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