RepositoryFactoryBeanSupport
1, create a factory, get entitymanager afterPropertiesSet
2. Get the factory proxy, including the query class strategy RepositoryQuery instance, that is, the @Query tag, and put it in the map initAndReturn
A RepositoryQuery instance holds a JpaQueryMethod instance, which in turn holds a Method instance. A marked method has a JpaQueryMethod and a RepositoryQuery
initAndReturn 中 new QueryExecutorMethodInterceptor ：
Iterable<Method> queryMethods = repositoryInformation.getQueryMethods();
for (Method method : queryMethods) {
RepositoryQuery query = lookupStrategy.resolveQuery(method, repositoryInformation, namedQueries); 创建RepositoryQuery
invokeListeners(query);
queries.put(method, query);
}
JpaQueryLookupStrategy：
public final RepositoryQuery resolveQuery(Method method, RepositoryMetadata metadata, NamedQueries namedQueries) {
return resolveQuery(new JpaQueryMethod(method, metadata, provider), em, namedQueries);
}
JpaQueryLookupStrategy内部类CreateIfNotFoundQueryLookupStrategy：
@Override
protected RepositoryQuery resolveQuery(JpaQueryMethod method, EntityManager em, NamedQueries namedQueries) {
try {
return lookupStrategy.resolveQuery(method, em, namedQueries);
} catch (IllegalStateException e) {
return createStrategy.resolveQuery(method, em, namedQueries);
}
}
JpaQueryLookupStrategy：private static class CreateQueryLookupStrategy extends AbstractQueryLookupStrategy {
@Override
protected RepositoryQuery resolveQuery(JpaQueryMethod method, EntityManager em, NamedQueries namedQueries) {
RepositoryQuery query = JpaQueryFactory.INSTANCE.fromQueryAnnotation(method, em, evaluationContextProvider);
if (null != query) {
return query;
}
query = JpaQueryFactory.INSTANCE.fromProcedureAnnotation(method, em);
if (null != query) {
return query;
}
String name = method.getNamedQueryName();
if (namedQueries.hasQuery(name)) {
return JpaQueryFactory.INSTANCE.fromMethodWithQueryString(method, em, namedQueries.getQuery(name),
evaluationContextProvider);
}
query = NamedQuery.lookupFrom(method, em);
if (null != query) {
return query;
}
throw new IllegalStateException(String.format(
"Did neither find a NamedQuery nor an annotated query for method %s!", method));
}
///这里决定用什么query
//SimpleJpaQuery
//NativeJpaQuery
//PartTreeJpaQuery
//NamedQuery
//StoredProcedureJpaQuery
JpaQueryFactory：
AbstractJpaQuery fromMethodWithQueryString(JpaQueryMethod method, EntityManager em, String queryString,
EvaluationContextProvider evaluationContextProvider) {
if (queryString == null) {
return null;
}
return method.isNativeQuery() ? new NativeJpaQuery(method, em, queryString, evaluationContextProvider, PARSER) : //
new SimpleJpaQuery(method, em, queryString, evaluationContextProvider, PARSER);
}
JpaQueryMethod 就是Repository接口中带有@Query注解方法的全部信息，包括注解，类名，实参等的存储类，所以Repository接口有多少个@Query注解方法，
就会包含多少个JpaQueryMethod实例被加入监听序列。实际运行时，一个RepositoryQuery实例持有一个JpaQueryMethod实例，JpaQueryMethod又持有一个Method实例。
工具类
Parameters是一个JpaParameters实例，存储的是参数名和参数值的键值对集合，JpaParameters还包含一个内部类JpaParameter，它extends Parameter；
JpaQueryExecution
抽象类和继承类也可以放在一个类中，继承类当作抽象类的内部类
spring-data-jpa使用了内部类来实现JpaQueryExecution的子类，也就是包含多个查询运行策略的（子）类。这么做其实是一种相对的封闭形式，
造成使用者只能实例化spring-data-jpa给出的这8种子类。你无法实现自己的JpaQueryExecution子类，在spring框架初始化时通过配置加载之。
AbstractJpaQuery 用来选择具体那种JpaQueryExecution子类
JpaQueryExecution内部类SingleEntityExecution：
static class SingleEntityExecution extends JpaQueryExecution {
@Override
protected Object doExecute(AbstractJpaQuery query, Object[] values) {
return query.createQuery(values).getSingleResult();/////这里正式使用orm联系数据库
}
}
SingleEntityExecution只覆盖了一个父类方法，而且实现里只有一条语句query.createQuery(values).getSingleResult();在执行这句时，
会调用底层数据库访问提供商，spring-data-jpa默认选择的是hibernate，执行数据库存取操作之后，将结果返回给这个JpaQueryExecution实例。
参考：
http://blog.csdn.net/gaolu/article/details/53415420