Oh!! Haha that is very good to know.
I am trying to implement a polyglot persistence solution that uses a RDBMS
and Neo4j JUST for its graphing / relationship capabilities. The nodes in
Neo4j would refer to objects that exist in the RDS and so the cross store
persistence support in SDN seemed like a great option. But now that I know
it may not work the way I expected, I'm wondering if you have any general
advice on how to use Neo4j with another datastore:
- What are some common patterns for including Neo4j in a polyglot setup?
- I'm specifically interested in understanding how one would keep data
in sync if Neo4j has references to data in other stores
- Use of XA transactions? Using jobs to sync things up later?
- Or is it more common to somehow model the domain so that data is
silo'ed into their individual stores? (not really sure this is possible
in
my use case)
Thank you!
On Wednesday, June 17, 2015 at 5:51:23 PM UTC-5, Michael Hunger wrote:
>
> I would say just ignore it.
>
> We had the idea a few years back but no one ever really used it, so it has
> been dormant ever since.
>
> JPA repos are not supported, it works with the entity manager methods as
> far as I can remember.
> And it was never meant or tested with a remote Neo4j service :)
>
> Why do you want to have this kind of setup?
>
> Michael
>
>
> Am 18.06.2015 um 00:13 schrieb Simon Lam <[email protected] <javascript:>
> >:
>
> Hello,
>
> I am a bit new to both Neo4j but this is more of an SDN specific question
> so I apologize in advance if this is the incorrect forum for this. I am
> working to get a minimal cross store (JPA + SDN) project set up and having
> some trouble with it. I've followed the instructions on the docs and have
> a config class that looks like:
>
> @Configuration
>
> @ComponentScan
>
> @EnableAutoConfiguration
>
> @EnableJpaRepositories
> @EnableTransactionManagement(mode = AdviceMode.ASPECTJ)
>
> @EnableJpaAuditing
> @EnableNeo4jRepositories
> public class Neo4jConfig extends CrossStoreNeo4jConfiguration {
>
> public Neo4jConfig() {
>
> setBasePackage("com.m1finance.entity");
> }
>
> @Bean
> public GraphDatabaseService graphDatabaseService() {
> return new SpringCypherRestGraphDatabase("
> http://localhost:7474/db/data/";, "user", "pw");
> }
> }
>
>
> So far I can verify in the logs that a chained tx manager is being used
> and that an initial REST call is definitely being made to my local Neo4j DB.
>
> I have a partial NodeEntity and would like to perform a sanity check by
> persisting it to an in memory H2 DB using a JPA respository call in a JUnit
> test. This seems to work fine with no errors (an ID is generated on the
> entity, etc) but absolutely nothing happens on the Neo4j side. Nothing in
> the test logs, nothing in the DB logs, no calls are being made. My
> questions are:
>
> - Is the basic assumption that saving the partial entity using a Jpa
> repo should create a graph node correct? Or am I supposed to be making
> individual calls (e.g. one to persist in my RDBMS and another in Neo4j)?
> - My research on the internet indicates that the cross store
> functionality is driven by aspects. Is this correct? I've included the
> plugin to download the required dependencies per the docs but are there
> any
> ways to verify that things are actually set up properly?
>
> Thank you!
>
> --
> You received this message because you are subscribed to the Google Groups
> "Neo4j" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected] <javascript:>.
> For more options, visit https://groups.google.com/d/optout.
>
>
>
--
You received this message because you are subscribed to the Google Groups
"Neo4j" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
