rtsi.com/2014.01.www.rtsi.com.tar/www.rtsi.com/OS9/OSK/DEMOS/UB_68020.LZH/mini_manual

Text File | 1996-06-26 | 46KB | 1,104 lines

                            UniBasic

                              DEMO


                             Preface

This archive or disk contains the demo version of UniBasic plus
many examples or its use.

This archive or disk may be freely distributed as long as ALL of
the files are kept intact and not modified or edited in any way.
If you add additional files to this archive or disk you must make
that very clear. We do not want to be blamed for any mistakes
other than our own.

The file 'ubdemo' is of particular concern to us. It is the
result of years of work and much money invested. It is provided
in a 'limited' form. We made a modification to limit the size of
the program it will compile. It will compile all of the example
programs included with the demo or a program of your own of about
a page or two. We did this by reducing the size of the symbol
table. Otherwise it is the same program as the one we hope you
will buy after you have tried out this demo. (The demo will
always be at version 1.0 while the one you buy will be the
current version.)

We are concerned because we know there are people out there that
will take this limitation as a challenge and try to defeat it. In
so doing they will have done the very same thing as if they broke
into our office and stole a copy of the program. Or gotten a copy
from a legitimate buyer. In all cases having a copy of UniBasic
without paying for it is stealing. There is no justification for
this. If you think we are charging too much for UniBasic then
don't buy it. Please don't use the price or any other excuse to
steal it.

Thank you
Computer Design Lab

Purchase UniBasic, at the address, phone number, or Compuserve address
listed below.

Several versions are available are present with more on the way. Those
currently available are:

MSDOS Version................................$99.00 US
Linux Version (personal use)................$149.00 US
Linux Version (commercial/multi-user).......$299.00 US
OS9/68000 Version (personal use)............$149.00 US
OS9/68000 Version (commercial/multi-user)...$299.00 US
OS9/68020 Version (personal use)............$149.00 US
OS9/68020 Version (commercial/multi-user)...$299.00 US

Make check or money order payable to Computer Design Lab in US funds.
Add $5.00 shipping/handling for US ground transportation.

UniBasic is shipped within 24 hours of order and can be shipped
overnight if required.

All versions come with a complete printed manual, reference card
and technical support from CDL.

To purchase your copy of UniBasic contact:

Computer Design Lab
RR1 Box 36
Rhineland, MO 65069
Tel: [phone]
CompuServe: 72762,3375


Limited warranty and limitation of remedies
------- -------- --- ---------- -- --------

Computer Design Labs is not responsible for any damage to the
user's computer system or data and in no event will Computer
Design Labs, its officers, directors, employees or agents be
responsible to the purchaser for any consequential, incidental,
or indirect damages (including damages for loss of business
profits, business interruption, loss of business information and
the like) arising out of the use or inability to use the Computer
Design Labs product, even if Computer Design Labs has been
advised of the possibility of such damages.

Unless explicitly stated in writing, Computer Design Labs does
not grant permission to distribute the software for profit in any
form, Non-profit distribution of the software is acceptable
without prior written notice, providing that the software is not
modified in any way, and the complete works of the software are
included in the distribution package.




                         UniBasic DEMO

                           VERSION 1.0

                           MiniManual





                    GENERAL FUNCTION and SYNTAX

UniBasic is a C code macro BASIC compiler for use on various operating
systems. This means that the user program written in BASIC (with or
without the UniBasic enhancements) is translated and compiled into C.
UniBasic incorporates many of the features of the newer structured
BASICs,allows the intermixing of C code, and supports a host of
language enhancements and compiler functions never before found in
any BASIC implementation.

The compiler features include macro capability, C function compatibility,
conditional compilation, compiler variables, object module linkage, and
a full set of compiler directives to manage these features. These
directives include #MACRO, #ENDM, #SET, #IF, #ENDIF as well as others.

The language enhancements include a full range of data types, structured
constructs (such as IF/ENDIF, WHILE/ENDWHILE), C language intermixing,
named labels (instead of line numbers), full access to operating system
calls, and full support of pointer variables, vector variables (based
variables), and memory buffers.

Since the output is C, the performance will be greater than interpreter or
"I-code" compiler implementations. The less experienced programmer can
immediately take advantage of the performance of UniBasic even before
learning the advanced features and enhancements.

The structure of UniBasic is slightly different than standard BASIC in that
labels, functions, and directives always begin in column #1 while statements
must be indented at least one space (or tab). This is quite similar to
the convention for some assemblers. Also as previously mentioned,
UniBasic uses labels rather than line numbers. If UniBasic encounters
a line number it will convert it to a pseudo label automatically which
greatly simplifies conversion of other BASIC programs to UniBasic.
Also since the style of UniBasic is somewhat akin to assembler,
multiple statements per line are not allowed. Instead the style is
to have the statement (or assembler instruction) followed by a
comment on each line for maximum readability. Unlike interpreters
and "I-code" compilers, comments and remarks have no effect on program
size or speed so their use is strongly encouraged. There are no default
data types which means that each variable must be dimensioned (declared)
before it can be used in a statement (or assembler instruction). Keywords
may be entered in any mix of upper and lower case while symbols are case
sensitive. For example, Test and test are two different symbols.

Since the macro capability allows both BASIC and C code within the macro,
macro libraries afford the effect of user-defined statements and
instructions.

The syntax of the various BASIC statements are more or less standard with any
other BASIC except for the requirement to indent a least one space or tab.
Labels, variables, macros, and any other symbols must begin with a letter
or underscore and may contain letters, numerals, or the underscore
character. Also, the last character may be a '$'. The maximum length is
28 characters. Since many system and compiler symbols begin with an
underscore, it is best to avoid the naming of symbols which begin with an
underscore. Each statement and compiler directive is explained in detail
in this manual as to syntax, function, and use.

UniBasic allows those who would prefer to program in BASIC to have the same
kind of power and utility as those who program in C and straight assembler
while maintaining the simplicity and straightforwardness of BASIC. Also the
linkage capability allows the combining of UniBasic modules with assembly
and "C" relocatable modules.

                             DATA TYPES

A brief description of data types is given in the DIM, TYPE, and PARAM
sections and a detailed coverage of pointer variables is given in the
POINTER VARIABLES section so much of what is said here may be redundant.
The mechanics of declaring variables and the various attributes assigned
to them is explained in the DIM, TYPE, and PARAM sections.

The first classification of a variable is whether it is a simple variable or
an array variable. An array is a linear table of the data type.
Multi-dimensional arrays are not allowed. UniBasic does no bounds checking
of arrays at run time. This may be an advantage or disadvantage depending
on whether the bounds violation is intentional or accidental. If it is
accidental it is possible to "clobber"adjacent data. Intentional violation
of array bounds is certainly not an example of structured programming, but
does allow the programmer to take advantage of knowing how the data is
stored and to, for example, treat a string as a byte array.

The second classification of a variable is whether it is local, global, or
external. Local variables are known only to the module in which it occurs.
Global variables are known to all modules in a set which will be linked
together. External variables refer to those which are declared to be
global by another module but are referred to by the current module.

The third classification of a variable is whether or not it is a pointer
variable . Pointer variables are variables in their own right and are a
special form of a 32 bit unsigned integer which have a special property
to index their value according to the data type which they are declared to
point to. The data which is accessed by a pointer variable is treated as
if it came from a standard variable of the type pointed to. There are
three functions which are used with pointers. The first is ADDR() and
is used to set the pointer to an address of a variable (presumably but not
necessarily of the same type as the pointer). The second is BUFADR() which
is used to set the pointer to the address of a memory buffer (see BUFFER,
BUF(), BUFSIZ() section). The third is INDEX() which is used to index the
pointer according to the function argument multiplied by the number of
bytes in the data type of the pointer.

The fourth classification of a variable is whether or not it is a vector.
Vectors (based variables) are similar to (but not the same as) "references"
in C++. Vectors are closely related to pointers in concept. Once a vector
has been declared with a DIM statement and subsequently assigned an address
with a SETVEC statement, it may be used as an ordinary variable. The vector,
then, has the power of a pointer without the need for pointer notation and
allows array sub scripting in buffers instead of pointer arithmetic.

Another type of variable is defined by the PARAM statement which must
immediately follow the occurrence of a function. The names of these
variables must not conflict with regular variable names, but are unique
within in the function in which declared. In other words, two are more
functions may declare the same PARAM names. These parameter variables as
well as labels within the function are removed from the symbol table
when the function body ends. These parameter variables "accept" values
passed to the function when the function is invoked (called). The
number of these PARAM statements must correspond to the number of
arguments of the caller.

                         USING the COMPILER

Once the program is written using an editor, the program is ready to be
compiled. This is accomplished by invoking the compile command as follows:

ub FileName.b {opts} {CompVar=Const...}

ub is the compiler's executable name. The compiler's executable object code
is normally stored in the CMDS directory (OS9), the usr/bin directory
(Linux), or as appropriate for the system being used.

If no options are given, UniBasic will compile the source (.b) file, creating
a C (.c) file. UniBasic then invokes the C compiler which in turn compiles
the C code to assembler and links the file into an executable object file.
This sequence of events may be modified or enhanced by the use of command
line options..

The options (opts) are as follows:

-a    Compile to assemblerfile(.a file OS9)(.s file Linux)
    (do not assemble and link)

-e    Compile with no output but check for errors. This is useful at
    various stages of program development. It allows the programmer
   to passively check for certain errors.

-f    Function mode. This option is used to compile functions or any
    non-mainline procedure    or subroutine. The file is processed to a
    relocatable assembler file (.r OS9) (.o Linux)

-g    Generate symbol module for the debugger. Automatically invokes the -g
    option to the    linker for later debugging purposes.

-r    Turn off remarks (comments) in -c option

-s    Symbol table dump to stdout

-v    Verbose mode. Reports detailed progress of compile process

-u    Unabridged hexadecimal symbol dump which also includes internal
    compiler symbols

-c    Compile to C file (.c) See -r option

-? Display options

                     WRITING and USING PROGRAMS

The default mode of the compiler is PROGRAM mode (also known as mainline).
To enter FUNCTION mode it is necessary to invoke a command line option,
but to enter PROGRAM mode no action is required.

The general form of a program is as follows:

* declare variables
 DIM A:LONG
 DIM B:LONG

* declare function
#FCN Square() INT

* program code body
 B=4
 A=Square(B)
 PRINT A+B
* END statement (optional)
 END

* (internal) function body
Square()
 PARAM X:LONG
 RETURN X*X

The program is written using your favorite editor. Remember that compiler
directives and labels always start in column one and that statements are
always indented at least one space or tab. When you write the program
remember to include a ".b" suffix to the program name.

To compile the program, enter the command to run the compiler.
(In this example the program name is "test")

ub test.b

If there are no errors reported, you may run the program by typing:

test (CR)

If the program does not perform as expected, it is sometimes useful to insert
PRINT statements as various points in the program to check its progress.
If the user is familiar with the Debug program, it is useful to compile the
program with the -g option which sets up a symbol module for the debugger.
If there are errors in your program that violate some rule or syntax of
the compiler, an English language indication of that error is reported.
If there are any errors at all, the compiler will not pass the output to the
C compiler (which will in turn invoke an assembler, optimizer and linker).
It is also possible to get an error from the C compiler, assembler, or
linker.Refer to the C compiler, assembler, or linker manual to resolve
these. The UniBasic compiler should "catch" most errors so that you will
not see C compiler, assembler, or linker errors in most cases. If you do,
you should report the error to CDL for possible resolution.

                   USING and DECLARING FUNCTIONS

UniBasic contains a large number of pre defined functions and the user may
write additional functions. Either compiler functions or user functions are
called and used in the same manner except that USER FUNCTIONS which do not
expect a return value (void functions) may be invoked by entering the
function name just as you would a statement name.

The pre defined compiler functions need not be declared. To declare a user
function we use the #FCN compiler directive. We must also "tell" the
compiler where to locate the function we just declared if the function is
external to our program. The optional ReturnType tells the compiler what
type of data to expect to be returned by the function. The three valid
return types are INT, FLOAT, and VOID. The default is INT. VOID means
return nothing. The function should either be in the RELS or LIB directory.
The number in parens tells the compiler how many arguments will be passed
(must be 1 thru 9). If a "c" (or nothing) is used for the number of
arguments, it indicates a variable number of arguments may be passed. If
"c" (or nothing) is used, then no checking is done as to number of
arguments. For example:

an internal fuction

#FCN DoIt() <ReturnType>

or an external function:

#FCN DoIt()<ReturnType>
#LIB LibName

The method used to set up functions and pass arguments is compatible with the
C compiler so it is possible to use C functions or to write functions for C.

If you forget to declare a function before using it, an error will be
reported by the compiler. If you forget the #LIB directive the linker will
return an error in the case of an external function.

It is common practice to build "header" files which declare functions and
tell the compiler where to find them. These files are then invoked with the
#USE directive.

The compiler will check the number of arguments declared against the number
of those given and will verify the data type (unless the number given is "c"
or nothing).

Since these user functions are modular in nature, their use can result in a
well structured program with good readability and if the functions are
general in nature they may be used by many programs (and programmers).

Functions which are included in the program (local functions) must be
declared just as external functions are with a #FCN directive. These need
no or #LIB directive. Symbol names used in these local functions are not
unique as in external functions, but are global within the program itself
(except for PARAM variables which are local to a function).

Local functions are written the same way as external functions. The function
name is not declared as a global label as in an external function, and as
mentioned above, variables are global with the rest of the program.

In the case of external functions, the function must be declared in both the
calling program and in the function itself.

WRITING and USING EXTERNAL FUNCTIONS

External functions are modules of previously compiled relocatable object
modules (ROF's).They may also be merged with other functions into a
library file.

Writing an external function with UniBasic is the same as writing any program
in UniBasic except the function is compiled with the -f option. Execution
begins in the function with the occurrence of the function name in column
1 followed by "()". The next line(s) are the PARAM statements which must
correspond to the number of arguments being passed to the function.

The variables and labels in an external function are independent of the
labels and variables in other functions and the calling programs unless
they are declared as EXTERNAL or GLOBAL.

Normally, a function is called with a list of arguments or parameters.
In UniBasic these are passed by value. To pass arguments by reference, the
arguments are passed as pointers. The pointer itself is passed by value but
the data it points to is said to be passed by reference. To pass arguments
to a function the calling program invokes the function in an expression
with the arguments specified in order within parentheses following the
function name. The compiler will put these arguments on the stack as
required automatically. The function extracts these arguments from
the stack when the function declaration is made as described above. The
general order of a function is as follows:

* declare local function variables (not used in this example)
 DIM x:BYTE
 DIM y:SHORT
 DIM z;LONG

FcnName()
 PARAM p:LONG
 PARAM q:LONG

* the body of the function goes here
* finish up with a RETURN
 RETURN p*q

To compile this function:

cb FcnName.b -f

The -f on the command line instructs the compiler to compile to a relocatable
file to be linked in later.

Functions can also share labels and variables with the calling program by
using GLOBAL labels and symbols. The calling program declares GLOBAL while
the function declares EXTERNAL or vice-versa.

                             C CODE

It is possible to intermingle C code in your Basic program. This may be done
on a single line basis by writing a ";" in column 1 followed by C code or on
a block basis by using the #C and #ENDC compiler directives with C code
between.

The C code may refer to Basic variables which have been declared but it is
important to know the details of how UniBasic handles the declaration in its
C output. For example, any complex data structure in Basic results in an
array of type char in the C output. Also, any pointer declaration in Basic
results in a long in C. Strings in Basic are null terminated (as in C) only
if the string is not "full". Members of structures are padded to an even
byte boundary when the next member isa short, long or float type or if it
is the last member in the structure. String declarations result in a char[],
but no "extra" byte is required as in C strings. Simple integer types and
float (real) do map as expected to comparable C types as do arrays of
these types.

While the compiler thoroughly checks the syntax and compiles BASIC statements
into C code, it does not process C statements to that degree. Since this is
the case, all errors involving C syntax and symbol problems (missing,
duplicate, etc.) will not show up until the C compiler processes the file.
The #SET compiler directive also outputs a C version of its result (#define)
so that the C compiler may apply this result to C code (if used).

An example of the use of C code in a BASIC program follows:

 REM Program to clear an entire string
 REM
 DIM A$:STRING[32]
 DIM I:LONG
 FOR I=1 TO 32
;A_Str[I-1]=0;
 NEXT I

Notice that since C does not allow a "$" in a symbol name, that UniBasic
changes the "$" to "_Str". Also notice that the string in Basic is an array
of type char in C with the first element addressed as0.

Due to the power and efficiency of UniBasic there should not be much (if any)
need to intermix C code, but the ability to do so is there if you need
(or want) it.

If you wish to use the C string functions on Basic strings, make sure the
basic string is at least one byte short of being "full" or DIM a byte
immediately following the DIM of the string and then initialize the
"tacked-on" byte to 0 (null).

There are several ways to include comments or remarks in a program. The first
and most obvious is the use of the REM statement which is standard BASIC.
The second is to follow the statement (on the same line) with a backslash
(\) followed by any comment desired. The third way is to start a line
(column #1) with a "*" character followed by any comment desired.

                         POINTER VARIABLES

The concept of pointer variables is foreign to BASIC (until now). Pointers
are a mainstay of the programming language "C". The challenge of adding
pointers to UniBasic was to provide the power of the concept but to do so
with a syntax that has a BASIC flavor rather than the cryptic style of "C".

Pointers are variables in their own right and are a special form of a 32 bit
unsigned integer which have a special property to index their value
according to the data type which they are declared to point to. For example
pointers can be assigned values and used in expressions just like the data
type long (integer). This allows the pointer to be set to any value from 0
to $ffffffff and this fact alone gives the pointer concept extreme power as
well as the power to cause a catastrophe if used carelessly. A more
"civilized" setting of a pointer would be to set it to the address of
another variable or to the address of a memory buffer (see BUFFER, BUFADR(),
BUFSIZ() section). When setting a pointer to another variable it is the data
type of the pointer which prevails when accessing data with the pointer (not
the data type of the variable being pointed to). Most "C" compilers give a
warning if a pointer is set to point to a variable with a different data
type than its own. UniBasic gives no such warning.

The functions ADDR() and INDEX() are intended solely for use with pointers
although the ADDR() function could be used to find the address of a variable
for any reason. Also the BUFADR() function is intended to set pointers to
the base address of memory buffers. The ADDR() function returns the memory
address of a variable given as the argument to the function. It is used
primarily to set a pointer to the absolute address of a variable. In the
case of an array, the address returned is the base address (address of the
first element of the array). The ADDR() function could also be used in
address calculations to determine the storage relationship of two or more
variables. To reference data pointed to by a pointer the pointer variable is
enclosed in square brackets i.e. [Ptr]. Data referenced by pointers must be
of a compatible type with the variable or constant used with it just as if
the data were to/from a standard variable (I.E. not pointed to by a
pointer). The INDEX function is used to point to different elements of the
object data (assuming the object is an array or is being treated as an
array). In "C" arithmetic operations on pointers take into account the data
type of the pointer. In UniBasic arithmetic operations on pointers do not
take the data type into account. The INDEX() function does take the
pointer's data type into account however. The arguments to the INDEX()
function are the pointer name and the index value. To index backward a
negative index value is specified. In the following example the INDEX()
function will be used on a pointer whose data type is INTEGER (32 bits or
4 bytes long). When we index the pointer by three it will point 12 bytes
past its previous setting (4 bytes times 3=12). The INC and DEC statements
also take into account the data size pointed to by a pointer. If Ptr is a
an INTEGER pointer then, INC Ptr will add 4 to the value of Ptr since an
INTEGER is 4 bytes in size.

 DIM Pointer:INTEGER ADDRESS
 DIM IntVar(30):INTEGER
 DIM Temp1:INTEGER
 DIM Temp2:INTEGER
Start
 Pointer=ADDR(IntVar) \ assigns base address of IntVar to Pointer
 Temp1=[Pointer] \ assigns the value of the first element to Temp1
 Pointer=INDEX(Pointer,3) \ Pointer now points to the fourth element
 Temp2=[Pointer] = assigns the value of the fourth element to Temp2
 DEC Pointer \ Pointer now points to the third element

                        COMPILER DIRECTIVES

                            #SET, #FIX

 The #SET compiler directive is used to define compiler variables and to set
 them to a numeric (long integer) value. The most common use of the compiler
 variable is to assign values to symbolic constants. Typically, since BASICs
 do not contain this feature, all constants must be literals (200, $0f,
 etc). Just as named variables such as Radius, Total, SubTotal, etc. are
 helpful and named labels such as GetInput, PrintList, etc. are more useful
 than line numbers, symbolic constants add much more meaning to the program
 than do abstract literal numbers. Any place that literal numeric constants
 are used (such as in expressions and DATA statements) symbolic constants
 may be used. The #Set directive may be used to set the same variable as
 many times as desired throughout the program. Thus when a compile variable
 is referenced the value last assigned by a #SET directive is the value
 referenced. Compiler variables are also used to control the flow of
 conditional compilation by being tested with the #IF compiler directive.
 The #FIX directive is a permanent #SET. The general form of the #SET
 compiler directive is as follows:

#SET CompVar=OPERAND1{OPERATOR|OPERAND2}

CompVar is the desired symbol name (the naming rules are the same as for
BASIC variable, labels, etc.). The name must be unique from any other
compiler variable, label, BASIC variable, macro, etc. or a symbol already
defined error will occur. The "=" is mandatory as is OPERAND1. OPERAND1 (and
OPERAND2 if used) must be either a compiler variable (previously created and
defined with a #SET directive) or a numeric literal (either a decimal
integer or Hexadecimal). Hexadecimal is indicated by a leading "$". The
OPERATOR must be a "+", "-", "*", "/","&", "|", "<<" or ">>" indicating
addition, subtraction, multiplication, division, AND, OR, left shift, or
right shift respectively. If the OPERATOR is used it must be immediately
followed by OPERAND2. Examples of the use of #SET directive follow:

#SET xyz=1 \ creates the compiler variable xyz and sets it to 1
 A=A+xyz \ implied LET statement equivalent to A=A+1
#SET xyz=$ff \ changes xyz to equal Hexadecimal $ff (255)
Label DATA "test",xyz \ equivalent to Label DATA "test",255
#SET Side=4 \ creates the compiler variable Side and sets it to 4
 Area=Side*Side \ implied LET statement equivalent to Area=4*4
#SET xyz=2 \ now equals 2
#SET abc=xyz \ creates the compiler var abc and sets it to the value of xyz
#FIX xyz=xyz+abc \ now equals 4 (permanently)
#IF xyz=5 \ example of testing a compiler variable for conditional compilation

The SET directive may also be used in the body of a macro declaration such as:

#MACRO MacroName \ declares a macro named MacroName
#SET CompVar=~0
 DATA CompVar
#ENDM \ defines end of macro declaration

MacroName
MacroName

Results in the following source code generation:

 DATA 1
 DATA 2

                          #IF, #ELSE, #ENDIF

These compiler directives are used to provide conditional compilation which
is a feature commonly found in assemblers and "C" compilers, but rarely in
Basic. Conditional compilation is useful when it is necessary to maintain
more than one version of a program which has many common sections but some
different sections. Conditional sections may be nested up to 20 deep.

The argument to the #IF directive is a simple condition of equality. The
general syntax is as follows:

 #IF CompVar=Const

CompVar is a compiler variable which my be declared by a #SET or #FIX
directive or on the compiler command line. Const is a constant which my be a
decimal or hexadecimal integer value or another compiler variable.

The most common way to use this feature is to assign a compiler variable at
the command line. For example:

cb testprog c=1

The #ELSE directive is optional and follows the body of code after the #IF
directive.

The #ENDIF directive ends the conditional section.

An example follows which would go with the command line example above:

#IF c=1
print "c=1"
#ELSE
print "c<>1"
#ENDIF

The above example is of no practical use, but shows the effect of conditional
compilation.

Also note that macros may be declared inside conditional sections which
effectively offers conditional macros. This is especially useful in writing
I/O drivers.

                              #C, #ENDC

These directives allow the use of C code blocks within a program or function.
After the #C directive, all input is simply passed through to the output
file until a #ENDC directive is encountered. C code may also be used on a
single line basis by writing a ";" in column 1. See section on C Code.

                             #MACRO, #ENDM

The #MACRO and #ENDM compiler directives are used to declare MACROS. The
purpose of macros is to allow the programmer to create complex blocks of
code and then use them as if they were a single instruction with a
descriptive name. At first glance the use of a macro might appear as a
subroutine. This is not the case, however, because a subroutine appears only
once in a program but may be called many times, while a macro is replicated
in the code each time it is used. Also the macro invocation allows
the specification of up to 9 arguments which are assigned to corresponding
variables in the macro. The general form of the macro declaration and
invocation are as follows:

Declaration:

#MACRO MacroName
 {body of macro}
#ENDM

Invocation:

 MacroName Arg#1,Arg#2,Arg#3,Arg#4,Arg#5,Arg#6,Arg#7,Arg#8,Arg#9

The macro declaration begins with the #MACRO compiler directive. All compiler
directives must have a "#" in character position #1 (no indentation
allowed). The body of the macro may contain no tilde characters (~) except
where an argument is being specified. The body of the macro willwhen
invoked be expanded to enter the source input stream, so the syntax within
the body must conform to compiler syntax rules. Any occurrence of the tilde
character without a single digit 0-9 immediately following will result in an
invalid macro argument error. Any occurrence of the tilde character
immediately followed by a single digit (0-9) will be replaced when the
macro is invoked with the corresponding invocation argument or in the case
of ~0 with the macro invocation number. Macros may NOT be nested. This means
that a macro declaration (#MACRO) may not appear in the body of a macro.
This will result in a nested macro error. The macro declarations are stored
in a fixed size macro buffer (20,000 bytes as of this writing) and if this
limit is exceeded a macro buffer full error will occur. The macro
declaration is terminated by an #ENDM compiler directive. An #ENDM directive
without a prior #MACRO directive will result in an #ENDM without #MACRO
error. If a macro is declared which already exists the result will be a
macro already defined error.

The macro invocation begins with the MacroName followed by a space followed
by up to 9 macro arguments. If the macro definition has no arguments then
none should be supplied in the invocation (the macro is invoked with the
MacroName only). The MacroName must NOT begin in character position #1 as it
will be taken for a Label. If there are less arguments in the macro
invocation than are used in the macro declaration a macro argument error
will occur. Macro invocation arguments are separated by a comma. NULL
arguments are specified by adjacent commas (I.E. no argument). Alpha-Numeric
strings (including "_","@", and "$") are valid arguments as are single
special characters such as "!", "#","%", etc. The SPACE and BACKSLASH
characters are NOT permitted. Quoted strings are passed as arguments with
the quotes themselves included in the argument.

Macros must be declared before they are invoked or the invocation will result
in an undefined symbol error.

Macro names must not conflict with other symbol names such as variables,
labels, compiler variables, etc. or a symbol already defined error will
occur.
#MACRO, #ENDM (cont)

The following example illustrates a simple macro declaration and invocation:

#MACRO GetName \ declares a macro named "GetName"
PRINT "Enter Name"
 INPUT ~1
 PRINT ~2
#ENDM \ defines end of macro declaration

 GetName UserName,"Thank You"

The invocation of the macro (GetName UserName) causes the generation of the
following lines in the source code stream:

 PRINT "Enter Name"
 INPUT UserName
 PRINT "Thank You"

The tilde character in the declaration followed by the numeral 1 indicates a
macro argument which is assigned a value during invocation. In this example
it is assigned "UserName" because it is argument #1 and UserName is the
first invocation argument. Accordingly the second argument (~2) is assigned
the quoted string "Thank You". Each macro may have up to 9 such arguments
which may appear any number of times in the macro declaration.

A special argument (~0) is also available in the declaration which is
replaced by the invocation number. For example, the first time a macro is
invoked the ~0 will be replaced with a "1", the second time with a "2" and
so on. This function is particularly useful in generating labels within a
macro. For Example:

#MACRO Demo \ declares a macro named "Demo"
Label~0 A=A+1
 IF A<200 THEN Label~0
#ENDM \ defines end of macro declaration

The first invocation of the Demo macro causes the generation of the following
lines in the source code stream:

Label1 A=A+1
 IF A<200 then Label1

The seventh invocation of the Demo macro causes the generation of the
following lines in thes ource code stream:

Label7 A=A+1
 IF A<200 THEN Label7

For the advanced user, the macro declaration body may contain a series of C
statements (or BASICstatements or both) with the effect of implementing
new statements not included in the language. This practice will eliminate
much of the "trickery" and "cheat coding" often found in BASIC programs and
result in much cleaner and readable code with faster execution speed. The
user may accumulate a MACRO library through freeware, shareware, and
purchased software.

             #USE, #SBRTN, #LIB, #FCN #STRUCT, #ENDSTRUCT

The #USE compiler directive is used to incorporate other files into the
source data stream. It is equivalent to the "include" directive in "C" and
is especially useful in declaring macros and functions. The USE directive
may be nested up to 20 deep. This is to say that a file called by #USE may
have a #USE directive itself. The general syntax of the #USE directive is as
follows:

#USE PathList

PathList is any legal path list to a file to be included in the source data
stream.

The #SBRTN compiler directive may be used to declare a subroutine. If the
subroutine is referenced by a GOSUB before it appears, no declaration is
required. even when not required, an ordered list of subroutines at the
beginning of a program is useful.

#SBRTN PathList

The #LIB compiler directive is used to declare any library files which may
contain relocatable files required by the program. Library files are merely
a collection of relocatable modules. The #LIB directive may also be used to
invoke C compiler options. The general syntax is as follows:

#LIB PathList

The #FCN compiler directive is used to declare functions. The name of the
function is declared as well as the number of arguments.The general syntax
is as follows:

#FCN FcnName(NumArgs)

NumArgs is a simple integer such as 1,2,3..etc. (or none)

The #STRUCT and #ENDSTRUCT are used to declare data structures just as the
TYPE statement does. The only difference is#STRUCT directive allows
declarations that are not limited by line length and the structure is more
clear to look at. For example:

#STRUCT UsrType
 Name:STRING[20}
 Address:STRING[30]
 ZipCode:BYTE
#ENDSTRUCT

The above example is equivalent to:

 TYPE UsrType=Name:STRING[20];Address:STRING[30];ZipCode:BYTE



******************** Important! READ THIS: *******************

Subroutines in UniBasic are actually void C functions. Unless the subroutine
is referenced by a GOSUB before it appears, the compiler will "think" it is
a label rather than a subroutine. Therefore a subroutine must be referenced
before it appears or, alternatively it may be declared with a #SBRTN
directive. Also the last statement of a subroutine must be a RETURN. Any
RETURN statement before the end of a subroutine must be inside a control
structure such as a FOR loop or WHILE/ENDWHILE block. Violations of these
rules will result in a "nesting" error.

The RETURN statement transfers control back to the caller and is also used to
return integer or floating point data to the caller in the case of integer
or floating point functions. For example:

 RETURN x+y

                           DIRECTIVES

#SET            #SET CompVar=Const
                #SET CompVar=Const+Const
                #SET CompVar=Const-Const
                #SET CompVar=Const*Const
                #SET CompVar=Const/Const

#FIX            #FIX CompVar=Const\ same as #SET but is permanent

#USE            #USE PathName

#IF                #IF CompVar=Const\ conditional compile
#ELSE            #ELSE\ conditional compile
#ENDIF            #ENDIF\ conditional compile

#MACRO            #MACRO MacroName
                Macro Body
#ENDM           #ENDM

#FCN            #FCN (NumArgs) <return data type> <modifier(s)>
                (NumArgs is number of arguments)
                (if NumArgs="c", NumArgs is variable)
                (return data type may be INT (default), FLOAT, or VOID)
                (modifiers may be POINTER, GLOBAL, and EXTERNAL)

#ROF            #ROF PathList\ relocatable file for linker

#LIB            #LIB PathList\ library file(s) for linker

#STRUCT         #STRUCT StructName\ same function as TYPE statement

#ENDSTRUCT      #ENDSTRUCT\ terminates structure declaration

#C              #C\ marks the beginning of a C code block

#ENDC           #ENDC\ marks the end of a C code block

#SBRTN          #SBRTN\ declares a subroutine

                        STATEMENTS

IF             IF CondExp THEN LABEL
            or
            IF CondExp THEN
                code body
            ENDIF
            or
            IF CondExp THEN
                code body
            ELSE
                code body
            ENDIF

WHILE         WHILE CondExp DO
                code body
            ENDWHILE

REPEAT         REPEAT
                code body
            UNTIL CondExp

LOOP         LOOP
                code body
            ENDLOOP

LOOP         LOOP COUNT=IntExp
                code body
            ENDLOOP

FOR         FOR Int=IntExp TO IntExp {STEP IntExp}
                code body
            Next Int

EXITIF         EXITIF CondExp <THEN>
            Falls thru to end of any of above structures

GOTO         GOTO Label
            GOTO [IntVar]

GOSUB         GOSUB Label
            GOSUB [IntVar]

RETURN         RETURN
            Return from subroutine or function

ON             ON Int GOTO
            ON Int GOSUB
            ON ERROR GOTO
            ON ERROR

STATUS:     STATUS \ optional prefix to any input/output statement
            which turns off error trapping

STATUS      STATUS\ contains status of last input/output operation
            -1 indicates error

READ         READ VAR,VAR...,VAR\ reads data statements
            READ #PathNum,StringVar

WRITE         WRITE #PathNum,StringVar

GET         GET #PathNum,Var\ can be simple var or whole array

PUT         PUT #PathNum,Var\ can be simple var or whole array

BUFREAD     BUFREAD #PathNum,Addr,Bytes

BUFWRITE     BUFWRITE #PathNum,Addr,Bytes

PRINT         PRINT Expr
            PRINT #PathNum,Expr
            PRINT USING(FormatString,Var,Var,...,Var)
            PRINT #PathNum USING(FormatString,Var,Var,...,Var)
            <<<<<<< left justified format field of 7 positions
            >>>>>> right justified format field of 6 positions
            ^^^^^ center justified format field of 5 positions

INPUT         INPUT Var
            INPUT "prompt",Var
            INPUT #PathNum,Var

FPRINT         Same as PRINT but sends no Line Feed

FINPUT         Same as input but sends no prompt

CREATE         CREATE #ByteVar,PathName     AccMode

OPEN         OPEN #ByteVar,PathName     AccMode

CLOSE         CLOSE #PathNum

SEEK         SEEK #PathNum,IntVal

DELETE         DELETE StringVal

LET         {LET} Var=Expr

POKE         POKE Addr,ByteVal

END         END

REM         REM\ Remark

SHELL         SHELL String

RESTORE     RESTORE Label

DATA         DATA Const,Const...,Const

BUFFER         BUFFER BufName=MemBytes\ opens buffer to size=MemBytes
            BUFFER BufName=0\ closes buffer
            (MemBytes=-1 opens largest buffer available)

DIM         DIM VarName[(ArraySize)}     DataType {modifier modifier}
            DataTypes      BOOLEAN, BYTE, SHORT, LONG, FLOAT, STRING,POINTER
            Modifiers      GLOBAL, EXTERNAL, ALIAS=, ADDRESS(POINTER),
            REMOTE, REGISTER, VECTOR
            DIM StringVarName{(ArraySize)}     STRING[StringSize] {modifier}

PARAM       Similar to DIM, immediately follows a function occurence
            with one PARAM to receive each argument to the function

TYPE         TYPE UserType=VarName{(ArraySize)}     DataType;VarName...etc.

INC         INC IntVar\ IntVar=IntVar+1
            INC PtrVar\ PtrVar=PtrVar+DataSize

DEC         DEC IntVar\ IntVar=IntVar-1
            DEC PtrVar\ PtrVar=PtrVar-DataSize

DIGITS         DIGITS NumConst\ # of digits for FLOAT (1-15) default 15

DECIMALS     NumConst\ # of decimal places for FLOAT (default 2)

BASE         BASE 0, BASE 1\ base of arrays (default  1)

ENDEXIT     ENDEXIT\ Terminates EXITIF CondExp THEN

SETVEC         SETVEC VectorVar=Addr\ Sets vector to address

EXCHANGE     EXCHANGE Var,Var\ vars must be same type and size

                FUNCTIONS and SPECIAL VARIABLES

EOF()            EOF(#PathNum)\ BOOLEAN result
LEN()            LEN(String)\ INT result
STR$()            STR$(NumVal)\ String result
ZSTR$()            ZSTR$(NumVal)\ String result zero padded
VAL()            VAL(StringVal)\ INT or FLOAT result
IVAL()          IVAL(StringVal)\ INT RESULT
FVAL()          FVAL(StringVal)\ FLOAT result
LAND()            LAND(IntVAl)\ INT result
LOR()            LOR(IntVal)\ INT result
LXOR()            LXOR(IntVal)\ INT result
LNOT()            LNOT(IntVal)\ INT result
MID$()            MID$(StringVal,Org,Bytes)\ String result
LEFT$()            LEFT$(StringVal,Bytes)\ String result
RIGHT$()        RIGHT$(StringVal,Bytes)\ String result
MOD()            MOD(IntVal,IntVal)\ INT result
CHR$()            CHR$(IntVal)\ String result
ASC()            ASC(StringVal)\ INT result
NOT()            NOT(BoolVal)\ Boolean result
BUFADR()        BUFADR(BufName)\ INT result (address of BufName)
BUFSIZ()        BUFSIZ(BufName)\ Int result (size of BufName)
PEEK()            PEEK(Addr)\ Byte result
ERR                ERR\ INT result
FILSIZ()        FILSIZ(#PathNum)\ INT result
POS                POS\ Int result
FILPOS()        FILPOS(#PathNum)\ INT result
INDEX()            INDEX(PtrVar,IndexVal)\ INT result
SIZE()            SIZE(VarName)\ INT result
TAB()            TAB(IntVal)\ INT result
ADDR()            ADDR(VarName)\ INT result
NextArg            StringVar=NextArg\    must be used in LET statement
                (Do once to get each arg)
                (When LEN(StringVar)=0 then all command line
                arguments are exhausted)
XferBytes        Total bytes transfered after I/O operation
THOUR()         THOUR(0)\ Hour BYTE result
TMIN()          TMIN(0)\ Minutes BYTE result
TSEC()          TSEC(0)\ Seconds BYTE result
TYEAR()         TYEAR(0) \ Year BYTE result
TMON()          TMON(0)\ Month BYTE result
TDAT()          TDAT(0)\ Date BYTE result
TDAY()          TDAY(0)\ DAY BYTE result 0=Sunday, 1=Monday, etc.
SQR()           SQR(FloatVal)\ Square Root FLOAT result
ABS()           ABS(NumExpr)\ Absolute Value INT or FLOAT result
INT()           INT(FloatVal)\ Integer Value of Float
SIN()           SIN(NumVal)\ FLOAT result
COS()           COS(NumVal)\ FLOAT result
TAN()           TAN(NumVal)\ FLOAT result
ASN()           ASN(NumVal)\ FLOAT result
ACS()           ACS(NumVal)\ FLOAT result
ATN()           ATN(NumVal)\ FLOAT result
LOG()           LOG(NumVal)\ FLOAT result
LOG10()         LOG10(NumVal)\ FLOAT result
HEX$()            HEX$(NumVal)\ String result
ZHEX$()            ZHEX$(NumVal)\ String result zero padded
BITTST()        BITTST(IntVal,IntConst)\ BOOLEAN result
BITCLR()        BITCLR(IntVal,IntConst)\ INT result
BITSET()        BITSET(IntVal,IntConst)\ INT result
BITCHG()        BITCHG(IntVal,IntConst)\ INT result
SHL()            SHL(IntVal,IntConst)\ INT result
SHR()            SHR(IntVal,IntConst)\ INT result
ROL()            ROL(IntVal,IntConst)\ INT result
ROR()            ROR(IntVal,IntConst)\ INT result
SUBSTR()        SUBSTR(Str1,Str2)\ Int result position of Str1 in Str2