Skip to content

Commit

Permalink
feat: constant keyword field (opensearch-project#12285)
Browse files Browse the repository at this point in the history
Constant keyword fields behave similarly to regular keyword fields, except that they are 
defined only in the index mapping, and all documents in the index appear to have the same 
value for the constant keyword field.

---------

Signed-off-by: Mohammad Hasnain <hasnain2808@gmail.com>
  • Loading branch information
hasnain2808 authored Apr 1, 2024
1 parent 62776d1 commit 1ec49bd
Show file tree
Hide file tree
Showing 7 changed files with 409 additions and 1 deletion.
1 change: 1 addition & 0 deletions CHANGELOG.md
Original file line number Diff line number Diff line change
Expand Up @@ -103,6 +103,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),

## [Unreleased 2.x]
### Added
- Constant Keyword Field ([#12285](https://github.com/opensearch-project/OpenSearch/pull/12285))
- Convert ingest processor supports ip type ([#12818](https://github.com/opensearch-project/OpenSearch/pull/12818))
- Add a counter to node stat api to track shard going from idle to non-idle ([#12768](https://github.com/opensearch-project/OpenSearch/pull/12768))
- Allow setting KEYSTORE_PASSWORD through env variable ([#12865](https://github.com/opensearch-project/OpenSearch/pull/12865))
Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,191 @@
/*
* SPDX-License-Identifier: Apache-2.0
*
* The OpenSearch Contributors require contributions made to
* this file be licensed under the Apache-2.0 license or a
* compatible open source license.
*/

package org.opensearch.index.mapper;

import org.apache.lucene.search.MatchAllDocsQuery;
import org.apache.lucene.search.Query;
import org.opensearch.OpenSearchParseException;
import org.opensearch.common.annotation.PublicApi;
import org.opensearch.common.regex.Regex;
import org.opensearch.index.fielddata.IndexFieldData;
import org.opensearch.index.fielddata.plain.ConstantIndexFieldData;
import org.opensearch.index.query.QueryShardContext;
import org.opensearch.search.aggregations.support.CoreValuesSourceType;
import org.opensearch.search.lookup.SearchLookup;

import java.io.IOException;
import java.util.Arrays;
import java.util.Collections;
import java.util.List;
import java.util.Map;
import java.util.function.Supplier;

/**
* Index specific field mapper
*
* @opensearch.api
*/
@PublicApi(since = "2.14.0")
public class ConstantKeywordFieldMapper extends ParametrizedFieldMapper {

public static final String CONTENT_TYPE = "constant_keyword";

private static final String valuePropertyName = "value";

/**
* A {@link Mapper.TypeParser} for the constant keyword field.
*
* @opensearch.internal
*/
public static class TypeParser implements Mapper.TypeParser {
@Override
public Mapper.Builder parse(String name, Map<String, Object> node, ParserContext parserContext) throws MapperParsingException {
if (!node.containsKey(valuePropertyName)) {
throw new OpenSearchParseException("Field [" + name + "] is missing required parameter [value]");
}
Object value = node.remove(valuePropertyName);
if (!(value instanceof String)) {
throw new OpenSearchParseException("Field [" + name + "] is expected to be a string value");
}
return new Builder(name, (String) value);
}
}

private static ConstantKeywordFieldMapper toType(FieldMapper in) {
return (ConstantKeywordFieldMapper) in;
}

/**
* Builder for the binary field mapper
*
* @opensearch.internal
*/
public static class Builder extends ParametrizedFieldMapper.Builder {

private final Parameter<String> value;

public Builder(String name, String value) {
super(name);
this.value = Parameter.stringParam(valuePropertyName, false, m -> toType(m).value, value);
}

@Override
public List<Parameter<?>> getParameters() {
return Arrays.asList(value);
}

@Override
public ConstantKeywordFieldMapper build(BuilderContext context) {
return new ConstantKeywordFieldMapper(
name,
new ConstantKeywordFieldMapper.ConstantKeywordFieldType(buildFullName(context), value.getValue()),
multiFieldsBuilder.build(this, context),
copyTo.build(),
this
);
}
}

/**
* Field type for Index field mapper
*
* @opensearch.internal
*/
@PublicApi(since = "2.14.0")
protected static final class ConstantKeywordFieldType extends ConstantFieldType {

protected final String value;

public ConstantKeywordFieldType(String name, String value) {
super(name, Collections.emptyMap());
this.value = value;
}

@Override
public String typeName() {
return CONTENT_TYPE;
}

@Override
protected boolean matches(String pattern, boolean caseInsensitive, QueryShardContext context) {
return Regex.simpleMatch(pattern, value, caseInsensitive);
}

@Override
public Query existsQuery(QueryShardContext context) {
return new MatchAllDocsQuery();
}

@Override
public IndexFieldData.Builder fielddataBuilder(String fullyQualifiedIndexName, Supplier<SearchLookup> searchLookup) {
return new ConstantIndexFieldData.Builder(fullyQualifiedIndexName, name(), CoreValuesSourceType.BYTES);
}

@Override
public ValueFetcher valueFetcher(QueryShardContext context, SearchLookup searchLookup, String format) {
if (format != null) {
throw new IllegalArgumentException("Field [" + name() + "] of type [" + typeName() + "] doesn't " + "support formats.");
}

return new SourceValueFetcher(name(), context) {
@Override
protected Object parseSourceValue(Object value) {
String keywordValue = value.toString();
return Collections.singletonList(keywordValue);
}
};
}
}

private final String value;

protected ConstantKeywordFieldMapper(
String simpleName,
MappedFieldType mappedFieldType,
MultiFields multiFields,
CopyTo copyTo,
ConstantKeywordFieldMapper.Builder builder
) {
super(simpleName, mappedFieldType, multiFields, copyTo);
this.value = builder.value.getValue();
}

public ParametrizedFieldMapper.Builder getMergeBuilder() {
return new ConstantKeywordFieldMapper.Builder(simpleName(), this.value).init(this);
}

@Override
protected void parseCreateField(ParseContext context) throws IOException {

final String value;
if (context.externalValueSet()) {
value = context.externalValue().toString();
} else {
value = context.parser().textOrNull();
}
if (value == null) {
throw new IllegalArgumentException("constant keyword field [" + name() + "] must have a value");
}

if (!value.equals(fieldType().value)) {
throw new IllegalArgumentException("constant keyword field [" + name() + "] must have a value of [" + this.value + "]");
}

}

@Override
public ConstantKeywordFieldMapper.ConstantKeywordFieldType fieldType() {
return (ConstantKeywordFieldMapper.ConstantKeywordFieldType) super.fieldType();
}

@Override
protected String contentType() {
return CONTENT_TYPE;
}
}
Original file line number Diff line number Diff line change
Expand Up @@ -46,6 +46,7 @@
import org.opensearch.index.mapper.BinaryFieldMapper;
import org.opensearch.index.mapper.BooleanFieldMapper;
import org.opensearch.index.mapper.CompletionFieldMapper;
import org.opensearch.index.mapper.ConstantKeywordFieldMapper;
import org.opensearch.index.mapper.DataStreamFieldMapper;
import org.opensearch.index.mapper.DateFieldMapper;
import org.opensearch.index.mapper.DocCountFieldMapper;
Expand Down Expand Up @@ -168,6 +169,7 @@ public static Map<String, Mapper.TypeParser> getMappers(List<MapperPlugin> mappe
mappers.put(FieldAliasMapper.CONTENT_TYPE, new FieldAliasMapper.TypeParser());
mappers.put(GeoPointFieldMapper.CONTENT_TYPE, new GeoPointFieldMapper.TypeParser());
mappers.put(FlatObjectFieldMapper.CONTENT_TYPE, FlatObjectFieldMapper.PARSER);
mappers.put(ConstantKeywordFieldMapper.CONTENT_TYPE, new ConstantKeywordFieldMapper.TypeParser());

for (MapperPlugin mapperPlugin : mapperPlugins) {
for (Map.Entry<String, Mapper.TypeParser> entry : mapperPlugin.getMappers().entrySet()) {
Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,114 @@
/*
* SPDX-License-Identifier: Apache-2.0
*
* The OpenSearch Contributors require contributions made to
* this file be licensed under the Apache-2.0 license or a
* compatible open source license.
*/

package org.opensearch.index.mapper;

import org.apache.lucene.index.IndexableField;
import org.opensearch.OpenSearchParseException;
import org.opensearch.common.CheckedConsumer;
import org.opensearch.common.compress.CompressedXContent;
import org.opensearch.common.xcontent.XContentFactory;
import org.opensearch.common.xcontent.json.JsonXContent;
import org.opensearch.core.common.bytes.BytesReference;
import org.opensearch.core.xcontent.MediaTypeRegistry;
import org.opensearch.core.xcontent.XContentBuilder;
import org.opensearch.index.IndexService;
import org.opensearch.plugins.Plugin;
import org.opensearch.test.InternalSettingsPlugin;
import org.opensearch.test.OpenSearchSingleNodeTestCase;
import org.junit.Before;

import java.io.IOException;
import java.util.Collection;

import static org.hamcrest.Matchers.containsString;

public class ConstantKeywordFieldMapperTests extends OpenSearchSingleNodeTestCase {

private IndexService indexService;
private DocumentMapperParser parser;

@Override
protected Collection<Class<? extends Plugin>> getPlugins() {
return pluginList(InternalSettingsPlugin.class);
}

@Before
public void setup() {
indexService = createIndex("test");
parser = indexService.mapperService().documentMapperParser();
}

public void testDefaultDisabledIndexMapper() throws Exception {

XContentBuilder mapping = XContentFactory.jsonBuilder()
.startObject()
.startObject("type")
.startObject("properties")
.startObject("field")
.field("type", "constant_keyword")
.field("value", "default_value")
.endObject()
.startObject("field2")
.field("type", "keyword")
.endObject();
mapping = mapping.endObject().endObject().endObject();
DocumentMapper mapper = parser.parse("type", new CompressedXContent(mapping.toString()));

MapperParsingException e = expectThrows(MapperParsingException.class, () -> mapper.parse(source(b -> {
b.field("field", "sdf");
b.field("field2", "szdfvsddf");
})));
assertThat(
e.getMessage(),
containsString(
"failed to parse field [field] of type [constant_keyword] in document with id '1'. Preview of field's value: 'sdf'"
)
);

final ParsedDocument doc = mapper.parse(source(b -> {
b.field("field", "default_value");
b.field("field2", "field_2_value");
}));

final IndexableField field = doc.rootDoc().getField("field");

// constantKeywordField should not be stored
assertNull(field);
}

public void testMissingDefaultIndexMapper() throws Exception {

final XContentBuilder mapping = XContentFactory.jsonBuilder()
.startObject()
.startObject("type")
.startObject("properties")
.startObject("field")
.field("type", "constant_keyword")
.endObject()
.startObject("field2")
.field("type", "keyword")
.endObject()
.endObject()
.endObject()
.endObject();

OpenSearchParseException e = expectThrows(
OpenSearchParseException.class,
() -> parser.parse("type", new CompressedXContent(mapping.toString()))
);
assertThat(e.getMessage(), containsString("Field [field] is missing required parameter [value]"));
}

private final SourceToParse source(CheckedConsumer<XContentBuilder, IOException> build) throws IOException {
XContentBuilder builder = JsonXContent.contentBuilder().startObject();
build.accept(builder);
builder.endObject();
return new SourceToParse("test", "1", BytesReference.bytes(builder), MediaTypeRegistry.JSON);
}
}
Loading

0 comments on commit 1ec49bd

Please sign in to comment.