The version of Apache log4j used by SoundHelix.
Clone
HTTPS:
git clone https://vervis.peers.community/repos/aEp6o
SSH:
git clone USERNAME@vervis.peers.community:aEp6o
Branches
Tags
- 1.3alpha-7
- CHAINSAW_2_SANDBOX_MERGE
- CORE_VERSION
- LEVEL_REPLACES_PRIORITY
- PREALPHA_1_3_AS_OF_2004_05_12
- PRE_CHAINSAW_MODEL_CONVERSION
- PRE_UGLI_MOVE
- TAG_CHAINSAW2_MOVE
- log4j-1.2.17
- log4j-1.2.17-rc1
- v1.3alpha8
- v1.3alpha8-temp
- v1_2_1
- v1_2_10-recalled
- v1_2_11
- v1_2_11_rc1
- v1_2_11rc3
- v1_2_12
- v1_2_12_rc1
- v1_2_12_rc2
- v1_2_12_rc3
- v1_2_12_rc4
- v1_2_12_rc5
- v1_2_12_rc6
- v1_2_13
- v1_2_13_rc1
- v1_2_13_rc2
- v1_2_13_site_update
- v1_2_14
- v1_2_14_maven
- v1_2_14_rc1
- v1_2_14_site_update
- v1_2_15
- v1_2_15_rc1
- v1_2_15_rc2
- v1_2_15_rc3
- v1_2_15_rc4
- v1_2_15_rc5
- v1_2_15_rc6
- v1_2_16
- v1_2_16_rc1
- v1_2_16_rc2
- v1_2_17
- v1_2_17-rc1
- v1_2_17_rc1
- v1_2_17_rc2
- v1_2_17_rc3
- v1_2_2
- v1_2_3
- v1_2_4
- v1_2_6
- v1_2_7
- v1_2_9
- v1_2_alpha0
- v1_2_alpha7
- v1_2beta1
- v1_2final
- v1_3alpha_1
- v1_3alpha_6
- v_1_0
- v_1_0_1
- v_1_0_4
- v_1_1
- v_1_1_1
- v_1_1_2
- v_1_1_3
- v_1_1_b1
- v_1_1b2
- v_1_1b3
- v_1_1b5
- v_1_1b6
- v_1_1b7
- v_1_2beta3
Hierarchy.java
/*
* Copyright (C) The Apache Software Foundation. All rights reserved.
*
* This software is published under the terms of the Apache Software
* License version 1.1, a copy of which has been included with this
* distribution in the LICENSE.APL file. */
// WARNING This class MUST not have references to the Category or
// WARNING RootCategory classes in its static initiliazation neither
// WARNING directly nor indirectly.
// Contributors:
// Luke Blanshard <luke@quiq.com>
// Mario Schomburg - IBM Global Services/Germany
// Anders Kristensen
// Igor Poteryaev
package org.apache.log4j;
import java.util.Hashtable;
import java.util.Enumeration;
import java.util.Vector;
import org.apache.log4j.spi.RootCategory;
import org.apache.log4j.spi.CategoryFactory;
import org.apache.log4j.or.RendererMap;
import org.apache.log4j.or.ObjectRenderer;
import org.apache.log4j.helpers.LogLog;
import org.apache.log4j.helpers.OptionConverter;
/**
This class is specialized in retreiving categories by name and
also maintaining the category hierarchy.
<p><em>The casual user should not have to deal with this class
firectly.</em> In fact, up until version 0.9.0, this class had
default package access. However, if you are in an environment where
multiple applications run in the same VM, then read on.
<p>The structure of the category hierachy is maintained by the
{@link #getInstance} method. The hierrachy is such that children
link to their parent but parents do not have any pointers to their
children. Moreover, categories can be instantiated in any order, in
particular decendant before ancestor.
<p>In case a decendant is created before a particular ancestor,
then it creates a provision node for the ancestor and adds itself
to the provision node. Other decendants of the same ancestor add
themselves to the previously created provision node.
<p>See the code below for further details.
@author Ceki Gülcü
*/
public class Hierarchy {
// DISABLE_OFF should be set to a value lower than all possible
// priorities.
static final int DISABLE_OFF = -1;
static final int DISABLE_OVERRIDE = -2;
static
private
CategoryFactory defaultFactory = new DefaultCategoryFactory();
Hashtable ht;
Category root;
RendererMap rendererMap;
int disable;
boolean emittedNoAppenderWarning = false;
boolean emittedNoResourceBundleWarning = false;
/**
Create a new Category hierarchy.
@param root The root of the new hierarchy.
*/
public
Hierarchy(Category root) {
ht = new Hashtable();
this.root = root;
// Don't disable any priority level by default.
disable = DISABLE_OFF;
this.root.setHierarchy(this);
rendererMap = new RendererMap();
}
/**
Add an object renderer for a specific class.
*/
public
void addRenderer(Class classToRender, ObjectRenderer or) {
rendererMap.put(classToRender, or);
}
/**
This call will clear all category definitions from the internal
hashtable. Invoking this method will irrevocably mess up the
category hiearchy.
<p>You should <em>really</em> know what you are doing before
invoking this method.
@since 0.9.0 */
public
void clear() {
//System.out.println("\n\nAbout to clear internal hash table.");
ht.clear();
}
/**
Check if the named category exists in the hirarchy. If so return
its reference, otherwise returns <code>null</code>.
@param name The name of the category to search for.
*/
public
Category exists(String name) {
Object o = ht.get(new CategoryKey(name));
if(o instanceof Category) {
return (Category) o;
} else {
return null;
}
}
/**
Similar to {@link #disable(Priority)} except that the priority
argument is given as a String. */
public
void disable(String priorityStr) {
if(disable != DISABLE_OVERRIDE) {
Priority p = Priority.toPriority(priorityStr, null);
if(p != null) {
disable = p.level;
} else {
LogLog.warn("Could not convert ["+priorityStr+"] to Priority.");
}
}
}
/**
Disable all logging requests of priority <em>equal to or
below</em> the priority parameter <code>p</code>, for
<em>all</em> categories in this hierarchy. Logging requests of
higher priority then <code>p</code> remain unaffected.
<p>Nevertheless, if the {@link
BasicConfigurator#DISABLE_OVERRIDE_KEY} system property is set to
"true" or any value other than "false", then logging requests are
evaluated as usual, i.e. according to the <a
href="../../../../manual.html#selectionRule">Basic Selection Rule</a>.
<p>The "disable" family of methods are there for speed. They
allow printing methods such as debug, info, etc. to return
immediately after an interger comparison without walking the
category hierarchy. In most modern computers an integer
comparison is measured in nanoseconds where as a category walk is
measured in units of microseconds.
<p>Other configurators define alternate ways of overriding the
disable override flag. See {@link PropertyConfigurator} and
{@link org.apache.log4j.xml.DOMConfigurator}.
@since 0.8.5 */
public
void disable(Priority p) {
if((disable != DISABLE_OVERRIDE) && (p != null)) {
disable = p.level;
}
}
/**
Disable all logging requests regardless of category and priority.
This method is equivalent to calling {@link #disable} with the
argument {@link Priority#FATAL}, the highest possible priority.
@since 0.8.5 */
public
void disableAll() {
disable(Priority.FATAL);
}
/**
Disable all logging requests of priority DEBUG regardless of
category. Invoking this method is equivalent to calling {@link
#disable} with the argument {@link Priority#DEBUG}.
@since 0.8.5 */
public
void disableDebug() {
disable(Priority.DEBUG);
}
/**
Disable all logging requests of priority INFO and below
regardless of category. Note that DEBUG messages are also
disabled.
<p>Invoking this method is equivalent to calling {@link #disable}
with the argument {@link Priority#INFO}.
@since 0.8.5 */
public
void disableInfo() {
disable(Priority.INFO);
}
/**
Undoes the effect of calling any of {@link #disable}, {@link
#disableAll}, {@link #disableDebug} and {@link #disableInfo}
methods. More precisely, invoking this method sets the Category
class internal variable called <code>disable</code> to its
default "off" value.
@since 0.8.5 */
public
void enableAll() {
disable = DISABLE_OFF;
}
/**
Override the shipped code flag if the <code>override</code>
parameter is not null.
<p>If <code>override</code> is null then there is nothing to do.
Otherwise, set Category.shippedCode to false if override has a
value other than "false".
*/
public
void overrideAsNeeded(String override) {
// If override is defined, any value other than false will be
// interpreted as true.
if(override != null) {
LogLog.debug("Handling non-null disable override directive: \""+
override +"\".");
if(OptionConverter.toBoolean(override, true)) {
LogLog.debug("Overriding all disable methods.");
disable = DISABLE_OVERRIDE;
}
}
}
/**
Return a new category instance named as the first parameter using
the default factory.
<p>If a category of that name already exists, then it will be
returned. Otherwise, a new category will be instantiated and
lthen inked with its existing ancestors as well as children.
@param name The name of the category to retreive.
*/
public
Category getInstance(String name) {
return getInstance(name, defaultFactory);
}
/**
Return a new category instance named as the first parameter using
<code>factory</code>.
<p>If a category of that name already exists, then it will be
returned. Otherwise, a new category will be instantiated by the
<code>factory</code> parameter and linked with its existing
ancestors as well as children.
@param name The name of the category to retreive.
@param factory The factory that will make the new category instance.
*/
public
Category getInstance(String name, CategoryFactory factory) {
//System.out.println("getInstance("+name+") called.");
CategoryKey key = new CategoryKey(name);
// Synchronize to prevent write conflicts. Read conflicts (in
// getChainedPriority method) are possible only if variable
// assignments are non-atomic.
Category category;
synchronized(ht) {
Object o = ht.get(key);
if(o == null) {
category = factory.makeNewCategoryInstance(name);
category.setHierarchy(this);
ht.put(key, category);
updateParents(category);
return category;
} else if(o instanceof Category) {
return (Category) o;
} else if (o instanceof ProvisionNode) {
//System.out.println("("+name+") ht.get(this) returned ProvisionNode");
category = factory.makeNewCategoryInstance(name);
category.setHierarchy(this);
ht.put(key, category);
updateChildren((ProvisionNode) o, category);
updateParents(category);
return category;
}
else {
// It should be impossible to arrive here
return null; // but let's keep the compiler happy.
}
}
}
/**
Returns all the currently defined categories in this hierarchy as
an {@link java.util.Enumeration Enumeration}.
<p>The root category is <em>not</em> included in the returned
{@link Enumeration}. */
public
Enumeration getCurrentCategories() {
// The accumlation in v is necessary because not all elements in
// ht are Category objects as there might be some ProvisionNodes
// as well.
Vector v = new Vector(ht.size());
Enumeration elems = ht.elements();
while(elems.hasMoreElements()) {
Object o = elems.nextElement();
if(o instanceof Category) {
v.addElement(o);
}
}
return v.elements();
}
public
boolean isDisabled(int level) {
return disable >= level;
}
/**
Get the renderer map for this hierarchy.
*/
public
RendererMap getRendererMap() {
return rendererMap;
}
/**
Get the root of this hierarchy.
@since 0.9.0
*/
public
Category getRoot() {
return root;
}
/**
Reset all values contained in this hierarchy instance to their
default. This removes all appenders from all categories, sets
the priority of all non-root categories to <code>null</code>,
sets their additivity flag to <code>true</code> and sets the priority
of the root category to {@link Priority#DEBUG DEBUG}. Moreover,
message disabling is set its default "off" value.
<p>Existing categories are not removed. They are just reset.
<p>This method should be used sparingly and with care as it will
block all logging until it is completed.</p>
@since version 0.8.5 */
public
void resetConfiguration() {
getRoot().setPriority(Priority.DEBUG);
root.setResourceBundle(null);
disable = Hierarchy.DISABLE_OFF;
// the synchronization is needed to prevent JDK 1.2.x hashtable
// surprises
synchronized(ht) {
shutdown(); // nested locks are OK
Enumeration cats = getCurrentCategories();
while(cats.hasMoreElements()) {
Category c = (Category) cats.nextElement();
c.setPriority(null);
c.setAdditivity(true);
c.setResourceBundle(null);
}
}
rendererMap.clear();
}
/**
Set the disable override value given a string.
@since 1.1
*/
public
void setDisableOverride(String override) {
if(OptionConverter.toBoolean(override, true)) {
LogLog.debug("Overriding disable.");
disable = DISABLE_OVERRIDE;
}
}
/**
Shutting down a hierarchy will <em>safely</em> close and remove
all appenders in all categories including the root category.
<p>Some appenders such as {@link org.apache.log4j.net.SocketAppender}
and {@link AsyncAppender} need to be closed before the
application exists. Otherwise, pending logging events might be
lost.
<p>The <code>shutdown</code> method is careful to close nested
appenders before closing regular appenders. This is allows
configurations where a regular appender is attached to a category
and again to a nested appender.
@since 1.0 */
public
void shutdown() {
Category root = getRoot();
// begin by closing nested appenders
root.closeNestedAppenders();
synchronized(ht) {
Enumeration cats = this.getCurrentCategories();
while(cats.hasMoreElements()) {
Category c = (Category) cats.nextElement();
c.closeNestedAppenders();
}
// then, remove all appenders
root.removeAllAppenders();
cats = this.getCurrentCategories();
while(cats.hasMoreElements()) {
Category c = (Category) cats.nextElement();
c.removeAllAppenders();
}
}
}
/**
This method loops through all the *potential* parents of
'cat'. There 3 possible cases:
1) No entry for the potential parent of 'cat' exists
We create a ProvisionNode for this potential parent and insert
'cat' in that provision node.
2) There entry is of type Category for the potential parent.
The entry is 'cat's nearest existing parent. We update cat's
parent field with this entry. We also break from the loop
because updating our parent's parent is our parent's
responsibility.
3) There entry is of type ProvisionNode for this potential parent.
We add 'cat' to the list of children for this potential parent.
*/
final
private
void updateParents(Category cat) {
String name = cat.name;
int length = name.length();
boolean parentFound = false;
//System.out.println("UpdateParents called for " + name);
// if name = "w.x.y.z", loop thourgh "w.x.y", "w.x" and "w", but not "w.x.y.z"
for(int i = name.lastIndexOf('.', length-1); i >= 0;
i = name.lastIndexOf('.', i-1)) {
String substr = name.substring(0, i);
//System.out.println("Updating parent : " + substr);
CategoryKey key = new CategoryKey(substr); // simple constructor
Object o = ht.get(key);
// Create a provision node for a future parent.
if(o == null) {
//System.out.println("No parent "+substr+" found. Creating ProvisionNode.");
ProvisionNode pn = new ProvisionNode(cat);
ht.put(key, pn);
} else if(o instanceof Category) {
parentFound = true;
cat.parent = (Category) o;
//System.out.println("Linking " + cat.name + " -> " + ((Category) o).name);
break; // no need to update the ancestors of the closest ancestor
} else if(o instanceof ProvisionNode) {
((ProvisionNode) o).addElement(cat);
} else {
Exception e = new IllegalStateException("unexpected object type " +
o.getClass() + " in ht.");
e.printStackTrace();
}
}
// If we could not find any existing parents, then link with root.
if(!parentFound)
cat.parent = root;
}
/**
We update the links for all the children that placed themselves
in the provision node 'pn'. The second argument 'cat' is a
reference for the newly created Category, parent of all the
children in 'pn'
We loop on all the children 'c' in 'pn':
If the child 'c' has been already linked to a child of
'cat' then there is no need to update 'c'.
Otherwise, we set cat's parent field to c's parent and set
c's parent field to cat.
*/
final
private
void updateChildren(ProvisionNode pn, Category cat) {
//System.out.println("updateChildren called for " + cat.name);
final int last = pn.size();
for(int i = 0; i < last; i++) {
Category c = (Category) pn.elementAt(i);
//System.out.println("Updating child " +p.name);
// Unless this child already points to a correct (lower) parent,
// make cat.parent point to c.parent and c.parent to cat.
if(!c.parent.name.startsWith(cat.name)) {
cat.parent = c.parent;
c.parent = cat;
}
}
}
}