Edit C:\apache-ant-1.8.0\docs\manual\projecthelper.html
<!-- Licensed to the Apache Software Foundation (ASF) under one or more contributor license agreements. See the NOTICE file distributed with this work for additional information regarding copyright ownership. The ASF licenses this file to You under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. --> <html> <head> <meta http-equiv="Content-Language" content="en-us"> <link rel="stylesheet" type="text/css" href="stylesheets/style.css"> <title>The Ant frontend: ProjectHelper</title> </head> <body> <h1>The Ant frontend: ProjectHelper</h1> <h2><a name="definition">What is a ProjectHelper?</a></h2> <p> The <code>ProjectHelper</code> in Ant is responsible for parsing the build file and creating java instances representing the build workflow. It also signals which kind of file it can parse, and which file name it expects as default input file. </p> <p> Ant' default <code>ProjectHelper</code> (<code>org.apache.tools.ant.helper.ProjectHelper2</code>) parses the usual build.xml files. And if no build file is specified on the command line, it will expect to find a file named <code>build.xml</code>. </p> <p> The immediate benefit of a such abstraction it that it is possible to make Ant understand other kind of descriptive languages than XML. Some experiments have been done around a pure java frontend, and a groovy one too (ask the dev mailing list for further info about these). </p> <h2><a name="repository">How is Ant is selecting the proper ProjectHelper</a></h2> <p> Ant knows about several implementations of <code>ProjectHelper</code> and has to decide which to use for each build file. </p> <p>At startup Ant lists the all implementations found and keeps them in the same order they've been found in an internal 'repository': <ul> <li>the first to be searched for is the one declared by the system property <code>org.apache.tools.ant.ProjectHelper</code> (see <a href="running.html#sysprops">Java System Properties</a>);</li> <li>then it searches with its class loader for a <code>ProjectHelper</code> service declarations in the META-INF: it searches in the classpath for a file <code>META-INF/services/org.apache.tools.ant.ProjectHelper</code>. This file will just contain the fully qualified name of the implementation of <code>ProjectHelper</code> to instanciate;</li> <li>it will also search with the system class loader for <code>ProjectHelper</code> service declarations in the META-INF;</li> <li>last but not least it will add its default <code>ProjectHelper</code> that can parse classical build.xml files.</li> </ul> In case of an error while trying to instanciate a <code>ProjectHelper</code>, Ant will log an error but won't stop. If you want further debugging info about the <code>ProjectHelper</code> internal 'repository', use the <b>system</b> property <code>ant.project-helper-repo.debug</code> and set it to <code>true</code>; the full stack trace will then also be printed. </p> <p> When Ant is expected to parse a file, it will ask the <code>ProjectHelper</code> repository to find an implementation that will be able to parse the input file. Actually it will just iterate over the ordered list and the first implementation that returns <code>true</code> to <code>supportsBuildFile(File buildFile)</code> will be selected. </p> <p> When Ant is started and no input file has been specified, it will search for a default input file. It will iterate over list of <code>ProjectHelper</code>s and will select the first one that expects a default file that actually exist. </p> <h2><a name="writing">Writing your own ProjectHelper</a></h2> <p> The class <code>org.apache.tools.ant.ProjectHelper</code> is the API expected to be implemented. So write your own <code>ProjectHelper</code> by extending that abstract class. You are then expected to implement at least the function <code>parse(Project project, Object source)</code>. Note also that your implementation will be instanciated by Ant, and it is expecting a default constructor with no arguments. </p> <p> There are some functions that will help you define what your helper is capable of and what is is expecting: <ul> <li><code>getDefaultBuildFile()</code>: defines which file name is expected if none provided</li> <li><code>supportsBuildFile(File buildFile)</code>: defines if your parser can parse the input file</li> <li><code>canParseAntlibDescriptor(URL url)</code>: whether your implementation is capable of parsing a given Antlib descriptor. The base class returns <code>false</code></li> <li><code>parseAntlibDescriptor(Project containingProject, URL source)</code>: invoked to actually parse the Antlib descriptor if your implementation returned <code>true</code> for the previous method.</li> </ul> </p> <p> Now that you have your implementation ready, you have to declare it to Ant. Two solutions here: <ul> <li>use the system property <code>org.apache.tools.ant.ProjectHelper</code> (see also the <a href="running.html#sysprops">Java System Properties</a>);</li> <li>use the service file in META-INF: in the jar you will build with your implementation, add a file <code>META-INF/services/org.apache.tools.ant.ProjectHelper</code>. And then in this file just put the fully qualified name of your implementation</li> </ul> </p> </body> </html>
Ms-Dos/Windows
Unix
Write backup
jsp File Browser version 1.2 by
www.vonloesch.de