XMLConfiguration.java

/*
 * 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.
 */

package org.apache.commons.configuration2;

import javax.xml.parsers.DocumentBuilder;
import javax.xml.parsers.DocumentBuilderFactory;
import javax.xml.parsers.ParserConfigurationException;
import javax.xml.transform.OutputKeys;
import javax.xml.transform.Result;
import javax.xml.transform.Source;
import javax.xml.transform.Transformer;
import javax.xml.transform.dom.DOMSource;
import javax.xml.transform.stream.StreamResult;
import java.io.IOException;
import java.io.InputStream;
import java.io.Reader;
import java.io.StringReader;
import java.io.StringWriter;
import java.io.Writer;
import java.net.URL;
import java.util.ArrayList;
import java.util.Collection;
import java.util.Collections;
import java.util.HashMap;
import java.util.Iterator;
import java.util.Map;

import org.apache.commons.configuration2.convert.ListDelimiterHandler;
import org.apache.commons.configuration2.ex.ConfigurationException;
import org.apache.commons.configuration2.io.ConfigurationLogger;
import org.apache.commons.configuration2.io.FileLocator;
import org.apache.commons.configuration2.io.FileLocatorAware;
import org.apache.commons.configuration2.io.InputStreamSupport;
import org.apache.commons.configuration2.resolver.DefaultEntityResolver;
import org.apache.commons.configuration2.tree.ImmutableNode;
import org.apache.commons.configuration2.tree.NodeTreeWalker;
import org.apache.commons.configuration2.tree.ReferenceNodeHandler;
import org.apache.commons.lang3.StringUtils;
import org.apache.commons.lang3.mutable.MutableObject;
import org.w3c.dom.Attr;
import org.w3c.dom.CDATASection;
import org.w3c.dom.Document;
import org.w3c.dom.Element;
import org.w3c.dom.NamedNodeMap;
import org.w3c.dom.Node;
import org.w3c.dom.NodeList;
import org.w3c.dom.Text;
import org.xml.sax.EntityResolver;
import org.xml.sax.InputSource;
import org.xml.sax.SAXException;
import org.xml.sax.SAXParseException;
import org.xml.sax.helpers.DefaultHandler;

/**
 * <p>
 * A specialized hierarchical configuration class that is able to parse XML documents.
 * </p>
 * <p>
 * The parsed document will be stored keeping its structure. The class also tries to preserve as much information from
 * the loaded XML document as possible, including comments and processing instructions. These will be contained in
 * documents created by the {@code save()} methods, too.
 * </p>
 * <p>
 * Like other file based configuration classes this class maintains the name and path to the loaded configuration file.
 * These properties can be altered using several setter methods, but they are not modified by {@code save()} and
 * {@code load()} methods. If XML documents contain relative paths to other documents (e.g. to a DTD), these references
 * are resolved based on the path set for this configuration.
 * </p>
 * <p>
 * By inheriting from {@link AbstractConfiguration} this class provides some extended functionality, e.g. interpolation
 * of property values. Like in {@link PropertiesConfiguration} property values can contain delimiter characters (the
 * comma ',' per default) and are then split into multiple values. This works for XML attributes and text content of
 * elements as well. The delimiter can be escaped by a backslash. As an example consider the following XML fragment:
 * </p>
 *
 * <pre>
 * &lt;config&gt;
 *   &lt;array&gt;10,20,30,40&lt;/array&gt;
 *   &lt;scalar&gt;3\,1415&lt;/scalar&gt;
 *   &lt;cite text="To be or not to be\, this is the question!"/&gt;
 * &lt;/config&gt;
 * </pre>
 *
 * <p>
 * Here the content of the {@code array} element will be split at the commas, so the {@code array} key will be assigned
 * 4 values. In the {@code scalar} property and the {@code text} attribute of the {@code cite} element the comma is
 * escaped, so that no splitting is performed.
 * </p>
 * <p>
 * The configuration API allows setting multiple values for a single attribute, e.g. something like the following is
 * legal (assuming that the default expression engine is used):
 * </p>
 *
 * <pre>
 * XMLConfiguration config = new XMLConfiguration();
 * config.addProperty(&quot;test.dir[@name]&quot;, &quot;C:\\Temp\\&quot;);
 * config.addProperty(&quot;test.dir[@name]&quot;, &quot;D:\\Data\\&quot;);
 * </pre>
 *
 * <p>
 * However, in XML such a constellation is not supported; an attribute can appear only once for a single element.
 * Therefore, an attempt to save a configuration which violates this condition will throw an exception.
 * </p>
 * <p>
 * Like other {@code Configuration} implementations, {@code XMLConfiguration} uses a {@link ListDelimiterHandler} object
 * for controlling list split operations. Per default, a list delimiter handler object is set which disables this
 * feature. XML has a built-in support for complex structures including list properties; therefore, list splitting is
 * not that relevant for this configuration type. Nevertheless, by setting an alternative {@code ListDelimiterHandler}
 * implementation, this feature can be enabled. It works as for any other concrete {@code Configuration} implementation.
 * </p>
 * <p>
 * Whitespace in the content of XML documents is trimmed per default. In most cases this is desired. However, sometimes
 * whitespace is indeed important and should be treated as part of the value of a property as in the following example:
 * </p>
 *
 * <pre>
 *   &lt;indent&gt;    &lt;/indent&gt;
 * </pre>
 *
 * <p>
 * Per default the spaces in the {@code indent} element will be trimmed resulting in an empty element. To tell
 * {@code XMLConfiguration} that spaces are relevant the {@code xml:space} attribute can be used, which is defined in
 * the <a href="http://www.w3.org/TR/REC-xml/#sec-white-space">XML specification</a>. This will look as follows:
 * </p>
 *
 * <pre>
 *   &lt;indent <strong>xml:space=&quot;preserve&quot;</strong>&gt;    &lt;/indent&gt;
 * </pre>
 *
 * <p>
 * The value of the {@code indent} property will now contain the spaces.
 * </p>
 * <p>
 * {@code XMLConfiguration} implements the {@link FileBasedConfiguration} interface and thus can be used together with a
 * file-based builder to load XML configuration files from various sources like files, URLs, or streams.
 * </p>
 * <p>
 * Like other {@code Configuration} implementations, this class uses a {@code Synchronizer} object to control concurrent
 * access. By choosing a suitable implementation of the {@code Synchronizer} interface, an instance can be made
 * thread-safe or not. Note that access to most of the properties typically set through a builder is not protected by
 * the {@code Synchronizer}. The intended usage is that these properties are set once at construction time through the
 * builder and after that remain constant. If you wish to change such properties during life time of an instance, you
 * have to use the {@code lock()} and {@code unlock()} methods manually to ensure that other threads see your changes.
 * </p>
 * <p>
 * More information about the basic functionality supported by {@code XMLConfiguration} can be found at the user's guide
 * at <a href="https://commons.apache.org/proper/commons-configuration/userguide/howto_basicfeatures.html"> Basic
 * features and AbstractConfiguration</a>. There is also a separate chapter dealing with
 * <a href="commons.apache.org/proper/commons-configuration/userguide/howto_xml.html"> XML Configurations</a> in
 * special.
 * </p>
 *
 * @since commons-configuration 1.0
 */
public class XMLConfiguration extends BaseHierarchicalConfiguration implements FileBasedConfiguration, FileLocatorAware, InputStreamSupport {
    /** Constant for the default indent size. */
    static final int DEFAULT_INDENT_SIZE = 2;

    /** Constant for output property name used on a transformer to specify the indent amount. */
    static final String INDENT_AMOUNT_PROPERTY = "{http://xml.apache.org/xslt}indent-amount";

    /** Constant for the default root element name. */
    private static final String DEFAULT_ROOT_NAME = "configuration";

    /** Constant for the name of the space attribute. */
    private static final String ATTR_SPACE = "xml:space";

    /** Constant for an internally used space attribute. */
    private static final String ATTR_SPACE_INTERNAL = "config-xml:space";

    /** Constant for the xml:space value for preserving whitespace. */
    private static final String VALUE_PRESERVE = "preserve";

    /** Schema Langauge key for the parser */
    private static final String JAXP_SCHEMA_LANGUAGE = "http://java.sun.com/xml/jaxp/properties/schemaLanguage";

    /** Schema Language for the parser */
    private static final String W3C_XML_SCHEMA = "http://www.w3.org/2001/XMLSchema";

    /** Stores the name of the root element. */
    private String rootElementName;

    /** Stores the public ID from the DOCTYPE. */
    private String publicID;

    /** Stores the system ID from the DOCTYPE. */
    private String systemID;

    /** Stores the document builder that should be used for loading. */
    private DocumentBuilder documentBuilder;

    /** Stores a flag whether DTD or Schema validation should be performed. */
    private boolean validating;

    /** Stores a flag whether DTD or Schema validation is used */
    private boolean schemaValidation;

    /** The EntityResolver to use */
    private EntityResolver entityResolver = new DefaultEntityResolver();

    /** The current file locator. */
    private FileLocator locator;

    /**
     * Creates a new instance of {@code XMLConfiguration}.
     */
    public XMLConfiguration() {
        initLogger(new ConfigurationLogger(XMLConfiguration.class));
    }

    /**
     * Creates a new instance of {@code XMLConfiguration} and copies the content of the passed in configuration into this
     * object. Note that only the data of the passed in configuration will be copied. If, for instance, the other
     * configuration is a {@code XMLConfiguration}, too, things like comments or processing instructions will be lost.
     *
     * @param c the configuration to copy
     * @since 1.4
     */
    public XMLConfiguration(final HierarchicalConfiguration<ImmutableNode> c) {
        super(c);
        rootElementName = c != null ? c.getRootElementName() : null;
        initLogger(new ConfigurationLogger(XMLConfiguration.class));
    }

    /**
     * Returns the name of the root element. If this configuration was loaded from a XML document, the name of this
     * document's root element is returned. Otherwise it is possible to set a name for the root element that will be used
     * when this configuration is stored.
     *
     * @return the name of the root element
     */
    @Override
    protected String getRootElementNameInternal() {
        final Document doc = getDocument();
        if (doc == null) {
            return rootElementName == null ? DEFAULT_ROOT_NAME : rootElementName;
        }
        return doc.getDocumentElement().getNodeName();
    }

    /**
     * Sets the name of the root element. This name is used when this configuration object is stored in an XML file. Note
     * that setting the name of the root element works only if this configuration has been newly created. If the
     * configuration was loaded from an XML file, the name cannot be changed and an {@code UnsupportedOperationException}
     * exception is thrown. Whether this configuration has been loaded from an XML document or not can be found out using
     * the {@code getDocument()} method.
     *
     * @param name the name of the root element
     */
    public void setRootElementName(final String name) {
        beginRead(true);
        try {
            if (getDocument() != null) {
                throw new UnsupportedOperationException("The name of the root element " + "cannot be changed when loaded from an XML document!");
            }
            rootElementName = name;
        } finally {
            endRead();
        }
    }

    /**
     * Returns the {@code DocumentBuilder} object that is used for loading documents. If no specific builder has been set,
     * this method returns <b>null</b>.
     *
     * @return the {@code DocumentBuilder} for loading new documents
     * @since 1.2
     */
    public DocumentBuilder getDocumentBuilder() {
        return documentBuilder;
    }

    /**
     * Sets the {@code DocumentBuilder} object to be used for loading documents. This method makes it possible to specify
     * the exact document builder. So an application can create a builder, configure it for its special needs, and then pass
     * it to this method.
     *
     * @param documentBuilder the document builder to be used; if undefined, a default builder will be used
     * @since 1.2
     */
    public void setDocumentBuilder(final DocumentBuilder documentBuilder) {
        this.documentBuilder = documentBuilder;
    }

    /**
     * Returns the public ID of the DOCTYPE declaration from the loaded XML document. This is <b>null</b> if no document has
     * been loaded yet or if the document does not contain a DOCTYPE declaration with a public ID.
     *
     * @return the public ID
     * @since 1.3
     */
    public String getPublicID() {
        beginRead(false);
        try {
            return publicID;
        } finally {
            endRead();
        }
    }

    /**
     * Sets the public ID of the DOCTYPE declaration. When this configuration is saved, a DOCTYPE declaration will be
     * constructed that contains this public ID.
     *
     * @param publicID the public ID
     * @since 1.3
     */
    public void setPublicID(final String publicID) {
        beginWrite(false);
        try {
            this.publicID = publicID;
        } finally {
            endWrite();
        }
    }

    /**
     * Returns the system ID of the DOCTYPE declaration from the loaded XML document. This is <b>null</b> if no document has
     * been loaded yet or if the document does not contain a DOCTYPE declaration with a system ID.
     *
     * @return the system ID
     * @since 1.3
     */
    public String getSystemID() {
        beginRead(false);
        try {
            return systemID;
        } finally {
            endRead();
        }
    }

    /**
     * Sets the system ID of the DOCTYPE declaration. When this configuration is saved, a DOCTYPE declaration will be
     * constructed that contains this system ID.
     *
     * @param systemID the system ID
     * @since 1.3
     */
    public void setSystemID(final String systemID) {
        beginWrite(false);
        try {
            this.systemID = systemID;
        } finally {
            endWrite();
        }
    }

    /**
     * Returns the value of the validating flag.
     *
     * @return the validating flag
     * @since 1.2
     */
    public boolean isValidating() {
        return validating;
    }

    /**
     * Sets the value of the validating flag. This flag determines whether DTD/Schema validation should be performed when
     * loading XML documents. This flag is evaluated only if no custom {@code DocumentBuilder} was set.
     *
     * @param validating the validating flag
     * @since 1.2
     */
    public void setValidating(final boolean validating) {
        if (!schemaValidation) {
            this.validating = validating;
        }
    }

    /**
     * Returns the value of the schemaValidation flag.
     *
     * @return the schemaValidation flag
     * @since 1.7
     */
    public boolean isSchemaValidation() {
        return schemaValidation;
    }

    /**
     * Sets the value of the schemaValidation flag. This flag determines whether DTD or Schema validation should be used.
     * This flag is evaluated only if no custom {@code DocumentBuilder} was set. If set to true the XML document must
     * contain a schemaLocation definition that provides resolvable hints to the required schemas.
     *
     * @param schemaValidation the validating flag
     * @since 1.7
     */
    public void setSchemaValidation(final boolean schemaValidation) {
        this.schemaValidation = schemaValidation;
        if (schemaValidation) {
            this.validating = true;
        }
    }

    /**
     * Sets a new EntityResolver. Setting this will cause RegisterEntityId to have no effect.
     *
     * @param resolver The EntityResolver to use.
     * @since 1.7
     */
    public void setEntityResolver(final EntityResolver resolver) {
        this.entityResolver = resolver;
    }

    /**
     * Returns the EntityResolver.
     *
     * @return The EntityResolver.
     * @since 1.7
     */
    public EntityResolver getEntityResolver() {
        return this.entityResolver;
    }

    /**
     * Returns the XML document this configuration was loaded from. The return value is <b>null</b> if this configuration
     * was not loaded from a XML document.
     *
     * @return the XML document this configuration was loaded from
     */
    public Document getDocument() {
        final XMLDocumentHelper docHelper = getDocumentHelper();
        return docHelper != null ? docHelper.getDocument() : null;
    }

    /**
     * Returns the helper object for managing the underlying document.
     *
     * @return the {@code XMLDocumentHelper}
     */
    private XMLDocumentHelper getDocumentHelper() {
        final ReferenceNodeHandler handler = getReferenceHandler();
        return (XMLDocumentHelper) handler.getReference(handler.getRootNode());
    }

    /**
     * Returns the extended node handler with support for references.
     *
     * @return the {@code ReferenceNodeHandler}
     */
    private ReferenceNodeHandler getReferenceHandler() {
        return getSubConfigurationParentModel().getReferenceNodeHandler();
    }

    /**
     * Initializes this configuration from an XML document.
     *
     * @param docHelper the helper object with the document to be parsed
     * @param elemRefs a flag whether references to the XML elements should be set
     */
    private void initProperties(final XMLDocumentHelper docHelper, final boolean elemRefs) {
        final Document document = docHelper.getDocument();
        setPublicID(docHelper.getSourcePublicID());
        setSystemID(docHelper.getSourceSystemID());

        final ImmutableNode.Builder rootBuilder = new ImmutableNode.Builder();
        final MutableObject<String> rootValue = new MutableObject<>();
        final Map<ImmutableNode, Object> elemRefMap = elemRefs ? new HashMap<>() : null;
        final Map<String, String> attributes = constructHierarchy(rootBuilder, rootValue, document.getDocumentElement(), elemRefMap, true, 0);
        attributes.remove(ATTR_SPACE_INTERNAL);
        final ImmutableNode top = rootBuilder.value(rootValue.getValue()).addAttributes(attributes).create();
        getSubConfigurationParentModel().mergeRoot(top, document.getDocumentElement().getTagName(), elemRefMap, elemRefs ? docHelper : null, this);
    }

    /**
     * Helper method for building the internal storage hierarchy. The XML elements are transformed into node objects.
     *
     * @param node a builder for the current node
     * @param refValue stores the text value of the element
     * @param element the current XML element
     * @param elemRefs a map for assigning references objects to nodes; can be <b>null</b>, then reference objects are
     *        irrelevant
     * @param trim a flag whether the text content of elements should be trimmed; this controls the whitespace handling
     * @param level the current level in the hierarchy
     * @return a map with all attribute values extracted for the current node; this map also contains the value of the trim
     *         flag for this node under the key {@value #ATTR_SPACE}
     */
    private Map<String, String> constructHierarchy(final ImmutableNode.Builder node, final MutableObject<String> refValue, final Element element,
        final Map<ImmutableNode, Object> elemRefs, final boolean trim, final int level) {
        final boolean trimFlag = shouldTrim(element, trim);
        final Map<String, String> attributes = processAttributes(element);
        attributes.put(ATTR_SPACE_INTERNAL, String.valueOf(trimFlag));
        final StringBuilder buffer = new StringBuilder();
        final NodeList list = element.getChildNodes();
        boolean hasChildren = false;

        for (int i = 0; i < list.getLength(); i++) {
            final org.w3c.dom.Node w3cNode = list.item(i);
            if (w3cNode instanceof Element) {
                final Element child = (Element) w3cNode;
                final ImmutableNode.Builder childNode = new ImmutableNode.Builder();
                childNode.name(child.getTagName());
                final MutableObject<String> refChildValue = new MutableObject<>();
                final Map<String, String> attrmap = constructHierarchy(childNode, refChildValue, child, elemRefs, trimFlag, level + 1);
                final boolean childTrim = Boolean.parseBoolean(attrmap.remove(ATTR_SPACE_INTERNAL));
                childNode.addAttributes(attrmap);
                final ImmutableNode newChild = createChildNodeWithValue(node, childNode, child, refChildValue.getValue(), childTrim, attrmap, elemRefs);
                if (elemRefs != null && !elemRefs.containsKey(newChild)) {
                    elemRefs.put(newChild, child);
                }
                hasChildren = true;
            } else if (w3cNode instanceof Text) {
                final Text data = (Text) w3cNode;
                buffer.append(data.getData());
            }
        }

        boolean childrenFlag = false;
        if (hasChildren || trimFlag) {
            childrenFlag = hasChildren || attributes.size() > 1;
        }
        final String text = determineValue(buffer.toString(), childrenFlag, trimFlag);
        if (!text.isEmpty() || (!childrenFlag && level != 0)) {
            refValue.setValue(text);
        }
        return attributes;
    }

    /**
     * Determines the value of a configuration node. This method mainly checks whether the text value is to be trimmed or
     * not. This is normally defined by the trim flag. However, if the node has children and its content is only whitespace,
     * then it makes no sense to store any value; this would only scramble layout when the configuration is saved again.
     *
     * @param content the text content of this node
     * @param hasChildren a flag whether the node has children
     * @param trimFlag the trim flag
     * @return the value to be stored for this node
     */
    private static String determineValue(final String content, final boolean hasChildren, final boolean trimFlag) {
        final boolean shouldTrim = trimFlag || (StringUtils.isBlank(content) && hasChildren);
        return shouldTrim ? content.trim() : content;
    }

    /**
     * Helper method for initializing the attributes of a configuration node from the given XML element.
     *
     * @param element the current XML element
     * @return a map with all attribute values extracted for the current node
     */
    private static Map<String, String> processAttributes(final Element element) {
        final NamedNodeMap attributes = element.getAttributes();
        final Map<String, String> attrmap = new HashMap<>();

        for (int i = 0; i < attributes.getLength(); ++i) {
            final org.w3c.dom.Node w3cNode = attributes.item(i);
            if (w3cNode instanceof Attr) {
                final Attr attr = (Attr) w3cNode;
                attrmap.put(attr.getName(), attr.getValue());
            }
        }

        return attrmap;
    }

    /**
     * Creates a new child node, assigns its value, and adds it to its parent. This method also deals with elements whose
     * value is a list. In this case multiple child elements must be added. The return value is the first child node which
     * was added.
     *
     * @param parent the builder for the parent element
     * @param child the builder for the child element
     * @param elem the associated XML element
     * @param value the value of the child element
     * @param trim flag whether texts of elements should be trimmed
     * @param attrmap a map with the attributes of the current node
     * @param elemRefs a map for assigning references objects to nodes; can be <b>null</b>, then reference objects are
     *        irrelevant
     * @return the first child node added to the parent
     */
    private ImmutableNode createChildNodeWithValue(final ImmutableNode.Builder parent, final ImmutableNode.Builder child, final Element elem,
        final String value, final boolean trim, final Map<String, String> attrmap, final Map<ImmutableNode, Object> elemRefs) {
        final ImmutableNode addedChildNode;
        final Collection<String> values;

        if (value != null) {
            values = getListDelimiterHandler().split(value, trim);
        } else {
            values = Collections.emptyList();
        }

        if (values.size() > 1) {
            final Map<ImmutableNode, Object> refs = isSingleElementList(elem) ? elemRefs : null;
            final Iterator<String> it = values.iterator();
            // Create new node for the original child's first value
            child.value(it.next());
            addedChildNode = child.create();
            parent.addChild(addedChildNode);
            XMLListReference.assignListReference(refs, addedChildNode, elem);

            // add multiple new children
            while (it.hasNext()) {
                final ImmutableNode.Builder c = new ImmutableNode.Builder();
                c.name(addedChildNode.getNodeName());
                c.value(it.next());
                c.addAttributes(attrmap);
                final ImmutableNode newChild = c.create();
                parent.addChild(newChild);
                XMLListReference.assignListReference(refs, newChild, null);
            }
        } else if (values.size() == 1) {
            // we will have to replace the value because it might
            // contain escaped delimiters
            child.value(values.iterator().next());
            addedChildNode = child.create();
            parent.addChild(addedChildNode);
        } else {
            addedChildNode = child.create();
            parent.addChild(addedChildNode);
        }

        return addedChildNode;
    }

    /**
     * Checks whether an element defines a complete list. If this is the case, extended list handling can be applied.
     *
     * @param element the element to be checked
     * @return a flag whether this is the only element defining the list
     */
    private static boolean isSingleElementList(final Element element) {
        final Node parentNode = element.getParentNode();
        return countChildElements(parentNode, element.getTagName()) == 1;
    }

    /**
     * Determines the number of child elements of this given node with the specified node name.
     *
     * @param parent the parent node
     * @param name the name in question
     * @return the number of child elements with this name
     */
    private static int countChildElements(final Node parent, final String name) {
        final NodeList childNodes = parent.getChildNodes();
        int count = 0;
        for (int i = 0; i < childNodes.getLength(); i++) {
            final Node item = childNodes.item(i);
            if (item instanceof Element && name.equals(((Element) item).getTagName())) {
                count++;
            }
        }
        return count;
    }

    /**
     * Checks whether the content of the current XML element should be trimmed. This method checks whether a
     * {@code xml:space} attribute is present and evaluates its value. See
     * <a href="http://www.w3.org/TR/REC-xml/#sec-white-space"> http://www.w3.org/TR/REC-xml/#sec-white-space</a> for more
     * details.
     *
     * @param element the current XML element
     * @param currentTrim the current trim flag
     * @return a flag whether the content of this element should be trimmed
     */
    private static boolean shouldTrim(final Element element, final boolean currentTrim) {
        final Attr attr = element.getAttributeNode(ATTR_SPACE);

        if (attr == null) {
            return currentTrim;
        }
        return !VALUE_PRESERVE.equals(attr.getValue());
    }

    /**
     * Creates the {@code DocumentBuilder} to be used for loading files. This implementation checks whether a specific
     * {@code DocumentBuilder} has been set. If this is the case, this one is used. Otherwise a default builder is created.
     * Depending on the value of the validating flag this builder will be a validating or a non validating
     * {@code DocumentBuilder}.
     *
     * @return the {@code DocumentBuilder} for loading configuration files
     * @throws ParserConfigurationException if an error occurs
     * @since 1.2
     */
    protected DocumentBuilder createDocumentBuilder() throws ParserConfigurationException {
        if (getDocumentBuilder() != null) {
            return getDocumentBuilder();
        }
        final DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
        if (isValidating()) {
            factory.setValidating(true);
            if (isSchemaValidation()) {
                factory.setNamespaceAware(true);
                factory.setAttribute(JAXP_SCHEMA_LANGUAGE, W3C_XML_SCHEMA);
            }
        }

        final DocumentBuilder result = factory.newDocumentBuilder();
        result.setEntityResolver(this.entityResolver);

        if (isValidating()) {
            // register an error handler which detects validation errors
            result.setErrorHandler(new DefaultHandler() {
                @Override
                public void error(final SAXParseException ex) throws SAXException {
                    throw ex;
                }
            });
        }
        return result;
    }

    /**
     * Creates and initializes the transformer used for save operations. This base implementation initializes all of the
     * default settings like indentation mode and the DOCTYPE. Derived classes may overload this method if they have
     * specific needs.
     *
     * @return the transformer to use for a save operation
     * @throws ConfigurationException if an error occurs
     * @since 1.3
     */
    protected Transformer createTransformer() throws ConfigurationException {
        final Transformer transformer = XMLDocumentHelper.createTransformer();

        transformer.setOutputProperty(OutputKeys.INDENT, "yes");
        transformer.setOutputProperty(INDENT_AMOUNT_PROPERTY, Integer.toString(DEFAULT_INDENT_SIZE));
        if (locator != null && locator.getEncoding() != null) {
            transformer.setOutputProperty(OutputKeys.ENCODING, locator.getEncoding());
        }
        if (publicID != null) {
            transformer.setOutputProperty(OutputKeys.DOCTYPE_PUBLIC, publicID);
        }
        if (systemID != null) {
            transformer.setOutputProperty(OutputKeys.DOCTYPE_SYSTEM, systemID);
        }

        return transformer;
    }

    /**
     * Creates a DOM document from the internal tree of configuration nodes.
     *
     * @return the new document
     * @throws ConfigurationException if an error occurs
     */
    private Document createDocument() throws ConfigurationException {
        final ReferenceNodeHandler handler = getReferenceHandler();
        final XMLDocumentHelper docHelper = (XMLDocumentHelper) handler.getReference(handler.getRootNode());
        final XMLDocumentHelper newHelper = docHelper == null ? XMLDocumentHelper.forNewDocument(getRootElementName()) : docHelper.createCopy();

        final XMLBuilderVisitor builder = new XMLBuilderVisitor(newHelper, getListDelimiterHandler());
        builder.handleRemovedNodes(handler);
        builder.processDocument(handler);
        initRootElementText(newHelper.getDocument(), getModel().getNodeHandler().getRootNode().getValue());
        return newHelper.getDocument();
    }

    /**
     * Sets the text of the root element of a newly created XML Document.
     *
     * @param doc the document
     * @param value the new text to be set
     */
    private void initRootElementText(final Document doc, final Object value) {
        final Element elem = doc.getDocumentElement();
        final NodeList children = elem.getChildNodes();

        // Remove all existing text nodes
        for (int i = 0; i < children.getLength(); i++) {
            final org.w3c.dom.Node nd = children.item(i);
            if (nd.getNodeType() == org.w3c.dom.Node.TEXT_NODE) {
                elem.removeChild(nd);
            }
        }

        if (value != null) {
            // Add a new text node
            elem.appendChild(doc.createTextNode(String.valueOf(value)));
        }
    }

    /**
     * {@inheritDoc} Stores the passed in locator for the upcoming IO operation.
     */
    @Override
    public void initFileLocator(final FileLocator loc) {
        locator = loc;
    }

    /**
     * Loads the configuration from the given reader. Note that the {@code clear()} method is not called, so the properties
     * contained in the loaded file will be added to the current set of properties.
     *
     * @param in the reader
     * @throws ConfigurationException if an error occurs
     * @throws IOException if an IO error occurs
     */
    @Override
    public void read(final Reader in) throws ConfigurationException, IOException {
        load(new InputSource(in));
    }

    /**
     * Loads the configuration from the given input stream. This is analogous to {@link #read(Reader)}, but data is read
     * from a stream. Note that this method will be called most time when reading an XML configuration source. By reading
     * XML documents directly from an input stream, the file's encoding can be correctly dealt with.
     *
     * @param in the input stream
     * @throws ConfigurationException if an error occurs
     * @throws IOException if an IO error occurs
     */
    @Override
    public void read(final InputStream in) throws ConfigurationException, IOException {
        load(new InputSource(in));
    }

    /**
     * Loads a configuration file from the specified input source.
     *
     * @param source the input source
     * @throws ConfigurationException if an error occurs
     */
    private void load(final InputSource source) throws ConfigurationException {
        if (locator == null) {
            throw new ConfigurationException(
                "Load operation not properly " + "initialized! Do not call read(InputStream) directly," + " but use a FileHandler to load a configuration.");
        }

        try {
            final URL sourceURL = locator.getSourceURL();
            if (sourceURL != null) {
                source.setSystemId(sourceURL.toString());
            }

            final DocumentBuilder builder = createDocumentBuilder();
            final Document newDocument = builder.parse(source);
            final Document oldDocument = getDocument();
            initProperties(XMLDocumentHelper.forSourceDocument(newDocument), oldDocument == null);
        } catch (final SAXParseException spe) {
            throw new ConfigurationException("Error parsing " + source.getSystemId(), spe);
        } catch (final Exception e) {
            this.getLogger().debug("Unable to load the configuration: " + e);
            throw new ConfigurationException("Unable to load the configuration", e);
        }
    }

    /**
     * Saves the configuration to the specified writer.
     *
     * @param writer the writer used to save the configuration
     * @throws ConfigurationException if an error occurs
     * @throws IOException if an IO error occurs
     */
    @Override
    public void write(final Writer writer) throws ConfigurationException, IOException {
        write(writer, createTransformer());
    }

    /**
     * Saves the configuration to the specified writer.
     *
     * @param writer the writer used to save the configuration.
     * @param transformer How to transform this configuration.
     * @throws ConfigurationException if an error occurs.
     * @since 2.7.0
     */
    public void write(final Writer writer, final Transformer transformer) throws ConfigurationException {
        final Source source = new DOMSource(createDocument());
        final Result result = new StreamResult(writer);
        XMLDocumentHelper.transform(transformer, source, result);
    }

    /**
     * Validate the document against the Schema.
     *
     * @throws ConfigurationException if the validation fails.
     */
    public void validate() throws ConfigurationException {
        beginWrite(false);
        try {
            final Transformer transformer = createTransformer();
            final Source source = new DOMSource(createDocument());
            final StringWriter writer = new StringWriter();
            final Result result = new StreamResult(writer);
            XMLDocumentHelper.transform(transformer, source, result);
            final Reader reader = new StringReader(writer.getBuffer().toString());
            final DocumentBuilder builder = createDocumentBuilder();
            builder.parse(new InputSource(reader));
        } catch (final SAXException | IOException | ParserConfigurationException pce) {
            throw new ConfigurationException("Validation failed", pce);
        } finally {
            endWrite();
        }
    }

    /**
     * A concrete {@code BuilderVisitor} that can construct XML documents.
     */
    static class XMLBuilderVisitor extends BuilderVisitor {
        /** Stores the document to be constructed. */
        private final Document document;

        /** The element mapping. */
        private final Map<Node, Node> elementMapping;

        /** A mapping for the references for new nodes. */
        private final Map<ImmutableNode, Element> newElements;

        /** Stores the list delimiter handler . */
        private final ListDelimiterHandler listDelimiterHandler;

        /**
         * Creates a new instance of {@code XMLBuilderVisitor}.
         *
         * @param docHelper the document helper
         * @param handler the delimiter handler for properties with multiple values
         */
        public XMLBuilderVisitor(final XMLDocumentHelper docHelper, final ListDelimiterHandler handler) {
            document = docHelper.getDocument();
            elementMapping = docHelper.getElementMapping();
            listDelimiterHandler = handler;
            newElements = new HashMap<>();
        }

        /**
         * Processes the specified document, updates element values, and adds new nodes to the hierarchy.
         *
         * @param refHandler the {@code ReferenceNodeHandler}
         */
        public void processDocument(final ReferenceNodeHandler refHandler) {
            updateAttributes(refHandler.getRootNode(), document.getDocumentElement());
            NodeTreeWalker.INSTANCE.walkDFS(refHandler.getRootNode(), this, refHandler);
        }

        /**
         * Updates the current XML document regarding removed nodes. The elements associated with removed nodes are removed from
         * the document.
         *
         * @param refHandler the {@code ReferenceNodeHandler}
         */
        public void handleRemovedNodes(final ReferenceNodeHandler refHandler) {
            for (final Object ref : refHandler.removedReferences()) {
                if (ref instanceof Node) {
                    final Node removedElem = (Node) ref;
                    removeReference((Element) elementMapping.get(removedElem));
                }
            }
        }

        /**
         * {@inheritDoc} This implementation ensures that the correct XML element is created and inserted between the given
         * siblings.
         */
        @Override
        protected void insert(final ImmutableNode newNode, final ImmutableNode parent, final ImmutableNode sibling1, final ImmutableNode sibling2,
            final ReferenceNodeHandler refHandler) {
            if (XMLListReference.isListNode(newNode, refHandler)) {
                return;
            }

            final Element elem = document.createElement(newNode.getNodeName());
            newElements.put(newNode, elem);
            updateAttributes(newNode, elem);
            if (newNode.getValue() != null) {
                final String txt = String.valueOf(listDelimiterHandler.escape(newNode.getValue(), ListDelimiterHandler.NOOP_TRANSFORMER));
                elem.appendChild(document.createTextNode(txt));
            }
            if (sibling2 == null) {
                getElement(parent, refHandler).appendChild(elem);
            } else if (sibling1 != null) {
                getElement(parent, refHandler).insertBefore(elem, getElement(sibling1, refHandler).getNextSibling());
            } else {
                getElement(parent, refHandler).insertBefore(elem, getElement(parent, refHandler).getFirstChild());
            }
        }

        /**
         * {@inheritDoc} This implementation determines the XML element associated with the given node. Then this element's
         * value and attributes are set accordingly.
         */
        @Override
        protected void update(final ImmutableNode node, final Object reference, final ReferenceNodeHandler refHandler) {
            if (XMLListReference.isListNode(node, refHandler)) {
                if (XMLListReference.isFirstListItem(node, refHandler)) {
                    final String value = XMLListReference.listValue(node, refHandler, listDelimiterHandler);
                    updateElement(node, refHandler, value);
                }
            } else {
                final Object value = listDelimiterHandler.escape(refHandler.getValue(node), ListDelimiterHandler.NOOP_TRANSFORMER);
                updateElement(node, refHandler, value);
            }
        }

        private void updateElement(final ImmutableNode node, final ReferenceNodeHandler refHandler, final Object value) {
            final Element element = getElement(node, refHandler);
            updateElement(element, value);
            updateAttributes(node, element);
        }

        /**
         * Updates the node's value if it represents an element node.
         *
         * @param element the element
         * @param value the new value
         */
        private void updateElement(final Element element, final Object value) {
            Text txtNode = findTextNodeForUpdate(element);
            if (value == null) {
                // remove text
                if (txtNode != null) {
                    element.removeChild(txtNode);
                }
            } else {
                final String newValue = String.valueOf(value);
                if (txtNode == null) {
                    txtNode = document.createTextNode(newValue);
                    if (element.getFirstChild() != null) {
                        element.insertBefore(txtNode, element.getFirstChild());
                    } else {
                        element.appendChild(txtNode);
                    }
                } else {
                    txtNode.setNodeValue(newValue);
                }
            }
        }

        /**
         * Updates the associated XML elements when a node is removed.
         *
         * @param element the element to be removed
         */
        private void removeReference(final Element element) {
            final org.w3c.dom.Node parentElem = element.getParentNode();
            if (parentElem != null) {
                parentElem.removeChild(element);
            }
        }

        /**
         * Helper method for accessing the element of the specified node.
         *
         * @param node the node
         * @param refHandler the {@code ReferenceNodeHandler}
         * @return the element of this node
         */
        private Element getElement(final ImmutableNode node, final ReferenceNodeHandler refHandler) {
            final Element elementNew = newElements.get(node);
            if (elementNew != null) {
                return elementNew;
            }

            // special treatment for root node of the hierarchy
            final Object reference = refHandler.getReference(node);
            final Node element;
            if (reference instanceof XMLDocumentHelper) {
                element = ((XMLDocumentHelper) reference).getDocument().getDocumentElement();
            } else if (reference instanceof XMLListReference) {
                element = ((XMLListReference) reference).getElement();
            } else {
                element = (Node) reference;
            }
            return element != null ? (Element) elementMapping.get(element) : document.getDocumentElement();
        }

        /**
         * Helper method for updating the values of all attributes of the specified node.
         *
         * @param node the affected node
         * @param elem the element that is associated with this node
         */
        private static void updateAttributes(final ImmutableNode node, final Element elem) {
            if (node != null && elem != null) {
                clearAttributes(elem);
                for (final Map.Entry<String, Object> e : node.getAttributes().entrySet()) {
                    if (e.getValue() != null) {
                        elem.setAttribute(e.getKey(), e.getValue().toString());
                    }
                }
            }
        }

        /**
         * Removes all attributes of the given element.
         *
         * @param elem the element
         */
        private static void clearAttributes(final Element elem) {
            final NamedNodeMap attributes = elem.getAttributes();
            for (int i = 0; i < attributes.getLength(); i++) {
                elem.removeAttribute(attributes.item(i).getNodeName());
            }
        }

        /**
         * Returns the only text node of an element for update. This method is called when the element's text changes. Then all
         * text nodes except for the first are removed. A reference to the first is returned or <b>null</b> if there is no text
         * node at all.
         *
         * @param elem the element
         * @return the first and only text node
         */
        private static Text findTextNodeForUpdate(final Element elem) {
            Text result = null;
            // Find all Text nodes
            final NodeList children = elem.getChildNodes();
            final Collection<org.w3c.dom.Node> textNodes = new ArrayList<>();
            for (int i = 0; i < children.getLength(); i++) {
                final org.w3c.dom.Node nd = children.item(i);
                if (nd instanceof Text) {
                    if (result == null) {
                        result = (Text) nd;
                    } else {
                        textNodes.add(nd);
                    }
                }
            }

            // We don't want CDATAs
            if (result instanceof CDATASection) {
                textNodes.add(result);
                result = null;
            }

            // Remove all but the first Text node
            for (final org.w3c.dom.Node tn : textNodes) {
                elem.removeChild(tn);
            }
            return result;
        }
    }
}