Compatibility of Types for C and Pascal

Table 6-1 and Table 6-2 list the default sizes and alignments of compatible types for C and Pascal.
Table 6-1 C and Pascal Size and Alignment of Compatible Types

Pascal Type C Type Size (bytes) Alignment (bytes)

double double 8 8

longreal double 8 8

real double 8 8

single float 4 4

shortreal float 4 4

integer16 short int 2 2

integer32 int 4 4

integer int 4 4

-128..127 char 1 1

boolean char 1 1

alfa char a[10] 10 1

char char 1 1

string char a[80] 80 1

varying[n] of char struct{int, char[n]} - 4

record struct/union - Same as element type

array array - Same as element type

variant record struct/union - -

fields in packed record bit field - -

Table 6-2 C and Pascal Size and Alignment of Compatible Types with -xl -

Pascal Type C Type Size (bytes) Alignment (bytes)

real float 4 4

integer short int 2 2

C has several assumptions about strings. All C strings are:
 Passed by reference since C strings are arrays
 Terminated by a null byte
 Located in static variable storage
You can satisfy these assumptions as follows:
 Pass by reference by making the strings var, in, out, or in out parameters.
 Provide the null byte explicitly before passing a string to C. Pascal guarantees the null byte only if the
string is a constant. The null byte is not required by the ISO Pascal Standard.
Array Indexes
Pascal array indexes can start at any integer; C array indexes always start at zero.
Aggregate Types
Aggregate types include arrays, varying arrays, sets, strings, alphas, records, and variant records.
Pascal aggregate types may require alignment and layout adjustment to match C unions, structures, and
arrays. Pascal aggregate types can be any of the following: arrays, varying arrays, sets, strings, alphas,
records, or variant records.
However, you can use the -calign option to eliminate some of these differences. With -calign, the following
assumptions are made:
 Pascal records have the same data layout as C structures.
 Arrays have the same data layout in both languages. However, if you use the -xl option in addition to
-calign, boolean arrays with an odd number of elements are different.
 Pascal variants are the same as C unions.
Incompatibilities
This section describes the incompatibilities between C and Pascal types.
Enumerated Types
C enumerated types are incompatible with Pascal enumerated types. Pascal enumerated types are represented
internally by sequences of integral values starting with 0. Storage is allocated for a variable of an enumerated
type as if the type was a subrange of integer. For example, an enumerated type of fewer than 128 elements is
treated as 0..127, which, according to the rules above, is equivalent to a char in C.
C enumerated types are allocated a full word and can take on arbitrary integer values.
Pascal Set Types
In Pascal, a set type is implemented as a bit vector, which is similar to a C short-word array, where each
short-word contains two bytes. Thus, sets are bit-vectors, and they are allocated in multiples of 16 bits. To
find out the size of a set, enter the following code:
ceiling( ord( highest_element ) / 16 )
Direct access to individual elements of a set is highly machine-dependent and should be avoided.

General Parameter Passing in C and Pascal

A few general rules apply to parameter passing:
 C passes all arrays by reference since C strings are arrays.
 C passes all structures by value.
 In C, if you want to pass anything else by reference, then you must explicitly prepend the reference
ampersand (&), or pass an explicit pointer.
 Pascal passes all parameters by value unless you explicitly state that they are var, in out, or out
parameters, in which case they are passed by reference.

Procedure Calls: C-Pascal

Here are examples of how a C main program calls a Pascal procedure:
The Pascal procedure, Samp, in the file Samp.p. Note the procedure Samp(var i: integer; var r: real);
procedure definition. begin i := 9; r := 9.9 end; { Samp }

The C main program, SampMain.c. Note the #include <stdio.h>

procedure definition and call. extern void Samp(int *, double *);
int main (void)
{ int i ; double d ; Samp(&i, &d) ;
printf ("%d %3.1f \n", i, d) ; }

The commands to compile and execute hostname% pc -c Samp.p

Samp.p and SampMain.c hostname% cc Samp.o SampMain.c
hostname% a.out 9 9.9
Variable Parameters
Pascal passes all variable parameters by reference, which C can do, too.
Simple Types without -xl
Without -xl, simple types match, as in the following example:
The Pascal procedure, SimVar.p procedure SimVar(var t, f: boolean; var c: char; var si:
integer16; var i: integer; var sr: shortreal;var r: real);
begin t := true; f := false; c := 'z'; si := 9;
i := 9; sr := 9.9; r := 9.9; end; { SimVar }

The C main program, SimVarMain.c #include <stdio.h>

extern void SimVar(char *, char *, char *, short *, int *, float *, double
*);
int main(void)
{ char t, f, c;
short si;
int i;
float sr;
double r;
SimVar(&t, &f, &c, &si, &i, &sr, &r);
printf(" %08o %08o %c %d %d %3.1f %3.1f \n",
t, f, c, si, i, sr, r);}

The commands to compile and execute SimVar.p and SimVarMain.c hostname% pc -c SimVar.p
hostname% cc SimVar.o SimVarMain.c
hostname% a.out
00000001 00000000 z 9 9 9.9 9.9
Simple Types with -xl
With the -xl option, the Pascal real must be paired with a C float, and the Pascal integer must be paired with a
C short int.
Strings of Characters
The C counterpart to the Pascal alfa and string types are arrays; C passes all arrays by reference. The C
counterpart to the Pascal varying is a structure; C passes structures by value.
Before you call Pascal with a null varying string, set the byte count to zero because that is what Pascal
assumes about such strings.
C can pass a structure consisting of a four-byte integer and an array of characters to a Pascal procedure,
expecting a var parameter that is a variable-length string.
See the following example:
The Pascal procedure, StrVar.p type

TVarStr = varying [25] of char;

procedure StrVar( var a: alfa; var s: string; var v: TVarStr);
begin a := 'abcdefghi' + chr(0);
s := 'abcdefghijklmnopqrstuvwxyz' + chr(0);
v := 'varstr' + chr(0); end; { StrVar }

The C main program, StrVarMain.c #include <stdio.h>

#include <string.h>
struct TVarLenStr { int nbytes; char a[25];};
extern void StrVar(char *, char *, struct TVarLenStr *);
int main(void)

{ struct TVarLenStr vls;

 char s10[10], s80[80], s25[25];
vls.nbytes = 0;
StrVar(s10, s80, &vls);
strncpy(s25, vls.a, vls.nbytes);
printf(" s10 = '%s' \n s80 = '%s' \n s25 = '%s' \n",
s10, s80, s25);
printf(" strlen(s25) = %d \n", strlen(s25));}

The commands to compile and hostname% pc -c StrVar.p hostname% cc StrVar.o StrVarMain.c -lpc
execute StrVar.p and StrVarMain.c hostname% a.out s10='abcdefghi'
s80='abcdefghijklmnopqrtstuvwxyz' s25='varstr' strlen(s25)=6

Fixed Arrays
For a fixed array parameter, pass the same type and size by reference, as shown in the following example:
The Pascal procedure, FixVec.p type
VecTyp = array [0..8] of integer;
procedure FixVec(var V: TVec; var Sum: integer);
var i: integer;
begin Sum := 0; for i := 0 to 8 do Sum := Sum + V[i]
end; { FixVec }

The C main program, FixVecMain.c #include <stdio.h>

extern void FixVec(int [], int *);
int main(void)
{ int Sum; static int a[] = {0,1,2,3,4,5,6,7,8}; FixVec(a, &Sum);
printf(" %d \n", Sum); }

The commands to compile and hostname% pc -c -calign FixVec.p

execute FixVec.p and FixVecMain.c hostname% cc FixVec.o FixVecMain.c -lpc
hostname% a.out 36

Although it does not apply in this example, arrays of aggregates in Pascal have, by default, a size that is
always a multiple of four bytes. When you use the -calign option to compile the Pascal code, that difference
with C is eliminated.
The following example illustrates this point. The string 'Sunday' only gets through to the C main program
when you compile the Pascal routine using
-calign.
The Pascal procedure, DaysOfWeek.p type
TDay= array [0..8] of char;
TWeek = array [0..6] of day;
TYear = array [0..51] of week;
procedure DaysOfWeek(var Y: TYear);
begin v[1][1] := 'Sunday'; end;
The C main program, #include <stdio.h>
DaysOfWeekMain.c extern void DaysOfWeek(char [][7][9]);
int main(void)
{ char Year[52][7][9]; DaysOfWeek(Year);
printf(" Day = '%s' \n", Year[1][1]);}

The commands to compile and execute hostname% pc -c DaysOfWeek.p

DaysOfWeek.p and DaysOfWeekMain.c without hostname% cc DaysOfWeek.o DaysOfWeekMain.c –
-calign lpc hostname% a.out Day = ''

The commands to compile and execute DaysOfWeek.p and hostname% pc -c -calign DaysOfWeek.p
DaysOfWeekMain.c with -calign hostname% cc DaysOfWeek.o
DaysOfWeekMain.c -lpc
hostname% a.out
day = 'Sunday '

The univ Arrays

You can pass any size array to a Pascal procedure expecting a univ array, although there is no special gain in
doing so, because there is no type or size checking for separate compilations. However, if you want to use an
existing Pascal procedure that has a univ array, you can do so. All univ arrays that are in, out, in out, or var
parameters pass by reference.
The Pascal procedure, UniVec.p, which Type TVec = array [0..9] of integer;
defines a 10-element array procedure UniVec(var V: univ TVec; in Last: integer;
var Sum: integer);
var i: integer;
begin Sum := 0;
for i := 0 to Last do Sum := Sum + V[i];
end; { UniVec }

The C main program, UniVecMain.c, which #include <stdio.h>

passes a 3-element array to the Pascal extern void UniVec(int *, int, int *);
procedure written to do a 10-element array int main(void)
{ int Sum;
static int a[] = {7, 8, 9};
UniVec(a, 2, &Sum);
printf(" %d \n", Sum);}

The commands to compile and execute UniVec.p and hostname% pc -c -calign UniVec.p
UniVecMain.c with -calign hostname% cc UniVec.o UniVecMain.c -lpc
hostname% a.out 24

Conformant Arrays
For single-dimension conformant arrays, pass upper and lower bounds, placed after the declared parameter
list. If the array is multidimensional, pass element widths as well, one element width for each dimension,
except the last one.
See this example:
function ip(var x: array [lb..ub: integer] of real): real;
extern double ip(double [], int, int); ... double v1[10]; double z; z = ip(v1, 0, 9); ...

One bounds pair may apply to several arrays if they are declared in the same parameter group:
function ip(var x,y:array[lb..ub:integer] of real):real;
... double v1[10], v2[10] ; extern double ip() ; double z ; z = ip ( v1, v2, 0, 9 ) ; ...

With multidimensional arrays, for all dimensions but the last one, pass the low bound, high bound, and
element width.
Examples of single-dimension, multidimension, and array-of-character conformant arrays follow.
Conformant arrays are included here only because they are a relatively standard feature; there are usually
more efficient and simpler ways to do the same thing.
Example 1: Single-Dimension Array

The Pascal procedure, IntCA.p. Pascal procedure IntCA(var a: array [lb..ub: integer] of integer);
passes the bounds pair. begin a[1] := 1; a[2] := 2 end; { IntCA }

The C main program, IntCAMain.c #include <stdio.h> extern void IntCA(int [], int, int);
int main(void)
{ int k ; static int s[] = { 0, 0, 0 };
IntCA (s, 0, sizeof(s)-1); for (k=0 ; k < 3 ; k++)
printf(" %d \n", s[k]); }

The commands to compile and hostname% pc -c -calign IntCA.p

execute IntCA.p and IntCAMain.c hostname% cc IntCA.o IntCAMain.c –lpc hostname% a.out
with -calign 0 1 2

Example 2: Multi-Dimension Array

The Pascal procedure, RealCA.p. procedure RealCA(var A: array [r1..r2: integer] of

Pascal passes low bound, high array [c1..c2: integer] of real);
bound, and element width. var col, row: integer;
begin for row := r1 to r2 do
for col := c1 to c2 do
if row = col then A[row, col] := 1.0
else A[row, col] := 0.0
end; { RealCA }

The C main program, RealCAMain.c. Array #include <stdio.h> #define NC 3 #define NR 2

M has 2 rows of 3 columns each. c1 and c2 extern void RealCA(double [][NC], int, int, int, int, int);
are the first and last columns. r1 and r2 are the int main(void)
first and last rows. wc is the width of a {double M[NR][NC]; int col, c1, c2, row, r1, r2, wc, wr;
column element (smallest) and is equal to c1 = 0; r1 = 0; c2 = NC - 1; r2 = NR - 1;
sizeof( M[0][0] ). wr is the width of a row wc = sizeof(M[0][0]); wr = (c2 - c1 + 1) * wc;
element (next largest) and is equal to RealCA(M, r1, r2, wr, c1, c2);
(c2-c1+1) * wc. for (row = r1; row <= r2; row++) { printf("\n");
for (col = c1; col <= c2; col++)
 printf("%4.1f", M[row][col]); };
printf("\n");}

The commands to compile and execute hostname% pc -c -calign RealCA.p

RealCA.p and RealCAMain.c with –calign hostname% cc RealCA.o RealCAMain.c -lpc
hostname% a.out 1.0 0.0 0.0 0.0 1.0 0.0

If wc is the width of the smallest element, as determined by sizeof(), then the width of the next largest
element is the number of those smaller elements in the next larger element multiplied by wc.
width of next largest element = (ub - lb + 1) * wc
In general, (lb, ub, wc) are the bounds and element width of the next lower dimension of the array. This
definition is recursive.
Example 3: Array of Characters

The Pascal procedure, ChrCAVar.p procedure ChrCAVar(var a: array [lb..ub: integer] of char);
begin a[0] := 'T'; a[13] := 'o'; end; { ChrCAVar }

The C main program, #include <stdio.h> extern void ChrCAVar(char [], int, int);
ChrCAVarMain.c. For C, the lower int main(void)
bound is always 0. { static char s[] = "this is a string" ;
ChrCAVar( s, 0, sizeof(s)-1) ; /*(s, lower, upper)*/
printf("%11s \n", s) ;}

The commands to compile and hostname% pc -c -calign ChrCAVar.p

execute ChrCAVar.p and hostname% cc ChrCAVar.o ChrCAVarMain.c -lpc
ChrCAVarMain.c hostname% a.out This is a string

Records and Structures

In most cases, a Pascal record describes the same objects as its C structure equivalent, provided that the
components have compatible types and are declared in the same order. The compatibility of the types
depends mostly on size and alignment. For more information on size and alignments of simple components,
see "Compatibility of Types for C and Pascal" on page 94.
By default, the alignment of a record is always four bytes and the size of a record is always a multiple of four
bytes. However, when you use -calign in compiling the Pascal code, the size and alignment of the Pascal
record matches the size and alignment of the equivalent C structure.
A Pascal record of an integer and a character string matches a C structure of the same constructs, as follows:
The Pascal procedure, StruChr.p. It Type TLenStr = record
is safer for the Pascal procedure to nbytes: integer;
explicitly provide the null byte and chrstr: array [0..24] of char
include it in the count before the end;
string is passed to C. procedure StruChr(var v: TLenStr);
begin v.NBytes := 14; v.ChrStr := 'St. Petersburg' + chr(0); end;

The C main program, StruChrMain.c #include <stdio.h> #include <string.h> struct TVarLenStr {
int NBytes; char a[25];};
extern void StruChr(struct TVarLenStr *);
int main(void)
 { struct TVarLenStr vls; char s25[25]; vls.NBytes = 0;
StruChr(&vls); strncpy(s25, vls.a, vls.NBytes);
printf(" s25 = '%s' \n", s25);
printf(" strlen(s25) = %d \n", strlen(s25));}

The commands to compile and hostname% pc -c StruChr.p

execute StruChr.p and StruChrMain.c hostname% cc StruChr.o StruChrMain.c -lpc
hostname% a.out s25='St. Petersburg' strlen(s25) = 13

The record in the example above has, by default, the same size and alignment as the equivalent C record.
Some records, though, are laid out differently unless you use the -calign option.
Consider this example:
The Pascal routine, DayWeather.p Type TDayWeather = record
TDay: array [0..8] of char;
TWeather: array [0..20] of char;
end;
TDayWeatherArray = array [0..1] of TDayWeather;
procedure DayWeather(var W: TDayWeatherArray;
var WeatherSize: integer);
begin W[1].TDay := 'Sunday' + chr(0);
W[1].TWeather := 'Sunny' + chr(0);
WeatherSize := 5; end; { StruChr }

The C main program, #include <stdio.h> #include <string.h> struct TDayRec {

DayWeatherMain.c char TDay[9]; char TWeather[21];};
extern void DayWeather(struct TDayRec [], int *);
int main(void)
{ char s25[25]; char t25[25]; struct TDayRec dr[2];
int nbytes = 0; DayWeather(dr, &nbytes);
strncpy(s25, dr[1].TDay, 6); printf(" day = '%s' \n", s25);
strncpy(t25, dr[1].TWeather, nbytes); printf(" weather = '%s' \n", t25);}

When you compile the Pascal routine without using the -calign option, the program does not work correctly.
The commands to compile and hostname% pc -c DayWeather.p
execute DayWeather.p and hostname% cc DayWeather.o DayWeatherMain.c -lpc
DayWeatherMain.c without -calign hostname% a.out day = '' weather = ' sun'

The commands to compile and execute hostname% pc -calign -c DayWeather.p

DayWeather.p and DayWeatherMain.c hostname% cc DayWeather.o DayWeatherMain.c -lpc
with -calign hostname% a.out day = 'Sunday' weather = 'sunny'

Variant Records
C equivalents of variant records can sometimes be constructed, although there is some variation with
architecture and sometimes a need to adjust alignment. You can avoid the need to adjust alignment by using
the -calign option.
The Pascal procedure, VarRec.p type
 vr = record
case tag: char of
'a': (ch1, ch2: char);
'b': (flag: boolean);
'K': (ALIGN: integer);
end;
procedure VarRec(var x: vr);
begin if x.ch1 = 'a' then x.ch2 := 'Z' end; { VarRec }

The C main program, VarRecMain.c #include <stdio.h>

struct vlr { char tag; union { struct {char ch1, ch2;}a_var;
struct {char flag; }b_var; struct {int ALIGN;}c_var; }var_part;};
extern void VarRec(struct vlr *); int main(void)
{ struct vlr *x; x = (struct vlr *)malloc(sizeof(struct vlr));
x->tag = 'a'; x->var_part.a_var.ch1 = 'a';
x->var_part.a_var.ch2 = 'b'; VarRec(x);
printf(" %c \n", x->var_part.a_var.ch2);}

The commands to compile and hostname% pc -c -calign VarRec.p

execute VarRec.p and VarRecMain.c hostname% cc VarRec.o VarRecMain.c -lpc
hostname% a.out Z

Pascal Set Type

In Pascal, a set type is implemented as a bit vector, which is similar to a C short-word array. Direct access to
individual elements of a set is highly machine-dependent and should be avoided.
In Pascal, bits are numbered within a byte from the most significant to least, as shown in Table 6-3.
Table 6-3 Set Implementation

Set Bit Numbering

set+3: 31, 30, 29, 28, 27, 26, 25, 24

set+2: 23, 22, 21, 20, 19, 18, 17, 16

set+1: 15, 14, 13, 12, 11, 10, 9, 8

set+0: 7, 6, 5, 4, 3, 2, 1, 0

In C, a set could be described as a short-word array beginning at an even address. With the current set
representation, it does not matter what the lower-bound value is.
The nth element in a set [lower...upper] can be tested as follows:
#define LG2BITSLONG 5 /* log2( bits in long word) */
#define LG2BITSWORD 4 /* log2( bits in short word) */
#define MSKBITSLONG 0x1f
#define MSKBITSHORT 0x0
short *setptr; /* set as array of shorts */ int upper; /* upper bound of the set */
int elem; /* ordinal value of set element */ int i;
if ( setptr[elem >> LG2BITSWORD] & (1 << (elem & MSKBITSWORD)) ) {
/* elem is in set */ }
Pascal intset Type
The Pascal intset type is predefined as set of [0..127]. A variable of this type takes 16 bytes of storage.
The Pascal procedure, IntSetVar.p, which has an intset of the elements procedure IntSetVar(var s: intset);
[1, 3, 7, 8] begin
s := [1, 3, 7, 8]
end; { IntSetVar }

The C main program, #include <stdio.h> extern void IntSetVar(unsigned int *); int main(void)
IntSetVarMain.c { int w ; unsigned int *p, *s ; s = (unsigned int *) malloc(16);
IntSetVar(s) ; for (w = 0, p = s ; w < 4 ; w++, p++)
printf("%012o %3d \n", *p, w); printf(" 110 001 010 (binary, word 4) \n");
printf(" 876 543 210 (bits, word 4)" \n"); }

The commands to compile and execute hostname% pc -c IntSetVar.p

IntSetVar.p and IntSetVarMain.c. The hostname% cc IntSetVar.o IntSetVarMain.c -lpc
output of this example depends on the hostname% a.out 000000000000 0
architecture of your machine. 000000000000 1 000000000000 2 000000000612 3
110 001 010 (binary, word 4) 876 543 210 (bits, word 4)
Value Parameters
There are three types of value parameters in Pascal.
Simple Types without -xl
Without -xl, simple types match, as in the following example:
The Pascal procedure, SimVal. p. t, f, c, procedure SimVal( t, f: boolean; c: char; si: integer16;
i, r, and s are value parameters. i: integer; sr: shortreal; r: real; var reply: integer);
begin Reply := 0;
if t then Reply := Reply + 1;
if not f then Reply := Reply + 8;
if c='z' then Reply := Reply + 64;
if si=9 then Reply := Reply + 512;
if i=9 then Reply := Reply + 4096;
if sr=shortreal(9.9) then Reply := Reply + 32768;
if r=9.9 then Reply := Reply + 262144;
end; { SimVal }

The C main program, SimValMain.c #include <stdio.h> extern void SimVal(char, char, char, short,
int, float, double, int *); int main(void)
{ char t = 1, f = 0; char c = 'z';
short si = 9; int i = 9; float sr = 9.9;
double r = 9.9; int args;
SimVal(t, f, c, si, i, sr, r, &args);
printf(" args = %06o \n", args);

The commands to compile and hostname% pc -c SimVal.p hostname% cc SimVal.o SimValMain.c –lpc
execute SimVal.p and SimValMain.c hostname% a.out args=111111

If no function prototype is provided for SimVal in SimValMain.c, then sr:shortreal must be changed to sr:real
in SimVal.p. This change is necessary because in C, a float is promoted to double in the absence of function
prototypes. In -xl mode, change sr:shortreal to sr:longreal.
Simple Types with -xl
With -xl, the Pascal real must be paired with a C float, and the Pascal integer must be paired with a C short
int.
Arrays
Since C cannot pass arrays by value, it cannot pass strings of characters, fixed arrays, or univ arrays by value.
Conformant Arrays
Pascal passes all value parameters on the stack or in registers, except for value conformant array parameters,
which are handled by creating a copy in the caller environment and passing a pointer to the copy. In addition,
the bounds of the array must be passed (see "Conformant Arrays" on page 105).
This example is the same as the single-dimension example in "Conformant Arrays," except that the var prefix
is deleted.
The Pascal procedure, ChrCAVal.p procedure ChrCAVal(a: array [lb..ub: integer] of char);
begin a[0] := 'T'; a[13] := 'o'; end; { ChrCAVal }

The C main program, #include <stdio.h> extern void ChrCAVal(char [], int, int);
ChrCAValMain.c int main(void) { static char s[] = "This is a string";
ChrCAVal(s, 0, sizeof(s) -1); printf(" %11s \n", s); }

The commands to compile and execute hostname% pc -c -calign ChrCAVal.p

ChrCAVal.p and ChrCAValMain.c with hostname% cc ChrCAVal.o ChrCAValMain.c -lpc
-calign hostname% a.out This is a string

Function Return Values

Function return values match types the same as with parameters, and they pass in much the same way.
Simple Types
The simple types pass in a straightforward way, as follows:
The Pascal function, RetReal.p function RetReal(x: real): real;
begin RetReal := x + 1.0 end; { RetReal }

The C main program, RetRealMain.c #include <stdio.h> extern double RetReal(double); int main(void)
{ double r, s; r = 2.0; s = RetReal(r); printf(" %f \n", s);}

The commands to compile and execute hostname% pc -c RetReal.p hostname% cc RetReal.o RetRealMain.c
RetReal.p and RetRealMain.c hostname% a.out 3.000000

Input and Output

If your C main program calls a Pascal procedure that does I/O, then include the following code before you
call the Pascal procedure:
PASCAL_IO_INIT();

Also, in the C main program just before exit, add the following line:
PASCAL_IO_DONE();
See this example:
The Pascal procedure, pasc_read.p procedure pasc_read;
var Tfile : text; data : integer;
begin writeln ('In Pascal procedure');
reset(Tfile, 'data.txt');
 while not eof(Tfile) do begin
readln (Tfile,data) end;
writeln ('At end of Pascal procedure',data) end;

The C main program, main.c #include <stdio.h> extern void pasc_read(); int main(void)
{FILE *ptr; printf ("Calling Pascal routine\n");PASCAL_IO_INIT();
pasc_read(); PASCAL_IO_DONE(); printf ("After Pascal routine\n");}

The commands to compile hostname% pc pasc_read.p -c hostname% cc -g main.c pasc_read.o -lpc

and execute pasc_read.p and hostname% a.out Calling Pascal routine In Pascal procedure
main.c At end of Pascal procedure 1 After Pascal routine

Procedure Calls: Pascal-C

This section parallels the section, "Procedure Calls: C-Pascal" on page 97. Earlier comments and restrictions
also apply here.
Variable Parameters
Pascal passes all variable parameters by reference, which C can do, too.
Simple Types
Simple types pass in a straightforward manner, as follows:
The C function, SimRef.c void SimRef( char *t, char *f, char *c, short *si, int *i, float *sr,
double *r)
{ *t = 1; *f = 0; *c = 'z'; *si = 9; *i = 9; *sr = 9.9; *r = 9.9;}

The Pascal main program SimRefMain(output);

program, SimRefMain.p var t, f: boolean; c: char; si: integer16; i: integer; sr: shortreal; r: real;
procedure SimRef(var t, f: boolean; var c: char; var si: integer16;
var i: integer; var sr: shortreal; var r: real);
external c;
begin SimRef(t, f, c, si, i, sr, r); writeln(t, f: 6, c: 2, si: 2, i: 2, sr :4:1, r :4:1);
end. { SimRefMain }

The commands to compile and execute hostname% cc -c SimRef.c hostname% pc SimRef.o SimRefMain.p
SimRef.c and SimRefMain.p hostname% a.out true false z 9 9 9.9 9.9

Strings of Characters
The alfa and string types pass simply; varying strings are more complicated. All pass by reference.
The C #include <string.h> struct TVarLenStr { int nbytes; char a[26];
function, }; void StrVar(char *s10, char *s80, struct TVarLenStr *vls) {
StrVar.c static char ax[11] = "abcdefghij"; static char sx[81] = "abcdefghijklmnopqrstuvwxyz";
static char vx[6] = "varstr"; strncpy(s10, ax, 11); strncpy(s80, sx, 80); strncpy(vls->a, vx, 6);
vls->nbytes = 6;}

The Pascal main program StrVarMain(output);

program, type TVarStr = varying[26] of char;
StrVarMain.p var a: alfa; s: string; v: TVarstr;
procedure StrVar(var a: alfa; var s: string; var v: TVarStr); external c;
 begin StrVar(a, s, v); writeln(a); writeln(s); writeln(v);
writeln(' length(v) = ', length(v) :2);end. { StrVarMain }

The commands to compile hostname% cc -c StrVar.c hostname% pc StrVar.o StrVarMain.p

and execute StrVar.c and hostname% a.out abcdefghij abcdefghijklmnopqrtstuvwxyz
StrVarMain.p varstr length(v) = 6

Avoid constructs that rely on strings being in static variable storage. For example, you could use mktemp(3)
in Pascal as follows:
Incorrect use of string in static variable storage tmp := mktemp('/tmp/eph.xxxxxx')

This use is incorrect, since mktemp()modifies its argument. Instead, use the C library routine strncpy() (see
string(3)) to copy the string constant to a declared char array variable, as in:
Correct solution using the C program Use_mktemp ;
library routine strncpy() procedure strncpy( var dest: univ string ; var srce: univ string ;
length: integer) ; external c ;
procedure mktemp(var dest: univ string); external c;
... var path: string ;
begin ... strncpy( path, '/tmp/eph.xxxxxx', sizeof(path)) ;
mktemp( path ) ; ... end .

Fixed Arrays
For a fixed-array parameter, pass the same type and size, as in this example:
The C function, FixVec.c void FixVec(int V[9], int *Sum)
{ int i; *Sum = 0; for (i = 0; i <= 8; i++) *Sum = *Sum + V[i];}

The Pascal main program FixVecMain(output);

program, FixVecMain.p type TVec = array [0..8] of integer;
var V: TVec := [0, 1, 2, 3, 4, 5, 6, 7, 8]; Sum: integer;
procedure FixVec(var XV: TVec; var XSum: integer); external c;
begin FixVec(V, Sum); writeln(Sum: 3); end. { FixVecMain }

The commands to compile and hostname% cc -c FixVec.c hostname% pc -calign FixVec.o

execute FixVec.c and FixVecMain.p FixVecMain.p hostname% a.out 36

The -calign option is not needed for this example, but may be necessary if the array parameter is an array of
aggregates.
The univ Arrays
The univ arrays that are in, out, in out, or var parameters pass by reference.
Here is an example:
The C function, UniVec.c void UniVec(int V[3], int Last, int *Sum)
{ int i; *Sum = 0; for (i = 0; i <= Last; i++) *Sum += V[i];}

The Pascal main program UniVecMain(output);

program, UniVecMain.p type TVec = array [0..9] of integer;
 var Sum: integer; V: array [0..2] of integer;
procedure UniVec(var V: univ TVec; in Last: integer; var Sum: integer);
external c;
begin V[0] := 7; V[1] := 8; V[2] := 9; UniVec(V, 2, Sum);
writeln(Sum);end. { UniVecMain }

The commands to compile and hostname% cc -c UniVec.c

execute UniVec.c and hostname% pc -calign UniVec.o UniVecMain.p
UniVecMain.p hostname% a.out 24
The -calign option is not needed for this example, but may be necessary if the array parameter is an array of
aggregates.
Conformant Arrays
For single-dimension conformant arrays, pass upper and lower bounds placed after the declared parameter
list. If the array is multidimensional, pass element widths as well, one element width for each dimension,
except the last one. Chapter 8, "The FORTRAN-Pascal Interface," has an example of multidimensional
conformant array passing.
The following example uses a single-dimension array:
The C function, IntCA.c void IntCA(int a[], int lb, int ub)
{int k; for (k=0; k <= ub - lb; k++) a[k] = 4;}

The Pascal main program, program IntCAMain(output);

IntCAMain.p. Note that var s: array [1..3] of integer; i: integer;
what Pascal passes as s, is procedure IntCA(var a: array [lb..ub: integer] of integer);
received in C as a, lb, ub. external c;
begin IntCA(s); for i := 1 to 3 do write(s[i]);
writeln end. { IntCAMain }

The commands to compile and execute hostname% cc -c IntCA.c

IntCA.c and IntCAMain.p hostname% pc -calign IntCA.o IntCAMain.p
hostname% a.out 4 4 4
The -calign option is not needed for this example, but may be necessary if the array parameter is an array of
aggregates.
Records and Structures
In most cases, a Pascal record describes the same objects as its C structure equivalent, provided that the
components have compatible types and are declared in the same order. For more information, see
"Compatibility of Types for C and Pascal" on page 94.
Records that contain aggregates may differ because aggregates in C and Pascal sometimes have different
sizes and alignments. If you compile the Pascal code with the -calign option, the differences are eliminated.
A Pascal record of an integer and a character string matches a C structure of an integer and an array of char
values, as follows:
The C function, StruChr.c #include <string.h> struct TVarLenStr { int nbytes; char a[26];};
void StruChr(struct TVarLenStr *v) { strncpy(v->a, "strvar", 6);
v->nbytes = 6;}

The Pascal main program, program StruChrMain(output);

StruChrMain.p type TVarLenStr = record
nbytes: integer; a: array [0..25] of char
end;
var vls: TVarLenStr; i: integer;
procedure StruChr(var vls: TVarLenStr); external c;
 begin StruChr(vls); write(' string=''');
for i := 0 to vls.nbytes - 1 do write(vls.a[i]);
writeln(''''); writeln(' length = ', vls.nbytes) end. { StruChrMain }

The commands to compile and execute hostname% cc -c StruChr.c hostname% pc -calign StruChr.o
StruChr.c and StruChrMain.p StruChrMain.p hostname% a.out string=' strvar' length= 6

Variant Records
C equivalents of variant records can sometimes be constructed, although there is some variation with the
architecture, and sometimes you have to adjust the alignment.
Following are some examples:
The C function, VarRec.c struct vlr { char tag; union {struct { char ch1, ch2;} a_var;struct {
char flag;}b_var;struct { int ALIGN; }c_var;} var_part;};
void VarRec(struct vlr *x) {if (x->var_part.a_var.ch1 == 'a')
x->var_part.a_var.ch2 = 'Z';}

The Pascal main program VarRecMain;

program, VarRecMain.p type vr = record
case tag: char of
'a': (ch1, ch2: char);
'b': (flag: boolean);
'K': (ALIGN: integer)
end;
var x: vr;
procedure VarRec(var d: vr); external c;
begin x.tag := 'a'; x.ch1 := 'a'; x.ch2 := 'b'; VarRec(x);
writeln(x.ch2) end. { VarRecMain }

The commands to compile and execute hostname% cc -c VarRec.c hostname% pc -calign VarRec.o
VarRec.c and VarRecMain.p VarRecMain.p hostname% a.out Z

The -calign option is not needed in the previous example, but may be necessary if the record contains
aggregates.
Non-Pascal Procedures
When you use the -xl option in compiling Pascal code, you can use the nonpascal keyword to declare that an
external procedure is written in another language. This keyword generally causes everything to be passed by
reference.
See this example:
The C function, NonPas.c. In the function for_C, s #include <stdio.h> void for_C(char *s, int len)
is a pointer (declared var in the procedure {int i; for (i = 0; i < len; i++) putchar(s[i]);
declaration), and len is not a pointer (not declared putchar('\n');}
var in the procedure declaration). In the function void for_NonPascal(char *s, int *len)
for_nonpascal, s is still a pointer (though not { int i; for (i = 0; i < *len; i++) putchar(s[i]);
declared var in the procedure declaration), and len putchar('\n');}
is now a pointer (though not declared var).

The Pascal main program NonPasMain; var s: string;

program, procedure for_C(var s: string; len: integer); external c;
NonPasMain.p procedure for_NonPascal(var s: string; len: integer); nonpascal;
 begin s :='Hello from Pascal'; for_C(s, 18); for_NonPascal(s, 18);
end. { NonPasMain }

The commands to compile and hostname% cc -c NonPas.c hostname% pc NonPas.o NonPasMain.p

execute NonPas.c and NonPasMain.p hostname% a.out Hello from Pascal Hello from Pascal

Value Parameters
In general, Pascal passes value parameters in registers or on the stack, widening to a full word if necessary.
Simple Types
With value parameters, simple types match, as in the following example:
The C function, SimVal.c void SimVal(char t,char f, char c, short si, int i, float sr, double r,
int *reply)
{*reply = 0; if (t) *reply += 01; if (!f) *reply += 010;
if (c == 'z') *reply += 0100; if (si == 9) *reply += 01000;
if (i == 9) *reply += 010000; if (sr ==(float)9.9) *reply += 0100000;
if (r == 9.9) *reply +=01000000;}

The Pascal main program SimVal(output);

program, SimValMain.p var t: boolean:= true; f: boolean := false; c: char si: integer16:= 9;
i: integer:= 9; sr: shortreal := 9.9; r: double := 9.9; args: integer;
procedure SimVal(t, f: boolean;c: char; si: integer16; i: integer; sr: shortreal;
r: double; var Reply: integer); external c;
begin SimVal(t, f, c, si, i, sr, r, args); writeln(' args = ', args :6 oct);end.

The commands to compile and hostname% cc -c SimVal.c hostname% pc SimVal.o SimValMain.p

execute SimVal.c and SimValMain.p hostname% a.out args=111111
Function Return Values
Function return values match types in the same manner as with parameters, and they pass in much the same
way. See "Variable Parameters" on page 98. The following example shows how to pass simple types.
The C function, RetReal.c double RetReal(double *x) return(*x + 1.0);}

The Pascal main program, program RetRealMain;

RetRealMain.p var r, s: real;
function RetReal(var x: real): real; external c;
begin r := 2.0; s := RetReal(r); writeln(r: 4: 1, s: 4: 1) end.

The commands to compile and hostname% cc -c RetReal.c hostname% pc RetReal.o RetRealMain.p

execute RetReal.c and RetRealMain.p hostname% a.out 2.0 3.0

Parameters That Are Pointers to Procedures

Pascal has a special type that is a pointer to a procedure. A variable of this type can be used as a parameter, as
follows:
The C function, ProcPar.c #include <string.h> void proc_c (void (*p)()) /* a pointer to procedure argument
*/
{ char *s ; s = "Called from C";
(*p)( s, strlen(s)); /* Call the Pascal routine */ }
The Pascal main program, program ProcParMain;
ProcParMain.p, which calls the C type { Declare a procedure pointer type. }
procedure, proc_c, passing it the address proc_ptr = ^procedure(var s: string; i: integer);
of the Pascal procedure, proc_pas. The {Declare an external C procedure which takes a procedure
C procedure assigns a value to the string argument.}
s, and calls the procedure whose pointer procedure proc_c(p: proc_ptr); external c;
it just received. Then the Pascal procedure proc_pas(var cstr: string; strlen: integer);
procedure, proc_pas, writes a literal var i: integer;
constant and the string it just received. begin write('Hello from PROC_PASCAL: ');
for i := 1 to strlen do write(cstr[i])
writeln; end; { proc_pas }
begin proc_c(addr(proc_pas)) end. { ProcParMain }

The commands to compile and execute hostname% cc -c ProcPar.c

ProcPar.c and ProcParMain.p hostname% pc ProcPar.o ProcParMain.p
hostname% a.out Hello from PROC_PASCAL: Called from C

Procedures and Functions as Parameters

It is probably clearer to pass a pointer to a procedure than to pass the procedure name itself. See "Procedure
Calls: Pascal-C" on page 121.
A procedure or function passed as an argument is associated with a static link to its lexical parent's activation
record. When an outer block procedure or function is passed as an argument, Pascal passes a null pointer in
the position normally occupied by the passed routine's static link. So that procedures and functions can be
passed to other languages as arguments, the static links for all procedure or function arguments are placed
after the end of the conformant array bounds pairs, if any.
Routines in other languages can be passed to Pascal; a dummy argument must be passed in the position
normally occupied by the passed routine's static link. If the passed routine is not a Pascal routine, the
argument is used only as a placeholder.

Global Variables in C and Pascal

If the types are compatible, a global variable can be shared between C and Pascal.
An example:
The Pascal procedure, GloVar.p Var Year: integer; procedure GloVar; begin
Year := 2001 end; { GloVar }

The C main program, #include <stdio.h> extern void GloVar(); int Year;
GloVarMain.c int main(void) { Year = 2042; GloVar(); printf( " %d \n", Year ) ; }

The commands to compile and execute GloVar.p and hostname% pc -c GloVar.p

GloVarMain.c without -xl. With -xl, the Pascal integer must hostname% cc GloVar.o GloVarMain.c
be paired with a C short int and declared public since the hostname% a.out
default visibility is private. 2001

File-Passing Between Pascal and C

You can pass a file pointer from Pascal to C, then have C do the I/O, as in:
The C procedure, #include <stdio.h> void UseFilePtr (FILE *ptr)
UseFilePtr.c { { /* Write to the file: */ fprintf( ptr, "[1] Passing the file descriptor \n") ;
fprintf( ptr, "[2] and writing information \n") ; fprintf( ptr, "[3] to a file \n");}
The Pascal main program, program UseFilePtrMain;
UseFilePtrMain.p var f: text; cfile: univ_ptr;
procedure UseFilePtr(cf: univ_ptr); external c;
begin rewrite(f, 'myfile.data'); { Make the file. } cfile := getfile(f);
{ Get a file pointer. } UseFilePtr(cfile); { Call the C function. }end.

The commands to compile and hostname% cc -c UseFilePtr.c hostname% pc UseFilePtr.o UseFilePtrMain.p

execute UseFilePtc.c and hostname% a.out hostname% cat myfile.data
UseFilePtrMain.p [1] Passing the file descriptor [2] and writing information [3] to a file

Hello, we meet again! Nice to see ya! I know you're eager to finish this last chapter of the first lesson. It's still
about the file. It is similar to the previous chapter. If you don't understand chapter 13, you'd better not carry
on, but re-learn it instead. This time, we will discuss :
1. Typed files
2. Untyped files
3. File commands
4. Directory commands
There are two kinds of binary files :
1. Typed files
2. Untyped files
Typed file means that the file has a uniform format throughout its contents. This kind of file includes
databases, because all of them contains the data records. Simply said, file of records. Untyped file means that
the file contains no uniform data. Although you may have seen records stored in this kind of file, that file
contains some additional information that may be different record structure. Simply said, file with no distinct
records.
First, we discuss typed files. Suppose you define a record like this :
type
Temployee = record name : string[20]; address : string[40];
phone : string[15]; age : byte; salary : longint; end;
Typed file of THAT record is defined like this :
Var F : file of Temployee;
The steps of using typed file is just the same as using text file.
1. You associate it with file name using assign.
2. Open it, using reset, OR Create it, using rewrite.
3. Use it.
Writeln in text file MUST BE CHANGED into Write and Readln with Read respectively.
4. Close it using close.
All error handling and IOResult use is all the same, so that I don't have to re-mention it all over again.
The difference is : If you open typed file with reset it doesn't mean that you can only read it (just in the text
files), but you may write on it and modify it. The command rewrite is still the same, create a new one,
discarding the file previous contents. Then, look at this example :

{ A crude database recording }

uses crt;
type Temployee = record
name : string[20]; address : string[40]; phone : string[15]; age : byte; salary : longint; end;
var F : file of Temployee; c : char; r : Temployee; s : string;
begin clrscr; write('Input file name to record databases : '); readln(s);
assign(F,s); { Associate it } rewrite(F); { Create it }
repeat clrscr;
write('Name = '); readln(r.name); { Input data } write('Address = '); readln(r.address);
 write('Phone = '); readln(r.phone); write('Age = '); readln(r.age);
write('Salary = '); readln(r.salary); write(F,r); { Write data to file }
write('Input data again (Y/N) ?');
repeat c:=upcase(readkey); { Ask user : Input again or not }
until c in ['Y','N'];
writeln(c);
until c='N'; close(F);
end.

Easy, right ? After creating database, display it. Modify the above program to read the file contents. This is
the hint :
1. Change rewrite to reset.
2. After the second clrscr (inside repeat..until block), add : read(F,r);
3. Remove the line "write(F,r)"
That's all. You may alter the displayed message to the appropriate one. Run it and see how it's done. Good !
You've done it !
Now, it's time to understand file pointer. In order to know the current position of the file, Pascal use a file
pointer. It's simply points to the next record or byte to read. To move the file pointer, use seek :
seek(F,recordno);
The recordno simply said the record number. If you want to read the tenth record of the file at any instant, use
this :
seek(F,9); { Data record number started from 0 }
read(F,r); { r is the record variable }
You may conclude that it is easy to access the records. Say the record number, seek it, and read it. Any record
number in range could be accessed. In range means not exceeding the maximum number of record inside that
file. Therefore, it is called Random File Access.
In the other hand, text files could not behave like that. So that it requires to be handled sequentially.
Therefore, there comes the jargon Sequential File Access.
Append DOES NOT work in typed files or untyped files. It is specially designed for text files. Then how can
we append data to typed files ? Easy. Follow these steps :
1. Open the file with reset.
2. Move the file pointer after the last record using seek.
Reset causes file opened but the file pointer points to the first record. How can we know the number of
records that is stored inside a file ? Number of records can be calculated as follows :
totalrecord := filesize(f);
Here is an example of a 'crude' database. It creates a new database if it is not exist, otherwise it appends data.

{ A crude database recording }

uses crt;
type Temployee = record
name : string[20]; address : string[40]; phone : string[15];
age : byte; salary : longint; end;
var F : file of Temployee; c : char; r : Temployee; s : string; n : integer;
begin clrscr; write('Input file name to record databases : '); readln(s);
assign(F,s); { Associate it } {$I-} reset(F); { First, open it } {$I+}
n:=IOResult; if n<>0 then { If it's doesn't exist then }
begin {$I-} rewrite(F); { Create it }
{$I+} n:=IOResult; if n<>0 then begin writeln('Error creating file !'); halt; end;
end
else { If it exists then }
seek(F,filesize(F)); { Move file pointer to the last record }

repeat
: : : { All remains the same } : :
Now, how can we delete a data ? The only routine that Pascal provides is Truncate. It deletes all data starting
from where file pointer points to the end of file. You may wonder how to delete a single data record. This is
how : Suppose the record number you want to delete is stored in n.
for i:=n to totalrecord-1 do
begin seek(f,i); read(f,r); seek(f,i-1); write(f,r); end;
seek(f,totalrecord-1); truncate(f); dec(totalrecord);
Yes, you move the next record to the deleted record. The second next to the next and so on until the end of
data. After that, you can safely truncate the last record, since the last record is already stored in record
number totalrecord-1 and the last record would be a mere duplicate. Last step you must make is that you must
adjust the totalrecord to comply with present situation (after deletion).
Easy, right ? Oh, yes ! I forgot to mention : Flush cannot be applied to binary files. It's just for text files.
It is unpractical to always having file pointer tracked. You can obtain the file pointer position by using filepos
:
n:=filepos(F);
N will hold the current file position (the record number).
That's all about typed files. You may want to see this program for better details. Run it and learn how it
works.
{ A crude database recording }
uses crt;
type Temployee = record
name : string[20]; address : string[40]; phone : string[15];
age : byte; salary : longint; end;
var F : file of Temployee; c : char; r : Temployee; s : string; n : integer;
begin clrscr; write('Input file name to record databases : '); readln(s);
assign(F,s); { Associate it } {$I-} reset(F); { First, open it } {$I+}
n:=IOResult; if n<>0 then { If it's doesn't exist then }
begin {$I-} rewrite(F); { Create it }
{$I+} n:=IOResult; if n<>0 then
begin writeln('Error creating file !'); halt; end;
end
else begin { If it exists then } n:=filesize(F); { Calculate total record }
seek(F,n); { Move file pointer PAST the last record }
end;
repeat clrscr; writeln('File position : ',filepos(f));
write('Name = '); readln(r.name); { Input data }
write('Address = '); readln(r.address); write('Phone = '); readln(r.phone);
write('Age = '); readln(r.age); write('Salary = '); readln(r.salary);
write(F,r); { Write data to file } write('Input data again (Y/N) ?');
repeat
c:=upcase(readkey); { Ask user : Input again or not }
until c in ['Y','N']; writeln(c);
until c='N'; close(F); end.

Before you fully understand typed files, DO NOT CONTINUE to untyped one, it will just make you more
confused.
Text files are usually used for INI files or setting files. Or, if your game needs special setup, you can use this
skill to modify AUTOEXEC.BAT, CONFIG.SYS or even *.INI in WINDOWS directory. Typed files are
usually done for recording high scores of your game, while untyped ones are for reading your game data :
pictures, sounds, etc. Serious applications make an extensive use of file. Databases usually use typed files.
Text files are used for making memos. Untyped ones is for reading pictures and sounds, for perhaps, you
want to make presentations or just displaying the company logo.
Untyped Files
Now, we're going to discuss the untyped files. The basics is all the same. Imagine you have a typed file, but
the record is one byte long. The declaration is a bit different from typed files :
var
F : file;
This declare F as untyped files. Assign and Close are still the same, but Reset and Rewrite are a little bit
different. Instead of just passing the F, you have to specify the "record size" for that file like this:
reset(F,1);
Rewrite is just similar. What is the "record size" for? It is used to specify the number of bytes each time you
read or write from that file. Let me explain that a bit later.
The command write and read cannot be used in untyped file. Use blockwrite and blockread instead. Here is
the syntax :
blockread (f,buffer,count,actual);
blockwrite(f,buffer,count,actual);

f is the file variable.

buffer is your own buffer, not Pascal's.
count is the number of records you want to read/write.
actual is the number of bytes that has been read/written.
The parameter count in both commands will determine how many records (not bytes) you will read or write.
Effectively, the total number of bytes (read or written) is count*record_size bytes. This is why you must set
the record size in the first place because Pascal doesn't know that.
In untyped files, you must prepare a buffer. A buffer can be records, arrays, or even pointers. Usually,
programmers use array instead of records. But, usually, if the file has certain structure, like graphic formats,
programmers use records. Let's first concern about array as the buffer. Suppose I declared buffer as array of
bytes :
Var buffer : array[1..2048] of byte; count, actual : word; f : file;
Since we have an array of bytes for the buffer, we have to set the record size to 1. We can simply consider
that each array element counts as one byte. If you have array of integers or words, you may as well set the
record size into 2.
Reading is done by this :
: reset(f,1); { Setting record size to 1 } : :
count:=sizeof(buffer); blockread(f,buffer,count,actual);
This example will read 1*2048 bytes because record size is set to 1 and the size of the buffer (specified in
count) variable is 2048.
Variable actual holds the number of bytes that is actually read from the disk. Likewise, writing to disk is done
through blockwrite :
: rewrite(f,1); : : count:=sizeof(buffer); blockwrite(f,buffer,count,actual);
You can even specify the number of bytes you want to read. Suppose you want to read just 512 bytes from a
file : blockread(f,buffer,512,actual);
Writing 512 bytes is just similar to reading. Now, how if the buffer is a record ? Suppose I declare the
record : type THeader = record
tag : string[4]; width, depth : word; bitperpixel : byte; end;
var hdr : THeader;
That kind of header is one example of reading picture file header. Usually, after reset, programmer has to
read the header to check validity. Reading the header can be done by blockread :
blockread(f,hdr,sizeof(hdr),actual);
The operator sizeof returns the number of bytes occupied by the operand or parameter automatically (so that
you don't have to count it manually). If the file is good, the header is fully read. That can be checked by this :
if actual=sizeof(header) then { The file has a good header }
:
:
But .... wait ! I saw somebody using untyped file with write and read. Well, that kind of person treating typed
file as untyped one. That causes a LOT of pain. That's why I'm not going to teach it. But, if you insist, you
can write me.
That's all about untyped files.
File Commands
Now, we're going to discuss other file commands :
1. Rename
2. Erase
3. Getenv
4. FSearch, FExpand and FSplit
5. FindFirst and FindNext
6. UnpackTime and PackTime
7. GetFTime and SetFTime
8. GetFAttr and SetFAttr
9. DiskFree and DiskSize
Number 3 through 9 need DOS unit
Rename, just like its name is to rename files. You must assign the old file name to a file variable, (the file is
not necessarily be opened) then use rename :
assign(f,oldname);
rename(f,newname);
Erase, is to erase files. You assign the file you want to erase, then erase it. You may NOT open the file. If
you've already opened it, close it first before erasing !
assign(f,filename);
erase(f);
You have used the crt unit so long and nothing else. Now, it's time to corporate DOS unit, so you'll probably
do this :
uses crt, dos;
You need no files to add as crt and dos are both in SYSTEM.TPL. SYSTEM.TPL is always loaded when
Pascal starts. Why do we need DOS unit ? Well many of file routines (that has been mentioned as number 3
thru 9) is in DOS unit. Also, interrupt handling, system time and other handy things reside in it. Let's cover
the file-handling routines.
Getenv
Getenv fetches the environment string of DOS. Go to command prompt of DOS, then type SET then press
Enter. DOS displays all the environment string, such as PATH, PROMPT, etc. This is example of how to get
the PATH contents (s is a string) :
s:=Getenv('PATH');
Getting PROMPT is similar : s:=Getenv('PROMPT');
FSearch
FSearch do searching files in a specified directory. Suppose you want to search for FORMAT.COM in DOS
and WINDOWS directory :

uses dos;
var s : string;
begin s:=FSearch('FORMAT.COM','C:\DOS;C:\WINDOWS');
if s='' then writeln('FORMAT.COM not found')
else writeln('FORMAT.COM found in ',s);
end.

When found, s returns complete path and filename, otherwise empty. You may extend the directory list (the
second parameter of FSearch) using semicolon such as :
... FSearch( ... , 'C:\DOS;C:\WINDOWS;C:\SCAN;C:\TOOL');
You may wonder that you can even search a file in the PATH environment variable. Yes, you COULD ! Do it
like this :
... FSearch( ... , getenv('PATH'));
FExpand
FExpand expands a simple file name into a full name (drive, full directory, and the file name itself). It is
especially useful when user inputs a relative directory, like this (s is a string) :
s:=FExpand('..\README');
S will be like this (for example) : 'C:\PASCAL\LESSON.1\README'
FSplit
It is just contrary to FExpand, splits a fully qualified name into directory, file name, and extension. Example :
Var s : string; d : dirstr; n : namestr; e : extstr; : : s:='C:\WINDOWS\WIN.INI';
fsplit(s,d,n,e); { d = 'C:\WINDOWS\', n = 'WIN', e = '.INI' }
Look at this program for better details.
uses dos;
var s : string; d : dirstr; n : namestr; e : extstr;
begin s:=FSearch('FORMAT.COM',getenv('PATH'));
if s='' then begin writeln('FORMAT.COM not found'); halt; end;
writeln('FORMAT.COM found in ',s); fsplit(s,d,n,e); writeln(d,' ',n,' ',e);
end.

FindFirst and FindNext

FindFirst and FindNext is used just like dir command in DOS. It uses the TSearchRec record tag. The syntax
of them :
findfirst(qualifier,attribute,searchrec);
findnext(searchrec);
qualifier can be '*.*' or '*.PAS' or any wildcard, just like the parameter in dir command. The attribute can be :
Attribute If you want to search for
ReadOnly read-only files
Hidden hidden files
SysFile system files
VolumeID disk volume label
Directory directories
Archive archive files
AnyFile any files
TSearchrec is defined as follows :
type
TSearchRec = record
Fill: array[1..21] of Byte; Attr: Byte; Time: Longint;
Size: Longint; Name: array[0..12] of Char; end;
Suppose I have the variable s declared as TSearchrec, and I'm using the findfirst to search '*.EXE' with any
attribute :
findfirst('*.EXE',AnyFile,s);
s.name holds the name s.size holds the size in bytes
s.time holds the date and time when the file is created s.attr holds the attribute
You should never touch the fill field. It is classified. I personally don't know what does it for, only folks in
Microsoft did, perhaps.
FindFirst is used to INITIATE the search. FindNext is used to do the subsequent search. You may repeat
FindNext as many as you want to retrieve all the filenames you desire. If there are no file names left, the
variable DosError is set to 18. For safety reason, you could repeat findnext while the DosError remains 0,
like this :

uses dos;
var s : TSearchrec; q : string;
begin write ('Enter qualifier to search = '); readln(q);
findfirst(q,AnyFile,s); while DosError=0 do begin
writeln(s.name); findnext(s); end;
end.

You may wonder how file date and time can be depicted as long integer. DOS is actually packed them. So,
after the call of findfirst and findnext, the s.time is in a packed form. How to unpack them ? Use unpacktime.
The syntax is :
unpacktime(s.time,dt);
s.time is the packed form, dt is the unpacked form. Dt is a record of DateTime, defined as follows :
type DateTime = record
Year,Month,Day,Hour, Min,Sec: Word;
end;
So, after unpacktime : dt.year holds the year dt.month holds the month, and so on (respectively).
Look at the following example, that display the file with its size and its date/time :

uses dos;
var dt : DateTime s : TSearchrec; q : string;
begin write ('Enter qualifier to search = '); readln(q);
findfirst(q,AnyFile,s); while DosError=0 do begin
unpacktime(s.time,dt); write (s.name :15, s.size:8,' ');
write (dt.month:2,'/',dt.day:2,'/',dt.year:4,' '); writeln(dt.hour :2,':',dt.min:2,':',dt.sec);
findnext(s); end;
end.

How about the attribute ? Is it packed too ? Not exactly. To detect each attribute you must do this :
if s.attr and ReadOnly = ReadOnly then write('Read only ');
if s.attr and Hidden = Hidden then write('Hidden ');
if s.attr and SysFile = SysFile then write('System ');
and so on.
You can do that detection routine to any attribute name shown above EXCEPT for AnyFile.
How can I search for ReadOnly or Hidden files only, but not archives and system files ? Do this at findfirst :
findfirst(q,ReadOnly or Hidden,s);
You may combine the attribute with OR, not and. Also, you may not combine AnyFile with others, because
the combination would not take effect.
Actually, processing attributes is a bit-wise operation, which I haven't taught you yet (later in lesson 2). But I
made it as simple as possible, so that you can understand.
PackTime, GetFTime, and SetFTime
UnpackTime has been described above. Now you may wonder how to PackTime. PackTime is used like this :
packtime(dt,n);
dt is DateTime variable (unpacked time), n is a long integer holds packed time. What does it for ? It's for
setting the file time by setftime :
setftime(f,n);
F is any file variable (text, typed or untyped), n is the packed form of date/time. Don't forget to assign f with
a file name first.
GetFTime returns the date/time of the file :
getftime(f,n);
F and n is the same as in setftime. After getftime, don't forget to unpack it first with unpacktime.
GetFAttr and SetFAttr
Likewise, you can set or get a file's attribute using setfattr and getfattr. Suppose f is any file variable (text,
typed, or untyped), and n is a word variable :
getfattr(f,n);
This will get the file attribute in n. Don't forget to associate f with a file name first. Detecting the attribute is
just the same as shown in findfirst and findnext. Setting attribute takes the same syntax. Suppose you want to
set the file as hidden and read-only :
n:=ReadOnly or Hidden;
setfattr(f,n);
DiskFree and DiskSize
DiskFree returns the number of bytes free in a specified drive. DiskSize returns the disk total size of a drive.
Both take a byte as parameter :
n:=DiskFree(d); { d is a byte, n is a longint }
n:=DiskSize(d);
d is the drive qualifier. D = 0 means that you want to search info in the current drive, d = 1 for drive A, d = 2
for drive B, d = 3 for drive C and so on. If n = -1, means that you have input an invalid drive.
Directory commands
We'll discuss these directory commands :
1. Chdir
2. Mkdir
3. Rmdir
4. Getdir
Chdir, Mkdir, and Rmdir is just the same as DOS command cd, md, and rd. All of them takes a string
parameter :
chdir('\DOS'); { same as cd\dos } mkdir('\TEMP'); { same as md\temp }
rmdir('\TEMP'); { same as rd\temp }
Getdir is used to get the directory, just like you type cd in DOS prompt, then press enter. The difference is
that getdir can be used to get the current directory of any drive, not just current drive as cd does. Getdir
syntax is like this :
getdir(d,s); { d is a byte, s is a string }
d is the drive qualifier. D = 0 means that you want to search info in the current drive, d = 1 for drive A, d = 2
for drive B, d = 3 for drive C and so on. S will return the directory. If d is invalid s just say the root directory.
Suppose d:=23 (drive W) is invalid, s will return 'W:\' as it were a root directory.
Phew ! A tedious lesson has finished. That's all folks ! If you DO understand what I've taught so long, then
congratulations ! If you might want to continue the second lesson, wait for another issue ! So long !