Home | History | Annotate | Line # | Download | only in mkindex
mkindex.c revision 1.5 
      1 /*	$NetBSD: mkindex.c,v 1.5 1999/09/10 00:18:21 jsm Exp $	*/
      2 
      3 /*-
      4  * Copyright (c) 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  * Barry Brachman.
      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 #ifndef lint
     40 static char copyright[] =
     41 "@(#) Copyright (c) 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[] = "@(#)mkindex.c	8.1 (Berkeley) 6/11/93";
     48 #else
     49 static char rcsid[] = "$NetBSD: mkindex.c,v 1.5 1999/09/10 00:18:21 jsm Exp $";
     50 #endif
     51 #endif /* not lint */
     52 
     53 #include <err.h>
     54 #include <stdio.h>
     55 
     56 #include "bog.h"
     57 
     58 char *nextword __P((FILE *, char *, int *, int *));
     59 
     60 int
     61 main(void)
     62 {
     63 	int clen, rlen, prev;
     64 	long off, start;
     65 	char buf[MAXWORDLEN + 1];
     66 
     67 	prev = '\0';
     68 	off = start = 0L;
     69 	while (nextword(stdin, buf, &clen, &rlen) != NULL) {
     70 		if (*buf != prev) {
     71 			if (prev != '\0')
     72 				printf("%c %6ld %6ld\n", prev, start, off - 1);
     73 			prev = *buf;
     74 			start = off;
     75 		}
     76 		off += clen + 1;
     77 	}
     78 	printf("%c %6ld %6ld\n", prev, start, off - 1);
     79 	fflush(stdout);
     80 	if (ferror(stdout))
     81 		err(1, "writing standard output");
     82 	exit(0);
     83 }
     84 
     85 /*
     86  * Return the next word in the compressed dictionary in 'buffer' or
     87  * NULL on end-of-file
     88  * Also set clen to the length of the compressed word (for mkindex) and
     89  * rlen to the strlen() of the real word
     90  */
     91 char *
     92 nextword(fp, buffer, clen, rlen)
     93 	FILE *fp;
     94 	char *buffer;
     95 	int *clen, *rlen;
     96 {
     97 	int ch, pcount;
     98 	char *p, *q;
     99 	static char buf[MAXWORDLEN + 1];
    100 	static int first = 1;
    101 	static int lastch = 0;
    102 
    103    	if (first) {
    104 		if ((pcount = getc(fp)) == EOF)
    105 			return (NULL);
    106 		first = 0;
    107 	}
    108 	else if ((pcount = lastch) == EOF)
    109 		return (NULL);
    110 
    111 	p = buf + (*clen = pcount);
    112 
    113 	while ((ch = getc(fp)) != EOF && ch >= 'a')
    114 			*p++ = ch;
    115 		lastch = ch;
    116 	*p = '\0';
    117 
    118 	*rlen = (int) (p - buf);
    119 	*clen = *rlen - *clen;
    120 
    121 	p = buf;
    122 	q = buffer;
    123 	while ((*q++ = *p) != '\0') {
    124 		if (*p++ == 'q')
    125 			*q++ = 'u';
    126 	}
    127 	return (buffer);
    128 }
    129