Question

我有以下结构：消息（消息表）可能有多个字段（字段表），每个字段可能有多个子字段（存储在相同的“字段”表中，唯一的区别是字段中没有值“ParentField”栏目。子字段可能有子字段等，但这并不重要。

当我检索10条每条有10个字段的消息，每个字段有20个子字段时，我可以从日志文件中看到NHibernate生成2000个SQL调用。有没有办法优化它？

谢谢！

这是NHibernate生成的2000个SQL语句之一：

SELECT   fieldresul0_.MessageResults_ID as MessageR6___2_, 
         fieldresul0_.ID as ID2_, 
         fieldresul0_.ID as ID5_1_, 
         fieldresul0_.Field_ID as Field2_5_1_, 
         fieldresul0_.Name as Name5_1_, 
         fieldresul0_.Value as Value5_1_, 
         fieldresul0_.MessagePosition as MessageP5_5_1_, 
         fieldresul0_.MessageResults_ID as MessageR6_5_1_, 
         fieldresul0_.ParentField_ID as ParentFi7_5_1_, 
         fieldresul1_.ID as ID5_0_, 
         fieldresul1_.Field_ID as Field2_5_0_, 
         fieldresul1_.Name as Name5_0_, 
         fieldresul1_.Value as Value5_0_, 
         fieldresul1_.MessagePosition as MessageP5_5_0_, 
         fieldresul1_.MessageResults_ID as MessageR6_5_0_, 
         fieldresul1_.ParentField_ID as ParentFi7_5_0_ 
FROM     FieldResults fieldresul0_ 
         LEFT OUTER JOIN FieldResults fieldresul1_ 
                      ON fieldresul0_.ParentField_ID=fieldresul1_.ID 
WHERE    fieldresul0_.MessageResults_ID=@p0 
ORDER BY fieldresul0_.MessagePosition

这是ActiveRecord生成的映射文件：

<?xml version="1.0" encoding="utf-16"?> 
    <hibernate-mapping  auto-import="true" default-lazy="false" xmlns:xsd="http://www.w3.org/2001/XMLSchema" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="urn:nhibernate-mapping-2.2">  
      <class name="FieldResult, Data" table="FieldResults"> 
        <id name="ID" access="property" column="ID" type="Int32" unsaved-value="0"> 
          <generator class="native">    
            <param name="sequence">FieldResults_ID</param>  
          </generator>  
        </id>   
        <property name="FieldID" access="property" type="String">   
          <column name="Field_ID"/> 
        </property> 
        <property name="Name" access="property" type="String">  
          <column name="Name"/> 
        </property> 
        <property name="DisplayValue" access="property" type="String">  
          <column name="Value"/>    
        </property> 
        <property name="MessagePosition" access="property" type="Int32">    
          <column name="MessagePosition"/>  
        </property> 
        <many-to-one name="ParentMessage" access="property" class="MessageResult, Data" column="MessageResults_ID" />   
        <many-to-one name="ParentField" access="property" class="FieldResult, Data" column="ParentField_ID" />  
        <bag name="Children" access="property" table="FieldResults" lazy="false" cascade="all" order-by="Field_ID"> 
          <key column="ParentField_ID" />   
          <one-to-many class="FieldResult, Data" /> 
        </bag>  
      </class>  
    </hibernate-mapping>`

Answer 1

我猜这是一个选择n + 1问题。您是否尝试过急切加载或自定义HQL查询？如果你没有

，我会发布一些示例HQL来解决这个问题

Answer 2

有一种方法可以帮助nhibernate执行更高效的连接。明确告诉查询处理器包含带有“join fetch”的查找表，并通过HQL和Session.CreateQuery（）获取结果

from FieldRestults fr 
left join fetch fr.ParentMessage 
left join fetch fr.ParentField

在我正在调查的另一个问题中优化一个包。

Castle ActiveRecord / NHibernate优化

2 个答案: