一、引言

问题：在代码编写的过程中，数据值的校验在JavaEE三层架构（展示层、业务层、数据访问层）均有涉及，各层的校验需求又是不尽相同的，因此往往会伴随着代码冗余，重复的校验逻辑出现在三层代码中。

简介：JSR-303 Bean-Validator伴随着此类问题应运而生，Hibernate-Validator就是一个遵循JSR-303规范的优秀实践。

图片截取自Hibernate Validator 官方pdf文档。

Hibernate-Validator提供了较为完善、便捷的校验方式解决了代码校验问题。常用于提供字段的校验，比如字段非空、字段长度限制、邮箱验证等等。

二、测试实例

下面引用Hibernate-Validator源码中ValidatorTest测试类下的方法为例：

class A {@NotNullString b;@ValidC c;@ValidD d;
}
@GroupSequence({ TestGroup.class, C.class })
class C {@Pattern(regexp = "[0-9]+", groups = TestGroup.class)@Length(min = 2)String id;C(String id) {this.id = id;}
}
@Test
public void testValidate() {ValidatorFactory factory = configuration.buildValidatorFactory(); Validator validator = factory.getValidator();A testInstance = new A();testInstance.c = new C( "aaa" );Set<ConstraintViolation<SerializableClass>> constraintViolations = validator.validate( testInstance );Set<ConstraintViolation<A>> constraintViolations = validator.validateProperty( testInstance, "c.id" );Set<ConstraintViolation<A>> constraintViolations = validator.validateValue( A.class, "c.id", "aaa" );
}

由上例可以看出，Hibernate-Validator主要是将JSR-303内置的约束以及Hibernate-Validator扩展的自定义约束通过注解的形式添加到属性上，来实现对对象内属性的校验功能。下面我们将通过上述例子拆分及解析校验实现的流程及细节，本文使用Hibernate-Validator下的7.0.1.Final版本解释源码。

三. Hibernate-Validator具体实现

1. 获取Validator实例

想要对对象进行校验，首先需要获得一个Validator校验器实例，由上述代码可知

ValidatorFactory factory = configuration.buildValidatorFactory();
Validator validator = factory.getValidator();

校验器获取方法在代码实现上较为简单，主要是遵循Java内部的SPI机制实现Java提供的相关接口实现。

主要实现分为两步：

a. validatorFactory 工厂创建

public static ValidatorFactory buildDefaultValidatorFactory() {return byDefaultProvider().configure().buildValidatorFactory();
}// 构建相关配置文件
public Configuration<?> configure() {// ValidationProviderResolver 的主要作用是确定环境下可用的 ValidationProvider 列表// 其唯一实现是 jakarta.validation.Validation 下的内部类 DefaultValidationProviderResolver ValidationProviderResolver resolver = this.resolver == null ? this.getDefaultValidationProviderResolver() : this.resolver;List validationProviders;try {// ValidationProvider的作用是提供程序校验器，通过其下的buildValidatorFactory方法构建ValidatorFactory工厂// Hibernate-Validator对其提供了唯一的实现类HibernateValidator构建并返回工厂实现ValidatorFactoryImplvalidationProviders = resolver.getValidationProviders();} catch (ValidationException var6) {throw var6;} catch (RuntimeException var7) {throw new ValidationException("Unable to get available provider resolvers.", var7);}if (validationProviders.isEmpty()) {String msg = "Unable to create a Configuration, because no Jakarta Bean Validation provider could be found. Add a provider like Hibernate Validator (RI) to your classpath.";throw new NoProviderFoundException(msg);} else {try {// 从 ValidationProvider 中获取相关通用配置Configuration<?> config = ((ValidationProvider)resolver.getValidationProviders().get(0)).createGenericConfiguration(this);return config;} catch (RuntimeException var5) {throw new ValidationException("Unable to instantiate Configuration.", var5);}}
}// 根据配置文件构建用于创建Validator的工厂并初始化相关参数
public final ValidatorFactory buildValidatorFactory() {loadValueExtractorsFromServiceLoader();// 解析Validation.xml，如何存在则解析其配置parseValidationXml();for ( ValueExtractorDescriptor valueExtractorDescriptor : valueExtractorDescriptors.values() ) {validationBootstrapParameters.addValueExtractorDescriptor( valueExtractorDescriptor );}ValidatorFactory factory = null;try {// 判断是否指定了Validator，指定了Validator则直接返回Factoryif ( isSpecificProvider() ) {factory = validationBootstrapParameters.getProvider().buildValidatorFactory( this );}else {// 判断是否指定了 ValidatorProviderfinal Class<? extends ValidationProvider<?>> providerClass = validationBootstrapParameters.getProviderClass();if ( providerClass != null ) {for ( ValidationProvider<?> provider : providerResolver.getValidationProviders() ) {if ( providerClass.isAssignableFrom( provider.getClass() ) ) {factory = provider.buildValidatorFactory( this );break;}}if ( factory == null ) {throw LOG.getUnableToFindProviderException( providerClass );}}else {List<ValidationProvider<?>> providers = providerResolver.getValidationProviders();assert providers.size() != 0; // I run therefore I am// 构建 ValidatorFactory 时初始化了很多对象，包括方法校验配置、分组排序、注解到注解校验实现的映射等factory = providers.get( 0 ).buildValidatorFactory( this );}}}finally {// close all input streams opened by this configurationfor ( InputStream in : configurationStreams ) {try {in.close();}catch (IOException io) {LOG.unableToCloseInputStream();}}}return factory;
}

b. 获取Validator实例

@Override
// getValidator 是 ValidatorFactoryImpl 下最重要的方法，用于获取 Validator 实例
public Validator getValidator() {return createValidator(constraintCreationContext.getConstraintValidatorManager().getDefaultConstraintValidatorFactory(),constraintCreationContext,validatorFactoryScopedContext,methodValidationConfiguration);
}

2. Validator校验

由实例中列举的validator校验方法可知，Validator校验分为三种类型，分别是：

validate：针对所有属性进行校验；
validateProperty：针对某一个具体属性进行校验；
validateValue：针对某一个具体属性及特定值进行校验。

本文选用validate的相关校验流程及源码进行解析，源码功能使用代码注释的方式解释。

validate具体校验流程如下所示：

执行ValidatorImpl下的validate方法，封装validationContext（校验器上下文）、valueContext（校验对象上下文）及validationOrder（校验群组及排序信息）数据内容，方便后续调用。

// 校验对象下的所有约束是否正常
@Override
public final <T> Set<ConstraintViolation<T>> validate(T object, Class<?>... groups) {// 校验对象是否为空Contracts.assertNotNull(object, MESSAGES.validatedObjectMustNotBeNull());// 校验Class Group是否为空sanityCheckGroups(groups);@SuppressWarnings("unchecked")// 获取对象的ClassClass<T> rootBeanClass = (Class<T>) object.getClass();// 通过反射获取对象内所有Field、Method、Class相关的元数据信息BeanMetaData<T> rootBeanMetaData = beanMetaDataManager.getBeanMetaData(rootBeanClass);// 判断是否含有约束，无约束则直接返回if (!rootBeanMetaData.hasConstraints()) {return Collections.emptySet();}// 将对象元数据和实际对象内容封装成一个完成的ValidationContextBaseBeanValidationContext<T> validationContext = getValidationContextBuilder().forValidate(rootBeanClass,rootBeanMetaData, object);// 接口分组排序，分组排序决定了哪个Group先校验，哪个Group后校验，用于处理有先后关系的校验ValidationOrder validationOrder = determineGroupValidationOrder(groups);// 填充校验对象上下文信息BeanValueContext<?, Object> valueContext = ValueContexts.getLocalExecutionContextForBean(validatorScopedContext.getParameterNameProvider(), object, validationContext.getRootBeanMetaData(),PathImpl.createRootPath());// 根据校验文件必要信息进行校验操作return validateInContext(validationContext, valueContext, validationOrder);
}

执行ValidatorImpl下的validateInContext方法，根据传入对象的群组、级联信息分别进行校验

private <T, U> Set<ConstraintViolation<T>> validateInContext(BaseBeanValidationContext<T> validationContext,BeanValueContext<U, Object> valueContext, ValidationOrder validationOrder) {// 如果Bean为null则直接返回空if (valueContext.getCurrentBean() == null) {return Collections.emptySet();}// 取出当前Bean元数据BeanMetaData<U> beanMetaData = valueContext.getCurrentBeanMetaData();if (beanMetaData.isDefaultGroupSequenceRedefined()) {validationOrder.assertDefaultGroupSequenceIsExpandable(beanMetaData.getDefaultGroupSequence(valueContext.getCurrentBean()));}// 校验当前对象内的约束内容Iterator<Group> groupIterator = validationOrder.getGroupIterator();while (groupIterator.hasNext()) {Group group = groupIterator.next();valueContext.setCurrentGroup(group.getDefiningClass());// 校验当前对象validateConstraintsForCurrentGroup(validationContext, valueContext);// 判断约束校验是否失败if (shouldFailFast(validationContext)) {return validationContext.getFailingConstraints();}}// 级联校验，即校验一个对象内可能存在的需要校验的其他对象groupIterator = validationOrder.getGroupIterator();while (groupIterator.hasNext()) {Group group = groupIterator.next();valueContext.setCurrentGroup(group.getDefiningClass());// 校验级联对象validateCascadedConstraints(validationContext, valueContext);// 判断约束校验是否失败if (shouldFailFast(validationContext)) {return validationContext.getFailingConstraints();}}// 如果需要校验的对象包含GroupSequece，按次序校验则走以下方法Iterator<Sequence> sequenceIterator = validationOrder.getSequenceIterator();while (sequenceIterator.hasNext()) {Sequence sequence = sequenceIterator.next();for (GroupWithInheritance groupOfGroups : sequence) {int numberOfViolations = validationContext.getFailingConstraints().size();for (Group group : groupOfGroups) {valueContext.setCurrentGroup(group.getDefiningClass());// 校验对象validateConstraintsForCurrentGroup(validationContext, valueContext);// 判断约束校验是否失败if (shouldFailFast(validationContext)) {return validationContext.getFailingConstraints();}// 校验级联对象validateCascadedConstraints(validationContext, valueContext);// 判断约束校验是否失败if (shouldFailFast(validationContext)) {return validationContext.getFailingConstraints();}}if (validationContext.getFailingConstraints().size() > numberOfViolations) {break;}}}return validationContext.getFailingConstraints();
}

根据校验属性的Group执行相应校验方法

private void validateConstraintsForCurrentGroup(BaseBeanValidationContext<?> validationContext,BeanValueContext<?, Object> valueContext) {if (!valueContext.validatingDefault()) {// 如果校验的group非Default Group，走该校验方法validateConstraintsForNonDefaultGroup(validationContext, valueContext);}else {// 如果校验的group为Default Group，走该校验方法validateConstraintsForDefaultGroup(validationContext, valueContext);}
}

解析当前对象的继承结构，根据继承结构和默认的GroupSequence进行校验

// 默认约束校验
private <U> void validateConstraintsForDefaultGroup(BaseBeanValidationContext<?> validationContext,BeanValueContext<U, Object> valueContext) {// 获取currentBeanMetaData对象校验当前对象内容final BeanMetaData<U> beanMetaData = valueContext.getCurrentBeanMetaData();final Map<Class<?>, Class<?>> validatedInterfaces = new HashMap<>();// 评估类的继承结构中每个类的约束，依次校验该类及其父类约束for (Class<? super U> clazz : beanMetaData.getClassHierarchy()) {BeanMetaData<? super U> hostingBeanMetaData = beanMetaDataManager.getBeanMetaData(clazz);boolean defaultGroupSequenceIsRedefined = hostingBeanMetaData.isDefaultGroupSequenceRedefined();// 如果默认的GroupSequence被重新定义，说明需要对属性校验排序处理，则走以下方法进行校验if (defaultGroupSequenceIsRedefined) {Iterator<Sequence> defaultGroupSequence = hostingBeanMetaData.getDefaultValidationSequence(valueContext.getCurrentBean());Set<MetaConstraint<?>> metaConstraints = hostingBeanMetaData.getMetaConstraints();while (defaultGroupSequence.hasNext()) {// 区分GroupSequence进行校验for (GroupWithInheritance groupOfGroups : defaultGroupSequence.next()) {boolean validationSuccessful = true;for (Group defaultSequenceMember : groupOfGroups) {// 遍历GroupSequence下的Group进行校验validationSuccessful = validateConstraintsForSingleDefaultGroupElement(validationContext,valueContext, validatedInterfaces, clazz, metaConstraints, defaultSequenceMember)&& validationSuccessful;}validationContext.markCurrentBeanAsProcessed(valueContext);if (!validationSuccessful) {break;}}}}// fast path in case the default group sequence hasn't been redefinedelse {// 查询当前Bean下可以直接进行校验的属性Set<MetaConstraint<?>> metaConstraints = hostingBeanMetaData.getDirectMetaConstraints();// 执行校验validateConstraintsForSingleDefaultGroupElement(validationContext, valueContext, validatedInterfaces,clazz, metaConstraints, Group.DEFAULT_GROUP);validationContext.markCurrentBeanAsProcessed(valueContext);}// 层级中的所有约束均完成校验，则停止校验if (defaultGroupSequenceIsRedefined) {break;}}
}

循环校验对象的每个属性

private <U> boolean validateConstraintsForSingleDefaultGroupElement(BaseBeanValidationContext<?> validationContext,ValueContext<U, Object> valueContext, final Map<Class<?>, Class<?>> validatedInterfaces, Class<? super U> clazz,Set<MetaConstraint<?>> metaConstraints, Group defaultSequenceMember) {boolean validationSuccessful = true;valueContext.setCurrentGroup(defaultSequenceMember.getDefiningClass());// 循环校验对象的每个属性for (MetaConstraint<?> metaConstraint : metaConstraints) {// HV-466, an interface implemented more than one time in the hierarchy has to be validated only one// time. An interface can define more than one constraint, we have to check the class we are validating.final Class<?> declaringClass = metaConstraint.getLocation().getDeclaringClass();if (declaringClass.isInterface()) {Class<?> validatedForClass = validatedInterfaces.get(declaringClass);if (validatedForClass != null && !validatedForClass.equals(clazz)) {continue;}validatedInterfaces.put(declaringClass, clazz);}// 校验类非interface接口则正常校验boolean tmp = validateMetaConstraint(validationContext, valueContext, valueContext.getCurrentBean(),metaConstraint);// 快速失败if (shouldFailFast(validationContext)) {return false;}validationSuccessful = validationSuccessful && tmp;}return validationSuccessful;
}

判断对象是否满足校验条件并对对象进行校验

private boolean validateMetaConstraint(BaseBeanValidationContext<?> validationContext,ValueContext<?, Object> valueContext, Object parent, MetaConstraint<?> metaConstraint) {BeanValueContext.ValueState<Object> originalValueState = valueContext.getCurrentValueState();valueContext.appendNode(metaConstraint.getLocation());boolean success = true;// 判断待校验文件是否满足校验条件，如对象属性分组数及校验数是否匹配，校验分组是否与当前分组匹配等if (isValidationRequired(validationContext, valueContext, metaConstraint)) {if (parent != null) {valueContext.setCurrentValidatedValue(valueContext.getValue(parent, metaConstraint.getLocation()));}// 对象属性校验success = metaConstraint.validateConstraint(validationContext, valueContext);validationContext.markConstraintProcessed(valueContext.getCurrentBean(), valueContext.getPropertyPath(),metaConstraint);}// 对象属性校验完成后重置校验数据valueContext.resetValueState(originalValueState);return success;
}

具体对象值提取校验逻辑

public boolean validateConstraint(ValidationContext<?> validationContext, ValueContext<?, Object> valueContext) {boolean success = true;// 提取集合内容器循环校验if (valueExtractionPath != null) {Object valueToValidate = valueContext.getCurrentValidatedValue();if (valueToValidate != null) {TypeParameterValueReceiver receiver = new TypeParameterValueReceiver(validationContext, valueContext,valueExtractionPath);ValueExtractorHelper.extractValues(valueExtractionPath.getValueExtractorDescriptor(), valueToValidate,receiver);success = receiver.isSuccess();}}// 常规校验流程else {success = doValidateConstraint(validationContext, valueContext);}return success;
}

失败异常处理

public final boolean validateConstraints(ValidationContext<?> validationContext, ValueContext<?, ?> valueContext) {List<ConstraintValidatorContextImpl> violatedConstraintValidatorContexts = new ArrayList<>(5);// 属性值执行校验validateConstraints(validationContext, valueContext, violatedConstraintValidatorContexts);// 如果violatedConstraintValidatorContexts集合不为空，则说明校验失败，校验失败属性处理if (!violatedConstraintValidatorContexts.isEmpty()) {for (ConstraintValidatorContextImpl constraintValidatorContext : violatedConstraintValidatorContexts) {for (ConstraintViolationCreationContext constraintViolationCreationContext : constraintValidatorContext.getConstraintViolationCreationContexts()) {validationContext.addConstraintFailure(valueContext, constraintViolationCreationContext,constraintValidatorContext.getConstraintDescriptor());}}return false;}return true;
}

获取属性校验对应校验器及构建约束校验器上下文

protected void validateConstraints(ValidationContext<?> validationContext, ValueContext<?, ?> valueContext,Collection<ConstraintValidatorContextImpl> violatedConstraintValidatorContexts) {if (LOG.isTraceEnabled()) {LOG.tracef("Validating value %s against constraint defined by %s.", valueContext.getCurrentValidatedValue(),descriptor);}// 获取对象相关的Validator校验器，注解与校验器的映射在ValidatorImpl初始化时就已经取出，存在校验上下文对象中。// 通过AbstractConstraintValidatorManagerImpl下的findMatchingValidatorDescriptor查询并匹配到对应属性。ConstraintValidator<B, ?> validator = getInitializedConstraintValidator(validationContext, valueContext);// 构建约束校验器上下文ConstraintValidatorContextImpl constraintValidatorContext = validationContext.createConstraintValidatorContextFor(descriptor, valueContext.getPropertyPath());// 校验，如果校验失败则加入集合violatedConstraintValidatorContextsif (validateSingleConstraint(valueContext, constraintValidatorContext, validator).isPresent()) {violatedConstraintValidatorContexts.add(constraintValidatorContext);}
}

通过筛选出的validator完成校验

protected final <V> Optional<ConstraintValidatorContextImpl> validateSingleConstraint(ValueContext<?, ?> valueContext, ConstraintValidatorContextImpl constraintValidatorContext,ConstraintValidator<A, V> validator) {boolean isValid;try {// 获取当前要校验的对象值value@SuppressWarnings("unchecked")V validatedValue = (V) valueContext.getCurrentValidatedValue();// 使用注解和属性类型对应的Validator进行校验isValid = validator.isValid(validatedValue, constraintValidatorContext);}catch (RuntimeException e) {if (e instanceof ConstraintDeclarationException) {throw e;}throw LOG.getExceptionDuringIsValidCallException(e);}if (!isValid) {// We do not add these violations yet, since we don't know how they are// going to influence the final boolean evaluationreturn Optional.of(constraintValidatorContext);}return Optional.empty();
}

上述代码即是从ValidatorImpl下的validator方法开始到实现注解属性校验的基础校验过程源码解析。
validateProperty和validateValue两种方法与validate方法差异不大，此处不进行过多赘述。

当校验对象中存在嵌套对象需要校验时，此时需要使用@Valid注解标注出需要校验的对象，通过级联校验功能进行校验。

private <T, U> Set<ConstraintViolation<T>> validateInContext(BaseBeanValidationContext<T> validationContext,BeanValueContext<U, Object> valueContext, ValidationOrder validationOrder) {// 级联校验，即校验一个对象内可能存在的需要校验的其他对象groupIterator = validationOrder.getGroupIterator();while (groupIterator.hasNext()) {Group group = groupIterator.next();valueContext.setCurrentGroup(group.getDefiningClass());// 校验级联对象validateCascadedConstraints(validationContext, valueContext);// 判断约束校验是否失败if (shouldFailFast(validationContext)) {return validationContext.getFailingConstraints();}}return validationContext.getFailingConstraints();
}
private void validateCascadedConstraints(BaseBeanValidationContext<?> validationContext,ValueContext<?, Object> valueContext) {// 获取当前对象下的可校验内容Validatable validatable = valueContext.getCurrentValidatable();BeanValueContext.ValueState<Object> originalValueState = valueContext.getCurrentValueState();// 校验当前对象下注释了@Valid的属性或方法参数for (Cascadable cascadable : validatable.getCascadables()) {valueContext.appendNode(cascadable);// 确认级联校验对象处于可校验状态if (isCascadeRequired(validationContext, valueContext.getCurrentBean(), valueContext.getPropertyPath(),cascadable.getConstraintLocationKind())) {Object value = getCascadableValue(validationContext, valueContext.getCurrentBean(), cascadable);CascadingMetaData cascadingMetaData = cascadable.getCascadingMetaData();if (value != null) {CascadingMetaData effectiveCascadingMetaData = cascadingMetaData.addRuntimeContainerSupport(valueExtractorManager, value.getClass());// 校验被@Valid注释的级联属性，递归validateInContext方法实现递归级联校验if (effectiveCascadingMetaData.isCascading()) {validateCascadedAnnotatedObjectForCurrentGroup(value, validationContext, valueContext,effectiveCascadingMetaData);}// Container集合数据特殊处理if (effectiveCascadingMetaData.isContainer()) {ContainerCascadingMetaData containerCascadingMetaData = effectiveCascadingMetaData.as(ContainerCascadingMetaData.class);if (containerCascadingMetaData.hasContainerElementsMarkedForCascading()) {// 校验Container内元素约束validateCascadedContainerElementsForCurrentGroup(value, validationContext, valueContext,containerCascadingMetaData.getContainerElementTypesCascadingMetaData());}}}}// 重置valueContext值内容valueContext.resetValueState(originalValueState);}
}
private void validateCascadedAnnotatedObjectForCurrentGroup(Object value,BaseBeanValidationContext<?> validationContext, ValueContext<?, Object> valueContext,CascadingMetaData cascadingMetaData) {Class<?> originalGroup = valueContext.getCurrentGroup();Class<?> currentGroup = cascadingMetaData.convertGroup(originalGroup);// 判断数据是否已经校验过或者校验是否已经失败if (validationContext.isBeanAlreadyValidated(value, currentGroup, valueContext.getPropertyPath())|| shouldFailFast(validationContext)) {return;}ValidationOrder validationOrder = validationOrderGenerator.getValidationOrder(currentGroup,currentGroup != originalGroup);BeanValueContext<?, Object> cascadedValueContext = buildNewLocalExecutionContext(valueContext, value);// 递归校验validateInContext(validationContext, cascadedValueContext, validationOrder);
}

3. Validator功能拓展

Validator除基础的校验流程外，为了强化功能以满足更多场景的需要，校验流程中还提供了诸如Group（群组）、Sequence（优先级）等功能，下面将解读这些功能在校验中的相关类源码。

Group（群组）

a. 为类中的属性约束定义群组

@Data
public class Driver extends Person {@Min(value = 18, message = "You have to be 18 to drive a car", groups = DriverChecks.class)public int age;public Driver(String name) {super(name);}
}

b. 封装校验群组及校验顺序信息到ValidationOrder对象中

public ValidationOrder getValidationOrder(Collection<Class<?>> groups) {// 如果Group为空则抛出异常if ( groups == null || groups.size() == 0 ) {throw LOG.getAtLeastOneGroupHasToBeSpecifiedException();}// Group集合下仅包含默认Group则直接返回if ( groups.size() == 1 && groups.contains( Default.class ) ) {return ValidationOrder.DEFAULT_GROUP;}// 所有的group类均为接口类，如果存在非接口Group则抛出异常for ( Class<?> clazz : groups ) {if ( !clazz.isInterface() ) {throw LOG.getGroupHasToBeAnInterfaceException( clazz );}}// 根据集合groups填充validationOrder群组及排序信息DefaultValidationOrder validationOrder = new DefaultValidationOrder();for ( Class<?> clazz : groups ) {if ( Default.class.equals( clazz ) ) { validationOrder.insertGroup( Group.DEFAULT_GROUP );}else if ( isGroupSequence( clazz ) ) {insertSequence( clazz, clazz.getAnnotation( GroupSequence.class ).value(), true, validationOrder );}else {Group group = new Group( clazz );validationOrder.insertGroup( group );insertInheritedGroups( clazz, validationOrder );}}return validationOrder;
}

c. validateInContext方法通过遍历ValidationOrder对象中的分组信息和序列信息进行校验

// 校验当前对象内的约束内容
Iterator<Group> groupIterator = validationOrder.getGroupIterator();
while (groupIterator.hasNext()) {…………
}
// 如果需要校验的对象包含GroupSequece，按次序校验则走以下方法
Iterator<Sequence> sequenceIterator = validationOrder.getSequenceIterator();
while (sequenceIterator.hasNext()) {…………
}

d. 后续指定Group后直接执行循环校验，校验分组与指定分组相同的属性。

private boolean isValidationRequired(BaseBeanValidationContext<?> validationContext,ValueContext<?, ?> valueContext, MetaConstraint<?> metaConstraint) {if (!validationContext.appliesTo(metaConstraint)) {return false;}if (validationContext.hasMetaConstraintBeenProcessed(valueContext.getCurrentBean(),valueContext.getPropertyPath(), metaConstraint)) {return false;}// 分组校验，如果需要校验的属性分组与当前正在校验分组不同，则不进行校验if (!metaConstraint.getGroupList().contains(valueContext.getCurrentGroup())) {return false;}return isReachable(validationContext, valueContext.getCurrentBean(), valueContext.getPropertyPath(),metaConstraint.getConstraintLocationKind());
}

分组校验通过对Validator注解增加分组的方式，限定了校验需要生效的范围，在调用validator方法时指定本次校验的分组，即可指定哪些方法需要校验。

Sequence（序列）

用户可以通过GroupSequence的方式指定分组的优先级，高优先级的分组可以优先执行校验逻辑。

private void insertSequence(Class<?> sequenceClass, Class<?>[] sequenceElements, boolean cache, DefaultValidationOrder validationOrder) {Sequence sequence = cache ? resolvedSequences.get( sequenceClass ) : null;if ( sequence == null ) {// 解析校验对象的GroupSequence校验优先级数据sequence = resolveSequence( sequenceClass, sequenceElements, new ArrayList<Class<?>>() );// 如果存在分组继承关系，则检查分组继承关系新增Groupsequence.expandInheritedGroups();// 缓存校验优先级信息if ( cache ) {final Sequence cachedResolvedSequence = resolvedSequences.putIfAbsent( sequenceClass, sequence );if ( cachedResolvedSequence != null ) {sequence = cachedResolvedSequence;}}}validationOrder.insertSequence( sequence );
}

4. 自定义约束

除了Hibernate-Validator提供的约束外，用户在开发过程中经常需要基于自身业务需求自定义约束。

基于Hibernate-Validator框架拓展validator约束功能较为简单，仅需实现其自定义约束的两个部分，注解及其注解的约束实现。

注解是否为约束注解的解析实现在BeanMetaData对象的构造过程中，具体实现在org.hibernate.validator.internal.metadata.provider.AnnotationMetaDataProvider类下的findConstraintAnnotations方法中：

protected <A extends Annotation> List<ConstraintDescriptorImpl<?>> findConstraintAnnotations(Member member,A annotation,ElementType type) {if ( annotation.annotationType().getPackage().getName().equals( "jdk.internal" ) ) {return Collections.emptyList();}List<ConstraintDescriptorImpl<?>> constraintDescriptors = newArrayList();List<Annotation> constraints = newArrayList();Class<? extends Annotation> annotationType = annotation.annotationType();if ( constraintHelper.isConstraintAnnotation( annotationType ) ) {constraints.add( annotation );}else if ( constraintHelper.isMultiValueConstraint( annotationType ) ) {constraints.addAll( constraintHelper.getConstraintsFromMultiValueConstraint( annotation ) );}for ( Annotation constraint : constraints ) {final ConstraintDescriptorImpl<?> constraintDescriptor = buildConstraintDescriptor(member, constraint, type);constraintDescriptors.add( constraintDescriptor );}return constraintDescriptors;
}

保证注解满足constraintHelper.isConstraintAnnotation方法中的要求即可，包括其中必要参数及注解的校验。

public boolean isConstraintAnnotation(Class<? extends Annotation> annotationType) {// 查询是否是Hibernate-Validator自身约束，是则直接返回trueif (isBuiltinConstraint(annotationType)) {return true;}// 自定义约束检查注解及参数是否合法if (annotationType.getAnnotation(Constraint.class) == null) {return false;}return externalConstraints.computeIfAbsent(annotationType, a -> {assertMessageParameterExists(a);assertGroupsParameterExists(a);assertPayloadParameterExists(a);assertValidationAppliesToParameterSetUpCorrectly(a);assertNoParameterStartsWithValid(a);return Boolean.TRUE;});
}

约束的实现类则可直接在注解类中通过 @Constraint(validatedBy = {……}) 指定，本文在此不进行过多赘述。

5. 结语

Hibernate-Validator是一种较为优雅的数据校验实现方式，在日常的工作中已经得到了广泛的应用，希望本文能够帮助使用者更好更完善的从底层源码的角度了解Hibernate-Validator的功能，如果本文存在错误或遗漏请与作者联系，谢谢

Hibernate Validator源码解析相关推荐

springboot源码解析autoconfigure之WebMvcAutoConfiguration
2019独角兽企业重金招聘Python工程师标准>>> 说在前面本次开始spring-boot-autoconfigure源码解析之WebMvcAutoConfiguration ...
clickhouse原理解析与开发实战 pdf_重识SSM，“超高频面试点+源码解析+实战PDF”，一次性干掉全拿走...
重识SSM,"超高频面试点"+"源码解析"+"实战PDF",一次性干掉全拿走!! 01 超高频面试点知识篇 1.1 Spring超高频面试点 ...
php的lumen框架,Lumen框架“服务容器”源码解析
1.服务容器 "服务容器"是Lumen框架整个系统功能调度配置的核心,它提供了整个框架运行过程中的一系列服务."服务容器"就是提供服务(服务可以理解为系统运行中 ...
SpringDataJPA+Hibernate框架源码剖析（六）@PersistenceContext和@Autowired注入EntityManager的区别
SpringDataJPA+Hibernate框架源码剖析系列文章: SpringDataJPA+Hibernate框架源码剖析(一)框架介绍 SpringDataJPA+Hibernate框架源码剖 ...
Tomcat8源码解析
Tomcat8源码解析 Tomcat总体架构 Connector:开启Socket并监听客户端请求,返回响应数据: Container:负责具体的请求处理: 一个Service负责维护多个Connec ...
Spring源码深度解析(郝佳)-Spring 常用注解使用及源码解析
我们在看Spring Boot源码时,经常会看到一些配置类中使用了注解,本身配置类的逻辑就比较复杂了,再加上一些注解在里面,让我们阅读源码更加难解释了,因此,这篇博客主要对配置类上的一些注解的使用 ...
Spring源码深度解析(郝佳)-学习-源码解析-Spring MVC（三）-Controller 解析
在之前的博客中Spring源码深度解析(郝佳)-学习-源码解析-Spring MVC(一),己经对 Spring MVC 的框架做了详细的分析,但是有一个问题,发现举的例子不常用,因为我们在实际开发项 ...
01.Fabric源码解析---线头（王雅震）
Fabric源码解析1--线头 Getting Started 简单提一下Fabric说明文档中的Getting Started部分.说明文档下载地址在 http://hyperledger-fabr ...
谷歌BERT预训练源码解析（二）：模型构建
版权声明:本文为博主原创文章,遵循 CC 4.0 BY-SA 版权协议,转载请附上原文出处链接和本声明. 本文链接:https://blog.csdn.net/weixin_39470744/arti ...

Hibernate Validator源码解析