Managing Indices
Learn how to create, update and delete indices with CBElasticsearch
Elasticsearch documents are stored in an "index", with the document structure defined by a "mapping". An Elasticsearch index is a JSON document store, and the mapping is a JSON configuration which defines the data type Elasticsearch should use for each document field.
By default, Elasticsearch will dynamically generate these index mapping when a document is saved to the index. See Dynamic Mappings in Elasticsearch for more details.
To retrieve a list of all indices on the connected cluster, use the client
getIndices method:var indexMap = getInstance( "[email protected]" ).getIndices();
This will return a struct of all indices ( with the names as keys ), which will provide additional information on each index, such as:
- Any assigned aliases
- The number of documents in the index
- The size of the storage space used for the index in bytes
The
IndexBuilder model assists with the creation and mapping of indices. Mappings define the allowable data types within your documents and allow for better and more accurate search aggregations. Let's say we have a book model that we intend to make searchable via a bookshop index. Let's go ahead and create the index using the IndexBuilder:var indexBuilder = getInstance( "[email protected]" ).new( "bookshop" ).save();
This will create an empty index which we can begin populating with documents.
To avoid the inherent troubles with dynamic mappings, you can define an explicit mapping using the
properties argument:getInstance( "[email protected]" )
.new(
name = "bookshop",
properties = {
"title" : { "type" : "text" },
"summary" : { "type" : "text" },
"description" : { "type" : "text" },
// denotes a nested struct with additional keys
"author" : { "type" : "object" },
// date with specific format type
"publishDate" : {
"type" : "date",
// our format will be = yyyy-mm-dd
"format" = "strict_date"
},
"edition" : { "type" : "integer" },
"ISBN" : { "type" : "integer" }
}
)
.save();
While it is not required that you explicitly define an index mapping, it is highly recommended since Elasticsearch's assumptions about the incoming document data may not always be correct. This leads to issues where the Elasticsearch-generated mapping is wrong and prevents further data from being indexed if it does not match the expected data type.
In the previous examples, we've created the index and mapping from the IndexBuilder itself. If we wish, we could instead pass the
IndexBuilder object to the [email protected] instance's applyIndex( required IndexBuilder indexBuilder ) method:var myNewIndex = indexBuilder.new( "bookshop" )
.populate( getInstance( "[email protected]" ).getConfig() );
getInstance( "[email protected]" ).applyIndex( myNewIndex );
So far we've passed a simple struct of field mappings in to the index properties. If we wanted to add additional settings or configure replicas and shards, we could pass a more comprehensive struct, including a range of settings to the
new() method to do so:indexBuilder.new(
"bookshop",
{
"settings" : {
"number_of_shards" : 10,
"number_of_replicas" : 2,
"auto_expand_replicas" : true,
"shard.check_on_startup" : "checksum"
},
"mappings" : {
"properties" : {
"title" : { "type" : "text" },
"summary" : { "type" : "text" },
"description" : { "type" : "text" },
// denotes a nested struct with additional keys
"author" : { "type" : "object" },
// date with specific format type
"publishDate" : {
"type" : "date",
// our format will be = yyyy-mm-dd
"format" : "strict_date"
},
"edition" : { "type" : "integer" },
"ISBN" : { "type" : "integer" }
}
}
}
);
The
IndexBuilder model also provides a patch() convenience method for updating the mapping or settings on an index:indexBuilder.patch(
index = "bookshop",
settings = {
"number_of_shards" : 10,
"number_of_replicas" : 2,
"auto_expand_replicas" : true,
"shard.check_on_startup": "checksum"
}
);
Here's a quick example of using
indexBuilder.patch() to add two new fields to an existing reviews index:indexBuilder.patch(
index = "reviews",
properties = {
"authorName" : { "type" : "text" },
"helpfulRating": { "type" : "integer" }
}
);
To retreive a list of all settings for an index you may use the
getSettings method on the client.var indexSettings = getInstance( "[email protected]" ).getSettings( "bookshop" )
To retreive a list of the configured mappings for an index you may use the
getMappings method on the client.var mappings = getInstance( "[email protected]" ).getMappings( "reviews" );
On occasion, you may need to ensure the index is updated in real time (immediately and synchronously). This can be done via the
refreshIndex() client method:var mappings = getInstance( "[email protected]" ).refreshIndex( "reviews" );
You can refresh multiple indices at once:
var mappings = getInstance( "[email protected]" ).refreshIndex( [ "reviews", "books" ] );
// OR
var mappings = getInstance( "[email protected]" ).refreshIndex( "reviews,books" );
as well as pass supported query parameters to the refresh endpoint. This can be useful when using wildcards in the index/alias names:
var mappings = getInstance( "[email protected]" ).refreshIndex(
[ "reviews", "book*" ],
{ "ignore_unavailable" : true }
);
To retrieve statistics on an index, use the
getIndexStats() method:var mappings = getInstance( "[email protected]" ).getIndexStats( "reviews" );
You can retrieve particular statistics metrics:
var mappings = getInstance( "[email protected]" )
.getIndexStats( "reviews", [ "indexing", "search" ] );
Or all metrics:
var mappings = getInstance( "[email protected]" )
.getIndexStats( "reviews", [ "_all" ] );
You can even retrieve all metrics on all indices by skipping the
indexName parameter entirely:var mappings = getInstance( "[email protected]" ).getIndexStats();
Finally, you can pass a struct of parameters to fine-tune the statistics result:
var mappings = getInstance( "[email protected]" )
.getIndexStats(
"reviews",
[ "_all" ],
{ "level" : "shards", "fields" : "title,createdTime" }
);
Elasticsearch allows mapping runtime fields, which are fields calculated at search time and returned in the
"fields" array.var script = getInstance( "[email protected]" )
.formatToPainless("
if( doc['summary'].value.contains('love') ){ emit('😍');}
if( doc['summary'].value.contains('great') ){ emit('🚀');}
if( doc['summary'].value.contains('hate') ){ emit('😡');}
if( doc['summary'].value.contains('broke') ){ emit('💔');}
");
getInstance( "[email protected]" )
.patch( "reviews", {
"mappings" : {
"runtime" : {
"summarized_emotions" : {
"type" : "text",
"script" : {
"source" : script
}
}
}
}
} );
This
summarized_emotions field can then be retrieved during a search to display an array of emotions matching the review summary.All good things must come to an end, eh? You can use
Client.deleteIndex() to delete an existing index:getInstance( "[email protected]" ).deleteIndex( "reviews" )
Or you can use
IndexBuilder.delete():IndexBuilder.new( "reviews" ).delete();
Deprecation notice: Custom index types are deprecated since Elasticsearch v7.0, and should no longer be used. Only a single type will be accepted in future releases.
Last modified 1mo ago