SoCalculator Class Reference
[Engines]

A general-purpose calculator. More...

#include <Inventor/engines/SoCalculator.h>

Inheritance diagram for SoCalculator:

Public Member Functions
virtual SoType	getTypeId () const
	SoCalculator ()
Static Public Member Functions
static SoType	getClassTypeId ()
Public Attributes
SoMFFloat	a
SoMFFloat	b
SoMFFloat	c
SoMFFloat	d
SoMFFloat	e
SoMFFloat	f
SoMFFloat	g
SoMFFloat	h
SoMFVec3f	A
SoMFVec3f	B
SoMFVec3f	C
SoMFVec3f	D
SoMFVec3f	E
SoMFVec3f	F
SoMFVec3f	G
SoMFVec3f	H
SoMFString	expression
SoEngineOutput	oa
SoEngineOutput	ob
SoEngineOutput	oc
SoEngineOutput	od
SoEngineOutput	oA
SoEngineOutput	oB
SoEngineOutput	oC
SoEngineOutput	oD

Detailed Description

A general-purpose calculator.

This engine is a general-purpose calculator. The calculator operates on floating-point values and 3D floating-point vectors. The engine takes up to eight inputs of each type (SoMFFloat and SoMFVec3f), and produces up to four outputs of each type.

Each input field (a-h, A-H) can have multiple values, allowing the engine to evaluate the expression with different values in parallel. Some inputs may have more values than others. In such cases, the last value of the shorter inputs will be repeated as necessary.

The expression input string specifies the expression to be evaluated. An expression can consist of multiple subexpressions. Several subexpressions can be specified in one string, separated by semicolons (;). Alternatively, the subexpressions can be stored in separate strings in the multiple-valued input field.

Each subexpression is of the form:

      \<lhs\> = \<rhs\>

The <lhs> can be any one of the outputs or a temporary variable. The engine provides 8 temporary floating-point variables (ta, tb, tc, td, te, tf, tg, and th), and 8 temporary vector variables (tA, tB, tC, tD, tE, tF, tG, and tH). You can assign a value to one component of a vector output (A-H) or a vector variable ( tA - tH ) by using the [ ] operator. For example, oA[0] = <rhs>, will evaluate the right hand side and assign the value to the first component of the output vector oA.

The <rhs> supports arithmetic, logical and conditional operators. They are:

      (unary) !, -
      (binary) +, -, *, /, \%, <, > <=, >=, ==, !=, &&, ||
      (ternary) ? :

The ternary operator is a conditional operator. For example, a ? b : c evaluates to b if a != 0, and to c if a==0.

Valid operands for the <rhs> include the inputs, outputs, temporary variables, and their components (e.g. oA[0]). Operands can also be numeric constants (e.g. 1.0), pre-defined named constants, or pre-defined functions.

The named constants are:

      MAXFLOAT
      MINFLOAT
      M_E
      M_LOG2E
      M_LOG10E
      M_LN2
      M_LN10
      M_PI
      M_SQRT2 = sqrt(2)
      M_SQRT1_2 = sqrt(1/2)

Most of the pre-defined functions come from the math library:

      cos, sin, tan,
      acos, asin, atan, atan2,
      cosh, sinh, tanh,
      sqrt, pow, exp, log, log10,
      ceil, floor, fabs, fmod.

Other functions are defined by SoCalculator. They are:

      rand(f) - Random number generator
      cross(v1, v2) - Vector cross
     product
      dot(v1, v2) - Vector dot product
      length(v) - Vector length
      normalize(v) - Normalize vector
      vec3f(f1, f2, f3) - Generate a vector from 3 floats

The subexpressions are evaluated in order, so a variable set in the <lhs> of an earlier expression may be used in the <rhs> of a later expression.

Note, when the input has multiple values, all the subexpressions specified in the expression are applied to all the multiple input values. This is unlike the SoBoolOperation engine, where each operation is applied only to the corresponding entries of the input data. Note also, that even though the inputs and outputs can have multiple values the [ ] operator is only for indexing into the values of a single vector. It does not index into the multiple values of a field. For example, if the floating-point input field a has two values: 1.0, and 2.0, then the expression

      "oA[0]=a; oA[1]=a; oA[2]=0.0"

will produce two output vectors in oA: (1.0, 1.0, 0.0) and (2.0, 2.0, 0.0).

Examples of expressions:

      "ta = oA[0]*floor(a)"
      "tb = (a+b)*sin(M_PI)"
      "oA = vec3f(ta, tb, ta+tb)"
      "oB = normalize(oA)"
      "ta = a; tb = sin(ta); oA = vec3f(ta, tb, 0)"