[LUCENE-7704] SysnonymGraphFilter doesn't respect ignoreCase parameter

• Type: Bug
• Status: Resolved
• Priority: Minor
• Resolution: Fixed
• Affects Version/s: 6.4.1
• Fix Version/s: 6.5, 7.0
• Component/s: modules/analysis
• Lucene Fields: New
• Assignee: Unassigned
• Reporter: Sebastian Yonekura Baeza
• Created: 22/Feb/17 18:23
• Updated: 28/Nov/24 21:04
• Resolved: 23/Feb/17 14:05

Description

Hi, it seems that SynonymGraphFilter doesn't respect ignoreCase parameter. In particular this test doesn't pass:

UppercaseSynonymMapTest.java

package com.mapcity.suggest.lucene;

import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.analysis.TokenStream;
import org.apache.lucene.analysis.Tokenizer;
import org.apache.lucene.analysis.core.WhitespaceTokenizer;
import org.apache.lucene.analysis.synonym.SynonymGraphFilter;
import org.apache.lucene.analysis.synonym.SynonymMap;
import org.apache.lucene.util.CharsRef;
import org.apache.lucene.util.CharsRefBuilder;
import org.junit.Test;

import java.io.IOException;

import static org.apache.lucene.analysis.BaseTokenStreamTestCase.assertTokenStreamContents;

/**
 * @author Sebastian Yonekura
 *         Created on 22-02-17
 */
public class UppercaseSynonymMapTest {

    @Test
    public void analyzerTest01() throws IOException {
        // This passes
        testAssertMapping("word", "synonym");
        // this one not
        testAssertMapping("word".toUpperCase(), "synonym");
    }

    private void testAssertMapping(String inputString, String outputString) throws IOException {
        SynonymMap.Builder builder = new SynonymMap.Builder(false);
        CharsRef input = SynonymMap.Builder.join(inputString.split(" "), new CharsRefBuilder());
        CharsRef output = SynonymMap.Builder.join(outputString.split(" "), new CharsRefBuilder());
        builder.add(input, output, true);
        Analyzer analyzer = new CustomAnalyzer(builder.build());
        TokenStream tokenStream = analyzer.tokenStream("field", inputString);
        assertTokenStreamContents(tokenStream, new String[]{
                outputString, inputString
        });
    }

    static class CustomAnalyzer extends Analyzer {
        private SynonymMap synonymMap;

        CustomAnalyzer(SynonymMap synonymMap) {
            this.synonymMap = synonymMap;
        }

        @Override
        protected TokenStreamComponents createComponents(String s) {
            Tokenizer tokenizer = new WhitespaceTokenizer();
            TokenStream tokenStream = new SynonymGraphFilter(tokenizer, synonymMap, true); // Ignore case True
            return new TokenStreamComponents(tokenizer, tokenStream);
        }
    }
}