boneill@hawk.ulowell.edu (SoftXc Coordinator) (03/26/88)
--------------Cut Here-----------
: This is a shar archive. Extract with sh, not csh.
echo x - arcpack.c
cat > arcpack.c << '!Funky!Stuff!'
static char *RCSid = "$Header: arcpack.c,v 1.2 86/07/15 07:53:48 turner Exp $";
/*
* $Log: arcpack.c,v $
* Hack-attack 1.3 86/12/20 01:23:45 wilhite@usceast.uucp
* Bludgeoned into submission for VAX 11/780 BSD4.2
* (ugly code, but fewer core dumps)
*
* Revision 1.2 86/07/15 07:53:48 turner
*
*
* Revision 1.1 86/06/26 15:00:37 turner
* initial version
*
*
*/
/* ARC - Archive utility - ARCPACK
$define(tag,$$segment(@1,$$index(@1,=)+1))#
$define(version,Version $tag(
TED_VERSION DB =3.37), created on $tag(
TED_DATE DB =02/03/86) at $tag(
TED_TIME DB =22:58:01))#
$undefine(tag)#
$version
(C) COPYRIGHT 1985 by System Enhancement Associates; ALL RIGHTS RESERVED
By: Thom Henderson
Description:
This file contains the routines used to compress a file
when placing it in an archive.
Language:
Computer Innovations Optimizing C86
*/
#include <stdio.h>
#include "arc.h"
/* stuff for non-repeat packing */
#define DLE 0x90 /* repeat sequence marker */
static unsigned char state; /* current packing state */
/* non-repeat packing states */
#define NOHIST 0 /* don't consider previous input*/
#define SENTCHAR 1 /* lastchar set, no lookahead yet */
#define SENDNEWC 2 /* run over, send new char next */
#define SENDCNT 3 /* newchar set, send count next */
/* packing results */
static long stdlen; /* length for standard packing */
static INT crcval; /* CRC check value */
INT pack(f,t,hdr) /* pack file into an archive */
FILE *f, *t; /* source, destination */
struct heads *hdr; /* pointer to header data */
{
INT c; /* one character of stream */
long ncrlen; /* length after packing */
long huflen; /* length after squeezing */
long lzwlen; /* length after crunching */
long pred_sq(), file_sq(); /* stuff for squeezing */
long pred_cm(); /* dynamic crunching cleanup */
char tnam[STRLEN]; /* temporary name buffer */
char *makefnam(); /* filename fixer upper */
FILE *crn = NULL; /* temporary crunch file */
INT getch();
INT getc_ncr();
INT putc_pak();
/* first pass - see which method is best */
if(!nocomp) /* if storage kludge not active */
{ if(note)
{ printf(" analyzing, "); fflush(stdout);}
if(arctemp) /* use temp area if specified */
sprintf(tnam,"%s.CRN",arctemp);
else makefnam("$ARCTEMP.CRN",arcname,tnam);
#if MSDOS
crn = fopen(tnam,"wrb");
#endif
#if BSD | ST
crn = fopen(tnam,"w+");
#endif
state = NOHIST; /* initialize ncr packing */
stdlen = ncrlen = 0; /* reset size counters */
crcval = 0; /* initialize CRC check value */
setcode(); /* initialize encryption */
init_cm(f,crn); /* initialize for crunching */
init_sq(); /* initialize for squeeze scan */
while((c=getc_ncr(f))!=EOF) /* for each byte of file */
{ ncrlen++; /* one more packed byte */
scan_sq(c); /* see what squeezing can do */
putc_cm(c,crn); /* see what crunching can do */
}
huflen = pred_sq(); /* finish up after squeezing */
lzwlen = pred_cm(crn); /* finish up after crunching */
}
else /* else kludge the method */
{ stdlen = 0; /* make standard look best */
ncrlen = huflen = lzwlen = 1;
}
/* standard set-ups common to all methods */
fseek(f,0L,0); /* rewind input */
hdr->crc = crcval; /* note CRC check value */
hdr->length = stdlen; /* set actual file length */
state = NOHIST; /* reinitialize ncr packing */
setcode(); /* reinitialize encryption */
/* choose and use the shortest method */
if(stdlen<=ncrlen && stdlen<=huflen && stdlen<=lzwlen)
{ if(kludge) /*DEBUG*/
printf("(%ld) ",lzwlen-stdlen);
if(note)
{ printf("storing, "); fflush(stdout);}/* store w/out compression */
hdrver = 2; /* note packing method */
stdlen = crcval = 0; /* recalc these for kludge */
while((c=getch(f))!=EOF) /* store it straight */
putc_pak(c,t);
hdr->crc = crcval;
hdr->length = hdr->size = stdlen;
}
else if(ncrlen<huflen && ncrlen<lzwlen)
{ if(kludge) /*DEBUG*/
printf("(%ld) ",lzwlen-ncrlen);
if(note)
{ printf("packing, ");
fflush(stdout);} /* pack w/repeat suppression */
hdrver = 3; /* note packing method */
hdr->size = ncrlen; /* set data length */
while((c=getc_ncr(f))!=EOF)
putc_pak(c,t);
}
else if(huflen<lzwlen)
{ if(kludge) /*DEBUG*/
printf("(%ld) ",lzwlen-huflen);
if(note)
{ printf("squeezing, "); fflush(stdout);}
hdrver = 4; /* note packing method */
hdr->size = file_sq(f,t); /* note final size */
}
else
{ if(kludge) /*DEBUG*/
printf("(%ld) ",huflen-lzwlen);
if(note)
{ printf("crunching, "); fflush(stdout);}
hdrver = 8;
hdr->size = lzwlen; /* size should not change */
if(crn) /* if temp was created */
{ fseek(crn,0L,0); /* then copy over crunched temp */
while((c=fgetc(crn))!=EOF)
putc_tst(c,t);
}
else /* else re-crunch */
{ init_cm(f,t);
while((c=getc_ncr(f))!=EOF)
putc_cm(c,t);
pred_cm(t); /* finish up after crunching */
}
}
/* standard cleanups common to all methods */
if(crn) /* get rid of crunch temporary */
{ fclose(crn);
if(unlink(tnam) && warn)
{ printf("Cannot delete temporary file %s\n",tnam);
nerrs++;
}
}
if(note)
printf("done.\n");
}
/* Non-repeat compression - text is passed through normally, except that
a run of more than two is encoded as:
<char> <DLE> <count>
Special case: a count of zero indicates that the DLE is really a DLE,
not a repeat marker.
*/
INT getc_ncr(f) /* get bytes with collapsed runs */
FILE *f; /* file to get from */
{
static INT lastc; /* value returned on last call */
static INT repcnt; /* repetition counter */
static INT c; /* latest value seen */
switch(state) /* depends on our state */
{
case NOHIST: /* no relevant history */
state = SENTCHAR;
return lastc = getch(f); /* remember the value next time */
case SENTCHAR: /* char was sent. look ahead */
switch(lastc) /* action depends on char */
{
case DLE: /* if we sent a real DLE */
state = NOHIST; /* then start over again */
return 0; /* but note that the DLE was real */
case EOF: /* EOF is always a special case */
return EOF;
default: /* else test for a repeat */
for(repcnt=1; (c=getch(f))==lastc && repcnt<255; repcnt++)
; /* find end of run */
switch(repcnt) /* action depends on run size */
{
case 1: /* not a repeat */
return lastc = c; /* but remember value next time */
case 2: /* a repeat, but too short */
state = SENDNEWC; /* send the second one next time */
return lastc;
default: /* a run - compress it */
state = SENDCNT; /* send repeat count next time */
return DLE; /* send repeat marker this time */
}
}
case SENDNEWC: /* send second char of short run */
state = SENTCHAR;
return lastc = c;
case SENDCNT: /* sent DLE, now send count */
state = SENDNEWC;
return repcnt;
default:
abort("Bug - bad ncr state\n");
}
}
static INT getch(f) /* special get char for packing */
FILE *f; /* file to get from */
{
INT c; /* a char from the file */
if((c=fgetc(f))!=EOF) /* if not the end of file */
{ crcval = addcrc(crcval,c); /* then update CRC check value */
stdlen++; /* and bump length counter */
}
return c;
}
INT putc_pak(c,f) /* put a packed byte into archive */
char c; /* byte to put */
FILE *f; /* archive to put it in */
{
putc_tst(code(c),f); /* put encoded byte, with checks */
}
!Funky!Stuff!
echo x - arcrun.c
cat > arcrun.c << '!Funky!Stuff!'
static char *RCSid = "$Header: arcrun.c,v 1.2 86/07/15 07:53:55 turner Exp $";
/*
* $Log: arcrun.c,v $
* Hack-attack 1.3 86/12/20 01:23:45 wilhite@usceast.uucp
* Bludgeoned into submission for VAX 11/780 BSD4.2
* (ugly code, but fewer core dumps)
*
* Revision 1.2 86/07/15 07:53:55 turner
*
*
* Revision 1.1 86/06/26 15:00:40 turner
* initial version
*
*
*/
/* ARC - Archive utility - ARCRUN
$define(tag,$$segment(@1,$$index(@1,=)+1))#
$define(version,Version $tag(
TED_VERSION DB =1.17), created on $tag(
TED_DATE DB =02/03/86) at $tag(
TED_TIME DB =22:59:06))#
$undefine(tag)#
$version
(C) COPYRIGHT 1985 by System Enhancement Associates; ALL RIGHTS RESERVED
By: Thom Henderson
Description:
This file contains the routines used to "run" a file
which is stored in an archive. At present, all we really do
is (a) extract a temporary file, (b) give its name as a system
command, and then (c) delete the file.
Language:
Computer Innovations Optimizing C86
*/
#include <stdio.h>
#include "arc.h"
INT runarc(num,arg) /* run file from archive */
INT num; /* number of arguments */
char *arg[]; /* pointers to arguments */
{
struct heads hdr; /* file header */
INT run; /* true to run current file */
INT did[MAXARG]; /* true when argument was used */
INT n; /* index */
char *makefnam(); /* filename fixer */
char buf[STRLEN]; /* filename buffer */
FILE *fopen(); /* file opener */
INT runfile();
for(n=0; n<num; n++) /* for each argument */
did[n] = 0; /* reset usage flag */
rempath(num,arg); /* strip off paths */
openarc(0); /* open archive for reading */
if(num) /* if files were named */
{ while(readhdr(&hdr,arc)) /* while more files to check */
{ run = 0; /* reset run flag */
for(n=0; n<num; n++) /* for each template given */
{ if(match(hdr.name,makefnam(arg[n],".*",buf)))
{ run = 1; /* turn on run flag */
did[n] = 1; /* turn on usage flag */
break; /* stop looking */
}
}
if(run) /* if running this one */
runfile(&hdr); /* then do it */
else /* else just skip it */
fseek(arc,hdr.size,1);
}
}
else while(readhdr(&hdr,arc)) /* else run all files */
runfile(&hdr);
closearc(0); /* close archive after changes */
if(note)
{ for(n=0; n<num; n++) /* report unused args */
{ if(!did[n])
{ printf("File not found: %s\n",arg[n]);
nerrs++;
}
}
}
}
static INT runfile(hdr) /* run a file */
struct heads *hdr; /* pointer to header data */
{
FILE *tmp, *fopen(); /* temporary file */
char buf[STRLEN], *makefnam(); /* temp file name, fixer */
char sys[STRLEN]; /* invocation command buffer */
char *dir, *gcdir(); /* directory stuff */
/* Replaced $ARCTEMP as the "system" call didn't approve. */
/* makefnam("$ARCTEMP",hdr->name,buf); */
sprintf(buf,"%s.RUN",arctemp);
/* if(!strcmp(buf,"$ARCTEMP.BAS"))
* strcpy(sys,"BASICA $ARCTEMP");
*
* else if(!strcmp(buf,"$ARCTEMP.BAT")
* || !strcmp(buf,"$ARCTEMP.COM")
* || !strcmp(buf,"$ARCTEMP.EXE"))
* strcpy(sys,"$ARCTEMP");
*
* else
* { if(warn)
* { printf("File %s is not a .BAS, .BAT, .COM, or .EXE\n",
* hdr->name);
* nerrs++;
* }
* fseek(arc,hdr->size,1);
* return;
* }
*/
if(warn)
if(tmp=fopen(buf,"r"))
abort("Temporary file %s already exists",buf);
if(!(tmp=fopen(buf,"w")))
abort("Unable to create temporary file %s",buf);
if(note)
printf("Invoking file: %s\n",hdr->name);
dir = gcdir(""); /* see where we are */
unpack(arc,tmp,hdr); /* unpack the entry */
fclose(tmp); /* release the file */
chmod(buf,"700"); /* make it exec'able */
system(buf); /* try to invoke it */
chdir(dir); free(dir); /* return to whence we started */
if(unlink(buf) && warn)
{ printf("Cannot unsave temporary file %s\n",buf);
nerrs++;
}
}
!Funky!Stuff!
echo x - arcs.h
cat > arcs.h << '!Funky!Stuff!'
/*
* $Header: arcs.h,v 1.2 86/07/15 07:54:02 turner Exp $
*/
/*
* $Log: arcs.h,v $
* Revision 1.2 86/07/15 07:54:02 turner
*
*
* Revision 1.1 86/06/26 15:01:27 turner
* initial version
*
*
*/
/* ARC - Archive utility - Archive file header format
Version 2.12, created on 12/17/85 at 14:40:26
(C) COPYRIGHT 1985 by System Enhancement Associates; ALL RIGHTS RESERVED
By: Thom Henderson
Description:
This file defines the format of an archive file header, excluding
the archive marker and the header version number.
Each entry in an archive begins with a one byte archive marker,
which is set to 26. The marker is followed by a one byte
header type code, from zero to 7.
If the header type code is zero, then it is an end marker, and
no more data should be read from the archive.
If the header type code is in the range 2 to 7, then it is
followed by a standard archive header, which is defined below.
If the header type code is one, then it is followed by an older
format archive header. The older format header does not contain
the true length. A header should be read for a length of
sizeof(struct heads)-sizeof(long). Then set length equal to size
and change the header version to 2.
Programming note:
The crc value given in the header is based on the unpacked data.
Language:
Computer Innovations Optimizing C86
*/
struct heads /* archive entry header format */
{ char name[FNLEN]; /* file name */
long size; /* size of file, in bytes */
unsigned INT date; /* creation date */
unsigned INT time; /* creation time */
INT crc; /* cyclic redundancy check */
long length; /* true file length */
} ;
!Funky!Stuff!
echo x - arcsq.c
cat > arcsq.c << '!Funky!Stuff!'
static char *RCSid = "$Header: arcsq.c,v 1.2 86/07/15 07:54:05 turner Exp $";
/*
* $Log: arcsq.c,v $
* Hack-attack 1.3 86/12/20 01:23:45 wilhite@usceast.uucp
* Bludgeoned into submission for VAX 11/780 BSD4.2
* (ugly code, but fewer core dumps)
*
* Revision 1.2 86/07/15 07:54:05 turner
*
*
* Revision 1.1 86/06/26 15:00:48 turner
* initial version
*
*
*/
/* ARC - Archive utility - ARCSQ
$define(tag,$$segment(@1,$$index(@1,=)+1))#
$define(version,Version $tag(
TED_VERSION DB =3.10), created on $tag(
TED_DATE DB =01/30/86) at $tag(
TED_TIME DB =20:10:46))#
$undefine(tag)#
$version
(C) COPYRIGHT 1985 by System Enhancement Associates; ALL RIGHTS RESERVED
By: Thom Henderson
Description:
This file contains the routines used to squeeze a file
when placing it in an archive.
Language:
Computer Innovations Optimizing C86
Programming notes:
Most of the routines used for the Huffman squeezing algorithm
were lifted from the SQ program by Dick Greenlaw, as adapted
to CI-C86 by Robert J. Beilstein.
*/
#include <stdio.h>
#include "arc.h"
/* stuff for Huffman squeezing */
#define TRUE 1
#define FALSE 0
#define ERROR (-1)
#define SPEOF 256 /* special endfile token */
#define NOCHILD (-1) /* marks end of path through tree */
#define NUMVALS 257 /* 256 data values plus SPEOF*/
#define NUMNODES (NUMVALS+NUMVALS-1) /* number of nodes */
#define MAXCOUNT (unsigned INT) 65535 /* biggest unsigned integer */
/* The following array of structures are the nodes of the
binary trees. The first NUMVALS nodes become the leaves of the
final tree and represent the values of the data bytes being
encoded and the special endfile, SPEOF.
The remaining nodes become the internal nodes of the final tree.
*/
struct nd /* shared by unsqueezer */
{ unsigned INT weight; /* number of appearances */
INT tdepth; /* length on longest path in tree */
INT lchild, rchild; /* indices to next level */
} node[NUMNODES]; /* use large buffer */
static INT dctreehd; /* index to head of final tree */
/* This is the encoding table:
The bit strings have first bit in low bit.
Note that counts were scaled so code fits unsigned integer.
*/
static INT codelen[NUMVALS]; /* number of bits in code */
static unsigned INT code[NUMVALS]; /* code itself, right adjusted */
static unsigned INT tcode; /* temporary code value */
static long valcount[NUMVALS]; /* actual count of times seen */
/* Variables used by encoding process */
static INT curin; /* value currently being encoded */
static INT cbitsrem; /* # of code string bits left */
static unsigned INT ccode; /* current code right justified */
INT init_sq() /* prepare for scanning pass */
{
INT i; /* node index */
/* Initialize all nodes to single element binary trees
with zero weight and depth.
*/
for(i=0; i<NUMNODES; ++i)
{ node[i].weight = 0;
node[i].tdepth = 0;
node[i].lchild = NOCHILD;
node[i].rchild = NOCHILD;
}
for(i=0; i<NUMVALS; i++)
valcount[i] = 0;
}
INT scan_sq(c) /* add a byte to the tables */
INT c; /* byte to add */
{
unsigned INT *wp; /* speeds up weight counting */
/* Build frequency info in tree */
if(c == EOF) /* it's traditional */
c = SPEOF; /* dumb, but traditional */
if(*(wp = &node[c].weight) != MAXCOUNT)
++(*wp); /* bump weight counter */
valcount[c]++; /* bump byte counter */
}
long pred_sq() /* predict size of squeezed file */
{
INT i;
INT btlist[NUMVALS]; /* list of intermediate b-trees */
INT listlen; /* length of btlist */
unsigned INT ceiling; /* limit for scaling */
long size = 0; /* predicted size */
INT numnodes; /* # of nodes in simplified tree */
INT scale();
INT heap();
INT bld_tree();
INT buildenc();
INT init_enc();
scan_sq(EOF); /* signal end of input */
ceiling = MAXCOUNT;
/* Keep trying to scale and encode */
do
{ scale(ceiling);
ceiling /= 2; /* in case we rescale */
/* Build list of single node binary trees having
leaves for the input values with non-zero counts
*/
for(i=listlen=0; i<NUMVALS; ++i)
{ if(node[i].weight != 0)
{ node[i].tdepth = 0;
btlist[listlen++] = i;
}
}
/* Arrange list of trees into a heap with the entry
indexing the node with the least weight at the top.
*/
heap(btlist,listlen);
/* Convert the list of trees to a single decoding tree */
bld_tree(btlist,listlen);
/* Initialize the encoding table */
init_enc();
/* Try to build encoding table.
Fail if any code is > 16 bits long.
*/
} while(buildenc(0,dctreehd) == ERROR);
/* Initialize encoding variables */
cbitsrem = 0; /* force initial read */
curin = 0; /* anything but endfile */
for(i=0; i<NUMVALS; i++) /* add bits for each code */
size += valcount[i] * codelen[i];
size = (size+7)/8; /* reduce to number of bytes */
numnodes = dctreehd<NUMVALS ? 0 : dctreehd-(NUMVALS-1);
size += sizeof(INT) + 2*numnodes*sizeof(INT);
return size;
}
/* The count of number of occurrances of each input value
have already been prevented from exceeding MAXCOUNT.
Now we must scale them so that their sum doesn't exceed
ceiling and yet no non-zero count can become zero.
This scaling prevents errors in the weights of the
interior nodes of the Huffman tree and also ensures that
the codes will fit in an unsigned integer. Rescaling is
used if necessary to limit the code length.
*/
static INT scale(ceil)
unsigned INT ceil; /* upper limit on total weight */
{
register INT i,c;
INT ovflw, divisor;
unsigned INT w, sum;
unsigned char increased; /* flag */
do
{ for(i=sum=ovflw=0; i<NUMVALS; ++i)
{ if(node[i].weight > (ceil-sum))
++ovflw;
sum += node[i].weight;
}
divisor = ovflw + 1;
/* Ensure no non-zero values are lost */
increased = FALSE;
for(i=0; i<NUMVALS; ++i)
{ w = node[i].weight;
if(w<divisor && w!=0)
{ /* Don't fail to provide a code if it's used at all */
node[i].weight = divisor;
increased = TRUE;
}
}
} while(increased);
/* Scaling factor choosen, now scale */
if(divisor>1)
for(i=0; i<NUMVALS; ++i)
node[i].weight /= divisor;
}
/* heap() and adjust() maintain a list of binary trees as a
heap with the top indexing the binary tree on the list
which has the least weight or, in case of equal weights,
least depth in its longest path. The depth part is not
strictly necessary, but tends to avoid long codes which
might provoke rescaling.
*/
static INT heap(list,length)
INT list[], length;
{
register INT i;
INT adjust();
for(i=(length-2)/2; i>=0; --i)
adjust(list,i,length-1);
}
/* Make a heap from a heap with a new top */
static INT adjust(list,top,bottom)
INT list[], top, bottom;
{
register INT k, temp;
INT cmptrees();
k = 2 * top + 1; /* left child of top */
temp = list[top]; /* remember root node of top tree */
if(k<=bottom)
{ if(k<bottom && cmptrees(list[k],list[k+1]))
++k;
/* k indexes "smaller" child (in heap of trees) of top */
/* now make top index "smaller" of old top and smallest child */
if(cmptrees(temp,list[k]))
{ list[top] = list[k];
list[k] = temp;
/* Make the changed list a heap */
adjust(list,k,bottom); /* recursive */
}
}
}
/* Compare two trees, if a > b return true, else return false.
Note comparison rules in previous comments.
*/
static INT cmptrees(a,b)
INT a, b; /* root nodes of trees */
{
if(node[a].weight > node[b].weight)
return TRUE;
if(node[a].weight == node[b].weight)
if(node[a].tdepth > node[b].tdepth)
return TRUE;
return FALSE;
}
/* HUFFMAN ALGORITHM: develops the single element trees
into a single binary tree by forming subtrees rooted in
interior nodes having weights equal to the sum of weights of all
their descendents and having depth counts indicating the
depth of their longest paths.
When all trees have been formed into a single tree satisfying
the heap property (on weight, with depth as a tie breaker)
then the binary code assigned to a leaf (value to be encoded)
is then the series of left (0) and right (1)
paths leading from the root to the leaf.
Note that trees are removed from the heaped list by
moving the last element over the top element and
reheaping the shorter list.
*/
static INT bld_tree(list,len)
INT list[];
INT len;
{
register INT freenode; /* next free node in tree */
register struct nd *frnp; /* free node pointer */
INT lch, rch; /* temps for left, right children */
INT i;
INT maxchar();
/* Initialize index to next available (non-leaf) node.
Lower numbered nodes correspond to leaves (data values).
*/
freenode = NUMVALS;
while(len>1)
{ /* Take from list two btrees with least weight
and build an interior node pointing to them.
This forms a new tree.
*/
lch = list[0]; /* This one will be left child */
/* delete top (least) tree from the list of trees */
list[0] = list[--len];
adjust(list,0,len-1);
/* Take new top (least) tree. Reuse list slot later */
rch = list[0]; /* This one will be right child */
/* Form new tree from the two least trees using
a free node as root. Put the new tree in the list.
*/
frnp = &node[freenode]; /* address of next free node */
list[0] = freenode++; /* put at top for now */
frnp->lchild = lch;
frnp->rchild = rch;
frnp->weight = node[lch].weight + node[rch].weight;
frnp->tdepth = 1 + maxchar(node[lch].tdepth, node[rch].tdepth);
/* reheap list to get least tree at top */
adjust(list,0,len-1);
}
dctreehd = list[0]; /* head of final tree */
}
static INT maxchar(a,b)
{
return a>b ? a : b;
}
static INT init_enc()
{
register INT i;
/* Initialize encoding table */
for(i=0; i<NUMVALS; ++i)
codelen[i] = 0;
}
/* Recursive routine to walk the indicated subtree and level
and maintain the current path code in bstree. When a leaf
is found the entire code string and length are put into
the encoding table entry for the leaf's data value .
Returns ERROR if codes are too long.
*/
static INT buildenc(level,root)
INT level; /* level of tree being examined, from zero */
INT root; /* root of subtree is also data value if leaf */
{
register INT l, r;
l = node[root].lchild;
r = node[root].rchild;
if(l==NOCHILD && r==NOCHILD)
{ /* Leaf. Previous path determines bit string
code of length level (bits 0 to level - 1).
Ensures unused code bits are zero.
*/
codelen[root] = level;
code[root] = tcode & (((unsigned INT)~0) >> (16-level));
return (level>16) ? ERROR : NULL;
}
else
{ if(l!=NOCHILD)
{ /* Clear path bit and continue deeper */
tcode &= ~(1 << level);
if(buildenc(level+1,l)==ERROR)
return ERROR; /* pass back bad statuses */
}
if(r!=NOCHILD)
{ /* Set path bit and continue deeper */
tcode |= 1 << level;
if(buildenc(level+1,r)==ERROR)
return ERROR; /* pass back bad statuses */
}
}
return NULL; /* it worked if we reach here */
}
static INT put_int(n,f) /* output an integer */
INT n; /* integer to output */
FILE *f; /* file to put it to */
{
putc_pak(n&0xff,f); /* first the low byte */
putc_pak(n>>8,f); /* then the high byte */
}
/* Write out the header of the compressed file */
static long wrt_head(ob)
FILE *ob;
{
register INT l,r;
INT i, k;
INT numnodes; /* # of nodes in simplified tree */
/* Write out a simplified decoding tree. Only the interior
nodes are written. When a child is a leaf index
(representing a data value) it is recoded as
-(index + 1) to distinguish it from interior indexes
which are recoded as positive indexes in the new tree.
Note that this tree will be empty for an empty file.
*/
numnodes = dctreehd<NUMVALS ? 0 : dctreehd-(NUMVALS-1);
put_int(numnodes,ob);
for(k=0, i=dctreehd; k<numnodes; ++k, --i)
{ l = node[i].lchild;
r = node[i].rchild;
l = l<NUMVALS ? -(l+1) : dctreehd-l;
r = r<NUMVALS ? -(r+1) : dctreehd-r;
put_int(l,ob);
put_int(r,ob);
}
return sizeof(INT) + numnodes*2*sizeof(INT);
}
/* Get an encoded byte or EOF. Reads from specified stream AS NEEDED.
There are two unsynchronized bit-byte relationships here.
The input stream bytes are converted to bit strings of
various lengths via the static variables named c...
These bit strings are concatenated without padding to
become the stream of encoded result bytes, which this
function returns one at a time. The EOF (end of file) is
converted to SPEOF for convenience and encoded like any
other input value. True EOF is returned after that.
*/
static INT gethuff(ib) /* Returns bytes except for EOF */
FILE *ib;
{
INT rbyte; /* Result byte value */
INT need, take; /* numbers of bits */
rbyte = 0;
need = 8; /* build one byte per call */
/* Loop to build a byte of encoded data.
Initialization forces read the first time.
*/
loop:
if(cbitsrem>=need) /* if current code is big enough */
{ if(need==0)
return rbyte;
rbyte |= ccode << (8-need); /* take what we need */
ccode >>= need; /* and leave the rest */
cbitsrem -= need;
return rbyte & 0xff;
}
/* We need more than current code */
if(cbitsrem>0)
{ rbyte |= ccode << (8-need); /* take what there is */
need -= cbitsrem;
}
/* No more bits in current code string */
if(curin==SPEOF)
{ /* The end of file token has been encoded. If
result byte has data return it and do EOF next time.
*/
cbitsrem = 0;
return (need==8) ? EOF : rbyte + 0;
}
/* Get an input byte */
if((curin=getc_ncr(ib)) == EOF)
curin = SPEOF; /* convenient for encoding */
ccode = code[curin]; /* get the new byte's code */
cbitsrem = codelen[curin];
goto loop;
}
/* This routine is used to perform the actual squeeze operation. It can
only be called after the file has been scanned. It returns the true
length of the squeezed entry.
*/
long file_sq(f,t) /* squeeze a file into an archive */
FILE *f; /* file to squeeze */
FILE *t; /* archive to receive file */
{
INT c; /* one byte of squeezed data */
long size; /* size after squeezing */
size = wrt_head(t); /* write out the decode tree */
while((c=gethuff(f))!=EOF)
{ putc_pak(c,t);
size++;
}
return size; /* report true size */
}
!Funky!Stuff!
echo x - arcsvc.c
cat > arcsvc.c << '!Funky!Stuff!'
static char *RCSid = "$Header: arcsvc.c,v 1.2 86/07/15 07:54:19 turner Exp $";
/*
* $Log: arcsvc.c,v $
* Hack-attack 1.3 86/12/20 01:23:45 wilhite@usceast.uucp
* Bludgeoned into submission for VAX 11/780 BSD4.2
* (ugly code, but fewer core dumps)
*
* Revision 1.2 86/07/15 07:54:19 turner
*
*
* Revision 1.1 86/06/26 15:00:57 turner
* initial version
*
*
*/
/* ARC - Archive utility - ARCSVC
$define(tag,$$segment(@1,$$index(@1,=)+1))#
$define(version,Version $tag(
TED_VERSION DB =2.15), created on $tag(
TED_DATE DB =12/17/85) at $tag(
TED_TIME DB =15:17:16))#
$undefine(tag)#
$version
(C) COPYRIGHT 1985 by System Enhancement Associates; ALL RIGHTS RESERVED
By: Thom Henderson
Description:
This file contains service routines needed to maintain an archive.
Language:
Computer Innovations Optimizing C86
*/
#include <stdio.h>
#include "arc.h"
INT openarc(chg) /* open archive */
INT chg; /* true to open for changes */
{
FILE *fopen(); /* file opener */
if(!(arc=fopen(arcname,"r")))
{ if(chg)
printf("Creating new archive: %s\n",arcname);
else abort("Cannot read archive: %s",arcname);
}
if(chg) /* if opening for changes */
if(!(new=fopen(newname,"w")))
abort("Cannot create archive copy: %s",newname);
}
INT closearc(chg) /* close an archive */
INT chg; /* true if archive was changed */
{
if(arc) /* if we had an initial archive */
fclose(arc); /* then close it */
if(chg) /* if things have changed */
{ setstamp(new,arcdate,arctime);/* archive matches newest file */
fclose(new); /* close the new copy */
if(arc) /* if we had an original archive */
{ if(keepbak) /* if a backup is wanted */
{ unlink(bakname); /* erase any old copies */
if(rename(arcname,bakname))
abort("Cannot rename %s to %s",arcname,bakname);
printf("Keeping backup archive: %s\n",bakname);
}
else if(unlink(arcname))
abort("Cannot delete old archive: %s",arcname);
}
if(rename(newname,arcname))
abort("Cannot rename %s to %s",newname,arcname);
}
}
/* CRC computation logic
The logic for this method of calculating the CRC 16 bit polynomial
is taken from an article by David Schwaderer in the April 1985
issue of PC Tech Journal.
*/
static INT crctab[] = /* CRC lookup table */
{ 0x0000, 0xC0C1, 0xC181, 0x0140, 0xC301, 0x03C0, 0x0280, 0xC241,
0xC601, 0x06C0, 0x0780, 0xC741, 0x0500, 0xC5C1, 0xC481, 0x0440,
0xCC01, 0x0CC0, 0x0D80, 0xCD41, 0x0F00, 0xCFC1, 0xCE81, 0x0E40,
0x0A00, 0xCAC1, 0xCB81, 0x0B40, 0xC901, 0x09C0, 0x0880, 0xC841,
0xD801, 0x18C0, 0x1980, 0xD941, 0x1B00, 0xDBC1, 0xDA81, 0x1A40,
0x1E00, 0xDEC1, 0xDF81, 0x1F40, 0xDD01, 0x1DC0, 0x1C80, 0xDC41,
0x1400, 0xD4C1, 0xD581, 0x1540, 0xD701, 0x17C0, 0x1680, 0xD641,
0xD201, 0x12C0, 0x1380, 0xD341, 0x1100, 0xD1C1, 0xD081, 0x1040,
0xF001, 0x30C0, 0x3180, 0xF141, 0x3300, 0xF3C1, 0xF281, 0x3240,
0x3600, 0xF6C1, 0xF781, 0x3740, 0xF501, 0x35C0, 0x3480, 0xF441,
0x3C00, 0xFCC1, 0xFD81, 0x3D40, 0xFF01, 0x3FC0, 0x3E80, 0xFE41,
0xFA01, 0x3AC0, 0x3B80, 0xFB41, 0x3900, 0xF9C1, 0xF881, 0x3840,
0x2800, 0xE8C1, 0xE981, 0x2940, 0xEB01, 0x2BC0, 0x2A80, 0xEA41,
0xEE01, 0x2EC0, 0x2F80, 0xEF41, 0x2D00, 0xEDC1, 0xEC81, 0x2C40,
0xE401, 0x24C0, 0x2580, 0xE541, 0x2700, 0xE7C1, 0xE681, 0x2640,
0x2200, 0xE2C1, 0xE381, 0x2340, 0xE101, 0x21C0, 0x2080, 0xE041,
0xA001, 0x60C0, 0x6180, 0xA141, 0x6300, 0xA3C1, 0xA281, 0x6240,
0x6600, 0xA6C1, 0xA781, 0x6740, 0xA501, 0x65C0, 0x6480, 0xA441,
0x6C00, 0xACC1, 0xAD81, 0x6D40, 0xAF01, 0x6FC0, 0x6E80, 0xAE41,
0xAA01, 0x6AC0, 0x6B80, 0xAB41, 0x6900, 0xA9C1, 0xA881, 0x6840,
0x7800, 0xB8C1, 0xB981, 0x7940, 0xBB01, 0x7BC0, 0x7A80, 0xBA41,
0xBE01, 0x7EC0, 0x7F80, 0xBF41, 0x7D00, 0xBDC1, 0xBC81, 0x7C40,
0xB401, 0x74C0, 0x7580, 0xB541, 0x7700, 0xB7C1, 0xB681, 0x7640,
0x7200, 0xB2C1, 0xB381, 0x7340, 0xB101, 0x71C0, 0x7080, 0xB041,
0x5000, 0x90C1, 0x9181, 0x5140, 0x9301, 0x53C0, 0x5280, 0x9241,
0x9601, 0x56C0, 0x5780, 0x9741, 0x5500, 0x95C1, 0x9481, 0x5440,
0x9C01, 0x5CC0, 0x5D80, 0x9D41, 0x5F00, 0x9FC1, 0x9E81, 0x5E40,
0x5A00, 0x9AC1, 0x9B81, 0x5B40, 0x9901, 0x59C0, 0x5880, 0x9841,
0x8801, 0x48C0, 0x4980, 0x8941, 0x4B00, 0x8BC1, 0x8A81, 0x4A40,
0x4E00, 0x8EC1, 0x8F81, 0x4F40, 0x8D01, 0x4DC0, 0x4C80, 0x8C41,
0x4400, 0x84C1, 0x8581, 0x4540, 0x8701, 0x47C0, 0x4680, 0x8641,
0x8201, 0x42C0, 0x4380, 0x8341, 0x4100, 0x81C1, 0x8081, 0x4040
};
INT addcrc(crc,c) /* update a CRC check */
INT crc; /* running CRC value */
unsigned char c; /* character to add */
{
return (((crc>>8)&0x00ff) ^ crctab[(crc^c)&0x00ff]) & 0x0000ffff;
}
!Funky!Stuff!
echo x - arctst.c
cat > arctst.c << '!Funky!Stuff!'
static char *RCSid = "$Header: arctst.c,v 1.1 86/06/26 15:01:02 turner Exp $";
/*
* $Log: arctst.c,v $
* Hack-attack 1.3 86/12/20 01:23:45 wilhite@usceast.uucp
* Bludgeoned into submission for VAX 11/780 BSD4.2
* (ugly code, but fewer core dumps)
*
* Revision 1.1 86/06/26 15:01:02 turner
* initial version
*
*
*/
/* ARC - Archive utility - ARCTST
$define(tag,$$segment(@1,$$index(@1,=)+1))#
$define(version,Version $tag(
TED_VERSION DB =2.12), created on $tag(
TED_DATE DB =02/03/86) at $tag(
TED_TIME DB =23:00:40))#
$undefine(tag)#
$version
(C) COPYRIGHT 1985 by System Enhancement Associates; ALL RIGHTS RESERVED
By: Thom Henderson
Description:
This file contains the routines used to test archive integrity.
Language:
Computer Innovations Optimizing C86
*/
#include <stdio.h>
#include "arc.h"
INT tstarc() /* test integrity of an archive */
{
struct heads hdr; /* file header */
long arcsize, ftell(); /* archive size */
openarc(0); /* open archive for reading */
fseek(arc,0L,2); /* move to end of archive */
arcsize = ftell(arc); /* see how big it is */
fseek(arc,0L,0); /* return to top of archive */
printf ("Testing archive...\n");
while(readhdr(&hdr,arc))
{ if(ftell(arc)+hdr.size>arcsize)
{ printf("Archive truncated in file %s\n",hdr.name);
nerrs++;
break;
}
else
{ printf(" File: %-12s ",hdr.name);
fflush(stdout);
if(unpack(arc,NULL,&hdr))
nerrs++;
else printf("okay\n");
}
}
if(nerrs<1)
printf("No errors detected\n");
else if(nerrs==1)
printf("One error detected\n");
else printf("%d errors detected\n",nerrs);
}
!Funky!Stuff!
echo x - arcunp.c
cat > arcunp.c << '!Funky!Stuff!'
static char *RCSid = "$Header: arcunp.c,v 1.2 86/07/15 07:54:25 turner Exp $";
/*
* $Log: arcunp.c,v $
* Hack-attack 1.3 86/12/20 01:23:45 wilhite@usceast.uucp
* Bludgeoned into submission for VAX 11/780 BSD4.2
* (ugly code, but fewer core dumps)
*
* Revision 1.2 86/07/15 07:54:25 turner
*
*
* Revision 1.1 86/06/26 15:01:04 turner
* initial version
*
*
*/
/* ARC - Archive utility - ARCUNP
$define(tag,$$segment(@1,$$index(@1,=)+1))#
$define(version,Version $tag(
TED_VERSION DB =3.16), created on $tag(
TED_DATE DB =02/03/86) at $tag(
TED_TIME DB =23:01:16))#
$undefine(tag)#
$version
(C) COPYRIGHT 1985 by System Enhancement Associates; ALL RIGHTS RESERVED
By: Thom Henderson
Description:
This file contains the routines used to expand a file
when taking it out of an archive.
Language:
Computer Innovations Optimizing C86
*/
#include <stdio.h>
#include "arc.h"
/* stuff for repeat unpacking */
#define DLE 0x90 /* repeat byte flag */
static INT state; /* repeat unpacking state */
/* repeat unpacking states */
#define NOHIST 0 /* no relevant history */
#define INREP 1 /* sending a repeated value */
static INT crcval; /* CRC check value */
static long size; /* bytes to read */
INT unpack(f,t,hdr) /* unpack an archive entry */
FILE *f, *t; /* source, destination */
struct heads *hdr; /* pointer to file header data */
{
INT c; /* one char of stream */
INT putc_unp();
INT putc_ncr();
INT getc_unp();
/* setups common to all methods */
crcval = 0; /* reset CRC check value */
size = hdr->size; /* set input byte counter */
state = NOHIST; /* initial repeat unpacking state */
setcode(); /* set up for decoding */
/* use whatever method is appropriate */
switch(hdrver) /* choose proper unpack method */
{
case 1: /* standard packing */
case 2:
while((c=getc_unp(f))!=EOF)
putc_unp(c,t);
break;
case 3: /* non-repeat packing */
while((c=getc_unp(f))!=EOF)
putc_ncr(c,t);
break;
case 4: /* Huffman squeezing */
init_usq(f);
while((c=getc_usq(f))!=EOF)
putc_ncr(c,t);
break;
case 5: /* Lempel-Zev compression */
init_ucr(0);
while((c=getc_ucr(f))!=EOF)
putc_unp(c,t);
break;
case 6: /* Lempel-Zev plus non-repeat */
init_ucr(0);
while((c=getc_ucr(f))!=EOF)
putc_ncr(c,t);
break;
case 7: /* L-Z plus ncr with new hash */
init_ucr(1);
while((c=getc_ucr(f))!=EOF)
putc_ncr(c,t);
break;
case 8: /* dynamic Lempel-Zev */
decomp(f,t);
break;
default: /* unknown method */
if(warn)
{ printf("I don't know how to unpack file %s\n",hdr->name);
printf("I think you need a newer version of ARC\n");
nerrs++;
}
fseek(f,hdr->size,1); /* skip over bad file */
return 1; /* note defective file */
}
/* cleanups common to all methods */
if((crcval&0xffff)!=(hdr->crc&0x0000ffff))
{ if(warn)
{ printf("WARNING: File %s fails CRC check\n",hdr->name);
nerrs++;
}
return 1; /* note defective file */
}
return 0; /* file is okay */
}
/* This routine is used to put bytes in the output file. It also
performs various housekeeping functions, such as maintaining the
CRC check value.
*/
static INT putc_unp(c,t) /* output an unpacked byte */
char c; /* byte to output */
FILE *t; /* file to output to */
{
crcval = addcrc(crcval,c); /* update the CRC check value */
putc_tst(c,t);
}
/* This routine is used to decode non-repeat compression. Bytes are
passed one at a time in coded format, and are written out uncoded.
The data is stored normally, except that runs of more than two
characters are represented as:
<char> <DLE> <count>
With a special case that a count of zero indicates a DLE as data,
not as a repeat marker.
*/
INT putc_ncr(c,t) /* put NCR coded bytes */
unsigned char c; /* next byte of stream */
FILE *t; /* file to receive data */
{
static INT lastc; /* last character seen */
switch(state) /* action depends on our state */
{
case NOHIST: /* no previous history */
if(c==DLE) /* if starting a series */
state = INREP; /* then remember it next time */
else putc_unp(lastc=c,t); /* else nothing unusual */
return;
case INREP: /* in a repeat */
if(c) /* if count is nonzero */
while(--c) /* then repeatedly ... */
putc_unp(lastc,t); /* ... output the byte */
else putc_unp(DLE,t); /* else output DLE as data */
state = NOHIST; /* back to no history */
return;
default:
abort("Bad NCR unpacking state (%d)",state);
}
}
/* This routine provides low-level byte input from an archive. This
routine MUST be used, as end-of-file is simulated at the end of
the archive entry.
*/
INT getc_unp(f) /* get a byte from an archive */
FILE *f; /* archive file to read */
{
if(!size) /* if no data left */
return EOF; /* then pretend end of file */
size--; /* deduct from input counter */
return code(fgetc(f)); /* and return next decoded byte */
}
!Funky!Stuff!
echo x - arcusq.c
cat > arcusq.c << '!Funky!Stuff!'
static char *RCSid = "$Header: arcusq.c,v 1.2 86/07/15 07:54:30 turner Exp $";
/*
* $Log: arcusq.c,v $
* Hack-attack 1.3 86/12/20 01:23:45 wilhite@usceast.uucp
* Bludgeoned into submission for VAX 11/780 BSD4.2
* (ugly code, but fewer core dumps)
*
* Revision 1.2 86/07/15 07:54:30 turner
*
*
* Revision 1.1 86/06/26 15:01:07 turner
* initial version
*
*
*/
/* ARC - Archive utility - ARCUSQ
$define(tag,$$segment(@1,$$index(@1,=)+1))#
$define(version,Version $tag(
TED_VERSION DB =3.13), created on $tag(
TED_DATE DB =01/30/86) at $tag(
TED_TIME DB =20:11:42))#
$undefine(tag)#
$version
(C) COPYRIGHT 1985 by System Enhancement Associates; ALL RIGHTS RESERVED
By: Thom Henderson
Description:
This file contains the routines used to expand a file
which was packed using Huffman squeezing.
Most of this code is taken from an USQ program by Richard
Greenlaw, which was adapted to CI-C86 by Robert J. Beilstein.
Language:
Computer Innovations Optimizing C86
*/
#include <stdio.h>
#include "arc.h"
/* stuff for Huffman unsqueezing */
#define ERROR (-1)
#define SPEOF 256 /* special endfile token */
#define NUMVALS 257 /* 256 data values plus SPEOF */
EXTERN struct nd /* decoding tree */
{ INT child[2]; /* left, right */
} node[NUMVALS]; /* use large buffer */
static INT bpos; /* last bit position read */
static INT curin; /* last byte value read */
static INT numnodes; /* number of nodes in decode tree */
static INT get_int(f) /* get an integer */
FILE *f; /* file to get it from */
{
INT i;
i = getc_unp(f);
return (short)(i | (getc_unp(f)<<8));
}
INT init_usq(f) /* initialize Huffman unsqueezing */
FILE *f; /* file containing squeezed data */
{
INT i; /* node index */
bpos = 99; /* force initial read */
numnodes = get_int(f);
if(numnodes<0 || numnodes>=NUMVALS)
abort("File has an invalid decode tree");
/* initialize for possible empty tree (SPEOF only) */
node[0].child[0] = -(SPEOF + 1);
node[0].child[1] = -(SPEOF + 1);
for(i=0; i<numnodes; ++i) /* get decoding tree from file */
{ node[i].child[0] = get_int(f);
node[i].child[1] = get_int(f);
}
}
INT getc_usq(f) /* get byte from squeezed file */
FILE *f; /* file containing squeezed data */
{
INT i; /* tree index */
/* follow bit stream in tree to a leaf */
for(i=0; i>=0; ) /* work down(up?) from root */
{ if(++bpos>7)
{ if((curin=getc_unp(f)) == ERROR)
return(ERROR);
bpos = 0;
/* move a level deeper in tree */
i = node[i].child[1&curin];
}
else i = node[i].child[1 & (curin >>= 1)];
}
/* decode fake node index to original data value */
i = -(i + 1);
/* decode special endfile token to normal EOF */
i = (i==SPEOF) ? EOF : i;
return i;
}
!Funky!Stuff!
echo x - arcvax.c
cat > arcvax.c << '!Funky!Stuff!'
#include <stdio.h>
#include <ctype.h>
#include "arc.h"
/*
* Misc routines to emulate IBM MSDOS functions under BSD
*
* Hack-attack 1.3 86/12/20 01:23:45 wilhite@usceast.uucp
* Bludgeoned into submission for VAX 11/780 BSD4.2
* (ugly code, but fewer core dumps)
*/
INT upper(string)
char *string;
{
char *p;
for(p = string; *p != NULL; p++)
if(islower(*p))
*p = toupper(*p);
}
char *setmem(dest,size,c)
char *dest,c;
INT size;
{
INT i;
for(i = 0; i < size; dest[i] = c, i++);
return(&dest[0]);
}
char *gcdir(dirname)
char *dirname;
{
if(dirname == NULL || strlen(dirname) == 0)
dirname = (char *)malloc(1024);
getwd(dirname);
}
INT abort(s,arg1,arg2,arg3)
char *s;
{
fprintf(stderr,"ARC: ");
fprintf(stderr,s,arg1,arg2,arg3);
fprintf(stderr,"\n");
#if BSD
perror("BSD");
#endif
exit(1);
}
!Funky!Stuff!
exit 0
--
---------------------
Robert Wilhite
akgua!usceast!wilhite