The source data is as follows:
The required effect is to obtain the latest data based on target_id grouping, which is:
There are currently two options in mind:
Option 1: Find the largest data ID through subqueries or joins in the same table
There is another way to write it:
<code>select * from (select * from track where type='task' and target_id in(...) ORDER BY time DESC ) as temp GROUP BY target_id</code>
Option 2: Query in two steps, first query the maximum ID in php, and then query the list data through the ID array
What I want to ask is is there any other simpler way to deal with this problem?
This kind of demand should be relatively common!
=====Attach structure and data=====
<code>DROP TABLE IF EXISTS `track`; CREATE TABLE `track` ( `id` int(11) NOT NULL AUTO_INCREMENT, `type` varchar(50) NOT NULL DEFAULT '' COMMENT 'task => 任务跟进,project => 项目跟进 ', `target_id` int(11) DEFAULT '0' COMMENT '跟进目标ID', `user_id` int(11) DEFAULT '0' COMMENT '跟进用户', `user_name` varchar(100) DEFAULT '' COMMENT '跟进用户名称', `content` varchar(500) DEFAULT '' COMMENT '跟进内容', `time` int(11) DEFAULT '0' COMMENT '跟进时间', PRIMARY KEY (`id`) ) ENGINE=InnoDB AUTO_INCREMENT=10 DEFAULT CHARSET=utf8 COMMENT='跟进记录表'; -- ---------------------------- -- Records of track -- ---------------------------- INSERT INTO `track` VALUES ('1', 'task', '67', '1', '超级管理员', '无所谓...', '1467774850'); INSERT INTO `track` VALUES ('2', 'task', '67', '1', '超级管理员', 'TTTT', '1467777620'); INSERT INTO `track` VALUES ('7', 'task', '67', '1', '超级管理员', '只耗损', '1468288894'); INSERT INTO `track` VALUES ('8', 'task', '34', '1', '超级管理员', 'STS', '1468288917'); INSERT INTO `track` VALUES ('9', 'task', '34', '1', '超级管理员', '吊顶', '1468288954');</code>
The source data is as follows:
The required effect is to obtain the latest data based on target_id grouping, which is:
There are currently two options in mind:
Option 1: Find the largest data ID through subqueries or joins in the same table
There is another way to write it:
<code>select * from (select * from track where type='task' and target_id in(...) ORDER BY time DESC ) as temp GROUP BY target_id</code>
Option 2: Query in two steps, first query the maximum ID in php, and then query the list data through the ID array
What I want to ask is is there any other simpler way to deal with this problem?
This kind of demand should be relatively common!
=====Attach structure and data=====
<code>DROP TABLE IF EXISTS `track`; CREATE TABLE `track` ( `id` int(11) NOT NULL AUTO_INCREMENT, `type` varchar(50) NOT NULL DEFAULT '' COMMENT 'task => 任务跟进,project => 项目跟进 ', `target_id` int(11) DEFAULT '0' COMMENT '跟进目标ID', `user_id` int(11) DEFAULT '0' COMMENT '跟进用户', `user_name` varchar(100) DEFAULT '' COMMENT '跟进用户名称', `content` varchar(500) DEFAULT '' COMMENT '跟进内容', `time` int(11) DEFAULT '0' COMMENT '跟进时间', PRIMARY KEY (`id`) ) ENGINE=InnoDB AUTO_INCREMENT=10 DEFAULT CHARSET=utf8 COMMENT='跟进记录表'; -- ---------------------------- -- Records of track -- ---------------------------- INSERT INTO `track` VALUES ('1', 'task', '67', '1', '超级管理员', '无所谓...', '1467774850'); INSERT INTO `track` VALUES ('2', 'task', '67', '1', '超级管理员', 'TTTT', '1467777620'); INSERT INTO `track` VALUES ('7', 'task', '67', '1', '超级管理员', '只耗损', '1468288894'); INSERT INTO `track` VALUES ('8', 'task', '34', '1', '超级管理员', 'STS', '1468288917'); INSERT INTO `track` VALUES ('9', 'task', '34', '1', '超级管理员', '吊顶', '1468288954');</code>
Put the data with the largest ID into the temporary table
<code>CREATE TEMPORARY TABLE tmp_id(`id` int(11) not null,PRIMARY KEY (`id`) ) </code>
Then
<code>INSERT INTO tmp_id SELECT max(`id`) as id FROM track GROUP BY target_id </code>
Then just join it
Finally, it is recommended to delete the temporary table explicitly
<code>DROP TEMPORARY TABLE IF EXISTS tmp_id</code>
I think plan 1A is pretty good, but why do we need to write in condition for target_id? If we don’t write it, the same result will be the same.
Generally, option 1 is used. There is another way to write option 1
<code>select * from track where id in(select substring_index(group_concat(id order by id desc),',',1) as maxid from track group by target_id);</code>