之前在项目需要实现一个功能——将xml文件映射成实体，然后对映射的实体进行逻辑处理，最后保存到数据库中；由于xml结构的数据是结构化的数据，所以需要保证保存的数据具有正确的主外键关联。如下所示，是一个需要保存到数据库的xml文件。当映射成对应的实体school和student的时候，我们需要知道“school-one”下面有哪些学生，“school-two”下面有哪些学生，这个时候想到了使用树形结构来保存实体，让实体之间依然存在关联关系。

<school-inf><msg>2017-10-1XX省学校信息总汇</msg><schools><school><name>school-one</name><students><student>Jack</student><student>Rose</student><student>Jon</student></students></school><school><name>school-two</name><students><student>Bob</student><student>Alisa</student></students></school></schools>
</school-inf>
复制代码

树形工具

以下是树形工具类的实现，包含了树形节点类和树形结构类，由于代码中注释已经比较全面，所以不做过多的说明。

树形节点类BeanTreeNode.java

每一个节点对应一个实体，节点包含了实体信息，为了保证实体之间的关联关系，需要留有父节点信息，所有的子节点信息。由此推断出，节点的主要成员有

• 父节点信息
• 所有子节点信息
• 当前实体信息

为了方便操作，我还多增加了id和pid（parent id）,以及节点类型（nodeType）。对id的相关操作我并没有添加，如果需要可以自行添加。

import java.util.ArrayList;
import java.util.List;
import java.util.UUID;/*** 实体树形结构点* BeanTreeNode* @author BrightLoong* @version 1.0**/
public class BeanTreeNode {/**标识id*/private String id;/**父id标识，为了方便获取冗余出来*/private String pid;/**父节点*/private BeanTreeNode parentNode;/**节点类型*/private String nodeType;/**节点值*/private Object bean;/**子节点*/private List<BeanTreeNode> childNodes;/*** @param parentNode* @param nodeType* @param bean* @param childNodes*/public BeanTreeNode(BeanTreeNode parentNode, String nodeType, Object bean) {this.parentNode = parentNode;this.nodeType = nodeType;this.bean = bean;this.childNodes = new ArrayList<BeanTreeNode>();this.id = UUID.randomUUID().toString().replaceAll("-", "");if (parentNode != null) {this.pid = parentNode.getId();}}/*** @return the nodeType*/public String getNodeType() {return nodeType;}/*** @param nodeType the nodeType to set*/public void setNodeType(String nodeType) {this.nodeType = nodeType;}/*** @return the parentNode*/public BeanTreeNode getParentNode() {return parentNode;}/*** @param parentNode the parentNode to set*/public void setParentNode(BeanTreeNode parentNode) {this.parentNode = parentNode;}/*** @return the bean*/public Object getBean() {return bean;}/*** @param bean the bean to set*/public void setBean(Object bean) {this.bean = bean;}/*** @return the childNodes*/public List<BeanTreeNode> getChildNodes() {return childNodes;}/*** @param childNodes the childNodes to set*/public void setChildNodes(List<BeanTreeNode> childNodes) {this.childNodes = childNodes;}/*** @return the id*/public String getId() {return id;}/*** @param id the id to set*/public void setId(String id) {this.id = id;}/*** @return the pid*/public String getPid() {return pid;}/*** @param pid the pid to set*/public void setPid(String pid) {this.pid = pid;}/*** 是否具有子节点* @return true or false*/public boolean haveChild() {return !CollectionUtils.isEmpty(childNodes);}
}
复制代码

树形结构类BeanTree.java

BeanTree.java里面包含了如下的一些常用操作：

• 返回根节点
• 返回最后添加节点
• 判断是否具有子节点
• 添加节点
• 移动节点到其他节点下
• 获取对应nodeType的所有节点或实体
• 根据实体获取节点
• 获取父节点
• 转化为map结构

代码如下

import java.util.ArrayList;
import java.util.Collection;
import java.util.HashMap;
import java.util.List;
import java.util.Map;import org.apache.commons.collections.CollectionUtils;/*** 实体树形结构* BeanTree* @author BrightLoong* @version 1.0**/
public class BeanTree {/**根节点*/private BeanTreeNode root;/*** 最新添加的节点*/private BeanTreeNode currentNode;/*** @return the currentNode*/public BeanTreeNode getCurrentNode() {return currentNode;}/*** @return the root*/public BeanTreeNode getRoot() {return root;}/*** 判断节点是否有子节点.* @param node 要判断的节点* @return true or false*/public boolean haveChild(BeanTreeNode node) {return CollectionUtils.isEmpty(node.getChildNodes());}/*** 在父节点上面添加节点，并返回天添加的节点.* @param parentNode 父节点* @param bean 要添加的bean* @param nodeType 节点类型* @return 返回包含bean的节点*/public BeanTreeNode addNode(BeanTreeNode parentNode, Object bean, String nodeType) {BeanTreeNode node;if (bean == null) {return null;}//如果没有父节点说明为root根节点if (parentNode == null) {node = root = new BeanTreeNode(null, nodeType, bean);} else {//创建子节点，并添加到父节点上node = new BeanTreeNode(parentNode, nodeType, bean);parentNode.getChildNodes().add(node);}currentNode = node;return node;}/*** 将当期bean-sBean,以及sBean下的子Bean，挂到dBean下* @param sBean 源Bean* @param dBean 目的父Bean*/public void removeTo(Object sBean, Object dBean) {BeanTreeNode sNode = getNodeByBean(sBean);BeanTreeNode dNode = getNodeByBean(dBean);removeTo(sNode, dNode);}/*** 将当期node-sNode,以及sNode下的子Node，挂到dNode下* @param sNode 源node* @param dNode 目的父node*/public void removeTo(BeanTreeNode sNode, BeanTreeNode dNode) {//从当前父节点移除sNodesNode.getParentNode().getChildNodes().remove(sNode);//将sNode移到dNode下dNode.getChildNodes().add(sNode);//修改sNode的父Id和父节点sNode.setPid(dNode.getId());sNode.setParentNode(dNode);}/*** 获取父bean.* @param bean 子bean* @return 返回父bean*/public Object getParentBean(Object bean) {return getNodeByBean(bean).getParentNode().getBean();}/*** 根据传入的bean获取bean下面对应类型的子bean.* @param bean 当前bean* @param nodeType 节点类型* @return 子bean的集合*/public List<Object> getBeanListByBeanAndNodeType(Object bean, String nodeType) {BeanTreeNode node = getNodeByBean(bean);return getBeanListByNodeType(node, nodeType);}/*** 根据传入的bean获取包含bean的Node节点* @param node 当前node* @param bean 要查找的bean* @return node节点*/public BeanTreeNode getNodeByBean(BeanTreeNode node, Object bean) {BeanTreeNode resultNode = null;if (node.getBean().equals(bean)) {resultNode = node;return resultNode;} else {for (BeanTreeNode tempNode : node.getChildNodes()) {resultNode = getNodeByBean(tempNode, bean);if (resultNode != null) {break;}}}return resultNode;}/*** 根据传入的bean获取root节点下包含bean的Node节点* @param bean 要查找的bean* @return node节点*/public BeanTreeNode getNodeByBean(Object bean) {return getNodeByBean(root, bean);}/*** 根据节点类型返回当前节点下对应节点类型的bean的list集合.* 默认如果当前节点满足类型，那么当前节点不会存在相同类型的子节点* @param node 当前节点* @param nodeType 节点类型* @return*/@SuppressWarnings("unchecked")public <T> List<T> getBeanListByNodeType(BeanTreeNode node, String nodeType) {List<T> beanList = new ArrayList<T>();if (node.getNodeType().equals(nodeType)) {beanList.add((T)node.getBean());} else {for (BeanTreeNode tempNode : node.getChildNodes()) {beanList.addAll((Collection<? extends T>) getBeanListByNodeType(tempNode, nodeType));}}return beanList;}/*** 根据节点类型返回根节点下对应节点类型的bean的list集合.* @param nodeType 节点类型* @return*/public <T> List<T> getBeanListByNodeType(String nodeType) {return getBeanListByNodeType(root, nodeType);}/*** 从root节点开始获取对应nodeType的node.* @param nodeType 节点类型* @return nodeType类型的节点集合*/public List<BeanTreeNode> getNodeListByNodeType(String nodeType) {return getNodeListByNodeType(root, nodeType);}/*** 从node节点开始获取对应nodeType的node.* @param node node节点* @param nodeType 节点类型* @return nodeType类型的节点集合*/public List<BeanTreeNode> getNodeListByNodeType(BeanTreeNode node, String nodeType) {List<BeanTreeNode> nodeList = new ArrayList<BeanTreeNode>();if(node==null){return nodeList;  }if (nodeType.equals(node.getNodeType())) {nodeList.add(node);} else {for (BeanTreeNode tempNode : node.getChildNodes()) {nodeList.addAll(getNodeListByNodeType(tempNode, nodeType));  }}return nodeList;}/*** 将树形结构转化为map.* @return*/public Map<String, List<Object>> toMap() {return toMap(root);}/*** 将对应节点及其子节点转化为map.* @param node 树节点* @return 转化后的map*/public Map<String, List<Object>> toMap(BeanTreeNode node) {Map<String, List<Object>> map = new HashMap<String, List<Object>>();toMap(node, map);return map;}/*** 根据传入的nodeType删除对应的节点以及其所有子节点.* @param nodeType*/public void delNodeByNodeType(String nodeType) {delNodeByNodeType(root, nodeType);}/*** 删除node节点下，类型为nodeType的节点和所有子节点* @param node* @param nodeType*/public void delNodeByNodeType(BeanTreeNode node, String nodeType) {List<BeanTreeNode> nodeList = getNodeListByNodeType(node, nodeType);for (BeanTreeNode beanTreeNode : nodeList) {beanTreeNode.getParentNode().getChildNodes().remove(beanTreeNode);}}/*** 从树结构里面删除bean和相关node.* @param bean bean*/public void delNodeByBean(Object bean) {BeanTreeNode node = getNodeByBean(bean);BeanTreeNode parentNode = node.getParentNode();List<BeanTreeNode> childNodes = parentNode.getChildNodes();Iterator<BeanTreeNode> it = childNodes.iterator();while (it.hasNext()) {BeanTreeNode beanTreeNode = it.next();if (node == beanTreeNode) {it.remove();}}}/*** 根据class返回对应的beanList.* @param cls class* @return beanList*/public <T> List<Object> getBeanListByClass(Class<T> cls) {return getBeanListByClass(root, cls);}/*** 根据class返回对应的beanList.* @param node 节点* @param cls class* @return beanList*/public <T> List<Object> getBeanListByClass(BeanTreeNode node, Class<T> cls) {List<Object> beanList = new ArrayList<Object>();Object bean = node.getBean();if (cls.isAssignableFrom(bean.getClass())) {beanList.add(bean);}List<BeanTreeNode> childNodes = node.getChildNodes();for (BeanTreeNode beanTreeNode : childNodes) {beanList.addAll(getBeanListByClass(beanTreeNode, cls));}return beanList;}/*** 将对应节点及其子节点转化为map.* @param node 树节点* @param map 用来保存结果的map*/private void toMap(BeanTreeNode node, Map<String, List<Object>> map) {String key = node.getNodeType();Object bean = node.getBean();if (map.containsKey(key)) {map.get(key).add(bean);} else {List<Object> list = new ArrayList<Object>();list.add(bean);map.put(key, list);}for (BeanTreeNode tempNode : node.getChildNodes()) {toMap(tempNode, map);}}
}
复制代码

测试树形工具

使用上面的xml进行测试，这里就不再做xml映射，假设存在上面xml所示的所有实体，“school-one”和“school-two”以及5个student，看看能否构造出想要的结构，测试类代码如下。

class SchoolInf {private String msg;public SchoolInf(String msg) {this.msg = msg;}
}class Student {private String name;public Student(String name) {this.name = name;}
}class School {private String name;public School(String name) {this.name = name;}
}public class Test {public static void main(String[] args) {SchoolInf schoolInf = new SchoolInf("2017-10-1XX省学校信息总汇");School school_one = new School("school-one");School school_two = new School("school-two");Student Jack = new Student("Jack");Student Rose = new Student("Rose");Student Jon = new Student("Jon");Student Bob = new Student("Bob");Student Alisa = new Student("Alisa");BeanTree tree = new BeanTree();BeanTreeNode root = tree.addNode(null, schoolInf, "root");BeanTreeNode school_node1 = tree.addNode(root, school_one, "school");BeanTreeNode school_node2 = tree.addNode(root, school_two, "school");tree.addNode(school_node1, Jack, "root");tree.addNode(school_node1, Rose, "root");tree.addNode(school_node1, Jon, "root");tree.addNode(school_node2, Bob, "root");tree.addNode(school_node2, Alisa, "root");System.out.println("end");}
}复制代码

我们通过调试观察树结构变量“tree”的值如下：

可以看出来能够构造出正确的结构，BeanTree中其他的一些方法这里就不在一一测试了。

更新记录

• 2018/1/10，在BeanTree中添加更多的操作方法。