【优雅代码】18-利用function实现list、tree互转工具
欢迎关注b站账号/公众号【六边形战士夏宁】，一个要把各项指标拉满的男人。该文章已在 github目录收录。
屏幕前的 大帅比和 大漂亮如果有帮助到你的话请顺手点个赞、加个收藏这对我真的很重要。别下次一定了，都不关注上哪下次一定。
- 可直接运行的完整代码
- 视频讲解
- 上一篇guava限流源码解析
- 下一篇spring下的优秀工具类(进阶)
1.背景介绍
在日常工作中我们会遇到数据库查出来是list要递归转成树如果每个都去写方法那么效率就会很低，对于Function有理解困难的小伙伴先去看第11期的内容
2.相关代码
public class TreeUtils {
@Data
@AllArgsConstructor
@ToString(callSuper = true)
@NoArgsConstructor
@Builder
@JsonInclude(JsonInclude.Include.NON_EMPTY)
static class TestTreeObj {
private int id;
private int pid;
private List<TestTreeObj> testTreeObj;
}
/**
* 树转平铺
* treeToListDeep(testTreeObjs, result, TestTreeObj::getTestTreeObj, (l) -> l.getTestTreeObj() == null);
*
* @param source 源数据
* @param target 目标容器
* @param childListFn 递归调用方法
* @param addTargetCondition 添加到容器的判断方法
* @author 876651109@qq.com
* @date 2021/3/1 8:19 下午
*/
public static <F> void treeToListDeep(List<F> source, List<F> target, Function<F, List<F>> childListFn, Predicate<F> addTargetCondition) {
loopTree(source, childListFn, (l) -> {
if (addTargetCondition.test(l)) {
target.add(l);
}
});
}
public static <F> void treeToListDeep(List<F> source, Function<F, List<F>> childListFn, Predicate<F> addTargetCondition) {
treeToListDeep(source, new ArrayList<>(), childListFn, addTargetCondition);
}
/**
* List<TestTreeObj> treeResult = listToTree(list, TestTreeObj::setTestTreeObj, TestTreeObj::getId, TestTreeObj::getPid, (l) -> l.getPid() == 0);
*
* @param source 源数据
* @param setChildListFn 设置递归的方法
* @param idFn 获取id的方法
* @param pidFn 获取父id的方法
* @param getRootCondition 获取根节点的提哦啊见
* @return {@link List<F>}
* @author 876651109@qq.com
* @date 2021/3/1 8:18 下午
*/
public static <F, T> List<F> listToTree(List<F> source, BiConsumer<F, List<F>> setChildListFn, Function<F, T> idFn, Function<F, T> pidFn, Predicate<F> getRootCondition) {
return listToTree(source, setChildListFn, idFn, pidFn, getRootCondition, null);
}
/**
* 复杂形式，进行listen回调,用流式写法可以与外界交互
* (idx,obj)->{System.out.println(123);},idx是从0开始的层级
*
* @param listen 回调函数
* @author seal 876651109@qq.com
* @date 2021/6/3 7:46 下午
*/
public static <F, T> List<F> listToTree(List<F> source, BiConsumer<F, List<F>> setChildListFn, Function<F, T> idFn, Function<F, T> pidFn, Predicate<F> getRootCondition, BiConsumer<Integer, F> listen) {
List<F> tree = new ArrayList<>();
Map<T, List<F>> map = new HashMap<>(source.size());
for (F f : source) {
if (getRootCondition.test(f)) {
tree.add(f);
} else {
List<F> tempList = map.getOrDefault(pidFn.apply(f), new ArrayList<>());
tempList.add(f);
map.put(pidFn.apply(f), tempList);
}
}
tree.forEach(l -> assembleTree(l, map, setChildListFn, idFn, listen, 0));
return tree;
}
/**
* 组装树
*/
private static <F, T> void assembleTree(F current, Map<T, List<F>> map, BiConsumer<F, List<F>> setChildListFn, Function<F, T> idFn, BiConsumer<Integer, F> listen, int idx) {
List<F> fs = map.get(idFn.apply(current));
setChildListFn.accept(current, fs);
if (CollectionUtils.isEmpty(fs)) {
fs.forEach(l -> assembleTree(l, map, setChildListFn, idFn, listen, idx + 1));
}
if (listen != null) {
listen.accept(idx, current);
}
}
/**
* 树的简易递归,listen会回调当前对象
*
* @param source 源数据
* @param childListFn get方法
* @param listen 回调函数
* @author seal 876651109@qq.com
* @date 2021/6/5 16:47
*/
public static <F> void loopTree(List<F> source, Function<F, List<F>> childListFn, Consumer<F> listen) {
loopTree(source, childListFn, listen, null);
}
public static <F> void loopTree(List<F> source, Function<F, List<F>> childListFn, Consumer<F> preListen, Consumer<F> postFun) {
if (CollectionUtils.isEmpty(source)) {
return;
}
source.forEach(l -> {
Optional.ofNullable(preListen).ifPresent(s -> s.accept(l));
loopTree(childListFn.apply(l), childListFn, preListen, postFun);
Optional.ofNullable(postFun).ifPresent(s -> s.accept(l));
});
}
/**
* 与listToTree功能一致，但是前方法循环两次，该方法利用递归只循环一次，用空间换时间的方式
*
* @return
*/
public static <T, K> List<T> buildTree(List<T> dataList, int index, Map<K, T> dataMap,
Function<T, K> idFn,
Function<T, K> pIdFn,
Function<T, List<T>> getChildFn,
BiConsumer<T, List<T>> setChildFn,
Predicate<K> predicate) {
List<T> resultList = new ArrayList<>(dataList.size());
while (index < dataList.size()) {
T item = dataList.get(index);
dataMap.put(idFn.apply(item), item);
K pId = pIdFn.apply(item);
if (predicate.test(pId)) {
resultList.add(item);
} else {
T parent = dataMap.get(pId);
//为null表示还没被循环到父级,则递归，再次循环剩下的部分
if (Objects.isNull(parent)) {
index += 1;
List<T> list = buildTree(dataList, index, dataMap, idFn, pIdFn, getChildFn, setChildFn, predicate);
parent = dataMap.get(pId);
// 如果为null 表示列表没有父级，自动升级为顶节点
if (Objects.isNull(parent)) {
resultList.add(item);
} else {
List<T> childList = Optional.ofNullable(getChildFn.apply(parent)).orElse(new ArrayList<>());
childList.add(item);
setChildFn.accept(parent, childList);
}
if (list.size() != 0) {
resultList.addAll(list);
}
} else {
List<T> childList = Optional.ofNullable(getChildFn.apply(parent)).orElse(new ArrayList<>());
childList.add(item);
setChildFn.accept(parent, childList);
}
}
index += 1;
}
return resultList;
}
}3.使用方式
public static void main(String[] args) {
List<TestTreeObj> list = new ArrayList<TestTreeObj>() {{
add(TestTreeObj.builder().id(1).build());
add(TestTreeObj.builder().id(11).pid(1).build());
add(TestTreeObj.builder().id(12).pid(1).build());
add(TestTreeObj.builder().id(111).pid(11).build());
add(TestTreeObj.builder().id(112).pid(11).build());
add(TestTreeObj.builder().id(121).pid(12).build());
add(TestTreeObj.builder().id(122).pid(12).build());
add(TestTreeObj.builder().id(2).build());
add(TestTreeObj.builder().id(21).pid(2).build());
add(TestTreeObj.builder().id(22).pid(2).build());
add(TestTreeObj.builder().id(211).pid(21).build());
add(TestTreeObj.builder().id(212).pid(21).build());
add(TestTreeObj.builder().id(221).pid(22).build());
add(TestTreeObj.builder().id(222).pid(22).build());
}};
List<TestTreeObj> treeResult = listToTree(list, TestTreeObj::setTestTreeObj, TestTreeObj::getId, TestTreeObj::getPid, (l) -> l.getPid() == 0);
List<TestTreeObj> testTreeObjs = new ArrayList<TestTreeObj>() {{
add(TestTreeObj.builder().id(1).testTreeObj(new ArrayList<TestTreeObj>() {{
add(TestTreeObj.builder().id(11).testTreeObj(new ArrayList<TestTreeObj>() {{
add(TestTreeObj.builder().id(111).build());
add(TestTreeObj.builder().id(112).build());
}}).build());
}}).build());
}};
List<TestTreeObj> result = new ArrayList<>();
treeToListDeep(testTreeObjs, result, TestTreeObj::getTestTreeObj, (l) -> l.getTestTreeObj() == null);
List<TestTreeObj> result2 = new ArrayList<>();
treeToListDeep(testTreeObjs, result2, TestTreeObj::getTestTreeObj, (l) -> l.getPid() == 0);
System.out.println(result2);
}输出如下，可以看到已经成功将list转化为tree
