Make "german2" an alias for "german" snowball stemmer

modules/analysis-common/src/main/java/org/elasticsearch/analysis/common/StemmerTokenFilterFactory.java

@@ -48,9 +48,13 @@
 import org.apache.lucene.analysis.snowball.SnowballFilter;
 import org.apache.lucene.analysis.sv.SwedishLightStemFilter;
 import org.elasticsearch.common.Strings;
+import org.elasticsearch.common.logging.DeprecationCategory;
+import org.elasticsearch.common.logging.DeprecationLogger;
 import org.elasticsearch.common.settings.Settings;
 import org.elasticsearch.env.Environment;
 import org.elasticsearch.index.IndexSettings;
+import org.elasticsearch.index.IndexVersion;
+import org.elasticsearch.index.IndexVersions;
 import org.elasticsearch.index.analysis.AbstractTokenFilterFactory;
 import org.tartarus.snowball.ext.ArmenianStemmer;
 import org.tartarus.snowball.ext.BasqueStemmer;
@@ -82,10 +86,14 @@ public class StemmerTokenFilterFactory extends AbstractTokenFilterFactory {
 
     private static final TokenStream EMPTY_TOKEN_STREAM = new EmptyTokenStream();
 
-    private String language;
+    private final String language;
+    private final IndexVersion version;
+
+    private static final DeprecationLogger DEPRECATION_LOGGER = DeprecationLogger.getLogger(StemmerTokenFilterFactory.class);
 
     StemmerTokenFilterFactory(IndexSettings indexSettings, Environment environment, String name, Settings settings) throws IOException {
         super(name, settings);
+        this.version = indexSettings.getIndexVersionCreated();
         this.language = Strings.capitalize(settings.get("language", settings.get("name", "porter")));
         // check that we have a valid language by trying to create a TokenStream
         create(EMPTY_TOKEN_STREAM).close();
@@ -167,7 +175,12 @@ public TokenStream create(TokenStream tokenStream) {
             } else if ("german".equalsIgnoreCase(language)) {
                 return new SnowballFilter(tokenStream, new GermanStemmer());
             } else if ("german2".equalsIgnoreCase(language)) {
-                // TODO Lucene 10 upgrade: how about bw comp for users relying on german2 stemmer that is now folded into german stemmer?
+                DEPRECATION_LOGGER.critical(
+                    DeprecationCategory.ANALYSIS,
+                    "german2_stemmer_deprecation",
+                    "The 'german2' stemmer has been deprecated and folded into the 'german' Stemmer. "
+                        + "Replace all usages of 'german2' with 'german'."
+                );
                 return new SnowballFilter(tokenStream, new GermanStemmer());
             } else if ("light_german".equalsIgnoreCase(language) || "lightGerman".equalsIgnoreCase(language)) {
                 return new GermanLightStemFilter(tokenStream);

modules/analysis-common/src/test/java/org/elasticsearch/analysis/common/StemmerTokenFilterFactoryTests.java

@@ -8,6 +8,7 @@
  */
 package org.elasticsearch.analysis.common;
 
+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;
@@ -16,6 +17,7 @@
 import org.elasticsearch.common.settings.Settings;
 import org.elasticsearch.env.Environment;
 import org.elasticsearch.index.IndexVersion;
+import org.elasticsearch.index.IndexVersions;
 import org.elasticsearch.index.analysis.AnalysisTestsHelper;
 import org.elasticsearch.index.analysis.IndexAnalyzers;
 import org.elasticsearch.index.analysis.NamedAnalyzer;
@@ -103,4 +105,40 @@ public void testMultipleLanguagesThrowsException() throws IOException {
         );
         assertEquals("Invalid stemmer class specified: [english, light_english]", e.getMessage());
     }
+
+    public void testGermanAndGerman2Stemmer() throws IOException {
+        IndexVersion v = IndexVersionUtils.randomVersionBetween(random(), IndexVersions.UPGRADE_TO_LUCENE_10_0_0, IndexVersion.current());
+        Analyzer analyzer = createGermanStemmer("german", v);
+        assertAnalyzesTo(analyzer, "Buecher Bücher", new String[] { "Buch", "Buch" });
+
+        analyzer = createGermanStemmer("german2", v);
+        assertAnalyzesTo(analyzer, "Buecher Bücher", new String[] { "Buch", "Buch" });
+        assertWarnings(
+            "The 'german2' stemmer has been deprecated and folded into the 'german' Stemmer. "
+                + "Replace all usages of 'german2' with 'german'."
+        );
+    }
+
+    public Analyzer createGermanStemmer(String variant, IndexVersion v) throws IOException {
+
+        Settings settings = Settings.builder()
+            .put("index.analysis.filter.my_german.type", "stemmer")
+            .put("index.analysis.filter.my_german.language", variant)
+            .put("index.analysis.analyzer.my_german.tokenizer", "whitespace")
+            .put("index.analysis.analyzer.my_german.filter", "my_german")
+            .put(SETTING_VERSION_CREATED, v)
+            .put(Environment.PATH_HOME_SETTING.getKey(), createTempDir().toString())
+            .build();
+
+        ESTestCase.TestAnalysis analysis = AnalysisTestsHelper.createTestAnalysisFromSettings(settings, PLUGIN);
+        TokenFilterFactory tokenFilter = analysis.tokenFilter.get("my_german");
+        assertThat(tokenFilter, instanceOf(StemmerTokenFilterFactory.class));
+        Tokenizer tokenizer = new WhitespaceTokenizer();
+        tokenizer.setReader(new StringReader("Buecher oder Bücher"));
+        TokenStream create = tokenFilter.create(tokenizer);
+        assertThat(create, instanceOf(SnowballFilter.class));
+        IndexAnalyzers indexAnalyzers = analysis.indexAnalyzers;
+        NamedAnalyzer analyzer = indexAnalyzers.get("my_german");
+        return analyzer;
+    }
 }