Define the _id field as id for an object

[agindre]

Actually, I have to implement my own json parser to handle the elasticsearch json because I want to use the _id as value for my object's variable id. For me, it's the easy way to have an auto-increment id from elasticsearch that I can use as identifier for my doc.

The other way is :

1. create the document form java
2. get the latest id by sending a request to ES server
3. insert the id into the document
4. index the doc into ES

Here a good example to do so by using the _version as auto-increment : http://blogs.perl.org/users/clinton_gormley/2011/10/elasticsearchsequence---a-blazing-fast-ticket-server.html

[ferhatsb]

Hi Alex,

Sorry for late response, did you check @JestId annotation? Does it fit for you case?

I think, we are not clearly understand the exact problem. Can you please gist your modifications?

Ferhat

[agindre]

Hi Ferhat,

Thank you for you answer. My ticket is not clear, sorry for that :s
What I meant is that I'm looking for an easy way to insert my document into elasticsearch with an id that is generated by elasticsearch. Let me give you an example.

Let's say that I've this document into my scala code :

case class Test (
 @JestId
 id: Int = 1,
 content: String = "lopsum",
 author: String = "Me"
)

If I well understood the usage of @JestId when I use Index index = new Index.Builder(source).index("app").type("document").build();, it will index a document in elasticsearch as follow :

{
  "id" : 1,
  "content" : "lopsum",
 "author" : "Me"
}

and if I want to get it with Get get = new Get.Builder("1").index("app").type("document").build();, I will receive something like this :

{
  "_index" : "jobzazjswwmi7cqb",
  "_type" : "document",
  "_id" : "1",
  "_version" : 2,
  "exists" : true, "_source" : {"id" : 1, "content": "lopsum", "author": "Me"} 
}

What I want to do is that I want to insert into elasticsearch without to specify the id of my document to let elasticsearch manage the auto-increment. So I have the same case class Test but I made my own parser to index the document as follow :

{
 "content" : "lopsum",
 "author" : "Me"
}

and my own parser to make the binding with my class as follow :

Test.id = elasticSearch.answer._id
Test.content = elasticSearch.answer._source.content
Test.author = elasticSearch.answer._source.author

Without this, if I want to use the index method from Jest, I have to get the lastest Id from my index/type (eg: 1 in my example) then set this value + 1 into my Test.id then index it with Jest.

Here a link to show want I'm looking for and a possible solution :
http://blogs.perl.org/users/clinton_gormley/2011/10/elasticsearchsequence---a-blazing-fast-ticket-server.html

[ferhatsb]

Hi Alex,

Thanks for information. I have some improvements and now results are;

Mark your id field with JestId but set it to null (or just be sure it is null global variables etc.).

case class Test (
 @JestId
 id: String = null,
 content: String = "lopsum",
 author: String = "Me"
)

If you index this Test object

Test source = new Test();
Index index = new Index.Builder(source).index("app").type("document").build();

Id of this document will automatically created by ES.

Now if we search from ES;

 Search search = new Search(query)
 JestResult result = client.execute(search);
 List<Test> articleResult = result.getSourceAsObjectList(Test.class);

List of Test objects' id values will be auto generated values.

Does it cover your case ?

Ferhat

[agindre]

Ferhat,

That's exactly what I was looking for !

Thank you !

[ferhatsb]

Excellent! You are welcome.

We are getting close to release 0.0.2.