com.tangosol.dev.introspect.ClassPathResourceDiscoverer Maven / Gradle / Ivy
Show all versions of coherence Show documentation
/*
* Copyright (c) 2000, 2020, Oracle and/or its affiliates.
*
* Licensed under the Universal Permissive License v 1.0 as shown at
* http://oss.oracle.com/licenses/upl.
*/
package com.tangosol.dev.introspect;
import com.tangosol.io.ClassLoaderAware;
import com.tangosol.net.CacheFactory;
import com.tangosol.util.Base;
import com.tangosol.util.Filter;
import com.tangosol.util.SimpleEnumerator;
import java.io.File;
import java.io.FileFilter;
import java.io.IOException;
import java.net.JarURLConnection;
import java.net.URI;
import java.net.URISyntaxException;
import java.net.URL;
import java.net.URLConnection;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collection;
import java.util.Enumeration;
import java.util.HashSet;
import java.util.Iterator;
import java.util.List;
import java.util.Set;
import java.util.jar.JarEntry;
import java.util.jar.JarFile;
import java.util.regex.Pattern;
/**
* ClassPathResourceDiscoverer is {@link ResourceDiscoverer} implementation
* that uses its {@link ClassLoader} ({@link #getContextClassLoader()}) to
* determine appropriate artifacts based on the discriminator passed to
* {@link #discoverResource(String, String)}. The resource discovery
* mechanism employed by this class resides in
* {@link #findUrls(String, String)} and by default uses
* {@link ClassLoader#getResources(String)}. This has some limitations
* including the inability refer to a resource prefix in the case of a jar.
* Thus the discriminator must be a resource that can be referenced in all
* artifacts that may have resources complying to the regular expression.
*
* Each of these derived artifacts
* ({@link URL}s) are evaluated against a resource discoverer chain
* {@link #getDiscovererChain()}.
*
* Each resource discoverer is given the opportunity discover all
* leaf-resources in the current artifact. ResourceDiscoverer's that
* implement {@link Filter} are provided the URL of the artifact to either
* disregard the artifact or confirm acceptance to interrogate said artifact.
* The chain of ResourceDiscoverer's are provided both the URL
* and a string expression. The expression is the same passed to
* {@link #discoverResource(String, String)} and could represent a package.
*
* @author hr 2011.10.18
*
* @since Coherence 12.1.2
*
* @see ResourceDiscoverer
* @see InformedResourceDiscoverer
* @see FileBasedResourceDiscoverer
* @see JarFileBasedResourceDiscoverer
*/
public class ClassPathResourceDiscoverer
extends AbstractFilteredResourceDiscoverer
{
// ----- constructors ---------------------------------------------------
/**
* Construct a ClassPathResourceDiscoverer with default configuration.
*/
public ClassPathResourceDiscoverer()
{
this(null, null);
}
/**
* Construct a ClassPathResourceDiscoverer with the provided list of
* {@link ResourceDiscoverer}s and a set of acceptable file types.
*
* @param setFileTypes the set of acceptable file types
* @param listDiscoverers the {@link ResourceDiscoverer} chain to call
* for each discovered resource
*/
public ClassPathResourceDiscoverer(Set setFileTypes, List> listDiscoverers)
{
m_listDiscoverers = listDiscoverers == null ? m_listDiscoverers : listDiscoverers;
m_setFileTypes = setFileTypes == null ? m_setFileTypes : setFileTypes;
initialize();
}
// ----- AbstractFilteredResourceDiscoverer methods ---------------------
/**
* Returns an {@link Enumeration} of classpath relative resources that
* comply to the provided {@code sExpression}. This implementation uses
* a chain of {@link ResourceDiscoverer}s to determine leaf-resources.
*
* Note:A null or empty discriminator will result in using the
* {@literal sExpression} as the discriminator. This default logic can
* be overridden by providing a custom implementation of
* {@link #findUrls(String, String)}.
*
* @param sDiscriminator a string representing the resource that allows
* the discovery of all artifacts (jar files or
* directories in the class path) that should be
* interrogated for resources complying to the
* {@literal sExpression}
* @param sExpression the regular expression applied to all child
* resources of the discovered resource via
* {@literal sDiscriminator}
*
* @return an Enumeration of resources that comply to provided
* {@literal sExpression}
*/
@Override
public Collection discoverResource(String sDiscriminator, String sExpression)
{
sExpression = sExpression == null ? "" : sExpression;
Set setResources = new HashSet<>();
String sAppendExpr = buildFileTypeExpression();
try
{
for (Enumeration enumResources = findUrls(sDiscriminator, sExpression); enumResources.hasMoreElements(); )
{
URL url = enumResources.nextElement();
for (ResourceDiscoverer discoverer : getDiscovererChain())
{
if (discoverer instanceof Filter && !((Filter) discoverer).evaluate(url))
{
continue;
}
Enumeration enumStringPaths = discoverer.discover(".*" + sExpression + ".*" + sAppendExpr, url);
while (enumStringPaths != null && enumStringPaths.hasMoreElements())
{
URI uri = toURI(enumStringPaths.nextElement());
if (uri != null)
{
setResources.add(uri);
}
}
}
}
}
catch (IOException e)
{
throw Base.ensureRuntimeException(e);
}
return setResources;
}
/**
* {@inheritDoc}
*/
@Override
protected Set getPermittedProtocols()
{
return new HashSet();
}
/**
* Based on the set of file types this implementation was instantiated
* with build a regular expression that results in an OR of the file
* types.
*
* @return regular expression permitting certain file types
*/
protected String buildFileTypeExpression()
{
Set setFileTypes = m_setFileTypes;
StringBuilder sbldrExpr = new StringBuilder("(");
for (Iterator iterFileTypes = setFileTypes.iterator(); iterFileTypes.hasNext(); )
{
String sFileType = iterFileTypes.next();
sbldrExpr.append("\\.")
.append(sFileType)
.append(iterFileTypes.hasNext() ? "|" : ")$");
}
return sbldrExpr.length() == 1 ? "" : sbldrExpr.toString();
}
/**
* Based on the provided {@literal sDiscriminator} and
* {@literal sExpression} derive an {@link Enumeration} of {@link URL}s
* that are likely candidates of satisfying {@literal sExpression}. These
* {@link URL}s will be passed down a chain of {@link ResourceDiscoverer}s
* for interrogation.
*
* @param sDiscriminator a string representing the resource that allows
* the discovery of all artifacts (jar files or
* directories in the class path) that should be
* interrogated for resources complying to the
* {@literal sExpression}
* @param sExpression the regular expression applied to all child
* resources of the discovered resource via
* {@literal sDiscriminator}
*
* @return an Enumeration or URLs that may comply to the {@literal sExpression}
*
* @throws IOException
*/
protected Enumeration findUrls(String sDiscriminator, String sExpression) throws IOException
{
return getContextClassLoader().getResources(sDiscriminator == null
|| sDiscriminator.isEmpty() ? sExpression : sDiscriminator);
}
// ----- ClassLoaderHolder methods --------------------------------------
/**
* Ensure all {@link ClassLoaderAware} resources this class is aware of
* are updated with the new {@link ClassLoader}.
*
* @param loader the context ClassLoader for this object
*/
@Override
public void setContextClassLoader(ClassLoader loader)
{
super.setContextClassLoader(loader);
List> listDiscoverers = getDiscovererChain();
if (listDiscoverers == null)
{
return;
}
for (ResourceDiscoverer discoverer : listDiscoverers)
{
discoverer.setContextClassLoader(getContextClassLoader());
}
}
// ----- accessors ------------------------------------------------------
/**
* Returns a list of {@link ResourceDiscoverer}s this class is configured
* to use.
*
* @return a list of ResourceDiscoverer's this class is configured
* to use
*/
public List> getDiscovererChain()
{
return m_listDiscoverers;
}
/**
* Sets a list of {@link ResourceDiscoverer}s this class is configured
* to use.
*
* @param listDiscoverers the list of ResourceDiscoverer's this class
* should use to determine leaf-resources
*/
public void setDiscoverers(List> listDiscoverers)
{
Base.azzert(listDiscoverers != null, "ClassPathResourceDiscoverer requires a non-null set of child discoverers");
m_listDiscoverers = listDiscoverers;
}
// ----- helpers --------------------------------------------------------
/**
* Based on the state of the class initialize it with sensible defaults.
*/
protected void initialize()
{
List> listDiscoverers = m_listDiscoverers;
if (listDiscoverers.size() == 0)
{
listDiscoverers.add(new FileBasedResourceDiscoverer());
listDiscoverers.add(new JarFileBasedResourceDiscoverer());
}
}
// ----- inner class: InformedResourceDiscoverer ------------------------
/**
* An InformedResourceDiscoverer is provided a number of {@link URL}s to
* be interrogated by a chain of {@link ResourceDiscoverer}
* implementations.
*
* This ResourceDiscoverer implementation is useful when the
* resources are known upfront precluding the need of a discovery
* mechanism such as a {@link ClassLoader#getResources(String)} request.
*/
public static class InformedResourceDiscoverer
extends ClassPathResourceDiscoverer
{
// ----- constructors -----------------------------------------------
/**
* Construct as InformedResourceDiscoverer with the provided array of
* {@link URL}s.
*
* @param aUrl array of URLs passed down the chain of
* {@link ResourceDiscoverer}s
*/
public InformedResourceDiscoverer(URL[] aUrl)
{
this(null, null, aUrl);
}
/**
* Construct an InformedResourceDiscoverer with the passed chain of
* {@link ResourceDiscoverer}s and an array of {@link URL}s.
*
* @param setFileTypes the set of acceptable file types
* @param listDiscoverers chain of ResourceDiscoverer's
* @param aUrl array of URLs passed down the chain
* of ResourceDiscoverer's
*/
public InformedResourceDiscoverer(Set setFileTypes, List> listDiscoverers, URL[] aUrl)
{
super(setFileTypes, listDiscoverers);
m_aUrl = aUrl;
}
/**
* {@inheritDoc}
*/
@Override
protected Enumeration findUrls(String sDiscriminator, String sExpression) throws IOException
{
return new SimpleEnumerator(m_aUrl);
}
// ----- data members -----------------------------------------------
/**
* Array of {@link URL}s passed down the chain of
* {@link ResourceDiscoverer}s.
*/
protected URL[] m_aUrl;
}
// ----- inner class: FileBasedResourceDiscoverer -----------------------
/**
* A {@link ResourceDiscoverer} implementation that works against "file"
* based resources, such that the {@link URL} provided has a "file"
* protocol. This implementation will traverse the directory tree
* discovering all complying leaf-resources.
*
* @author hr 2011.10.18
*/
protected static class FileBasedResourceDiscoverer
extends AbstractFilteredResourceDiscoverer
{
// ----- AbstractFilteredResourceDiscoverer methods -----------------
/**
* {@inheritDoc}
*/
protected Set getPermittedProtocols()
{
return s_setProtocols;
}
// ----- ResourceDiscoverer interface -------------------------------
/**
* Starting from the {@code urlRoot} traverse the directory tree
* evaluating all leaf-resources against {@code sFileExpression}
* returning complying resources. This method will return the full
* path of the resource.
*
* @param sFileExpression the expression that each leaf-resource
* must match
* @param urlRoot the root element to start traversing from
*
* @return a number of resources that complied with the
* {@code sFileExpression} and belong to {@code urlRoot}
*/
@Override
public Collection discoverResource(String sFileExpression, URL urlRoot)
{
Set setResources = new HashSet<>();
URI uriRoot = toURI(urlRoot);
if (uriRoot == null)
{
CacheFactory.log("FileBasedResourceDiscoverer requires a non-null and non-empty root directory",
CacheFactory.LOG_WARN);
return setResources;
}
File root = ensureDirectory(new File(uriRoot));
if (root == null)
{
return setResources;
}
if (root.isFile() && root.canRead())
{
setResources.add(uriRoot);
return setResources;
}
Pattern regEx = Pattern.compile(sFileExpression);
findFiles(root, regEx, setResources);
return setResources;
}
// ----- helpers ----------------------------------------------------
/**
* From the current directory ({@code root}), find all files that
* match the {@code regEx} parameter adding them to the
* {@code setResources}. This method will be called recursively if
* any of the files discovered are directories.
*
* @param root the directory to start from
* @param regEx the pattern to match file names against
* @param setResources the set of resources to add complying files
* to
*/
protected void findFiles(File root, final Pattern regEx, Set setResources)
{
File[] complyingFiles = root.listFiles(new FileFilter()
{
public boolean accept(File pathname)
{
// ensure path separators use the java & unix format
String sPath = pathname.getAbsolutePath();
return pathname.isDirectory() || regEx.matcher(sPath.contains("\\")
? sPath.replaceAll("\\\\", "/") : sPath).matches();
}
});
for (int i = 0; i< complyingFiles.length; ++i)
{
File complyingFile = complyingFiles[i];
if (complyingFile.isFile())
{
setResources.add(complyingFile.toURI());
}
else if (complyingFile.isDirectory())
{
if (ensureDirectory(complyingFile) == null)
{
continue;
}
findFiles(complyingFile, regEx, setResources);
}
}
}
/**
* Based on {@code root} ensure it is a directory and can be
* read. Display a warning message if not the case and return null.
*
* @param root the file resource
*
* @return the file reference
*/
protected File ensureDirectory(File root)
{
if (!root.isDirectory())
{
CacheFactory.log(String.format(
"FileBasedResourceDiscoverer root directory should be a valid directory [rootDir: %s]",
root.getAbsolutePath()), CacheFactory.LOG_WARN);
return null;
}
if (!root.canRead())
{
CacheFactory.log(String.format(
"FileBasedResourceDiscoverer root directory should be a readable directory [rootDir: %s]",
root.getAbsolutePath()), CacheFactory.LOG_WARN);
return null;
}
return root;
}
// ----- constants --------------------------------------------------
/**
* All protocols this {@link ResourceDiscoverer} can operate against.
*/
protected static final Set s_setProtocols = new HashSet(Arrays.asList("file"));
}
// ----- inner class: JarFileBasedResourceDiscoverer --------------------
/**
* A {@link ResourceDiscoverer} implementation that works against "jar",
* "zip", "wsjar", "code-source", based resources, such that the
* {@link URL} provided has one of the mentioned protocols. This
* implementation will traverse the directory tree discovering all
* complying leaf-resources.
*
* @author hr 2011.10.18
*/
protected static class JarFileBasedResourceDiscoverer
extends AbstractFilteredResourceDiscoverer
{
// ----- AbstractFilteredResourceDiscoverer methods -----------------
/**
* {@inheritDoc}
*/
@Override
protected Set getPermittedProtocols()
{
return s_setProtocols;
}
/**
* Starting from the {@code urlRoot} traverse the directory tree
* evaluating all leaf-resources against {@code sFileExpression}
* returning complying resources. This method will return the full
* path of the resource.
*
* @param urlRoot the root element to start traversing from
* @param sFileExpression the expression that each leaf-resource
* must match
*
* @return a number of resources that complied with the
* {@code sFileExpression} and belong to {@code urlRoot}
*/
@Override
public Collection discoverResource(String sFileExpression, URL urlRoot)
{
Set setResources = new HashSet<>();
Pattern regEx = Pattern.compile(sFileExpression);
try
{
URLConnection urlCon = urlRoot.openConnection();
JarFile jarFile;
if (urlCon instanceof JarURLConnection)
{
JarURLConnection jarUrl = (JarURLConnection) urlCon;
jarUrl.setUseCaches(false);
jarFile = jarUrl.getJarFile();
}
else
{
// attempt to create a File reference removing any notation
// specific to the jar protocol
String sFullLocation = urlRoot.getFile();
int iEmbeddedRef = sFullLocation.indexOf("!/");
String sLocation = iEmbeddedRef == -1 ? sFullLocation : sFullLocation.substring(0, iEmbeddedRef);
// try to create a URI out using the location
// if not possible assume the format of sLocation to be
// "file:/foo" or "someother:/"
// in case of "/dir1/dir:2/" pass this as the fs location
try
{
jarFile = new JarFile(new URI(sLocation.replace(" ", "%20")).getSchemeSpecificPart());
}
catch (URISyntaxException e)
{
int iColon = sLocation.indexOf(":");
jarFile = new JarFile(iColon == -1 || sLocation.substring(0, iColon).contains(File.separator)
? sLocation : sLocation.substring(iColon + 1));
}
}
// scan the entries adding all matched values to the results
for (Enumeration enumEntries = jarFile.entries(); enumEntries.hasMoreElements(); )
{
JarEntry entry = enumEntries.nextElement();
String sResource = entry.getName();
if (regEx.matcher(sResource).matches())
{
URI uri = toURI(new URL(urlRoot, "/" + sResource));
if (uri != null)
{
setResources.add(uri);
}
}
}
}
catch (IOException e)
{
CacheFactory.log(String.format(
"JarFileBasedResourceDiscoverer could not open a connection to the resource [url: %s]"
+ ", hence this resource will not be used",
urlRoot.toExternalForm()), CacheFactory.LOG_WARN);
}
return setResources;
}
// ----- constants --------------------------------------------------
/**
* All protocols this {@link ResourceDiscoverer} can operate against.
*/
protected static final Set s_setProtocols = new HashSet(Arrays.asList("jar", "zip", "wsjar", "code-source"));
}
// ----- data member ----------------------------------------------------
/**
* The set of accepted file types to return as a result of
* {@link #discoverResource(String, String)}.
*/
protected Set m_setFileTypes = new HashSet(Arrays.asList("class"));
/**
* A {@link List} of {@link ResourceDiscoverer}s the
* ClassPathResourceDiscoverer will use to discover leaf-resources from
* different types of resources.
*/
protected List> m_listDiscoverers = new ArrayList>(2);
}