UnionCombiner.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.tree;

import java.util.LinkedList;
import java.util.List;

/**
 * <p>
 * A specialized implementation of the {@code NodeCombiner} interface that constructs a union from two passed in node
 * hierarchies.
 * </p>
 * <p>
 * The given source hierarchies are traversed, and their nodes are added to the resulting structure. Under some
 * circumstances two nodes can be combined rather than adding both. This is the case if both nodes are single children
 * (no lists) of their parents and do not have values. The corresponding check is implemented in the
 * {@code findCombineNode()} method.
 * </p>
 * <p>
 * Sometimes it is not possible for this combiner to detect whether two nodes can be combined or not. Consider the
 * following two node hierarchies:
 * </p>
 *
 * <pre>
 * Hierarchy 1:
 *
 * Database
 *   +--Tables
 *        +--Table
 *             +--name [users]
 *             +--fields
 *                   +--field
 *                   |    +--name [uid]
 *                   +--field
 *                   |    +--name [usrname]
 *                     ...
 * </pre>
 *
 * <pre>
 * Hierarchy 2:
 *
 * Database
 *   +--Tables
 *        +--Table
 *             +--name [documents]
 *             +--fields
 *                   +--field
 *                   |    +--name [docid]
 *                   +--field
 *                   |    +--name [docname]
 *                     ...
 * </pre>
 *
 * <p>
 * Both hierarchies contain data about database tables. Each describes a single table. If these hierarchies are to be
 * combined, the result should probably look like the following:
 * </p>
 *
 * <pre>
 * Database
 *   +--Tables
 *        +--Table
 *        |    +--name [users]
 *        |    +--fields
 *        |          +--field
 *        |          |    +--name [uid]
 *        |            ...
 *        +--Table
 *             +--name [documents]
 *             +--fields
 *                   +--field
 *                   |    +--name [docid]
 *                     ...
 * </pre>
 *
 * <p>
 * i.e. the {@code Tables} nodes should be combined, while the {@code Table} nodes should both be added to the resulting
 * tree. From the combiner's point of view there is no difference between the {@code Tables} and the {@code Table} nodes
 * in the source trees, so the developer has to help out and give a hint that the {@code Table} nodes belong to a list
 * structure. This can be done using the {@code addListNode()} method; this method expects the name of a node, which
 * should be treated as a list node. So if {@code addListNode("Table");} was called, the combiner knows that it must not
 * combine the {@code Table} nodes, but add it both to the resulting tree.
 * </p>
 * <p>
 * Another limitation is the handling of attributes: Attributes can only have a single value. So if two nodes are to be
 * combined which both have an attribute with the same name, it is not possible to construct a proper union attribute.
 * In this case, the attribute value from the first node is used.
 * </p>
 *
 * @since 1.3
 */
public class UnionCombiner extends NodeCombiner {
    /**
     * Combines the given nodes to a new union node.
     *
     * @param node1 the first source node
     * @param node2 the second source node
     * @return the union node
     */
    @Override
    public ImmutableNode combine(final ImmutableNode node1, final ImmutableNode node2) {
        final ImmutableNode.Builder result = new ImmutableNode.Builder();
        result.name(node1.getNodeName());

        // attributes of the first node take precedence
        result.addAttributes(node2.getAttributes());
        result.addAttributes(node1.getAttributes());

        // Check if nodes can be combined
        final List<ImmutableNode> children2 = new LinkedList<>(node2.getChildren());
        node1.forEach(child1 -> {
            final ImmutableNode child2 = findCombineNode(node1, node2, child1);
            if (child2 != null) {
                result.addChild(combine(child1, child2));
                children2.remove(child2);
            } else {
                result.addChild(child1);
            }
        });

        // Add remaining children of node 2
        children2.forEach(result::addChild);

        return result.create();
    }

    /**
     * <p>
     * Tries to find a child node of the second source node, with which a child of the first source node can be combined.
     * During combining of the source nodes an iteration over the first source node's children is performed. For each child
     * node it is checked whether a corresponding child node in the second source node exists. If this is the case, these
     * corresponding child nodes are recursively combined and the result is added to the combined node. This method
     * implements the checks whether such a recursive combination is possible. The actual implementation tests the following
     * conditions:
     * </p>
     * <ul>
     * <li>In both the first and the second source node there is only one child node with the given name (no list
     * structures).</li>
     * <li>The given name is not in the list of known list nodes, i.e. it was not passed to the {@code addListNode()}
     * method.</li>
     * <li>None of these matching child nodes has a value.</li>
     * </ul>
     * <p>
     * If all of these tests are successful, the matching child node of the second source node is returned. Otherwise the
     * result is <strong>null</strong>.
     * </p>
     *
     * @param node1 the first source node
     * @param node2 the second source node
     * @param child the child node of the first source node to be checked
     * @return the matching child node of the second source node or <strong>null</strong> if there is none
     */
    protected ImmutableNode findCombineNode(final ImmutableNode node1, final ImmutableNode node2, final ImmutableNode child) {
        if (child.getValue() == null && !isListNode(child) && HANDLER.getChildrenCount(node1, child.getNodeName()) == 1
            && HANDLER.getChildrenCount(node2, child.getNodeName()) == 1) {
            final ImmutableNode child2 = HANDLER.getChildren(node2, child.getNodeName()).get(0);
            if (child2.getValue() == null) {
                return child2;
            }
        }
        return null;
    }
}