Chapter 4
				 - continued -
			     - Part 2 of 2 parts -
                                    of the
                            Turbo Pascal Reference

			  The System Library Reference


This chapter is part of the Turbo Pascal Reference electronic freeware book (C)
Copyright 1992 by Ed Mitchell.  This freeware book contains supplementary
material to Borland Pascal Developer's Guide, published by Que Corporation,
1992.  However, Que Corporation has no affiliation with nor responsibility for
the content of this free book.  Please see Chapter 1 of the Turbo Pascal
Reference for important information about your right to distribute and use this
material freely.  If you find this material of use, I would appreciate your
purchase of one my books, such as the Borland Pascal Developer's Guide or
Secrets of the Borland C++ Masters, Sams Books, 1992.  Thank you.


IOResult function
----------------------------------------------------------------
Declaration:  
     function IOResult : Word;

Example:
     repeat
       Writeln( 'Enter name of file to open:' );
       Readln( FileName );
       {$I-}
       Assign( F, FileName );
       Reset( F );
       {$I+} { I usually leave $I- set for the duration of my
              programs }
       ErrorCode := IOResult;
       if ErrorCode <> 0   then
       begin
          Writeln('Unable to open file ', FileName );
       end;
     until  ErrorCode = 0;

Purpose:
     IOResult returns the error code of the last input/output operation.  To
use IOResult, you must disable automatic IOResult checking by writing {$I-}
before performing an I/O operation.  Then, after each I/O operation you can
manually check the result code returned by IOResult, and take appropriate
action.
     Normally, you should copy the value of IOResult to a temporary variable,
like this:

     ErrorCode := IOResult;
     If ErrorCode <> 0 then {take action}

While you can check IOResult directly, IOResult resets the last error condition
to zero when called.  This means if you check the value of  IOResult in an
if-then statement, and then in the process of handling the error you again
reference IOResult, the value returned will now be 0! Trying to track this type
of problem is a bit maddening until you realize that IOResult returns the
current error code and then resets itself to zero.



Keep procedure
----------------------------------------------------------------
Declaration:  
     procedure Keep(ExitCode: Word);

Unit:       Dos

Purpose:
     Terminates a program but leaves the program resident in memory.  Such a
program is a TSR program, which must latch itself onto a interrupt to be
"revivied", and which must  be aware of the other considerations that apply to
a TSR program (such as avoiding DOS reentrant code and so on).  The ExitCode
parameter value is passed to the Halt standard procedure for indicating an
error condition.  DO NOT use this procedure unless you know what you are doing!
 You can learn how to write a TSR program entirely in Turbo Pascal by reading
Chapter 9, "Special Programming Techniques:  8087 Usage, Interrupts and TSRs",
in the Borland Pascal Developer's Guide, from Que Books.



KeyPressed function
----------------------------------------------------------------
Declaration:  
     function KeyPressed : Boolean;

Unit:       Crt

Example:

     repeat
       { Do any operations repeatedly here }
     until KeyPressed;

Purpose:
     KeyPressed returns True if a keyboard character has been typed, False
otherwise.  KeyPressed is useful for executing a section of code until a
character is typed at the keyboard.  KeyPressed only tells you that 1 or more
characters have been typed; you need to call Read, Readln or ReadKey to read
the input.
See ReadKey



Length function
----------------------------------------------------------------
Declaration:  
     function Length ( S : String );

Example:

     Writeln( Length ('ABCDEFGHIJKLMNOPQRSTUVWXYZ') );

outputs,

     26

Purpose:
     Length(S) returns the current value of the length byte of S, which is the
actual length of the string that is stored in S.  SizeOf returns the total
storage occupied by S.  If, for example, S is declared as,
     var
       S : String[80];
then SizeOf(S) returns 81 (80 bytes plus the length byte), and Length(S)
returns a value from 0 to 80, depending upon the current contents of the
string.
See SizeOf



Ln function
----------------------------------------------------------------
Declaration:  
     function Ln( X : Real );

Purpose:
     Returns the natural logarithm of X.  To compute the base 10 logarithm of
an expression, define a new function like this:
     function Log10 ( X : Real ): Real;
     begin
       Log10 := Ln(X) / Ln(10);
     end;
See Exp


Lo function
----------------------------------------------------------------
Declaration:  
     function Lo(X): Byte;

Purpose:
     Returns the low byte of X, where X is either an Integer or Word value. 
Thefunction Hi(X) returns the corresponding high byte.
See Hi, Swap



LowVideo procedure
----------------------------------------------------------------
Declaration:  
     procedure LowVideo;

Unit:       Crt

Purpose:
     Performs the inverse of HighVideo by subtracting 8 from the text
attribute's foreground color, translating color values in the range 8 to 15
down to 0 to 7.  See HighVideo for more details.
See HighVideo, NormVideo, TextColor



Lst file
----------------------------------------------------------------
Declaration:  
     Lst : Text;

Unit:       Printer

Purpose:
     The Printer unit provides a special Lst file identifier for easy printer
output.  All you need do to use Lst, is use the Printer unit in your program or
unit's uses statement.  Thereafter, to send output to the printer, specify Lst
as the file identifier in your Write or Writeln statements.  You do not need to
open or close the printer as this is handled automatically by the Printer
unit's internal initialization section.
See Write, Writeln



Mark procedure
----------------------------------------------------------------
Declaration:  
     procedure Mark (var P: Pointer);

Purpose:
     Copies the current value of the heap pointer to P.  Mark, together with
Release, enable you to dynamically allocate a group of objects, and then
dispose of them in a single call to Release.  See "The Use of Mark and Release
Procedures" in Chapter 3, for full details.



MaxAvail function
----------------------------------------------------------------
Declaration:  
     function MaxAvail : Longint;

Purpose:
     During program execution, the heap memory area may become fragmented with
blocks allocated here and there.  By calling MaxAvail you can determine the
size, in bytes, of the largest free block available for allocation.  A related
function, MemAvail, returns the total number of free bytes available in the
heap.



MemAvail function
----------------------------------------------------------------
Declaration:  
     function MemAvail : Longint;

Purpose:
     Returns the total number of free bytes in the heap.  Note that you may not
be able to allocate a dynamic variable as large as MemAvail because the heap
may be fragmented.  You can determine the largest possible free block by
calling MaxAvail.
     A common use for MemAvail is to periodically check your memory allocation
and deallocation.  If you find that MemAvail is constantly increasing durng
program execution, you may be dynamically allocating a variable and never
disposing of the memory block.  Eventually, your program will run out of memory
and be halted.
     You may also wish to place MemAvail checks at key points within your
program.  For example, before calling a procedure that makes a large number of
dynamic memory allocations, save the contents of MemAvail in another variable. 
Then, upon return from the procedure, compare MemAvail to the saved value.  If
they are not the same, this is a sign that somewhere within the called
procedure, or perhaps a procedure called by it, memory is not being properly
disposed.



Mkdir procedure
----------------------------------------------------------------
Declaration:  
     procedure MkDir(S: String);

Example:  
var
     DirName : String;
     ErrorCode : Integer;
begin
     repeat
       Write('Enter name of directory to make: ');
       Readln( DirName );
       if  DirName <> ''  then
       begin
          {$I-}
          MkDir ( DirName );
          ErrorCode := IOResult;
          if ErrorCode <> 0 then
            Writeln('Error creating directory ', DirName )
          else
            Writeln('Directory ', DirName, 
              ' has been created.');
       end;
     until DirName = '';
end.

Purpose:
     To create or make a subdirectory, call MkDir, setting the parameter S to
the name of the directory to create.  S may specify a full path, such as
'C:\TP\UTILS\GRAPHICS' or a simple directory name such as 'GRAPHICS' which
creates subdirectory 'GRAPHICS' in the current subdirectory.  
     The directory name specified may not be the name of an existing file.  You
can check for errors by setting {$I-} and examining the value returned by
IOResult.



Move procedure
----------------------------------------------------------------
Declaration:  
     procedure Move (var Source, Dest; Count : Word);

Example:
   1  { DEMOMOVE.PAS }
   2  program MoveDemo;
   3  { Demonstrates how Move procedure can be used to move large numbers
   4  of bytes around inside an array.  This demo stores lines of text
   5  into a large TextArray.  New lines are added by calling AddLine, which
   6  calls MakeAHole to slide the existing data in the TextArray to make
   7  room for new text.  AddLine then uses Move to copy the string data
   8  into TextArray.  Data is fetched using the procedure GetLine.
   9    The variable CurSize keeps track of how many bytes of TextArray are
  10  currently in use.  CurLocation keeps track of the current position; if we
  11  were writing a text editor based on this approach, CurLocation would
index
  12  into TextArray where the editor's cursor appears on the screen.  In other
  13  word, the next new line that we type would be inserted at CurLocation.
  14  }
  15  
  16  const
  17    MaxSize = 32000;
  18    MARKER = 13;
  19  
  20  var
  21    TextArray : Array[0..MaxSize] of Char;
  22    CurSize : Word;
  23    CurLocation : Word;
  24  
  25  
  26  procedure MakeAHole ( Address : Word; Size : Word );
  27  { Slides data in the TextArray sideways, leaving a "hole" to fill
  28  with new data }
  29  begin
  30    Move( TextArray[Address], TextArray[Address+Size], CurSize - Address +
1 );
  31  end;
  32  
  33  
  34  
  35  procedure AddLine ( var Address : Word; S : String );
  36  { Adds the contents of S to the TextArray, starting at location Address }
  37  var
  38    NumBytes : Integer;
  39  begin
  40    NumBytes := Length(S) + 1;
  41    S[0] := Chr(MARKER);
  42    MakeAHole ( Address, NumBytes );
  43    Move( S[0], TextArray[Address], NumBytes );
  44    CurSize := CurSize + NumBytes;
  45    Address := Address + NumBytes;
  46  end;
  47  
  48  
  49  
  50  procedure GetLine ( Address : Word; var S : String );
  51  { On entry, Address points to a MARKER character in the text array. 
GetLine
  52  copies the text between this marker and the next marker (or end of array)
  53  to S. }
  54  var
  55    TempAddr : Word;
  56  begin
  57    TempAddr := Address + 1;
  58    while (TempAddr <= CurSize) and
  59          (TextArray[TempAddr] <> Chr(MARKER)) do
  60      Inc(TempAddr);
  61    Move ( TextArray[Address+1], S[1], TempAddr - Address );
  62    S[0] := chr( TempAddr - Address );
  63  end;
  64  
  65  
  66  procedure MoveAhead (var Address : Word );
  67  { Where address points to some location in TextArray, MoveAhead positions
  68  Address to the index location of the next Marker character. }
  69  begin
  70    if  TextArray[Address] = chr(MARKER)  then
  71      Inc(Address);
  72    while (Address <= CurSize) and
  73          (TextArray[Address] <> Chr(MARKER)) do
  74      Inc(Address);
  75  end;
  76  
  77  
  78  var
  79    I : Integer;
  80    ALine : String;
  81  
  82  begin
  83    CurSize := 0;
  84    CurLocation := 0;
  85  
  86    { Add some sample data into TextArray }
  87    AddLine ( CurLocation, 'Line 1 I''m the first line of data');
  88    AddLine ( CurLocation, 'Line 2 and I''m the second!');
  89    AddLine ( CurLocation, 'Line 3 The quick brown fox jumped over the');
  90    AddLine ( CurLocation, 'Line 4 lazy brown dog and fell and hurt
herself.');
  91  
  92    CurLocation := 0;
  93    MoveAhead ( CurLocation );
  94    AddLine ( CurLocation, 'One more line!!!!!');
  95  
  96    CurLocation := 0;
  97    for I := 1 to 5 do
  98    begin
  99      GetLine ( CurLocation, ALine );
 100      Writeln( ALine );
 101      MoveAhead ( CurLocation );
 102    end;
 103  
 104    readln;
 105  end.

Purpose:
     Move copies a block of bytes of Count size from Source to Dest.  Move uses
optimized assembly language to provide extremely fast copies of potentially
large amounts of data and is particular useful for sliding values around inside
large text arrays, such as might be used in a text editor (see example, above).

Important! Range Checking!
     Move does not perform range checking.  You may, for instance, move 30,000
bytes into a 255 character string.  And Move will go ahead and do just that,
writing over portions of your run-time stack, other variables, and generally
anything else that gets in the way.  So be careful!



MsDos procedure
----------------------------------------------------------------
Declaration:  
     procedure MsDos (var Regs: Registers );

Unit:       Dos

Example:
     uses
       Dos;
     var
       Reg : Registers;
     begin
       { Call the DOS Int 21H Function 0EH Select Disk function
       to determine
       the maximum number of drives permitted for the system;
       this
       is equivalent to the number of logical drives on the
       system, or
       the LASTDRIVE value in the CONFIG.SYS file, or a default
       of 5
       if LASTDRIVE (approximately) }
       Reg.AH := $0E;
       Reg.DL := $00;
       MsDos( Reg );
       Writeln( 'Total Drives=', Reg.AL );
     end.

Purpose:
     MsDos access the DOS-functions available through software interrupt 21H. 
While most of the especially useful DOS-level functions are available through
Turbo Pascal library routines such as GetFTime, SetDate, and so on, you can
access some of the other functions using the MsDos procedure call mechanism. 
The variable parameter Regs defines a case variant record structure for
representing the CPU registers, where Register is defined as:

type
     Registers = record
       case Integer of
          0: (AX, BX, CX, DX, BP, SI, DI, DS, ES, Flags: Word);
          1: (Al, AH, BL, BH, CL, CH, DL, DH: Byte);
     end;

Note that the SS and SP registers are not available in this structure and
cannot be used in conjunction with MsDos calls.  For complete details on DOS
function calls, see DOS Programmer's Reference, by Terry Dettmann and Jim Kyle
(Que Books, 1989).



New procedure
----------------------------------------------------------------
Declaration:  
     procedure New( var P: Pointer [, Init: constructor ] );

Example:
     See "The Pointer type" in Chapter 3 of the Turbo Pascal Reference.

Purpose:
     Dynamic variable allocations and dynamic object creations are performed
with New.  New allocates a memory block, from the heap memory area, and sets P
to point to the memory block.  If P points to an object, you can optionally add
the name of the object's constructor method and it will be called immediately
after the memory is allocated.  This provides a short hand method of
simultaneously allocating and initializing the object's fields and virtual
method table.
     The New procedure may fail if there is insufficient memory in the heap to
allocate the new block.  Normally, this results in a run-time error that
terminates the program.  However, by setting the HeapError variable to point to
your own heap error handler, your program can take steps to gracefully recover
from out of memory conditions.  See HeapError in this section, and chapter 3,
"Pointers and Memory Management" for tips on trapping the out of memory error
condition.
     You must make certain that you dispose of allocated memory blocks once
they are no longer needed (see Dispose).  If you fail to dispose of a memory
block it continues to occupy memory space and can contribute to your program's
running of out memory.
     Since a pointer variable "points" to a memory block, you reference the
thing that P points to by typing P^.  This is described in chapter 3.  
See Dispose, FreeMem, GetMem



NormVideo procedure
----------------------------------------------------------------
Declaration:  
     procedure NormVideo;

Unit:       Crt

Purpose:
     Resets the current text attribute to the original text attribute that was
in use when the program was launched.  Specifically, when the program is
started, the value of the text attribute where the cursor is located is saved. 
You can use NormVideo to restore this original text attribute at program exit. 
By doing so, the screen image will be reset to whatever screen attributes were
in effect when the program was started (or specifically, the screen attribute
at the cursor's location when the program was started).
See HighVideo, LowVideo, TextBackground, Textcolor



NoSound procedure
----------------------------------------------------------------
Declaration:  
     procedure NoSound;

Unit:       Crt

Purpose:
     NoSound is used in conjunction with the Sound procedure.  Sound turns on
the speaker and emits a user selected tone.  To turn off the speaker, call
NoSound.
See Sound



Odd function
----------------------------------------------------------------
Declaration:  
     function Odd( X: Longint ): Boolean;

Purpose:
     Odd returns True if its parameter is an odd number, or False if its
parameter is an even number.  Sort of an odd function.



Ofs function
----------------------------------------------------------------
Declaration:  
     function Ofs( X ): Word;

Purpose:
     Where X is any type of variable, Ofs(X) produces the offset portion of the
variable's address.  All 80x86 addresses consist of a segment:offset word pair.
The two values are added together to produce the actual physical memory
address.  To find the corresponding segment value, call Seg(X).
See Addr, Seg



Ord function
----------------------------------------------------------------
Declaration:  
     function Ord( X ): Longint;

Example:
     var
       I : Integer;
       S : String;
     ...
     { Convert all upper case letters to lower case letters }
     for I := 1 to Length(S) do
       if (S[I]>='A') and (S[I]<='Z')  then
          S[I] := Chr( Ord(S[I]) + 32 );

Purpose:
     Ord translates its parameter X to a Longint value reprsenting X's
ordinality.  For example, Ord('A') returns a value of 65, because A is the
sixty-fifth value in the set of characters (this also coincides with 'A's ASCII
code value).  The sample code shown above uses Ord to convert a character value
to its ASCII code.  Since the capital letter 'A' has an ASCII value of 65, and
the lower case letter 'A' has an ASCII code of 97, adding 32 to the ordinal
value of a character is equivlant to converting the character to lower case. 
Chr converts the integer result back to a character value.
If you have an ordinal type declared as,
     type
       DaysOfTheWeek= (Sun,Mon,Tue,Wed,Thu,Fri,Sat);
     var
       Day : DaysOfTheWeek;

And then the code, 

     Day := Tue;
     Writeln(Ord(Day));

The output will be:
     2
since the ordinal values of DaysOfTheWeek are:
     Sun  0
     Mon  1
     Tue  2
     Wed  3
     Thu  4
     Fri    5
     Sat    6



PackTime procedure
----------------------------------------------------------------
Declaration:  
     procedure PackTime(var DT: DateTime; var Time: Longint);

Unit:       Dos

Example:
     uses dos;
     var
       F : Text;
       LFileInfo : Longint;
       DFileInfo : DateTime; { declared in Dos unit }
     begin
       Assign( F, 'FSEARCH.PAS' );
       Reset( F );
       GetFTime ( F, LFileInfo );
       UnpackTime ( LFileInfo, DFileInfo );
       with DFileInfo do
          Writeln(Month, '-', Day, '-', Year,
            '  ',Hour:2,':',Min:2,':',Sec:2);

       { Change file's date to 1992 }
       DFileInfo.Year := 1992;
       PackTime ( DFileInfo, LFileInfo );
       SetFTime ( F, LFileInfo );
       Readln;
      end.

Purpose:
     PackTime converts a date and time record DT, containing individual values
for Year, Month, Day, Hour, Min and Sec, into a 4 byte Longint reprsentation
and returns the result in parameter variable Time.
     PackTime is used with UnpackTime, for encoding and decoding the date and
time data returned or set by GetFTime and SetFTime, respectively. You can write
a program, such as the sample program above, to pre- or post-date files, as
needed.  When that important project needs to be done on a Friday and you find
yourself feverishly coding through late weekend hours, you still might be able
to convince your boss that you really did finish the job on Friday!
See GetFTime, SetFTime, UnpackTime



ParamCount function
----------------------------------------------------------------
Declaration:  
     function ParamCount : Word;

Purpose:
     When you start up a program from the DOS command line, you can add a
number of comand line parameters that are accessible from your program (see
ParamStr).  ParamCount returns the total number of command line parameters
entered on the command line.



ParamStr function
----------------------------------------------------------------
Declaration:  
     function ParamStr( Index: Word ) : String;

Purpose:
     When you start up a program from the DOS command line, you can add a
number of comand line parameters that are accessible from your program.   These
command line parameters might be filenames, starting directories or any other
values that your application requires.  For example, the command line compiler
TPC has a large variety of command line options that may be specified when
starting the compiler.  Inside your application, you can fetch each of the
command line parameters by calling ParamStr and setting Index to choose a
particular parameter.
     Normally, you should restrict the value of Index to the range of 1 to
ParamCount, where ParamCount returns the total number of command line
parameters.  Invalid values for Index result in a null string being returned by
ParamStr.  On DOS 3.0 and later, ParamStr(0) provides the name of the program
that is currently executing.  ParamStr(0) is especially useful when an
application must find the subdirectory from where it was launched.
See ParamCount



Pi function
----------------------------------------------------------------
Declaration:  function Pi : Real;

Purpose:
     Pi returns the constant value 3.1415926535897932385, where the exact
precision depends on the math emulation or math processor mode that is in use.



Port/PortW pseudo arrays
----------------------------------------------------------------
Declaration:  
When used for output:
     Port[ Index ] := <byte valued expression>
     PortW[ Index ] := <word valued expression>
       
When used for input:
     <Result> := Port[ Index ];
     <Result> := PortW[ Index ];

Example:
     { Assign the value of Databyte to I/O address $3F8 }
     Port[$3F8] := Databyte;

Purpose:
     Port and PortW act like arrays whose index corresponds to a hardware I/O
port.  The index value ranges from 0 up to 65,535.  By using Port for byte
values and PortW for word values, these pseudo-array variables are used to
write high level language implementations of hardware device drivers, such as
serial port access routines. PortW is used to access a word value.
     Port and PortW are not really arrays but Pascal level access mechanisms
for the low-level data ports.  As such, Port and PortW should only be used to
the left of an assignment statement or as a component in an expression.  They
cannot be used as variable parameters to any procedure or function.



Pos function
----------------------------------------------------------------
Declaration:  
     function Pos ( Substr, S: String ): Byte;

Example:

     { This examples uses the Pos() function to implement
       a string replacement routine }

     function Replace ( SearchFor, ReplaceWith, 
            InThis : String ): String;
     { Finds every occurrence of SearchFor within InThis, and
     changes each occurrence to the value of ReplaceWith.
     This function solves the problem recursively.  Once it finds
     an occurrence of SearchFor, it makes the first replacement
     and then calls itself on the remainder of the string.}
     var
       Index : Integer;
     begin
       Index := Pos (SearchFor, InThis);
       if  Index > 0  then
          Replace := Copy ( InThis, 1, 
                 Pos(Searchfor, InThis) - 1) +
                 ReplaceWith +
                 Replace( SearchFor, ReplaceWith,
                 Copy( InThis, Index + Length(Searchfor),255 ));
       else
          Replace := InThis;
     end;

Example Usage of Replace:

     Writeln(
     Replace('e', '*',
        'The quick brown fox jumped over the lazy dog.');

outputs,

     Th* quick brown fox jump*d ov*r the lazy dog.

Purpose:
     When you need to search for a substring within another string, use the Pos
function.  Pos scans through string S searching for an occurrence of Substr. 
Iffound, Pos returns the character index in S where Substr begins. If not
found, Pos returns 0.
     The example function Replace, above, is a handy routine to perform a "find
and replace" operation on strings.  Replace uses Pos to find the first
occurence of the Searchfor string, replaces it with ReplaceWith, and then calls
Replace recursively to change the remainder of the string.



Pred function
----------------------------------------------------------------
Declaration:  
     function Pred( X ) : <same type as X>;

Example:
     type
       DaysOfTheWeek= (Sun,Mon,Tue,Wed,Thu,Fri,Sat);
     var
       Day : DaysOfTheWeek;
     ...
     Day := Tue;
     Day := Pred( Day );      { Sets Day to Mon }

Purpose:
     Pred(X) sets X to the predecessor value in the ordinal valued set from
which X is derived.   In the case of an integer or word value, Pred is
equivalent to the Dec function.  Some additional examples:
     Pred(10) = 9
     Pred('B') = 'A'
See Dec, Inc, Succ



PrefixSeg variable
----------------------------------------------------------------
Declaration:  
     PrefixSeg: Word = 0;

Purpose:
     PrefixSeg contains the segment address of the Program Segment Prefix
(PSP).  The PSP contains information about the program loaded in memory,
including the command line that was used to load the program (so the program
can access command line parameters), pointers to the memory blocks used by the
program and other information.  PrefixSeg is used in the example TSR program
shown in Chapter 11 of this book.  For further details about the data contained
in the PSP, consult Appendix F of DOS Programmer's Reference, 2nd edition, Que
Books, 1989.  



Ptr function
----------------------------------------------------------------
Declaration:  
     function Ptr( Seg, Ofs : Word) : Pointer;

Example:
     { Using Ptr to manufacture a pointer into the DOS data area
     to obtain the number of Kbytes of low memory available to
     DOS.  On most systems today, this will always 
     be 640k bytes. }
     var
       DOSMemory : word;
     begin
       DOSMemory:= Word(Ptr( $40, 19 )^);
       Writeln('Low DOS Memory, in k bytes=', DOSMemory);
     end.

Purpose:
     Ptr is used to create a pointer to a specific memory address, specified by
Seg:Ofs.



Random function
----------------------------------------------------------------
Declaration:  
     function Random [ (Range: Word) ] : Word
or
     function Random [ (Range: Word) ] : Real;

Example:
     { On an EGA or VGA monitor, randomly rearrange the colors in
     the palette }
     Randomize;
     repeat
       SetPalette( Random(GetMaxColors), Random(GetMaxColors));
       Delay ( 50 );
     until Keypressed;

Purpose:
     Random produces a pseudo-random sequence of numbers.  In its default
usage, Random returns a Real random value between 0 and 1 ( 0<= Random < 1). 
Each subsequent call to Random returns another number in the pseudo-random
sequence.
     If you specify a Range value, Random returns a Word value betetween 0 up
to Range ( 0<= Random < Range).  For instance, Random(10) returns random values
in the range of 0 to 9.
     The specific sequence generated by Random depends upon the initial value
of the RandSeed variable, which is set by Randomize.  If you do not call
Randomize at the start of your program, every time that your program executes
it will generate the same random number sequence.



Randomize procedure
----------------------------------------------------------------
Declaration:  
     procedure Randomize;

Purpose:
     Randomize sets the RandSeed variable to a value translated from the
current system clock time.  Generally you will wish to call Randomize at the
start of your program.  If you do not call Randomize, your program will
generate the same random number sequence each time that it is executed.   You
can select a specific random number sequence by initializing the RandSeed
variable to a value of your own choosing.  See the description of RandSeed.

     

RandSeed variable
----------------------------------------------------------------
Declaration:  
     RandSeed : Longint;

Example:

     RandSeed := 99879;
     repeat
       SetPalette( Random(GetMaxColors), Random(GetMaxColors) );
       Delay ( 50 );
     until Keypressed;

Purpose:
     RandSeed contains the random number generator seed value.  By explicitly
setting RandSeed to a value of your choosing, you can repeatedly generate the
same random number sequence.  



Read procedure
----------------------------------------------------------------
Purpose:
     The Read procedure inputs values from the keyboard or text file and
assigns the values to variables.  When used on typed files, Read reads the next
file component into a variable.
     By setting {$I-}, you can manually check for errors that may occur when
reading data by checking the result code from IOResult.

Text files
Declaration:  
     procedure Read( [var F: Text; ] V1 [, V2, ..., Vn ] );

Description:
     When reading data from a text file or the system input (keyboard) file,
the behavior of Read depends on the data type of the variables to receive the
data and the position of the file pointer (for example, at end of file).  The
basic operation is to copy data from the input, convert if necessary to the
appropriate format, and place into one of the parameter variables.  Read may
have from 1 to many variables specified, and the variable type may be different
for each of the variables.  For example, to read 2 integers and a single real
value, write:
     var
       N1, N2 : Integer;
       X : Real;
     ...
     Read( N1, N2, X );

You should type the response to this input using blanks, tabs or the Enter key
to separate each of the values, such as,
     57 68 2345.987
Interestingly, you may use the Enter key to separate each value, such as,
     57 <Enter>
     68 <Enter>
     2345.987 <Enter>

Important!  The Dangling Enter key!
     In the previous example, the last <Enter> remains in the input buffer. 
Forthis reason, when reading input from the keyboard, it is recommended that
you use the Readln procedure instead of Read.  Readln behaves identically to
Read except that Readln swallows and disgards the Enter key.
     From a user interface perspective, it would be preferable to allow numeric
input values to be separated by the comma "," character.  Unfortunately, Readln
does not support this capability.
     The details of how Read performs on the various data types are described
in the following sections.  Also see the closely related procedure, Readln, and
the output procedures Write and Writeln.


String
     Read copies input characters to the string variable, until it encounters
an end-of-line or end-of-file condition.  The end-of-line character is never
included in the data copied to the string variable.  When reading data from a
text file, the next Read operation begins at the previous end-of-line
character.  Thinking that it is at the end of another line, Read returns a null
string.  The result of this is that Read only reads the first line and then
gets stuck on the end-of-line character.  The solution is to use Readln for
reading successive lines of text.
     If the input line exceeds the defined maximum length for the string
variable, then only that portion of the text that will fit is read into the
string variable.  The remaining text out to the next end-of-line character is
disgarded.

Integer
     When reading integer values to an integer variable, Read skips over
leading blanks and tabs and then reads until encountering the next blank, tab,
end-of-line character or end-of-file.  The characters that is has read are then
converted to an integer value.  If these characters do not correspond to the
proper format for an integer, an error condition occurs.  You can trap these
errors by setting {$I-} and checking the value of IOResult.  
     If the end-of-file is reached while reading the characters that make up an
integer value, then the integer value is set to 0.

Real
     When reading real number values to a real variable, Read skips over
leading blanks and tabs and then reads characters until encountering the next
blank, tab, end-of-line character or end-of-file.  The characters that it has
read, assuming that they correspond to the proper format for a real constant
are then converted to a real value and assigned to the variable.  If the
characters have an invalid format for a real constant, then an error occurs.

Char
     Read (Ch), where Ch is a Char data type, reads the next single character
from the file and copies the character to Ch.  If the file is positioned at the
end of file, Ch is set to Chr(26) to mark the end-of-file.  If the file pointer
is positioned on an end-of-line character, then Ch is set to Chr(13).  Note
that while you can use Read(Ch) to read an individual character from the
standard input or keyboard, you must still press the Enter key to terminate the
input.  If you wish to read invididual keystrokes without having to press
Enter, use the ReadKey function.


Typed files
Declaration:  
     procedure Read ( F, V1 [, V2, ..., Vn ] );

Description:
     In this form, Read is used to read individual recordor (or file
components) from a disk file.  You can read the data in the file sequentially,
by specifying multiple data values, or by repeatedly calling Read.  After each
record is read, the file pointer is advanced to the next record in the file. 
Toread records at random locations, call Seek to reposition the file pointer to
a new location in the file.  As with all I/O operations, you can manually check
for I/O errors by setting the {$I-} compiler option and checking the result
value returned by IOResult.
     For further information on record file operations, see chapter 3, "Disk
File operations".
See Seek, Write



ReadKey function
----------------------------------------------------------------
Declaration:  
     function ReadKey : Char;

Unit:       Crt

Example:

   1  { KEYTEST.PAS }
   2  Program KeyTest;
   3  Uses
   4    Crt;
   5  Var
   6   Code : Integer;
   7  
   8  
   9  Function GetKey : Integer;
  10  { Pauses for input of a single keystroke from the keyboard and returns
  11    the ASCII value.  In the case where an Extended keyboard key is
pressed,
  12    GetKey returns the ScanCode + 256.  The Turbo Pascal ReadKey function
  13    is called to perform the keystroke input.  This routine returns a 0
  14    when an Extended key has been typed (e.g. left or right arrow) and we
  15    must read the next byte to determined the Scan code.
  16  }
  17  Var
  18    Ch : Char;
  19  Begin
  20    Ch := ReadKey;
  21    If Ord(Ch) <> 0 Then
  22      GetKey := Ord(Ch)   { Return normal ASCII value }
  23    Else
  24    Begin
  25      { Read the DOS Extended SCAN code that follows }
  26      GetKey := Ord(ReadKey) + 256;
  27    End;
  28  End;{GetKey}
  29  
  30  Begin
  31  Repeat
  32    Code := GetKey;
  33    Writeln(Code);
  34  Until Code = 27;
  35  End.

Purpose:
     If a key has already been entered, ReadKey returns the key that was typed.
Otherwise, ReadKey waits for a key to be pressed on the keyboard.  In both
cases, the key that is entered is not echoed to the screen.
     Normal characters having ASCII codes from 1 to 255 are returned by
ReadKey.  However, the IBM PC supports additional character codes in order to
recognize the function keys, Ins, Del, and so on.  When one of these extended
keyboard characters is typed at the keyboard, ReadKey returns Chr(0) and you
must then call ReadKey a second time to read the actual scan code.  The
function GetKey, shown in the example above, provides a simple routine to read
both normal and extended IBM keyboard values.



Readln procedure
----------------------------------------------------------------
Declaration:  
     procedure Readln( [var F: Text; ] [V1 , V2, ..., Vn ] );

Purpose:
     Readln is identical to Read except that after reading the input data,
Readln causes the file pointer to advance to the start of the next line (just
after the previous end-of-line marker).
     Unlike Read, you can call Readln with no parameters and Readln disgards
all input characters up to the next end-of-line.  Readln, without parameters,
is often used to wait for the user to press the Enter key, like this:
     Write('Press Enter to continue.');
     Readln;



Release procedure
----------------------------------------------------------------
Declaration:  
     procedure Release (var P: Pointer);

Purpose:
     Release resets the top of heap pointer to the value in P, which should
have been initialized previously with a call to Mark.  See "The Use of Mark and
Release Procedures" in Chapter 3, "The Turbo Pascal Language" of the Turbo
Pascal Reference.



Rename procedure
----------------------------------------------------------------
Declaration:  
     procedure Rename(var F; Newname : String );

Example:
     var
       OldName: String;
       NewName: String;
       F: File;
     begin
       Write('Old filename: ');
       Readln( OldName );
       Write('New filename: ');
       Readln( NewName );
       Assign(F, OldName);
       Rename(F, NewName );
     end.

Purpose:
     With a filename assigned to file identifier F (using Assign), call Rename
to change the name of the file to Newname.  F must not be open.  Interestingly,
if NewName contains a directory name in addition to the filename, you can use
Rename to move a file from one directory to another directory, with or without
a name change.   Further, this move is nearly instantaneous, is handled
internally by DOS shifting its internal file directory around, not by actually
copying the entire file from one directory to another. For example, to move
MYFILE from the current subdirectory to the \TOOLS subdirectory, write,
     Assign( F, 'MYFILE' );
     Rename( F, '\TOOLS\MYFILE' );



Reset procedure
----------------------------------------------------------------
Declaration:  
     procedure Reset (var F [: File; RecSize: Word] );

Example:
     See "Disk file operations" in Chapter 3, "The Turbo Pascal Language".

Purpose:
     Reset opens an existing file of any type for subsequent reading or writing
of data and positions the file pointer to the beginning of the file.  In the
case of text files, Reset sets the file to read-only; if you wish to write to a
text file, use either Append or Rewrite to open the file.  Prior to calling
Reset, you should assign a filename to the file identifier by calling Assign (a
null filename '' may be used to open the standard input file).
     The parameter RecSize may only be used with untyped files only and
specifies a record size to be used when reading or writing blocks of data.  If
unspecified, the default record size is 128 bytes.  See BlockRead/BlockWrite
for additional details on block I/O.
     If the file does not exist, Reset produces an I/O error that can be
trapped by setting the {$I-} compiler directive and manually checking the
IOResult return value.
See Append, Assign, Close, Rewrite, Truncate



Rewrite procedure
----------------------------------------------------------------
Declaration:  
     procedure Rewrite (var F [: File; RecSize: Word] );

Example:
     See "Disk file operations" in Chapter 3, "The Turbo Pascal Language" of
The Turbo Pascal Reference.

Purpose:
     Rewrite creates a new file.  Prior to calling Rewrite, you should call
Assign to associate a filename (a null filename '' specifies the standard
output file) with the file identifier F.  Rewrite creates a new disk file,
erasing any previous file with the same name, and in the case of a text file,
makes the file available for write-only access.
     For untyped files, you can specify a record size parameter in RecSize (see
BlockRead/BlockWrite).  If unspecified, the default record size is 128 bytes.
See Append, Assign, Close, Reset, Truncate



RmDir procedure
----------------------------------------------------------------
Declaration:  
     procedure RmDir ( S: String );

Purpose:
     Use RmDir to delete an existing - and empty of any files - subdirectory. 
Sspecifies the name of the directory to delete.  If the directory still
contains files, or if the directory does not exist, an I/O error occurs (use
the {$I-} compiler directive and IOResult to check for errors).
See MkDir



Round function
----------------------------------------------------------------
Declaration:  
     procedure Round ( X: Real );

Example:
     {Round the value of Pi to the nearest's thousandth's place }
     Writeln( Round ( Pi * 1000 ) / 1000.0 );

Purpose:
     Where X is a real value, X is rounded to the nearest integer value.  When
X is halfway between two integer values (e.g. 3.5), then X is rounded as
follows:
       If X > 0, then X is rounded up to the next integer value.  Examples: 3.5
       becomes 4;  7.5 becomes 8.
       if X < 0, then X is rounded down to the next integer value.  Examples: 
       -3.5 becomes -4; -7.5 becomes 8.
The code shown in the example above illustrates a method of using Round to
round numbers to a particular number of decimal places.  In the example, the
value of Pi, 3.1415926535.... is multiple by 1000, producing 3141.5926535.... 
This value is then rounded to 3142.0, and divided by 1000.0, producing 3.142 as
the value of Pi rounded to the thousandth's place.
See Int, Trunc



RunError procedure
----------------------------------------------------------------
Declaration:  
     procedure RunError [ ( ErrorCode: Byte ) ];

Example:
     
     {$IFDEF Validate}
     if (Temperature < 0.0) or (Temperature > 212.0)  then
       RunError( TempOutOfRange );
     {$ENDIF}

Purpose:
     RunError, like the Halt procedure, stops program execution.  RunError,
however, generates a run-time error by setting ErrorAddr equal to the address
of the statement that called RunError, and if specified, sets ExitCode to
ErrorCode.
     A good use for RunError is to assist in debugging your programs by placing
internal consistency checks into your code.  For example, inside your code, you
may wish to place special code to insure that values are always within their
specified range, that pointers are non-nil, and so on.  An invalid value
indicates that your program is not operating correctly, so you can call
RunError to force an immediate halt to program execution.  Normally, you should
place this type of check code inside of conditional compilation statements so
that you can remove it from the final version of your program.
See ErrorAddr, ExitCode, ExitProc, Halt



Seek procedure
----------------------------------------------------------------
Declaration:  
     procedure Seek(var F; N: Longint);

Example:
     Also see "Random Access Data Files" in Chapter 3, "The Turbo Pascal
Language".

     type
       TDataRecord = record
          Name        : String[20];
          PhoneNumber : String[14];
          Age         : Integer;
          Available   : Boolean;
       end;
     var
       RandomFile  : File of TDataRecord;
       ...
     Write('Enter number of record to edit:  ');
     Readln( RecordNum );
     Seek( RandomFile, RecordNum );
     Read( RandomFile, DataRecord);
     if  DataRecord.Available  then
       Writeln('Record #', RecordNum, ' does not contain any data.');

Purpose:
     When performing random accesss to a disk file, Seek is used to position
the file pointer of file F, to the Nth component of the file.  Another way of
saying that is that Seek positions to the Nth record in the file.  The next
read or write operation will be made to record N.
     When you need to add new records to the end of a random access file, you
can use Seek to position to the last record, plus 1.  This way the next write
will append a new record to the end of the file.  A convenient way to access
the last record is to write,
     Seek ( F, FileSize(F) );
since FileSize returns the current filesize in terms of file components or
records.
See FilePos, FileSize, Read, Reset, Rewrite, Write



SeekEof function
----------------------------------------------------------------
Declaration:  
     function SeekEof [  ( var F: Text ); ] 

Purpose:
     Text files often contain extra blanks, tabs or blank lines after the last
useful data contained in the file.  For this reason, its possible that the
Eof(F) function will return False, even though the next Read (F) will encounter
an end of file condition.  The solution is to call SeekEof instead of Eof. 
SeekEof skips ahead, past any blanks, tabs or blank lines to determine if the
current file pointer is at the logical end of file.
     Make sure that you use SeekEof only on open text files, and no other file
types.  To manually check for file errors, set the {$I-} compiler directive and
check IOResult.
See Eof



SeekEoln function
----------------------------------------------------------------
Declaration:  function SeekEoln[ (var F: Text); ]
Purpose:
     SeekEoln exists for the same reason as SeekEof (See SeekEof).  SeekEoln
skips over trailing blanks and tabs in a line to locate the end-of-line marker.



Seg function
----------------------------------------------------------------
Declaration:  
     function Seg(X): Word;

Purpose:
     Computes and returns the segment part of the segment:offset address of X,
where X is any identifier, including variables, procedures and functions.
See Addr, Ofs



SetCBreak procedure
----------------------------------------------------------------
Declaration:  
     procedure SetCBreak( Break: Boolean);

Unit:       Dos

Purpose:
     The DOS operating system periodically checks for a the Ctrl-Break key
combination being pressed during program execution.  As you know, pressing
Ctrl-Break interrupts the flow of program execution.  However, the actual
processing of the Ctrl-Break key depends on the state of a DOS internal
variable.  GetCBreak returns the current state of this flag:  if set to True,
DOS checks for Ctrl-Break at every DOS system call; if False, DOS checks only
during input/output operations to the console (CON:), printer (LPT:) or
communications ports (COMx:).  
     You can set the state of the Ctrl-Break flag by calling SetCBreak and
setting Break to True, to enable checking at each DOS system call; or False to
restrict checking to input/output operations.
See CheckBreak, GetCBreak



SetDate procedure
----------------------------------------------------------------
Declaration:  
     procedure SetDate( Year, Month, Day: Word );

Unit:       Dos

Purpose:
     Sets the DOS system date to the Year, Month and Day parameters, where Year
should be in the range 1980 to 2099, Month in the range 1 to 12, and Day in the
range 1 to 31.  SetDate does not return an error status; if the date values are
erroneous, the new date is not set.
See GetDate, GetTime, SetTime



SetFAttr procedure
----------------------------------------------------------------
Declaration:  
     procedure SetFAttr (var F; Attr: Word );

Unit:       Dos

Example:
Assign(F, 'SCRATCH.DAT' );
GetFAttr( F, Attr );
SetFAttr( F, Attr or Hidden ); { Make the file a hidden file }

Purpose:
     Every file on the system has associated with it a set of file attributes. 
These attributes indicate if the file is read-only, hidden, or perhaps a
directory name.  You can obtain a file's attribute settings by calling
GetFAttr, where F is a file variable to which a filename has been assigned with
Assign (but should not be open).
     Using SetFAttr, you can change the file's attributes, as shown in the
example above.  The bit pattern used for attributes is determined by these
constants:  
     ReadOnly    =  $01;
     Hidden   =  $02;
     SysFile     =  $04;
     VolumeID =  $08;
     Directory      = $10;
     Archive     =  $20;
     AnyFile     = $3F;
AnyFile is typically used with FindFirst and FindNext and doesn't really have
any use here except that it can be used to match any file attributes with an
and operation.  
     Combinations of attributes are set by or'ing the values together, as shown
in the example above.  To get and set the file's access date and time, see
GetFTime and SetFTime.

     

SetFTime procedure
----------------------------------------------------------------
Declaration:  
     procedure SetFTime( var F; Time: Longint );

Unit:       Dos

Example:

     Assign( F, 'FSEARCH.PAS' );
     Reset( F );
     GetFTime ( F, LFileInfo );
     UnpackTime ( LFileInfo, DFileInfo );
     with DFileInfo do
       Writeln(Month, '-', Day, '-', Year,
          '  ',Hour:2,':',Min:2,':',Sec:2);

     { Change file's date to 1992 }
     DFileInfo.Year := 1992;
     PackTime ( DFileInfo, LFileInfo );
     SetFTime ( F, LFileInfo );

Purpose:
     Every file has associated with it, a date and time stamp indicating when
the file was last written to.  You can fetch this date and time information by
calling GetFTime (see GetFTime).  GetFTime returns the date and file time
packed into a Longint value (see UnpackTime for converting the packed time
format into individual date and time components).
     The example, above, shows how to use SetFDate to output a new date and
time stamp to the open file associated with F.  DFileInfo is declared as a DT
record, where DT defines the fields, Month, Day, Year, Hour, Min, Sec.  After
setting the field values, you must use PackTime to convert the record into a
packed Loingint, prior to calling SetFDate.
See GetFAttr, GetFTime, PackTime, SetFAttr, UnpackTime



SetIntVec procedure
----------------------------------------------------------------
Declaration:  
     procedure SetIntVec( IntNo: Byte; Vector: Pointer);

Purpose:
     Places an address, Vector, in to the interrupt vector table at entry
number IntNo, where IntNo is a value from 0 to 255.
See GetIntVec



SetTextBuf procedure
----------------------------------------------------------------
Declaration:  
     procedure SetTextBuf(var F: Text; var Buf [; Size: Word ] );

Example:
     (See GetMem for an example of a dynamically allocated text buffer).

     Program SetTextBufDemo;
     var
       F : Text;
       Buffer : Array[0..1023] of Char;
       TextLine : String;

     begin
       Assign( F, 'TEMP.PAS' );
       SetTextBuf( F, Buffer );
       Reset( F );

       while not Eof(F) do
       begin
          Readln( F, TextLine );
          Writeln( TextLine );
       end;

       Close( F );
     end.

Purpose:
     By default, all text files use an internal 128-byte buffer for reading and
writing textual data to and from the disk.  For faster text file performance,
programs can create their own text file buffer, of any appropriate size, by
calling SetTextBuf and passing the name of the buffer as the Buf variable
parameter.  By creating a larger text file buffer, such as 1,024 bytes, you
will generally see a significant improvement in performance.  You don't need to
specify the Size parameter; by default, SetTextBuf sets the internal text
buffer size to SizeOf(Buf).  The new  buffer will be used until you close file
F, or open a new file using the F file id.  After closing the file, be certain
to properly dispose of the buffer, particularly if the buffer is a dynamic
variable.

Important!  Call SetTextBuf Before Calling Append, Reset, or Rewrite
     As a rule of thumb, always call SetTextBuf just prior to calling Append,
Reset or Rewrite.  While you can call SetTextBuf after opening a file, the act
of switching text buffers can cause a potential loss of data.  So just don't do
that!
See "Disk file operations" in Chapter 3, or Append, Reset, Rewrite



SetTime procedure
----------------------------------------------------------------
Declaration:  
     procedure SetTime(Hour, Minute, Second, Sec100: Word);

Unit:       Dos

Purpose:
     Use SetTime to set the system clock.  If any of the values are out of the
permissible ranges (0 to 23 for Hour, 0 to 59 for Minute and Second, and 0 to
100 for Sec100), the time is not changed.
See GetTime



SetVerify procedure
----------------------------------------------------------------
Declaration:  
     procedure SetVerify(Verify: Boolean);

Unit:       Dos

Purpose:
     By calling SetVerify with Verify set to True, DOS will subsequently verify
the accuracy of all disk write operations (by rereading the data it just
wrote).  Since this is a time consuming operation, many applications will
choose not to use this feature.  To turn off automatic verification, call
SetVerify with Verify to False.



Sin function
----------------------------------------------------------------
Declaration:  
     function Sin ( X: Real ) : Real;

Example:

     GetAspectRatio ( Xasp, Yasp );
     { Compute (X,Y) 10% further out than the radius. }
     X := Round( 1.1 * Radius * Cos ( AngleInRadians ) );
     Y := - Round( 1.1 * Radius * Sin ( AngleInRadians ) *
            (Xasp/Yasp));

Purpose:
     Computes and returns the sine of X, where X is an angle measurement in
radians.  Note that many of the graphics routines use an angle measurement in
degrees. To convert from degrees to radians, divide the degree measurement by
180 / Pi.
See ArcCos/ArcSin, ArcTan, Cos, Tan



SizeOf function
----------------------------------------------------------------
Declaration:  
     function SizeOf( X ) : Word;

Purpose:
     SizeOf computes and returns the total memory storage, in bytes, of its
parameter X, where X is either a variable or a data type.  SizeOf is frequently
used when allocating memory blocks dynamically (such as GetMem ( PBuffer,
SizeOf(Buffer) )) and elsewhere.



Sound procedure
----------------------------------------------------------------
Declaration:  
     procedure Sound(Hz : Word);

Unit:       Crt

Purpose:
     Activates the PC's speaker, setting it to a tone having the frequency (in
hertz or cycles per second) specified by Hz.  The only way to turn off the
speaker, once activated, is to call NoSound.  Between calls to Sound and
NoSound, use the Delay procedure to leave the speaker turned on for a set
amount of time.
See Delay, Sound



SPtr function
----------------------------------------------------------------
Declaration:  
     function SPtr : Word;

Purpose:
     Returns the value of 80x86 SP (stack pointer) register.  SP is always
relative to the SS (Stack segment) register.
See SSeg



Sqr function
----------------------------------------------------------------
Declaration:  
     function Sqr( X ) : <type of X>

Purpose:
     Returns X squared, which is equivalent to X * X.



Sqrt function
----------------------------------------------------------------
Declaration:  
     function Sqrt(X: Real) : Real;

Purpose:
     Where X is greater than or equal to 0, Sqrt(X) returns the square root of
X.  Sqrt(X) if not defined for negatives values of X.



SSeg function
----------------------------------------------------------------
Declaration:  
     function SSeg: Word;

Purpose:
     Returns the current value of the SS or Stack Segment register of the CPU.
See SPtr



Str procedure
----------------------------------------------------------------
Declaration:  
     procedure Str(X [: Width [:Decimals] ]; var S: String);

Purpose:
     When you need to convert a byte, integer, word, longint, or one of the
real data types to a string representation, use the Str procedure, specifying
an optional field Width and number of Decimals of accuracy.  For example,
     Str( 3.14159, S );
sets S equal to '3.1415900000E+00'.  To convert the value to a decimal
notation, you must specify the Width and Decimals parameters, where Width
specifies the overall length of the resulting string, and Decimals specifies
the number of decimal places to be included in the result.  For example,
     Str( 3.14159:5:2, S );
sets S equal to '3.14'.  Changing Decimals from 2 to 4 sets S equal to
'3.1416'.  Note:  The value of X is rounded, as needed, to fit within the
specified width and decimal places.
     To convert the string representation of a number, say '3.14159' to the
internal numeric format, use the Val procedure.



Succ function
----------------------------------------------------------------
Declaration:  
     function Succ(X) : <same type as X>;

Purpose:
     Where X is any ordinal value, Succ(X) returns the successor of X.  For
example, the Succ(9) is 10.  See Pred for an example involving ordinal types.
See Inc, Pred



Swap function
----------------------------------------------------------------
Declaration:  
     function Swap(X) : <same type as X>;

Purpose:
     When X is either an integer or word value, Swap(X) returns the result of
swapping the high and low order bytes of X.  You can examine the high or low
order bytes of an integer or word value using the Hi and Lo functions.



SwapVectors procedure
----------------------------------------------------------------
Declaration:  
     procedure SwapVectors;

Unit:       Dos

Example:
     See Exec.

Purpose:
     SwapVectors is normally used before and after calling the Exec procedure,
which is used to launch another application.  Essentially, SwapVectors results
in saving the state of the interrupt vectors before a call to Exec, and
restoring the vector table upon return.  This helps prevent problems that may
occur if your application depends on specific settings in the interrupt vector
table - settings that may be changed by the application which is launched.



Test8087 variable
----------------------------------------------------------------
Declaration:  
     Test8087: Byte = 0;

Purpose:
     When a Turbo Pascal program starts up, the initialization code determines
if the 8087 math coprocessor is available for use, and sets Test8087 as
follows:

Value  Meaning
0      No math coprocessor found.
1      8087 available.
2      80287 available.
3      80387 available.



TextAttr variable
----------------------------------------------------------------
Declaration:  
     TextAttr: Byte;

Unit:       Crt

Example:
     uses Crt;
     begin
       TextColor(3);
       TextBackground(5);
       Writeln('Foreground color=', TextAttr and 15);
       Writeln('Background color=', (TextAttr shr 4) and 15);
       Readln;
     end.

Purpose:
     Stores the currently defined text attributes (color, background color) in
internal format.



TextBackground procedure
----------------------------------------------------------------
Declaration:  
     procedure TextBackground( Color: Byte );

Unit:       Crt

Purpose:
     Text written to the display has two color attributes:  a foreground color
and a background color.  TextBackground sets the current background color
attribute to the value of Color, such that all text output following the call
to TextBackground will be drawn with the specified background color.  Color
should have a value from 0 to 7, to choose one of the 8 possible background
colors.  You may substitute one of the following constants for your color
choice:
     Black       0
     Blue     1
     Green       2
     Cyan     3
     Red      4
     Magenta     5
     Brown       6
     LightGray      7



TextColor procedure
----------------------------------------------------------------
Declaration:  
     procedure TextColor(Color: Byte);

Unit:       Crt

Purpose:
     Text written to the display has two color attributes, a foreground color
and a background color.  TextColor sets the foreground color attribute to the
value of Color, so that all subsequent text output will be written in the
selected color, using the current TextBackground color.  Foreground color
choices may range from 0 to 15, with values 8 through 15 referred to as "high
intensity" equivalents of colors 0 to 7.  You may use the following constants
to select the color choice:
     
     Black       0
     Blue        1
     Green       2
     Cyan        3
     Red         4
     Magenta     5
     Brown       6
     LightGray   7
     DarkGray    8
     LightBlue   9
     LightGreen  10
     LightCyan   11
     LightRed    12
     LightMagenta   13
     Yellow      14
     White       15



TextMode procedure
----------------------------------------------------------------
Declaration:  
     procedure TextMode(Mode: Word);

Unit:       Crt

Example:
     TextMode ( CO80 );

Purpose:
     Most of today's video displays support a variety of text operating modes,
including both 40 and 80 column wide displays, color and black & white display
modes, and for EGA and VGA displays, the ability to display 43 or 50 lines of
text.  The TextMode procedure selects the appropriate display mode by setting
Mode to one of the following constants:

     BW40        Selects B&W in 40 x 25 line display mode
     BW80        Selects B&W in 80 x25  line display mode
     Mono        Selects B&W in 80 x25 line on monochrome
                 display adaptor card
     CO40        Selects color in 40 x 25 line display mode
     CO80        Selects color in 80 x 25 line display mode
     C40         Same as CO40
     C80         Same as CO80
     LastMode    Restores the previous video mode prior to
                 program startup.

The selected text mode takes effect immediately and results in the screen being
cleared.
     LastMode is a variable and it stores the contents of the video mode at the
time the program was started.  If your program changes video modes during
program execution, you may wish to save the value of LastMode so that you can
restore the appropriate video mode at program termination (also see ExitProc).
     You may also use TextMode to select 43 or 50 line mode for use on EGA/VGA
displays by adding or or'ing the constant Font8x8 (where Font8x8 = 256 and
selects a small font for use in the higher resolution modes) to the mode
selection constant, like this:
     TextMode( LastMode and 255  or Font8x8 );
In this form, the current video mode (such as 80 column color) is retained but
the number of lines is increased to 43 or 50 lines, depending on your display. 
You may explicitly select both a video mode and 43 or 50 line mode by writing:
     TextMode( CO80 + Font8x8 );
To switch back to 25 line mode and maintain the the other mode settings, you
can write,
     TextMode( LastMode and 255 );
or specifically set a new mode, such as,
     TextMode( BW80 );
You can also use the Lo() and Hi() functions to check the value of LastMode. 
Hi(LastMode) returns Font8x8 if 43/50 line mode is in effect.  Lo(LastMode)
returns the current video mode.



Trunc function
----------------------------------------------------------------
Declaration:  
     function Trunc( X: Real ): Longint;

Purpose:
     Trunc converts the real value X to a longint integer value by disgarding
the fractional portion of X.  For example, Trunc(3.14159) results in 3.  When
converting real values to Longint, be aware that the range of valid real
numbers exceeds the permissible range of Longint, and it is possible to receive
an out of range run-time error.
See Frac, Int, Round



Truncate procedure
----------------------------------------------------------------
Declaration:  
     procedure Truncate(var F);

Purpose:
     When using any open file, other than text files, Truncate(F) disgards all
current data in the file from the current file pointer position on to the end
of the file, effectively making the current file position the new end of file.



Unpacktime procedure
----------------------------------------------------------------
Declaration:  
     procedure UnpackTime(Time: Longint; var DT: DateTime);

Unit:       Dos

Example:
     uses dos;
     var
       F : Text;
       LFileInfo : Longint;
       DFileInfo : DateTime; { declared in Dos unit }
     begin
       Assign( F, 'FSEARCH.PAS' );
       Reset( F );
       GetFTime ( F, LFileInfo );
       UnpackTime ( LFileInfo, DFileInfo );
       with DFileInfo do
          Writeln(Month, '-', Day, '-', Year,
            '  ',Hour:2,':',Min:2,':',Sec:2);
       Readln;
     end.

Purpose:
     The procedures GetFTime, SetFTime, FindFirst and FindNext all use a
special packed date/time format that packs the date and time information into a
Longint value.  To convert the Longint packed representation into individual
date and time components, use UnPackTime.  To convert back to packed format,
use PackTime.
     UnPackTime converts its Time parameter value into the contents of the DT
record parameter variable.  DT contains separate fields for Year, Month, Day,
Hour, Min and Sec, each a Word type value.  You can use these fields to display
the date or time of a file in traditional date or time formats.



UpCase function
----------------------------------------------------------------
Declaration:  
     function UpCase (Ch: Char): Char;

Example:
     function Uppercase( S : String ): String;
     var
       I : Integer;
     begin
       for I := 1 to Length(S) do
          S[I] := Upcase(S[I]);
       Uppercase := S;
     end;

     begin

       Writeln(Uppercase('Now is the time for all good men and women...'));
       Readln;

     end.

Purpose:
     UpCase converts its parameter Ch to uppercase, (if Ch is a letter from 'a'
to 'z') or makes no changes if Ch is not a lower case letter, and returns the
new value as its return result.  The example above shows a simple function to
convert an entire string to upper case.



Val procedure
----------------------------------------------------------------
Declaration:  
     procedure Val(S: String; var V; var Code: Integer);

Example:
     uses Crt;
     var
       S : String;
       Result : Real;
       ErrCode : Integer;
     begin
       ClrScr;
       Gotoxy(1,10);
       Write('Enter number: ');
       Readln(S);
       {$R-}
       Val( S, Result, ErrCode );
       if ErrCode = 0 then
          Writeln(S, ' converted successfully to ', Result )
       else
       begin
          Gotoxy( 14+ErrCode, 11 );
          Write('^ invalid character in number.');
       end;
       Readln;
     end.

Purpose:
     Val converts a string containing a number, into its internal numeric
format for the appropriate variable type, such as integer or real.  The string
to convert is passed as parameter S and is returned in variable parameter V. 
     The string to convert must contain a valid number for the type of the
variable parameter V.  If an invalid character is detected during the
conversion, Code is set to the index of the offending character within string
S.  The example code above uses the returned error code to display a pointer to
the invalid character.
     When converting numbers, its possible for the number in S to be out of
range of the permissible values for variable V.  For example, attempting to
convert the string value shown below results in a range violation:
     var
       W : Word;
     ...
     Val('8799343.5', W, ErrCode );

To avoid out of range errors, you can select the {$R-} compiler directive to
disable automatic range checking.  A more thorough procedure, however, is to
convert the value first to a temporary Real or Longint variable and then use an
if-then statement to insure that the result falls within the valid range for
the target variable.
     


WhereX function
----------------------------------------------------------------
Declaration:  
     function WhereX : Byte;

Unit:       Crt

Purpose:
     Returns the current X coordinate of the text cursor.  WhereY returns the
corresponding Y coordinate.



WhereY function
----------------------------------------------------------------
Declaration:  
     function WhereY: Byte;

Unit:       Crt

Purpose:
     Returns the current Y coordinate of the text cursor.  



WindMin/WindMax variables
----------------------------------------------------------------
Declaration:  
     WindMin, WindMax : Word;

Unit:       Crt

Purpose:
     WindMin stores the coordinate of the upper left corner and WindMax the
lower right corner of a window defined by the Window procedure.  The values are
stored such that the X value is stored in the low byte and Y in the high byte. 
Note that in WindMin and WindMax, the stored values are 0-relative, and not
1-relative as used in the Window procedure.  This means that, for example,
WindMax may contain an X value of 24 (instead of 25) and a Y value of 79
(instead of 80).



Window procedure
----------------------------------------------------------------
Declaration:  
     procedure Window(X1, Y1, X2, Y2: Byte);

Unit:       Crt

Purpose:
     When your program starts running, the default text window is the entire
screen, from (1,1) to (80,25).  However, you can redefine this text window to a
smaller region of the screen, and thereafter, all subsequent output and related
screen I/O statements are performed relative to the location of the window. 
Theparameters (X1,Y1) define the upper left corner of the window, and (X2,Y2)
define the lower right corner of the window.  For example, to define a narrow
window occupying just the right half of the screen, you may write,
     Window(40, 1, 80, 25 );
After the call to Window, subsequent I/O will be relative to this window.  A
Gotoxy(10,10) positions the cursor to the physical location (40+10, 1+10).
     You can use the Window statement for a variety of purposes because it
effects all of the following routines:
     ClrEol, ClrScr, DelLine, GotoXY, InsLine, WhereX, WhereY, Read, Readln,
Write, Writeln
     You can, for instance, create a small viewing window on the screen for
storing a list of items.  Through the use of the DelLine and InsLine
procedures, you can make text within the viewing window scroll up or down, as
required.
Important!  Window Doesn't do Windows!
     Window does not provide for overlapping windows nor does it save the
region behind the active window.  Window merely provides a simple way of
restricting your screen output to specific regions on the screen.
See WindMin/WindMax



Write procedure
----------------------------------------------------------------
Purpose:
     When used with text output files, Write outputs one or more values to the
screen or to a text file.  When used with typed files, Write outputs its
parameter values to the next file component (or record) in the file.  For typed
files only, Seek may be used to position the file pointer to specific record
location prior to calling Write.

Write with Text Files
Declaration:  
     procedure Write( [ var F: Text; ] V1 [, V2, ..., Vn ] );

Description:
     Write outputs its parameters to the screen or to the text file F, if F is
specified.  For numeric parameters, a MinWidth and a DecPlaces parameter may
specify the minimum character width allocated to the item, and a number of
decimal places, respectively, using the format:
     Write( V1: MinWidth : DecPlaces );
MinWidth and DecPlaces are optional, however MinWidth must be specified in
order to specify DecPlaces, and DecPlaces may only be specified for real data
types.  If the length of the item being output is less than the MinWidth
specified, then leading blanks are placed before the item so that it will
occupy at least MinWidth characters.  For example,
     Write('/', '*' : 5, '/');
displays,
     /     */

Formatting Trick
     Many applications need to output large groups of blanks between output
items, such as when formatting a table of values.  While there are several ways
to accomplish this, such as placing Write(' ') inside a for loop, the easiest
method uses the MinWidth parameter to specify the number of blanks desired,
like this, which outputs 40 blanks:
     Write(' ':40);

Output formatting for each data type is described in the following sections.

String
     String values are copied directly to the output, byte by byte.  If
MinWidth is specified, sufficient leading blanks are output, if needed, to make
the string occupy exactly MinWidth character spaces.  If the string length
exceeds MinWidth, then the entire string is output.

Integer, Word, Longint
     Integer values are converted from their internal format to a textual
representation.  If MinWidth is specified, sufficient leading blanks are
output, if needed, to make the textual representation of the integer as wide as
MinWidth.

Real
     Real values are normally output in exponential notation, such that the
statement,
     Writeln(Pi)
outputs,
     3.1415926536E+00
When MinWidth is not specified for a real value, a default MinWidth of 17 is
assumed.  In the exponential floating point format, if MinWidth is less than 8,
MinWidth is set to 8.  To output a real value in familiar fixed point notation,
such as,
     3.1416
you must specify a MinWidth and DecPlaces parameter.  For example,
     Writeln(Pi:7:4);
outputs,
     3.1416
Note that this may resulting in rounding of the value, as needed.

Char
     Individual character values are output directly.  If MinWidth is
specified, the MinWidth-1 leading blanks are placed before the character value.

Boolean
     Boolean values are displayed as either True or False, according to the
value of the boolean expression.


Write with Typed Files
Declaration:  
     procedure Write( var F; V1, [, V2, ..., Nn ] );

Description:
     When used with typed files, Write outputs one or more components to the
file, and positions the file pointer to the next record in the file.  The Seek
procedure may be used to position the file pointer prior to writing the record.
See  "Disk file operations" in Chapter 3, "The Turbo Pascal Language" for
complete details on sequential and random typed file access.



Writeln procedure
----------------------------------------------------------------
Declaration:  
     procedure Writeln( [var F:Text; ] [V1 [V2, ..., Vn] ] );

Purpose:
     For use in displaying output to the screen or writing output to text
files, Writeln is equivalent to Write, except that it also outputs an
end-of-line character following the parameters.