com.sapportals.wcm.service.indexmanagement
Interface ITaskQueueHandler

[contained in: com.sap.km.cm.service.par - km.shared.service.indexmanagement_api.jar]

public interface ITaskQueueHandler

The handler puts index management tasks into the task queue of the task queue service and reads index management tasks from the task queue to execute them.
ITaskQueueHandler is mainly responsible for indexing, reindexing and deltaindexing of documents. If no TaskQueueService is active the methods are immediately executed. Otherwise tasks are enqueued in a task queue. The task queue is checked according to a time intervall, which could be adjusted in the system configuration. This way, it is possible to execute different tasks on different machines in a distributed CM scenario. Internally this interface is also used by particular indexes , when the corresponding methods are called.

See Also:

IIndex, AbstractIndex

Method Summary

void	activateSchedulerEntry(java.lang.String schedulerEntryId) Activates a newly created scheduler entry for the current system of the task queue reader.
void	createTaxonomyHierarchyByCopiing(java.lang.String taxonomyName, ICollection existingHierarchy, IClassificationIndex index) The taxonomy hierarchy is created by copiing an existing hierarchy.
void	deindexDocuments(IIndex index, IRidList deindexRids, boolean removeClassification) Deindexes a list of document IDs from an index.
void	deltaReindex(IIndex index) Performs indexing of the documents that are not yet indexed or have been changed since the last crawl.
void	executeTask(ITask task) Runs an index management task of the task queue service.
void	indexDocument(IIndex index, IResource document, IResourceEvent event) Indexes a document into an index.
void	reindex(IIndex index, boolean disablePostClearTasks) Clears the index and starts putting all documents into the index again.
void	startDeindexCrawler(IIndex index, RID startRid) Runs a deindex crawl as a separate thread.
void	startDeindexCrawler(IIndex index, RID startRid, IResourceEvent event, RID eventPreRid, RID eventNewRid, java.lang.String crawlerProfileId) Runs a deindex crawl as a separate thread.
void	startDeltaCrawler(IIndex index, RID startRid) Runs an delta index crawl as a separate thread.
void	startIndexCrawler(IIndex index, RID startRid, IResourceEvent event) Runs an index crawl as a separate thread.
void	startTrainingCrawler(java.lang.String taxonomyName, ICollection taxonomyCollection, IClassificationIndex index, boolean isSearchRelevant) Runs a taxonomy training crawler as a separate thread.
void	updateCategoryContents(java.lang.String taxonomyName, boolean dirtyOnly, IIndex index, IResourceContext context, boolean markCategoriesUpdated) Starts a process that checks and updates categories of a taxonomy where the query is not in sync with the documents of category.

Method Detail

startIndexCrawler

public void startIndexCrawler(IIndex index,
                              RID startRid,
                              IResourceEvent event)
                       throws WcmException

Runs an index crawl as a separate thread.
An index crawl indexes all documents that are crawled.

Parameters:

index - index where the documents have to be indexed into

startRid - start resource of the crawl

event - event that caused the index request; can be null

Throws:

WcmException - exception raised in failure situation

startDeltaCrawler

public void startDeltaCrawler(IIndex index,
                              RID startRid)
                       throws WcmException

Runs an delta index crawl as a separate thread.
An delta index crawl indexes all changed documents since the last crawl.

Parameters:

index - index where the documents have to be indexed into

startRid - start resource of the crawl

Throws:

WcmException - exception raised in failure situation

startTrainingCrawler

public void startTrainingCrawler(java.lang.String taxonomyName,
                                 ICollection taxonomyCollection,
                                 IClassificationIndex index,
                                 boolean isSearchRelevant)
                          throws WcmException

Runs a taxonomy training crawler as a separate thread.
A taxonomy training crawl copies an existing folder hierarchy as a taxonomy and uses documents inside to train the categories.

Parameters:

taxonomyName - name of taxonomy

taxonomyCollection - start folder for the crawl; represents the taxonomy root

index - index where the documents have to be indexed into

isSearchRelevant - if true the start folder is attached to the index, if false the crawl is only used to build the taxonomy.

Throws:

WcmException - exception raised in failure situation

createTaxonomyHierarchyByCopiing

public void createTaxonomyHierarchyByCopiing(java.lang.String taxonomyName,
                                             ICollection existingHierarchy,
                                             IClassificationIndex index)
                                      throws WcmException

The taxonomy hierarchy is created by copiing an existing hierarchy.

Parameters:

taxonomyName - name of taxonomy

existingHierarchy - root folder of the hierarchy to copy

index - classification index where the taxonomy will be created

startDeindexCrawler

public void startDeindexCrawler(IIndex index,
                                RID startRid)
                         throws WcmException

Runs a deindex crawl as a separate thread.
An deindex crawl removes all crawled documents from the index.

Parameters:

index - index where the documents have to be deindexed from

startRid - start resource ID of the crawl

Throws:

WcmException - exception raised in failure situation

startDeindexCrawler

public void startDeindexCrawler(IIndex index,
                                RID startRid,
                                IResourceEvent event,
                                RID eventPreRid,
                                RID eventNewRid,
                                java.lang.String crawlerProfileId)
                         throws WcmException

Runs a deindex crawl as a separate thread.
An deindex crawl removes all crawled documents from the index.

Parameters:

index - index where the documents have to be deindexed from

startRid - start resource ID of the crawl

event - event that caused the deindex request; can be null

eventPreRid - resource ID of the events resource before an action like rename or move.

eventNewRid - current resource ID of the events resource

crawlerProfileId - crawler profile id of the data source as fallback if the data source is not available anymore when requesting the crawler profile id after a detach.

Throws:

WcmException - exception raised in failure situation

indexDocument

public void indexDocument(IIndex index,
                          IResource document,
                          IResourceEvent event)
                   throws WcmException

Indexes a document into an index.

Parameters:

index - index where the documents have to be indexed into

document - document to index

event - event that caused the index request; can be null

Throws:

WcmException - exception raised in failure situation

reindex

public void reindex(IIndex index,
                    boolean disablePostClearTasks)
             throws WcmException

Clears the index and starts putting all documents into the index again.
It may be necessary to reindex an index may be needed if a CM repository that is assigned to the index now points to another underlying repository. This can be caused by configuration changes. If this is the case, the documents in the index are no longer accessible from CM and must be removed from the index. The documents of the new target repository also have to be indexed.

Parameters:

index - index to reindex

disablePostClearTasks - if true, the tasks that can be performed by the index implementation after the index has been cleared are skipped.
Example: Do not train the classification index before reindexing all documents.

Throws:

WcmException - exception raised in failure situation

deltaReindex

public void deltaReindex(IIndex index)
                  throws WcmException

Performs indexing of the documents that are not yet indexed or have been changed since the last crawl. Additionally, it deindexes documents from the index that were indexed but have since been deleted. In contrast to the reindex method, this call does not clear the index.

Throws:

WcmException - exception raised in failure situation

deindexDocuments

public void deindexDocuments(IIndex index,
                             IRidList deindexRids,
                             boolean removeClassification)
                      throws WcmException

Deindexes a list of document IDs from an index.

Parameters:

index - index where the documents have to be deindexed from

deindexRids - list of resource IDs to deindex

removeClassification - if true the classification if the documents is deleted, if false the classification remains.

Throws:

WcmException - exception raised in failure situation

activateSchedulerEntry

public void activateSchedulerEntry(java.lang.String schedulerEntryId)
                            throws WcmException

Activates a newly created scheduler entry for the current system of the task queue reader.

Parameters:

schedulerEntryId - id of the scheduler entry

Throws:

WcmException - exception raised in failure situation

updateCategoryContents

public void updateCategoryContents(java.lang.String taxonomyName,
                                   boolean dirtyOnly,
                                   IIndex index,
                                   IResourceContext context,
                                   boolean markCategoriesUpdated)
                            throws WcmException

Starts a process that checks and updates categories of a taxonomy where the query is not in sync with the documents of category.

Parameters:

taxonomyName - name of the taxonomy to check

dirtyOnly - if false all categories are checked, if false only categories that are out of sync (dirty) are updated.

index - index where the taxonomy is defined

context - user context

markCategoriesUpdated - if false the updated categories remain dirty

Throws:

WcmException - exception raised in failure situation

executeTask

public void executeTask(ITask task)
                 throws WcmException

Runs an index management task of the task queue service.
This method is called by the task queue reader of the index management service.

Parameters:

task - task queue task

Throws:

WcmException - exception raised in failure situation