1 /* uniq -- remove duplicate lines from a sorted file
2 Copyright (C) 1986, 1991, 1995 Free Software Foundation, Inc.
4 This program is free software; you can redistribute it and/or modify
5 it under the terms of the GNU General Public License as published by
6 the Free Software Foundation; either version 2, or (at your option)
9 This program is distributed in the hope that it will be useful,
10 but WITHOUT ANY WARRANTY; without even the implied warranty of
11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 GNU General Public License for more details.
14 You should have received a copy of the GNU General Public License
15 along with this program; if not, write to the Free Software
16 Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */
18 /* Written by Richard Stallman and David MacKenzie. */
22 /* Get isblank from GNU libc. */
27 #include <sys/types.h>
34 # define UINT_MAX ((unsigned int) ~(unsigned int) 0)
38 # define INT_MAX ((int) (UINT_MAX >> 1))
42 #include "linebuffer.h"
46 /* Undefine, to avoid warning about redefinition on some systems. */
48 #define min(x, y) ((x) < (y) ? (x) : (y))
50 /* The name this program was run with. */
53 /* Number of fields to skip on each line when doing comparisons. */
54 static int skip_fields;
56 /* Number of chars to skip after skipping any fields. */
57 static int skip_chars;
59 /* Number of chars to compare; if 0, compare the whole lines. */
60 static int check_chars;
64 count_occurrences, /* -c Print count before output lines. */
65 count_none /* Default. Do not print counts. */
68 /* Whether and how to precede the output lines with a count of the number of
69 times they occurred in the input. */
70 static enum countmode countmode;
74 output_repeated, /* -d Only lines that are repeated. */
75 output_unique, /* -u Only lines that are not repeated. */
76 output_all /* Default. Print first copy of each line. */
79 /* Which lines to output. */
80 static enum output_mode mode;
82 /* If nonzero, display usage information and exit. */
85 /* If nonzero, print the version on standard output then exit. */
86 static int show_version;
88 static struct option const longopts[] =
90 {"count", no_argument, NULL, 'c'},
91 {"repeated", no_argument, NULL, 'd'},
92 {"unique", no_argument, NULL, 'u'},
93 {"skip-fields", required_argument, NULL, 'f'},
94 {"skip-chars", required_argument, NULL, 's'},
95 {"check-chars", required_argument, NULL, 'w'},
96 {"help", no_argument, &show_help, 1},
97 {"version", no_argument, &show_version, 1},
105 fprintf (stderr, _("Try `%s --help' for more information.\n"),
110 Usage: %s [OPTION]... [INPUT [OUTPUT]]\n\
114 Discard all but one of successive identical lines from INPUT (or\n\
115 standard input), writing to OUTPUT (or standard output).\n\
117 -c, --count prefix lines by the number of occurrences\n\
118 -d, --repeated only print duplicate lines\n\
119 -f, --skip-fields=N avoid comparing the first N fields\n\
120 -s, --skip-chars=N avoid comparing the first N characters\n\
121 -u, --unique only print unique lines\n\
122 -w, --check-chars=N compare no more than N characters in lines\n\
125 --help display this help and exit\n\
126 --version output version information and exit\n\
128 A field is a run of whitespace, than non-whitespace characters.\n\
129 Fields are skipped before chars. \n\
135 /* Given a linebuffer LINE,
136 return a pointer to the beginning of the line's field to be compared. */
139 find_field (const struct linebuffer *line)
142 register char *lp = line->buffer;
143 register int size = line->length;
146 for (count = 0; count < skip_fields && i < size; count++)
148 while (i < size && ISBLANK (lp[i]))
150 while (i < size && !ISBLANK (lp[i]))
154 for (count = 0; count < skip_chars && i < size; count++)
160 /* Return zero if two strings OLD and NEW match, nonzero if not.
161 OLD and NEW point not to the beginnings of the lines
162 but rather to the beginnings of the fields to compare.
163 OLDLEN and NEWLEN are their lengths. */
166 different (const char *old, const char *new, int oldlen, int newlen)
172 if (oldlen > check_chars)
173 oldlen = check_chars;
174 if (newlen > check_chars)
175 newlen = check_chars;
177 order = memcmp (old, new, min (oldlen, newlen));
179 return oldlen - newlen;
183 /* Output the line in linebuffer LINE to stream STREAM
184 provided that the switches say it should be output.
185 If requested, print the number of times it occurred, as well;
186 LINECOUNT + 1 is the number of times that the line occurred. */
189 writeline (const struct linebuffer *line, FILE *stream, int linecount)
191 if ((mode == output_unique && linecount != 0)
192 || (mode == output_repeated && linecount == 0))
195 if (countmode == count_occurrences)
196 fprintf (stream, "%7d\t", linecount + 1);
198 fwrite (line->buffer, sizeof (char), line->length, stream);
202 /* Process input file INFILE with output to OUTFILE.
203 If either is "-", use the standard I/O stream for it instead. */
206 check_file (const char *infile, const char *outfile)
210 struct linebuffer lb1, lb2;
211 struct linebuffer *thisline, *prevline, *exch;
212 char *prevfield, *thisfield;
213 int prevlen, thislen;
216 if (!strcmp (infile, "-"))
219 istream = fopen (infile, "r");
221 error (1, errno, "%s", infile);
223 if (!strcmp (outfile, "-"))
226 ostream = fopen (outfile, "w");
228 error (1, errno, "%s", outfile);
233 initbuffer (thisline);
234 initbuffer (prevline);
236 if (readline (prevline, istream) == 0)
238 prevfield = find_field (prevline);
239 prevlen = prevline->length - (prevfield - prevline->buffer);
241 while (!feof (istream))
243 if (readline (thisline, istream) == 0)
245 thisfield = find_field (thisline);
246 thislen = thisline->length - (thisfield - thisline->buffer);
247 if (!different (thisfield, prevfield, thislen, prevlen))
251 writeline (prevline, ostream, match_count);
257 prevfield = thisfield;
262 writeline (prevline, ostream, match_count);
265 if (ferror (istream) || fclose (istream) == EOF)
266 error (1, errno, _("error reading %s"), infile);
268 if (ferror (ostream) || fclose (ostream) == EOF)
269 error (1, errno, _("error writing %s"), outfile);
276 main (int argc, char **argv)
279 char *infile = "-", *outfile = "-";
281 program_name = argv[0];
282 setlocale (LC_ALL, "");
283 bindtextdomain (PACKAGE, LOCALEDIR);
284 textdomain (PACKAGE);
290 countmode = count_none;
292 while ((optc = getopt_long (argc, argv, "0123456789cdf:s:uw:", longopts,
310 skip_fields = skip_fields * 10 + optc - '0';
314 countmode = count_occurrences;
318 mode = output_repeated;
321 case 'f': /* Like '-#'. */
324 if (xstrtol (optarg, NULL, 10, &tmp_long, NULL) != LONGINT_OK
325 || tmp_long <= 0 || tmp_long > INT_MAX)
326 error (1, 0, _("invalid number of fields to skip: `%s'"),
328 skip_fields = (int) tmp_long;
332 case 's': /* Like '+#'. */
335 if (xstrtol (optarg, NULL, 10, &tmp_long, NULL) != LONGINT_OK
336 || tmp_long <= 0 || tmp_long > INT_MAX)
337 error (1, 0, _("invalid number of bytes to skip: `%s'"),
339 skip_chars = (int) tmp_long;
344 mode = output_unique;
350 if (xstrtol (optarg, NULL, 10, &tmp_long, NULL) != LONGINT_OK
351 || tmp_long <= 0 || tmp_long > INT_MAX)
352 error (1, 0, _("invalid number of bytes to compare: `%s'"),
354 check_chars = (int) tmp_long;
365 printf ("uniq - %s\n", PACKAGE_VERSION);
372 if (optind >= 2 && strcmp (argv[optind - 1], "--") != 0)
374 /* Interpret non-option arguments with leading `+' only
375 if we haven't seen `--'. */
376 while (optind < argc && argv[optind][0] == '+')
378 char *opt_str = argv[optind++];
380 if (xstrtol (opt_str, NULL, 10, &tmp_long, NULL) != LONGINT_OK
381 || tmp_long <= 0 || tmp_long > INT_MAX)
382 error (1, 0, _("invalid number of bytes to compare: `%s'"),
384 skip_chars = (int) tmp_long;
389 infile = argv[optind++];
392 outfile = argv[optind++];
395 usage (1); /* Extra arguments. */
397 check_file (infile, outfile);