Add auto syllabification for other languages (#30)

Add auto-syllabification for English, Polish and Spanish with possibility to easily add other languages if necessary.

.gitignore

@@ -2,3 +2,4 @@
 build/
 doc/
 /target/
+.idea

pom.xml

@@ -18,6 +18,7 @@
 		<junit.version>5.9.2</junit.version>
 		<mp3agic.version>0.8.1</mp3agic.version>
 		<richtextfx.version>0.11.0</richtextfx.version>
+		<hyphenator.version>1.0.0</hyphenator.version>
 	</properties>
 
 	<build>
@@ -282,6 +283,7 @@
 						<modulePath>${settings.localRepository}\org\openjfx\javafx-base\${javafx.version}</modulePath>
 						<modulePath>${settings.localRepository}\org\openjfx\javafx-graphics\${javafx.version}</modulePath>
 						<modulePath>${settings.localRepository}\org\controlsfx\controlsfx\${controlsfx.version}</modulePath>
+						<modulePath>${settings.localRepository}\io\github\nianna\hyphenator\${hyphenator.version}</modulePath>
 						<modulePath>${project.jars.outputDirectory}</modulePath>
 					</modulePaths>
 					<javaOptions>
@@ -330,6 +332,12 @@
 
 	<dependencies>
 
+		<dependency>
+			<groupId>io.github.nianna</groupId>
+			<artifactId>hyphenator</artifactId>
+			<version>${hyphenator.version}</version>
+		</dependency>
+
 		<dependency>
 			<groupId>com.googlecode.soundlibs</groupId>
 			<artifactId>jlayer</artifactId>

src/main/java/com/github/nianna/karedi/context/SyllabizerContext.java

@@ -1,32 +1,44 @@
 package com.github.nianna.karedi.context;
 
+import com.github.nianna.karedi.I18N;
 import com.github.nianna.karedi.song.Song;
 import com.github.nianna.karedi.song.tag.TagKey;
 import com.github.nianna.karedi.syllabizer.Syllabizer;
 import com.github.nianna.karedi.syllabizer.SyllabizerFactory;
+import com.github.nianna.karedi.syllabizer.SyllabizerInitializationFailedException;
 import com.github.nianna.karedi.util.Language;
 import javafx.beans.property.ReadOnlyObjectProperty;
 
-import java.util.HashMap;
-import java.util.Map;
 import java.util.Optional;
+import java.util.concurrent.ConcurrentHashMap;
+import java.util.concurrent.ConcurrentMap;
+import java.util.logging.Logger;
 
 public class SyllabizerContext {
 
-    private static final Map<Language, Syllabizer> SYLLABIZERS = new HashMap<>();
+    private static final Logger LOGGER = Logger.getLogger(SyllabizerContext.class.getName());
+
+    private static final ConcurrentMap<Language, Syllabizer> SYLLABIZERS = new ConcurrentHashMap<>();
 
     private final ReadOnlyObjectProperty<Song> activeSongProperty;
 
     SyllabizerContext(ActiveSongContext activeSongContext) {
         this.activeSongProperty = activeSongContext.activeSongProperty();
-        SyllabizerFactory.supportedLanguages()
-                .forEach(language -> SYLLABIZERS.put(language, SyllabizerFactory.createFor(language)));
     }
 
     public Optional<Syllabizer> findSyllabizer() {
         return Optional.of(activeSongProperty.get())
                 .flatMap(song -> song.getTagValue(TagKey.LANGUAGE))
                 .flatMap(Language::parse)
-                .map(SYLLABIZERS::get);
+                .map(language -> SYLLABIZERS.computeIfAbsent(language, this::createSyllabizer));
+    }
+
+    private Syllabizer createSyllabizer(Language language) {
+        try {
+            return SyllabizerFactory.createFor(language);
+        } catch (SyllabizerInitializationFailedException exception) {
+            LOGGER.severe(I18N.get("syllabizer.init_failed"));
+            return SyllabizerFactory.createNoopSyllabizer();
+        }
     }
 }

src/main/java/com/github/nianna/karedi/syllabizer/EnglishSyllabizer.java

@@ -0,0 +1,9 @@
+package com.github.nianna.karedi.syllabizer;
+
+class EnglishSyllabizer extends PatternSyllabizer {
+
+    EnglishSyllabizer() {
+        super("hyph_en_US.dic", ".*[aeiou].*|.+y");
+    }
+
+}

src/main/java/com/github/nianna/karedi/syllabizer/JapaneseSyllabizer.java

@@ -4,7 +4,7 @@
 import java.util.List;
 import java.util.Set;
 
-class JapaneseSyllabizer implements Syllabizer {
+class JapaneseSyllabizer extends Syllabizer {
 
     private static final Set<Character> CONSONANTS = Set.of(
             'b', 'c', 'd', 'f', 'g', 'h', 'j', 'k', 'l', 'm', 'n', 'p', 'q', 'r', 's', 't', 'v', 'w', 'x', 'y', 'z'
@@ -23,12 +23,7 @@ class JapaneseSyllabizer implements Syllabizer {
      * </p>
      */
     @Override
-    public List<String> syllabize(String input) {
-        String normalized = input.replaceAll("\\R", " ").stripTrailing();
-        return syllabizeNormalized(normalized);
-    }
-
-    private List<String> syllabizeNormalized(String input) {
+    protected List<String> syllabizeNormalized(String input) {
         List<String> results = new LinkedList<>();
         int nextSyllableStartIndex = 0;
         boolean canNextSyllableBeFinished = false;

src/main/java/com/github/nianna/karedi/syllabizer/NoopSyllabizer.java

@@ -0,0 +1,11 @@
+package com.github.nianna.karedi.syllabizer;
+
+import java.util.List;
+
+public class NoopSyllabizer extends Syllabizer {
+
+    @Override
+    protected List<String> syllabizeNormalized(String input) {
+        return List.of(" " + input);
+    }
+}

src/main/java/com/github/nianna/karedi/syllabizer/PatternSyllabizer.java

@@ -0,0 +1,36 @@
+package com.github.nianna.karedi.syllabizer;
+
+import com.github.nianna.karedi.util.ResourceUtils;
+import io.github.nianna.api.HyphenatedText;
+import io.github.nianna.api.Hyphenator;
+import io.github.nianna.api.HyphenatorProperties;
+
+import java.util.List;
+
+abstract class PatternSyllabizer extends Syllabizer {
+
+    private final Hyphenator hyphenator;
+
+    private final SyllablesSanitizer sanitizer;
+
+    PatternSyllabizer(String dictionaryName, String syllablesPattern) {
+        try {
+            List<String> patterns = ResourceUtils.readLines("/syllabizer/%s".formatted(dictionaryName));
+            HyphenatorProperties properties = new HyphenatorProperties(1, 1);
+            hyphenator = new Hyphenator(patterns, properties);
+            sanitizer = new SyllablesSanitizer(syllablesPattern);
+        } catch (Exception e) {
+            e.printStackTrace();
+            throw new SyllabizerInitializationFailedException(e);
+        }
+    }
+
+    @Override
+    public List<String> syllabizeNormalized(String input) {
+        HyphenatedText text = hyphenator.hyphenateText(input);
+        String separator = "\t";
+        String textToSplit = " " + text.read(separator + " ", separator);
+        return sanitizer.sanitize(textToSplit.split(separator));
+    }
+
+}

src/main/java/com/github/nianna/karedi/syllabizer/PolishSyllabizer.java

@@ -0,0 +1,9 @@
+package com.github.nianna.karedi.syllabizer;
+
+class PolishSyllabizer extends PatternSyllabizer {
+
+    PolishSyllabizer() {
+        super("hyph_pl_PL.dic", ".*[aąeęioóuy].*");
+    }
+
+}

src/main/java/com/github/nianna/karedi/syllabizer/SpanishSyllabizer.java

@@ -0,0 +1,9 @@
+package com.github.nianna.karedi.syllabizer;
+
+class SpanishSyllabizer extends PatternSyllabizer {
+
+    SpanishSyllabizer() {
+        super("hyph_es_ES.dic", ".*[aeiouáéíóú].*");
+    }
+
+}

src/main/java/com/github/nianna/karedi/syllabizer/Syllabizer.java

@@ -2,7 +2,15 @@
 
 import java.util.List;
 
-public interface Syllabizer {
+public abstract class Syllabizer {
 
-    List<String> syllabize(String input);
+    public final List<String> syllabize(String input) {
+        String normalized = input.replaceAll("\\s+", " ").strip();
+        if (normalized.isEmpty()) {
+            return List.of();
+        }
+        return syllabizeNormalized(normalized);
+    }
+
+    protected abstract List<String> syllabizeNormalized(String input);
 }

src/main/java/com/github/nianna/karedi/syllabizer/SyllabizerFactory.java

@@ -2,25 +2,23 @@
 
 import com.github.nianna.karedi.util.Language;
 
-import java.util.Set;
-
-import static com.github.nianna.karedi.util.Language.JAPANESE;
-
 public class SyllabizerFactory {
 
     private SyllabizerFactory() {
 
     }
 
     public static Syllabizer createFor(Language language) {
-        if (language == JAPANESE) {
-            return new JapaneseSyllabizer();
-        }
-        return null;
+        return switch (language) {
+            case ENGLISH -> new EnglishSyllabizer();
+            case JAPANESE -> new JapaneseSyllabizer();
+            case POLISH -> new PolishSyllabizer();
+            case SPANISH, ESPANOL -> new SpanishSyllabizer();
+            default -> null;
+        };
     }
 
-    public static Set<Language> supportedLanguages() {
-        return Set.of(JAPANESE);
+    public static Syllabizer createNoopSyllabizer() {
+        return new NoopSyllabizer();
     }
-
 }

src/main/java/com/github/nianna/karedi/syllabizer/SyllabizerInitializationFailedException.java

@@ -0,0 +1,9 @@
+package com.github.nianna.karedi.syllabizer;
+
+public class SyllabizerInitializationFailedException extends IllegalStateException {
+
+    public SyllabizerInitializationFailedException(Throwable cause) {
+        super(cause);
+    }
+
+}

src/main/java/com/github/nianna/karedi/syllabizer/SyllablesSanitizer.java

@@ -0,0 +1,35 @@
+package com.github.nianna.karedi.syllabizer;
+
+import java.util.ArrayList;
+import java.util.List;
+
+class SyllablesSanitizer {
+
+    private final String syllablesPattern;
+
+    SyllablesSanitizer(String syllablesPattern) {
+        this.syllablesPattern = syllablesPattern;
+    }
+
+    List<String> sanitize(String[] syllables) {
+        ArrayList<String> result = new ArrayList<>();
+        for (int i = 0; i < syllables.length; i++) {
+            String syllable = syllables[i];
+            if (syllable.toLowerCase().strip().matches(syllablesPattern)) {
+                result.add(syllable);
+            } else {
+                if (result.isEmpty() || syllable.startsWith(" ")) {
+                    if (i < syllables.length - 1) {
+                        syllables[i + 1] = syllable + syllables[i + 1];
+                    } else {
+                        result.add(syllable);
+                    }
+                } else {
+                    result.set(result.size() - 1, result.get(result.size() - 1) + syllable);
+                }
+            }
+        }
+        return result;
+    }
+
+}

src/main/java/com/github/nianna/karedi/util/ResourceUtils.java

@@ -0,0 +1,23 @@
+package com.github.nianna.karedi.util;
+
+import java.nio.charset.StandardCharsets;
+import java.util.ArrayList;
+import java.util.List;
+import java.util.Scanner;
+
+public class ResourceUtils {
+
+    private ResourceUtils() {
+
+    }
+
+    public static List<String> readLines(String path) {
+        List<String> lines = new ArrayList<>();
+        try (Scanner scan = new Scanner(ResourceUtils.class.getResourceAsStream(path), StandardCharsets.UTF_8)) {
+            while (scan.hasNextLine()) {
+                lines.add(scan.nextLine());
+            }
+        }
+        return lines;
+    }
+}

src/main/java/module-info.java

@@ -11,6 +11,7 @@
     requires org.fxmisc.richtext;
     requires org.fxmisc.flowless;
     requires wellbehavedfx;
+    requires nianna.hyphenator;
 
     opens com.github.nianna.karedi to javafx.graphics;
     opens com.github.nianna.karedi.controller to javafx.fxml;

src/main/resources/messages.properties

@@ -250,6 +250,8 @@ songsaver.export.fail=Export failed - file not found
 songsaver.save.success=Save finished
 songsaver.save.fail=Save failed
 
+syllabizer.init_failed=Failed to initialize auto-syllabizer
+
 player.midi.unavailable=Midi sequencer unavailable
 player.midi.invalid_data=Invalid midi sequence requested
 player.mp3.fail=Playback failed

src/main/resources/messages_en_GB.properties

@@ -250,6 +250,8 @@ songsaver.export.fail=Export failed - file not found
 songsaver.save.success=Save finished
 songsaver.save.fail=Save failed
 
+syllabizer.init_failed=Failed to initialize auto-syllabizer
+
 player.midi.unavailable=Midi sequencer unavailable
 player.midi.invalid_data=Invalid midi sequence requested
 player.mp3.fail=Playback failed

src/main/resources/messages_pl_PL.properties

@@ -253,6 +253,8 @@ songsaver.export.fail=Eksportowanie nie powiod\u0142o si\u0119
 songsaver.save.success=Zapisano zmiany
 songsaver.save.fail=Zapisanie zmian nie powiod\u0142o si\u0119
 
+syllabizer.init_failed=Inicjalizacja sylabizatora nie powiod\u0142a si\u0119
+
 player.midi.unavailable=Sekwencera midi niedost\u0119pny
 player.midi.invalid_data=B\u0142edna sekwencja midi
 player.mp3.fail=B\u0142\u0105d odtwarzania