v02i070: sq - squeeze a dictionary

[Mike Wexler]

Submitted-By: "Mike Wexler" <mikew at wyse1.wyse.com>

Archive-Name: sq


Comp.sources.misc: Volume 2, Issue 70
Submitted-By: "Mike Wexler" <mikew at wyse1.wyse.com>
Archive-Name: sq

[A couple programs that sat around until I had time to fix 'em up.  ++bsa]

The following program compresses dictionaries.  It works in combination
with compress and sort.  You should have both of these for optimal use.
I have included a simple demo.  Have fun...
#--------------------------------CUT HERE-------------------------------------
#! /bin/sh
#
# This is a shell archive.  Save this into a file, edit it
# and delete all lines above this comment.  Then give this
# file to sh by executing the command "sh file".  The files
# will be extracted into the current directory owned by
# you with default permissions.
#
# The files contained herein are:
#
# -rw-r--r--   1 allbery  System       449 Mar  5 18:43 Makefile
# -rw-r--r--   1 allbery  System       451 Mar  5 18:43 Makefile.orig
# -rw-r--r--   1 allbery  System       546 Mar  5 18:43 sq.c
# -rw-r--r--   1 allbery  System       606 Mar  5 18:43 sq.l
# -rw-r--r--   1 allbery  System       468 Mar  5 18:43 unsq.c
#
echo 'x - Makefile'
if test -f Makefile; then echo 'shar: not overwriting Makefile'; else
sed 's/^X//' << '________This_Is_The_END________' > Makefile
XCFLAGS=-O
XBINDIR=/usr/new
XMANDIR=/usr/man/manl
XDICT=/usr/dict/words
Xall: sq unsq
Xsq: sq.c
X	$(CC) $(CFLAGS) -o sq sq.c
Xunsq: unsq.c
X	$(CC) $(CFLAGS) -o unsq unsq.c
Xinstall:
X	cp sq unsq $(BINDIR)
X	cp sq.l $(MANDIR)
Xshar:
X	shar sq.l sq.c unsq.c Makefile > sq.shar
Xdemo:
X	sort $(DICT) | sq | compress > dict.sq.Z
X	compress < $(DICT) > dict.Z
X	ls -l $(DICT) dict.Z dict.sq.Z
X	compress -d < dict.sq.Z | unsq | sort -df > dict.copy
X	diff $(DICT) dict.copy
________This_Is_The_END________
if test `wc -c < Makefile` -ne 449; then
	echo 'shar: Makefile was damaged during transit (should have been 449 bytes)'
fi
fi		; : end of overwriting check
echo 'x - Makefile.orig'
if test -f Makefile.orig; then echo 'shar: not overwriting Makefile.orig'; else
sed 's/^X//' << '________This_Is_The_END________' > Makefile.orig
XCFLAGS=-O
XBINDIR=/usr/local/bin
XMANDIR=/usr/man/manl
XDICT=/usr/dict/words
Xall: sq unsq
Xsq: sq.c
X	$(CC) $(CFLAGS) -o sq sq.c
Xunsq: unsq.c
X	$(CC) $(CFLAGS) -o unsq unsq.c
Xinstall:
X	cp sq unsq $BINDIR
X	cp sq.l $MANDIR
Xshar:
X	shar sq.l sq.c unsq.c Makefile > sq.shar
Xdemo:
X	sort $(DICT) | sq | compress > dict.sq.Z
X	compress < $(DICT) > dict.Z
X	ls -l $(DICT) dict.Z dict.sq.Z
X	compress -d < dict.sq.Z | unsq | sort -df > dict.copy
X	diff $(DICT) dict.copy
________This_Is_The_END________
if test `wc -c < Makefile.orig` -ne 451; then
	echo 'shar: Makefile.orig was damaged during transit (should have been 451 bytes)'
fi
fi		; : end of overwriting check
echo 'x - sq.c'
if test -f sq.c; then echo 'shar: not overwriting sq.c'; else
sed 's/^X//' << '________This_Is_The_END________' > sq.c
X#include <stdio.h>
X
Xmain()
X{
X  char word[257];
X  static char prev[257]="";
X  char outword[258];
X
X  while (gets(word)!=NULL) {
X    trunc(word,prev);
X  }
Xexit(0);
X}
X
Xtrunc(word,prev) 
Xchar *word;
Xchar *prev;
X{
X  unsigned char same_count;
X  char *wordp;
X  char *prevp;
X
X  wordp=word;
X  prevp=prev;
X  for (same_count=0;*wordp==*prevp++;++wordp,++same_count);
X  if (same_count>255) {
X    fprintf(stderr,"same count exceeded 255 characters, aborted");
X    exit(1);
X  }
X  fwrite(&same_count,1,1,stdout);
X  puts(wordp);
X  strcpy(prev,word);
X  return;
X}
X
________This_Is_The_END________
if test `wc -c < sq.c` -ne 546; then
	echo 'shar: sq.c was damaged during transit (should have been 546 bytes)'
fi
fi		; : end of overwriting check
echo 'x - sq.l'
if test -f sq.l; then echo 'shar: not overwriting sq.l'; else
sed 's/^X//' << '________This_Is_The_END________' > sq.l
X.TH SQ 1 LOCAL
X.\" $Header: unshar.man,v 1.1 87/02/27 13:45:34 rs Exp $
X.SH NAME
Xsq \- squeeze a sorted word list
Xunsq \- unsqueeze a sorted word list
X.SH SYNOPSIS
X.B sq
X< infile > outfile
X.PP
X.B unsq
X< infile > outfile
X.SH DESCRIPTION
X.I sq
Xcompresses a sorted list of words(a dictionary). 
XFor example:
X.RS
Xsort /usr/dict/words | sq | compress > words.sq.Z
X.RE
Xwill compress dict by about a factor of 4.
X.PP
X.I unsq
Xuncompress the output of
X.I sq.
XFor example:
X.RS
Xcompress -d < words.sq.Z | unsq | sort -f -o words
X.RE
Xwill uncompress a dictionary compressed with
X.I sq.
X.SH SEE ALSO
Xcompress(1), sort.
________This_Is_The_END________
if test `wc -c < sq.l` -ne 606; then
	echo 'shar: sq.l was damaged during transit (should have been 606 bytes)'
fi
fi		; : end of overwriting check
echo 'x - unsq.c'
if test -f unsq.c; then echo 'shar: not overwriting unsq.c'; else
sed 's/^X//' << '________This_Is_The_END________' > unsq.c
X#include <stdio.h>
X
Xmain()
X{
X  char word[257];
X  static char prev[257]="";
X  unsigned char count;
X  char outword[258];
X
X  while (!expand(word,prev)) {
X    puts(word);
X  }
Xexit(0);
X}
X
Xexpand(word,prev) 
Xchar *word;
Xchar *prev;
X{
X  unsigned char same_count;
X  char *wordp;
X  char *prevp;
X
X  fread(&same_count,1,1,stdin);
X  prevp=prev;
X  wordp=word;
X  while (same_count--)
X    *wordp++=(*prevp++);
X  if (gets(wordp)==NULL) return(1);
X  strcpy(prev,word);
X  return(0);
X}
X
________This_Is_The_END________
if test `wc -c < unsq.c` -ne 468; then
	echo 'shar: unsq.c was damaged during transit (should have been 468 bytes)'
fi
fi		; : end of overwriting check
exit 0