file.c revision 210389 
1/*	$OpenBSD: file.c,v 1.11 2010/07/02 20:48:48 nicm Exp $	*/
2
3/*-
4 * Copyright (c) 1999 James Howard and Dag-Erling Co�dan Sm�rgrav
5 * Copyright (C) 2008-2009 Gabor Kovesdan <gabor@FreeBSD.org>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 * 1. Redistributions of source code must retain the above copyright
12 *    notice, this list of conditions and the following disclaimer.
13 * 2. Redistributions in binary form must reproduce the above copyright
14 *    notice, this list of conditions and the following disclaimer in the
15 *    documentation and/or other materials provided with the distribution.
16 *
17 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
18 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20 * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
21 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
22 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
23 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
24 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
25 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
26 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
27 * SUCH DAMAGE.
28 */
29
30#include <sys/cdefs.h>
31__FBSDID("$FreeBSD: head/usr.bin/grep/file.c 210389 2010-07-22 19:11:57Z gabor $");
32
33#include <sys/param.h>
34#include <sys/types.h>
35#include <sys/stat.h>
36
37#include <bzlib.h>
38#include <err.h>
39#include <errno.h>
40#include <stdio.h>
41#include <stdlib.h>
42#include <string.h>
43#include <unistd.h>
44#include <wchar.h>
45#include <wctype.h>
46#include <zlib.h>
47
48#include "grep.h"
49
50static char	 fname[MAXPATHLEN];	/* file name */
51
52#define		 MAXBUFSIZ	(16 * 1024)
53#define		 PREREAD_M	0.2
54
55/* Some global variables for the buffering and reading. */
56static char	*lnbuf;
57static size_t	 lnbuflen;
58static unsigned char *binbuf;
59static int	 binbufsiz;
60unsigned char	*binbufptr;
61static int	 bzerr;
62
63#define iswbinary(ch)	(!iswspace((ch)) && iswcntrl((ch)) && \
64			    (ch != L'\b') && (ch != L'\0'))
65
66/*
67 * Returns a single character according to the file type.
68 * Returns -1 on failure.
69 */
70int
71grep_fgetc(struct file *f)
72{
73	unsigned char c;
74
75	switch (filebehave) {
76	case FILE_STDIO:
77		return (fgetc(f->f));
78	case FILE_GZIP:
79		return (gzgetc(f->gzf));
80	case FILE_BZIP:
81		BZ2_bzRead(&bzerr, f->bzf, &c, 1);
82		if (bzerr == BZ_STREAM_END)
83			return (-1);
84		else if (bzerr != BZ_SEQUENCE_ERROR && bzerr != BZ_OK)
85			errx(2, "%s", getstr(2));
86		return (c);
87	}
88	return (-1);
89}
90
91/*
92 * Returns true if the file position is a EOF, returns false
93 * otherwise.
94 */
95int
96grep_feof(struct file *f)
97{
98
99	switch (filebehave) {
100	case FILE_STDIO:
101		return (feof(f->f));
102	case FILE_GZIP:
103		return (gzeof(f->gzf));
104	case FILE_BZIP:
105		return (bzerr == BZ_STREAM_END);
106	}
107	return (1);
108}
109
110/*
111 * At the first call, fills in an internal buffer and checks if the given
112 * file is a binary file and sets the binary flag accordingly.  Then returns
113 * a single line and sets len to the length of the returned line.
114 * At any other call returns a single line either from the internal buffer
115 * or from the file if the buffer is exhausted and sets len to the length
116 * of the line.
117 */
118char *
119grep_fgetln(struct file *f, size_t *len)
120{
121	struct stat st;
122	size_t bufsiz, i = 0;
123	int ch = 0;
124
125	/* Fill in the buffer if it is empty. */
126	if (binbufptr == NULL) {
127
128		/* Only pre-read to the buffer if we need the binary check. */
129		if (binbehave != BINFILE_TEXT) {
130			if (f->stdin)
131				st.st_size = MAXBUFSIZ;
132			else if (stat(fname, &st) != 0)
133				err(2, NULL);
134
135			bufsiz = (MAXBUFSIZ > (st.st_size * PREREAD_M)) ?
136			    (st.st_size / 2) : MAXBUFSIZ;
137
138			binbuf = grep_malloc(sizeof(char) * bufsiz);
139
140			while (i < bufsiz) {
141				ch = grep_fgetc(f);
142				if (ch == EOF)
143					break;
144				binbuf[i++] = ch;
145			}
146
147			f->binary = memchr(binbuf, (filebehave != FILE_GZIP) ?
148			    '\0' : '\200', i - 1) != NULL;
149		}
150		binbufsiz = i;
151		binbufptr = binbuf;
152	}
153
154	/* Read a line whether from the buffer or from the file itself. */
155	for (i = 0; !(grep_feof(f) &&
156	    (binbufptr == &binbuf[binbufsiz])); i++) {
157		if (binbufptr == &binbuf[binbufsiz]) {
158			ch = grep_fgetc(f);
159		} else {
160			ch = binbufptr[0];
161			binbufptr++;
162		}
163		if (i >= lnbuflen) {
164			lnbuflen *= 2;
165			lnbuf = grep_realloc(lnbuf, ++lnbuflen);
166		}
167		if ((ch == '\n') || (ch == EOF)) {
168			lnbuf[i] = '\0';
169			break;
170		} else
171			lnbuf[i] = ch;
172	}
173	if (grep_feof(f) && (i == 0) && (ch != '\n'))
174		return (NULL);
175	*len = i;
176	return (lnbuf);
177}
178
179/*
180 * Opens the standard input for processing.
181 */
182struct file *
183grep_stdin_open(void)
184{
185	struct file *f;
186
187	snprintf(fname, sizeof fname, "%s", getstr(1));
188
189	f = grep_malloc(sizeof *f);
190
191	if ((f->f = fdopen(STDIN_FILENO, "r")) != NULL) {
192		f->stdin = true;
193		return (f);
194	}
195
196	free(f);
197	return (NULL);
198}
199
200/*
201 * Opens a normal, a gzipped or a bzip2 compressed file for processing.
202 */
203struct file *
204grep_open(const char *path)
205{
206	struct file *f;
207
208	snprintf(fname, sizeof fname, "%s", path);
209
210	f = grep_malloc(sizeof *f);
211
212	f->stdin = false;
213	switch (filebehave) {
214	case FILE_STDIO:
215		if ((f->f = fopen(path, "r")) != NULL)
216			return (f);
217		break;
218	case FILE_GZIP:
219		if ((f->gzf = gzopen(fname, "r")) != NULL)
220			return (f);
221		break;
222	case FILE_BZIP:
223		if ((f->bzf = BZ2_bzopen(fname, "r")) != NULL)
224			return (f);
225		break;
226	}
227
228	free(f);
229	return (NULL);
230}
231
232/*
233 * Closes a normal, a gzipped or a bzip2 compressed file.
234 */
235void
236grep_close(struct file *f)
237{
238
239	switch (filebehave) {
240	case FILE_STDIO:
241		fclose(f->f);
242		break;
243	case FILE_GZIP:
244		gzclose(f->gzf);
245		break;
246	case FILE_BZIP:
247		BZ2_bzclose(f->bzf);
248		break;
249	}
250
251	/* Reset read buffer for the file we are closing */
252	binbufptr = NULL;
253	free(binbuf);
254
255}
256