffer, st.st_size, MADV_SEQUENTIAL);
206			}
207		}
208	}
209
210	if ((buffer == NULL) || (buffer == MAP_FAILED))
211		buffer = grep_malloc(MAXBUFSIZ);
212
213	/* Fill read buffer, also catches errors early */
214	if (bufrem == 0 && grep_refill(f) != 0)
215		goto error2;
216
217	/* Check for binary stuff, if necessary */
218	if (binbehave != BINFILE_TEXT && fileeol != '\0' &&
219	    memchr(bufpos, '\0', bufrem) != NULL)
220		f->binary = true;
221
222	return (f);
223
224error2:
225	close(f->fd);
226error1:
227	free(f);
228	return (NULL);
229}
230
231/*
232 * Closes a file.
233 */
234void
235grep_close(struct file *f)
236{
237
238	close(f->fd);
239
240	/* Reset read buffer and line buffer */
241	if (filebehave == FILE_MMAP) {
242		munmap(buffer, fsiz);
243		buffer = NULL;
244	}
245	bufpos = buffer;
246	bufrem = 0;
247
248	free(lnbuf);
249	lnbuf = NULL;
250	lnbuflen = 0;
251}
252