The XML connector has a requirement that it stream the XML documents it is parsing because
large docs can blow out memory. In order to satisfy the streaming and caching
requirements, I am chunking the xml into nodes under a node representing the response as
it passes into the parser. So a document is represented in the cache by a FQN (the ID)
and it's children (the content).
I assume in the API you're describing the Object returned from getFromCache is the
actual cached object, and not a Node of any sort, right? I would prefer to have the
hierarchical structure of the cache exposed in the API, it simplifies access in ways that
would otherwise have to be accomplished through more state in the connector. In this
case, the connector would have to map a document key to multiple cache items.
~jd
----- "Ramesh Reddy (JIRA)" <jira-events(a)lists.jboss.org> wrote:
[
https://jira.jboss.org/jira/browse/TEIID-444?page=com.atlassian.jira.plug...
]
Ramesh Reddy updated TEIID-444:
-------------------------------
Fix Version/s: 6.1.0
Affects Version/s: 6.0.0
To support providing the Cache services at connector level, Teiid will
provide following two methods on the "ExecutionContext" interface in
the Connector API
-- Object getFromCache(CacheScope scope, Object key);
-- void storeInCache(CacheScope scope, Object key, Object value);
Where CacheScope is defined as
public enum CacheScope {
REQUEST, // retained until end of every top level request
SESSION, // retained until end of user session termination
VDB, // retained for life of VDB
SERVICE, // retained for life of the service
GLOBAL; // common across system, query engine life cycle
}
except for REQUEST, the items placed in all other scopes will be
replicated across the cluster in server mode. In embedded, no
replication is available.
> Expose the cache to connectors.
> -------------------------------
>
> Key: TEIID-444
> URL:
https://jira.jboss.org/jira/browse/TEIID-444
> Project: Teiid
> Issue Type: Feature Request
> Components: Connector API
> Affects Versions: 6.0.0
> Reporter: John Doyle
> Assignee: Ramesh Reddy
> Fix For: 6.1.0
>
>
> The XML-Relational connectors have historically had their own cache.
I'm modifying the connectors to use JBoss cache and it would benefit
users if this connector and other connectors with need of a cache
could use the teiid cache to ease cache configuration and monitoring.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the
administrators:
https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
http://www.atlassian.com/software/jira
_______________________________________________
teiid-issues mailing list
teiid-issues(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/teiid-issues
--
John Doyle
Senior Software Engineer
JBoss, a division of Red Hat
Office: 978.392.3916
Email: jdoyle(a)redhat.com