locate/bigram.c

   1 /* bigram -- list bigrams for locate
   2    Copyright (C) 1994 Free Software Foundation, Inc.
   3
   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)
   7    any later version.
   8
   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.
  13
  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., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301,
  17    USA.
  18 */
  19
  20 /* Usage: bigram < text > bigrams
  21    Use `code' to encode a file using this output.
  22
  23    Read a file from stdin and write out the bigrams (pairs of
  24    adjacent characters), one bigram per line, to stdout.  To reduce
  25    needless duplication in the output, it starts finding the
  26    bigrams on each input line at the character where that line
  27    first differs from the previous line (i.e., in the ASCII
  28    remainder).  Therefore, the input should be sorted in order to
  29    get the least redundant output.
  30
  31    Written by James A. Woods <jwoods@adobe.com>.
  32    Modified by David MacKenzie <djm@gnu.ai.mit.edu>.  */
  33
  34 #include <config.h>
  35 #include <stdio.h>
  36
  37 #if defined(HAVE_STRING_H) || defined(STDC_HEADERS)
  38 #include <string.h>
  39 #else
  40 #include <strings.h>
  41 #endif
  42
  43 #ifdef STDC_HEADERS
  44 #include <stdlib.h>
  45 #endif
  46 #include <sys/types.h>
  47
  48 #include <getline.h>
  49 #include <xalloc.h>
  50 #include "closeout.h"
  51
  52 /* The name this program was run with.  */
  53 char *program_name;
  54
  55 /* Return the length of the longest common prefix of strings S1 and S2. */
  56
  57 static int
  58 prefix_length (char *s1, char *s2)
  59 {
  60   register char *start;
  61
  62   for (start = s1; *s1 == *s2 && *s1 != '\0'; s1++, s2++)
  63     ;
  64   return s1 - start;
  65 }
  66
  67 int
  68 main (int argc, char **argv)
  69 {
  70   char *path;                   /* The current input entry.  */
  71   char *oldpath;                /* The previous input entry.  */
  72   size_t pathsize, oldpathsize; /* Amounts allocated for them.  */
  73   int line_len;                 /* Length of input line.  */
  74
  75   program_name = argv[0];
  76   (void) argc;
  77   atexit (close_stdout);
  78
  79   pathsize = oldpathsize = 1026; /* Increased as necessary by getline.  */
  80   path = xmalloc (pathsize);
  81   oldpath = xmalloc (oldpathsize);
  82
  83   /* Set to empty string, to force the first prefix count to 0.  */
  84   oldpath[0] = '\0';
  85
  86   while ((line_len = getline (&path, &pathsize, stdin)) > 0)
  87     {
  88       register int count;       /* The prefix length.  */
  89       register int j;           /* Index into input line.  */
  90
  91       path[line_len - 1] = '\0'; /* Remove the newline. */
  92
  93       /* Output bigrams in the remainder only. */
  94       count = prefix_length (oldpath, path);
  95       for (j = count; path[j] != '\0' && path[j + 1] != '\0'; j += 2)
  96         {
  97           putchar (path[j]);
  98           putchar (path[j + 1]);
  99           putchar ('\n');
 100         }
 101
 102       {
 103         /* Swap path and oldpath and their sizes.  */
 104         char *tmppath = oldpath;
 105         size_t tmppathsize = oldpathsize;
 106         oldpath = path;
 107         oldpathsize = pathsize;
 108         path = tmppath;
 109         pathsize = tmppathsize;
 110       }
 111     }
 112
 113   free (path);
 114   free (oldpath);
 115
 116   return 0;
 117 }