Apache BVal (源码)是实体数据验证 Java Bean Validation 的参考实现。Apache BVal 提供了 JSR 303 规范中所有内置 constraint 的实现,用于对 Bean 中的字段的值进行约束定义、描述和验证。若单单说 JSR 规范大渣可能还不清楚,但做过 POJO 的 Hibernate Validator 注解的朋友就知道是啥,——那为什么不使用主流的 Hibernate Validator 呢?因为这货净是个压缩包都已经 13mb 了(尽管可以有文档、源码其他在内),BVal 才只有 400 多 kb,而我只需要服务端验证而已,——天真的孩纸伤不起啊。俺的 ORM 也是 Mybatis 的,务求尽可能地轻量级。
Spring MVC 3.x 虽然自带了验证器 Validatior,可以在控制器中对表单提交的数据进行验证,但这个验证器是极其弱,因为你需要完全手工编码 if (null) else warn("不能空数据"),——太变态了(入下面的例子)——我们需要框架来减轻费时耗力的劳动,于是这类验证框架出现了。
import org.springframework.validation.Errors; import org.springframework.validation.ValidationUtils; import org.springframework.validation.Validator; public class UserValidator implements Validator { public boolean supports(Class<?> clazz) { // TODO Auto-generated method stub return User.class.equals(clazz); } public void validate(Object obj, Errors errors) { // TODO Auto-generated method stub ValidationUtils.rejectIfEmpty(errors, "username", null, "Username is empty."); User user = (User) obj; if (null == user.getPassword() || "".equals(user.getPassword())) errors.rejectValue("password", null, "Password is empty."); } }
而我们理想的是这样的,在 POJO 身上声明验证条件的注解(Fields or Method 均可):
import javax.validation.constraints.Min; import javax.validation.constraints.NotNull; public class User {@NotNull(message="用户名不能为空")private String username; private String password; private int age; public String getUsername() { return username; } public void setUsername(String username) { this.username = username; } @NotNull(message="密码不能为null") public String getPassword() { return password; } public void setPassword(String password) { this.password = password; } @Min(value=10, message="年龄的最小值为10") public int getAge() { return age; } public void setAge(int age) { this.age = age; } }
好了,现在准备一下 BVal jar 包:
加入测试用例:
import static org.junit.Assert.assertNotNull; import java.util.Set; import javax.validation.ConstraintViolation; import javax.validation.Validation; import javax.validation.Validator; import javax.validation.ValidatorFactory; import org.apache.bval.jsr.ApacheValidationProvider; import org.junit.Test; import com.ajaxjs.business.model.News; public class TestValidation { @Test public void testValid() { News news = new News(); ValidatorFactory avf = Validation.byProvider(ApacheValidationProvider.class).configure().buildValidatorFactory(); Validator validator = avf.getValidator(); //news.setName("dsdsa"); Set<ConstraintViolation<News>> constraintViolations = validator.validate(news); for (ConstraintViolation<News> constraintViolation : constraintViolations) { System.out.println(constraintViolation.getMessage()); } assertNotNull(constraintViolations); } }
注解说明
@Vaild 是用于嵌套另外一个 pojo 的,如:
@NotNull @Valid private Person driver;那么,
Car car = new Car(null, "", 1); Person driver = new Person(); car.setDriver(driver);
Annotation | Supported data types | Use | Hibernate metadata impact |
---|---|---|---|
@AssertFalse |
Boolean , boolean |
Checks that the annotated element is false |
None |
@AssertTrue |
Boolean , boolean |
Checks that the annotated element is true |
None |
@DecimalMax(value=, inclusive=) |
BigDecimal , BigInteger , CharSequence , byte , short , int , long and the respective wrappers of the primitive types; Additionally supported by HV: any sub-type of Number |
Checks whether the annotated value is less than the specified maximum, when inclusive=false . Otherwise whether the value is less than or equal to the specified maximum. The parameter value is the string representation of the max value according to the BigDecimal string representation. |
None |
@DecimalMin(value=, inclusive=) |
BigDecimal , BigInteger , CharSequence , byte , short , int , long and the respective wrappers of the primitive types; Additionally supported by HV: any sub-type of Number |
Checks whether the annotated value is larger than the specified minimum, when inclusive=false . Otherwise whether the value is larger than or equal to the specified minimum. The parameter value is the string representation of the min value according to the BigDecimal string representation. |
None |
@Digits(integer=, fraction=) |
BigDecimal , BigInteger , CharSequence , byte , short , int , long and the respective wrappers of the primitive types; Additionally supported by HV: any sub-type of Number |
Checks whether the annoted value is a number having up to integer digits and fraction fractional digits |
Defines column precision and scale |
@Future |
java.util.Date , supported by HV, if the Joda Time date/time API is on the class path: any implementations of ReadablePartial and ReadableInstant |
Checks whether the annotated date is in the future | None |
@Max(value=) |
BigDecimal , BigInteger , byte , short , int , long and the respective wrappers of the primitive types; Additionally supported by HV: any sub-type of CharSequence (the numeric value represented by the character sequence is evaluated), any sub-type of Number |
Checks whether the annotated value is less than or equal to the specified maximum | Adds a check constraint on the column |
@Min(value=) |
BigDecimal , BigInteger , byte , short , int , long and the respective wrappers of the primitive types; Additionally supported by HV: any sub-type of CharSequence (the numeric value represented by the char sequence is evaluated), any sub-type of Number |
Checks whether the annotated value is higher than or equal to the specified minimum | Adds a check constraint on the column |
@NotNull |
Any type | Checks that the annotated value is not null. |
Column(s) are not nullable |
@Null |
Any type | Checks that the annotated value is null |
None |
@Past |
java.util.Date , java.util.Calendar ; Additionally date/time API is on the class path: any implementations of ReadablePartial and ReadableInstant |
Checks whether the annotated date is in the past | None |
@Pattern(regex=, flag=) |
CharSequence |
Checks if the annotated string matches the regular expression regex considering the given flag match |
None |
@Size(min=, max=) |
CharSequence , Collection , Map and arrays |
Checks if the annotated element's size is between min and max (inclusive) | Column length will be set to max |
@Valid |
Any non-primitive type | Performs validation recursively on the associated object. If the object is a collection or an array, the elements are validated recursively. If the object is a map, the value elements are validated recursively. |
翻译是:
Constraint | 详细信息 |
---|---|
@Null |
被注释的元素必须为 null |
@NotNull |
被注释的元素必须不为 null |
@AssertTrue |
被注释的元素必须为 true |
@AssertFalse |
被注释的元素必须为 false |
@Min(value) |
被注释的元素必须是一个数字,其值必须大于等于指定的最小值 |
@Max(value) |
被注释的元素必须是一个数字,其值必须小于等于指定的最大值 |
@DecimalMin(value) |
被注释的元素必须是一个数字,其值必须大于等于指定的最小值 |
@DecimalMax(value) |
被注释的元素必须是一个数字,其值必须小于等于指定的最大值 |
@Size(max, min) |
被注释的元素的大小必须在指定的范围内 |
@Digits (integer, fraction) |
被注释的元素必须是一个数字,其值必须在可接受的范围内 |
@Past |
被注释的元素必须是一个过去的日期 |
@Future |
被注释的元素必须是一个将来的日期 |
@Pattern(value) |
被注释的元素必须符合指定的正则表达式 |
Apache BVal 提供额外的注解,在 org.apache.bval.constraints.* 下。
注入到 Spring
如果用 Apache BVal 结合 Spring 是怎么做的呢?首先在 MVC 的 xml 配置文件中加入以下:
<mvc:annotation-driven validator="validator"/> <!-- 数据验证 Validator bean --> <bean id="validator" class="org.springframework.validation.beanvalidation.LocalValidatorFactoryBean"> <property name="providerClass" value="org.apache.bval.jsr.ApacheValidationProvider" /> </bean> <!-- // -->
这是一种全局的配置方式,我们注入 Bval 的验证器,另外,还要让 MVC 打开注解驱动。加上了 <mvn:annotation-driven/> 之后 Spring 会自动检测 classpath 下的 JSR-303 提供者并自动启用对 JSR-303 的支持。然后就可以在 POJO 中添加注解,而且要在控制器中声明 bean 的验证,如下例的 @Valid T news,否则 Bval 不会起作用。
/** * 新建 * @return */ @RequestMapping(method = RequestMethod.POST) public String create(@Valid T news, BindingResult result,Model model) { System.out.println("新建"); if (result.hasErrors()) { LOGGER.info("create error!"); }else{ LOGGER.info("create ok!"); } news.setService(getService()); try { getService().create(news); model.addAttribute("newlyId", news.getId()); } catch (ServiceException e) { model.addAttribute("errMsg", e.toString()); } return "common/entity/json_cud"; }
一定要注意的是,控制器方法的参数顺序。Binding Result 必须在 Bean 后面。这是 Spring MVC 的约定。MVC 对控制器其他参数的顺序没什么规定,唯独这个 BindingResult 作了如此规定。目的是为了可以允许有多个 bean,于是也就有多个 BindingResult。
怎么处理错误就不详细说了,不同场景下要求不同。
自定义验证
可否自定义验证条件?我还没试,应该可以参考 Hibernate Validator 的做法。下面这篇文章说得很详细。
《Springmvc validator 验证的使用》JSR 303 - Bean Validation 介绍及最佳实践
lk
调用 Apache BVal
07-18 补充如下:
可否脱离 Spring 运行?本身 BVal 就是一个独立的框架,因此答案不仅是可以的,而且非常简单。下面使用工厂和单例两种模式,返回一个校验器。
import javax.validation.Validation; import javax.validation.Validator; import javax.validation.ValidatorFactory; import org.apache.bval.jsr.ApacheValidationProvider; public enum MyValidatorFactory { SINGLE_INSTANCE { // BVal 与 JSR 接口结合,返回 ValidatorFactory 工厂 ValidatorFactory avf = Validation.byProvider(ApacheValidationProvider.class).configure().buildValidatorFactory(); @Override public Validator getValidator() { return avf.getValidator(); } }; /** * 返回一个校验器 * @return 校验器 */ public abstract Validator getValidator(); }
这里一点技巧就是通过 Java 枚举来实现单例。
好了,拿一个 Bean 来测试下,分别有通过和报错两种情况的测试。
import java.util.Set; import static org.junit.Assert.*; import org.junit.*; import javax.validation.ConstraintViolation; import javax.validation.Validator; import com.ajaxjs.framework.model.MyValidatorFactory; import com.ajaxjs.framework.model.News; public class TestValid { @Test public void testQuery() { Validator v = MyValidatorFactory.SINGLE_INSTANCE.getValidator(); News news = new News(); news.setName("标题"); Set<ConstraintViolation<News>> result = v.validate(news); assertTrue("应该通过校验", result.isEmpty()); News news2 = new News(); result = v.validate(news2); assertTrue("应该不通过校验", !result.isEmpty()); System.out.println(result.size()); for(ConstraintViolation<News> r : result) { System.out.println(r.getMessage());// 什么错? System.out.println(r.getPropertyPath());// 哪个字段错? } } }
主要通过返回 ConstraintViolation 的 isEmpty() 方法来判断是否有错误。