postgresql實現樹狀查詢

postgresql實現樹狀查詢

先建立一張表TREE_TEST，表中有三個字段，分別是id，name和parent_id

CREATE TABLE TREE_TEST (
 ID INTEGER PRIMARY KEY,
 NAME VARCHAR(32),
 PARENT_ID INTEGER REFERENCES TREE_TEST(ID)
);

 

而後插入幾條測試數據

INSERT INTO TREE_TEST(ID, NAME, PARENT_ID) VALUES(1, 'TREE_1', NULL);
INSERT INTO TREE_TEST(ID, NAME, PARENT_ID) VALUES(11, 'TREE_11', 1);
INSERT INTO TREE_TEST(ID, NAME, PARENT_ID) VALUES(12, 'TREE_12', 1);
INSERT INTO TREE_TEST(ID, NAME, PARENT_ID) VALUES(13, 'TREE_13', 1);
INSERT INTO TREE_TEST(ID, NAME, PARENT_ID) VALUES(111, 'TREE_111', 11);
INSERT INTO TREE_TEST(ID, NAME, PARENT_ID) VALUES(121, 'TREE_121', 12);
INSERT INTO TREE_TEST(ID, NAME, PARENT_ID) VALUES(122, 'TREE_122', 12);
INSERT INTO TREE_TEST(ID, NAME, PARENT_ID) VALUES(131, 'TREE_131', 13);
INSERT INTO TREE_TEST(ID, NAME, PARENT_ID) VALUES(132, 'TREE_132', 13);
INSERT INTO TREE_TEST(ID, NAME, PARENT_ID) VALUES(133, 'TREE_133', 13);

 

使用postgre中的recursive實現遞歸查詢

WITH RECURSIVE T(ID,NAME,PARENT_ID) AS(
    SELECT ID,NAME,PARENT_ID FROM TREE_TEST WHERE ID =13
    UNION ALL
  SELECT T1.ID,T1.NAME,T1.PARENT_ID
   FROM TREE_TEST T1 JOIN T ON T1.PARENT_ID=T.ID  
)
SELECT ID,NAME,PARENT_ID FROM T

這句話執行的步驟是：

一、先執行語句SELECT ID,NAME,PARENT_ID FROM TREE_TEST WHERE ID =13

二、把上一步執行的結果做爲表T，而後用該表T與TREE_TEST實現UNION ALL查詢，並把查詢的結果做爲T，繼續這種UNION ALL，直到查詢完全部的子級葉子節點

三、執行最後面一句sql：SELECT ID,NAME,PARENT_ID FROM T

該語句是查詢出id爲13和13下全部的子集，查詢結果以下：

id    name    parent_id
13    TREE_13        1
131    TREE_131    13
132    TREE_132    13
133    TREE_133    13

固然也能夠把某個子級的深度查詢出來，sql以下：

WITH RECURSIVE T (ID, NAME, PARENT_ID, PATH, DEPTH)  AS (
    SELECT ID, NAME, PARENT_ID, ARRAY[ID] AS PATH, 1 AS DEPTH
    FROM TREE_TEST
    WHERE PARENT_ID IS NULL

    UNION ALL

    SELECT  D.ID, D.NAME, D.PARENT_ID, T.PATH || D.ID, T.DEPTH + 1 AS DEPTH
    FROM TREE_TEST D
    JOIN T ON D.PARENT_ID = T.ID
    )
    SELECT ID, NAME, PARENT_ID, PATH, DEPTH FROM T
ORDER BY PATH;

 

該語句查詢結果以下：

id        name        parent_id    path                depth 1        TREE_1                            {1}                    1 11        TREE_11        1                {1,11}                2 111        TREE_111    11            {1,11,111}        3 12        TREE_12        1            {1,12}                2 121        TREE_121    12            {1,12,121}        3 122        TREE_122    12            {1,12,122}        3 13        TREE_13        1            {1,13}                2 131        TREE_131    13            {1,13,131}        3 132        TREE_132    13            {1,13,132}        3 133        TREE_133    13            {1,13,133}        3