Variables
Creating variables
Let us assume that our set of rules consists of the following two rules:
1
2
IF the age of the applicant is LESS than 18
3
AND his place of residence IS NOT Dortmund
4
THEN You must be at least 18 years old and live in Dortmund.
5
6
7
IF the age of the applicant is LESS than 18
8
OR his professional experience is SHORTER than 5 years
9
THEN You must be at least 18 years old and have work experience
10
of at least 5 years
11
Copied!
In both rules the query appears whether the user is already 18 years old. If we have 10, 20 or even 100 rules in which this query is needed, it becomes clear that such duplications unnecessarily inflate the clarity and complexity of the entire set of rules.
At this point, variables provide a remedy. We can store the condition with the age in our own variable:
1
the age of the applicant is LESS than 18 years old
2
AS Underage
Copied!
The variable has a variable indicator AS. Everything that precedes the
ASis the value of a variable. The text after the keyword ASis its name.Now that we have swapped the condition into a variable, we can use it in the appropriate places in our rules. And this is how our whole set of rules looks like:
1
2
the age of the applicant is LESS than 18 years old
3
AS underage
4
5
IF the applicant is underage
6
AND his residence IS NOT Dortmund
7
THEN You must be at least 18 years old and come from Dortmund.
8
9
10
IF the applicant is underage
11
OR his professional experience is LESS than 5 years
12
THEN You must be at least 18 years old and have work experience
13
of at least 5 years
14
Copied!
Remember to always define variables before you use them. This means that the definition of a variable always needs to be anywhere above every place it is used in.
A further advantage when using variables is the additional semantics or domain specificity. We have packaged a complex and somewhat formal expression behind a simple semantically named variable. The one who creates the rules can influence the readability of his rules at this point.
Data type of a variable
Variables not only consist of a value and a name, they also have a data type. This data type is automatically derived from the value of a variable. Let's take another look at the upper condition:
1
the age of the applicant is LESS than 18 years old
2
AS underage
Copied!
The value of this variable is a comparison operation. The result of a comparison operation is always true or false, i.e. a truth value. The data type of this variable is therefore boolean.
We can create variables with different data types, such as a numeric variable:
1
18
2
AS Number18
Copied!
We can create variables with arithmetic operations
1
20 - 18
2
AS professional experience
Copied!
Or a simple variable with a static text of type String
1
The fastest person in the world
2
AS Superman
Copied!
For example, variables can contain the following schema attributes:
YAML
JSON
1
age: 0
2
residence : "text"
Copied!
1
{
2
age: 0,
3
residence : "text"
4
}
Copied!
And here is the access to the attributes of the schema:
1
Age - 18
2
AS professional experience
3
4
Residence IS Cologne
5
AS Cologne Boy
Copied!
With the help of variables, complex rules can be simplified by outsourcing the frequently used or particularly complicated expressions. Thus, it is possible to give simple semantic names to complex formal expressions. This not only increases the readability of the rules, but also their maintainability.
Last modified 2yr ago
Copy link