1 /* $OpenBSD: cut.c,v 1.23 2015/12/02 00:56:46 schwarze 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 #include <assert.h> 37 #include <ctype.h> 38 #include <err.h> 39 #include <errno.h> 40 #include <limits.h> 41 #include <locale.h> 42 #include <stdio.h> 43 #include <stdlib.h> 44 #include <string.h> 45 #include <unistd.h> 46 47 char dchar[5]; 48 int dlen; 49 50 int bflag; 51 int cflag; 52 int dflag; 53 int fflag; 54 int nflag; 55 int sflag; 56 57 void b_cut(FILE *, char *); 58 void c_cut(FILE *, char *); 59 void f_cut(FILE *, char *); 60 void get_list(char *); 61 void usage(void); 62 63 int 64 main(int argc, char *argv[]) 65 { 66 FILE *fp; 67 void (*fcn)(FILE *, char *); 68 int ch, rval; 69 70 setlocale(LC_CTYPE, ""); 71 72 if (pledge("stdio rpath", NULL) == -1) 73 err(1, "pledge"); 74 75 dchar[0] = '\t'; /* default delimiter */ 76 dchar[1] = '\0'; 77 dlen = 1; 78 79 while ((ch = getopt(argc, argv, "b:c:d:f:sn")) != -1) 80 switch(ch) { 81 case 'b': 82 get_list(optarg); 83 bflag = 1; 84 break; 85 case 'c': 86 get_list(optarg); 87 cflag = 1; 88 break; 89 case 'd': 90 if ((dlen = mblen(optarg, MB_CUR_MAX)) == -1) 91 usage(); 92 assert(dlen < sizeof(dchar)); 93 (void)memcpy(dchar, optarg, dlen); 94 dchar[dlen] = '\0'; 95 dflag = 1; 96 break; 97 case 'f': 98 get_list(optarg); 99 fflag = 1; 100 break; 101 case 'n': 102 nflag = 1; 103 break; 104 case 's': 105 sflag = 1; 106 break; 107 case '?': 108 default: 109 usage(); 110 } 111 argc -= optind; 112 argv += optind; 113 114 if (bflag + cflag + fflag != 1 || 115 (nflag && !bflag) || 116 ((dflag || sflag) && !fflag)) 117 usage(); 118 119 if (MB_CUR_MAX == 1) { 120 nflag = 0; 121 if (cflag) { 122 bflag = 1; 123 cflag = 0; 124 } 125 } 126 127 fcn = fflag ? f_cut : (cflag || nflag) ? c_cut : b_cut; 128 129 rval = 0; 130 if (*argv) 131 for (; *argv; ++argv) { 132 if (strcmp(*argv, "-") == 0) 133 fcn(stdin, "stdin"); 134 else { 135 if ((fp = fopen(*argv, "r"))) { 136 fcn(fp, *argv); 137 (void)fclose(fp); 138 } else { 139 rval = 1; 140 warn("%s", *argv); 141 } 142 } 143 } 144 else { 145 if (pledge("stdio rpath", NULL) == -1) 146 err(1, "pledge"); 147 148 fcn(stdin, "stdin"); 149 } 150 exit(rval); 151 } 152 153 int autostart, autostop, maxval; 154 155 char positions[_POSIX2_LINE_MAX + 1]; 156 157 void 158 get_list(char *list) 159 { 160 int setautostart, start, stop; 161 char *pos; 162 char *p; 163 164 /* 165 * set a byte in the positions array to indicate if a field or 166 * column is to be selected; use +1, it's 1-based, not 0-based. 167 * This parser is less restrictive than the Draft 9 POSIX spec. 168 * POSIX doesn't allow lists that aren't in increasing order or 169 * overlapping lists. We also handle "-3-5" although there's no 170 * real reason too. 171 */ 172 while ((p = strsep(&list, ", \t"))) { 173 setautostart = start = stop = 0; 174 if (*p == '-') { 175 ++p; 176 setautostart = 1; 177 } 178 if (isdigit((unsigned char)*p)) { 179 start = stop = strtol(p, &p, 10); 180 if (setautostart && start > autostart) 181 autostart = start; 182 } 183 if (*p == '-') { 184 if (isdigit((unsigned char)p[1])) 185 stop = strtol(p + 1, &p, 10); 186 if (*p == '-') { 187 ++p; 188 if (!autostop || autostop > stop) 189 autostop = stop; 190 } 191 } 192 if (*p) 193 errx(1, "[-bcf] list: illegal list value"); 194 if (!stop || !start) 195 errx(1, "[-bcf] list: values may not include zero"); 196 if (stop > _POSIX2_LINE_MAX) 197 errx(1, "[-bcf] list: %d too large (max %d)", 198 stop, _POSIX2_LINE_MAX); 199 if (maxval < stop) 200 maxval = stop; 201 for (pos = positions + start; start++ <= stop; *pos++ = 1) 202 ; 203 } 204 205 /* overlapping ranges */ 206 if (autostop && maxval > autostop) 207 maxval = autostop; 208 209 /* set autostart */ 210 if (autostart) 211 memset(positions + 1, '1', autostart); 212 } 213 214 /* ARGSUSED */ 215 void 216 b_cut(FILE *fp, char *fname) 217 { 218 int ch, col; 219 char *pos; 220 221 for (;;) { 222 pos = positions + 1; 223 for (col = maxval; col; --col) { 224 if ((ch = getc(fp)) == EOF) 225 return; 226 if (ch == '\n') 227 break; 228 if (*pos++) 229 (void)putchar(ch); 230 } 231 if (ch != '\n') { 232 if (autostop) 233 while ((ch = getc(fp)) != EOF && ch != '\n') 234 (void)putchar(ch); 235 else 236 while ((ch = getc(fp)) != EOF && ch != '\n') 237 ; 238 } 239 (void)putchar('\n'); 240 } 241 } 242 243 void 244 c_cut(FILE *fp, char *fname) 245 { 246 static char *line = NULL; 247 static size_t linesz = 0; 248 ssize_t linelen; 249 char *cp, *pos, *maxpos; 250 int len; 251 252 while ((linelen = getline(&line, &linesz, fp)) != -1) { 253 if (line[linelen - 1] == '\n') 254 line[linelen - 1] = '\0'; 255 256 cp = line; 257 pos = positions + 1; 258 maxpos = pos + maxval; 259 while(pos < maxpos && *cp != '\0') { 260 len = mblen(cp, MB_CUR_MAX); 261 if (len == -1) 262 len = 1; 263 pos += nflag ? len : 1; 264 if (pos[-1] == '\0') 265 cp += len; 266 else 267 while (len--) 268 putchar(*cp++); 269 } 270 if (autostop) 271 puts(cp); 272 else 273 putchar('\n'); 274 } 275 } 276 277 void 278 f_cut(FILE *fp, char *fname) 279 { 280 static char *line = NULL; 281 static size_t linesz = 0; 282 ssize_t linelen; 283 char *sp, *ep, *pos, *maxpos; 284 int output; 285 286 while ((linelen = getline(&line, &linesz, fp)) != -1) { 287 if (line[linelen - 1] == '\n') 288 line[linelen - 1] = '\0'; 289 290 if ((ep = strstr(line, dchar)) == NULL) { 291 if (!sflag) 292 puts(line); 293 continue; 294 } 295 296 pos = positions + 1; 297 maxpos = pos + maxval; 298 output = 0; 299 sp = line; 300 for (;;) { 301 if (*pos++) { 302 if (output) 303 fputs(dchar, stdout); 304 while (sp < ep) 305 putchar(*sp++); 306 output = 1; 307 } else 308 sp = ep; 309 if (*sp == '\0' || pos == maxpos) 310 break; 311 sp += dlen; 312 if ((ep = strstr(sp, dchar)) == NULL) 313 ep = strchr(sp, '\0'); 314 } 315 if (autostop) 316 puts(sp); 317 else 318 putchar('\n'); 319 } 320 } 321 322 void 323 usage(void) 324 { 325 (void)fprintf(stderr, 326 "usage: cut -b list [-n] [file ...]\n" 327 " cut -c list [file ...]\n" 328 " cut -f list [-s] [-d delim] [file ...]\n"); 329 exit(1); 330 } 331