I'm new to MongoDB and I need food for thought.
My project is a ddd-/clean architecture-like project, so i want my domain framework free.
Now I have repositories in my domain with some methods like loadAggregate(...) saveAggregate(...) etc.
And additionally there are query repositories with other data structure to read the data.
Now I need to implement that repositories and I want to use MongoDB.
But I don't know, what is the best strategy.
My Rest-Framework is Spring, so I can use String Data MongoDB. But how can I connect my simple repositories with the spring repositories?
Or I can use MongoDB natively or with KMongo...
Which choice is the best solution to start with?
I hope, my trouble became clear.
Big thx!
I want to know if there's a way to configure the datasource for Ignite as Elastic Search. I was browsing the web. But I did not find a solution.
I want to implement this integration for a Java application.
If I understand your idea correctly there's a way to do it. As far as I can see Elasticsearch supports SQL table-like data access and it's available through jdbc connection. From the Ignite's side we have 3rd party persistance, it uses jdbc to connect to an underlying store system. To be honest I haven't tested it but I suppose it should work.
Also I need mention that you can use GridGain WebConsole to generate simple Ignite project from existing jdbc connection. This functionality could be found on Configuration tab -> Create Cluster Configuration.
Searched over the net but unable to find the satisfying approach.
I am new to spring boot and aware of starter dependancies,
I want to develop a springboot app using elastic search as a storage system.
Wherever i searched i found that somewhere my service class will have to implement some interface from springframework for ES crud operations.
Is there any other way without implementing or extending the components.
I myself want to create transport client and want to query ES by my code or methods not by overidden ones.
Please if you ahve ever seen any projects you can redirect me to that link .
Thanks.
Assuming I understand you correctly, you can use the Elasticsearch REST client: https://www.elastic.co/guide/en/elasticsearch/client/java-rest/current/java-rest-low.html
You supply the JSON entities for the queries and parse the responses yourself. Its pretty basic in what it does, so you're not dependent on a lot of third party stuff to perform operations.
I have an LDAP-Server that contains a large set of user data and would like to import this into an Apache Solr index. The question is not about whether this is a good idea or not (as discussed here). I need this kind of architecture as one of our production systems depends on a Solr index of our ldap data.
I'm considering different options to do so, but I'm not sure which one should be preferred:
Option 1: Use the Apache Solr DataImportHandler:
This seems to be the most straight forward Solr way of doing so. Unfortunately there does not seem to be DataSource available that would work with LDAP.
I tried to combine the JdbcDataSource with the JDBC-LDAP-Bridge. In theory that might probably work but the driver looks quite dated (latest Version from 2007).
Another Option might be to write a custom LdapDataSource using some of the LDAP-Libraries for Java (probably Spring LDAP, directly via JNDI or something similar?).
Option 2: Build a custom Feeder:
Another option might be to write a standalone service/script that bridges between the two services. However that feels a bit like reinventing the wheel.
Option 3: Something I haven't thought of yet:
Maybe there are additional options here that I simply haven't discovered yet.
Solved it by writing a custom LDAP DataSource for the Solr DataImportHandler.
It's not as hard as it sounds. The JdbcDataSource can be used as a template for writing your custom DataSource, so basically you just have to rewrite that one Java-Class for the LDAP protocol.
For accessing the LDAP-Client there are numerous options, such as plain JNDI, UnboundID LDAP SDK, Apache LDAP API, OpenDJ LDAP SDK or OpenLDAP JLDAP (there are probably more but I only had a look at those).
I went for UnboundID LDAP due to its well documented API and full support for LDAPv3.
Afterwards it is just a matter of referencing the datasource from the data-config.xml.
A nice side-effect of this setup is, that you can use all the goodies that the Solr DataImportHandler provides while indexing the LDAP server (Entity Processors and Transformers). This makes it easy to map the data structure between LDAP and the Solr Index.
I'm looking for resources showing how to integrate MongoDB with Hibernate (preferably from within spring) so that I can switch between a RDBMS and a NoSql alternative: does anyone have experience doing this?
You can't easily do this. The point of Hibernate is to map Java Objects to a relational database. Although Hibernate abstracts a lot of details away you still need to understand how relational databases work with things such as foreign and primary keys, and the performance implications of queries you run. MongoDB requires an entire different way of designing your database focusing on objects instead of columns and tables. while you may be able to create a Hibernate dialect for MongoDB creating a design that would work on both a relational database and a NoSql database will give you a design that works poorly on both.
What about Hibernate OGM? It provides JPA for No-SQL databases.
Migration would be easier if you use Spring MongoTemplate (similar to HibernateTemplate). Among its features is support for JPA annotations (although, I'm not sure to what extent).
See more: http://www.springsource.org/spring-data/mongodb
You'll need the following:
Add spring-data-mongodb JAR to your project (available in maven
central).
Add mongo-java-driver JAR to your project (available in
maven central).
Use the provided MongoTemplate class in a similar
manner to HibernateTemplate. E.g.:
mongoTemplate.findById(id, MyClass.class);
mongoTemplate.insert(myObject);
Here's a concrete example with code: use-spring-and-hibernate-with-mongodb
If you are using Java then you can use Hibernate OGM it provides Java Persistence support for NoSQL databases.
For more details visit http://hibernate.org/ogm/
There is nice work done earlier as:
http://pragmaticintegrator.wordpress.com/2011/07/14/use-spring-and-hibernate-with-mongodb/
http://pragmaticintegrator.wordpress.com/2011/07/27/unit-test-your-springhibernate-and-mongodb-setup/#comments
refer to these links. it will be helpful to you.
There is also kundera, which uses JPA-annotations to read/write your object from/to a mongodb. If you ara familiar with hibernate, it should be quite straightformard to use.
I recently tried Morphia, which takes the same approach, but with its own annotations.
It works fine
May this blog helps: http://drorbr.blogspot.com/2010/02/migrating-springhibernate-application.html
Here Dror Bereznitsky describes nicely how to integrate a sping/hibernate based solution with mongodb.
For the sake of completeness, PlayORM also supports MongoDB now. PlayORM is an object NoSQL mapping solution so you can write POJO’s and let it deal with all the details of marshalling/unmarshalling to MongoDB. Visit its documentation here
I think Hibernate provides desired functionality. Take a look at this, found on their official website:
Mixing several NoSQL datastores in one application, e.g. use Neo4j for your friendship graph and MongoDB for your blog posts. Or mix NoSQL and relational databases.
reference
Well just to give you an example, I am doing somehting simmilar. In ColdFusion, Hibernate is integrated and in order to save your Hibernate Object, you hvae to do EntitySave(Obj). However what we have done is build the Orm object, and then use a mongoDB Coldfusion component and just save the object by going mongo.Save(obj,collectionName).