PostgreSQL实现递归查询
2024-04-23 23:29:24
使用 WITH RECURSIVE 实现递归查询
SQL查询中WITH xxx as () 是对一个查询子句做别名，同时数据库会对该子句生成临时表（WITH子句只能被SELECT查询块引用）。
WITH RECURSIVE 则是一个递归的查询子句，他会把查询出来的结果再次代入到查询子句中继续查询。
创建表并插入数据
create table city
(id int4,pid int4,name varchar(50)
);INSERT INTO "city" ("id", "pid", "name") VALUES (1000, 0, '广东省');
INSERT INTO "city" ("id", "pid", "name") VALUES (1001, 0, '江苏省');
INSERT INTO "city" ("id", "pid", "name") VALUES (1002, 1000, '广州市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1003, 1000, '深圳市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1004, 1000, '佛山市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1005, 1000, '东莞市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1006, 1000, '珠海市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1007, 1000, '中山市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1008, 1000, '惠州市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1009, 1000, '汕头市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1010, 1000, '江门市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1011, 1000, '湛江市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1012, 1000, '肇庆市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1013, 1000, '梅州市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1014, 1000, '茂名市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1015, 1000, '阳江市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1016, 1000, '清远市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1017, 1000, '韶关市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1018, 1000, '揭阳市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1019, 1000, '汕尾市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1020, 1000, '潮州市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1021, 1000, '河源市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1022, 1000, '云浮市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1023, 1001, '南京市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1024, 1001, '苏州市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1025, 1001, '无锡市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1026, 1001, '南通市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1027, 1001, '常州市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1028, 1001, '徐州市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1029, 1001, '盐城市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1030, 1001, '扬州市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1031, 1001, '泰州市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1032, 1001, '镇江市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1033, 1001, '淮安市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1034, 1001, '连云港市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1035, 1001, '宿迁市');
INSERT INTO "city" ("id", "pid", "name") VALUES (1036, 1002, '荔湾区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1037, 1002, '越秀区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1038, 1002, '海珠区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1039, 1002, '天河区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1040, 1002, '白云区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1041, 1002, '黄埔区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1042, 1002, '番禺区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1043, 1002, '花都区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1044, 1002, '萝岗区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1045, 1002, '南沙区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1046, 1002, '增城区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1047, 1002, '从化区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1048, 1024, '姑苏区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1049, 1024, '虎丘区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1050, 1024, '吴中区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1051, 1024, '相城区');
INSERT INTO "city" ("id", "pid", "name") VALUES (1052, 1024, '吴江区');
WITH RECURSIVE 向下递归查询示例
以广东省为根节点，向下查询城市和市区
with recursive c as (select c1.id as id,c1.pid as pid,c1.name as name,1::integer as ll from city c1where c1.id = 1000union allselect c1.id as id,c1.pid as pid,c1.name as name,c.ll + 1 as ll from c,city c1where c1.pid = c.id
)
select c.id,c.pid,c.name,c.ll
from c
order by c.id
递归向下查询结果如下图：
二级联动效果查询
with recursive c as (select c1.id as id,c1.pid as pid,cast(c1.name as varchar(100)) as name,1::integer as ll from city c1where c1.id = 1000union allselect c1.id as id,c1.pid as pid,cast(c.name|| '>' || c1.name as varchar(100)) as name,c.ll + 1 as ll from city c1inner join c on c.id = c1.pidwhere c1.pid = c.id
)
select c.id,c.pid,c.name,c.ll
from c
order by c.id
查询结果如下图：
以越秀区为节点向上递归查询
with recursive c as (select c1.id as id,c1.pid as pid,c1.name as name,1::integer as ll from city c1where c1.id = 1037union allselect c1.id as id,c1.pid as pid,c1.name as name,c.ll + 1 as ll from c,city c1where c1.id = c.pid
)
select c.id,c.pid,c.name,c.ll
from c
order by c.id
递归向上查询结果如下图：
PostgreSQL实现递归查询相关推荐
- PostgreSQL的递归查询（RECURSIVE）
PostgreSQL的递归查询(RECURSIVE) 实验环境 操作系统:windows 10 家庭中文版数据库系统: PostgreSQL 9.6.2 说明 遇到树形结构的数据时,oracle可以使 ...
- [postgresql]postgresql的递归查询sql实例
--1.递归获取组织树 --创建表 create table public.org_test_t(id int ,name varchar(300),parentid int); --初始化数据 in ...
- PostgreSQL Oracle 兼容性之 - INDEX SKIP SCAN (递归查询变态优化) 非驱动列索引扫描优化...
标签 PostgreSQL , Oracle , index skip scan , 非驱动列条件 , 递归查询 , 子树 背景 对于输入条件在复合索引中为非驱动列的,如何高效的利用索引扫描? 在Or ...
- 数据库选型十八摸 之 PostgreSQL - 致 架构师、开发者
标签 PostgreSQL , 数据库特性 , 数据库应用场景分析 , 数据库选型 背景 数据库对于一家企业来说,相比其他基础组件占据比较核心的位置. 有很多企业由于最初数据库选型问题,导致一错再错, ...
- PostgreSQL 图式搜索(graph search)实践 - 百亿级图谱，毫秒响应
标签 PostgreSQL , CTE , 递归查询 , cycle , depth , loop , deep , level , 层级 , array , row array , JSON 背景 ...
- PostgreSQL WITH（WITH RECURSIVE） 查询表达式
PostgreSQL9.6中对WITH查询有如下描述 WITH提供了一种方式来书写在一个大型查询中使用的辅助语句.这些语句通常被称为公共表 表达式或CTE,它们可以被看成是定义只在一个查询中存在的临时 ...
- 【阿里云技术】PostgreSQL的20+种玩法
作者 德哥,周正中,阿里云高级技术专家,PostgreSQL 中国社区发起人之一,PostgreSQL 象牙塔发起人之一,DBA+社群联合发起人之一,10余项数据库相关专利,曾就职于斯凯网络,负责数据 ...
- PostgreSQL递归查询
出处:http://www.2cto.com/database/201309/242797.html PostgreSQL递归查询 数据库中的数据存在父子关系(单继承,每一条记录只有一个父亲). 如 ...
- PostgreSQL 中的递归查询 与oracle 的比较
PostgreSQL 中的递归查询,2种方法: 1.用with decursive WITH RECURSIVE d AS (SELECT d1.id,d1.parent_id,d1.caption ...
最新文章
- 营销自动化的4大预测分析错误
- 《编程导论（Java）#183;1.4.1 范式》
- [NOI2018]冒泡排序
- 第 11 章 日志管理 - 089 - 初探 ELK
- Java 函数引用 替代方案
- C++里vector::erase函数
- mysql5.6设置日志路径_mysql5.6.12切换binlog二进制日志路径_MySQL
- 最高响应比优先算法（HRRF）及例题详解
- 如何绘制用例图 - How to Draw Use Case Diagram
- 使用BeautifulSoup爬取百度图片
- 跑得快，打不死！清华大学开发“小强”机器人，壮汉狂踩也挡不住前进步伐
- 手机如何将图片缩小到20k？怎么指定压缩图片大小？
- 详解Java编码与解码以及常见的编码表，灵活处理乱码问题
- 项目实训--Unity多人游戏开发（八、3D音效融合AudioMixer、统一的音频播放系统）
- overleaf 中相关的问题
- 自定义拍照时 拍照界面_拍照时图片比例怎么选？比构图还要提前一步的摄影攻略要做好...
- Qt加载高德在线地图
- StackWalker 堆栈打印
- 好的数据库面试题集合
- 怎么看曲线有没有斜渐近线_关于曲线的斜渐近线