usr.bin/cut/cut.c

   1 /*      $NetBSD: cut.c,v 1.24 2007/12/15 19:44:49 perry 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. Neither the name of the University nor the names of its contributors
  19  *    may be used to endorse or promote products derived from this software
  20  *    without specific prior written permission.
  21  *
  22  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  23  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  24  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  25  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  26  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  27  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  28  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  29  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  30  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  31  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  32  * SUCH DAMAGE.
  33  */
  34
  35 #include <sys/cdefs.h>
  36 #ifndef lint
  37 __COPYRIGHT("@(#) Copyright (c) 1989, 1993\
  38  The Regents of the University of California.  All rights reserved.");
  39 #endif /* not lint */
  40
  41 #ifndef lint
  42 #if 0
  43 static char sccsid[] = "@(#)cut.c       8.3 (Berkeley) 5/4/95";
  44 #endif
  45 __RCSID("$NetBSD: cut.c,v 1.24 2007/12/15 19:44:49 perry Exp $");
  46 #endif /* not lint */
  47
  48 #include <ctype.h>
  49 #include <err.h>
  50 #include <errno.h>
  51 #include <limits.h>
  52 #include <locale.h>
  53 #include <stdio.h>
  54 #include <stdlib.h>
  55 #include <string.h>
  56 #include <unistd.h>
  57 #include <util.h>
  58 #include <wchar.h>
  59 #include <sys/param.h>
  60
  61 static int bflag;
  62 static int      cflag;
  63 static char     dchar;
  64 static int      dflag;
  65 static int      fflag;
  66 static int      sflag;
  67
  68 static void     b_cut(FILE *, const char *);
  69 static void     c_cut(FILE *, const char *);
  70 static void     f_cut(FILE *, const char *);
  71 static void     get_list(char *);
  72 static void     usage(void) __dead;
  73
  74 int
  75 main(int argc, char *argv[])
  76 {
  77         FILE *fp;
  78         void (*fcn)(FILE *, const char *);
  79         int ch;
  80
  81         fcn = NULL;
  82         (void)setlocale(LC_ALL, "");
  83
  84         dchar = '\t';                   /* default delimiter is \t */
  85
  86         /* Since we don't support multi-byte characters, the -c and -b
  87            options are equivalent, and the -n option is meaningless. */
  88         while ((ch = getopt(argc, argv, "b:c:d:f:sn")) != -1)
  89                 switch(ch) {
  90                 case 'b':
  91                         fcn = b_cut;
  92                         get_list(optarg);
  93                         bflag = 1;
  94                         break;
  95                 case 'c':
  96                         fcn = c_cut;
  97                         get_list(optarg);
  98                         cflag = 1;
  99                         break;
 100                 case 'd':
 101                         dchar = *optarg;
 102                         dflag = 1;
 103                         break;
 104                 case 'f':
 105                         get_list(optarg);
 106                         fcn = f_cut;
 107                         fflag = 1;
 108                         break;
 109                 case 's':
 110                         sflag = 1;
 111                         break;
 112                 case 'n':
 113                         break;
 114                 case '?':
 115                 default:
 116                         usage();
 117                 }
 118         argc -= optind;
 119         argv += optind;
 120
 121         if (fflag) {
 122                 if (cflag || bflag)
 123                         usage();
 124         } else if ((!cflag && !bflag) || dflag || sflag)
 125                 usage();
 126         else if (bflag && cflag)
 127                 usage();
 128
 129         if (*argv)
 130                 for (; *argv; ++argv) {
 131                         if (strcmp(*argv, "-") == 0)
 132                                 fcn(stdin, "stdin");
 133                         else {
 134                                 if ((fp = fopen(*argv, "r")) == NULL)
 135                                         err(1, "%s", *argv);
 136                                 fcn(fp, *argv);
 137                                 (void)fclose(fp);
 138                         }
 139                 }
 140         else
 141                 fcn(stdin, "stdin");
 142         return 0;
 143 }
 144
 145 static size_t autostart, autostop, maxval;
 146
 147 static char *positions = NULL;
 148 static size_t numpositions = 0;
 149 #define ALLOC_CHUNK     _POSIX2_LINE_MAX        /* malloc granularity */
 150
 151 static void
 152 get_list(char *list)
 153 {
 154         size_t setautostart, start, stop;
 155         char *pos;
 156         char *p;
 157
 158         if (positions == NULL) {
 159                 numpositions = ALLOC_CHUNK;
 160                 positions = ecalloc(numpositions, sizeof(*positions));
 161         }
 162
 163         /*
 164          * set a byte in the positions array to indicate if a field or
 165          * column is to be selected; use +1, it's 1-based, not 0-based.
 166          * This parser is less restrictive than the Draft 9 POSIX spec.
 167          * POSIX doesn't allow lists that aren't in increasing order or
 168          * overlapping lists.  We also handle "-3-5" although there's no
 169          * real reason too.
 170          */
 171         for (; (p = strtok(list, ", \t")) != NULL; list = NULL) {
 172                 setautostart = start = stop = 0;
 173                 if (*p == '-') {
 174                         ++p;
 175                         setautostart = 1;
 176                 }
 177                 if (isdigit((unsigned char)*p)) {
 178                         start = stop = strtol(p, &p, 10);
 179                         if (setautostart && start > autostart)
 180                                 autostart = start;
 181                 }
 182                 if (*p == '-') {
 183                         if (isdigit((unsigned char)p[1]))
 184                                 stop = strtol(p + 1, &p, 10);
 185                         if (*p == '-') {
 186                                 ++p;
 187                                 if (!autostop || autostop > stop)
 188                                         autostop = stop;
 189                         }
 190                 }
 191                 if (*p)
 192                         errx(1, "[-cf] list: illegal list value");
 193                 if (!stop || !start)
 194                         errx(1, "[-cf] list: values may not include zero");
 195                 if (stop + 1 > numpositions) {
 196                         size_t newsize;
 197                         newsize = roundup(stop + 1, ALLOC_CHUNK);
 198                         positions = erealloc(positions, newsize);
 199                         (void)memset(positions + numpositions, 0,
 200                             newsize - numpositions);
 201                         numpositions = newsize;
 202                 }
 203                 if (maxval < stop)
 204                         maxval = stop;
 205                 for (pos = positions + start; start++ <= stop; pos++)
 206                         *pos = 1;
 207         }
 208
 209         /* overlapping ranges */
 210         if (autostop && maxval > autostop)
 211                 maxval = autostop;
 212
 213         /* set autostart */
 214         if (autostart)
 215                 (void)memset(positions + 1, '1', autostart);
 216 }
 217
 218 static void
 219 /*ARGSUSED*/
 220 f_cut(FILE *fp, const char *fname __unused)
 221 {
 222         int ch, field, isdelim;
 223         char *pos, *p, sep;
 224         int output;
 225         size_t len;
 226         char *lbuf, *tbuf;
 227
 228         for (sep = dchar, tbuf = NULL; (lbuf = fgetln(fp, &len)) != NULL;) {
 229                 output = 0;
 230                 if (lbuf[len - 1] != '\n') {
 231                         /* no newline at the end of the last line so add one */
 232                         if ((tbuf = (char *)malloc(len + 1)) == NULL)
 233                                 err(1, NULL);
 234                         (void)memcpy(tbuf, lbuf, len);
 235                         tbuf[len++] = '\n';
 236                         lbuf = tbuf;
 237                 }
 238                 for (isdelim = 0, p = lbuf;; ++p) {
 239                         ch = *p;
 240                         /* this should work if newline is delimiter */
 241                         if (ch == sep)
 242                                 isdelim = 1;
 243                         if (ch == '\n') {
 244                                 if (!isdelim && !sflag)
 245                                         (void)fwrite(lbuf, len, 1, stdout);
 246                                 break;
 247                         }
 248                 }
 249                 if (!isdelim)
 250                         continue;
 251
 252                 pos = positions + 1;
 253                 for (field = maxval, p = lbuf; field; --field, ++pos) {
 254                         if (*pos) {
 255                                 if (output++)
 256                                         (void)putchar(sep);
 257                                 while ((ch = *p++) != '\n' && ch != sep)
 258                                         (void)putchar(ch);
 259                         } else {
 260                                 while ((ch = *p++) != '\n' && ch != sep)
 261                                         continue;
 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                 (void)putchar('\n');
 276                 if (tbuf) {
 277                         free(tbuf);
 278                         tbuf = NULL;
 279                 }
 280         }
 281         if (tbuf)
 282                 free(tbuf);
 283 }
 284
 285 static void
 286 usage(void)
 287 {
 288         (void)fprintf(stderr, "Usage:\tcut -b list [-n] [file ...]\n"
 289             "\tcut -c list [file1 ...]\n"
 290             "\tcut -f list [-d delim] [-s] [file ...]\n");
 291         exit(1);
 292 }
 293
 294 /* make b_put(): */
 295 #define CUT_BYTE 1
 296 #include "x_cut.c"
 297 #undef CUT_BYTE
 298
 299 /* make c_put(): */
 300 #define CUT_BYTE 0
 301 #include "x_cut.c"
 302 #undef CUT_BYTE