I'm using spring integration to store data in a mongo database. I'm using the java classes (MongoDbStoringMessageHandler), not the xml configuration and I can't find the way to get the results when adding some data in the database...
Is it possible ? How ?
The MongoDbStoringMessageHandler is a one-way component and it doesn't return anything.
Consider to use a MongoDbOutboundGateway instead with the CollectionCallback injected where you can perform an updateMany() and get UpdateResult as a reply from this gateway.
See more info in the Reference Manual: https://docs.spring.io/spring-integration/reference/html/mongodb.html#mongodb-outbound-gateway
UPDATE
but I don't know what parameter to put for the function to insert the Message payload.... Since there is no reference of the message in the ServiceActivator
Oh! I see. That a bug. We can't get access to the message from that context. Please, raise a JIRA on the matter: https://jira.spring.io/projects/INT/
Meanwhile as a workaround I suggest you ti write a custom POJO with injected MongoOperations and ther you can build any possible logic against a requestMessage.
The JIRA is here: https://jira.spring.io/browse/INT-4570
Related
Me using spring-kafka in spring-boot application to send the data topic.
I need to fetch the data from a oracle table and send it.
I fetch List from oracle table. How to send them to topic ?
i.e.
Is there any way to send them as a List ? if yes how ?
If yes, then how to deserialize it at consumer side ?
Is it possible to send data like a streaming fashion using spring-book and spring-kafka ? if yes any more info or sample/snippet plz ...
How to handle partitionKey if I send List at a time?
Currently I am sending individual Company obj hence have key defined as below
companyKafkaTemplate.send(COMPANY_TOPIC,this.getKey(company), company);
For a List serialization and deserialization I would suggest to use a JSON support in Spring Kafka: https://docs.spring.io/spring-kafka/docs/2.2.7.RELEASE/reference/html/#serdes
For a streaming I would suggest to take a look into a Reactive support in Spring Kafka, based on the Reactor Kafka project: https://github.com/reactor/reactor-kafka
For that purpose we provide a ReactiveKafkaProducerTemplate and ReactiveKafkaConsumerTemplate.
I have a Spring application with a JPA Repository. Now I'd like to add some validations. I found several solutions but none works perfect and I don't know which methode should be used in which case:
1.) Im using validation annotations (e.g. #NotNull) in my model object. But this generates a not usefull rest error response like ""Internal Server Error","message":"Could not commit JPA transaction; "
2.) I'm using the 'Validator' interface for custom validations but I get the spring "compiler" error "Validator has incorrect spelling"
3.) Some tutorials use 'ConstraintValidator' interface
4.) Some tutorials write custom rest methods for validations.
When should I use which and how can I solve my problems?
This is how you can manage it and easy is:
Define all your custom message in /messages/messages.properties under resources folder.
so if you error property is: error.user.name = User name can not be null.
then call your specific property in your pojo for that property.
#NotNull(message = "error.user.name")
Let say I have these configurations in my xml,
<int-sftp:outbound-channel-adapter id="sftpOutbound"
channel="sftpChannel"
auto-create-directory="true"
remote-directory="/path/to/remote/directory/"
session-factory="cachingSessionFactory">
<int-sftp:request-handler-advice-chain>
<int:retry-advice />
</int-sftp:request-handler-advice-chain>
</int-sftp:outbound-channel-adapter>
How can I retrieve the attributes i.e, remote-directory in Java class ?
I tried to use context.getBean('sftpOutbound') but it returns EventDrivenConsumer class which doesn't have methods to get the configurations.
I'm using spring-integration-sftp v 4.0.0.
I am actually more concerned with why you wan to access it. I mean the remote directory and other attributes will come with the headers of each message, so you will have access to it at the Message level, but not at the level of Event Driven Consumer and that is by design, hence my question.
I want to build an Apache camel application to download a Jira
issue report, parse it, and store it into a .csv file.
I'm new at Apache camel, I do believe the jira here should be an endpoint, how to setup this configuration, I want to set is as from:("Jira") to (csv file).
I believe it could be something like this:
public void configure() throws Exception {
from("jira://pullRequestComment?ProjectKey=CAMEL-0000&IssueTypeId=1&IssueSummary=title")
.process(new MyLogProcessor())
.bean(new MyTransformer(),"transformContent")
.to("file:D:/camel/output");
}
I tried the above code, I got an exception for java conversion type.
Exception:
The JIRA component returns Java objects from the JIRA REST API. You need to either:
Support passing in the object type to your processor class as a method argument
Convert the JIRA Java Object to something else, then pass into your processor
BTW- The JIRA component caches "seen" data to know what is "new" to pass into the route. For really busy JIRA servers, this looks and acts like a memory leak so you'll need to be mindful to manage that scenario
The pullRequestComment endpoint is for a producer endpoint (i.e. it can only go in to("jira:pullRequestComment?..."). Since you want to poll for new comments, you should use the newComment endpoint. So your route would look something like:
from("jira:newComment?serverUrl={host}&username={username}password={password}")
.process(new MyLogProcessor())
.bean(new MyTransformer(),"transformContent")
.to("file:D:/camel/output");
Note that this endpoint returns an object of type com.atlassian.jira.rest.client.domain.Comment, so in MyLogProcessor if you do exchange.getIn().getBody(), it will return an object of type Comment (or maybe a List if there are multiple objects, you'll have to test this).
If you want to post a pull request comment, then you can use the pullRequestComment endpoint like the following:
from("direct://some/uri/name")
.header("ProjectKey", "CAMEL-0000")
.header("IssueTypeId", 1L)
.header("IssueSummary", "title")
.to("jira:pullRequestComment?serverUrl={host}&username={username}password={password}")
.... // More processing here
Then if you invoke the route from("direct://some/uri/name"), it should post the comment that's in the exchange body.
Consider a Person entity with a property name that is annotated as #NotNull. Then a simple PersonRepository and this repo exposed with Spring Data Rest.
When I POST to create a new Person, if the name property is null a ValidationException occurs as expected. But what I actually get on the client is an Internal Server Error (500) and the message is a TransactionSystemException that happened much later in the exception chain.
What I'd expect to get is a Bad Request (400) with the actual ValidationException and all it's useful information so the client can know what's wrong with the posted data.
There seems to be a way to attach custom validators with SDR as explained here. But the thing is, this is not a custom validator, it's a standard bean validation that happens when the repository is asked to save data. So I'm not really sure how those two come together.
So questions:
What are my options to let the client know what's wrong with the submitted data when using SDR?. Things like what fields are invalid and what's the error for each field would be awesome.
Are there any examples about this anywhere?
Thanks a lot.
What you need is a proper ExceptionHandler, it will handle back end exceptions and send meaningful rich messages (json/xml) to the front end client.
Take a look a this git repository