mathops Module

Liviu Chircu

OpenSIPS Solutions

Edited by

Liviu Chircu

Edited by

Stephane Alnet

Revision History
Revision $Revision$$Date$

Table of Contents

1. Admin Guide
1.1. Overview
1.2. Dependencies
1.2.1. OpenSIPS Modules
1.2.2. External Libraries or Applications
1.3. Exported Parameters
1.3.1. decimal_digits (integer)
1.4. Exported Functions
1.4.1. math_eval(expression, result_pvar)
1.4.2. math_rpn(expression, result_pvar)
1.4.3. math_trunc(number, result_pvar)
1.4.4. math_floor(number, result_pvar)
1.4.5. math_ceil(number, result_pvar)
1.4.6. math_round(number, result_pvar[, decimals])
1.4.7. math_round_sf(number, result_pvar, figures)

List of Examples

1.1. Setting the decimal_digits module parameter
1.2. math_eval usage
1.3. math_rpn usage
1.4. math_trunc usage
1.5. math_floor usage
1.6. math_ceil usage
1.7. math_round usage
1.8. math_round_sf usage

Chapter 1. Admin Guide

1.1. Overview

The mathops module provides a series of functions which enable various floating point operations at OpenSIPS script level.

1.2. Dependencies

1.2.1. OpenSIPS Modules

The following modules must be loaded before this module:

  • No dependencies on other OpenSIPS modules..

1.2.2. External Libraries or Applications

The following libraries or applications must be installed before running OpenSIPS with this module loaded:

  • None.

1.3. Exported Parameters

1.3.1. decimal_digits (integer)

The precision of the results returned by all the module functions. The higher the “decimal_digits” value, the more decimal digits the results will have.

Default value is “6”.

Example 1.1. Setting the decimal_digits module parameter

modparam("mathops", "decimal_digits", 10)

1.4. Exported Functions

1.4.1.  math_eval(expression, result_pvar)

The function evaluates a given expression and writes the results in the output pseudo-variable. The expression may contain any number of pseudo variables.

Currently allowed syntax for specifying an expression:

  • Nested parantheses

  • binary + - / * operators

Meaning of the parameters is as follows:

  • expression - String containing a mathematical expression. It can also include pseudo variables. The expression parameter can only be given as a string.

  • result_pvar - pseudo-variable which will hold the result of the evaluation. Specified as a quoted string.

This function can be used from any route.

Example 1.2. math_eval usage

...
# Compute some random math expression

$avp(1) = "3.141592";
$avp(2) = "2.71828";
$avp(3) = "123.45678";

if (math_eval("$avp(1) * ($avp(3) - ($avp(1) - $avp(2))) / $avp(3)", "$avp(result)")) {
	xlog("Result of expression: $avp(result)\n");
} else {
	xlog("Math eval failed!\n");
}

...

1.4.2.  math_rpn(expression, result_pvar)

The function evaluates a given RPN expression and writes the results in the output pseudo-variable. The expression may contain any number of pseudo variables.

The expression is specified as a Reverse Polish Notation script. Values are pushed onto a stack, while operations are executed on that stack. The following operations are supported:

  • binary operators: + - / * mod pow

  • unary functions: neg exp ln log10 abs sqrt cbrt floor ceil round nearbyint trunc

    neg will change the sign of the top of the stack

    ln is natural logarithm; abs is absolute value; other functions are standard C functions

  • constants: e pi

  • stack manipulations commands: drop dup swap

Meaning of the parameters is as follows:

  • expression - String containing a RPN expression. It can also include pseudo variables. The expression parameter can only be given as a string.

  • result_pvar - pseudo-variable which will hold the result of the evaluation. Specified as a quoted string.

This function can be used from any route.

Example 1.3. math_rpn usage

$avp(1) = "3";

if (math_rpn("1 $avp(1) swap swap dup drop / exp ln 1 swap /", "$avp(result)")) {
	xlog("Result of expression: $avp(result)\n");
} else {
	xlog("RPN eval failed!\n");
}

/* This example RPN script will push 1 then 3 onto the stack, then do a couple no-ops
(exchange the two values twice, duplicate one of them then drop the duplicate),
compute the division of 1 by 3, then do another no-op (exponentiation then logarithm), and
finally compute 1 divided by the result, giving 3 as the result. */

1.4.3.  math_trunc(number, result_pvar)

Truncation of a number towards zero. This means that trunc(3.7) = 3.0 and trunc(-2.9) = -2.0.

Meaning of the parameters is as follows:

  • number - Number to be truncated. The number parameter can have the following types:

    • string - statically given

    • pvar - value of an existing pseudo-variable (as string value - it makes no sense to truncate integers)

  • result_pvar - pseudo-variable which will hold the result of the evaluation. Specified as a quoted string.

This function can be used from any route.

Example 1.4. math_trunc usage

...
# Truncate a random number

$avp(1) = "3.141492";

if (math_trunc("$avp(1)", "$avp(result)")) {
	xlog("Truncate result: $avp(result)\n");
} else {
	xlog("Truncate failed!\n");
}
...

1.4.4.  math_floor(number, result_pvar)

Truncates a number, always towards -infinity. This means that floor(3.7) = 3.0 and floor(-2.9) = -3.0

Meaning of the parameters is as follows:

  • number - Number to be truncated. The number parameter can have the following types:

    • string - statically given

    • pvar - value of an existing pseudo-variable (as string value - it makes no sense to truncate integers)

  • result_pvar - pseudo-variable which will hold the result of the evaluation. Specified as a quoted string.

This function can be used from any route.

Example 1.5. math_floor usage

...
# Truncate a random number

$avp(1) = "3.141492";

if (math_floor("$avp(1)", "$avp(result)")) {
	xlog("Floor result: $avp(result)\n");
} else {
	xlog("Floor operation failed!\n");
}
...

1.4.5.  math_ceil(number, result_pvar)

Truncates a number, always towards +infinity. This means that ceil(3.2) = 4.0 and ceil(-2.9) = -2.0

Meaning of the parameters is as follows:

  • number - Number to be truncated. The number parameter can have the following types:

    • string - statically given

    • pvar - value of an existing pseudo-variable (as string value - it makes no sense to truncate integers)

  • result_pvar - pseudo-variable which will hold the result of the evaluation. Specified as a quoted string.

This function can be used from any route.

Example 1.6. math_ceil usage

...
# Truncate a random number

$avp(1) = "3.141492";

if (math_ceil("$avp(1)", "$avp(result)")) {
	xlog("Ceil result: $avp(result)\n");
} else {
	xlog("Ceil operation failed!\n");
}
...

1.4.6.  math_round(number, result_pvar[, decimals])

The round function returns the nearest integer, and tie-breaking is done away from zero. Examples: round(1.2) = 1.0, round(0.5) = 1.0, round(-0.5) = -1.0

By default, the function returns an integer. An additional parameter controls the number of decimal digits of the initial number which will be kept. The rounding will then be done using the remaining decimal digits, and the result will be a float value, represented as a string.

Meaning of the parameters is as follows:

  • number - Number to be rounded. The number parameter can have the following types:

    • string - statically given

    • pvar - value of an existing pseudo-variable (as string value - it makes no sense to truncate integers)

  • result_pvar - pseudo-variable which will hold the result of the evaluation. Specified as a quoted string.

  • decimals - (pvar / integer as a string) which further improves the precision of the rounding.

This function can be used from any route.

Example 1.7. math_round usage

...
# Rounding PI

$avp(1) = "3.141492";

if (math_round("$avp(1)", "$avp(result)")) {

	# result should be: 3
	xlog("Round result: $avp(result)\n");
} else {
	xlog("Round operation failed!\n");
}

...

if (math_round("$avp(1)", "$avp(result)", "4")) {

	# result should be: "3.1415"
	xlog("Round result: $avp(result)\n");
} else {
	xlog("Round operation failed!\n");
}
...

1.4.7.  math_round_sf(number, result_pvar, figures)

To give a simple explanation, rounding to N significant figures is done by first obtaining the number resulted from keeping N significant figures (0 padded if necessary), then adjusting it if the N+1'th digit is greater or equal to 5.

Some examples:

  • round_sf(17892.987, 1) = 20000

    round_sf(17892.987, 2) = 18000

    round_sf(17892.987, 3) = 17900

    round_sf(17892.987, 4) = 17890

    round_sf(17892.987, 5) = 17893

    round_sf(17892.987, 6) = 17893.0

    round_sf(17892.987, 7) = 17892.99

Meaning of the parameters is as follows:

  • number - Number to be rounded. The number parameter can have the following types:

    • string - statically given

    • pvar - value of an existing pseudo-variable (as string value - it makes no sense to truncate integers)

  • result_pvar - pseudo-variable which will hold the result of the evaluation. Specified as a quoted string.

  • figures - (pvar / integer as a string) which further improves the precision of the rounding.

This function can be used from any route.

Example 1.8. math_round_sf usage

...
# Rounding PI

$avp(1) = "3.141492";

if (math_round_sf("$avp(1)", "$avp(result)", "4")) {

	# result should be: "3.141"
	xlog("Round result: $avp(result)\n");
} else {
	xlog("Round operation failed!\n");
}

...