Add Character.toTitleCase. More efficient implementation for Character.digit

classlib/src/main/java/org/teavm/classlib/impl/JCLPlugin.java

@@ -210,6 +210,8 @@ public class JCLPlugin implements TeaVMPlugin {
                 new CharacterMetadataGenerator());
         reg.register(new MethodReference(Character.class, "obtainClasses", StringResource.class),
                 new CharacterMetadataGenerator());
+        reg.register(new MethodReference(Character.class, "acquireTitleCaseMapping", StringResource.class),
+                new CharacterMetadataGenerator());
     }
 
     @PlatformMarker

classlib/src/main/java/org/teavm/classlib/impl/unicode/UnicodeHelper.java

@@ -44,21 +44,58 @@ public final class UnicodeHelper {
         }
     }
 
-    public static String encodeIntByte(int[] data) {
+    public static String encodeIntPairsDiff(int[] data) {
         StringBuilder sb = new StringBuilder();
-        Base46.encode(sb, data.length);
+        Base46.encodeUnsigned(sb, data.length / 2);
-        for (int i = 0; i < data.length; i++) {
+        int lastKey = 0;
-            Base46.encode(sb, data[i]);
+        int lastValue = 0;
+        for (int i = 0; i < data.length; i += 2) {
+            int key = data[i];
+            int value = data[i + 1];
+            Base46.encode(sb, key - lastKey);
+            Base46.encode(sb, value - lastValue);
+            lastKey = key;
+            lastValue = value;
         }
         return sb.toString();
     }
 
-    public static int[] decodeIntByte(String text) {
+    public static int[] decodeIntPairsDiff(String text) {
         CharFlow flow = new CharFlow(text.toCharArray());
-        int sz = Base46.decode(flow);
+        int sz = Base46.decodeUnsigned(flow);
-        int[] data = new int[sz];
+        int[] data = new int[sz * 2];
+        int j = 0;
+        int lastKey = 0;
+        int lastValue = 0;
         for (int i = 0; i < sz; i++) {
-            data[i] = Base46.decode(flow);
+            lastKey += Base46.decode(flow);
+            lastValue += Base46.decode(flow);
+            data[j++] = lastKey;
+            data[j++] = lastValue;
+        }
+        return data;
+    }
+
+    public static String encodeIntDiff(int[] data) {
+        StringBuilder sb = new StringBuilder();
+        Base46.encodeUnsigned(sb, data.length);
+        int last = 0;
+        for (int i = 0; i < data.length; i++) {
+            int v = data[i];
+            Base46.encode(sb, v - last);
+            last = v;
+        }
+        return sb.toString();
+    }
+
+    public static int[] decodeIntDiff(String text) {
+        CharFlow flow = new CharFlow(text.toCharArray());
+        int sz = Base46.decodeUnsigned(flow);
+        int[] data = new int[sz];
+        int last = 0;
+        for (int i = 0; i < sz; i++) {
+            last += Base46.decode(flow);
+            data[i] = last;
         }
         return data;
     }

classlib/src/main/java/org/teavm/classlib/impl/unicode/UnicodeSupport.java

@@ -31,6 +31,7 @@ public final class UnicodeSupport {
     private static volatile CountDownLatch latch = new CountDownLatch(1);
     private static int[] digitValues;
     private static byte[] classes;
+    private static int[] titleCaseMapping;
     private static Map<String, Byte> classMap = new HashMap<>();
 
     static {
@@ -72,6 +73,7 @@ public final class UnicodeSupport {
     private static void parseUnicodeData() {
         IntegerArray digitValues = new IntegerArray(4096);
         IntegerArray classes = new IntegerArray(65536);
+        IntegerArray titleCaseMapping = new IntegerArray(256);
         try (BufferedReader reader = new BufferedReader(new InputStreamReader(UnicodeHelper.class
                 .getResourceAsStream("UnicodeData.txt")))) {
             while (true) {
@@ -94,6 +96,14 @@ public final class UnicodeSupport {
                 }
                 Byte charClass = classMap.get(fields[2]);
                 classes.add(charClass != null ? charClass.intValue() : 0);
+
+                if (!fields[14].isEmpty()) {
+                    int titleCaseCode = parseHex(fields[14]);
+                    if (fields[12].isEmpty() || parseHex(fields[12]) != titleCaseCode) {
+                        titleCaseMapping.add(charCode);
+                        titleCaseMapping.add(titleCaseCode);
+                    }
+                }
             }
         } catch (IOException e) {
             throw new RuntimeException("Error reading unicode data", e);
@@ -120,6 +130,7 @@ public final class UnicodeSupport {
         for (int i = 0; i < classes.size(); ++i) {
             UnicodeSupport.classes[i] = (byte) classes.get(i);
         }
+        UnicodeSupport.titleCaseMapping = titleCaseMapping.getAll();
     }
 
     private static String[] splitLine(String line) {
@@ -200,4 +211,9 @@ public final class UnicodeSupport {
         ensureUnicodeData();
         return classes;
     }
+
+    public static int[] getTitleCaseMapping() {
+        ensureUnicodeData();
+        return titleCaseMapping;
+    }
 }

classlib/src/main/java/org/teavm/classlib/java/io/TPrintStream.java

@@ -202,6 +202,16 @@ public class TPrintStream extends TFilterOutputStream {
         printSB();
     }
 
+    public void println(char c) {
+        sb.append(c);
+        printSB();
+    }
+
+    public void println(boolean b) {
+        sb.append(b);
+        printSB();
+    }
+
     public void println(String s) {
         sb.append(s).append('\n');
         printSB();

classlib/src/main/java/org/teavm/classlib/java/lang/CharacterMetadataGenerator.java

@@ -31,6 +31,8 @@ public class CharacterMetadataGenerator implements MetadataGenerator {
                 return generateObtainDigitMapping(context);
             case "obtainClasses":
                 return generateObtainClasses(context);
+            case "acquireTitleCaseMapping":
+                return generateObtainTitleCaseMapping(context);
             default:
                 return null;
         }
@@ -38,7 +40,7 @@ public class CharacterMetadataGenerator implements MetadataGenerator {
 
     private Resource generateObtainDigitMapping(MetadataGeneratorContext context) {
         StringResource res = context.createResource(StringResource.class);
-        res.setValue(UnicodeHelper.encodeIntByte(UnicodeSupport.getDigitValues()));
+        res.setValue(UnicodeHelper.encodeIntPairsDiff(UnicodeSupport.getDigitValues()));
         return res;
     }
 
@@ -47,4 +49,10 @@ public class CharacterMetadataGenerator implements MetadataGenerator {
         res.setValue(UnicodeHelper.compressRle(UnicodeSupport.getClasses()));
         return res;
     }
+
+    private Resource generateObtainTitleCaseMapping(MetadataGeneratorContext context) {
+        StringResource res = context.createResource(StringResource.class);
+        res.setValue(UnicodeHelper.encodeIntDiff(UnicodeSupport.getTitleCaseMapping()));
+        return res;
+    }
 }

classlib/src/main/java/org/teavm/classlib/java/lang/TCharacter.java

@@ -91,6 +91,7 @@ public class TCharacter extends TObject implements TComparable<TCharacter> {
     public static final int SIZE = 16;
     static final int ERROR = 0xFFFFFFFF;
     private static int[] digitMapping;
+    private static int[] titleCaseMapping;
     private static UnicodeHelper.Range[] classMapping;
     private char value;
     private static TCharacter[] characterCache = new TCharacter[128];
@@ -266,6 +267,37 @@ public class TCharacter extends TObject implements TComparable<TCharacter> {
     @Unmanaged
     private static native int toUpperCaseSystem(int codePoint);
 
+    public static int toTitleCase(int codePoint) {
+        int[] mapping = getTitleCaseMapping();
+        int l = 0;
+        int u = (mapping.length / 2) - 1;
+        while (u >= l) {
+            int idx = (l + u) / 2;
+            int val = mapping[idx * 2];
+            if (codePoint > val) {
+                l = idx + 1;
+            } else if (codePoint < val) {
+                u = idx - 1;
+            } else {
+                return mapping[idx * 2 + 1];
+            }
+        }
+        return toUpperCase(codePoint);
+    }
+
+    public static char toTitleCase(char c) {
+        return (char) toTitleCase((int) c);
+    }
+
+    private static int[] getTitleCaseMapping() {
+        if (titleCaseMapping == null) {
+            titleCaseMapping = UnicodeHelper.decodeIntDiff(acquireTitleCaseMapping().getValue());
+        }
+        return titleCaseMapping;
+    }
+
+    private static native StringResource acquireTitleCaseMapping();
+
     public static int digit(char ch, int radix) {
         return digit((int) ch, radix);
     }
@@ -317,7 +349,7 @@ public class TCharacter extends TObject implements TComparable<TCharacter> {
 
     private static int[] getDigitMapping() {
         if (digitMapping == null) {
-            digitMapping = UnicodeHelper.decodeIntByte(obtainDigitMapping().getValue());
+            digitMapping = UnicodeHelper.decodeIntPairsDiff(obtainDigitMapping().getValue());
         }
         return digitMapping;
     }