1 /* $OpenBSD: cut.c,v 1.13 2008/06/27 08:02:13 sobrado Exp $ */ 2 /* $NetBSD: cut.c,v 1.9 1995/09/02 05:59:23 jtc Exp $ */ 3 4 /* 5 * Copyright (c) 1989, 1993 6 * The Regents of the University of California. All rights reserved. 7 * 8 * This code is derived from software contributed to Berkeley by 9 * Adam S. Moskowitz of Menlo Consulting and Marciano Pitargue. 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 * 3. Neither the name of the University nor the names of its contributors 20 * may be used to endorse or promote products derived from this software 21 * without specific prior written permission. 22 * 23 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 24 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 25 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 26 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 27 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 28 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 29 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 30 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 31 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 32 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 33 * SUCH DAMAGE. 34 */ 35 36 #ifndef lint 37 static char copyright[] = 38 "@(#) Copyright (c) 1989, 1993\n\ 39 The Regents of the University of California. All rights reserved.\n"; 40 #endif /* not lint */ 41 42 #ifndef lint 43 #if 0 44 static char sccsid[] = "@(#)cut.c 8.3 (Berkeley) 5/4/95"; 45 #endif 46 static char rcsid[] = "$OpenBSD: cut.c,v 1.13 2008/06/27 08:02:13 sobrado Exp $"; 47 #endif /* not lint */ 48 49 #include <ctype.h> 50 #include <err.h> 51 #include <errno.h> 52 #include <limits.h> 53 #include <locale.h> 54 #include <stdio.h> 55 #include <stdlib.h> 56 #include <string.h> 57 #include <unistd.h> 58 59 int cflag; 60 char dchar; 61 int dflag; 62 int fflag; 63 int sflag; 64 65 void c_cut(FILE *, char *); 66 void f_cut(FILE *, char *); 67 void get_list(char *); 68 void usage(void); 69 70 int 71 main(int argc, char *argv[]) 72 { 73 FILE *fp; 74 void (*fcn)(FILE *, char *); 75 int ch; 76 77 setlocale (LC_ALL, ""); 78 79 dchar = '\t'; /* default delimiter is \t */ 80 81 /* Since we don't support multi-byte characters, the -c and -b 82 options are equivalent, and the -n option is meaningless. */ 83 while ((ch = getopt(argc, argv, "b:c:d:f:sn")) != -1) 84 switch(ch) { 85 case 'b': 86 case 'c': 87 fcn = c_cut; 88 get_list(optarg); 89 cflag = 1; 90 break; 91 case 'd': 92 dchar = *optarg; 93 dflag = 1; 94 break; 95 case 'f': 96 get_list(optarg); 97 fcn = f_cut; 98 fflag = 1; 99 break; 100 case 's': 101 sflag = 1; 102 break; 103 case 'n': 104 break; 105 case '?': 106 default: 107 usage(); 108 } 109 argc -= optind; 110 argv += optind; 111 112 if (fflag) { 113 if (cflag) 114 usage(); 115 } else if (!cflag || dflag || sflag) 116 usage(); 117 118 if (*argv) 119 for (; *argv; ++argv) { 120 if (!(fp = fopen(*argv, "r"))) 121 err(1, "%s", *argv); 122 fcn(fp, *argv); 123 (void)fclose(fp); 124 } 125 else 126 fcn(stdin, "stdin"); 127 exit(0); 128 } 129 130 int autostart, autostop, maxval; 131 132 char positions[_POSIX2_LINE_MAX + 1]; 133 134 void 135 get_list(char *list) 136 { 137 int setautostart, start, stop; 138 char *pos; 139 char *p; 140 141 /* 142 * set a byte in the positions array to indicate if a field or 143 * column is to be selected; use +1, it's 1-based, not 0-based. 144 * This parser is less restrictive than the Draft 9 POSIX spec. 145 * POSIX doesn't allow lists that aren't in increasing order or 146 * overlapping lists. We also handle "-3-5" although there's no 147 * real reason too. 148 */ 149 while ((p = strsep(&list, ", \t"))) { 150 setautostart = start = stop = 0; 151 if (*p == '-') { 152 ++p; 153 setautostart = 1; 154 } 155 if (isdigit(*p)) { 156 start = stop = strtol(p, &p, 10); 157 if (setautostart && start > autostart) 158 autostart = start; 159 } 160 if (*p == '-') { 161 if (isdigit(p[1])) 162 stop = strtol(p + 1, &p, 10); 163 if (*p == '-') { 164 ++p; 165 if (!autostop || autostop > stop) 166 autostop = stop; 167 } 168 } 169 if (*p) 170 errx(1, "[-cf] list: illegal list value"); 171 if (!stop || !start) 172 errx(1, "[-cf] list: values may not include zero"); 173 if (stop > _POSIX2_LINE_MAX) 174 errx(1, "[-cf] list: %d too large (max %d)", 175 stop, _POSIX2_LINE_MAX); 176 if (maxval < stop) 177 maxval = stop; 178 for (pos = positions + start; start++ <= stop; *pos++ = 1) 179 ; 180 } 181 182 /* overlapping ranges */ 183 if (autostop && maxval > autostop) 184 maxval = autostop; 185 186 /* set autostart */ 187 if (autostart) 188 memset(positions + 1, '1', autostart); 189 } 190 191 /* ARGSUSED */ 192 void 193 c_cut(FILE *fp, char *fname) 194 { 195 int ch, col; 196 char *pos; 197 198 for (;;) { 199 pos = positions + 1; 200 for (col = maxval; col; --col) { 201 if ((ch = getc(fp)) == EOF) 202 return; 203 if (ch == '\n') 204 break; 205 if (*pos++) 206 (void)putchar(ch); 207 } 208 if (ch != '\n') { 209 if (autostop) 210 while ((ch = getc(fp)) != EOF && ch != '\n') 211 (void)putchar(ch); 212 else 213 while ((ch = getc(fp)) != EOF && ch != '\n') 214 ; 215 } 216 (void)putchar('\n'); 217 } 218 } 219 220 void 221 f_cut(FILE *fp, char *fname) 222 { 223 int ch, field, isdelim; 224 char *pos, *p, sep; 225 int output; 226 size_t len; 227 char *lbuf, *tbuf; 228 229 for (sep = dchar, tbuf = NULL; (lbuf = fgetln(fp, &len));) { 230 output = 0; 231 if (lbuf[len - 1] != '\n') { 232 /* no newline at the end of the last line so add one */ 233 if ((tbuf = (char *)malloc(len + 1)) == NULL) 234 err(1, NULL); 235 memcpy(tbuf, lbuf, len); 236 tbuf[len] = '\n'; 237 lbuf = tbuf; 238 } 239 for (isdelim = 0, p = lbuf;; ++p) { 240 ch = *p; 241 /* this should work if newline is delimiter */ 242 if (ch == sep) 243 isdelim = 1; 244 if (ch == '\n') { 245 if (!isdelim && !sflag) 246 (void)fwrite(lbuf, len, 1, stdout); 247 break; 248 } 249 } 250 if (!isdelim) 251 continue; 252 253 pos = positions + 1; 254 for (field = maxval, p = lbuf; field; --field, ++pos) { 255 if (*pos) { 256 if (output++) 257 (void)putchar(sep); 258 while ((ch = *p++) != '\n' && ch != sep) 259 (void)putchar(ch); 260 } else 261 while ((ch = *p++) != '\n' && ch != sep) 262 ; 263 if (ch == '\n') 264 break; 265 } 266 if (ch != '\n') { 267 if (autostop) { 268 if (output) 269 (void)putchar(sep); 270 for (; (ch = *p) != '\n'; ++p) 271 (void)putchar(ch); 272 } else 273 for (; (ch = *p) != '\n'; ++p) 274 ; 275 } 276 (void)putchar('\n'); 277 } 278 if (tbuf) 279 free(tbuf); 280 } 281 282 void 283 usage(void) 284 { 285 (void)fprintf(stderr, 286 "usage: cut -b list [-n] [file ...]\n" 287 " cut -c list [file ...]\n" 288 " cut -f list [-s] [-d delim] [file ...]\n"); 289 exit(1); 290 } 291