首页 1 2 3 4 5 6 7

Spring @Conditional注解源码分析

本文通过阅读@Conditional注解、Condition接口、ConditionEvaluator类以及@ConditionalOnProperty(Spring Boot提供)注解源码，深入分析Spring Conditional的实现原理。

源码版本

由于不同版本的spring代码实现细节可能存在差异，此处记录一下本文使用的源码版本：

spring 5.2.12.RELEASE
spring-boot 2.3.8.RELEASE

@Conditional注解

标注在类或方法上，当所有指定Condition都匹配时，才允许向spring容器注册组件。

如果一个@Configuration类标注了@Conditional注解，则与该类关联的所有@Bean方法、@Import注解和@ComponentScan注解都将受指定Condition约束。

注解定义：

@Target({ElementType.TYPE, ElementType.METHOD})
public @interface Conditional {

	/**
	 * 当所有指定Condition都匹配时，才允许向spring容器注册组件。
	 *
	 * 这个泛型的意思是接受所有Condition接口的实现类。
	 */
	Class<? extends Condition>[] value();
}

Condition接口

匹配条件。

在注册BeanDefinition之前立即检查条件，并且可以根据当时可以确定的任何条件取消组件注册。

Condition必须遵循与BeanFactoryPostProcessor相同的限制，并注意不要与bean实例交互。如果需要对与@Configuration bean交互的Condition进行更细粒度的控制，请考虑实现ConfigurationCondition接口。

接口定义：

public interface Condition {

	/**
	 * 确定条件是否匹配。
	 */
	boolean matches(ConditionContext context, AnnotatedTypeMetadata metadata);
}

ConditionContext接口

用于获取BeanDefinitionRegistry、BeanFactory、Environment等。

只有一个ConditionEvaluator.ConditionContextImpl实现类(一个内部类)：

class ConditionEvaluator {

	private final ConditionContextImpl context;

	public ConditionEvaluator(BeanDefinitionRegistry registry,
			Environment environment, ResourceLoader resourceLoader) {
		this.context = new ConditionContextImpl(registry, environment, resourceLoader);
	}

ConditionEvaluator类

用于处理Conditional相关注解。

核心的逻辑都在这个类里面：

public boolean shouldSkip(AnnotatedTypeMetadata metadata, ConfigurationPhase phase) {
    // 1. 判断目标组件是否被Conditional标注
	if (metadata == null || !metadata.isAnnotated(Conditional.class.getName())) {
		return false;
	}

	if (phase == null) {
		if (metadata instanceof AnnotationMetadata &&
				ConfigurationClassUtils.isConfigurationCandidate((AnnotationMetadata) metadata)) {
			return shouldSkip(metadata, ConfigurationPhase.PARSE_CONFIGURATION);
		}
		return shouldSkip(metadata, ConfigurationPhase.REGISTER_BEAN);
	}

    // 2. 获取到所有的Condition并实例化
	List<Condition> conditions = new ArrayList<>();
	for (String[] conditionClasses : getConditionClasses(metadata)) {
		for (String conditionClass : conditionClasses) {
			Condition condition = getCondition(conditionClass, this.context.getClassLoader());
			conditions.add(condition);
		}
	}

    // 3. Condition排序
	AnnotationAwareOrderComparator.sort(conditions);

    // 4. Condition匹配验证
	for (Condition condition : conditions) {
		ConfigurationPhase requiredPhase = null;
		if (condition instanceof ConfigurationCondition) {
			requiredPhase = ((ConfigurationCondition) condition).getConfigurationPhase();
		}
		if ((requiredPhase == null || requiredPhase == phase) &&
            !condition.matches(this.context, metadata)) {
			return true;
		}
	}

	return false;
}

核心功能4步：

判断目标组件是否被Conditional标注
获取到所有的Condition并实例化
Condition排序
Condition匹配验证

下文将展开说明这4个步骤。

判断目标组件是否被Conditional标注

AnnotatedTypeMetadata中封装了目标组件的注解元信息，可以通过他获取到目标组件的注解相关信息，比如是否被某个注解标注、某个注解的属性等。

AnnotatedTypeMetadata中定义了该方法的实现方式：

default boolean isAnnotated(String annotationName) {
    // 通过MergedAnnotations来获取注解标注状态
	return getAnnotations().isPresent(annotationName);
}

我们在之前的《Spring-@Bean注解源码分析》中介绍过，此处使用的是StandardAnnotationMetadata实现类，而该类中的MergedAnnotations是使用以下方式创建和获取的：

public StandardAnnotationMetadata(Class<?> introspectedClass, boolean nestedAnnotationsAsMap) {
	super(introspectedClass);
    // 这里使用的是TypeMappedAnnotations实现类
	this.mergedAnnotations = MergedAnnotations.from(introspectedClass,
			SearchStrategy.INHERITED_ANNOTATIONS, RepeatableContainers.none());
	this.nestedAnnotationsAsMap = nestedAnnotationsAsMap;
}

@Override
public MergedAnnotations getAnnotations() {
	return this.mergedAnnotations;
}

所以isAnnotated的核心判断逻辑在TypeMappedAnnotations的isPresent方法中：

public boolean isPresent(String annotationType) {
	if (this.annotationFilter.matches(annotationType)) {
		return false;
	}
	return Boolean.TRUE.equals(scan(annotationType,
			IsPresent.get(this.repeatableContainers, this.annotationFilter, false)));
}

scan方法的详细逻辑还是使用java反射的Class.getDeclaredAnnotations()方法来实现的，此处不展开说明，后续会有专门的章节介绍。

获取到所有的Condition并实例化

// 2. 获取到所有的Condition并实例化
List<Condition> conditions = new ArrayList<>();
// 获取Condition集
for (String[] conditionClasses : getConditionClasses(metadata)) {
	for (String conditionClass : conditionClasses) {
        // 创建Condition实例
		Condition condition = getCondition(conditionClass, this.context.getClassLoader());
		conditions.add(condition);
	}
}

// 这个方法从AnnotatedTypeMetadata中获取所有Conditional注解指定的Condition类名集
private List<String[]> getConditionClasses(AnnotatedTypeMetadata metadata) {
	MultiValueMap<String, Object> attributes =
        metadata.getAllAnnotationAttributes(Conditional.class.getName(), true);
	Object values = (attributes != null ? attributes.get("value") : null);
	return (List<String[]>) (values != null ? values : Collections.emptyList());
}

AnnotatedTypeMetadata接口getAllAnnotationAttributes方法，这是一个default方法，用于获取指定注解的Attribute集：

default MultiValueMap<String, Object> getAllAnnotationAttributes(
		String annotationName, boolean classValuesAsString) {

	Adapt[] adaptations = Adapt.values(classValuesAsString, true);
	return getAnnotations().stream(annotationName) // 获取到所有的注解信息
			.filter(MergedAnnotationPredicates.unique(MergedAnnotation::getMetaTypes)) // 过滤
			.map(MergedAnnotation::withNonMergedAttributes)
			.collect(MergedAnnotationCollectors.toMultiValueMap(map ->
					map.isEmpty() ? null : map, adaptations));
}

getAnnotations()返回的是TypeMappedAnnotations类型对象，他的stream方法：

public <A extends Annotation> Stream<MergedAnnotation<A>> stream(String annotationType) {
	if (this.annotationFilter == AnnotationFilter.ALL) {
		return Stream.empty();
	}
    // 这里使用java api创建Stream
	return StreamSupport.stream(spliterator(annotationType), false);
}

private <A extends Annotation> Spliterator<MergedAnnotation<A>> spliterator(Object annotationType) {
	return new AggregatesSpliterator<>(annotationType, getAggregates());
}

private List<Aggregate> getAggregates() {
	List<Aggregate> aggregates = this.aggregates;
	if (aggregates == null) {
        // 这里扫描所有注解
		aggregates = scan(this, new AggregatesCollector());
		if (aggregates == null || aggregates.isEmpty()) {
			aggregates = Collections.emptyList();
		}
		this.aggregates = aggregates;
	}
	return aggregates;
}

// 这里扫描所有注解
private <C, R> R scan(C criteria, AnnotationsProcessor<C, R> processor) {
	if (this.annotations != null) {
		R result = processor.doWithAnnotations(criteria, 0, this.source, this.annotations);
		return processor.finish(result);
	}
	if (this.element != null && this.searchStrategy != null) {
		return AnnotationsScanner.scan(criteria, this.element, this.searchStrategy, processor);
	}
	return null;
}

scan方法里面会通过递归方式使用java反射api从组件类getAnnotations以便获取到所有的注解信息，代码较多，此处不展开记录。

Condition排序

判断Condition类实现了PriorityOrdered接口，并使用PriorityOrdered进行排序
判断Condition类实现了Ordered接口，并使用Ordered进行排序
从Condition类获取Order注解，并使用Order注解的值进行排序

AnnotationAwareOrderComparator.sort(conditions);

public static void sort(List<?> list) {
	if (list.size() > 1) {
		list.sort(INSTANCE); // 这里的Comparator使用的是AnnotationAwareOrderComparator实现类
	}
}

AnnotationAwareOrderComparator实现类：

public class AnnotationAwareOrderComparator extends OrderComparator {

	@Override
	protected Integer findOrder(Object obj) {
        // 先使用父类方法获取排序，其实就是判断Ordered实现并获取排序，在本例中显然获取不到
		Integer order = super.findOrder(obj);
		if (order != null) {
			return order;
		}
        // 从类标注的注解中获取排序
		return findOrderFromAnnotation(obj);
	}

	private Integer findOrderFromAnnotation(Object obj) {
		AnnotatedElement element = (obj instanceof AnnotatedElement ?
                                    (AnnotatedElement) obj : obj.getClass());
		MergedAnnotations annotations = MergedAnnotations.from(element, SearchStrategy.TYPE_HIERARCHY);
        // 从类标注的注解中获取排序，比如Order注解，此处不展开记录了
		Integer order = OrderUtils.getOrderFromAnnotations(element, annotations);
		if (order == null && obj instanceof DecoratingProxy) {
			return findOrderFromAnnotation(((DecoratingProxy) obj).getDecoratedClass());
		}
		return order;
	}

	@Override
	public Integer getPriority(Object obj) {
		if (obj instanceof Class) {
			return OrderUtils.getPriority((Class<?>) obj);
		}
		Integer priority = OrderUtils.getPriority(obj.getClass());
		if (priority == null  && obj instanceof DecoratingProxy) {
			return getPriority(((DecoratingProxy) obj).getDecoratedClass());
		}
		return priority;
	}
}

public class OrderComparator implements Comparator<Object> {

	@Override
	public int compare(Object o1, Object o2) {
		return doCompare(o1, o2, null);
	}

	private int doCompare(Object o1, Object o2, OrderSourceProvider sourceProvider) {
        // 优先级排序判断
		boolean p1 = (o1 instanceof PriorityOrdered);
		boolean p2 = (o2 instanceof PriorityOrdered);
		if (p1 && !p2) {
			return -1;
		} else if (p2 && !p1) {
			return 1;
		}

        // 分别获取到order并比较
		int i1 = getOrder(o1, sourceProvider);
		int i2 = getOrder(o2, sourceProvider);
		return Integer.compare(i1, i2);
	}

	private int getOrder(Object obj, OrderSourceProvider sourceProvider) {
		Integer order = null;
		if (obj != null && sourceProvider != null) {
			Object orderSource = sourceProvider.getOrderSource(obj);
			if (orderSource != null) {
				if (orderSource.getClass().isArray()) {
					for (Object source : ObjectUtils.toObjectArray(orderSource)) {
						order = findOrder(source);
						if (order != null) {
							break;
						}
					}
				} else {
					order = findOrder(orderSource);
				}
			}
		}
        // 直接执行这里
		return (order != null ? order : getOrder(obj));
	}

	protected int getOrder(Object obj) {
		if (obj != null) {
            // 继续调用findOrder获取排序
            // 在此处场景下调用AnnotationAwareOrderComparator的findOrder方法
			Integer order = findOrder(obj);
			if (order != null) {
				return order;
			}
		}
		return Ordered.LOWEST_PRECEDENCE;
	}

	protected Integer findOrder(Object obj) {
		return (obj instanceof Ordered ? ((Ordered) obj).getOrder() : null);
	}

	public Integer getPriority(Object obj) {
		return null;
	}
}

Condition匹配验证

for (Condition condition : conditions) {
	ConfigurationPhase requiredPhase = null;
	if (condition instanceof ConfigurationCondition) {
		requiredPhase = ((ConfigurationCondition) condition).getConfigurationPhase();
	}
	if ((requiredPhase == null || requiredPhase == phase) &&
        !condition.matches(this.context, metadata)) {
		return true;
	}
}

@ConditionalOnProperty注解和OnPropertyCondition类

@Conditional that checks if the specified properties have a specific value. By default the properties must be present in the Environment and not equal to false. The havingValue() and matchIfMissing() attributes allow further customizations.

@Conditional(OnPropertyCondition.class)
public @interface ConditionalOnProperty {

	String[] value() default {};

	String prefix() default "";

	String[] name() default {};

	String havingValue() default "";

	boolean matchIfMissing() default false;
}

OnPropertyCondition从env中检测指定env参数是否配置了指定的值，只有满足时才允许装配目标组件。

如何判断被@Conditional注解标注

前面已经介绍，spring是使用(AnnotatedTypeMetadata)metadata.isAnnotated(Conditional.class.getName())来判断组件被@Conditional注解标注的。但是@Conditional标注在@ConditionalOnProperty注解上，使用普通的(Class)clazz.isAnnotationPresent(Conditional.class)方式无法判断，那么metadata.isAnnotated方法是如何判断的呢？

从上面@ConditionalOnProperty注解定义可以知道，目标组件类标注了@ConditionalOnProperty注解，@ConditionalOnProperty注解又标注了@Conditional注解，这显然是一个递归，只有从目标组件类开始递归解析，就可以解析出目标组件类上的所有注解，以下是一个我自己编写的简单示例：

@EnableAspectJAutoProxy
@Configuration
@ComponentScan("org.net5ijy.mybatis.test.config.aop")
@ConditionalOnAop
public class ServiceAopConfig {}

@Retention(RetentionPolicy.RUNTIME)
@Target({ ElementType.TYPE, ElementType.METHOD })
@Documented
@Conditional(AopCondition.class)
public @interface ConditionalOnAop {}

@Test
public void testAllResolveAnnotations() {

  // 不解析以下三个注解
  Set<Class<?>> exclude = new HashSet<Class<?>>() {{
    add(Target.class);
    add(Retention.class);
    add(Documented.class);
  }};

  // 存放所有注解
  List<Annotation> annotationList = new ArrayList<>();

  // 首先获取一次组件类直接标注的注解
  Annotation[] annotations = ServiceAopConfig.class.getDeclaredAnnotations();

  // 递归解析
  while (annotations.length > 0) {
    annotationList.addAll(Arrays.asList(annotations));

    // 递归解析
    List<Annotation> tmp = new ArrayList<>();
    for (Annotation annotation : annotations) {
      Annotation[] list = annotation.annotationType().getDeclaredAnnotations();
      for (Annotation a : list) {
        if (!exclude.contains(a.annotationType())) {
          tmp.add(a);
        }
      }
    }

    annotations = tmp.toArray(new Annotation[0]);
  }
  // 循环结束之后，ServiceAopConfig直接和间接标注的所有注解就都解析出来了

  for (Annotation annotation : annotationList) {
    System.out.println(annotation);
  }
}

AnnotatedTypeMetadata接口和StandardAnnotationMetadata类

上文介绍了，spring是使用(AnnotatedTypeMetadata)metadata.isAnnotated(Conditional.class.getName())来判断组件被@Conditional注解标注的。

public interface AnnotatedTypeMetadata {

	/**
	 * Return annotation details based on the direct annotations of the underlying element.
	 * @return merged annotations based on the direct annotations
	 * @since 5.2
	 */
	MergedAnnotations getAnnotations();

	/**
	 * Determine whether the underlying element has an
	 * annotation or meta-annotation of the given type defined.
	 * @return whether a matching annotation is defined
	 */
	default boolean isAnnotated(String annotationName) {
		return getAnnotations().isPresent(annotationName);
	}

    // ...
}

在我们分析的场景，此处使用的是StandardAnnotationMetadata实现类。StandardAnnotationMetadata实现类getAnnotations()返回TypeMappedAnnotations类型对象。

TypeMappedAnnotations类isPresent方法：

public boolean isPresent(String annotationType) {
	if (this.annotationFilter.matches(annotationType)) {
		return false;
	}
    // 在scan方法中
	return Boolean.TRUE.equals(scan(annotationType,
			IsPresent.get(this.repeatableContainers, this.annotationFilter, false)));
}

private <C, R> R scan(C criteria, AnnotationsProcessor<C, R> processor) {
	if (this.annotations != null) {
		R result = processor.doWithAnnotations(criteria, 0, this.source, this.annotations);
		return processor.finish(result);
	}
	if (this.element != null && this.searchStrategy != null) {
        // 执行到这里
		return AnnotationsScanner.scan(criteria, this.element, this.searchStrategy, processor);
	}
	return null;
}

AnnotationsScanner.scan方法：

static <C, R> R scan(C context, AnnotatedElement source, SearchStrategy searchStrategy,
		AnnotationsProcessor<C, R> processor) {
	R result = process(context, source, searchStrategy, processor);
	return processor.finish(result);
}

// 层层调用到processElement方法
private static <C, R> R processElement(C context, AnnotatedElement source,
		AnnotationsProcessor<C, R> processor) {

	try {
		R result = processor.doWithAggregate(context, 0);
        // 继续调用processor.doWithAnnotations方法
        // 这里的processor是TypeMappedAnnotations.IsPresent类型对象
		return (result != null ? result : processor.doWithAnnotations(
			context, 0, source, getDeclaredAnnotations(source, false)));
	} catch (Throwable ex) {
		AnnotationUtils.handleIntrospectionFailure(source, ex);
	}
	return null;
}

TypeMappedAnnotations.IsPresent类doWithAnnotations方法：

public Boolean doWithAnnotations(Object requiredType, int aggregateIndex,
		Object source, Annotation[] annotations) {
	for (Annotation annotation : annotations) {
		if (annotation != null) {
			Class<? extends Annotation> type = annotation.annotationType();
			if (type != null && !this.annotationFilter.matches(type)) {
				if (type == requiredType || type.getName().equals(requiredType)) {
					return Boolean.TRUE;
				}
				Annotation[] repeatedAnnotations =
						this.repeatableContainers.findRepeatedAnnotations(annotation);
				if (repeatedAnnotations != null) {
					Boolean result = doWithAnnotations(
							requiredType, aggregateIndex, source, repeatedAnnotations);
					if (result != null) {
						return result;
					}
				}
				if (!this.directOnly) {
                    // 这里会进行递归解析
					AnnotationTypeMappings mappings = AnnotationTypeMappings.forAnnotationType(type);
					for (int i = 0; i < mappings.size(); i++) {
						AnnotationTypeMapping mapping = mappings.get(i);
						if (isMappingForType(mapping, this.annotationFilter, requiredType)) {
							return Boolean.TRUE;
						}
					}
				}
			}
		}
	}
	return null;
}

AnnotationTypeMappings.forAnnotationType方法：

static AnnotationTypeMappings forAnnotationType(Class<? extends Annotation> annotationType) {
	return forAnnotationType(annotationType, AnnotationFilter.PLAIN);
}

static AnnotationTypeMappings forAnnotationType(
		Class<? extends Annotation> annotationType, AnnotationFilter annotationFilter) {
	return forAnnotationType(annotationType, RepeatableContainers.standardRepeatables(), annotationFilter);
}

static AnnotationTypeMappings forAnnotationType(Class<? extends Annotation> annotationType,
		RepeatableContainers repeatableContainers, AnnotationFilter annotationFilter) {
	return new AnnotationTypeMappings(repeatableContainers, annotationFilter, annotationType);
}

AnnotationTypeMappings对象：

private AnnotationTypeMappings(RepeatableContainers repeatableContainers,
		AnnotationFilter filter, Class<? extends Annotation> annotationType) {
	this.repeatableContainers = repeatableContainers;
	this.filter = filter;
	this.mappings = new ArrayList<>();
	addAllMappings(annotationType);
	this.mappings.forEach(AnnotationTypeMapping::afterAllMappingsSet);
}

private void addAllMappings(Class<? extends Annotation> annotationType) {
	Deque<AnnotationTypeMapping> queue = new ArrayDeque<>();
	addIfPossible(queue, null, annotationType, null);
    // 递归解析
	while (!queue.isEmpty()) {
		AnnotationTypeMapping mapping = queue.removeFirst();
		this.mappings.add(mapping);
		addMetaAnnotationsToQueue(queue, mapping);
	}
}

private void addMetaAnnotationsToQueue(Deque<AnnotationTypeMapping> queue, AnnotationTypeMapping source) {
	Annotation[] metaAnnotations = 
        AnnotationsScanner.getDeclaredAnnotations(source.getAnnotationType(), false);
	// ...
}

static Annotation[] getDeclaredAnnotations(AnnotatedElement source, boolean defensive) {
	boolean cached = false;
	Annotation[] annotations = declaredAnnotationCache.get(source);
	if (annotations != null) {
		cached = true;
	} else {
        // 这里使用java反射api获取类标注的注解
		annotations = source.getDeclaredAnnotations();
		if (annotations.length != 0) {
			// ...
		}
	}
	if (!defensive || annotations.length == 0 || !cached) {
		return annotations;
	}
	return annotations.clone();
}

OnPropertyCondition类

class OnPropertyCondition extends SpringBootCondition {
	@Override
	public ConditionOutcome getMatchOutcome(ConditionContext context, AnnotatedTypeMetadata metadata) {
        // 获取所有的@ConditionalOnProperty注解属性
		List<AnnotationAttributes> allAnnotationAttributes = annotationAttributesFromMultiValueMap(
				metadata.getAllAnnotationAttributes(ConditionalOnProperty.class.getName()));
		List<ConditionMessage> noMatch = new ArrayList<>();
		List<ConditionMessage> match = new ArrayList<>();
        // 遍历@ConditionalOnProperty注解属性
		for (AnnotationAttributes annotationAttributes : allAnnotationAttributes) {
			ConditionOutcome outcome = determineOutcome(annotationAttributes, context.getEnvironment());
			(outcome.isMatch() ? match : noMatch).add(outcome.getConditionMessage());
		}
		if (!noMatch.isEmpty()) {
			return ConditionOutcome.noMatch(ConditionMessage.of(noMatch));
		}
		return ConditionOutcome.match(ConditionMessage.of(match));
	}

	private ConditionOutcome determineOutcome(AnnotationAttributes annotationAttributes,
                                              PropertyResolver resolver) {
		Spec spec = new Spec(annotationAttributes);
		List<String> missingProperties = new ArrayList<>();
		List<String> nonMatchingProperties = new ArrayList<>();
        // 这里在匹配env参数
        // Spec类封装了prefix, havingValue, names, matchIfMissing等配置
        // collectProperties方法将env参数与prefix, havingValue, names, matchIfMissing等进行匹配
		spec.collectProperties(resolver, missingProperties, nonMatchingProperties);
		if (!missingProperties.isEmpty()) {
			return ConditionOutcome.noMatch(...);
		}
		if (!nonMatchingProperties.isEmpty()) {
			return ConditionOutcome.noMatch(...);
		}
		return ConditionOutcome
				.match(ConditionMessage.forCondition(ConditionalOnProperty.class, spec).because("matched"));
	}
}

public abstract class SpringBootCondition implements Condition {

	@Override
	public final boolean matches(ConditionContext context, AnnotatedTypeMetadata metadata) {
		String classOrMethodName = getClassOrMethodName(metadata);
		try {
            // 此处需要子类实现
			ConditionOutcome outcome = getMatchOutcome(context, metadata);
			logOutcome(classOrMethodName, outcome);
			recordEvaluation(context, classOrMethodName, outcome);
			return outcome.isMatch();
		} catch (NoClassDefFoundError ex) {
			throw new IllegalStateException("", ex);
		} catch (RuntimeException ex) {
			throw new IllegalStateException("", ex);
		}
	}

	public abstract ConditionOutcome getMatchOutcome(
        ConditionContext context, AnnotatedTypeMetadata metadata);
}

(Spec)spec.collectProperties方法：

private void collectProperties(PropertyResolver resolver, List<String> missing, List<String> nonMatching) {
	for (String name : this.names) {
		String key = this.prefix + name;
		if (resolver.containsProperty(key)) {
			if (!isMatch(resolver.getProperty(key), this.havingValue)) {
				nonMatching.add(name);
			}
		} else {
			if (!this.matchIfMissing) {
				missing.add(name);
			}
		}
	}
}