Package org.apache.myfaces.view.facelets.util

Class Classpath

java.lang.Object

org.apache.myfaces.view.facelets.util.Classpath

public final class Classpath
extends Object

Version:

$Id$

Author:

Jacob Hookom, Roland Huss, Ales Justin (ales.justin@jboss.org)

Method Summary

Modifier and Type	Method	Description
static URL[]	search(ClassLoader loader, String prefix, String suffix)
static URL[]	search(String prefix, String suffix)
static Set<String>	searchResourceNames(ClassLoader loader, String prefix)	Collect, in a single classpath walk, the names of every resource entry whose name starts with the given prefix (e.g.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Details

search

public static URL[] search(String prefix,
 String suffix)
                    throws IOException

Throws:

IOException

search

public static URL[] search(ClassLoader loader,
 String prefix,
 String suffix)
                    throws IOException

Throws:

IOException

searchResourceNames

public static Set<String> searchResourceNames(ClassLoader loader,
 String prefix)
                                       throws IOException

Collect, in a single classpath walk, the names of every resource entry whose name starts with the given prefix (e.g. "META-INF/"). Unlike search(ClassLoader, String, String) this neither filters by suffix nor resolves URLs; it only enumerates jar/directory entries once and returns their names.

Callers that repeatedly search the same prefix for different suffixes (the startup config/taglib/contract providers all scan META-INF/) can walk the classpath once, filter the returned names by suffix, and resolve the handful of matches via ClassLoader.getResources(String) - instead of re-opening and re-enumerating every jar once per suffix. Keeping only names (not eagerly-built URLs) avoids allocating a URL per entry.

Throws:

IOException