1 /* $NetBSD: file.c,v 1.5 2011/02/16 18:35:39 joerg Exp $ */
3 /* $OpenBSD: file.c,v 1.11 2010/07/02 20:48:48 nicm Exp $ */
6 * Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
7 * Copyright (C) 2008-2010 Gabor Kovesdan <gabor@FreeBSD.org>
8 * Copyright (C) 2010 Dimitry Andric <dimitry@andric.com>
11 * Redistribution and use in source and binary forms, with or without
12 * modification, are permitted provided that the following conditions
14 * 1. Redistributions of source code must retain the above copyright
15 * notice, this list of conditions and the following disclaimer.
16 * 2. Redistributions in binary form must reproduce the above copyright
17 * notice, this list of conditions and the following disclaimer in the
18 * documentation and/or other materials provided with the distribution.
20 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
21 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
24 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
25 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
26 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
27 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
28 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
29 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
33 #include <sys/cdefs.h>
34 __FBSDID("$FreeBSD$");
36 #include <sys/param.h>
39 #include <sys/types.h>
59 #define MAXBUFSIZ (32 * 1024)
62 static gzFile gzbufdesc;
63 static lzma_stream lstrm = LZMA_STREAM_INIT;
65 static BZFILE* bzbufdesc;
68 static unsigned char *buffer;
69 static unsigned char *bufpos;
73 static unsigned char *lnbuf;
74 static size_t lnbuflen;
77 grep_refill(struct file *f)
81 if (filebehave == FILE_MMAP)
87 if (filebehave == FILE_GZIP) {
88 nr = gzread(gzbufdesc, buffer, MAXBUFSIZ);
90 } else if (filebehave == FILE_BZIP && bzbufdesc != NULL) {
93 nr = BZ2_bzRead(&bzerr, bzbufdesc, buffer, MAXBUFSIZ);
97 /* No problem, nr will be okay */
99 case BZ_DATA_ERROR_MAGIC:
101 * As opposed to gzread(), which simply returns the
102 * plain file data, if it is not in the correct
103 * compressed format, BZ2_bzRead() instead aborts.
105 * So, just restart at the beginning of the file again,
106 * and use plain reads from now on.
108 BZ2_bzReadClose(&bzerr, bzbufdesc);
110 if (lseek(f->fd, 0, SEEK_SET) == -1)
112 nr = read(f->fd, buffer, MAXBUFSIZ);
115 /* Make sure we exit with an error */
119 } else if ((filebehave == FILE_XZ) || (filebehave == FILE_LZMA)) {
120 lzma_action action = LZMA_RUN;
121 uint8_t in_buf[MAXBUFSIZ];
124 ret = (filebehave == FILE_XZ) ?
125 lzma_stream_decoder(&lstrm, UINT64_MAX,
127 lzma_alone_decoder(&lstrm, UINT64_MAX);
132 lstrm.next_out = buffer;
133 lstrm.avail_out = MAXBUFSIZ;
134 lstrm.next_in = in_buf;
135 nr = read(f->fd, in_buf, MAXBUFSIZ);
140 action = LZMA_FINISH;
143 ret = lzma_code(&lstrm, action);
145 if (ret != LZMA_OK && ret != LZMA_STREAM_END)
147 bufrem = MAXBUFSIZ - lstrm.avail_out;
150 nr = read(f->fd, buffer, MAXBUFSIZ);
160 grep_lnbufgrow(size_t newlen)
163 if (lnbuflen < newlen) {
164 lnbuf = grep_realloc(lnbuf, newlen);
172 grep_fgetln(struct file *f, size_t *lenp)
180 /* Fill the buffer, if necessary */
181 if (bufrem == 0 && grep_refill(f) != 0)
185 /* Return zero length to indicate EOF */
190 /* Look for a newline in the remaining part of the buffer */
191 if ((p = memchr(bufpos, '\n', bufrem)) != NULL) {
192 ++p; /* advance over newline */
201 /* We have to copy the current buffered data to the line buffer */
202 for (len = bufrem, off = 0; ; len += bufrem) {
203 /* Make sure there is room for more data */
204 if (grep_lnbufgrow(len + LNBUFBUMP))
206 memcpy(lnbuf + off, bufpos, len - off);
208 if (grep_refill(f) != 0)
211 /* EOF: return partial line */
213 if ((p = memchr(bufpos, '\n', bufrem)) == NULL)
215 /* got it: finish up the line (like code above) */
219 if (grep_lnbufgrow(len))
221 memcpy(lnbuf + off, bufpos, diff);
235 * Opens a file for processing.
238 grep_open(const char *path)
242 f = grep_malloc(sizeof *f);
243 memset(f, 0, sizeof *f);
245 /* Processing stdin implies --line-buffered. */
247 f->fd = STDIN_FILENO;
248 } else if ((f->fd = open(path, O_RDONLY)) == -1)
251 if (filebehave == FILE_MMAP) {
254 if ((fstat(f->fd, &st) == -1) || (st.st_size > OFF_MAX) ||
255 (!S_ISREG(st.st_mode)))
256 filebehave = FILE_STDIO;
258 int flags = MAP_PRIVATE | MAP_NOCORE | MAP_NOSYNC;
259 #ifdef MAP_PREFAULT_READ
260 flags |= MAP_PREFAULT_READ;
263 buffer = mmap(NULL, fsiz, PROT_READ, flags,
265 if (buffer == MAP_FAILED)
266 filebehave = FILE_STDIO;
270 madvise(buffer, st.st_size, MADV_SEQUENTIAL);
275 if ((buffer == NULL) || (buffer == MAP_FAILED))
276 buffer = grep_malloc(MAXBUFSIZ);
278 if (filebehave == FILE_GZIP &&
279 (gzbufdesc = gzdopen(f->fd, "r")) == NULL)
282 #ifndef WITHOUT_BZIP2
283 if (filebehave == FILE_BZIP &&
284 (bzbufdesc = BZ2_bzdopen(f->fd, "r")) == NULL)
288 /* Fill read buffer, also catches errors early */
289 if (bufrem == 0 && grep_refill(f) != 0)
292 /* Check for binary stuff, if necessary */
293 if (binbehave != BINFILE_TEXT && memchr(bufpos, '\0', bufrem) != NULL)
309 grep_close(struct file *f)
314 /* Reset read buffer and line buffer */
315 if (filebehave == FILE_MMAP) {
316 munmap(buffer, fsiz);