----------------------------------------------------------------------------------------------------------------------==
-----== solr 一对多数据导入solr处理
前言：
我们平时的业务中，很可能出现，一对多的情况，比如一个人有多个爱好，一个教练有多个特长，我们在筛选的时候，想通过一个条件筛选出所有有此条件的数据.
例如 A有[1,2,3]属性，B有[2,5,8]属性，C有[2,9,8]属性，如果我想筛选2属性，那么A、B、C三个数据，应该都得筛选出来；如果筛选2,8属性，那么结果应该是B、C。
这种情况，大多都会有三张表（我这里说的就是带关系表的这种情况），那么，我们该怎么写data-config.xml来方便数据导入呢？
下面，我们就来看看，要怎么处理！
1、数据表情景
-- 教练表
mysql> select id,name,gender,top,area_city,area_street,createtime,updatetime from coach limit 5;
+----+-----------+--------+-----+-----------+-------------+---------------------+---------------------+
| id | name | gender | top | area_city | area_street | createtime | updatetime |
+----+-----------+--------+-----+-----------+-------------+---------------------+---------------------+
| 1 | 王大伟 | 1 | 0 | 0 | 0 | 2019-01-13 15:44:32 | 2019-07-11 12:56:09 |
| 2 | 李佳钊 | 1 | 0 | 0 | 0 | 2019-01-13 15:44:32 | 2019-07-17 13:51:49 |
| 3 | 于春雷 | 1 | 0 | 0 | 0 | 2019-01-13 15:44:32 | 2019-07-11 12:56:09 |
| 4 | 张咪 | 2 | 0 | 0 | 0 | 2019-01-13 15:44:32 | 2019-07-11 12:56:09 |
| 11 | 陈云鹏 | 1 | 0 | 0 | 0 | 2019-01-13 15:44:32 | 2019-07-11 12:56:09 |
+----+-----------+--------+-----+-----------+-------------+---------------------+---------------------+
5 rows in set (0.00 sec)
-- 特长表
mysql> select * from specialty;
+-------------+--------------------+------+---------+----------+--------+---------------------+
| specialtyid | specialty_name | type | user_id | coach_id | sortid | createtime |
+-------------+--------------------+------+---------+----------+--------+---------------------+
| 1 | 减脂 | 2 | 0 | 0 | 1 | 2019-04-27 21:16:12 |
| 2 | 塑形 | 2 | 0 | 0 | 2 | 2019-04-27 21:16:14 |
| 3 | 增肌 | 2 | 0 | 0 | 3 | 2019-04-27 21:16:17 |
| 4 | 体态调整 | 2 | 0 | 0 | 0 | 2019-04-26 21:58:13 |
| 5 | 体能训练 | 2 | 0 | 0 | 0 | 2019-04-26 21:58:13 |
+-------------+--------------------+------+---------+----------+--------+---------------------+
15 rows in set (0.00 sec)
-- 教练特长关系表
mysql> select * from coach_specialty where coach_id in (1,2,3,4,11);
+----+----------+--------------+---------------------+
| id | coach_id | specialty_id | createtime |
+----+----------+--------------+---------------------+
| 1 | 1 | 2 | 2019-01-13 15:44:32 |
| 2 | 1 | 5 | 2019-01-13 15:44:32 |
| 3 | 2 | 3 | 2019-01-13 15:44:32 |
| 4 | 2 | 4 | 2019-01-13 15:44:32 |
| 5 | 3 | 3 | 2019-01-13 15:44:32 |
| 6 | 4 | 5 | 2019-01-13 15:44:32 |
+----+----------+--------------+---------------------+
6 rows in set (0.00 sec)
2、solr数据形式
我们最终要的结果是，每个教练，将自己所有的特长都跟到自己的对应的数据中，以数组形式存储。
3、solr中sql该如何写
这块是关键点，直接关乎solr中最终导入数据的格式，所以，一定要注意！关键操作如下：
---== data-config.xml文件
<document>
<entity name="coach" dataSource="source" pk="id"
query="select id,name,head_img,certificate,score,cityid,gym_id,area_city,area_district,area_street,longitude,latitude,concat(latitude,',',longitude) as location,updatetime,gender,status,is_have_course from coach"
deltaImportQuery="select id,name,cityid,gym_id,area_city,area_district,area_street,longitude,latitude,concat(latitude,',',longitude) as location,updatetime,gender,status,is_have_course from coach where id='${dataimporter.delta.id}'"
deltaQuery="select id from coach where updatetime > '${dataimporter.last_index_time}'" >
<field column="id" name="id" />
<field column="name" name="name" />
<field column="head_img" name="head_img" />
<field column="certificate" name="certificate" />
<field column="score" name="score" />
<field column="cityid" name="cityid" />
<field column="area_city" name="area_city" />
<field column="area_district" name="area_district" />
<field column="area_street" name="area_street" />
<field column="longitude" name="longitude" />
<field column="latitude" name="latitude" />
<field column="location" name="location" />
<field column="updatetime" name="updatetime" />
<field column="gender" name="gender" />
<field column="status" name="status" />
<field column="is_have_course" name="is_have_course" />
<entity name="coach_specialty"
query="select cs.specialty_id from coach_specialty cs where cs.coach_id='${coach.id}'"
deltaQuery="select coach_id from coach_specialty where createtime > '${dataimporter.last_index_time}'"
parentDeltaQuery="select id from coach where id='${coach.id}'">
<field column="specialty_id" name="specialty_id" />【关键点1】
<entity name="specialty"
query="SELECT specialtyid, specialty_name FROM `specialty` WHERE specialtyid = '${coach_specialty.specialty_id}'" >
<field column="specialty_name" name="specialty_name" />【关键点2】
</entity>
</entity>
</entity>
</document>
---== managed-schema 文件
<field name="specialty_id" type="pint" indexed="true" stored="true" multiValued="true" />
<field name="specialty_name" type="string" indexed="true" stored="true" multiValued="true" />
注意：如果数据是数组格式的，这里一定记得加【multiValued="true"】这个属性，切记！
4、保存，重启solr，并对数据进行导入，我这里作的是clean,并且，full-import。
5、用solr控制台，query一下，最终结果
{
"responseHeader":{
"status":0,
"QTime":0,
"params":{
"q":"*:*",
"_":"1565262090439"}},
"response":{"numFound":2939,"start":0,"docs":[
{
"gender":1,
"area_city":0,
"area_street":[0],
"latitude":"0E-7",
"certificate":0,
"cityid":201,
"is_have_course":1,
"area_district":0,
"score":5.0,
"name":"王大伟",
"location":"0.0000000,0.0000000",
"id":"1",
"updatetime":"2019-07-11T12:56:09Z",
"longitude":"0E-7",
"status":2,
"specialty_id":[2,
5],
"specialty_name":["塑形",
"体能训练"],
"_version_":1641298795321884672},
{
"gender":1,
"area_city":0,
"area_street":[0],
"latitude":"0E-7",
"certificate":1,
"cityid":201,
"is_have_course":1,
"area_district":0,
"score":5.0,
"name":"李佳钊",
"location":"0.0000000,0.0000000",
"id":"2",
"updatetime":"2019-07-17T13:51:49Z",
"longitude":"0E-7",
"status":1,
"specialty_id":[3,
4],
"specialty_name":["增肌",
"体态调整"],
"_version_":1641298795521114112},
{
"gender":1,
"area_city":0,
"area_street":[0],
"latitude":"0E-7",
"certificate":2,
"cityid":201,
"is_have_course":1,
"area_district":0,
"score":5.0,
"name":"于春雷",
"location":"0.0000000,0.0000000",
"id":"3",
"updatetime":"2019-07-11T12:56:09Z",
"longitude":"0E-7",
"status":2,
"specialty_id":[3],
"specialty_name":["增肌"],
"_version_":1641298795526356992}]
}
}
结语：
到这里，就基本上完事了，剩下的就是你代码里怎么处理搜索条件的拼接了。
目前solr问题，网上，并不是很多，所以，以分享精神，积极上贴，利己利人，这样solr会越用越顺利！
----------------------------------------------------------------------------------------------------------------------==
