Rediģēt

Kopīgot, izmantojot


Add arithmetic and logical operators to rules using Microsoft Rules Composer (Preview)

Applies to: Azure Logic Apps (Standard)

Important

This capability is in preview and is subject to the Supplemental Terms of Use for Microsoft Azure Previews.

This guide describes how to add arithmetic and logical operators to the rules in your ruleset using the Microsoft Rules Composer.

Prerequisites

Add an arithmetic operator to a rule

You can add an arithmetic operator to a condition or action in a rule. The following table describes the available arithmetic operators:

Arithmetic operator Description
Add The addition operator that adds arg1 to arg2.
Subtract The subtraction operator that subtracts arg1 from arg2.
Multiply The multiplication operator that multiplies arg1 by arg2.
Divide The division operator that divides arg1 by arg2.
Remainder The remainder operator that performs arg1 modulo arg2.
  1. In the Microsoft Rules Composer, load the XML file that contains the rule store you want to work on.

  2. In the RuleSet Explorer window, find and select the rule that you want.

  3. In the Facts Explorer window, select the Vocabularies tab.

  4. Under Vocabularies > Functions > Version 1.0, drag the arithmetic operator that you want to an argument in a condition in the conditions editor or an action in the actions editor.

  5. In the condition or action arguments, specify the values for left and right operands.

    • If the operands have different types, the rules engine performs automatic numeric promotion where the engine converts the smaller operand type to the larger operand type.

      For example, suppose you use the Add operator on an operand that has int type and an operand that has long type. Before the engine performs the Add operation, the engine converts the int type to the long type.

    • If the engine can promote both operands to a common type, the engine supports double promotion.

      For example, suppose you use the Add operator on an operand that has int type and an operand that has uint type. Before the engine performs the Add operation, the engine converts both operand types to the long type.

Add a logical operator to a rule

You can add a logical operator to a predicate in a condition. The following table describes the available logical operators:

Logical operator Description
AND Combine two or more predicates to form a logical AND expression. Returns true if both predicates evaluate to true. Otherwise, returns false.
OR Combine two or more predicates to form a logical OR expression. Returns true if one predicate evaluates to true. Otherwise, returns false.
NOT Negate a logical expression or predicate. Returns true if the predicate evaluates to false. Otherwise, returns false.
  1. In the Microsoft Rules Composer, load the XML file that contains the rule store you want to work on.

  2. In the RuleSet Explorer window, find and select the rule that you want.

  3. In the IF pane, which is the conditions editor, from the Conditions shortcut menu, select one of the following commands:

    Logical operator Description
    Add logical AND Combine two or more predicates to form a logical AND expression.
    Add logical OR Combine two or more predicates to form a logical OR expression.
    Add logical NOT Negate a logical expression or predicate.
  4. In the conditions editor, open the operator's shortcut menu, and add the predicates or nested logical operators that you want.

    If the operands have different types, the rules engine converts the type for one operand to match the type for the other operand, or converts the types for both operands to a common type before evaluating the expression.

Handle null values

The following section describes the expected behaviors for null values associated with different types and provides options for checking null or the existence of a specific field or member.

.NET classes

  • For types derived from the Object type, you can set their fields to null.

  • You can pass null as an argument for parameters that aren't value types, but you might get a runtime error, based on the member's implementation.

  • You can't use null for comparison if the return type isn't an Object type.

XML elements

  • An XML document never returns an XML value as null. Instead, this value is either an empty string or a "doesn't exist" error. For an empty string, an error might occur for the conversion of certain types, such as fields specified as an integer type when you build a rule.

  • The Microsoft Rules Composer doesn't allow you to set a field to null or to set a field type to Object.

  • Through the object model, you can set the type to Object. In this case, the returned value has the type to which the XPath evaluates, such as Float, Boolean, or String, based on the XPath expression.

Check for null or existence

When you write rules, you naturally want to check that a field exists before you compare its value. However, if the field is null or doesn't exist, comparing the value causes an error.

For example, suppose you have the following rule:

IF Product/Quantity Exists AND Product/Quantity > 1

If Product/Quantity doesn't exist, the rule throws an error. To work around this problem, you can pass a parent node to a helper method that returns the Product/Quantity value if that element exists, or return something else if that element doesn't exist.

The following example shows the updated and new helper method rule:

Rule 1

IF Exists(Product/Quantity) THEN Assert(CreateObject(typeof(Helper), Product/Quantity))

Rule 2

IF Helper.Value == X THEN...

As another possible solution, you can create a rule such as the following example:

IF Product/Quantity Exists THEN CheckQuantityAndDoSomething(Product/Quantity)

In the preceding example, the <CheckQuantityAndDoSomething> function checks the parameter value and executes if the condition is met.

Note

Alternatively, you can modify the XPath Field property for the XML fact to catch any errors, but this approach isn't recommended.