[
https://issues.apache.org/jira/browse/IGNITE-539?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Artem Shutak updated IGNITE-539:
--------------------------------
Affects Version/s: sprint-2
> Implement auto detection which caches should be used to execute SQL query
> -------------------------------------------------------------------------
>
> Key: IGNITE-539
> URL: https://issues.apache.org/jira/browse/IGNITE-539
> Project: Ignite
> Issue Type: Improvement
> Affects Versions: sprint-2
> Reporter: Artem Shutak
> Assignee: Sergi Vladykin
>
> Now, for execution of SQL query we have to have some instance of Ignite
> cache. But if we want to execute cross caches SQL query with full naming of
> tables we have to have some instance of cache.
> For example we have 2 caches: "partitioned" (with Person and Purchase tables)
> and "replicated" (with Product table). Then for execution query
> select firstName, lastName, Product.name from "partitioned".Person,
> "partitioned".Purchase, "replicated".Product where Person.id =
> Purchase.personId and Purchase.productId = Product.id;
> We have to use something like this:
> ignite.jcache("partitioned").queryFields(new SqlFieldsQuery(<sql_query>));
> It would be better if user have not to be care about which cache instance he
> should use. And use something like it:
> ignite.jcache().queryFields(new SqlFieldsQuery(<sql_query>));
> or
> ignite.query(<sql_query>);
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
