I'm trying to get working deletion of one document in spring data elasticsearch repository. And can't find way how to solve this error:
[userindex] RoutingMissingException[routing is required for
[userindex]/[address]/[12]
I have two linked documents:
#Document(indexName = "userindex", type = "user")
public class User {
#Field(index = FieldIndex.not_analyzed, type = FieldType.Long)
private Long userId;
...
}
#Document(indexName = "userindex", type = "address")
public class Address {
#Field(type = FieldType.String)
private String name;
#Field(index = FieldIndex.not_analyzed, type = FieldType.String)
private String addressId;
#Field(type = FieldType.String, store = true)
#Parent(type = "user")
private String parentId;
...
}
When I'm trying to delete one address via ElasticsearchCrudRepository<Address, Long> by using standard method delete(Long id) I receiving RoutingMissingException mentioned above.
If I'm trying to do it using ElasticSeach client, like this:
client.prepareDelete().setIndex("userindex")
.setType("address")
.setParent("user")
.setId(id.toString())
.execute().get();
everything works fine, but seems to me working directly with client is not the spring-data way.
Also I can't find any way how to customize delete method with annotation org.springframework.data.elasticsearch.annotations.Query.
I checked sources of org.springframework.data.elasticsearch.core.ElasticsearchTemplate and can't find any way how to add support for delete query.
Anybody knows how to solve it instead of using a client?
The version of spring-data-elasticsearch is 2.0.1
Update 03.05.2017
First of all, in my code was an error with my deletion, don't how it worked before, but it should be:
client.prepareDelete().setIndex("userindex")
.setType("address")
.setParent("500")
.setId(id.toString())
.execute().get();
Here 500 is parent id instead of type name.
And now about the spring-data way. There is no spring-data way in elasticsearch integration.
Proof:
DATAES-257
DATAES-331
If you want to do it the Spring way, you can use ElasticsearchTemplate which is much similar to RestTemplate.
ElasticsearchTemplate has deleteIndex() function which can delete the whole index. Also, you can do lots of other stuff with the template.
Example project with delete index is here: https://github.com/TechPrimers/spring-data-elastic-example-4/blob/master/src/main/java/com/techprimers/elastic/resource/SearchResource.java
Code:
#Autowired
ElasticsearchTemplate template;
template.deleteIndex(Users.class);
Related
I am learning elasticsearch with spring data so can someone help me understand better what elasticsearch query is doing here. I am trying to return back only a set of results based off of a certain value, in this case env. It seems to me that this JPQL query, is not making a difference to only return what I ask for. I have also used an #Query with no difference.
-- here is part of my repository class
public interface MyFormRepo extends ElasticsearchRepository<MyForm, String> {
//??? these function calls are not effecting my return
#Query("{\"bool\": {\"must\": [{\"match\": {\"env\": \"?0\"}}]}}")
Page<MyForm> getAllByEnv(String env, Pageable pageable);
Page<MyForm> findAllByEnv(String env, Pageable pageable);
-- Here is part of my entity class
#Document(indexName = "my_form")
public class MyForm {
#Id
private String id;
#Field(type = Text)
private String schema;
#Field(type = Long)
private long version;
#Field(type = Text)
private String env;
Here is what I understand. Elasticsearch has this concept called Fuzziness (https://www.elastic.co/guide/en/elasticsearch/reference/current/common-options.html#fuzziness) so in it's searches which is based on Levenshtein distance (https://en.wikipedia.org/wiki/Levenshtein_distance). Spring data does not allow us to modify this out of the box and is considered Fuzziness.AUTO by default. https://docs.spring.io/spring-data/elasticsearch/docs/current/reference/html/#elasticsearch.misc. As for the queries neither of them will do anything different. Both findByAllEnv and getAllBeEnv has a fuziness.AUTO, as for the #Query I found a good reason stated at this site: What is difference between match and bool must match query in Elasticsearch. What I ended up finding is that I must implemented a custom repo for that I have found this example/explaination How to query Elastic with Spring-data-elastic.
I'm running into a problem with my Elasticsearch Document index creation failing on startup with "java.lang.IllegalArgumentException: can't add a _parent field that points to an already existing type, that isn't already a parent". I'm not sure if this is due to a version upgrade or b/c I am starting with a brand new Elasticsearch server install.
Contrived example that shows what I'm seeing:
// UserSearchResult.java
#Document(indexName = "hr_index", type = "user")
public class UserSearchResult implements Serializable {
...
#Field(type=FieldType.keyword)
#Parent(type="department")
private String departmentCode;
...
}
// DepartmentSearchResult.java
#Document(indexName = "hr_index", type = "department")
public class DepartmentSearchResult implements Serializable {
...
}
When I start my application I get that exception. If I check the ElasticSearch server, I see the "hr_index" index and the "department" mapping, but the "user" mapping is not created.
If I understand the error, it's because "department" is being created and then when Spring tries to create "user" with "department" as its parent, it doesn't like that, since department wasn't previously marked as a parent when it was created.
Is there some way (via annotation?) to denote DepartmentSearchResult as being a parent when it's created somehow?
Or, is it possible to give a hint to Spring Data Elasticsearch as to what order it should create the indices/mappings? I have seen some other posts (Spring Data Elasticsearch Parent/Child Document Repositories / Test execution error) but disabling auto creation and then manually creating it myself (either as part of my Spring codebase or external to the app) seems kind of "un-Spring-y" to me?
Or, is there some other approach I should be taking?
(This is a working Spring application that had been using Spring 4.2.1 and Spring Data Release Train Gosling, that I'm attempting to upgrade to use Spring 5.0.0 and Spring Data Release Train Kay. As part of this I am starting with a fresh Elasticsearch install, and so I'm not sure if this error is coming from the upgrade or just b/c the install is clean).
In the SD ES, issues related to the parent-child relationship at now really poorly developed.
The problem is most likely due to the fact that you are using a clean installation of Elasticsearch. Before the update, the problem did not arise, because mappings have already been created. For the solution, you can use elasticsearchTemplate, which is part of SD ES, and ApplicationListener. It's simple. Just 3 steps.
Drop index in ES (it only needs one time):
curl -XDELETE [ES_IP]:9200/hr_index
Tell SD ES not to create indices and mappings automatically
// UserSearchResult.java
#Document(indexName = "hr_index", type = "user", createIndex = false)
public class UserSearchResult implements Serializable {
...
#Field(type=FieldType.keyword)
#Parent(type="department")
private String departmentCode;
...
}
// DepartmentSearchResult.java
#Document(indexName = "hr_index", type = "department", createIndex = false)
public class DepartmentSearchResult implements Serializable {
...
}
Add a ApplicationListener:
#Component
public class ApplicationStartupListener implements ApplicationListener<ContextRefreshedEvent> {
#Autowired
private ElasticsearchTemplate elasticsearchTemplate;
//Mapping for child must be created only if mapping for parents doesn't exist
#Override
public void onApplicationEvent(ContextRefreshedEvent event) {
elasticsearchTemplate.createIndex(DepartmentSearchResult.class);
try {
elasticsearchTemplate.getMapping(DepartmentSearchResult.class);
} catch (ElasticsearchException e) {
elasticsearchTemplate.putMapping(UserSearchResult.class);
elasticsearchTemplate.putMapping(DepartmentSearchResult.class);
}
}
}
P.S. Among other things, it is worth paying attention to the fact that with the release of ES 5.6, a process for removing types began. This inevitably entails the removal of the parent-child relationship. In one of the next releases of the SD ES, we will provide the opportunity to work with joins. Working with parent-child relationships is unlikely to be improved
I'm using Vaadin 7 + Grails 2.3, there is some questions
My domain classes
class Base {
private static final Date NULL_DATE = new Date(0)
Date createdAt = NULL_DATE;
Date updatedAt = NULL_DATE;
def beforeInsert(){
createdAt = new Date();
updatedAt = new Date();
}
def beforeUpdate(){
updatedAt = new Date();
}
static mapping = {
autoTimestamp true
}
}
abstract class Person extends Base{
String name;
String name2;
String phone1;
String phone2;
static constraints = {
name2 nullable:true
phone1 nullable:true
phone2 nullable:true
}
}
class Customer extends Person {
double credit;
}
THE PROBLEMS
PROBLEM 1
In my Vaadin class UI, if I try this
class MyUI extends UI {
#Override
protected void init(VaadinRequest vaadinRequest) {
Customer customer = new Customer()
customer.name="RODRIGO"
customer.save()
}
}
Show this error
Servlet.service() for servlet [VaadinServlet 0] in context with path [/AgileWeb] threw exception [com.vaadin.server.ServiceException: groovy.lang.MissingPropertyException: No such property: name for class: agileweb.Customer
Possible solutions: all] with root cause
Message: No such property: name for class: agileweb.Customer
Possible solutions: all
there is no "name" property? The class Customer extends Person that has this property.
PROBLEM 2
If I try this
Customer customer = new Customer()
Customer.setName("RODRIGO")
Customer.save()
Show thos error : Message: No Hibernate Session bound to thread, and configuration does not allow creation of non-transactional one here
I have seach about this error but I didn't understand to fix it, maybe I'm new with Grails
PROBLEM 3 - ACTUALLY IS A QUESTION
I know that I can use BeanItemContainer, just from List object, that is possible with no problem, but BeanItemContainer is not lazy load, I'd like to use HbnContainer addon (https://vaadin.com/directory#addon/hbncontainer) becase it just need a hibernate session, so How Can I get the "Hibernante session", is there any example and how to do it?
PROBLEM 4 - ACTUALLY IS A QUESTION (AGAIN)
Following this tutorial https://vaadin.com/wiki/-/wiki/Main/Vaadin%20on%20Grails%20-%20Database%20access
It works to save object in the database, but the questions
- Is it really necessary to create a service for each domain class? I have read that it's recomend to put domain logical in the services, I agree with this, but what about simple domain that no need any logical?
so, is there possible to create something like DAO for services? is there any service design to avoid repeted code just to save objects?
I know that are many questions but I think these questions are the same of others, I really want to use Vaadin + Grails to enjoy the better of both, but is not easy to me at the moment!
Before I start answering your question, let me question you domain model. Base class is generally fine, but I want to talk about Person and Customer. You may have good reasons why you picked up inheritance, but please think of composition. Maybe you could have Person that contains a reference to an enum, that states type of the person. You could start here with that: http://en.wikipedia.org/wiki/Composition_over_inheritance
I think you have a typo there. Call save() method on 'customer' not 'Customer', which is a class
When there is a request coming to Grails application, it opens a session and the session is available during that request. There is not this kind of behavior like that in Vaadin. You need to put it into a Service. Yes, you can make generic service to save an object
class GenericService { def save(def domain) { domain.save(failOnError:true) }}
You can get the session factory like this
import org.codehaus.groovy.grails.commons.ApplicationHolder as AH
def ctx = AH.application.mainContext
def sessionFactory = ctx.sessionFactory
or
ApplicationContext applicationContext = grailsApplication.mainContext
ConfigurableLocalSessionFactoryBean factory = applicationContext.getBean('&sessionFactory')
As I wrote, you could create GenericService or a service per domain object. Just to keep in mind that GenericService should only save an object and contain no other logic that would be specific for a domain object.
I tried to change the standard 'id' in grails:
calls Book {
String id
String title
static mapping {
id generator:'assigned'
}
}
unfortunately, I soon noticed that this breaks my bootstrap. Instead of
new Book (id:'some ISBN', title:'great book').save(flush:true, failOnError:true)
I had to use
def b = new Book(title:'great book')
b.id = 'some ISBN'
b.save(flush:true, failOnError:true)
otherwise I get an 'ids for this class must be manually assigned before calling save()' error.
but that's ok so far.
I then encountered the same problem in the save action of my bookController. But this time, the workaround didn't do the trick.
Any suggestions?
I known, I can rename the id, but then I will have to change all scaffolded views...
That's a feature of databinding. You don't want submitted data to be able to change managed fields like id and version, so the Map constructor that you're using binds all available properties except those two (it also ignores any value for class, metaClass, and a few others).
So there's a bit of a mismatch here since the value isn't managed by Hibernate/GORM but by you. As you saw the workaround is that you need to create the object in two steps instead of just one.
I can't replicate this problem (used Grails 2.0.RC1). I think it might be as simple as a missing equal sign on your static mapping = { (you just have static mapping {)
Here's the code for a domain object:
class Book {
String id
String name
static mapping = {
id generator:'assigned'
}
}
And inside BootStrap.groovy:
def init = { servletContext ->
new Book(name:"test",id:"123abc").save(failOnError:true)
}
And it works fine for me. I see the id as 123abc.
You need to set the bindable constraint to true for your id prop, e.g.
class Employee {
Long id
String name
static constraints = {
id bindable: true
}
}
Is it possible to explicitly set the id of a domain object in Grails' Bootstrap.groovy (or anywhere, for that matter)?
I've tried the following:
new Foo(id: 1234, name: "My Foo").save()
and:
def foo = new Foo()
foo.id = 1234
foo.name = "My Foo"
foo.save()
But in both cases, when I print out the results of Foo.list() at runtime, I see that my object has been given an id of 1, or whatever the next id in the sequence is.
Edit:
This is in Grails 1.0.3, and when I'm running my application in 'dev' with the built-in HSQL database.
Edit:
chanwit has provided one good solution below. However, I was actually looking for a way to set the id without changing my domain's id generation method. This is primarily for testing: I'd like to be able to set certain things to known id values either in my test bootstrap or setUp(), but still be able to use auto_increment or a sequence in production.
Yes, with manually GORM mapping:
class Foo {
String name
static mapping = {
id generator:'assigned'
}
}
and your second snippet (not the first one) will do the job (Id won't be assigned when passing it through constructor).
What I ended up using as a workaround was to not try and retrieve objects by their id. So for the example given in the question, I changed my domain object:
class Foo {
short code /* new field */
String name
static constraints = {
code(unique: true)
name()
}
}
I then used an enum to hold all of the possible values for code (which are static), and would retrieve Foo objects by doing a Foo.findByCode() with the appropriate enum value (instead of using Foo.get() with the id like I wanted to do previously).
It's not the most elegant solution, but it worked for me.
As an alternative, assuming that you're importing data or migrating data from an existing app, for test purposes you could use local maps within the Bootstrap file. Think of it like an import.sql with benefits ;-)
Using this approach:
you wouldn't need to change your domain constraints just for
testing,
you'll have a tested migration path from existing data, and
you'll have a good data slice (or full slice) for future integration tests
Cheers!
def init = { servletContext ->
addFoos()
addBars()
}
def foosByImportId = [:]
private addFoos(){
def pattern = ~/.*\{FooID=(.*), FooCode=(.*), FooName=(.*)}/
new File("import/Foos.txt").eachLine {
def matcher = pattern.matcher(it)
if (!matcher.matches()){
return;
}
String fooId = StringUtils.trimToNull(matcher.group(1))
String fooCode = StringUtils.trimToNull(matcher.group(2))
String fooName = StringUtils.trimToNull(matcher.group(3))
def foo = Foo.findByFooName(fooName) ?: new Foo(fooCode:fooCode,fooName:fooName).save(faileOnError:true)
foosByImportId.putAt(Long.valueOf(fooId), foo) // ids could differ
}
}
private addBars(){
...
String fooId = StringUtils.trimToNull(matcher.group(5))
def foo = foosByImportId[Long.valueOf(fooId)]
...
}