The use of the four basic binary operators taking BigDecimal operand(s) causes a compilation error in executable model.

Issue

Building a rule which is described in MVEL dialect and uses one of the four basic binary operators (+, -, *, /) taking one or two BigDecimal type value(s) as operand(s) with executable rule models enabled causes a compilation error.

For example, in the following rule,

package com.example.reproducer

import com.example.reproducer.Bus
import java.math.BigDecimal

dialect "mvel"

rule "bus3b"
    when
        $bus : Bus()
    then
        BigDecimal grossWeight = $bus.weight + $bus.capacity * 55 + $bus.maxLoadingWeight;    // ..... (*1)
        System.out.println("***** " + $bus + ", gross weight = " + grossWeight);
end

// $bus.weight and $bus.maxLoadingWeight are BigDecimal type properties, and $bus.capacity is an int type property of Bus class.

the expression $bus.weight + $bus.capacity * 55 in the statement (*1) performs an addition of BigDecimal type value ($bus.weight) and numeric value (the result of $bus.capacity * 55).

When I build a maven project which contains this rule with executable model enabled, a compilation error occurs with the message like below.

$ mvn clean compile -DgenerateModel=YES
...
[ERROR] /work2/testdir/reproducer_model_compiler_3b/target/generated-sources/drools-model-compiler/main/java/com/example/reproducer/P3B/LambdaConsequence3B6B2D15753A225FFA39E42DED953773.java:[18,61] bad operand types for binary operator '+'
  first type:  java.math.BigDecimal
  second type: int
...

This issue occurs for the following operations:

• BigDecimal_type_value op Numeric_value
• Numeric_value op BigDecimal_type_value
• BigDecimal_type_value op BigDecimal_type_value

where op is one of the binary operators +, -, *, /.