Evaluation

This section introduces programmatic use of SpEL’s interfaces and its expression language. The complete language reference can be found in the Language Reference.

The following code demonstrates how to use the SpEL API to evaluate the literal string expression, Hello World.

  • Java

ExpressionParser parser = new SpelExpressionParser();
Expression exp = parser.parseExpression("'Hello World'"); (1)
String message = (String) exp.getValue();
1 The value of the message variable is "Hello World".

The SpEL classes and interfaces you are most likely to use are located in the cn.taketoday.expression package and its sub-packages, such as spel.support.

The ExpressionParser interface is responsible for parsing an expression string. In the preceding example, the expression string is a string literal denoted by the surrounding single quotation marks. The Expression interface is responsible for evaluating the defined expression string. The two types of exceptions that can be thrown when calling parser.parseExpression(…​) and exp.getValue(…​) are ParseException and EvaluationException, respectively.

SpEL supports a wide range of features such as calling methods, accessing properties, and calling constructors.

In the following method invocation example, we call the concat method on the string literal, Hello World.

  • Java

ExpressionParser parser = new SpelExpressionParser();
Expression exp = parser.parseExpression("'Hello World'.concat('!')"); (1)
String message = (String) exp.getValue();
1 The value of message is now "Hello World!".

The following example demonstrates how to access the Bytes JavaBean property of the string literal, Hello World.

  • Java

ExpressionParser parser = new SpelExpressionParser();

// invokes 'getBytes()'
Expression exp = parser.parseExpression("'Hello World'.bytes"); (1)
byte[] bytes = (byte[]) exp.getValue();
1 This line converts the literal to a byte array.

SpEL also supports nested properties by using the standard dot notation (such as prop1.prop2.prop3) as well as the corresponding setting of property values. Public fields may also be accessed.

The following example shows how to use dot notation to get the length of a string literal.

  • Java

ExpressionParser parser = new SpelExpressionParser();

// invokes 'getBytes().length'
Expression exp = parser.parseExpression("'Hello World'.bytes.length"); (1)
int length = (Integer) exp.getValue();
1 'Hello World'.bytes.length gives the length of the literal.

The String’s constructor can be called instead of using a string literal, as the following example shows.

  • Java

ExpressionParser parser = new SpelExpressionParser();
Expression exp = parser.parseExpression("new String('hello world').toUpperCase()"); (1)
String message = exp.getValue(String.class);
1 Construct a new String from the literal and convert it to upper case.

Note the use of the generic method: public <T> T getValue(Class<T> desiredResultType). Using this method removes the need to cast the value of the expression to the desired result type. An EvaluationException is thrown if the value cannot be cast to the type T or converted by using the registered type converter.

The more common usage of SpEL is to provide an expression string that is evaluated against a specific object instance (called the root object). The following example shows how to retrieve the name property from an instance of the Inventor class and how to reference the name property in a boolean expression.

  • Java

// Create and set a calendar
GregorianCalendar c = new GregorianCalendar();
c.set(1856, 7, 9);

// The constructor arguments are name, birthday, and nationality.
Inventor tesla = new Inventor("Nikola Tesla", c.getTime(), "Serbian");

ExpressionParser parser = new SpelExpressionParser();

Expression exp = parser.parseExpression("name"); // Parse name as an expression
String name = (String) exp.getValue(tesla);
// name == "Nikola Tesla"

exp = parser.parseExpression("name == 'Nikola Tesla'");
boolean result = exp.getValue(tesla, Boolean.class);
	// result == true

Understanding EvaluationContext

The EvaluationContext interface is used when evaluating an expression to resolve properties, methods, or fields and to help perform type conversion. Infra provides two implementations.

  • SimpleEvaluationContext: Exposes a subset of essential SpEL language features and configuration options, for categories of expressions that do not require the full extent of the SpEL language syntax and should be meaningfully restricted. Examples include but are not limited to data binding expressions and property-based filters.

  • StandardEvaluationContext: Exposes the full set of SpEL language features and configuration options. You can use it to specify a default root object and to configure every available evaluation-related strategy.

SimpleEvaluationContext is designed to support only a subset of the SpEL language syntax. It excludes Java type references, constructors, and bean references. It also requires you to explicitly choose the level of support for properties and methods in expressions. By default, the create() static factory method enables only read access to properties. You can also obtain a builder to configure the exact level of support needed, targeting one or some combination of the following.

  • Custom PropertyAccessor only (no reflection)

  • Data binding properties for read-only access

  • Data binding properties for read and write

Type Conversion

By default, SpEL uses the conversion service available in Infra core (cn.taketoday.core.convert.ConversionService). This conversion service comes with many built-in converters for common conversions, but is also fully extensible so that you can add custom conversions between types. Additionally, it is generics-aware. This means that, when you work with generic types in expressions, SpEL attempts conversions to maintain type correctness for any objects it encounters.

What does this mean in practice? Suppose assignment, using setValue(), is being used to set a List property. The type of the property is actually List<Boolean>. SpEL recognizes that the elements of the list need to be converted to Boolean before being placed in it. The following example shows how to do so.

  • Java

class Simple {
  public List<Boolean> booleanList = new ArrayList<>();
}

Simple simple = new Simple();
simple.booleanList.add(true);

EvaluationContext context = SimpleEvaluationContext.forReadOnlyDataBinding().build();

// "false" is passed in here as a String. SpEL and the conversion service
// will recognize that it needs to be a Boolean and convert it accordingly.
parser.parseExpression("booleanList[0]").setValue(context, simple, "false");

// b is false
Boolean b = simple.booleanList.get(0);

Parser Configuration

It is possible to configure the SpEL expression parser by using a parser configuration object (cn.taketoday.expression.spel.SpelParserConfiguration). The configuration object controls the behavior of some of the expression components. For example, if you index into an array or collection and the element at the specified index is null, SpEL can automatically create the element. This is useful when using expressions made up of a chain of property references. If you index into an array or list and specify an index that is beyond the end of the current size of the array or list, SpEL can automatically grow the array or list to accommodate that index. In order to add an element at the specified index, SpEL will try to create the element using the element type’s default constructor before setting the specified value. If the element type does not have a default constructor, null will be added to the array or list. If there is no built-in or custom converter that knows how to set the value, null will remain in the array or list at the specified index. The following example demonstrates how to automatically grow the list.

  • Java

class Demo {
  public List<String> list;
}

// Turn on:
// - auto null reference initialization
// - auto collection growing
SpelParserConfiguration config = new SpelParserConfiguration(true, true);

ExpressionParser parser = new SpelExpressionParser(config);

Expression expression = parser.parseExpression("list[3]");

Demo demo = new Demo();

Object o = expression.getValue(demo);

// demo.list will now be a real collection of 4 entries
// Each entry is a new empty String

By default, a SpEL expression cannot contain more than 10,000 characters; however, the maxExpressionLength is configurable. If you create a SpelExpressionParser programmatically, you can specify a custom maxExpressionLength when creating the SpelParserConfiguration that you provide to the SpelExpressionParser. If you wish to set the maxExpressionLength used for parsing SpEL expressions within an ApplicationContext — for example, in XML bean definitions, @Value, etc. — you can set a JVM system property or Infra property named infra.context.expression.maxLength to the maximum expression length needed by your application (see Supported Infra Properties).

SpEL Compilation

Infra provides a basic compiler for SpEL expressions. Expressions are usually interpreted, which provides a lot of dynamic flexibility during evaluation but does not provide optimum performance. For occasional expression usage, this is fine, but, when used by other components such as Infra Integration, performance can be very important, and there is no real need for the dynamism.

The SpEL compiler is intended to address this need. During evaluation, the compiler generates a Java class that embodies the expression behavior at runtime and uses that class to achieve much faster expression evaluation. Due to the lack of typing around expressions, the compiler uses information gathered during the interpreted evaluations of an expression when performing compilation. For example, it does not know the type of a property reference purely from the expression, but during the first interpreted evaluation, it finds out what it is. Of course, basing compilation on such derived information can cause trouble later if the types of the various expression elements change over time. For this reason, compilation is best suited to expressions whose type information is not going to change on repeated evaluations.

Consider the following basic expression.

someArray[0].someProperty.someOtherProperty < 0.1

Because the preceding expression involves array access, some property de-referencing, and numeric operations, the performance gain can be very noticeable. In an example micro benchmark run of 50,000 iterations, it took 75ms to evaluate by using the interpreter and only 3ms using the compiled version of the expression.

Compiler Configuration

The compiler is not turned on by default, but you can turn it on in either of two different ways. You can turn it on by using the parser configuration process (discussed earlier) or by using a Infra property when SpEL usage is embedded inside another component. This section discusses both of these options.

The compiler can operate in one of three modes, which are captured in the cn.taketoday.expression.spel.SpelCompilerMode enum. The modes are as follows.

  • OFF (default): The compiler is switched off.

  • IMMEDIATE: In immediate mode, the expressions are compiled as soon as possible. This is typically after the first interpreted evaluation. If the compiled expression fails (typically due to a type changing, as described earlier), the caller of the expression evaluation receives an exception.

  • MIXED: In mixed mode, the expressions silently switch between interpreted and compiled mode over time. After some number of interpreted runs, they switch to compiled form and, if something goes wrong with the compiled form (such as a type changing, as described earlier), the expression automatically switches back to interpreted form again. Sometime later, it may generate another compiled form and switch to it. Basically, the exception that the user gets in IMMEDIATE mode is instead handled internally.

IMMEDIATE mode exists because MIXED mode could cause issues for expressions that have side effects. If a compiled expression blows up after partially succeeding, it may have already done something that has affected the state of the system. If this has happened, the caller may not want it to silently re-run in interpreted mode, since part of the expression may be run twice.

After selecting a mode, use the SpelParserConfiguration to configure the parser. The following example shows how to do so.

  • Java

SpelParserConfiguration config = new SpelParserConfiguration(SpelCompilerMode.IMMEDIATE,
    this.getClass().getClassLoader());

SpelExpressionParser parser = new SpelExpressionParser(config);

Expression expr = parser.parseExpression("payload");

MyMessage message = new MyMessage();

Object payload = expr.getValue(message);

When you specify the compiler mode, you can also specify a ClassLoader (passing null is allowed). Compiled expressions are defined in a child ClassLoader created under any that is supplied. It is important to ensure that, if a ClassLoader is specified, it can see all the types involved in the expression evaluation process. If you do not specify a ClassLoader, a default ClassLoader is used (typically the context ClassLoader for the thread that is running during expression evaluation).

The second way to configure the compiler is for use when SpEL is embedded inside some other component and it may not be possible to configure it through a configuration object. In such cases, it is possible to set the spel.compiler.mode property via a JVM system property (or via the InfraProperties mechanism) to one of the SpelCompilerMode enum values (off, immediate, or mixed).

Compiler Limitations

Infra does not support compiling every kind of expression. The primary focus is on common expressions that are likely to be used in performance-critical contexts. The following kinds of expressions cannot be compiled.

  • Expressions involving assignment

  • Expressions relying on the conversion service

  • Expressions using custom resolvers or accessors

  • Expressions using overloaded operators

  • Expressions using array construction syntax

  • Expressions using selection or projection

Compilation of additional kinds of expressions may be supported in the future.