.TH "cfloat" 3 "Nov 27, 1997"
.SH CLASS
.B
Float
.SH SUPERCLASS
CAObject
.SH MATURITY INDEX
Immature
.SH DESCRIPTION
.PP

It is planned to make this class evolve into an Objective C class for arbitrary precision floating-point arithmetic.  Currently, 
.B
Float 
is merely a Computer Algebra Kit compatible interface to plain C floating-point arithmetic.  Instances are object wrappers around the 
.B
float 
datatype; arithmetic is subject to the same limitations, round-off errors etc. as in C.
.PP

Support for floating-point arithmetic is minimal in the Computer Algebra Kit; it is possible to convert results that are obtained by exact arithmetic to a numerical approximation (see the method 
.B
asNumerical 
of Integer, Matrix, Polynomial etc.) for further processing by numerical analysis software.
.SH METHOD TYPES
.PP
.B
Creation
.RS 3
new
.br
newFloatValue:
.br
newStrValue:
.br
.RE
.PP
.B
Identity
.RS 3
hash
.br
isEqual:
.br
floatValue
.br
floatValue:
.br
.RE
.PP
.B
Comparing
.RS 3
sign
.br
compare:
.br
.RE
.PP
.B
Addition
.RS 3
zero
.br
isZero
.br
isOpposite:
.br
negate
.br
double
.br
add:
.br
subtract:
.br
.RE
.PP
.B
Multiplication
.RS 3
one
.br
minusOne
.br
isOne
.br
isMinusOne
.br
square
.br
multiply:
.br
inverse
.br
divide:
.br
.RE
.PP
.B
Expression Object
.RS 3
asExpression
.br
expression:
.br
.RE
.PP
.B
Printing
.RS 3
printsLeadingSign
.br
printToFile:
.br
.RE
.SH METHODS
.PP
.B
new
.RS 1
+
.B
new
.RE
.PP

Creates an floating-point object equal to zero.
.PP
.B
newFloatValue:
.RS 1
+
.B
newFloatValue:
(float)
.I
floatValue
.RE
.PP

Creates an object with value equal to 
.I
floatValue.
.PP
.B
newStrValue:
.RS 1
+
.B
newStrValue:
(STR)
.I
aString
.RE
.PP

Returns a new float object given a string 
.I
aString
, or returns 
.B
nil 
if the method can\&'t parse 
.I
aString.  
This method acts like the ANSI function 
.B
atof().
.PP
.B
hash
.RS 1
- (unsigned)
.B
hash
.RE
.PP

Returns a small integer that is the same for objects that are equal (in the sense of 
.B
isEqual:
).
.PP
.B
isEqual:
.RS 1
- (BOOL)
.B
isEqual:
.I
b
.RE
.PP
.B
floatValue
.RS 1
- (float)
.B
floatValue
.RE
.PP

Returns the value of the object.
.PP
.B
floatValue:
.RS 1
-
.B
floatValue:
(float)
.I
f
.RE
.PP

Creates an object with value 
.I
f.
.PP
.B
sign
.RS 1
- (int)
.B
sign
.RE
.PP

Returns plus one if the object is positive (greater than zero), zero if zero and minus one if negative (less than zero).
.PP
.B
compare:
.RS 1
- (int)
.B
compare:
.I
b
.RE
.PP

Returns -1, 0, or +1 if the object is respectively less than, equal to, or greater than 
.I
b.  
Returns 0 for pointer equal objects.
.PP
.B
zero
.RS 1
-
.B
zero
.RE
.PP

Returns a new reference to a shared float instance with value equal to zero.
.PP
.B
isZero
.RS 1
- (BOOL)
.B
isZero
.RE
.PP

Whether the object is equal to zero.
.PP
.B
isOpposite:
.RS 1
- (BOOL)
.B
isOpposite:
.I
b
.RE
.PP

Whether the object is the opposite of 
.I
b.
.PP
.B
negate
.RS 1
-
.B
negate
.RE
.PP

Returns the opposite of the object.
.PP
.B
double
.RS 1
-
.B
double
.RE
.PP

Returns a new object, equal to the object multiplied by two i.e., added to itself.
.PP
.B
add:
.RS 1
-
.B
add:
.I
b
.RE
.PP

Adds 
.I
b 
to the object.  Returns a new object.
.PP
.B
subtract:
.RS 1
-
.B
subtract:
.I
b
.RE
.PP

Subtracts 
.I
b 
from the object.  Returns a new object.
.PP
.B
one
.RS 1
-
.B
one
.RE
.PP

Returns a new reference to a shared float instance with value equal to one.
.PP
.B
minusOne
.RS 1
-
.B
minusOne
.RE
.PP

Returns a new reference to the float minus one.
.PP
.B
isOne
.RS 1
- (BOOL)
.B
isOne
.RE
.PP

Whether the object is equal to one.
.PP
.B
isMinusOne
.RS 1
- (BOOL)
.B
isMinusOne
.RE
.PP

Whether the object is equal to minus one.
.PP
.B
square
.RS 1
-
.B
square
.RE
.PP

Returns the square of the object i.e., a new object equal to the original object multiplied by itself.
.PP
.B
multiply:
.RS 1
-
.B
multiply:
.I
b
.RE
.PP

Returns a new object, the product of the object multiplied (to the right) by 
.I
b.  
If the objects are pointer equal, this method is equivalent to 
.B
square.
.PP
.B
inverse
.RS 1
-
.B
inverse
.RE
.PP

Returns the multiplicative inverse of the object (returns a new object).  Returns 
.B
nil 
if the object is equal to zero or cannot be inverted.
.PP
.B
divide:
.RS 1
-
.B
divide:
.I
b
.RE
.PP

Returns a new object, the 
.I
exact 
quotient of the object divided by 
.I
b.  
Returns 
.B
nil 
if 
.I
b 
is equal to zero or if the division is not exact.
.PP
.B
asExpression
.RS 1
-
.B
asExpression
.RE
.PP

Returns a new expression object, containing a copy of the receiving object.
.PP
.B
expression:
.RS 1
-
.B
expression:
.I
anExpression
.RE
.PP

Try to evaluate 
.I
anExpression 
as a floating-point number or return 
.B
nil 
otherwise (when 
.I
anExpression 
is symbolic, for example).
.PP
.B
printsLeadingSign
.RS 1
- (BOOL)
.B
printsLeadingSign
.RE
.PP

Returns YES if the number is negative.
.PP
.B
printToFile:
.RS 1
-
.B
printToFile:
(FILE *)
.I
aFile
.RE
.PP

Prints the floating-point number to 
.I
aFile 
according to the \&'%g\&' format of the ANSI function 
.B
printf().