jep-data-extension is a project that extends JEP, the powerful mathematical expression parser and evaluator for Java, introducing data manipulation capabilities with custom functions and operators ready to use, including RegEx, XPath, JSONPath and Web Services.

For details about JEP core features, access JEP documentation.

ℹ️ Find examples in the wiki.

JEP offers a good set of operators for common arithmetic and logical operations, for example: addition (+), subtraction (-), multiplication (*), division (/), modulus (%), power (^), "Boolean And" (&&), "Boolean Or" (||), and "Boolean Not" (!). These operators can be used when required and are supported by JEP.

Standard JEP relational operators accept only numerical variables (including string representation of non-complex, numerical values). These operators have been overloaded to allow comparison of dates, as well as string representations of valid dates:

• Less than (<)
• Less than or equal to (<=)
• Greater than (>)
• Greater than or equal to (<=)
• Equal (==)
• Not equal (!=)

This allows the usage of dates in expressions that can be evaluated to true or false. For example:

if("2017-03-11T10:15:00:123Z" < now(), "past", "not past")

Standard "element" operator ([]) was overloaded to support data extraction out of JSON arrays and other collections by index.

jsonArray[1] //extracts the first element of the given jsonArray

Note: Also achievable using the function get(jsonArray, 1).

If you are using Maven, add jep-data-extension as a dependency to your pom.xml file.

<dependency>
    <groupId>net.obvj</groupId>
    <artifactId>jep-data-extension</artifactId>
    <version>1.0.7</version>
</dependency>

If you use other dependency managers (such as Gradle, Grape, Ivy, etc.) click here.

1. Add jep-data-extensions to your class path

2. Put your source variables in a map:

   Map<String, Object> myVariables = new HashMap<>();
   myVariables.put("myText", "foo");

3. Use the JEPContextFactory class to create an extended JEP object with all available extensions:

   JEP jep = JEPContextFactory.newContext(myVariables);

   Note: Alternatively, you may use the JEPContextFactory.newContext() to receive a new context with no initial variables. You may add them later, calling jep.addVariable(String, Object).

4. Parse your expression:

   Node node = jep.parseExpresion("replace(myText, \"oo\", \"ee\")");

5. Evaluate the expression with JEP's evaluate method:

   String result = (String) jep.evaluate(node); //result = "fee"

6. Alternatively, you can use the assignment operator inside an expression. The new variable will be available in JEP context for reuse and can be retrieved using the getVarValue method:

   Node node = jep.parseExpression("myText = replace(\"fear\", \"f\", \"d\")"));
   jep.evaluate(node);
   String result = (String) jep.getVarValue("myText"); //result = "dear"

The ExpressionEvaluatorFacade is a convenient object to quickly parse a single expression. A single call to the evaluate method will create a new evaluation context and return the expression result directly. It accepts a map with initial variables for use in the operation.

1. Put your source variables in a map:

   Map<String, Object> myVariables = new HashMap<>();
   myVariables.put("date1", "2018-12-20T09:10:00.123456789Z");

2. Evaluate your expression:

   String expression = "if(date1 < now(), \"overdue\", \"not overdue\")";
   String result = (String) ExpressionEvaluatorFacade.evaluate(expression, myVariables); 
   //result = "overdue"

1. Run jep-data-extension JAR file from a terminal:

   $ java -jar jep-data-extension-1.0.2.jar
   

2. Consume the RESTful API with the httpGet function:

   JEP > dateTime = httpGet("http://date.jsontest.com/")
   {
      "time": "08:31:19 PM",
      "date": "05-12-2012"
   }
   

3. Filter the JSON using a JsonPath expression that extracts the date field:

   JEP > date = jsonpath(dateTime, "$.date")
   05-12-2012
   

4. Check the type of the generated date variable:

   JEP > class(date)
   java.lang.String
   

5. Convert the string to Date:

   JEP > date = str2date(date, "dd-MM-yyyy")
   Wed Dec 05 00:00:00 BRST 2012
   

6. Extract the week number from the converted date:

   JEP > isoWeekNumber(date)
   49