Package org.apache.jackrabbit.oak.plugins.index.property.strategy

Class ContentMirrorStoreStrategy

java.lang.Object

org.apache.jackrabbit.oak.plugins.index.property.strategy.ContentMirrorStoreStrategy

All Implemented Interfaces:

IndexStoreStrategy

public class ContentMirrorStoreStrategy
extends java.lang.Object
implements IndexStoreStrategy

An IndexStoreStrategy implementation that saves the nodes under a hierarchy that mirrors the repository tree.
This should minimize the chance that concurrent updates overlap on the same content node.

For example for a node that is under /test/node, the index structure will be /oak:index/index/test/node:

 
 /
   test
     node
   oak:index
     index
       test
         node
 
 

Field Summary

Fields

Modifier and Type	Field	Description
static int	TRAVERSING_WARN	logging a warning every oak.traversing.warn traversed nodes.

Constructor Summary

Constructors

Constructor	Description
ContentMirrorStoreStrategy()
ContentMirrorStoreStrategy(java.lang.String indexName)
ContentMirrorStoreStrategy(java.lang.String indexName, java.lang.String pathPrefix, boolean prependPathPrefix)	Constructs a ContentMirrorStoreStrategy

Method Summary

Modifier and Type	Method	Description
long	count(Filter filter, NodeState root, NodeState indexMeta, java.util.Set<java.lang.String> values, int max)	Count the occurrence of a given set of values.
long	count(NodeState root, NodeState indexMeta, java.util.Set<java.lang.String> values, int max)	Count the occurrence of a given set of values.
boolean	exists(org.apache.jackrabbit.guava.common.base.Supplier<NodeBuilder> index, java.lang.String key)	Check whether an entry for the given key exists.
java.lang.String	getIndexNodeName()
java.lang.Iterable<java.lang.String>	query(Filter filter, java.lang.String name, NodeState indexMeta, java.lang.Iterable<java.lang.String> values)	Search for a given set of values.
java.lang.Iterable<java.lang.String>	query(Filter filter, java.lang.String indexName, NodeState indexMeta, java.lang.String indexStorageNodeName, java.lang.Iterable<java.lang.String> values)
void	update(org.apache.jackrabbit.guava.common.base.Supplier<NodeBuilder> index, java.lang.String path, @Nullable java.lang.String indexName, @Nullable NodeBuilder indexMeta, java.util.Set<java.lang.String> beforeKeys, java.util.Set<java.lang.String> afterKeys)	Updates the index for the given path.

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

TRAVERSING_WARN

public static final int TRAVERSING_WARN

logging a warning every oak.traversing.warn traversed nodes. Default 10000

Constructor Detail

ContentMirrorStoreStrategy

public ContentMirrorStoreStrategy()

ContentMirrorStoreStrategy

public ContentMirrorStoreStrategy(java.lang.String indexName)

ContentMirrorStoreStrategy

public ContentMirrorStoreStrategy(java.lang.String indexName,
                                  java.lang.String pathPrefix,
                                  boolean prependPathPrefix)

Constructs a ContentMirrorStoreStrategy

Parameters:

indexName - name of sub node under which paths are stored

pathPrefix - path of the index in repository. Defaults to empty for indexes at root nodes i.e. those stored directly under '/oak:index'. For non root index its the path excluding the '/oak:index' node. For e.g. for index at '/content/oak:index/fooIndex' the pathPrefix would be '/content'. If this is appened to the paths returned by index then they would become absolute path in repository

prependPathPrefix - Should the path prefix be added to the query result

Method Detail

update

public void update(org.apache.jackrabbit.guava.common.base.Supplier<NodeBuilder> index,
                   java.lang.String path,
                   @Nullable
                   @Nullable java.lang.String indexName,
                   @Nullable
                   @Nullable NodeBuilder indexMeta,
                   java.util.Set<java.lang.String> beforeKeys,
                   java.util.Set<java.lang.String> afterKeys)

Description copied from interface: IndexStoreStrategy

Updates the index for the given path.

Specified by:

update in interface IndexStoreStrategy

Parameters:

index - the index node supplier

path - path stored in the index

indexName - the name of the index. May be null.

indexMeta - the definition of the index. May be null.

beforeKeys - keys that no longer apply to the path

afterKeys - keys that now do apply to the path

query

public java.lang.Iterable<java.lang.String> query(Filter filter,
                                                  java.lang.String indexName,
                                                  NodeState indexMeta,
                                                  java.lang.String indexStorageNodeName,
                                                  java.lang.Iterable<java.lang.String> values)

query

public java.lang.Iterable<java.lang.String> query(Filter filter,
                                                  java.lang.String name,
                                                  NodeState indexMeta,
                                                  java.lang.Iterable<java.lang.String> values)

Description copied from interface: IndexStoreStrategy

Search for a given set of values.

Specified by:

query in interface IndexStoreStrategy

Parameters:

filter - the filter (can optionally be used for optimized query execution)

name - the name of the index (for logging)

indexMeta - the index metadata node (may not be null)

values - values to look for (null to check for property existence)

Returns:

an iterator of paths

count

public long count(NodeState root,
                  NodeState indexMeta,
                  java.util.Set<java.lang.String> values,
                  int max)

Description copied from interface: IndexStoreStrategy

Count the occurrence of a given set of values. Used in calculating the cost of an index.

Specified by:

count in interface IndexStoreStrategy

Parameters:

root - the root node (may not be null)

indexMeta - the index metadata node (may not be null)

values - values to look for (null to check for property existence)

max - the maximum value to return

Returns:

the aggregated count of occurrences for each provided value

count

public long count(Filter filter,
                  NodeState root,
                  NodeState indexMeta,
                  java.util.Set<java.lang.String> values,
                  int max)

Description copied from interface: IndexStoreStrategy

Count the occurrence of a given set of values. Used in calculating the cost of an index.

Specified by:

count in interface IndexStoreStrategy

Parameters:

filter - the filter which can be used to estimate better cost

root - the root node (may not be null)

indexMeta - the index metadata node (may not be null)

values - values to look for (null to check for property existence)

max - the maximum value to return

Returns:

the aggregated count of occurrences for each provided value

exists

public boolean exists(org.apache.jackrabbit.guava.common.base.Supplier<NodeBuilder> index,
                      java.lang.String key)

Description copied from interface: IndexStoreStrategy

Check whether an entry for the given key exists.

Specified by:

exists in interface IndexStoreStrategy

Parameters:

index - the index node supplier

key - the key

Returns:

true if at least one entry exists

getIndexNodeName

public java.lang.String getIndexNodeName()

Specified by:

getIndexNodeName in interface IndexStoreStrategy