Package com.codemesh.terp.eval

The com.codemesh.terp.eval package contains most of the types that you would use when you embed terp into your own application.

See:
Description

Class Summary

PathTokenStream	A tokenizer for path expressions.
TerpContextImpl	The public portion of the context in which bindings and general evaluation settings are specified.
TerpEvaluatorImpl	The default evaluator for terp templates and expressions.
TerpParser	The parser that converts expression and template sources into abstract syntax trees.
TerpRegistrarImpl	The default implementation of a registrar for the terp framework.
TextTokenStream	A token stream that distinguishes between text and embedded tokens.
TokenStream	The baseclass for token-producing streams.

Package com.codemesh.terp.eval Description

The com.codemesh.terp.eval package contains most of the types that you would use when you embed terp into your own application. When you use terp programmatically, you need to first parse the template or expression using a TerpParser instance. You need to create a context in which the template or expression is going to be evaluated. The default context implemenetation is provided by the TerpContextImpl type. The parser and the context are then handed to a TerpEvaluatorImpl instance which performs the expansion or evaluation.

The following snippet demonstrates this usage in an expression evaluation:

TerpParser    parser = new TerpParser( "2*pi", TokenizerState.EXPRESSION );
TerpContext   ctx = new TerpContextImpl();
TerpEvaluator eval = TerpEvaluatorImpl.getInstance();
Object        result;

try {
    result = eval.evaluate( parser, ctx );
}
catch( TerpException te ) {
    // ...
}

The following snippet demonstrates this usage in a template expansion:

TerpParser    parser = new TerpParser( "The value of 2*pi is ${(2*pi)['%4.2f']}", TokenizerState.TEXT );
TerpContext   ctx = new TerpContextImpl();
TerpEvaluator eval = TerpEvaluatorImpl.getInstance();

try {
    eval.expand( parser, ctx );
    System.out.println( ctx.getAppendable().toString() );
}
catch( TerpException te ) {
    // ...
}

It is important to keep the following points in mind:

• You can reuse TerpParser instances, i.e. a template once parsed can be used for more than one expansion/evaluation. This can significantly improve performance.
• You do not have to create multiple instances of type TerpEvaluator. Using a single instance of type TerpEvaluatorImpl is totally sufficient and improves performance by reducing intialization activities.
• The TerpContextImpl type is the type that you will typically (re)create for every evaluation or expansion, though even it may be reused if you are careful.