myql 查询树形表结果:说说、说说的述评、评论的回复
myql 查询树形表结果:说说、说说的评论、评论的回复
有三张表关联表:
用户的说说表(ixt_customer_note)
说说的评论表(ixt_customer_note_comment)
评论的回复表(ixt_customer_note_reply)
说说表保存的是用户发表的说说信息,评论表保存的是用户对说说发表的评论信息,回复表保存的是用户对评论及回复的回复
要求查询三张表返回结果为树形结构,如下:
发表说说:1003
发表说说:1002
发表评论:comment1002_1
发表评论:comment1002_2
发表回复:reply_1002_1
发表回复:reply_1002_2
发表评论:comment1002_3
发表说说:1001
发表评论:comment1001_1
发表评论:comment1001_2
发表说说:1000
发表评论:comment1000_1
发表回复:reply_1000_1
发表回复:reply_1000_2
1、设计三张表及插入相关数据
ixt_customer_note 表结构: |
ixt_customer_note 表sql语句: DROP TABLE IF EXISTS `ixt_customer_note`; CREATE TABLE `ixt_customer_note` ( `id` varchar(50) NOT NULL COMMENT '主键UUID', `customerId` varchar(50) NOT NULL COMMENT '用户id', `content` varchar(500) NOT NULL COMMENT '说说内容', `createUser` varchar(50) DEFAULT NULL COMMENT '创建人ID', `createDate` datetime DEFAULT NULL COMMENT '创建时间', `updateUser` varchar(50) DEFAULT NULL COMMENT '更新人ID', `updateDate` datetime DEFAULT NULL COMMENT '更新时间', `deleteFlag` bit(1) NOT NULL DEFAULT b'0' COMMENT '删除标识:0未删除,1删除', PRIMARY KEY (`id`) ) ENGINE=InnoDB DEFAULT CHARSET=utf8; INSERT INTO `ixt_customer_note` VALUES ('1000', 'user1', '1000', null, '2015-10-01 21:18:24', null, null, ''); INSERT INTO `ixt_customer_note` VALUES ('1001', 'user1', '1001', null, '2015-10-06 21:18:19', null, null, ''); INSERT INTO `ixt_customer_note` VALUES ('1002', 'user1', '1002', null, '2015-10-14 22:05:04', null, null, ''); INSERT INTO `ixt_customer_note` VALUES ('1003', 'user1', '1003', null, '2015-10-15 21:18:12', null, null, ''); |
ixt_customer_note_comment 表结构: |
ixt_customer_note_comment 表sql语句: DROP TABLE IF EXISTS `ixt_customer_note_comment`; CREATE TABLE `ixt_customer_note_comment` ( `id` varchar(50) NOT NULL COMMENT '主键UUID', `customerId` varchar(50) NOT NULL COMMENT '评论用户ID', `dataId` varchar(50) NOT NULL COMMENT '被评论的说说ID', `content` varchar(1000) NOT NULL COMMENT '评论内容', `createUser` varchar(50) DEFAULT NULL COMMENT '创建人ID', `createDate` datetime DEFAULT NULL COMMENT '更新人ID', `updateUser` varchar(50) DEFAULT NULL COMMENT '更新时间', `updateDate` datetime DEFAULT NULL COMMENT '更新时间', `deleteFlag` bit(1) NOT NULL DEFAULT b'0' COMMENT '删除标识:0未删除,1删除', PRIMARY KEY (`id`) ) ENGINE=InnoDB DEFAULT CHARSET=utf8; INSERT INTO `ixt_customer_note_comment` VALUES ('1111', 'a1', '1001', INSERT INTO `ixt_customer_note_comment` VALUES ('1212', 'a2', '1001', INSERT INTO `ixt_customer_note_comment` VALUES ('2121', 'b3', '1002', INSERT INTO `ixt_customer_note_comment` VALUES ('321', 'b1', '1002', INSERT INTO `ixt_customer_note_comment` VALUES ('3221', 'c1', '1000', INSERT INTO `ixt_customer_note_comment` VALUES ('421', 'b2', '1002', |
ixt_customer_note_reply 表结构: |
ixt_customer_note_reply 表sql语句: DROP TABLE IF EXISTS `ixt_customer_note_reply`; CREATE TABLE `ixt_customer_note_reply` ( `id` varchar(50) NOT NULL COMMENT '主键UUID', `customerId` varchar(50) NOT NULL COMMENT '回复用户id', `commentDataId` varchar(50) DEFAULT NULL COMMENT '被回复的评论ID', `parentReplyDataId` varchar(50) DEFAULT NULL COMMENT '被回复的回复的id', `content` varchar(1000) NOT NULL COMMENT '回复内容', `createUser` varchar(50) DEFAULT NULL COMMENT '创建人ID', `createDate` datetime DEFAULT NULL COMMENT '更新人ID', `updateUser` varchar(50) DEFAULT NULL COMMENT '更新时间', `updateDate` datetime DEFAULT NULL COMMENT '更新时间', `deleteFlag` bit(1) NOT NULL DEFAULT b'0' COMMENT '删除标识:0未删除,1删除', PRIMARY KEY (`id`) ) ENGINE=InnoDB DEFAULT CHARSET=utf8; INSERT INTO `ixt_customer_note_reply` VALUES ('1212', 'v1', '3221', INSERT INTO `ixt_customer_note_reply` VALUES ('3121', 'v2', '3221', INSERT INTO `ixt_customer_note_reply` VALUES ('431', 'v3', '421', INSERT INTO `ixt_customer_note_reply` VALUES ('5231', 'v4', '421', |
2、分别查出三张表的数据:
2.1、查询用户说说表倒序显示
select createDate, id as dataId, customerId, concat('发表说说:',content) as content from ixt_customer_note order by createDate desc;
2.2、查询说说的评论正序显示
select nc.createDate, nc.dataId, nc.customerId,
concat('发表评论:',nc.content) as content from ixt_customer_note_comment nc
left join ixt_customer_note n on nc.dataId=n.id order by n.createDate
desc, nc.createDate asc;
2.3、查询说说的评论的回复正序显示
select nr.createDate, nc.dataId, nr.customerId,
concat('发表回复:',nr.content) as content from ixt_customer_note_reply nr
left join ixt_customer_note_comment nc on nr.commentDataId=nc.id left
join ixt_customer_note n on nc.dataId=n.id order by n.createDate desc,
nc.createDate asc, nr.createDate asc;
3、有了这三张表数据后,如何将他们显示为一张表,最终得到树形结构?
如果想要得到树形展示,可以考虑能否将三张表返回的结果合并为一张表,因为他们的结果合并在一起后正是我们需要的所有数据,只不过最终展示的效果要调整一下。
好的,先考虑合并用户说说及说说的评论,并按树形结构展示,这时我们应该使用 union关键字,求并集。观察一下,合并之后的结果集,应该先根据说说的发表时间倒序,再根据说说的评论的发表时间正序,所以写sql执行一下:
大致的语句为:select * from(说说的结果集 union 评论的结果集) as T order by 说说.createDate desc, 评论.createDate asc;
select * from((select createDate as createDate1, "" as createDate2,
id as dataId, customerId, concat('发表说说:',content) as content from
ixt_customer_note order by createDate desc) union (select n.createDate
as createDate1, nc.createDate as createDate2, nc.dataId, nc.customerId,
concat(' 发表评论:',nc.content) as content from ixt_customer_note_comment
nc left join ixt_customer_note n on nc.dataId=n.id order by
n.createDate desc, nc.createDate asc)) as T order by createDate1 desc,
createDate2 asc;
4、上面合并结果集是我们想要的结果,好的,再来合并回复结果集。合并之后的结果集应该按说说的发表时间倒序,再按评论的发表时间正序,再按回复的发表时间正序。为了区分出每条记录是哪张表的,我们在结果集中添加一个字段type,表示记录的类型:t1是说说,t2是评论,t3是回复。
sql语句:select * from(说说的结果集 union 评论的结果集 union 回复的结果集) as T order
by 说说.createDate desc, 评论.createDate asc, 回复.createDate asc;
select * from((select createDate as createDate1, "" as createDate2,
"" as createDate3, "t1" as type, id as dataId, customerId,
concat('发表说说:',content) as content from ixt_customer_note order by
createDate desc) union (select n.createDate as createDate1,
nc.createDate as createDate2, "" as createDate3, "t2" as type,
nc.dataId, nc.customerId, concat(' 发表评论:',nc.content) as content from
ixt_customer_note_comment nc left join ixt_customer_note n
on nc.dataId=n.id order by n.createDate desc, nc.createDate asc) union
(select n.createDate as createDate1, nc.createDate as createDate2,
nr.createDate as createDate3, "t3" as type, nc.dataId, nr.customerId,
concat(' 发表回复:',nr.content) as content from
ixt_customer_note_reply nr left join ixt_customer_note_comment nc on
nr.commentDataId=nc.id left join ixt_customer_note n on nc.dataId=n.id
order by n.createDate desc, nc.createDate asc, nr.createDate asc)) as T
order by createDate1 desc, createDate2 asc, createDate3 asc;
5、上面结果集是我们想要的,不过createDate最终应该只有一个,可以继续改进,将createDate合并为一列,说说显示createDate1,评论显示createDate2,回复显示createDate3。
改进后的语句如下:
select
if(T.type='t1',T.createDate1,(if(T.type='t2',T.createDate2,T.createDate3)))
as createDate, T.type, T.dataId, T.customerId, T.content from((select
createDate as createDate1, "" as createDate2, "" as createDate3, "t1"
as type,customerId, id as dataId, concat('发表说说:',content) as content
from ixt_customer_note order by createDate desc) union (select
n.createDate as createDate1, nc.createDate as createDate2, "" as
createDate3, "t2" as type, nc.customerId, nc.dataId, concat('
发表评论:',nc.content) as content from ixt_customer_note_comment nc left
join ixt_customer_note n on nc.dataId=n.id order by n.createDate desc,
nc.createDate asc) union (select n.createDate as createDate1,
nc.createDate as createDate2, nr.createDate as createDate3, "t3" as
type, nr.customerId, nc.dataId, concat(' 发表回复:',nr.content) as
content from ixt_customer_note_reply nr left join
ixt_customer_note_comment nc on nr.commentDataId=nc.id left join
ixt_customer_note n on nc.dataId=n.id order by n.createDate desc,
nc.createDate asc, nr.createDate asc)) as T order by createDate1 desc,
createDate2 asc, createDate3 asc;
myql 查询树形表结果:说说、说说的述评、评论的回复的更多相关文章
- myql查询创建表语句SHOW CREATE TABLE table_name
技术背景:刚开始学习MySQL时候,有时偷懒,会用SHOW CREATE TABLE 表名\G来复制表创建语句,可是当运行的时候总会因为"表名和列名上有单引号",提示语法错误不能运 ...
- oracle 树形表结构查询 排序
oracle 树形表结构排序 select * from Table start with parentid is null connect by prior id=parentid order SI ...
- SQL Server 树形表非循环递归查询
很多人可能想要查询整个树形表关联的内容都会通过循环递归来查...事实上在微软在SQL2005或以上版本就能用别的语法进行查询,下面是示例. --通过子节点查询父节点WITH TREE AS( ...
- sql查询指定表外键约束
//////////////////查询指定表外键约束select a.name as 约束名, object_name(b.parent_object_id) as 外键表, d.name as 外 ...
- SQL查询每个表的字段数量
--SQL查询每个表的字段数量select b.[name], count(*) As AllCount,ISNULL(ISNULL(sum(case when isnullable=0 then 1 ...
- SQL server 查询某个表在哪些存储过程(SP)中使用到
1.查询某个表被哪些存储过程(以下简称 SP)使用到 : select distinct object_name(id) from syscomments where id in (select ob ...
- SQL 查询所有表名、字段名、类型、长度、存储过程、视图
-- 获得存储过程创建语句 select o.xtype,o.name,cm.text from syscomments cm inner join sysobjects o on o.id=cm.i ...
- 【转】oracle查询用户表,函数,储存过程,
◆Oracle查询用户表空间:select * from user_all_tables ◆Oracle查询所有函数和储存过程:select * from user_source ◆Oracle查询所 ...
- SQL语句查询某表的所有字段及数据类型
SQL语句查询某表的所有字段及数据类型 SELECT name AS column_name , TYPE_NAME(system_type_id) AS column_type , max_leng ...
随机推荐
- jquery-三级联动
html <!DOCTYPE html> <html> <head> <meta charset=gbk /> <title>selectL ...
- eclipse连接多个git仓库方法
只需要在本地建立多个仓库就行,提交的时候一个本地仓库对应一个git仓库
- webView和js交互
与 js 交互 OC 调用 JS // 执行 js - (void)webViewDidFinishLoad:(UIWebView *)webView { NSString *title = [web ...
- Swift中的类型转换
写在前面:1,类型转换的两种方式 2,as!和as?的用法 3,类型判断中 is和===的用法 类型转换方式一,利用类型的构造器进行转换 let str = " var i = Int(st ...
- coreseek安装
一. Sphinx简介 Sphinx是由俄罗斯人Andrew Aksyonoff开发的一个全文检索引擎.意图为其他应用提供高速.低空间占用.高结果 相关度的全文搜索功能.Sphinx可以非常容易的与 ...
- ubuntu下非root用户下获得使用wireshark的权限
在非root用户下不能使用wireshark用来抓包,所以需要进行以下操作: sudo groupadd wireshark sudo chgrp wireshark /usr/bin/dumpcap ...
- ECSHOP 商品字段增加新字段的方法
结合ecshop后台“商品编辑”.“商品录入”来谈谈如何给ecshop商品增加一个新字段,假设我们将这个新字段命名为 new_add 1.首先要修改数据表结构,给表 ecs_goods 增加新字段:n ...
- c语言一些知识点的记录
1.extern关键字 extern关键字可以置于变量或者函数前面,用于告诉编译器此变量或函数定义于其他的模块.
- 在QtCreator中使用doxygen
接触Doxygen后,认识到其强大之处,一口气将之前的烂代码重构了一遍,所有的文件头,函数注释等等都是手动添加注释.在keil中可以看到其对JavaDoc风格的注释有高亮,非常好看.但是keil这个I ...
- EF Core 杂记
本系列文章,将介绍本人在学习和使用EF Core的过程中的收获与心得. 或许有的地方讲的错误 欢迎大家批评指出. 1.EF Core 数据库迁移(Migration)