Package org.alfresco.repo.search

Interface Indexer

All Known Subinterfaces:

LuceneIndexer

All Known Implementing Classes:

AbstractLuceneIndexerImpl, IndexerComponent, NoActionIndexer, NoIndexIndexer

public interface Indexer

This interface abstracts how indexing is used from within the node service implementation. It has to optionally offer transactional integration For example, the lucene indexer

Author:

andyh

Method Summary

Modifier and Type	Method	Description
void	createChildRelationship(org.alfresco.service.cmr.repository.ChildAssociationRef relationshipRef)	Create a refernce link between a parent and child.
void	createNode(org.alfresco.service.cmr.repository.ChildAssociationRef relationshipRef)	Create an index entry when a new node is created.
void	deleteChildRelationship(org.alfresco.service.cmr.repository.ChildAssociationRef relationshipRef)	Delete a relationship between a parent and child.
void	deleteIndex(org.alfresco.service.cmr.repository.StoreRef storeRef)	Delete the index for a store
void	deleteNode(org.alfresco.service.cmr.repository.ChildAssociationRef relationshipRef)	Delete a node entry from an index.
void	detectNodeChanges(org.alfresco.service.cmr.repository.NodeRef nodeRef, org.alfresco.service.cmr.search.SearchService searcher, java.util.Collection<org.alfresco.service.cmr.repository.ChildAssociationRef> addedParents, java.util.Collection<org.alfresco.service.cmr.repository.ChildAssociationRef> deletedParents, java.util.Collection<org.alfresco.service.cmr.repository.ChildAssociationRef> createdNodes, java.util.Collection<org.alfresco.service.cmr.repository.NodeRef> updatedNodes)	Does a database vs index comparison for the given created/updated/renamed/referenced nodeRef in order to determine the set of indexing operations required
void	flushPending()
void	setReadThrough(boolean isReadThrough)	Activates 'read through' behaviour for this indexer.
void	updateChildRelationship(org.alfresco.service.cmr.repository.ChildAssociationRef relationshipBeforeRef, org.alfresco.service.cmr.repository.ChildAssociationRef relationshipAfterRef)	Alter the relationship between parent and child nodes in the index.
void	updateNode(org.alfresco.service.cmr.repository.NodeRef nodeRef)	Update an index entry due to property changes on a node.

Method Detail

createNode

void createNode(org.alfresco.service.cmr.repository.ChildAssociationRef relationshipRef)

Create an index entry when a new node is created. A node is always created with a name in a given parent and so a relationship ref is required.

Parameters:

relationshipRef - ChildAssociationRef

updateNode

void updateNode(org.alfresco.service.cmr.repository.NodeRef nodeRef)

Update an index entry due to property changes on a node. There are no strucural impications from such a change.

Parameters:

nodeRef - NodeRef

deleteNode

void deleteNode(org.alfresco.service.cmr.repository.ChildAssociationRef relationshipRef)

Delete a node entry from an index. This implies structural change. The node will be deleted from the index. This will also remove any remaining refernces to the node from the index. The index has no idea of the primary link.

Parameters:

relationshipRef - ChildAssociationRef

createChildRelationship

void createChildRelationship(org.alfresco.service.cmr.repository.ChildAssociationRef relationshipRef)

Create a refernce link between a parent and child. Implies only (potential) structural changes

Parameters:

relationshipRef - ChildAssociationRef

updateChildRelationship

void updateChildRelationship(org.alfresco.service.cmr.repository.ChildAssociationRef relationshipBeforeRef,
                             org.alfresco.service.cmr.repository.ChildAssociationRef relationshipAfterRef)

Alter the relationship between parent and child nodes in the index. This can be used for:

1. rename,
2. move,
3. move and rename,
4. replace

This could be implemented as a delete and add but some implementations may be able to optimise this operation.

Parameters:

relationshipBeforeRef - ChildAssociationRef

relationshipAfterRef - ChildAssociationRef

deleteChildRelationship

void deleteChildRelationship(org.alfresco.service.cmr.repository.ChildAssociationRef relationshipRef)

Delete a relationship between a parent and child. This will remove a structural route through the index. The index has no idea of reference and primary relationships and will happily remove the primary relationship before refernces which could remain. Use delete to ensure all strctural references are removed or call this sure you are doing an unlink (remove a hard link in the unix file system world).

Parameters:

relationshipRef - ChildAssociationRef

detectNodeChanges

void detectNodeChanges(org.alfresco.service.cmr.repository.NodeRef nodeRef,
                       org.alfresco.service.cmr.search.SearchService searcher,
                       java.util.Collection<org.alfresco.service.cmr.repository.ChildAssociationRef> addedParents,
                       java.util.Collection<org.alfresco.service.cmr.repository.ChildAssociationRef> deletedParents,
                       java.util.Collection<org.alfresco.service.cmr.repository.ChildAssociationRef> createdNodes,
                       java.util.Collection<org.alfresco.service.cmr.repository.NodeRef> updatedNodes)

Does a database vs index comparison for the given created/updated/renamed/referenced nodeRef in order to determine the set of indexing operations required

Parameters:

nodeRef - the nodeRef to process

searcher - searcher to query the indexes

addedParents - set to add new secondary parent associations to

deletedParents - set to add removed secondary parent associations to

createdNodes - set to add created nodes to

updatedNodes - set to add updated nodes to

deleteIndex

void deleteIndex(org.alfresco.service.cmr.repository.StoreRef storeRef)

Delete the index for a store

Parameters:

storeRef - StoreRef

flushPending

void flushPending()

setReadThrough

void setReadThrough(boolean isReadThrough)

Activates 'read through' behaviour for this indexer. Rather than accessing the database through the current (potentially old) transaction, it will use a discrete read only transaction for each node it indexes. This avoids 'stale' nodes building up in the caches during long reindex runs.

Parameters:

isReadThrough - boolean