file.c revision 1.9
1 /* $NetBSD: file.c,v 1.9 2018/08/12 07:53:19 christos Exp $ */ 2 /* $FreeBSD: head/usr.bin/grep/file.c 211496 2010-08-19 09:28:59Z des $ */ 3 /* $OpenBSD: file.c,v 1.11 2010/07/02 20:48:48 nicm Exp $ */ 4 5 /*- 6 * Copyright (c) 1999 James Howard and Dag-Erling Codan Smrgrav 7 * Copyright (C) 2008-2010 Gabor Kovesdan <gabor (at) FreeBSD.org> 8 * Copyright (C) 2010 Dimitry Andric <dimitry (at) andric.com> 9 * All rights reserved. 10 * 11 * Redistribution and use in source and binary forms, with or without 12 * modification, are permitted provided that the following conditions 13 * are met: 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. 19 * 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 30 * SUCH DAMAGE. 31 */ 32 33 #if HAVE_NBTOOL_CONFIG_H 34 #include "nbtool_config.h" 35 #endif 36 37 #include <sys/cdefs.h> 38 __RCSID("$NetBSD: file.c,v 1.9 2018/08/12 07:53:19 christos Exp $"); 39 40 #include <sys/param.h> 41 #include <sys/types.h> 42 #include <sys/stat.h> 43 44 #include <err.h> 45 #include <errno.h> 46 #include <fcntl.h> 47 #include <stddef.h> 48 #include <stdlib.h> 49 #include <string.h> 50 #include <unistd.h> 51 #include <wchar.h> 52 #include <wctype.h> 53 54 #include "grep.h" 55 56 #define MAXBUFSIZ (32 * 1024) 57 #define LNBUFBUMP 80 58 59 static gzFile gzbufdesc; 60 #ifndef WITHOUT_BZ2 61 static BZFILE* bzbufdesc; 62 #endif 63 64 static unsigned char buffer[MAXBUFSIZ]; 65 static unsigned char *bufpos; 66 static size_t bufrem; 67 68 static unsigned char *lnbuf; 69 static size_t lnbuflen; 70 71 static inline int 72 grep_refill(struct file *f) 73 { 74 ssize_t nr; 75 int bzerr; 76 77 bufpos = buffer; 78 bufrem = 0; 79 80 if (filebehave == FILE_GZIP) { 81 nr = gzread(gzbufdesc, buffer, MAXBUFSIZ); 82 #ifndef WITHOUT_BZ2 83 } else if (filebehave == FILE_BZIP && bzbufdesc != NULL) { 84 nr = BZ2_bzRead(&bzerr, bzbufdesc, buffer, MAXBUFSIZ); 85 switch (bzerr) { 86 case BZ_OK: 87 case BZ_STREAM_END: 88 /* No problem, nr will be okay */ 89 break; 90 case BZ_DATA_ERROR_MAGIC: 91 /* 92 * As opposed to gzread(), which simply returns the 93 * plain file data, if it is not in the correct 94 * compressed format, BZ2_bzRead() instead aborts. 95 * 96 * So, just restart at the beginning of the file again, 97 * and use plain reads from now on. 98 */ 99 BZ2_bzReadClose(&bzerr, bzbufdesc); 100 bzbufdesc = NULL; 101 if (lseek(f->fd, 0, SEEK_SET) == -1) 102 return (-1); 103 nr = read(f->fd, buffer, MAXBUFSIZ); 104 break; 105 default: 106 /* Make sure we exit with an error */ 107 nr = -1; 108 } 109 #endif 110 } else 111 nr = read(f->fd, buffer, MAXBUFSIZ); 112 113 if (nr < 0) 114 return (-1); 115 116 bufrem = nr; 117 return (0); 118 } 119 120 static inline int 121 grep_lnbufgrow(size_t newlen) 122 { 123 124 if (lnbuflen < newlen) { 125 lnbuf = grep_realloc(lnbuf, newlen); 126 lnbuflen = newlen; 127 } 128 129 return (0); 130 } 131 132 char * 133 grep_fgetln(struct file *f, size_t *lenp) 134 { 135 unsigned char *p; 136 char *ret; 137 size_t len; 138 size_t off; 139 ptrdiff_t diff; 140 141 /* Fill the buffer, if necessary */ 142 if (bufrem == 0 && grep_refill(f) != 0) 143 goto error; 144 145 if (bufrem == 0) { 146 /* Return zero length to indicate EOF */ 147 *lenp = 0; 148 return ((char *)bufpos); 149 } 150 151 /* Look for a newline in the remaining part of the buffer */ 152 if ((p = memchr(bufpos, line_sep, bufrem)) != NULL) { 153 ++p; /* advance over newline */ 154 ret = (char *)bufpos; 155 len = p - bufpos; 156 bufrem -= len; 157 bufpos = p; 158 *lenp = len; 159 return (ret); 160 } 161 162 /* We have to copy the current buffered data to the line buffer */ 163 for (len = bufrem, off = 0; ; len += bufrem) { 164 /* Make sure there is room for more data */ 165 if (grep_lnbufgrow(len + LNBUFBUMP)) 166 goto error; 167 memcpy(lnbuf + off, bufpos, len - off); 168 off = len; 169 if (grep_refill(f) != 0) 170 goto error; 171 if (bufrem == 0) 172 /* EOF: return partial line */ 173 break; 174 if ((p = memchr(bufpos, line_sep, bufrem)) == NULL) 175 continue; 176 /* got it: finish up the line (like code above) */ 177 ++p; 178 diff = p - bufpos; 179 len += diff; 180 if (grep_lnbufgrow(len)) 181 goto error; 182 memcpy(lnbuf + off, bufpos, diff); 183 bufrem -= diff; 184 bufpos = p; 185 break; 186 } 187 *lenp = len; 188 return ((char *)lnbuf); 189 190 error: 191 *lenp = 0; 192 return (NULL); 193 } 194 195 static inline struct file * 196 grep_file_init(struct file *f) 197 { 198 199 if (filebehave == FILE_GZIP && 200 (gzbufdesc = gzdopen(f->fd, "r")) == NULL) 201 goto error; 202 203 #ifndef WITHOUT_BZ2 204 if (filebehave == FILE_BZIP && 205 (bzbufdesc = BZ2_bzdopen(f->fd, "r")) == NULL) 206 goto error; 207 #endif 208 209 /* Fill read buffer, also catches errors early */ 210 if (grep_refill(f) != 0) 211 goto error; 212 213 /* Check for binary stuff, if necessary */ 214 if (!nulldataflag && binbehave != BINFILE_TEXT && 215 memchr(bufpos, '\0', bufrem) != NULL) 216 f->binary = true; 217 218 return (f); 219 error: 220 close(f->fd); 221 free(f); 222 return (NULL); 223 } 224 225 /* 226 * Opens a file for processing. 227 */ 228 struct file * 229 grep_open(const char *path) 230 { 231 struct file *f; 232 233 f = grep_malloc(sizeof *f); 234 memset(f, 0, sizeof *f); 235 if (path == NULL) { 236 /* Processing stdin implies --line-buffered. */ 237 lbflag = true; 238 f->fd = STDIN_FILENO; 239 } else if ((f->fd = open(path, O_RDONLY)) == -1) { 240 free(f); 241 return (NULL); 242 } 243 244 return (grep_file_init(f)); 245 } 246 247 /* 248 * Closes a file. 249 */ 250 void 251 grep_close(struct file *f) 252 { 253 254 close(f->fd); 255 256 /* Reset read buffer and line buffer */ 257 bufpos = buffer; 258 bufrem = 0; 259 260 free(lnbuf); 261 lnbuf = NULL; 262 lnbuflen = 0; 263 } 264
Indexes created Tue Sep 23 22:09:46 GMT 2025