1 /* $NetBSD: cut.c,v 1.14 2002/05/12 21:28:50 wiz Exp $ */ 2 3 /* 4 * Copyright (c) 1989, 1993 5 * The Regents of the University of California. All rights reserved. 6 * 7 * This code is derived from software contributed to Berkeley by 8 * Adam S. Moskowitz of Menlo Consulting and Marciano Pitargue. 9 * 10 * Redistribution and use in source and binary forms, with or without 11 * modification, are permitted provided that the following conditions 12 * are met: 13 * 1. Redistributions of source code must retain the above copyright 14 * notice, this list of conditions and the following disclaimer. 15 * 2. Redistributions in binary form must reproduce the above copyright 16 * notice, this list of conditions and the following disclaimer in the 17 * documentation and/or other materials provided with the distribution. 18 * 3. All advertising materials mentioning features or use of this software 19 * must display the following acknowledgement: 20 * This product includes software developed by the University of 21 * California, Berkeley and its contributors. 22 * 4. Neither the name of the University nor the names of its contributors 23 * may be used to endorse or promote products derived from this software 24 * without specific prior written permission. 25 * 26 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 27 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 28 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 29 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 30 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 31 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 32 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 33 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 34 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 35 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 36 * SUCH DAMAGE. 37 */ 38 39 #include <sys/cdefs.h> 40 #ifndef lint 41 __COPYRIGHT("@(#) Copyright (c) 1989, 1993\n\ 42 The Regents of the University of California. All rights reserved.\n"); 43 #endif /* not lint */ 44 45 #ifndef lint 46 #if 0 47 static char sccsid[] = "@(#)cut.c 8.3 (Berkeley) 5/4/95"; 48 #endif 49 __RCSID("$NetBSD: cut.c,v 1.14 2002/05/12 21:28:50 wiz Exp $"); 50 #endif /* not lint */ 51 52 #include <ctype.h> 53 #include <err.h> 54 #include <errno.h> 55 #include <limits.h> 56 #include <locale.h> 57 #include <stdio.h> 58 #include <stdlib.h> 59 #include <string.h> 60 #include <unistd.h> 61 62 int cflag; 63 char dchar; 64 int dflag; 65 int fflag; 66 int sflag; 67 68 void c_cut __P((FILE *, char *)); 69 void f_cut __P((FILE *, char *)); 70 void get_list __P((char *)); 71 int main __P((int, char **)); 72 void usage __P((void)); 73 74 int 75 main(argc, argv) 76 int argc; 77 char *argv[]; 78 { 79 FILE *fp; 80 void (*fcn) __P((FILE *, char *)); 81 int ch; 82 83 fcn = NULL; 84 setlocale (LC_ALL, ""); 85 86 dchar = '\t'; /* default delimiter is \t */ 87 88 /* Since we don't support multi-byte characters, the -c and -b 89 options are equivalent, and the -n option is meaningless. */ 90 while ((ch = getopt(argc, argv, "b:c:d:f:sn")) != -1) 91 switch(ch) { 92 case 'b': 93 case 'c': 94 fcn = c_cut; 95 get_list(optarg); 96 cflag = 1; 97 break; 98 case 'd': 99 dchar = *optarg; 100 dflag = 1; 101 break; 102 case 'f': 103 get_list(optarg); 104 fcn = f_cut; 105 fflag = 1; 106 break; 107 case 's': 108 sflag = 1; 109 break; 110 case 'n': 111 break; 112 case '?': 113 default: 114 usage(); 115 } 116 argc -= optind; 117 argv += optind; 118 119 if (fflag) { 120 if (cflag) 121 usage(); 122 } else if (!cflag || dflag || sflag) 123 usage(); 124 125 if (*argv) 126 for (; *argv; ++argv) { 127 if (!(fp = fopen(*argv, "r"))) 128 err(1, "%s", *argv); 129 fcn(fp, *argv); 130 (void)fclose(fp); 131 } 132 else 133 fcn(stdin, "stdin"); 134 exit(0); 135 } 136 137 int autostart, autostop, maxval; 138 139 char positions[_POSIX2_LINE_MAX + 1]; 140 141 void 142 get_list(list) 143 char *list; 144 { 145 int setautostart, start, stop; 146 char *pos; 147 char *p; 148 149 /* 150 * set a byte in the positions array to indicate if a field or 151 * column is to be selected; use +1, it's 1-based, not 0-based. 152 * This parser is less restrictive than the Draft 9 POSIX spec. 153 * POSIX doesn't allow lists that aren't in increasing order or 154 * overlapping lists. We also handle "-3-5" although there's no 155 * real reason too. 156 */ 157 for (; (p = strtok(list, ", \t")) != NULL; list = NULL) { 158 setautostart = start = stop = 0; 159 if (*p == '-') { 160 ++p; 161 setautostart = 1; 162 } 163 if (isdigit((unsigned char)*p)) { 164 start = stop = strtol(p, &p, 10); 165 if (setautostart && start > autostart) 166 autostart = start; 167 } 168 if (*p == '-') { 169 if (isdigit((unsigned char)p[1])) 170 stop = strtol(p + 1, &p, 10); 171 if (*p == '-') { 172 ++p; 173 if (!autostop || autostop > stop) 174 autostop = stop; 175 } 176 } 177 if (*p) 178 errx(1, "[-cf] list: illegal list value\n"); 179 if (!stop || !start) 180 errx(1, "[-cf] list: values may not include zero\n"); 181 if (stop > _POSIX2_LINE_MAX) 182 errx(1, "[-cf] list: %d too large (max %d)\n", 183 stop, _POSIX2_LINE_MAX); 184 if (maxval < stop) 185 maxval = stop; 186 for (pos = positions + start; start++ <= stop; *pos++ = 1); 187 } 188 189 /* overlapping ranges */ 190 if (autostop && maxval > autostop) 191 maxval = autostop; 192 193 /* set autostart */ 194 if (autostart) 195 memset(positions + 1, '1', autostart); 196 } 197 198 /* ARGSUSED */ 199 void 200 c_cut(fp, fname) 201 FILE *fp; 202 char *fname; 203 { 204 int ch, col; 205 char *pos; 206 207 ch = 0; 208 for (;;) { 209 pos = positions + 1; 210 for (col = maxval; col; --col) { 211 if ((ch = getc(fp)) == EOF) 212 return; 213 if (ch == '\n') 214 break; 215 if (*pos++) 216 (void)putchar(ch); 217 } 218 if (ch != '\n') { 219 if (autostop) 220 while ((ch = getc(fp)) != EOF && ch != '\n') 221 (void)putchar(ch); 222 else 223 while ((ch = getc(fp)) != EOF && ch != '\n'); 224 } 225 (void)putchar('\n'); 226 } 227 } 228 229 void 230 f_cut(fp, fname) 231 FILE *fp; 232 char *fname; 233 { 234 int ch, field, isdelim; 235 char *pos, *p, sep; 236 int output; 237 char lbuf[_POSIX2_LINE_MAX + 1]; 238 239 for (sep = dchar; fgets(lbuf, sizeof(lbuf), fp);) { 240 output = 0; 241 for (isdelim = 0, p = lbuf;; ++p) { 242 if (!(ch = *p)) 243 errx(1, "%s: line too long.\n", fname); 244 /* this should work if newline is delimiter */ 245 if (ch == sep) 246 isdelim = 1; 247 if (ch == '\n') { 248 if (!isdelim && !sflag) 249 (void)printf("%s", lbuf); 250 break; 251 } 252 } 253 if (!isdelim) 254 continue; 255 256 pos = positions + 1; 257 for (field = maxval, p = lbuf; field; --field, ++pos) { 258 if (*pos) { 259 if (output++) 260 (void)putchar(sep); 261 while ((ch = *p++) != '\n' && ch != sep) 262 (void)putchar(ch); 263 } else { 264 while ((ch = *p++) != '\n' && ch != sep) 265 continue; 266 } 267 if (ch == '\n') 268 break; 269 } 270 if (ch != '\n') { 271 if (autostop) { 272 if (output) 273 (void)putchar(sep); 274 for (; (ch = *p) != '\n'; ++p) 275 (void)putchar(ch); 276 } else 277 for (; (ch = *p) != '\n'; ++p); 278 } 279 (void)putchar('\n'); 280 } 281 } 282 283 void 284 usage() 285 { 286 (void)fprintf(stderr, "usage:\tcut -b list [-n] [file ...]\n" 287 "\tcut -c list [file1 ...]\n" 288 "\tcut -f list [-d delim] [-s] [file ...]\n"); 289 exit(1); 290 } 291