[
https://issues.apache.org/jira/browse/IGNITE-32?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14230777#comment-14230777
]
Dmitriy Setrakyan commented on IGNITE-32:
-----------------------------------------
As an output, there should be the following structure
{noformat}
- ignite.xml (Ignite configuration, either new or existing)
|_ caches/
|_ personCache.xml (configuration for type Person based on
GridCacheQueryTypeMetadata)
|_ organizationCache.xml (configuration for type Organization based on
GridCacheQueryTypeMetadata)
|_ ...
{noformat}
We should either be able to generate a new XML configuration files ore merge
into existing ones.
Also, if user selects to generate classes as well, then we should generate
classes as follows:
{noformat}
some.domain.package/
Person.java
Organization.java
...
{noformat}
> Need to create a utility for auto-schema loading from databases
> ---------------------------------------------------------------
>
> Key: IGNITE-32
> URL: https://issues.apache.org/jira/browse/IGNITE-32
> Project: Ignite
> Issue Type: Task
> Components: UI
> Reporter: Dmitriy Setrakyan
> Priority: Critical
> Fix For: sprint-1
>
>
> Need to create utility for auto-schema loading:
> # Should be written in Java
> # Preferably no additional dependencies should be added to the project.
> The process is composed of 2 parts, schema mapping and automatic data loading.
> h3. Schema Mapping
> # Should be a standalone GUI utility
> # Users should select schemas, then tables from schemas, then fields from
> tables, and automatically generate Spring XML for SQL Query metadata
> configuration.
> # Should also optionally automatically generate POJO classes.
> # Should automatically map SQL table names to Java standard allowing users to
> provide customer Regex.
> # By default, every table should be mapped to a separate Cache, but user
> should be able to override this in any possible way.
> h3. Automatic {{GridCacheStore}} implementation
> # The new {{GridCacheStore}} implementation should read the XML configuration
> generated by *Schema Mapping* step and automatically implement all CRUD and
> TX operations out of the box.
> # The implementation of the {{loadCache(...)}} method should accept SQL for
> loading (or where clause) as a parameter. Users should be able to preload a
> portion of the table based on some search criteria on startup or at any point
> after.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
