Category : OS/2 Files
Archive   : ZOO21E.ZIP
Filename : PORTABLE.C

 
Output of file : PORTABLE.C contained in archive : ZOO21E.ZIP
#ifndef LINT
/* @(#) portable.c 2.24 88/08/24 01:22:06 */
static char sccsid[]="@(#) portable.c 2.24 88/08/24 01:22:06";
#endif /* LINT */

#include "options.h"
/*
Copyright (C) 1986, 1987 Rahul Dhesi -- All rights reserved
(C) Copyright 1988 Rahul Dhesi -- All rights reserved
*/
/**********************
portable.c contains functions needed to make Zoo portable to various
implementations of C.

Note: Provided a 2's complement machine is used, all functions in
this file are themselves machine-independent and need not be changed
when implementing Zoo on a different machine. Some code will choke
on 1's complement machines--I think.

For machine-dependent declarations see files "machine.h" and "options.h".

For machine-dependent functions see file "machine.c"
*/

#include "zoo.h"
#include "zooio.h"

#include "various.h"
#include "zoofns.h"

#include "machine.h"
#include "debug.h"
#include "assert.h"

#ifdef NEEDCTYP
#include /* for tolower() */
#endif


#include "portable.h"

#ifdef TRACE_IO
extern int verbose;
#endif

/* Functions defined for use within this file only. */
long to_long PARMS((BYTE[]));
int to_int PARMS((BYTE[]));
void b_to_zooh PARMS((struct zoo_header *, BYTE[]));
void b_to_dir PARMS((struct direntry *, BYTE[]));
int dir_to_b PARMS((BYTE[], struct direntry *));
void zooh_to_b PARMS((BYTE[], struct zoo_header *));
void splitlong PARMS((BYTE[], long));
void splitint PARMS((BYTE[], int));

#ifdef TRACE_IO
void show_h PARMS ((struct zoo_header *));
void show_dir PARMS ((struct direntry *));
#endif /* TRACE_IO */

extern unsigned int crccode;

/************************************************************************/
/* I/O functions */
/************************************************************************/

/* some functions get defined only if they aren't already macros */

#ifndef zooread
int zooread (file, buffer, count)
ZOOFILE file; char *buffer; int count;
{ return (fread (buffer, 1, count, file)); }
#endif /* zooread */

#ifndef FIZ
#ifndef zoowrite
int zoowrite (file, buffer, count)
ZOOFILE file; char *buffer; int count;
{
if (file == NULLFILE)
return (count);
else
return (fwrite (buffer, 1, count, file));
}
#endif /* zoowrite */

ZOOFILE zoocreate (fname)
char *fname;
{ return ((ZOOFILE) fopen (fname, Z_NEW)); }

#endif /* FIZ */

#ifndef zooseek
long zooseek (file, offset, whence)
ZOOFILE file; long offset; int whence;
{ return (fseek (file, offset, whence)); }
#endif /* zooseek */

ZOOFILE zooopen (fname, option)
char *fname; char *option;
{ return ((ZOOFILE) fopen (fname, option)); }

#ifndef zootell
long zootell (file)
ZOOFILE file;
{ return ftell (file); }
#endif /* zootell */

int zooclose (file)
ZOOFILE file;
{ return fclose (file); }

/**********************
low_ch() is a macro that returns a lowercased char; it may be
used with any char, whether or not it is uppercase. It will
be used below by one or two functions.
*/

#define low_ch(c) (isupper(c) ? tolower(c) : c)

/************************************************************************/
/*** Following are functions that make up for various implementations ***/
/*** of C not having certain library routines. ***/
/************************************************************************/

#ifndef FIZ
/**********************
str_lwr() converts a string to lowercase and returns a pointer to the string
*/
char *str_lwr (str)
char *str;
{
register char *s;
s = str;
while (*s != '\0') {
*s = toascii(*s);
*s = low_ch(*s);
s++;
}
return (str);
}

/**********************
str_icmp() compares strings just like strcmp() but it does it without regard to
case.
*/
int str_icmp (s1, s2)
register char *s1, *s2;
{
for ( ; low_ch(*s1) == low_ch(*s2); s1++, s2++)
if (*s1 == '\0')
return(0);
return(low_ch(*s1) - low_ch(*s2));
}

#ifdef NEED_MEMSET
/**********************
memset() it sets the first "count" bytes of "dest" to the character
"c" and returns a pointer to "dest".
*/
VOIDPTR memset (dest, c, count)
register VOIDPTR dest;
int c;
unsigned count;
{
register unsigned i;
for (i = 0; i < count; i++) {
*((char *) (dest + i)) = c;
}
return dest;
}
#endif /* NEED_MEMSET */

#ifdef NEED_MEMCPY
/**********************
memcpy() copies "count" bytes from "src" to "dest" and returns
a pointer to "dest". Not necessarily safe for overlapping moves. */

VOIDPTR memcpy(dest, src, count)
register VOIDPTR dest;
register VOIDPTR src;
unsigned count;
{
VOIDPTR savedest = dest;
while (count > 0) {
*((char *) dest++) = *((char *) src++);
count--;
}
}
#endif /* NEED_MEMCPY */

#ifndef FPUTCHAR
/**********************
fputchar() writes a character to stdout. It is identical to putchar
but is a function, not a macro.
*/
int fputchar (c)
int c;
{
return (fputc(c, stdout));
}
#endif /* FPUTCHAR */
#endif /* FIZ */

/***********************************************************************/
/*** Following are declarations and functions that are written in a ***/
/*** machine-independent way but they implement machine-dependent ***/
/*** activities ***/
/***********************************************************************/

#ifndef DIRECT_CONVERT
/**********************
to_long() converts four consecutive bytes, in order of increasing
significance, to a long integer. It is used to make Zoo independent of the
byte order of the system.
*/
long to_long(data)
BYTE data[];
{
return (long) ((unsigned long) data[0] | ((unsigned long) data[1] << 8) |
((unsigned long) data[2] << 16) | ((unsigned long) data[3] << 24));
}

#ifndef FIZ
/********************
splitlong() converts a long integer to four consecutive BYTEs in order
of increasing significance.
*/
void splitlong(bytes, bigword)
BYTE bytes[];
long bigword;
{
int i;
for (i = 0; i < 4; i++) {
bytes[i] = bigword & 0xff;
bigword = (unsigned long) bigword >> 8;
}
}
#endif /* FIZ */

/*******************
splitint() converts an integer to two consecutive BYTEs in order
of increasing significance.
*/
void splitint(bytes, word)
BYTE bytes[];
int word;
{
bytes[0] = word & 0xff;
word = (unsigned int) word >> 8;
bytes[1] = word & 0xff;
}

/**********************
to_int() converts two consecutive bytes, in order of increasing
significance, to an integer, in a machine-independent manner
*/
int to_int(data)
BYTE data[];
{
return (int) ((unsigned int) data[0] | ((unsigned int) data[1] << 8));
}

#else /* else of ifndef DIRECT_CONVERT */

long to_long(data)
BYTE data[];
{
return ( * (long *) data );
}

#ifndef FIZ
/********************
splitlong() converts a long integer to four consecutive BYTEs in order
of increasing significance.
*/
void splitlong(bytes, bigword)
BYTE bytes[];
long bigword;
{
* (long *) bytes = bigword;
}
#endif /* FIZ */

/*******************
splitint() converts an integer to two consecutive BYTEs in order
of increasing significance.
*/
void splitint(bytes, word)
BYTE bytes[];
int word;
{
* (int *) bytes = word;
}

/**********************
to_int() converts two consecutive bytes, in order of increasing
significance, to an integer.
*/
int to_int(data)
BYTE data[];
{
return (*(int *) data);
}

#endif /* ifndef DIRECT_CONVERT .. else ... */

#ifndef FIZ
/**********************
Function frd_zooh() reads the header of a Zoo archive in a machine-
independent manner, from a ZOOFILE.
*/
int frd_zooh(zoo_header, zoo_file)
struct zoo_header *zoo_header;
ZOOFILE zoo_file;
{
int status;
BYTE bytes[SIZ_ZOOH]; /* canonical header representation */
#ifdef TRACE_IO
if (verbose) {
printf("At file position [%8lx] ", ftell(zoo_file));
}
#endif
status = zooread (zoo_file, (char *) bytes, SIZ_ZOOH);
b_to_zooh (zoo_header, bytes); /* convert array to structure */
#ifdef TRACE_IO
if (verbose) {
printf("frd_zooh: reading\n");
show_h(zoo_header);
}
#endif
if (status < MINZOOHSIZ)
return (-1);
else
return (0);
}
#endif /* FIZ */

/**********************
Function frd_dir() reads a directory entry in a machine-independent manner,
from a ZOOFILE.
*/
int frd_dir(direntry, zoo_file)
struct direntry *direntry;
ZOOFILE zoo_file;
{
int status;
BYTE bytes[MAXDIRSIZE]; /* big enough to hold variable part too */

/* To simplify things, we read the maximum possible size of the
directory entry including the variable size and discard what is not
needed */
#ifdef TRACE_IO
if (verbose) {
printf("At file position [%8lx] ", ftell(zoo_file));
}
#endif
status = zooread (zoo_file, (char *) bytes, MAXDIRSIZE);
if (status < SIZ_DIR)
return (-1);
b_to_dir (direntry, bytes);
#ifdef TRACE_IO
if (verbose) {
printf("frd_dir: reading\n");
show_dir(direntry);
}
#endif
return (0);
}

#ifndef FIZ
/***********************
Function fwr_dir() writes a directory entry in a machine-independent manner
to a ZOOFILE. Return value is -1 on error, else 0.
*/
int fwr_dir(direntry, zoo_file)
struct direntry *direntry;
ZOOFILE zoo_file;
{
int size;
BYTE bytes[MAXDIRSIZE];
assert (direntry->type <= 2);
size = dir_to_b (bytes, direntry);
#ifdef TRACE_IO
if (verbose) {
printf("At file position [%8lx] ", ftell(zoo_file));
printf("fwr_dir: writing\n");
show_dir(direntry);
}
#endif

if (zoowrite (zoo_file, (char *) bytes, size) != size)
return (-1);
else
return (0);
}

/***********************
Function fwr_zooh() writes an archive header in a machine-independent manner
to a ZOOFILE. Return value is -1 if error else 0.
*/
int fwr_zooh(zoo_header, zoo_file)
struct zoo_header *zoo_header;
ZOOFILE zoo_file;
{
BYTE bytes[SIZ_ZOOH]; /* was SIZ_DIR -- probably a typo */
int hsize; /* how much to write -- depends on header type */
hsize = MINZOOHSIZ; /* in case it's an old type 0 header */
if (zoo_header->type > 0) /* but if it's a newer header... */
hsize = SIZ_ZOOH; /* ...size of new type 1 header */
zooh_to_b (bytes, zoo_header);
if (zoowrite (zoo_file, (char *) bytes, hsize) != hsize)
return (-1);
else
return (0);
}

/***********************
b_to_zooh() converts an array of BYTE to a zoo_header structure.
*/
void b_to_zooh (zoo_header, bytes)
struct zoo_header *zoo_header;
BYTE bytes[];
{
int i;
for (i = 0; i < SIZ_TEXT; i++) /* copy text */
zoo_header->text[i] = bytes[TEXT_I + i];
zoo_header->zoo_tag = to_long(&bytes[ZTAG_I]); /* copy zoo_tag */
zoo_header->zoo_start = to_long(&bytes[ZST_I]); /* copy zoo_start */
zoo_header->zoo_minus = to_long(&bytes[ZSTM_I]);
zoo_header->major_ver = bytes[MAJV_I]; /* copy versions */
zoo_header->minor_ver = bytes[MINV_I];
/* default is no archive comment and a header type of 0 */
zoo_header->type = 0;
zoo_header->acmt_pos = 0L;
zoo_header->acmt_len = 0;
zoo_header->vdata = 0;
if (zoo_header->zoo_start != FIXED_OFFSET) { /* if newer header */
zoo_header->type = bytes[HTYPE_I];
zoo_header->acmt_pos = to_long(&bytes[ACMTPOS_I]);
zoo_header->acmt_len = to_int(&bytes[ACMTLEN_I]);
zoo_header->vdata = bytes[HVDATA_I];
}
}

/***********************
zooh_to_b() converts a zoo_header structure to an array of BYTE.
*/
void zooh_to_b (bytes, zoo_header)
struct zoo_header *zoo_header;
BYTE bytes[];
{
int i;
for (i = 0; i < SIZ_TEXT; i++) /* copy text */
bytes[TEXT_I + i] = zoo_header->text[i];
splitlong (&bytes[ZTAG_I], zoo_header->zoo_tag);
splitlong (&bytes[ZST_I], zoo_header->zoo_start);
splitlong (&bytes[ZSTM_I], zoo_header->zoo_minus);
bytes[MAJV_I] = zoo_header->major_ver; /* copy versions */
bytes[MINV_I] = zoo_header->minor_ver;
bytes[HTYPE_I] = zoo_header->type; /* header type */
if (zoo_header->type > 0) {
splitlong (&bytes[ACMTPOS_I], zoo_header->acmt_pos); /* comment posn */
splitint (&bytes[ACMTLEN_I], zoo_header->acmt_len); /* comment len */
bytes[HVDATA_I] = zoo_header->vdata; /* version data */
}
} /* zooh_to_b() */

/************************
dir_to_b() converts a directory entry structure to an array of BYTE.
*/
int dir_to_b (bytes, direntry)
struct direntry *direntry;
BYTE bytes[];
{
int i;
int cursize;
int fixsize;
int totalsize;
splitlong(&bytes[DTAG_I], direntry->zoo_tag);
bytes[DTYP_I] = direntry->type ;
bytes[PKM_I] = direntry->packing_method ;
splitlong(&bytes[NXT_I], direntry->next);
splitlong(&bytes[OFS_I], direntry->offset);
splitint(&bytes[DAT_I], direntry->date);
splitint(&bytes[TIM_I], direntry->time);
splitint(&bytes[CRC_I], direntry->file_crc);
splitlong(&bytes[ORGS_I], direntry->org_size);
splitlong(&bytes[SIZNOW_I], direntry->size_now);
bytes[DMAJ_I] = direntry->major_ver;
bytes[DMIN_I] = direntry->minor_ver;
bytes[DEL_I] = direntry->deleted;
bytes[STRUC_I] = direntry->struc;
splitlong(&bytes[CMT_I], direntry->comment);
splitint(&bytes[CMTSIZ_I], direntry->cmt_size);
for (i = 0; i < FNM_SIZ; i++)
bytes[FNAME_I + i] = direntry->fname[i];
bytes[TZ_I] = NO_TZ; /* assume unknown */
bytes[NAMLEN_I] = 0;
bytes[DIRLEN_I] = 0;

cursize = SIZ_DIR; /* to count size of directory */
fixsize = SIZ_DIR; /* size of fixed part */
assert (direntry->type <= 2);
if (direntry->type == 2) { /* handle stuff relevant to type 2 */
cursize = SIZ_DIRL;
fixsize = SIZ_DIRL;
bytes[TZ_I] = direntry->tz;
assert(direntry->namlen < 256 && direntry->namlen >= 0);
cursize += 2; /* space for namlen and dirlen */
if (direntry->namlen != 0) {
bytes[NAMLEN_I] = direntry->namlen;
for (i = 0; i < direntry->namlen; i++)
bytes[LFNAME_I+i] = direntry->lfname[i];
cursize += direntry->namlen;
}
assert(direntry->dirlen < 256 && direntry->dirlen >= 0);
if (direntry->dirlen != 0) {
bytes[DIRLEN_I] = direntry->dirlen;
for (i = 0; i < direntry->dirlen; i++)
bytes[cursize+i] = direntry->dirname[i];
cursize += direntry->dirlen;
}
/* Can't store system id if no namlen & dirlen...BUG!...now fixed.
Fortunately, system_id was always 0 so far so it probably
got interpreted as namlen=0 and dirlen=0 (2 bytes) */
splitint(&bytes[cursize], direntry->system_id);
cursize += 2;
bytes[cursize] = direntry->fattr & 0xff; /* byte 0 */
splitint(&bytes[cursize+1], (int) (direntry->fattr >> 8)); /* 1 & 2 */
cursize += 3;
bytes[cursize] = (direntry->vflag & 0xff); /* version flag */
splitint(&bytes[cursize+1], direntry->version_no); /* version number */
cursize += 3;
}

splitint(&bytes[VARDIRLEN_I], direntry->var_dir_len);
assert(cursize ==
((bytes[DIRLEN_I] > 0 || bytes[NAMLEN_I] > 0) ? 2 : 0) +
fixsize + bytes[DIRLEN_I] + bytes[NAMLEN_I]
);

/* total size of dir entry is size of fixed part + size of var. part */
totalsize = fixsize + direntry->var_dir_len;

/* Do CRC assuming CRC field is zero, and stuff CRC into field. */
splitint(&bytes[DCRC_I], 0); /* fill with zeroes */
crccode = 0;
/* avoid mixing pointers to signed and unsigned char */
addbfcrc((char *) bytes, totalsize); /* update CRC */
splitint(&bytes[DCRC_I], crccode);

/* return total length of directory entry */
return (totalsize);

} /* dir_to_b() */
#endif /* FIZ */

/* b_to_dir() converts bytes to directory entry structure. The CRC of the
directory bytes, if any, is checked and a zero or nonzero value is returned
in direntry->dir_crc according as the check is good or bad */

void b_to_dir(direntry, bytes)
struct direntry *direntry;
BYTE bytes[];
{
int i;
int sysid_offs; /* temp variable */
unsigned int savecrc;
direntry->zoo_tag = to_long(&bytes[DTAG_I]);
direntry->type = bytes[DTYP_I];
direntry->packing_method = bytes[PKM_I];
direntry->next = to_long(&bytes[NXT_I]);
direntry->offset = to_long(&bytes[OFS_I]);
direntry->date = to_int(&bytes[DAT_I]);
direntry->time = to_int(&bytes[TIM_I]);
direntry->file_crc = to_int(&bytes[CRC_I]);
direntry->org_size = to_long(&bytes[ORGS_I]);
direntry->size_now = to_long(&bytes[SIZNOW_I]);
direntry->major_ver = bytes[DMAJ_I];
direntry->minor_ver = bytes[DMIN_I];
direntry->deleted = bytes[DEL_I];
direntry->struc = bytes[STRUC_I];
direntry->comment = to_long(&bytes[CMT_I]);
direntry->cmt_size = to_int(&bytes[CMTSIZ_I]);
/* for now, versions not implemented */
direntry->vflag = 0;
direntry->version_no = 0;
for (i = 0; i < FNM_SIZ; i++)
direntry->fname[i] = bytes[FNAME_I + i];

/* start by assuming variable part is zero bytes */
direntry->var_dir_len = direntry->dir_crc = 0;
direntry->namlen = direntry->dirlen = 0;
direntry->lfname[0] = direntry->dirname[0] = '\0';
direntry->tz = NO_TZ; /* assume unknown */
direntry->system_id = SYSID_NIX; /* default system_id if not present */
direntry->fattr = NO_FATTR; /* assume none */

assert (direntry->type <= 2);
if (direntry->type == 2) {
direntry->var_dir_len = to_int(&bytes[VARDIRLEN_I]);
assert(direntry->var_dir_len <= MAXDIRSIZE);
if (direntry->var_dir_len > MAXDIRSIZE)
direntry->var_dir_len = MAXDIRSIZE;
direntry->tz = bytes[TZ_I];
if (direntry->var_dir_len > 0)
direntry->namlen = bytes[NAMLEN_I];
if (direntry->var_dir_len > 1)
direntry->dirlen = bytes[DIRLEN_I];
for (i = 0; i < direntry->namlen; i++)
direntry->lfname[i] = bytes[LFNAME_I + i];
for (i = 0; i < direntry->dirlen; i++)
direntry->dirname[i] = bytes[DIRNAME_I + direntry->namlen + i];
sysid_offs = DIRNAME_I + direntry->namlen + i; /* offset of system id */
if (direntry->var_dir_len > direntry->namlen + direntry->dirlen + 2) {
direntry->system_id = to_int(&bytes[sysid_offs]);
}
if (direntry->var_dir_len > direntry->namlen + direntry->dirlen + 4) {
direntry->fattr = ((unsigned long) bytes[sysid_offs + 2]) |
((unsigned long) bytes[sysid_offs + 3] << 8) |
((unsigned long) bytes[sysid_offs + 4] << 16);
}
if (direntry->var_dir_len > direntry->namlen + direntry->dirlen + 7) {
direntry->vflag = bytes[sysid_offs + 5];
direntry->version_no = to_int(&bytes[sysid_offs + 6]);
}
/* do CRC calculation */
savecrc = (unsigned int) to_int(&bytes[DCRC_I]);
crccode = 0;
splitint(&bytes[DCRC_I], 0);
addbfcrc((char *) bytes, SIZ_DIRL + direntry->var_dir_len);
direntry->dir_crc = crccode - savecrc;
}
}

#ifdef FILTER
#define TWOBYTES 2 /* better than literal 2; figure out why */

/* rdint() reads two bytes from standard input in archive order */
int rdint (val)
unsigned int *val;
{
BYTE bytes[TWOBYTES];
if (zooread (STDIN, bytes, TWOBYTES) == TWOBYTES) {
*val = to_int(bytes);
return (0);
} else
return (1);
}

/* wrint() writes an unsigned int to standard output in archive order */
int wrint (val)
unsigned int val;
{
BYTE bytes[TWOBYTES];
splitint (bytes, val);
if (zoowrite (STDOUT, bytes, TWOBYTES) == TWOBYTES)
return (0);
else
return (1);
}
#endif /* FILTER */

#ifdef TRACE_IO
/* dump contents of archive header */
void show_h (zoo_header)
struct zoo_header *zoo_header;
{
int i;
printf ("Header text:\n");
for (i = 0; i < SIZ_TEXT; i++) { /* ASSUMES ASCII TEXT */
int c;
c = zoo_header->text[i];
if (c >= ' ' && c < 0x7f)
putchar (c);
else {
putchar ('^');
putchar (i & 0x40);
}
}
putchar('\n');
printf ("zoo_tag = [%8lx] zoo_start = [%8lx] zoo_minus = [%8lx]\n",
zoo_header->zoo_tag, zoo_header->zoo_start,
zoo_header->zoo_minus);
printf ("major_ver.minor_ver = [%d.%d]\n",
zoo_header->major_ver, zoo_header->minor_ver);
if (zoo_header->zoo_start != FIXED_OFFSET) {
printf ("type = [%d] ", zoo_header->type);
printf ("acmt_pos = [%8lx] acmt_len = [%4x] vdata = [%2x]",
zoo_header->acmt_pos, zoo_header->acmt_len, zoo_header->vdata);
printf ("\n");
}
printf ("---------\n");
}

/* dump contents of directory entry */
void show_dir (direntry)
struct direntry *direntry;
{
printf ("Directory entry for file [%s][%s]:\n",
direntry->fname, direntry->lfname);
printf ("tag = [%8lx] type = [%d] PM = [%d] Next = [%8lx] Offset = [%8lx]\n",
direntry->zoo_tag, (int) direntry->type,
(int) direntry->packing_method, direntry->next,
direntry->offset);
printf ("Orig size = [%ld] Size now = [%ld] dmaj_v.dmin_v = [%d.%d]\n",
direntry->org_size, direntry->size_now,
(int) direntry->major_ver, (int) direntry->minor_ver);
printf ("Struc = [%d] DEL = [%d] comment_offset = [%8lx] cmt_size = [%d]\n",
(int) direntry->struc, (int) direntry->deleted, direntry->comment,
direntry->cmt_size);
printf ("var_dir_len = [%d] TZ = [%d] dir_crc = [%4x]\n",
direntry->var_dir_len, (int) direntry->tz, direntry->dir_crc);
printf ("system_id = [%d] dirlen = [%d] namlen = [%d] fattr=[%24lx]\n",
direntry->system_id, direntry->dirlen, direntry->namlen, direntry->fattr);
printf ("vflag = [%4x] version_no = [%4x]\n",
direntry->vflag, direntry->version_no);
if (direntry->dirlen > 0)
printf ("dirname = [%s]\n", direntry->dirname);
printf ("---------\n");
}
#endif /* TRACE_IO */


  3 Responses to “Category : OS/2 Files
Archive   : ZOO21E.ZIP
Filename : PORTABLE.C

  1. Very nice! Thank you for this wonderful archive. I wonder why I found it only now. Long live the BBS file archives!

  2. This is so awesome! 😀 I’d be cool if you could download an entire archive of this at once, though.

  3. But one thing that puzzles me is the “mtswslnkmcjklsdlsbdmMICROSOFT” string. There is an article about it here. It is definitely worth a read: http://www.os2museum.com/wp/mtswslnk/