構造無限級樹並深度遍歷查找指定節點

目錄

• 前言
• 數據
• 實現
  • 構造無限級
  • 廣度優先遍歷
  • 深度優先遍歷
• 完整代碼

前言

最近接到一個關於樹的需求，想起了大學那會兒聽過俞敏洪有關樹的演講，老師重複在課上放了不少次，N年過去了印象仍是很深。
演講內容是介樣的，讓咱們一塊兒來喝湯~

人的生活方式有兩種，
第一種方式是像草同樣活着，
你儘管活着，
每一年還在成長，
可是你畢竟是一棵草，
你吸取雨露陽光，
可是卻長不大。
人們能夠踩過你，
可是人們不會由於你的痛苦，而他產生痛苦；
人們不會由於你被踩了，而來憐憫你，
由於人們自己就沒有看到你。
因此咱們每個人，
都應該像樹同樣的成長，
即便咱們如今什麼都不是，
可是隻要你有樹的種子，
即便你被踩到泥土中間，
你依然可以吸取泥土的營養，
本身成長起來。
當你長成參天大樹之後，
遙遠的地方，人們就能看到你；
走近你，你能給人一片綠色。
活着是美麗的風景，
死了依然是棟樑之才，
活着死了都有用。

回到今天的主題，需求是這樣的：構造無限級樹型結構，界面相似下邊這樣的結構

而後，還須要把全部的橙色標記的葉子節點從垂直方向篩選出來，最終效果就是獲得一個這樣順序的一級數組

KKK
QQQ
JJJ
HHH
GGG
LLL

須要根據AAA這個節點和他全部子節點的數據，構造出一個無限級樹型結構，以下

數據

• 頂級AAA數據

{"id":348019,"name":"AAA","is_leaf":2,"parent_id":347234,"type":0,"order":["348020"]}

• AAA全部的孩子節點

[{"id":348020,"name":"BBB","is_leaf":2,"parent_id":348019,"type":0,"order":["348037","348033","348021","348023","348034"]},{"id":348021,"name":"CCC","is_leaf":2,"parent_id":348020,"type":0,"order":["348022","348035"]},{"id":348022,"name":"DDD","is_leaf":2,"parent_id":348021,"type":0,"order":["348024","348032","348030"]},{"id":348024,"name":"EEE","is_leaf":2,"parent_id":348022,"type":0,"order":["348025","348036"]},{"id":348025,"name":"FFF","is_leaf":2,"parent_id":348024,"type":0,"order":["348031"]},{"id":348030,"name":"HHH","is_leaf":1,"parent_id":348022,"type":11,"order":[]},{"id":348023,"name":"GGG","is_leaf":1,"parent_id":348020,"type":11,"order":[]},{"id":348031,"name":"QQQ","is_leaf":1,"parent_id":348025,"type":11,"order":[]},{"id":348032,"name":"JJJ","is_leaf":1,"parent_id":348022,"type":11,"order":[]},{"id":348035,"name":"MMM\u65e0\u53f6\u5b50","is_leaf":2,"parent_id":348021,"type":0,"order":[]},{"id":348036,"name":"NNN\u65e0\u53f6\u5b50","is_leaf":2,"parent_id":348024,"type":0,"order":[]},{"id":348033,"name":"KKK","is_leaf":1,"parent_id":348020,"type":11,"order":[]},{"id":348037,"name":"OOO","is_leaf":2,"parent_id":348020,"type":0,"order":[]},{"id":348034,"name":"LLL","is_leaf":1,"parent_id":348020,"type":11,"order":[]}]

節點數據庫結構都同樣，核心是有一個parent_id，每一個節點會保存直屬葉子的排序order

實現

其實這裏考察的遞歸構造樹、廣度優先遍歷、深度優先遍歷
其實無論啥語言，思想都是這樣，由於這個需求是php系統寫的接口，就用php給出實現代碼，氣氛搞起來~

構造無限級

private function getTree($list, $node)
    {
        $tree = [];
        foreach ($list as $k => $v) {
            if ($v['parent_id'] == $node['id']) {
                $v['children'] = $this->getTree($list, $v);
                $tree[] = $v;
            }
        }

        if (empty($node['order'])) {
            return [];
        }
        //注意：爲了保證全部的葉子節點是根據已有的order排序，這裏咱們從新擺放一下
        $treeMap = [];
        foreach ($tree as $v) {
            $treeMap[$v['id']] = $v;//創建id和數據的map
        }
        //根據order從新排序
        $orderTree = [];
        foreach ($node['order'] as $id) {//根據order的順序從新擺放
            $orderTree[] = $treeMap[$id];
        }
        return $orderTree;
    }

咱們來使用一下，使用以下

最終輸出的json以下

能夠看到，已經構造出了一個無限級的樹，而且和咱們的界面是如出一轍的層級和順序

拉下來如今咱們要遍歷這棵樹，爲了獲取到全部is_leaft=1的節點，咱們先用廣度優先遍從來試一下，由於這個對於找節點來講性能比較高

廣度優先遍歷

代碼以下

private function bfsFindLeafs($node)
    {
        $result = [];//知足條件的組合
        $q = [];
        $q[] = $node;
        while (count($q) > 0) {
            $size = count($q);
            for ($i = 0; $i < $size; $i++) {
                $current = array_pop($q);
                //判斷節點若是是知足條件，加入路徑
                if ($current['is_leaf'] == 1) {
                    $result[] = $current;
                }
                //把全部的子節點加入隊列
                foreach ($current['children'] as $v) {
                    $q[] = $v;
                }
            }
        }
        return $result;
    }

咱們來使用一下

$getAllLeafs = $this->bfsFindLeafs($topNode);
echo json_encode($getAllLeafs);

運行後輸出

[
    {
        "id": 348034,
        "name": "LLL",
        "is_leaf": 1,
        "parent_id": 348020,
        "type": 11,
        "order": [],
        "children": []
    },
    {
        "id": 348023,
        "name": "GGG",
        "is_leaf": 1,
        "parent_id": 348020,
        "type": 11,
        "order": [],
        "children": []
    },
    {
        "id": 348030,
        "name": "HHH",
        "is_leaf": 1,
        "parent_id": 348022,
        "type": 11,
        "order": [],
        "children": []
    },
    {
        "id": 348032,
        "name": "JJJ",
        "is_leaf": 1,
        "parent_id": 348022,
        "type": 11,
        "order": [],
        "children": []
    },
    {
        "id": 348031,
        "name": "QQQ",
        "is_leaf": 1,
        "parent_id": 348025,
        "type": 11,
        "order": [],
        "children": []
    },
    {
        "id": 348033,
        "name": "KKK",
        "is_leaf": 1,
        "parent_id": 348020,
        "type": 11,
        "order": [],
        "children": []
    }
]

好，咱們已經能夠獲取到全部is_leaf爲1的節點了，可是這裏有一個問題，不是從上到下輸出的

LLL
GGG
HHH
JJJ
QQQ
KKK

這時候，咱們就須要一個深度優先遍從來實現了

深度優先遍歷

深度優先遍歷其實就是回滾算法的一種，爲了從上到下輸出，須要先序遍歷

private function dfsFindLeafs($node)
    {
        $trackList = [];
        foreach ($node['children'] as $v) {
            if ($v['is_leaf'] == 1) {
                $trackList[] = $v;
            }
            if (empty($v['children'])) {
                continue;
            }
            $trackList = array_merge($trackList,$this->dfsFindLeafs($v));
        }
        return $trackList;
    }

咱們再來運行一下

$getAllLeafs = $this->dfsFindLeafs($topNode);
echo json_encode($getAllLeafs);

輸出

[
    {
        "id": 348033,
        "name": "KKK",
        "is_leaf": 1,
        "parent_id": 348020,
        "type": 11,
        "order": [],
        "children": []
    },
    {
        "id": 348031,
        "name": "QQQ",
        "is_leaf": 1,
        "parent_id": 348025,
        "type": 11,
        "order": [],
        "children": []
    },
    {
        "id": 348032,
        "name": "JJJ",
        "is_leaf": 1,
        "parent_id": 348022,
        "type": 11,
        "order": [],
        "children": []
    },
    {
        "id": 348030,
        "name": "HHH",
        "is_leaf": 1,
        "parent_id": 348022,
        "type": 11,
        "order": [],
        "children": []
    },
    {
        "id": 348023,
        "name": "GGG",
        "is_leaf": 1,
        "parent_id": 348020,
        "type": 11,
        "order": [],
        "children": []
    },
    {
        "id": 348034,
        "name": "LLL",
        "is_leaf": 1,
        "parent_id": 348020,
        "type": 11,
        "order": [],
        "children": []
    }
]

順序爲

KKK
QQQ
JJJ
HHH
GGG
LLL

此次順序終於對了，oh yeah~，和咱們以前的界面垂直順序一致

完整代碼

下面給出完整代碼

<?php


/**
 * Class Test
 * @author chenqionghe
 */
class Test
{

    public function run()
    {
        $topNodeJson = '{"id":348019,"name":"AAA","is_leaf":2,"parent_id":347234,"type":0,"order":["348020"]}';
        $childrenJson = '[{"id":348020,"name":"BBB","is_leaf":2,"parent_id":348019,"type":0,"order":["348037","348033","348021","348023","348034"]},{"id":348021,"name":"CCC","is_leaf":2,"parent_id":348020,"type":0,"order":["348022","348035"]},{"id":348022,"name":"DDD","is_leaf":2,"parent_id":348021,"type":0,"order":["348024","348032","348030"]},{"id":348024,"name":"EEE","is_leaf":2,"parent_id":348022,"type":0,"order":["348025","348036"]},{"id":348025,"name":"FFF","is_leaf":2,"parent_id":348024,"type":0,"order":["348031"]},{"id":348030,"name":"HHH","is_leaf":1,"parent_id":348022,"type":11,"order":[]},{"id":348023,"name":"GGG","is_leaf":1,"parent_id":348020,"type":11,"order":[]},{"id":348031,"name":"QQQ","is_leaf":1,"parent_id":348025,"type":11,"order":[]},{"id":348032,"name":"JJJ","is_leaf":1,"parent_id":348022,"type":11,"order":[]},{"id":348035,"name":"MMM\u65e0\u53f6\u5b50","is_leaf":2,"parent_id":348021,"type":0,"order":[]},{"id":348036,"name":"NNN\u65e0\u53f6\u5b50","is_leaf":2,"parent_id":348024,"type":0,"order":[]},{"id":348033,"name":"KKK","is_leaf":1,"parent_id":348020,"type":11,"order":[]},{"id":348037,"name":"OOO","is_leaf":2,"parent_id":348020,"type":0,"order":[]},{"id":348034,"name":"LLL","is_leaf":1,"parent_id":348020,"type":11,"order":[]}]';
        $topNode = json_decode($topNodeJson, true);
        $childrenList = json_decode($childrenJson, true);
        $topNode['children'] = $this->getTree($childrenList, $topNode);
        $getAllLeafs = $this->dfsFindLeafs($topNode);
        echo json_encode($getAllLeafs);
    }


    /**
     * 遞歸構造無限級樹
     *
     * @param $list
     * @param $node
     * @return array
     */
    private function getTree($list, $node)
    {
        $tree = [];
        foreach ($list as $k => $v) {
            if ($v['parent_id'] == $node['id']) {
                $v['children'] = $this->getTree($list, $v);
                $tree[] = $v;
            }
        }
        if (empty($node['order'])) {
            return [];
        }
        //注意：爲了保證全部的葉子節點是根據已有的order排序，這裏咱們從新擺放一下
        $treeMap = [];
        foreach ($tree as $v) {
            $treeMap[$v['id']] = $v;//創建id和數據的map
        }
        //根據order從新排序
        $orderTree = [];
        foreach ($node['order'] as $id) {//根據order的順序從新擺放
            $orderTree[] = $treeMap[$id];
        }
        return $orderTree;
    }

    /**
     * 廣度優先遍歷獲取全部葉子
     * @param $node
     * @return array
     */
    private function bfsFindLeafs($node)
    {
        $result = [];//知足條件的組合
        $q = [];
        $q[] = $node;
        while (count($q) > 0) {
            $size = count($q);
            for ($i = 0; $i < $size; $i++) {
                $current = array_pop($q);
                //判斷節點若是是知足條件，加入路徑
                if ($current['is_leaf'] == 1) {
                    $result[] = $current;
                }
                //把全部的子節點加入隊列
                foreach ($current['children'] as $v) {
                    $q[] = $v;
                }
            }

        }
        return $result;
    }


    /**
     * 獲取全部的葉子路徑（深度度優先遍歷）
     */
    private function dfsFindLeafs($node)
    {
        $trackList = [];
        foreach ($node['children'] as $v) {
            if ($v['is_leaf'] == 1) {
                $trackList[] = $v;
            }
            if (empty($v['children'])) {
                continue;
            }
            $trackList = array_merge($trackList, $this->dfsFindLeafs($v));
        }
        return $trackList;
    }
}

(new Test())->run();

ok，到這裏咱們就已經學會如何構造無限級樹型結構，並學會用深度優先遍歷垂直輸出指定條件節點，還知道了怎麼用廣度優先遍歷，giao~