usr.bin/cut/cut.c

   1 /*
   2  * Copyright (c) 1989, 1993
   3  *      The Regents of the University of California.  All rights reserved.
   4  *
   5  * This code is derived from software contributed to Berkeley by
   6  * Adam S. Moskowitz of Menlo Consulting and Marciano Pitargue.
   7  *
   8  * Redistribution and use in source and binary forms, with or without
   9  * modification, are permitted provided that the following conditions
  10  * are met:
  11  * 1. Redistributions of source code must retain the above copyright
  12  *    notice, this list of conditions and the following disclaimer.
  13  * 2. Redistributions in binary form must reproduce the above copyright
  14  *    notice, this list of conditions and the following disclaimer in the
  15  *    documentation and/or other materials provided with the distribution.
  16  * 3. All advertising materials mentioning features or use of this software
  17  *    must display the following acknowledgement:
  18  *      This product includes software developed by the University of
  19  *      California, Berkeley and its contributors.
  20  * 4. Neither the name of the University nor the names of its contributors
  21  *    may be used to endorse or promote products derived from this software
  22  *    without specific prior written permission.
  23  *
  24  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  25  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  26  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  27  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  28  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  29  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  30  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  31  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  32  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  33  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  34  * SUCH DAMAGE.
  35  *
  36  * @(#) Copyright (c) 1989, 1993 The Regents of the University of California.  All rights reserved.
  37  * @(#)cut.c    8.3 (Berkeley) 5/4/95
  38  * $FreeBSD: src/usr.bin/cut/cut.c,v 1.9.2.3 2001/07/30 09:59:16 dd Exp $
  39  * $DragonFly: src/usr.bin/cut/cut.c,v 1.6 2008/03/09 16:05:39 swildner Exp $
  40  */
  41
  42 #include <ctype.h>
  43 #include <err.h>
  44 #include <limits.h>
  45 #include <locale.h>
  46 #include <stdio.h>
  47 #include <stdlib.h>
  48 #include <string.h>
  49 #include <unistd.h>
  50
  51 int     cflag;
  52 char    dchar;
  53 int     dflag;
  54 int     fflag;
  55 int     sflag;
  56 int     wflag;
  57
  58 static void     c_cut(FILE *, const char *);
  59 static void     f_cut(FILE *, const char *);
  60 static void     get_list(char *);
  61 static int      is_delim(int);
  62 static void     usage(void);
  63
  64 int
  65 main(int argc, char **argv)
  66 {
  67         FILE *fp;
  68         void (*fcn) (FILE *, const char *) = NULL;
  69         int ch;
  70
  71         fcn = NULL;
  72         setlocale (LC_ALL, "");
  73
  74         dchar = '\t';                   /* default delimiter is \t */
  75
  76         /* Since we don't support multi-byte characters, the -c and -b
  77            options are equivalent, and the -n option is meaningless. */
  78         while ((ch = getopt(argc, argv, "b:c:d:f:snw")) != -1)
  79                 switch(ch) {
  80                 case 'b':
  81                 case 'c':
  82                         fcn = c_cut;
  83                         get_list(optarg);
  84                         cflag = 1;
  85                         break;
  86                 case 'd':
  87                         dchar = *optarg;
  88                         dflag = 1;
  89                         break;
  90                 case 'f':
  91                         get_list(optarg);
  92                         fcn = f_cut;
  93                         fflag = 1;
  94                         break;
  95                 case 's':
  96                         sflag = 1;
  97                         break;
  98                 case 'n':
  99                         break;
 100                 case 'w':
 101                         wflag = 1;
 102                         break;
 103                 case '?':
 104                 default:
 105                         usage();
 106                 }
 107         argc -= optind;
 108         argv += optind;
 109
 110         if (fflag) {
 111                 if (cflag || (wflag && dflag))
 112                         usage();
 113         } else if (!cflag || dflag || sflag || wflag)
 114                 usage();
 115
 116         if (*argv)
 117                 for (; *argv; ++argv) {
 118                         if (!(fp = fopen(*argv, "r")))
 119                                 err(1, "%s", *argv);
 120                         fcn(fp, *argv);
 121                         (void)fclose(fp);
 122                 }
 123         else
 124                 fcn(stdin, "stdin");
 125         exit(0);
 126 }
 127
 128 size_t autostart, autostop, maxval;
 129
 130 char positions[_POSIX2_LINE_MAX + 1];
 131
 132 static void
 133 get_list(char *list)
 134 {
 135         size_t setautostart, start, stop;
 136         char *pos;
 137         char *p;
 138
 139         /*
 140          * set a byte in the positions array to indicate if a field or
 141          * column is to be selected; use +1, it's 1-based, not 0-based.
 142          * This parser is less restrictive than the Draft 9 POSIX spec.
 143          * POSIX doesn't allow lists that aren't in increasing order or
 144          * overlapping lists.  We also handle "-3-5" although there's no
 145          * real reason too.
 146          */
 147         for (; (p = strsep(&list, ", \t")) != NULL;) {
 148                 setautostart = start = stop = 0;
 149                 if (*p == '-') {
 150                         ++p;
 151                         setautostart = 1;
 152                 }
 153                 if (isdigit((unsigned char)*p)) {
 154                         start = stop = strtol(p, &p, 10);
 155                         if (setautostart && start > autostart)
 156                                 autostart = start;
 157                 }
 158                 if (*p == '-') {
 159                         if (isdigit((unsigned char)p[1]))
 160                                 stop = strtol(p + 1, &p, 10);
 161                         if (*p == '-') {
 162                                 ++p;
 163                                 if (!autostop || autostop > stop)
 164                                         autostop = stop;
 165                         }
 166                 }
 167                 if (*p)
 168                         errx(1, "[-cf] list: illegal list value");
 169                 if (!stop || !start)
 170                         errx(1, "[-cf] list: values may not include zero");
 171                 if (stop > _POSIX2_LINE_MAX)
 172                         errx(1, "[-cf] list: %ld too large (max %d)",
 173                             (long)stop, _POSIX2_LINE_MAX);
 174                 if (maxval < stop)
 175                         maxval = stop;
 176                 for (pos = positions + start; start++ <= stop; *pos++ = 1);
 177         }
 178
 179         /* overlapping ranges */
 180         if (autostop && maxval > autostop)
 181                 maxval = autostop;
 182
 183         /* set autostart */
 184         if (autostart)
 185                 memset(positions + 1, '1', autostart);
 186 }
 187
 188 /* ARGSUSED */
 189 static void
 190 c_cut(FILE *fp, const char *fname)
 191 {
 192         int ch, col;
 193         char *pos;
 194         fname = NULL;
 195
 196         ch = 0;
 197         for (;;) {
 198                 pos = positions + 1;
 199                 for (col = maxval; col; --col) {
 200                         if ((ch = getc(fp)) == EOF)
 201                                 return;
 202                         if (ch == '\n')
 203                                 break;
 204                         if (*pos++)
 205                                 (void)putchar(ch);
 206                 }
 207                 if (ch != '\n') {
 208                         if (autostop)
 209                                 while ((ch = getc(fp)) != EOF && ch != '\n')
 210                                         (void)putchar(ch);
 211                         else
 212                                 while ((ch = getc(fp)) != EOF && ch != '\n');
 213                 }
 214                 (void)putchar('\n');
 215         }
 216 }
 217
 218 static int
 219 is_delim(int ch)
 220 {
 221         if (wflag) {
 222                 if (ch == ' ' || ch == '\t')
 223                         return 1;
 224         } else {
 225                 if (ch == dchar)
 226                         return 1;
 227         }
 228         return 0;
 229 }
 230
 231 static void
 232 f_cut(FILE *fp, const char *fname __unused)
 233 {
 234         int ch, field, isdelim;
 235         char *pos, *p, sep;
 236         int output;
 237         char *lbuf, *mlbuf = NULL;
 238         size_t lbuflen;
 239
 240         sep = wflag ? ' ' : dchar;
 241         while ((lbuf = fgetln(fp, &lbuflen)) != NULL) {
 242                 /* Assert EOL has a newline. */
 243                 if (*(lbuf + lbuflen - 1) != '\n') {
 244                         /* Can't have > 1 line with no trailing newline. */
 245                         mlbuf = malloc(lbuflen + 1);
 246                         if (mlbuf == NULL)
 247                                 err(1, "malloc");
 248                         memcpy(mlbuf, lbuf, lbuflen);
 249                         *(mlbuf + lbuflen) = '\n';
 250                         lbuf = mlbuf;
 251                 }
 252                 output = 0;
 253                 for (isdelim = 0, p = lbuf;; ++p) {
 254                         ch = *p;
 255                         /* this should work if newline is delimiter */
 256                         if (is_delim(ch))
 257                                 isdelim = 1;
 258                         if (ch == '\n') {
 259                                 if (!isdelim && !sflag)
 260                                         (void)fwrite(lbuf, lbuflen, 1, stdout);
 261                                 break;
 262                         }
 263                 }
 264                 if (!isdelim)
 265                         continue;
 266
 267                 pos = positions + 1;
 268                 for (field = maxval, p = lbuf; field; --field, ++pos) {
 269                         if (*pos) {
 270                                 if (output++)
 271                                         (void)putchar(sep);
 272                                 while ((ch = *p++) != '\n' && !is_delim(ch))
 273                                         (void)putchar(ch);
 274                                 /* compress whitespace */
 275                                 if (wflag && ch != '\n')
 276                                         while (is_delim(*p)) p++;
 277                         } else {
 278                                 while ((ch = *p++) != '\n' && !is_delim(ch))
 279                                         continue;
 280                                 /* compress whitespace */
 281                                 if (wflag && ch != '\n')
 282                                         while (is_delim(*p)) p++;
 283                         }
 284                         if (ch == '\n')
 285                                 break;
 286                 }
 287                 if (ch != '\n') {
 288                         if (autostop) {
 289                                 if (output)
 290                                         (void)putchar(sep);
 291                                 for (; (ch = *p) != '\n'; ++p)
 292                                         (void)putchar(ch);
 293                         } else
 294                                 for (; (ch = *p) != '\n'; ++p);
 295                 }
 296                 (void)putchar('\n');
 297         }
 298         if (mlbuf != NULL)
 299                 free(mlbuf);
 300 }
 301
 302 static void
 303 usage(void)
 304 {
 305         (void)fprintf(stderr, "%s\n%s\n%s\n",
 306                 "usage: cut -b list [-n] [file ...]",
 307                 "       cut -c list [file ...]",
 308                 "       cut -f list [-s] [-w | -d delim] [file ...]");
 309         exit(1);
 310 }