solr - solr delta import 用更新的文档替换所有文档

Question

增量导入正在工作，但结果不是我们想要的。

Solr 使用完整导入替换所有添加的文档，并更新/添加已更新的文档，而不是仅更新/删除必要的文档。

是否有一些配置选项可以保留没有更新的文档？或者有人可以向我发送一个带有工作增量导入的示例数据配置，以便我可以从中学习并看看我做错了什么？

Answer 1

我的数据配置文件使事情变得更容易。

<dataConfig>
    <dataSource convertType="true" type="JdbcDataSource"
   driver="com.mysql.jdbc.Driver"
   url="jdbc:mysql://localhost/table1"
   user="user_db"
   password="password"/>

    <document name="content">

        <entity name="id" pk="id"
query="SELECT pp.*, pp.profile_positions_id as id, p.position as position_name, b1.name as branche_name, b2.name as sub_branche_name, pc.name as position_class_name,
p1.user_id as profile_user_id, p1.company_id as company_id, p1.active as profile_active, p1.visibility as profile_visibility, p1.profile_type as profile_type, p1.gender as profile_gender, p1.initials as profile_initials, p1.firstname as profile_firstname, p1.infix as profile_infix, p1.lastname as profile_lastname, p1.profile_description as profile_description, p1.email as profile_email, p1.zip as profile_zip, p1.lat as profile_lat, p1.lon as profile_lon, p1.city as profile_city, p1.travel_ability as profile_travel_ability, p1.status as profile_status, p1.deleted as profile_deleted, p1.deleted as profile_active,
c.company_name as company_name, c.businesstype as company_businesstype, c.amount_of_people as company_amount_of_people, c.visibility as company_visibility, c.deleted as company_deleted, c.valid as company_valid, pp.level as level,

u.deleted as user_deleted, pp.deleted as position_deleted

FROM profile_positions pp
    LEFT JOIN profiles p1 ON p1.profile_id = pp.profile_id
    LEFT JOIN companies c ON c.company_id = p1.company_id
    LEFT JOIN positions p ON p.position_id = pp.position_id
    LEFT JOIN branches b1 ON b1.branche_id = pp.branche_id
    LEFT JOIN branches b2 ON b2.branche_id = pp.sub_branche_id
    LEFT JOIN position_classes pc ON pc.position_class_id = pp.position_class
    LEFT JOIN users u ON u.user_id = p1.user_id;"

deltaImportQuery="SELECT pp.*, pp.profile_positions_id as id, p.position as position_name, b1.name as branche_name, b2.name as sub_branche_name, pc.name as position_class_name,
p1.user_id as profile_user_id, p1.company_id as company_id, p1.active as profile_active, p1.visibility as profile_visibility, p1.profile_type as profile_type, p1.gender as profile_gender, p1.initials as profile_initials, p1.firstname as profile_firstname, p1.infix as profile_infix, p1.lastname as profile_lastname, p1.profile_description as profile_description, p1.email as profile_email, p1.zip as profile_zip, p1.lat as profile_lat, p1.lon as profile_lon, p1.city as profile_city, p1.travel_ability as profile_travel_ability, p1.status as profile_status, p1.deleted as profile_deleted, p1.deleted as profile_active,
c.company_name as company_name, c.businesstype as company_businesstype, c.amount_of_people as company_amount_of_people, c.visibility as company_visibility, c.deleted as company_deleted, c.valid as company_valid, pp.level as level,

u.deleted as user_deleted, pp.deleted as position_deleted

FROM profile_positions pp
    LEFT JOIN profiles p1 ON p1.profile_id = pp.profile_id
    LEFT JOIN companies c ON c.company_id = p1.company_id
    LEFT JOIN positions p ON p.position_id = pp.position_id
    LEFT JOIN branches b1 ON b1.branche_id = pp.branche_id
    LEFT JOIN branches b2 ON b2.branche_id = pp.sub_branche_id
    LEFT JOIN position_classes pc ON pc.position_class_id = pp.position_class
    LEFT JOIN users u ON u.user_id = p1.user_id
WHERE pp.profile_positions_id='${dih.delta.id}';"

deltaQuery="select profile_positions_id from profile_positions pp

LEFT JOIN profiles p1 ON p1.profile_id = pp.profile_id

where updated_at &gt;= '${dih.last_index_time}' OR created_at &gt;= '${dih.last_index_time}' OR company_id IN(
SELECT company_id FROM companies where updated_at &gt;= '${dih.last_index_time}' OR created_at &gt;= '${dih.last_index_time}') OR user_id IN(
SELECT user_id FROM users where updated_at &gt;= '${dih.last_index_time}' OR created_at &gt;= '${dih.last_index_time}');">

   <field column="id" name="id" />
   <field column="profile_id" name="profile_id" />
   <field column="position_id" name="position_id" />

   <field column="position_name" name="position_name" />
   <field column="position_name" name="position_name_1" />
   <field column="position_name" name="position_name_2" />

   <field column="branche_id" name="branche_id" />
   <field column="branche_name" name="branche_name" />
   <field column="sub_branche_id" name="sub_branche_id" />
   <field column="sub_branche_name" name="sub_branche_name" />
   <field column="position_class" name="position_class" />
   <field column="position_class_name" name="position_class_name" />
   <field column="level" name="level" />
   <field column="experience" name="experience" />
   <field column="grade" name="grade" />
   <field column="position_deleted" name="position_deleted" />

<!--   Company -->
   <field column="profile_user_id" name="user_id" />
   <field column="company_id" name="company_id" />
   <field column="profile_active" name="active" />
   <field column="profile_visibility" name="visibility" />
   <field column="profile_type" name="profile_type" />
   <field column="profile_gender" name="gender" />
   <field column="profile_initials" name="initials" />
   <field column="profile_firstname" name="firstname" />
   <field column="profile_infix" name="infix" />
   <field column="profile_lastname" name="lastname" />

   <field column="profile_email" name="email" />
   <field column="profile_zip" name="zip" />

   <field column="profile_lat" name="latlng_0_coordinate" />
   <field column="profile_lon" name="latlng_1_coordinate" />

   <field column="profile_city" name="city" />
   <field column="profile_travel_ability" name="travel_ability" />
   <field column="profile_status" name="profile_status" />
   <field column="profile_active" name="profile_active" />
   <field column="profile_deleted" name="profile_deleted" />

   <field column="user_deleted" name="user_deleted" />

<!--    Profile -->
   <field column="company_name" name="company_name" />
   <field column="company_businesstype" name="businesstype" />
   <field column="company_amount_of_people" name="amount_of_people" />
   <field column="company_visibility" name="company_visibility" />
   <field column="company_valid" name="company_valid" />
   <field column="company_deleted" name="company_deleted" />

        </entity>
    </document>
</dataConfig>