I have and endpoint, where I validate received json document which contains collections of objects. I would like to only log these objects which don't pass a validation, when others i would like to store in db. Controller should return 200 OK in that situation. I was trying to use BindingResult object for this purpose. Unfortunately i always get a ConstraintViolationException. It seems that it validates it before it enter the method and throw exception. How can I force it to use BindingResult object ?
#RestController
#Validated
#RequestMapping(path = "/test")
class TestController {
#PostMapping(consumes = APPLICATION_JSON_VALUE)
public ResponseEntity<Void> addObjects(#RequestBody #Valid List<Document> objects, BindingResult bindingResult) {
if(bindingResult.hasErrors()){
}
return new ResponseEntity<Void>(HttpStatus.OK);
}
}
I'm using Spring Boot 1.5.9.RELEASE with Java 8
I've managed to solve it finally. Problem is with #Validated annotation on controller class. With this annotation spring do a validation on request and throw ConstraintViolationException. Without that, validation is triggered later and it results are stored in BindingResult object as expected
Could You please add the model classes with its annotations?
Remember that if You have any fields in Document class which are Your custom defined classes and You want it to be validated also then You have to decorate these fields with #Valid annotation too.
I am working on a Spring REST application.
This application has only REST controllers, no view part.
I want to know how can I validate a #RequestParam
For example
#RequestMapping(value = "", params = "from", method = RequestMethod.GET)
public List<MealReadingDTO> getAllMealReadingsAfter(#RequestParam(name = "from", required = true) Date fromDate) {
......
......
}
In the above example, my goal is to validate the Date. Suppose someone pass an invalid value, then I should be able to handle that situation.
Now it is giving and exception with 500 status.
PS
My question is not just about Date validation.
Suppose, there is a boolean parameter and someone passes tru instead of true by mistake, I should be able to handle this situation as well.
Thanks in advance :)
Spring will fail with an 500 status code, because it cannot parse the value.
The stages of request handling are:
receive request
identify endpoint
parse request params / body values and bind them to the detected objects
validate values if #Validated is used
enter method call with proper parameters
In your case the flow fails at the parse (3) phase.
Most probably you receive a BindException.
You may handle these cases by providing an exception handler for your controller.
#ControllerAdvice
public class ControllerExceptionHandler {
#ExceptionHandler(BindException.class)
#ResponseStatus(HttpStatus.BAD_REQUEST)
#ResponseBody
public YourErrorObject handleBindException(BindException e) {
// the details which field binding went wrong are in the
// exception object.
return yourCustomErrorData;
}
}
Otherwise when parsing is not functioning as expected (especially a hussle with Dates), you may want to add your custom mappers / serializers.
Most probably you have to configure Jackson, as that package is responsible for serializing / deserializing values.
I'm currently developing a Spring Boot application that exposes endpoints using #RestController and #RequestMapping annotations.
I recently discovered the concept of the projections as defined in Spring Data Rest (#Projection-annotated interfaces and #RepositoryRestResource-annotated JPA repositories) and I'd like to apply this concept to my existing services.
As I understand this post Spring Boot Projection with RestController, (please correct me if I'm wrong), #RestController and #RepositoryRestResource classes both define endpoints. So these annotations seem quite incompatible.
Is there a Spring component which can simply apply the projections concept to the #RestController endpoints ?
Is there a way to manually reroute requests from a endpoint to another ? (for example, using #RestController endpoints as some sort of proxy which performs controls or other operations before rerouting the request to the #RepositoryRestResource endpoints)
EDIT: I add a glimpse of the final code I'd like to have in the end.
#RestController
public class MyController {
#RequestMapping(value = "/elements/{id}", method = RequestMethod.GET)
public ResponseEntity<Element> getElements(
#PathVariable("id") Integer elementId,
#RequestParam("projection") String projection,
#RequestHeader(value = "someHeader") String header{
// [manual controls on the header then call to a service which returns the result]
}
}
#Entity
public class Element {
private Integer id;
private String shortField;
private String longField;
private List<SubElement> subElements;
// [Getters & setters]
}
#Projection(name = "light", types = {Element.class})
interface ElementLight {
public Integer getId();
public String getShortField();
}
If I call /elements/4, I'd get the complete Element having id = 4.
If I call /elements/4?projection=light, I'd get only the id and the short field of the Element having id = 4.
This answer gives some detail about how to create projection instances of your entities - https://stackoverflow.com/a/29386907/5371736
So depending on your projection parameter you could generate the given projections.
Hope this is what you are looking for.
Let say we have an API endpoint configured using Spring MVC and Spring Security. We would like to be able to handle pairs of #RequestMapping and #Secured annotations where the only #Secured annotation values differ from pair to pair. This way, we would be able to return a different response body depending on security rules for the same request.
This may allow our code to be more maintainable by avoiding to check for security rules directly into the method body.
With a not working example, here is what we would like to do :
#Controller
#RequestMapping("/api")
public class Controller {
#Secured ({"ROLE_A"})
#RequestMapping(value="{uid}", method=RequestMethod.GET)
#ResponseBody
public Response getSomething(#PathVariable("uid") String uid) {
// Returns something for users having ROLE_A
}
#Secured ({"ROLE_B"})
#RequestMapping(value="{uid}", method=RequestMethod.GET)
#ResponseBody
public Response getSomethingDifferent(#PathVariable("uid") String uid) {
// Returns something different for users having ROLE_B
}
}
How can we achieve this ?
And if this can be done: How the priority should be managed for a user who has both ROLE_A and ROLE_B ?
Assuming you are using Spring 3.1 (or up) together with the RequestMappingHandlerMapping (and RequestMappingHandlerAdapter) you can extend the request mapping mechanism. You can do this by creating your own implementation of the RequestCondition interface and extend the RequestMappingHandlerMapping to construct this based on the #Secured annotation on your method.
You would need to override the 'getCustomMethodCondition' method on the RequestMappingHandlerMapping and based on the Method and the existence of the #Secured annotation construct your custom implementation of the RequestCondition. All that information is then taken into account when matching incoming requests to methods.
Related answers (although not specific for #Secured annotations but the mechanism is the same) is also to be found here or here
I don't think you can do this in spring-mvc, since both routes have exactly the same #RequestMapping (#Secured) is not taken into account by the route engine of spring-mvc. The easiest solution would be to do this:
#Secured ({"ROLE_A", "ROLE_B"})
#RequestMapping(value="{uid}", method=RequestMethod.GET)
#ResponseBody
public Response getSomething(#PathVariable("uid") String uid, Principal p) {
// Principal p gets injected by spring
// and you need to cast it to check access roles.
if (/* p.hasRole("ROLE_A") */) {
return "responseForA";
} else if (/* p.hasRole("ROLE_B") */) {
return "responseForB";
} else {
// This is not really needed since #Secured guarantees that you don't get other role.
return 403;
}
}
However, I would change your design, since the response is different per role, why not have 2 separate request mappings with slightly different URLs? If at some point you have users with role A and B at the same time, you can't let the user choose what response to get (think, for example, of the public and private profiles of LinkedIn)
In the following example, the ScriptFile parameter is marked with an #Valid annotation.
What does #Valid annotation do?
#RequestMapping(value = "/scriptfile", method = RequestMethod.POST)
public String create(#Valid ScriptFile scriptFile, BindingResult result, ModelMap modelMap) {
if (scriptFile == null) throw new IllegalArgumentException("A scriptFile is required");
if (result.hasErrors()) {
modelMap.addAttribute("scriptFile", scriptFile);
modelMap.addAttribute("showcases", ShowCase.findAllShowCases());
return "scriptfile/create";
}
scriptFile.persist();
return "redirect:/scriptfile/" + scriptFile.getId();
}
It's for validation purposes.
Validation It is common to validate a
model after binding user input to it.
Spring 3 provides support for
declarative validation with JSR-303.
This support is enabled automatically
if a JSR-303 provider, such as
Hibernate Validator, is present on
your classpath. When enabled, you can
trigger validation simply by
annotating a Controller method
parameter with the #Valid annotation:
After binding incoming POST
parameters, the AppointmentForm will
be validated; in this case, to verify
the date field value is not null and
occurs in the future.
Look here for more info:
http://blog.springsource.com/2009/11/17/spring-3-type-conversion-and-validation/
Adding to above answers, take a look at following. AppointmentForm's date column is annotated with couple of annotations. By having #Valid annotation that triggers validations on the AppointmentForm (in this case #NotNull and #Future). These annotations could come from different JSR-303 providers (e.g, Hibernate, Spring..etc).
#RequestMapping(value = "/appointments", method = RequestMethod.POST)
public String add(#Valid AppointmentForm form, BindingResult result) {
....
}
static class AppointmentForm {
#NotNull #Future
private Date date;
}
#Valid in itself has nothing to do with Spring. It's part of Bean Validation specification(there are several of them, the latest one being JSR 380 as of second half of 2017), but #Valid is very old and derives all the way from JSR 303.
As we all know, Spring is very good at providing integration with all different JSRs and java libraries in general(think of JPA, JTA, Caching, etc.) and of course those guys took care of validation as well. One of the key components that facilitates this is MethodValidationPostProcessor.
Trying to answer your question - #Valid is very handy for so called validation cascading when you want to validate a complex graph and not just a top-level elements of an object. Every time you want to go deeper, you have to use #Valid. That's what JSR dictates. Spring will comply with that with some minor deviations(for example I tried putting #Validated instead of #Valid on RestController method and validation works, but the same will not apply for a regular "service" beans).
I wanted to add more details about how the #Valid works, especially in spring.
Everything you'd want to know about validation in spring is explained clearly and in detail in https://reflectoring.io/bean-validation-with-spring-boot/, but I'll copy the answer to how #Valid works incase the link goes down.
The #Valid annotation can be added to variables in a rest controller method to validate them. There are 3 types of variables that can be validated:
the request body,
variables within the path (e.g. id in /foos/{id}) and,
query parameters.
So now... how does spring "validate"? You can define constraints to the fields of a class by annotating them with certain annotations. Then, you pass an object of that class into a Validator which checks if the constraints are satisfied.
For example, suppose I had controller method like this:
#RestController
class ValidateRequestBodyController {
#PostMapping("/validateBody")
ResponseEntity<String> validateBody(#Valid #RequestBody Input input) {
return ResponseEntity.ok("valid");
}
}
So this is a POST request which takes in a request body, and we're mapping that request body to a class Input.
Here's the class Input:
class Input {
#Min(1)
#Max(10)
private int numberBetweenOneAndTen;
#Pattern(regexp = "^[0-9]{1,3}\\.[0-9]{1,3}\\.[0-9]{1,3}\\.[0-9]{1,3}$")
private String ipAddress;
// ...
}
The #Valid annotation will tell spring to go and validate the data passed into the controller by checking to see that the integer numberBetweenOneAndTen is between 1 and 10 inclusive because of those min and max annotations. It'll also check to make sure the ip address passed in matches the regular expression in the annotation.
side note: the regular expression isn't perfect.. you could pass in 3 digit numbers that are greater than 255 and it would still match the regular expression.
Here's an example of validating a query variable and path variable:
#RestController
#Validated
class ValidateParametersController {
#GetMapping("/validatePathVariable/{id}")
ResponseEntity<String> validatePathVariable(
#PathVariable("id") #Min(5) int id) {
return ResponseEntity.ok("valid");
}
#GetMapping("/validateRequestParameter")
ResponseEntity<String> validateRequestParameter(
#RequestParam("param") #Min(5) int param) {
return ResponseEntity.ok("valid");
}
}
In this case, since the query variable and path variable are just integers instead of just complex classes, we put the constraint annotation #Min(5) right on the parameter instead of using #Valid.
IIRC #Valid isn't a Spring annotation but a JSR-303 annotation (which is the Bean Validation standard). What it does is it basically checks if the data that you send to the method is valid or not (it will validate the scriptFile for you).
public String create(#Valid #NotNull ScriptFile scriptFile, BindingResult result, ModelMap modelMap) {
if (scriptFile == null) throw new IllegalArgumentException("A scriptFile is required");
I guess this #NotNull annotation is valid therefore if condition is not needed.
I think I know where your question is headed. And since this question is the one that pop ups in google's search main results, I can give a plain answer on what the #Valid annotation does.
I'll present 3 scenarios on how I've used #Valid
Model:
public class Employee{
private String name;
#NotNull(message="cannot be null")
#Size(min=1, message="cannot be blank")
private String lastName;
//Getters and Setters for both fields.
//...
}
JSP:
...
<form:form action="processForm" modelAttribute="employee">
<form:input type="text" path="name"/>
<br>
<form:input type="text" path="lastName"/>
<form:errors path="lastName"/>
<input type="submit" value="Submit"/>
</form:form>
...
Controller for scenario 1:
#RequestMapping("processForm")
public String processFormData(#Valid #ModelAttribute("employee") Employee employee){
return "employee-confirmation-page";
}
In this scenario, after submitting your form with an empty lastName field, you'll get an error page since you're applying validation rules but you're not handling it whatsoever.
Example of said error:
Exception page
Controller for scenario 2:
#RequestMapping("processForm")
public String processFormData(#Valid #ModelAttribute("employee") Employee employee,
BindingResult bindingResult){
return bindingResult.hasErrors() ? "employee-form" : "employee-confirmation-page";
}
In this scenario, you're passing all the results from that validation to the bindingResult, so it's up to you to decide what to do with the validation results of that form.
Controller for scenario 3:
#RequestMapping("processForm")
public String processFormData(#Valid #ModelAttribute("employee") Employee employee){
return "employee-confirmation-page";
}
#ExceptionHandler(MethodArgumentNotValidException.class)
#ResponseStatus(HttpStatus.BAD_REQUEST)
public Map<String, String> invalidFormProcessor(MethodArgumentNotValidException ex){
//Your mapping of the errors...etc
}
In this scenario you're still not handling the errors like in the first scenario, but you pass that to another method that will take care of the exception that #Valid triggers when processing the form model. Check this see what to do with the mapping and all that.
To sum up: #Valid on its own with do nothing more that trigger the validation of validation JSR 303 annotated fields (#NotNull, #Email, #Size, etc...), you still need to specify a strategy of what to do with the results of said validation.
Hope I was able to clear something for people that might stumble with this.
Just adding to the above answer, In a web application
#valid is used where the bean to be validated is also annotated with validation annotations e.g. #NotNull, #Email(hibernate annotation) so when while getting input from user the values can be validated and binding result will have the validation results.
bindingResult.hasErrors() will tell if any validation failed.
Another handy aspect of #Valid not mentioned above is that (ie: using Postman to test an endpoint) #Valid will format the output of an incorrect REST call into formatted JSON instead of a blob of barely readable text. This is very useful if you are creating a commercially consumable API for your users.