org.apache.metamodel.elasticsearch.nativeclient

Class ElasticSearchDataContext

java.lang.Object

org.apache.metamodel.AbstractDataContext

org.apache.metamodel.QueryPostprocessDataContext

org.apache.metamodel.elasticsearch.AbstractElasticSearchDataContext

org.apache.metamodel.elasticsearch.nativeclient.ElasticSearchDataContext

All Implemented Interfaces:

HasReadTypeConverters, DataContext, UpdateableDataContext

public class ElasticSearchDataContext
extends AbstractElasticSearchDataContext

DataContext implementation for ElasticSearch analytics engine. ElasticSearch has a data storage structure hierarchy that briefly goes like this:

• Index
• Document type (short: Type) (within an index)
• Documents (of a particular type)

When instantiating this DataContext, an index name is provided. Within this index, each document type is represented as a table. This implementation supports either automatic discovery of a schema or manual specification of a schema, through the SimpleTableDef class.

Field Summary

Fields inherited from class org.apache.metamodel.elasticsearch.AbstractElasticSearchDataContext

dynamicTableDefinitions, indexName, staticTableDefinitions, TIMEOUT_SCROLL

Fields inherited from class org.apache.metamodel.QueryPostprocessDataContext

INFORMATION_SCHEMA_NAME, SYSTEM_PROPERTY_CREATE_DEFAULT_TABLE_ALIAS

Constructor Summary

Constructors

Constructor and Description
ElasticSearchDataContext(org.elasticsearch.client.Client client, String indexName) Constructs a ElasticSearchDataContext and automatically detects the schema structure/view on all indexes (see detectTable(ClusterState, String, String)).
ElasticSearchDataContext(org.elasticsearch.client.Client client, String indexName, SimpleTableDef... tableDefinitions) Constructs a ElasticSearchDataContext.

Method Summary

Modifier and Type	Method and Description
protected SimpleTableDef[]	detectSchema() Performs an analysis of the available indexes in an ElasticSearch cluster instance and detects the elasticsearch types structure based on the metadata provided by the ElasticSearch java client.
static SimpleTableDef	detectTable(org.elasticsearch.cluster.ClusterState cs, String indexName, String documentType) Performs an analysis of an available index type in an ElasticSearch Client client and tries to detect the index structure based on the metadata provided by the java client.
protected Number	executeCountQuery(Table table, List<FilterItem> whereItems, boolean functionApproximationAllowed) Executes a simple count query, if possible.
protected Row	executePrimaryKeyLookupQuery(Table table, List<SelectItem> selectItems, Column primaryKeyColumn, Object keyValue) Executes a query which obtains a row by primary key (as defined by Column.isPrimaryKey()).
UpdateSummary	executeUpdate(UpdateScript update) Submits an UpdateScript for execution on the DataContext.
org.elasticsearch.client.Client	getElasticSearchClient() Gets the Client that this DataContext is wrapping.
protected DataSet	materializeMainSchemaTable(Table table, List<Column> columns, int maxRows) Executes a simple one-table query against a table in the main schema of the subclasses of this class.
protected DataSet	materializeMainSchemaTable(Table table, List<SelectItem> selectItems, List<FilterItem> whereItems, int firstRow, int maxRows) Execute a simple one-table query against a table in the main schema of the subclasses of this class.
protected void	onSchemaCacheRefreshed() Method invoked when schemas have been refreshed using AbstractDataContext.refreshSchemas().

Methods inherited from class org.apache.metamodel.elasticsearch.AbstractElasticSearchDataContext

getIndexName, getMainSchema, getMainSchemaName, limitMaxRowsIsSet, sortTables

Methods inherited from class org.apache.metamodel.QueryPostprocessDataContext

addConverter, executeQuery, getDefaultSchemaName, getSchemaByNameInternal, getSchemaNamesInternal, isMainSchemaTable, isScalarFunctionMaterialized, materializeFromItem, materializeMainSchemaTable, materializeMainSchemaTableSelect, materializeTable

Methods inherited from class org.apache.metamodel.AbstractDataContext

compileQuery, executeQuery, executeQuery, getColumnByQualifiedLabel, getDefaultSchema, getSchemaByName, getSchemaNames, getSchemas, getTableByQualifiedLabel, isQualifiedPathDelim, parseQuery, query, refreshSchemas

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apache.metamodel.DataContext

compileQuery, executeQuery, executeQuery, executeQuery, getColumnByQualifiedLabel, getDefaultSchema, getSchemaByName, getSchemaNames, getSchemas, getTableByQualifiedLabel, parseQuery, query, refreshSchemas

Constructor Detail

ElasticSearchDataContext

public ElasticSearchDataContext(org.elasticsearch.client.Client client,
                                String indexName,
                                SimpleTableDef... tableDefinitions)

Constructs a ElasticSearchDataContext. This constructor accepts a custom array of SimpleTableDefs which allows the user to define his own view on the indexes in the engine.

Parameters:

client - the ElasticSearch client

indexName - the name of the ElasticSearch index to represent

tableDefinitions - an array of SimpleTableDefs, which define the table and column model of the ElasticSearch index.

ElasticSearchDataContext

public ElasticSearchDataContext(org.elasticsearch.client.Client client,
                                String indexName)

Constructs a ElasticSearchDataContext and automatically detects the schema structure/view on all indexes (see detectTable(ClusterState, String, String)).

Parameters:

client - the ElasticSearch client

indexName - the name of the ElasticSearch index to represent

Method Detail

detectSchema

protected SimpleTableDef[] detectSchema()

Description copied from class: AbstractElasticSearchDataContext

Performs an analysis of the available indexes in an ElasticSearch cluster instance and detects the elasticsearch types structure based on the metadata provided by the ElasticSearch java client.

Specified by:

detectSchema in class AbstractElasticSearchDataContext

Returns:

a mutable schema instance, useful for further fine tuning by the user.

detectTable

public static SimpleTableDef detectTable(org.elasticsearch.cluster.ClusterState cs,
                                         String indexName,
                                         String documentType)
                                  throws Exception

Performs an analysis of an available index type in an ElasticSearch Client client and tries to detect the index structure based on the metadata provided by the java client.

Parameters:

cs - the ElasticSearch cluster

indexName - the name of the index

documentType - the name of the index type

Returns:

a table definition for ElasticSearch.

Throws:

Exception

materializeMainSchemaTable

protected DataSet materializeMainSchemaTable(Table table,
                                             List<SelectItem> selectItems,
                                             List<FilterItem> whereItems,
                                             int firstRow,
                                             int maxRows)

Description copied from class: QueryPostprocessDataContext

Execute a simple one-table query against a table in the main schema of the subclasses of this class. This default implementation will delegate to QueryPostprocessDataContext.materializeMainSchemaTable(Table, List, int, int) and apply WHERE item filtering afterwards.

Overrides:

materializeMainSchemaTable in class QueryPostprocessDataContext

Returns:

materializeMainSchemaTable

protected DataSet materializeMainSchemaTable(Table table,
                                             List<Column> columns,
                                             int maxRows)

Description copied from class: QueryPostprocessDataContext

Executes a simple one-table query against a table in the main schema of the subclasses of this class.

Specified by:

materializeMainSchemaTable in class QueryPostprocessDataContext

Parameters:

table - the table to query

columns - the columns of the table to query

maxRows - the maximum amount of rows needed or -1 if all rows are wanted.

Returns:

a dataset with the raw table/column content.

executePrimaryKeyLookupQuery

protected Row executePrimaryKeyLookupQuery(Table table,
                                           List<SelectItem> selectItems,
                                           Column primaryKeyColumn,
                                           Object keyValue)

Description copied from class: QueryPostprocessDataContext

Executes a query which obtains a row by primary key (as defined by Column.isPrimaryKey()). This method is provided to allow subclasses to optimize lookup queries since they are quite common and often a datastore can retrieve the row using some specialized means which is much more performant than scanning all records manually.

Overrides:

executePrimaryKeyLookupQuery in class QueryPostprocessDataContext

Parameters:

table - the table on which the lookup is requested.

selectItems - the items to select from the lookup query.

primaryKeyColumn - the column that is the primary key

keyValue - the primary key value that is specified in the lookup query.

Returns:

the row if the particular table, or null if not available.

executeCountQuery

protected Number executeCountQuery(Table table,
                                   List<FilterItem> whereItems,
                                   boolean functionApproximationAllowed)

Description copied from class: QueryPostprocessDataContext

Executes a simple count query, if possible. This method is provided to allow subclasses to optimize count queries since they are quite common and often a datastore can retrieve the count using some specialized means which is much more performant than counting all records manually.

Overrides:

executeCountQuery in class QueryPostprocessDataContext

Parameters:

table - the table on which the count is requested.

whereItems - a (sometimes empty) list of WHERE items.

functionApproximationAllowed - whether approximation is allowed or not.

Returns:

the count of the particular table, or null if not available.

executeUpdate

public UpdateSummary executeUpdate(UpdateScript update)

Description copied from interface: UpdateableDataContext

Submits an UpdateScript for execution on the DataContext. Since implementations of the DataContext vary quite a lot, there is no golden rule as to how an update script will be executed. But the implementors should strive towards handling an UpdateScript as a single transactional change to the data store.

Parameters:

update - the update script to execute

Returns:

a summary of the updates performed

onSchemaCacheRefreshed

protected void onSchemaCacheRefreshed()

Description copied from class: AbstractDataContext

Method invoked when schemas have been refreshed using AbstractDataContext.refreshSchemas(). Can be overridden to add callback functionality in subclasses.

Overrides:

onSchemaCacheRefreshed in class AbstractDataContext

getElasticSearchClient

public org.elasticsearch.client.Client getElasticSearchClient()

Gets the Client that this DataContext is wrapping.

Returns: