Merge pull request #200 from dynatrace-oss/errorprone

introduced errorprone, fixed hashTestObject4 jmh benchmark

README.md

@@ -184,7 +184,7 @@ However, this algorithm can be useful for file indexes, for example, to find ide
 ```java
 // create some file in the given path
 File file = path.resolve("test.txt").toFile();
-try (FileWriter fileWriter = new FileWriter(file)) {
+try (FileWriter fileWriter = new FileWriter(file, StandardCharsets.UTF_8)) {
     fileWriter.write("this is the file content");
 }
 

build.gradle

@@ -20,6 +20,7 @@ plugins {
 	id 'signing'
 	id 'io.github.gradle-nexus.publish-plugin' version '1.3.0'
 	id "com.palantir.revapi" version "1.7.0"
+	id "net.ltgt.errorprone" version "3.1.0"
 }
 
 static def readJavaLicense(licenseName) {
@@ -185,6 +186,7 @@ dependencies {
 	testImplementation group: 'net.openhft', name: 'zero-allocation-hashing', version: '0.16'
 	testImplementation group: 'com.appmattus.crypto', name: 'cryptohash', version: '0.10.1'
 	testImplementation group: 'org.greenrobot', name: 'essentials', version: '3.1.0'
+	errorprone("com.google.errorprone:error_prone_core:2.24.0")
 }
 
 jacocoTestReport {
@@ -383,4 +385,10 @@ nexusPublishing {
 
 if (file("extra-configuration.gradle").exists()) {
 	apply from: 'extra-configuration.gradle'
-}
+}
+
+tasks.withType(JavaCompile).configureEach {
+	options.compilerArgs << "-Werror"
+	options.errorprone.disableWarningsInGeneratedCode = false
+	// options.errorprone.enabled = false
+}

src/jmh/java/com/dynatrace/hash4j/distinctcount/HyperLogLogPerformanceTest.java

@@ -71,6 +71,7 @@ public void distinctCountAddWithMartingaleEstimator(AddState addState, Blackhole
     blackhole.consume(martingaleEstimator.getDistinctCountEstimate());
   }
 
+  @SuppressWarnings("ImmutableEnumChecker")
   public enum Estimator {
     MAXIMUM_LIKELIHOOD_ESTIMATOR(HyperLogLog.MAXIMUM_LIKELIHOOD_ESTIMATOR),
     CORRECTED_RAW_ESTIMATOR(HyperLogLog.CORRECTED_RAW_ESTIMATOR);

src/jmh/java/com/dynatrace/hash4j/distinctcount/UltraLogLogPerformanceTest.java

@@ -71,6 +71,7 @@ public void distinctCountAddWithMartingaleEstimator(AddState addState, Blackhole
     blackhole.consume(martingaleEstimator.getDistinctCountEstimate());
   }
 
+  @SuppressWarnings("ImmutableEnumChecker")
   public enum Estimator {
     MAXIMUM_LIKELIHOOD_ESTIMATOR(UltraLogLog.MAXIMUM_LIKELIHOOD_ESTIMATOR),
     OPTIMAL_FGRA_ESTIMATOR(UltraLogLog.OPTIMAL_FGRA_ESTIMATOR);

src/jmh/java/com/dynatrace/hash4j/hashing/AbstractPerformanceTest.java

@@ -1,5 +1,5 @@
 /*
- * Copyright 2022 Dynatrace LLC
+ * Copyright 2022-2023 Dynatrace LLC
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -281,8 +281,7 @@ private static String createRandomString(int minLen, int maxLen, SplittableRando
     int len = random.nextInt(minLen, maxLen + 1);
     StringBuilder sb = new StringBuilder();
     for (int i = 0; i < len; ++i) {
-      char c = 'a';
-      c += random.nextInt(0, 26);
+      char c = (char) ('a' + random.nextInt(0, 26));
       sb.append(c);
     }
     return sb.toString();
@@ -292,9 +291,8 @@ private static String createRandomGreekString(int minLen, int maxLen, Splittable
     int len = random.nextInt(minLen, maxLen + 1);
     StringBuilder sb = new StringBuilder();
     for (int i = 0; i < len; ++i) {
-      char c = 0x03B1;
-      c += random.nextInt(0, 24);
-      if (c >= 0x03C2) c += 1;
+      char c = (char) (0x03B1 + random.nextInt(0, 24));
+      if (c >= 0x03C2) c = (char) (c + 1);
       sb.append(c);
     }
     return sb.toString();
@@ -311,8 +309,6 @@ private static TestObject[] createTestObjects(
   }
 
   static {
-    final SplittableRandom random = new SplittableRandom(0);
-
     BYTE_ARRAYS_1 = createRandomByteArrays(NUM_OBJECTS, 1, 1, 0x035348bcb49493a4L);
     BYTE_ARRAYS_4 = createRandomByteArrays(NUM_OBJECTS, 1, 4, 0xcc6444ca02edfbd0L);
     BYTE_ARRAYS_16 = createRandomByteArrays(NUM_OBJECTS, 1, 16, 0x187c616cabc3e0a7L);
@@ -346,7 +342,7 @@ private static TestObject[] createTestObjects(
     TEST_OBJECTS1 = createTestObjects(NUM_OBJECTS, TestObject1::new, 0x37569b3107539e19L);
     TEST_OBJECTS2 = createTestObjects(NUM_OBJECTS, TestObject2::new, 0x892da841ae127839L);
     TEST_OBJECTS3 = createTestObjects(NUM_OBJECTS, TestObject3::new, 0xb443e2873a03f397L);
-    TEST_OBJECTS4 = createTestObjects(NUM_OBJECTS, TestObject3::new, 0x49952ea071f1cc0aL);
+    TEST_OBJECTS4 = createTestObjects(NUM_OBJECTS, TestObject4::new, 0x49952ea071f1cc0aL);
   }
 
   private void directBytesTest(byte[][] data, Blackhole blackhole) {

src/jmh/java/com/dynatrace/hash4j/hashing/AbstractZeroAllocationHashing64BitPerformanceTest.java

@@ -1,5 +1,5 @@
 /*
- * Copyright 2022 Dynatrace LLC
+ * Copyright 2022-2023 Dynatrace LLC
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -45,6 +45,7 @@ protected long hashCharsIndirect(String s) {
 
   protected abstract LongHashFunction createHashFunction();
 
+  @Override
   protected long hashObject(TestObject testObject) {
     try {
       ByteArrayDataOutput dataOutput = ByteStreams.newDataOutput();

src/jmh/java/com/dynatrace/hash4j/hashing/UnorderedHashTest.java

@@ -1,5 +1,5 @@
 /*
- * Copyright 2022 Dynatrace LLC
+ * Copyright 2022-2023 Dynatrace LLC
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -19,7 +19,6 @@
 import java.util.List;
 import java.util.SplittableRandom;
 import java.util.concurrent.ThreadLocalRandom;
-import java.util.function.ToLongFunction;
 import java.util.stream.Collectors;
 import org.openjdk.jmh.annotations.Benchmark;
 import org.openjdk.jmh.annotations.BenchmarkMode;
@@ -40,10 +39,13 @@ public class UnorderedHashTest {
   }
 
   private static final HashFunnel<Long> LONG_FUNNEL = (l, h) -> h.putLong(l);
-  private static final ToLongFunction<Long> LONG_2_HASH =
-      x -> Hashing.murmur3_128().hashToLong(x, LONG_FUNNEL);
+
+  private static final long long2Hash(long x) {
+    return Hashing.murmur3_128().hashToLong(x, LONG_FUNNEL);
+  }
+
   private static final HashFunnel<List<Long>> LIST_LONG_FUNNEL =
-      (l, h) -> h.putUnorderedIterable(l, LONG_2_HASH);
+      (l, h) -> h.putUnorderedIterable(l, UnorderedHashTest::long2Hash);
 
   @Benchmark
   @BenchmarkMode(Mode.AverageTime)

src/main/java/com/dynatrace/hash4j/distinctcount/HyperLogLog.java

@@ -322,7 +322,7 @@ public static int computeToken(long hashValue) {
    */
   @Override
   public HyperLogLog add(long hashValue, StateChangeObserver stateChangeObserver) {
-    int idx = (int) (hashValue >>> (-p));
+    int idx = (int) (hashValue >>> -p);
     int newValue = Long.numberOfLeadingZeros(~(~hashValue << p)) + 1;
     int oldValue = (int) ARRAY_HANDLER.update(state, idx, newValue, Math::max);
     if (stateChangeObserver != null && newValue > oldValue) {
@@ -601,7 +601,7 @@ private static final class MaximumLikelihoodEstimator implements Estimator {
     //
     // for a numerical evaluation see
     // https://www.wolframalpha.com/input?i=sqrt%28ln%282%29%2Fzeta%282%2C2%29%29
-    private static final double INV_SQRT_FISHER_INFORMATION = 1.0367047097785011;
+    private static final double INV_SQRT_FISHER_INFORMATION = 1.0367047097785012;
     private static final double ML_EQUATION_SOLVER_EPS =
         0.001 * INV_SQRT_FISHER_INFORMATION; // 0.1% of theoretical relative error
 

src/main/java/com/dynatrace/hash4j/distinctcount/UltraLogLog.java

@@ -245,10 +245,10 @@ public static int computeToken(long hashValue) {
   public UltraLogLog add(long hashValue, StateChangeObserver stateChangeObserver) {
     int q = Long.numberOfLeadingZeros(state.length - 1L); // q = 64 - p
     int idx = (int) (hashValue >>> q);
-    int nlz = Long.numberOfLeadingZeros(~(~hashValue << (-q))); // nlz in {0, 1, ..., 64-p}
+    int nlz = Long.numberOfLeadingZeros(~(~hashValue << -q)); // nlz in {0, 1, ..., 64-p}
     byte oldState = state[idx];
     long hashPrefix = unpack(oldState);
-    hashPrefix |= 1L << (nlz + (~q)); // (nlz + (~q)) = (nlz + p - 1) in {p-1, ... 63}
+    hashPrefix |= 1L << (nlz + ~q); // (nlz + (~q)) = (nlz + p - 1) in {p-1, ... 63}
     byte newState = pack(hashPrefix);
     state[idx] = newState;
     if (stateChangeObserver != null && newState != oldState) {
@@ -323,7 +323,7 @@ static long unpack(byte register) {
   // visible for testing
   static byte pack(long hashPrefix) {
     int nlz = Long.numberOfLeadingZeros(hashPrefix) + 1;
-    return (byte) (((-nlz) << 2) | ((hashPrefix << nlz) >>> 62));
+    return (byte) ((-nlz << 2) | ((hashPrefix << nlz) >>> 62));
   }
 
   /**
@@ -410,7 +410,7 @@ private static final class MaximumLikelihoodEstimator implements Estimator {
     //
     // for a numerical evaluation see
     // https://www.wolframalpha.com/input?i=3%2F2+*+ln%282%29+*+zeta%283%2C5%2F4%29%2F%28zeta%282%2C5%2F4%29%29%5E2
-    private static final double ML_BIAS_CORRECTION_CONSTANT = 0.48147376527720066;
+    private static final double ML_BIAS_CORRECTION_CONSTANT = 0.48147376527720065;
 
     // returns contribution to alpha, scaled by 2^64
     private static long contribute(int r, int[] b, int p) {
@@ -772,7 +772,7 @@ static double smallRangeEstimate(long c0, long c4, long c8, long c10, long m) {
       long alpha = m + 3 * (c0 + c4 + c8 + c10);
       long beta = m - c0 - c4;
       long gamma = 4 * c0 + 2 * c4 + 3 * c8 + c10;
-      double quadRootZ = (sqrt(beta * beta + 4 * alpha * gamma) - beta) / (2 * alpha);
+      double quadRootZ = (sqrt((double) (beta * beta + 4 * alpha * gamma)) - beta) / (2 * alpha);
       double rootZ = quadRootZ * quadRootZ;
       return rootZ * rootZ;
     }
@@ -781,7 +781,7 @@ static double largeRangeEstimate(long c4w0, long c4w1, long c4w2, long c4w3, lon
       long alpha = m + 3 * (c4w0 + c4w1 + c4w2 + c4w3);
       long beta = c4w0 + c4w1 + 2 * (c4w2 + c4w3);
       long gamma = m + 2 * c4w0 + c4w2 - c4w3;
-      return sqrt((sqrt(beta * beta + 4 * alpha * gamma) - beta) / (2 * alpha));
+      return sqrt((sqrt((double) (beta * beta + 4 * alpha * gamma)) - beta) / (2 * alpha));
     }
 
     // this is psi as defined in the paper divided by ETA_X

src/main/java/com/dynatrace/hash4j/hashing/AbstractHashStream.java

@@ -159,7 +159,7 @@ public HashStream putInts(int[] x, int off, int len) {
 
   @Override
   public HashStream putLong(long v) {
-    putInt((int) (v));
+    putInt((int) v);
     putInt((int) (v >> 32));
     return this;
   }

src/main/java/com/dynatrace/hash4j/hashing/AbstractKomihash.java

@@ -257,7 +257,7 @@ public HashStream64 putChars(CharSequence s) {
               remainingChars -= 32;
               off += 32;
             } while (remainingChars >= 32);
-            buffer[0] = (byte) (z);
+            buffer[0] = (byte) z;
           }
         }
       }

src/main/java/com/dynatrace/hash4j/hashing/AbstractWyhashFinal.java

@@ -222,7 +222,7 @@ protected static long[] makeSecret(long seed) {
         ok = true;
         seed += 0xa0761d6478bd642fL;
         secret[i] =
-            (c[(int) (Long.remainderUnsigned(wymix(seed, seed ^ 0xe7037ed1a0b428dbL), c.length))]
+            (c[(int) Long.remainderUnsigned(wymix(seed, seed ^ 0xe7037ed1a0b428dbL), c.length)]
                 & 0xFFL);
         if ((secret[i] & 1) == 0) {
           seed += 0x633acdbf4d2dbd49L; // = 7 * 0xa0761d6478bd642fL
@@ -232,7 +232,7 @@ protected static long[] makeSecret(long seed) {
         for (int j = 8; j < 64; j += 8) {
           seed += 0xa0761d6478bd642fL;
           secret[i] |=
-              (c[(int) (Long.remainderUnsigned(wymix(seed, seed ^ 0xe7037ed1a0b428dbL), c.length))]
+              (c[(int) Long.remainderUnsigned(wymix(seed, seed ^ 0xe7037ed1a0b428dbL), c.length)]
                       & 0xFFL)
                   << j;
         }
@@ -438,7 +438,7 @@ public HashStream64 putChars(CharSequence s) {
             setLong(buffer, 32, b4);
             setLong(buffer, 40, b5);
           }
-          buffer[0] = (byte) (z);
+          buffer[0] = (byte) z;
         }
       }
       while (remainingChars >= 4) {

src/main/java/com/dynatrace/hash4j/hashing/FarmHashNa.java

@@ -66,10 +66,9 @@ class FarmHashNa extends AbstractHasher64 {
   private static final long K1 = 0xb492b66fbe98f273L;
   protected static final long K2 = 0x9ae16a3b2f90404fL;
   private static final long K_MUL = 0x9ddfea08eb382d69L;
-  private static final long SEED = 81;
-  private static final long START_X = SEED * K2;
-  private static final long START_Y = SEED * K1 + 113;
-  private static final long START_Z = shiftMix(START_Y * K2 + 113) * K2;
+  private static final long START_X = 0x1529cba0ca458ffL;
+  private static final long START_Y = 0x226bb95b4e64b6d4L;
+  private static final long START_Z = 0x134a747f856d0526L;
 
   private static final FarmHashNa INSTANCE = new FarmHashNa();
 
@@ -733,6 +732,7 @@ private long naHashLen33To64(int bufferCount) {
           rotateRight(e + f, 43) + rotateRight(g, 30) + h, e + rotateRight(f + a, 18) + g, mul);
     }
 
+    @Override
     public long getAsLong() {
       return finalizeHash(processRemaining());
     }

src/main/java/com/dynatrace/hash4j/hashing/HashStream128.java

@@ -1,5 +1,5 @@
 /*
- * Copyright 2022 Dynatrace LLC
+ * Copyright 2022-2023 Dynatrace LLC
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -171,5 +171,6 @@ <T> HashStream128 putUnorderedIterable(
    *
    * @return this
    */
+  @Override
   HashStream128 reset();
 }

src/main/java/com/dynatrace/hash4j/hashing/HashStream64.java

@@ -1,5 +1,5 @@
 /*
- * Copyright 2022 Dynatrace LLC
+ * Copyright 2022-2023 Dynatrace LLC
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -171,5 +171,6 @@ <T> HashStream64 putUnorderedIterable(
    *
    * @return this
    */
+  @Override
   HashStream64 reset();
 }

src/main/java/com/dynatrace/hash4j/hashing/Komihash4_3.java

@@ -152,7 +152,7 @@ public long hashBytesToLong(byte[] input, int off, int len) {
     } else if (len > 3) {
       long fb = getInt(input, off) & 0xFFFFFFFFL;
       long y = getInt(input, off + len - 4);
-      fb |= (y << 32) >>> (-ml8);
+      fb |= (y << 32) >>> -ml8;
       fb |= 1L << ml8 << (y >>> 63);
       r2l ^= fb;
     } else if (len > 0) {
@@ -267,7 +267,7 @@ public long hashCharsToLong(CharSequence input) {
     } else if (len > 1) {
       long fb = getInt(input, off) & 0xFFFFFFFFL;
       long y = getInt(input, off + len - 2);
-      fb |= (y << 32) >>> (-ml8);
+      fb |= (y << 32) >>> -ml8;
       fb |= 1L << ml8 << (y >>> 63);
       r2l ^= fb;
     } else if (len > 0) {

src/main/java/com/dynatrace/hash4j/hashing/Komihash5_0.java

@@ -146,11 +146,11 @@ public long hashBytesToLong(byte[] input, int off, int len) {
     if (len > 7) {
       r2l ^= getLong(input, off);
       long y = getLong(input, off + len - 8);
-      r2h ^= (1L << ml8) | (y >>> 1 >>> (~ml8));
+      r2h ^= (1L << ml8) | (y >>> 1 >>> ~ml8);
     } else if (len > 3) {
       long mh = getInt(input, off + len - 4);
       long ml = getInt(input, off) & 0xFFFFFFFFL;
-      r2l ^= (1L << ml8) | ml | (mh << 32 >>> (-ml8));
+      r2l ^= (1L << ml8) | ml | (mh << 32 >>> -ml8);
     } else if (len > 0) {
       long m = (1L << ml8) | (input[off] & 0xFFL);
       if (len > 1) m |= (input[off + 1] & 0xFFL) << 8;
@@ -256,11 +256,11 @@ public long hashCharsToLong(CharSequence input) {
     if (len > 3) {
       r2l ^= getLong(input, off);
       long y = getLong(input, off + len - 4);
-      r2h ^= (1L << ml8) | (y >>> 1 >>> (~ml8));
+      r2h ^= (1L << ml8) | (y >>> 1 >>> ~ml8);
     } else if (len > 1) {
       long mh = getInt(input, off + len - 2);
       long ml = getInt(input, off) & 0xFFFFFFFFL;
-      r2l ^= (1L << ml8) | ml | (mh << 32 >>> (-ml8));
+      r2l ^= (1L << ml8) | ml | (mh << 32 >>> -ml8);
     } else if (len > 0) {
       long m = (1L << ml8) | input.charAt(off);
       r2l ^= m;