Skip to content

Commit

Permalink
token_count type : add an option to count tokens (fix #23227) (#24175)
Browse files Browse the repository at this point in the history 
Add option "enable_position_increments" with default value true.
If option is set to false, indexed value is the number of tokens
(not position increments count)
  • Loading branch information
@fbaligand @jimczi
fbaligand authored and jimczi committed Apr 20, 2017
1 parent 525101b commit 4a45579
Show file tree
Hide file tree
Showing 4 changed files with 108 additions and 44 deletions.
53 changes: 43 additions & 10 deletions core/src/main/java/org/elasticsearch/index/mapper/TokenCountFieldMapper.java
View file
Original file line number Diff line number Diff line change
Expand Up @@ -22,10 +22,8 @@
import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.analysis.TokenStream;
import org.apache.lucene.analysis.tokenattributes.PositionIncrementAttribute;
import org.apache.lucene.document.Field;
import org.apache.lucene.index.IndexOptions;
import org.apache.lucene.index.IndexableField;
import org.elasticsearch.Version;
import org.elasticsearch.common.settings.Settings;
import org.elasticsearch.common.xcontent.XContentBuilder;
import org.elasticsearch.index.analysis.NamedAnalyzer;
Expand All @@ -36,6 +34,7 @@
import java.util.Map;

import static org.elasticsearch.common.xcontent.support.XContentMapValues.nodeIntegerValue;
import static org.elasticsearch.common.xcontent.support.XContentMapValues.nodeBooleanValue;
import static org.elasticsearch.index.mapper.TypeParsers.parseField;

/**
Expand All @@ -47,10 +46,12 @@ public class TokenCountFieldMapper extends FieldMapper {

public static class Defaults {
public static final MappedFieldType FIELD_TYPE = new NumberFieldMapper.NumberFieldType(NumberFieldMapper.NumberType.INTEGER);
public static final boolean DEFAULT_POSITION_INCREMENTS = true;
}

public static class Builder extends FieldMapper.Builder<Builder, TokenCountFieldMapper> {
private NamedAnalyzer analyzer;
private boolean enablePositionIncrements = Defaults.DEFAULT_POSITION_INCREMENTS;

public Builder(String name) {
super(name, Defaults.FIELD_TYPE, Defaults.FIELD_TYPE);
Expand All @@ -66,18 +67,26 @@ public NamedAnalyzer analyzer() {
return analyzer;
}

public Builder enablePositionIncrements(boolean enablePositionIncrements) {
this.enablePositionIncrements = enablePositionIncrements;
return this;
}

public boolean enablePositionIncrements() {
return enablePositionIncrements;
}

@Override
public TokenCountFieldMapper build(BuilderContext context) {
setupFieldType(context);
return new TokenCountFieldMapper(name, fieldType, defaultFieldType,
context.indexSettings(), analyzer, multiFieldsBuilder.build(this, context), copyTo);
context.indexSettings(), analyzer, enablePositionIncrements, multiFieldsBuilder.build(this, context), copyTo);
}
}

public static class TypeParser implements Mapper.TypeParser {
@Override
@SuppressWarnings("unchecked")
public Mapper.Builder parse(String name, Map<String, Object> node, ParserContext parserContext) throws MapperParsingException {
public Mapper.Builder<?,?> parse(String name, Map<String, Object> node, ParserContext parserContext) throws MapperParsingException {
TokenCountFieldMapper.Builder builder = new TokenCountFieldMapper.Builder(name);
for (Iterator<Map.Entry<String, Object>> iterator = node.entrySet().iterator(); iterator.hasNext();) {
Map.Entry<String, Object> entry = iterator.next();
Expand All @@ -93,6 +102,9 @@ public Mapper.Builder parse(String name, Map<String, Object> node, ParserContext
}
builder.analyzer(analyzer);
iterator.remove();
} else if (propName.equals("enable_position_increments")) {
builder.enablePositionIncrements(nodeBooleanValue(propNode));
iterator.remove();
}
}
parseField(builder, name, node, parserContext);
Expand All @@ -104,11 +116,13 @@ public Mapper.Builder parse(String name, Map<String, Object> node, ParserContext
}

private NamedAnalyzer analyzer;
private boolean enablePositionIncrements;

protected TokenCountFieldMapper(String simpleName, MappedFieldType fieldType, MappedFieldType defaultFieldType,
Settings indexSettings, NamedAnalyzer analyzer, MultiFields multiFields, CopyTo copyTo) {
Settings indexSettings, NamedAnalyzer analyzer, boolean enablePositionIncrements, MultiFields multiFields, CopyTo copyTo) {
super(simpleName, fieldType, defaultFieldType, indexSettings, multiFields, copyTo);
this.analyzer = analyzer;
this.enablePositionIncrements = enablePositionIncrements;
}

@Override
Expand All @@ -124,7 +138,7 @@ protected void parseCreateField(ParseContext context, List<IndexableField> field
if (value == null) {
tokenCount = (Integer) fieldType().nullValue();
} else {
tokenCount = countPositions(analyzer, name(), value);
tokenCount = countPositions(analyzer, name(), value, enablePositionIncrements);
}

boolean indexed = fieldType().indexOptions() != IndexOptions.NONE;
Expand All @@ -138,19 +152,26 @@ protected void parseCreateField(ParseContext context, List<IndexableField> field
* @param analyzer analyzer to create token stream
* @param fieldName field name to pass to analyzer
* @param fieldValue field value to pass to analyzer
* @param enablePositionIncrements should we count position increments ?
* @return number of position increments in a token stream
* @throws IOException if tokenStream throws it
*/
static int countPositions(Analyzer analyzer, String fieldName, String fieldValue) throws IOException {
static int countPositions(Analyzer analyzer, String fieldName, String fieldValue, boolean enablePositionIncrements) throws IOException {
try (TokenStream tokenStream = analyzer.tokenStream(fieldName, fieldValue)) {
int count = 0;
PositionIncrementAttribute position = tokenStream.addAttribute(PositionIncrementAttribute.class);
tokenStream.reset();
while (tokenStream.incrementToken()) {
count += position.getPositionIncrement();
if (enablePositionIncrements) {
count += position.getPositionIncrement();
} else {
count += Math.min(1, position.getPositionIncrement());
}
}
tokenStream.end();
count += position.getPositionIncrement();
if (enablePositionIncrements) {
count += position.getPositionIncrement();
}
return count;
}
}
Expand All @@ -163,6 +184,14 @@ public String analyzer() {
return analyzer.name();
}

/**
* Indicates if position increments are counted.
* @return <code>true</code> if position increments are counted
*/
public boolean enablePositionIncrements() {
return enablePositionIncrements;
}

@Override
protected String contentType() {
return CONTENT_TYPE;
Expand All @@ -172,12 +201,16 @@ protected String contentType() {
protected void doMerge(Mapper mergeWith, boolean updateAllTypes) {
super.doMerge(mergeWith, updateAllTypes);
this.analyzer = ((TokenCountFieldMapper) mergeWith).analyzer;
this.enablePositionIncrements = ((TokenCountFieldMapper) mergeWith).enablePositionIncrements;
}

@Override
protected void doXContentBody(XContentBuilder builder, boolean includeDefaults, Params params) throws IOException {
super.doXContentBody(builder, includeDefaults, params);
builder.field("analyzer", analyzer());
if (includeDefaults || enablePositionIncrements() != Defaults.DEFAULT_POSITION_INCREMENTS) {
builder.field("enable_position_increments", enablePositionIncrements());
}
}

}
35 changes: 24 additions & 11 deletions core/src/test/java/org/elasticsearch/index/mapper/TokenCountFieldMapperIntegrationIT.java
View file
Original file line number Diff line number Diff line change
Expand Up @@ -131,6 +131,12 @@ private void init() throws IOException {
.field("analyzer", "standard")
.field("doc_values", true)
.endObject()
.startObject("token_count_without_position_increments")
.field("type", "token_count")
.field("analyzer", "english")
.field("enable_position_increments", false)
.field("store", true)
.endObject()
.endObject()
.endObject()
.endObject()
Expand Down Expand Up @@ -169,6 +175,7 @@ private SearchRequestBuilder searchByNumericRange(int low, int high) {
private SearchRequestBuilder prepareSearch() {
SearchRequestBuilder request = client().prepareSearch("test").setTypes("test");
request.addStoredField("foo.token_count");
request.addStoredField("foo.token_count_without_position_increments");
if (loadCountedFields) {
request.addStoredField("foo");
}
Expand All @@ -186,32 +193,38 @@ private void assertSearchReturns(SearchResponse result, String... ids) {
for (SearchHit hit : result.getHits()) {
String id = hit.getId();
if (id.equals("single")) {
assertSearchHit(hit, 4);
assertSearchHit(hit, new int[]{4}, new int[]{4});
} else if (id.equals("bulk1")) {
assertSearchHit(hit, 3);
assertSearchHit(hit, new int[]{3}, new int[]{3});
} else if (id.equals("bulk2")) {
assertSearchHit(hit, 5);
assertSearchHit(hit, new int[]{5}, new int[]{4});
} else if (id.equals("multi")) {
assertSearchHit(hit, 2, 7);
assertSearchHit(hit, new int[]{2, 7}, new int[]{2, 7});
} else if (id.equals("multibulk1")) {
assertSearchHit(hit, 1, 8);
assertSearchHit(hit, new int[]{1, 8}, new int[]{1, 8});
} else if (id.equals("multibulk2")) {
assertSearchHit(hit, 6, 10);
assertSearchHit(hit, new int[]{6, 10}, new int[]{3, 9});
} else {
throw new ElasticsearchException("Unexpected response!");
}
}
}

private void assertSearchHit(SearchHit hit, int... termCounts) {
private void assertSearchHit(SearchHit hit, int[] standardTermCounts, int[] englishTermCounts) {
assertThat(hit.field("foo.token_count"), not(nullValue()));
assertThat(hit.field("foo.token_count").getValues().size(), equalTo(termCounts.length));
for (int i = 0; i < termCounts.length; i++) {
assertThat((Integer) hit.field("foo.token_count").getValues().get(i), equalTo(termCounts[i]));
assertThat(hit.field("foo.token_count").getValues().size(), equalTo(standardTermCounts.length));
for (int i = 0; i < standardTermCounts.length; i++) {
assertThat((Integer) hit.field("foo.token_count").getValues().get(i), equalTo(standardTermCounts[i]));
}

assertThat(hit.field("foo.token_count_without_position_increments"), not(nullValue()));
assertThat(hit.field("foo.token_count_without_position_increments").getValues().size(), equalTo(englishTermCounts.length));
for (int i = 0; i < englishTermCounts.length; i++) {
assertThat((Integer) hit.field("foo.token_count_without_position_increments").getValues().get(i), equalTo(englishTermCounts[i]));
}

if (loadCountedFields && storeCountedFields) {
assertThat(hit.field("foo").getValues().size(), equalTo(termCounts.length));
assertThat(hit.field("foo").getValues().size(), equalTo(standardTermCounts.length));
}
}
}
52 changes: 35 additions & 17 deletions core/src/test/java/org/elasticsearch/index/mapper/TokenCountFieldMapperTests.java
View file
Original file line number Diff line number Diff line change
Expand Up @@ -24,23 +24,18 @@
import org.apache.lucene.analysis.MockTokenizer;
import org.apache.lucene.analysis.Token;
import org.apache.lucene.analysis.TokenStream;
import org.elasticsearch.Version;
import org.elasticsearch.cluster.metadata.IndexMetaData;
import org.elasticsearch.common.compress.CompressedXContent;
import org.elasticsearch.common.settings.Settings;
import org.elasticsearch.common.xcontent.XContentFactory;
import org.elasticsearch.index.IndexService;
import org.elasticsearch.plugins.Plugin;
import org.elasticsearch.test.ESSingleNodeTestCase;
import org.elasticsearch.test.InternalSettingsPlugin;
import org.elasticsearch.test.VersionUtils;

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

import static com.carrotsearch.randomizedtesting.RandomizedTest.getRandom;
import static org.hamcrest.Matchers.containsString;
import static org.hamcrest.Matchers.equalTo;

Expand Down Expand Up @@ -80,26 +75,49 @@ public void testMerge() throws IOException {
assertThat(((TokenCountFieldMapper) stage2.mappers().smartNameFieldMapper("tc")).analyzer(), equalTo("standard"));
}

public void testCountPositions() throws IOException {
// We're looking to make sure that we:
Token t1 = new Token(); // Don't count tokens without an increment
/**
* When position increments are counted, we're looking to make sure that we:
- don't count tokens without an increment
- count normal tokens with one increment
- count funny tokens with more than one increment
- count the final token increments on the rare token streams that have them
*/
public void testCountPositionsWithIncrements() throws IOException {
Analyzer analyzer = createMockAnalyzer();
assertThat(TokenCountFieldMapper.countPositions(analyzer, "", "", true), equalTo(7));
}

/**
* When position increments are not counted (only positions are counted), we're looking to make sure that we:
- don't count tokens without an increment
- count normal tokens with one increment
- count funny tokens with more than one increment as only one
- don't count the final token increments on the rare token streams that have them
*/
public void testCountPositionsWithoutIncrements() throws IOException {
Analyzer analyzer = createMockAnalyzer();
assertThat(TokenCountFieldMapper.countPositions(analyzer, "", "", false), equalTo(2));
}

private Analyzer createMockAnalyzer() {
Token t1 = new Token(); // Token without an increment
t1.setPositionIncrement(0);
Token t2 = new Token();
t2.setPositionIncrement(1); // Count normal tokens with one increment
t2.setPositionIncrement(1); // Normal token with one increment
Token t3 = new Token();
t2.setPositionIncrement(2); // Count funny tokens with more than one increment
int finalTokenIncrement = 4; // Count the final token increment on the rare token streams that have them
t2.setPositionIncrement(2); // Funny token with more than one increment
int finalTokenIncrement = 4; // Final token increment
Token[] tokens = new Token[] {t1, t2, t3};
Collections.shuffle(Arrays.asList(tokens), random());
final TokenStream tokenStream = new CannedTokenStream(finalTokenIncrement, 0, tokens);
// TODO: we have no CannedAnalyzer?
Analyzer analyzer = new Analyzer() {
@Override
public TokenStreamComponents createComponents(String fieldName) {
return new TokenStreamComponents(new MockTokenizer(), tokenStream);
}
};
assertThat(TokenCountFieldMapper.countPositions(analyzer, "", ""), equalTo(7));
@Override
public TokenStreamComponents createComponents(String fieldName) {
return new TokenStreamComponents(new MockTokenizer(), tokenStream);
}
};
return analyzer;
}

@Override
Expand Down
12 changes: 6 additions & 6 deletions docs/reference/mapping/types/token-count.asciidoc
View file
Original file line number Diff line number Diff line change
Expand Up @@ -48,12 +48,6 @@ GET my_index/_search
<2> The `name.length` field is a `token_count` <<multi-fields,multi-field>> which will index the number of tokens in the `name` field.
<3> This query matches only the document containing `Rachel Alice Williams`, as it contains three tokens.

[NOTE]
===================================================================
Technically the `token_count` type sums position increments rather than
counting tokens. This means that even if the analyzer filters out stop
words they are included in the count.
===================================================================

[[token-count-params]]
==== Parameters for `token_count` fields
Expand All @@ -68,6 +62,12 @@ The following parameters are accepted by `token_count` fields:
value. Required. For best performance, use an analyzer without token
filters.

`enable_position_increments`::

Indicates if position increments should be counted.
Set to `false` if you don't want to count tokens removed by analyzer filters (like <<analysis-stop-tokenfilter,`stop`>>).
Defaults to `true`.

<<mapping-boost,`boost`>>::

Mapping field-level query time boosting. Accepts a floating point number, defaults
Expand Down

0 comments on commit 4a45579

Please sign in to comment.