Hi,
I've added a flag to indicate wether the indexer should be called synchronous (before transaction commit) or asynchronous (in a separat thread after commit).
<listener classname="org.apache.slide.search.IndexTrigger">
<configuration>
<indexer classname="org.apache.slide.search.LoggingIndexer" uri="/files/articles" synchronous="true" />
</configuration>
</listener>
Regards, Daniel
Hi,
just checked in the event driven index trigger.
It can trigger indexers that implement the Indexer interface. The configuration of indexers is the same as for the extractor framework. I've implemented a simple LoggingIndexer that logs a message every time the index should be updated, created or dropped. This is a good starting point for custom implementations.
Indexer matching can be achieved via uri or content-type mapping (see extractor)
Notes: * The current namespace can be accessed via the given uri. * The index trigger relies on the event collectors, so they must be activated.
This is the part of the configuration in Domain.xml that activated the index trigger.
The custom indexers can be configured via Configurable interface (see my extractor posing for an example). Indexers and extractors can follow the singleton pattern if they should be accessable inside the vm.
<listener classname="org.apache.slide.search.IndexTrigger">
<configuration>
<indexer classname="org.apache.slide.search.LoggingIndexer" uri="/files/articles" />
</configuration>
</listener>
Regards, Daniel
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
