AVRO-3527: codegen equals and hashCode for Records

Update the compiler to generate the implementation of the `.equals()`
and `.hashCode() function, instead of relying on the
implementation of GenericData.  This improves the performance of
those functions significantly.

The generated implementations are factor 10 to 20 faster for
`.equals()` and a factor 5 to 10 for `.hashCode()`.

Result of Perf test before the change:

```
Benchmark              Mode  Cnt          Score             Error  Units
SpecficTest.equals    thrpt    3   12598610.194 +/-  11160265.279  ops/s
SpecficTest.hashCode  thrpt    3   24729446.862 +/-  29051332.794  ops/s
```

Results using generated functions:

```
Benchmark              Mode  Cnt          Score             Error  Units
SpecficTest.equals    thrpt    3  211314296.950 +/- 104154793.126  ops/s
SpecficTest.hashCode  thrpt    3  180349506.632 +/- 143639246.771  ops/s
```

Signed-off-by: Steven Aerts <steven.aerts@gmail.com>

Author: steven-aerts

lang/java/avro/src/main/java/org/apache/avro/generic/GenericData.java

@@ -1181,9 +1181,9 @@ protected int compare(Object o1, Object o2, Schema s, boolean equals) {
     case NULL:
       return 0;
     case STRING:
-      Utf8 u1 = o1 instanceof Utf8 ? (Utf8) o1 : new Utf8(o1.toString());
-      Utf8 u2 = o2 instanceof Utf8 ? (Utf8) o2 : new Utf8(o2.toString());
-      return u1.compareTo(u2);
+      CharSequence cs1 = o1 instanceof CharSequence ? (CharSequence) o1 : o1.toString();
+      CharSequence cs2 = o2 instanceof CharSequence ? (CharSequence) o2 : o2.toString();
+      return Utf8.compareSequences(cs1, cs2);
     default:
       return ((Comparable) o1).compareTo(o2);
     }

lang/java/avro/src/main/java/org/apache/avro/util/Utf8.java

@@ -237,4 +237,28 @@ public void readExternal(ObjectInput in) throws IOException, ClassNotFoundExcept
     setByteLength(in.readInt());
     in.readFully(bytes);
   }
+
+  public static int compareSequences(CharSequence cs1, CharSequence cs2) {
+    if (cs1 == cs2) {
+      return 0;
+    }
+
+    if (cs1 == null || cs2 == null) {
+      return cs1 == null ? 1 : -1;
+    }
+
+    if (cs1.getClass() == cs2.getClass() && cs1 instanceof Comparable) {
+      return ((Comparable<Object>) cs1).compareTo(cs2);
+    }
+
+    for (int i = 0, len = Math.min(cs1.length(), cs2.length()); i < len; i++) {
+      char a = cs1.charAt(i);
+      char b = cs2.charAt(i);
+      if (a != b) {
+        return a - b;
+      }
+    }
+
+    return cs1.length() - cs2.length();
+  }
 }

lang/java/avro/src/test/java/org/apache/avro/util/TestUtf8.java

@@ -98,6 +98,15 @@ public void testHashCodeReused() {
     assertEquals(3198781, u.hashCode());
   }
 
+  @Test
+  public void testHashCodeSameAsString() {
+    assertEquals("a".hashCode(), new Utf8("a").hashCode());
+    assertEquals("zz".hashCode(), new Utf8("zz").hashCode());
+    assertEquals("z".hashCode(), new Utf8("z").hashCode());
+    assertEquals("hello".hashCode(), new Utf8("hello").hashCode());
+    assertEquals("hell".hashCode(), new Utf8("hell").hashCode());
+  }
+
   @Test
   public void testSerialization() throws IOException, ClassNotFoundException {
     try (ByteArrayOutputStream bos = new ByteArrayOutputStream();

lang/java/compiler/src/main/java/org/apache/avro/compiler/specific/SpecificCompiler.java

@@ -1131,6 +1131,46 @@ public static String mangle(String word, Set<String> reservedWords, boolean isMe
     return word;
   }
 
+  public boolean canGenerateEqualsAndHashCode(Schema schema) {
+    return getUsedCustomLogicalTypeFactories(schema).isEmpty();
+  }
+
+  public boolean isPrimitiveType(Schema schema) {
+    return !isUnboxedJavaTypeNullable(schema) && getConvertedLogicalType(schema) == null;
+  }
+
+  public String hashCodeFor(Schema schema, String name) {
+    switch (javaUnbox(schema)) {
+    case "int":
+      return "Integer.hashCode(" + name + ")";
+    case "long":
+      return "Long.hashCode(" + name + ")";
+    case "float":
+      return "Float.hashCode(" + name + ")";
+    case "double":
+      return "Double.hashCode(" + name + ")";
+    case "boolean":
+      return "Boolean.hashCode(" + name + ")";
+    default:
+      // Hashcode of Union is expected to match ordinal
+      if (schema.getType() == Schema.Type.ENUM || ((schema.getType() == Schema.Type.UNION)
+          && (schema.getTypes().stream().anyMatch(t -> t.getType() == Schema.Type.ENUM)))) {
+        if (schema.getType() == Schema.Type.ENUM
+            || (schema.getTypes().size() == 2 && schema.getTypes().contains(NULL_SCHEMA))) {
+          return "(" + name + " == null ? 0 : ((java.lang.Enum) " + name + ").ordinal())";
+        } else {
+          return "(" + name + " == null ? 0 : " + name + " instanceof java.lang.Enum ? ((java.lang.Enum) " + name
+              + ").ordinal() : " + name + ".hashCode())";
+        }
+      }
+      return "(" + name + " == null ? 0 : " + name + ".hashCode())";
+    }
+  }
+
+  public boolean ignoredField(Field field) {
+    return field.order() == Field.Order.IGNORE;
+  }
+
   /**
    * Utility for use by templates. Return schema fingerprint as a long.
    */

lang/java/compiler/src/main/velocity/org/apache/avro/compiler/specific/templates/java/classic/record.vm

@@ -604,6 +604,47 @@ public class ${this.mangleTypeIdentifier($schema.getName())}#if ($schema.isError
     }
   }
 #end
+#if ($this.canGenerateEqualsAndHashCode($schema))
+
+  @Override
+  public int hashCode() {
+    int result = 1;
+#foreach ($field in $schema.getFields())
+#if (!${this.ignoredField($field)})
+#set ($n = ${this.mangle($field.name(), $schema.isError())})
+    result = 31 * result + ${this.hashCodeFor($field.schema(), $n)};
+#end
+#end
+    return result;
+  }
+
+  @Override
+  public boolean equals(Object o) {
+    if (this == o) {
+      return true;
+    }
+    if (!(o instanceof ${this.mangleTypeIdentifier($schema.getName())})) {
+      return false;
+    }
+    ${this.mangleTypeIdentifier($schema.getName())} other = (${this.mangleTypeIdentifier($schema.getName())}) o;
+#foreach ($field in $schema.getFields())
+#if (!${this.ignoredField($field)})
+#set ($n = ${this.mangle($field.name(), $schema.isError())})
+#set ($s = $field.schema())
+#if (${this.isPrimitiveType($s)})
+    if (this.$n != other.$n) {
+#elseif (${this.javaType($field.schema()).equals("java.lang.CharSequence")})
+    if (Utf8.compareSequences(this.$n, other.$n) != 0) {
+#else
+    if (!java.util.Objects.equals(this.$n, other.$n)) {
+#end
+      return false;
+    }
+#end
+#end
+    return true;
+  }
+#end
 }
 
 #macro( encodeVar $indent $var $s )

lang/java/tools/src/test/compiler/output-string/avro/examples/baseball/FieldTest.java

@@ -16,6 +16,8 @@
 @org.apache.avro.specific.AvroGenerated
 public class FieldTest extends org.apache.avro.specific.SpecificRecordBase implements org.apache.avro.specific.SpecificRecord {
   private static final long serialVersionUID = 4609235620572341636L;
+
+
   public static final org.apache.avro.Schema SCHEMA$ = new org.apache.avro.Schema.Parser().parse("{\"type\":\"record\",\"name\":\"FieldTest\",\"namespace\":\"avro.examples.baseball\",\"doc\":\"Test various field types\",\"fields\":[{\"name\":\"number\",\"type\":\"int\",\"doc\":\"The number of the player\"},{\"name\":\"last_name\",\"type\":{\"type\":\"string\",\"avro.java.string\":\"String\"}},{\"name\":\"timestamp\",\"type\":{\"type\":\"long\",\"logicalType\":\"timestamp-millis\"}},{\"name\":\"timestampMicros\",\"type\":{\"type\":\"long\",\"logicalType\":\"timestamp-micros\"}},{\"name\":\"timeMillis\",\"type\":{\"type\":\"int\",\"logicalType\":\"time-millis\"}},{\"name\":\"timeMicros\",\"type\":{\"type\":\"long\",\"logicalType\":\"time-micros\"}}]}");
   public static org.apache.avro.Schema getClassSchema() { return SCHEMA$; }
 
@@ -654,6 +656,48 @@ public FieldTest build() {
     READER$.read(this, SpecificData.getDecoder(in));
   }
 
+
+  @Override
+  public int hashCode() {
+    int result = 1;
+    result = 31 * result + Integer.hashCode(number);
+    result = 31 * result + (last_name == null ? 0 : last_name.hashCode());
+    result = 31 * result + (timestamp == null ? 0 : timestamp.hashCode());
+    result = 31 * result + (timestampMicros == null ? 0 : timestampMicros.hashCode());
+    result = 31 * result + (timeMillis == null ? 0 : timeMillis.hashCode());
+    result = 31 * result + (timeMicros == null ? 0 : timeMicros.hashCode());
+    return result;
+  }
+
+  @Override
+  public boolean equals(Object o) {
+    if (this == o) {
+      return true;
+    }
+    if (!(o instanceof FieldTest)) {
+      return false;
+    }
+    FieldTest other = (FieldTest) o;
+    if (this.number != other.number) {
+      return false;
+    }
+    if (!java.util.Objects.equals(this.last_name, other.last_name)) {
+      return false;
+    }
+    if (!java.util.Objects.equals(this.timestamp, other.timestamp)) {
+      return false;
+    }
+    if (!java.util.Objects.equals(this.timestampMicros, other.timestampMicros)) {
+      return false;
+    }
+    if (!java.util.Objects.equals(this.timeMillis, other.timeMillis)) {
+      return false;
+    }
+    if (!java.util.Objects.equals(this.timeMicros, other.timeMicros)) {
+      return false;
+    }
+    return true;
+  }
 }
 
 

lang/java/tools/src/test/compiler/output-string/avro/examples/baseball/Player.java

@@ -16,6 +16,8 @@
 @org.apache.avro.specific.AvroGenerated
 public class Player extends org.apache.avro.specific.SpecificRecordBase implements org.apache.avro.specific.SpecificRecord {
   private static final long serialVersionUID = 3865593031278745715L;
+
+
   public static final org.apache.avro.Schema SCHEMA$ = new org.apache.avro.Schema.Parser().parse("{\"type\":\"record\",\"name\":\"Player\",\"namespace\":\"avro.examples.baseball\",\"doc\":\"選手 is Japanese for player.\",\"fields\":[{\"name\":\"number\",\"type\":\"int\",\"doc\":\"The number of the player\"},{\"name\":\"first_name\",\"type\":{\"type\":\"string\",\"avro.java.string\":\"String\"}},{\"name\":\"last_name\",\"type\":{\"type\":\"string\",\"avro.java.string\":\"String\"}},{\"name\":\"position\",\"type\":{\"type\":\"array\",\"items\":{\"type\":\"enum\",\"name\":\"Position\",\"symbols\":[\"P\",\"C\",\"B1\",\"B2\",\"B3\",\"SS\",\"LF\",\"CF\",\"RF\",\"DH\"]}}}]}");
   public static org.apache.avro.Schema getClassSchema() { return SCHEMA$; }
 
@@ -582,6 +584,40 @@ public Player build() {
       }
     }
   }
+
+  @Override
+  public int hashCode() {
+    int result = 1;
+    result = 31 * result + Integer.hashCode(number);
+    result = 31 * result + (first_name == null ? 0 : first_name.hashCode());
+    result = 31 * result + (last_name == null ? 0 : last_name.hashCode());
+    result = 31 * result + (position == null ? 0 : position.hashCode());
+    return result;
+  }
+
+  @Override
+  public boolean equals(Object o) {
+    if (this == o) {
+      return true;
+    }
+    if (!(o instanceof Player)) {
+      return false;
+    }
+    Player other = (Player) o;
+    if (this.number != other.number) {
+      return false;
+    }
+    if (!java.util.Objects.equals(this.first_name, other.first_name)) {
+      return false;
+    }
+    if (!java.util.Objects.equals(this.last_name, other.last_name)) {
+      return false;
+    }
+    if (!java.util.Objects.equals(this.position, other.position)) {
+      return false;
+    }
+    return true;
+  }
 }
 
 

lang/java/tools/src/test/compiler/output/AddExtraOptionalGettersTest.java

@@ -16,6 +16,8 @@
 @org.apache.avro.specific.AvroGenerated
 public class AddExtraOptionalGettersTest extends org.apache.avro.specific.SpecificRecordBase implements org.apache.avro.specific.SpecificRecord {
   private static final long serialVersionUID = -3300987256178011215L;
+
+
   public static final org.apache.avro.Schema SCHEMA$ = new org.apache.avro.Schema.Parser().parse("{\"type\":\"record\",\"name\":\"AddExtraOptionalGettersTest\",\"namespace\":\"avro.examples.baseball\",\"doc\":\"Test that extra optional getters are added\",\"fields\":[{\"name\":\"name\",\"type\":\"string\"},{\"name\":\"favorite_number\",\"type\":[\"int\",\"null\"]}]}");
   public static org.apache.avro.Schema getClassSchema() { return SCHEMA$; }
 
@@ -423,6 +425,32 @@ public AddExtraOptionalGettersTest build() {
       }
     }
   }
+
+  @Override
+  public int hashCode() {
+    int result = 1;
+    result = 31 * result + (name == null ? 0 : name.hashCode());
+    result = 31 * result + (favorite_number == null ? 0 : favorite_number.hashCode());
+    return result;
+  }
+
+  @Override
+  public boolean equals(Object o) {
+    if (this == o) {
+      return true;
+    }
+    if (!(o instanceof AddExtraOptionalGettersTest)) {
+      return false;
+    }
+    AddExtraOptionalGettersTest other = (AddExtraOptionalGettersTest) o;
+    if (Utf8.compareSequences(this.name, other.name) != 0) {
+      return false;
+    }
+    if (!java.util.Objects.equals(this.favorite_number, other.favorite_number)) {
+      return false;
+    }
+    return true;
+  }
 }
 
 

lang/java/tools/src/test/compiler/output/NoSettersTest.java

@@ -16,6 +16,8 @@
 @org.apache.avro.specific.AvroGenerated
 public class NoSettersTest extends org.apache.avro.specific.SpecificRecordBase implements org.apache.avro.specific.SpecificRecord {
   private static final long serialVersionUID = 8604146783520861700L;
+
+
   public static final org.apache.avro.Schema SCHEMA$ = new org.apache.avro.Schema.Parser().parse("{\"type\":\"record\",\"name\":\"NoSettersTest\",\"namespace\":\"avro.examples.baseball\",\"doc\":\"Test that setters are omitted\",\"fields\":[{\"name\":\"name\",\"type\":\"string\"},{\"name\":\"favorite_number\",\"type\":[\"int\",\"null\"]}]}");
   public static org.apache.avro.Schema getClassSchema() { return SCHEMA$; }
 
@@ -381,6 +383,32 @@ public NoSettersTest build() {
       }
     }
   }
+
+  @Override
+  public int hashCode() {
+    int result = 1;
+    result = 31 * result + (name == null ? 0 : name.hashCode());
+    result = 31 * result + (favorite_number == null ? 0 : favorite_number.hashCode());
+    return result;
+  }
+
+  @Override
+  public boolean equals(Object o) {
+    if (this == o) {
+      return true;
+    }
+    if (!(o instanceof NoSettersTest)) {
+      return false;
+    }
+    NoSettersTest other = (NoSettersTest) o;
+    if (Utf8.compareSequences(this.name, other.name) != 0) {
+      return false;
+    }
+    if (!java.util.Objects.equals(this.favorite_number, other.favorite_number)) {
+      return false;
+    }
+    return true;
+  }
 }
 
 

lang/java/tools/src/test/compiler/output/OptionalGettersAllFieldsTest.java

@@ -16,6 +16,8 @@
 @org.apache.avro.specific.AvroGenerated
 public class OptionalGettersAllFieldsTest extends org.apache.avro.specific.SpecificRecordBase implements org.apache.avro.specific.SpecificRecord {
   private static final long serialVersionUID = 874861432798554536L;
+
+
   public static final org.apache.avro.Schema SCHEMA$ = new org.apache.avro.Schema.Parser().parse("{\"type\":\"record\",\"name\":\"OptionalGettersAllFieldsTest\",\"namespace\":\"avro.examples.baseball\",\"doc\":\"Test that optional getters are created for all fields\",\"fields\":[{\"name\":\"name\",\"type\":\"string\"},{\"name\":\"nullable_name\",\"type\":[\"string\",\"null\"]},{\"name\":\"favorite_number\",\"type\":[\"int\"]},{\"name\":\"nullable_favorite_number\",\"type\":[\"int\",\"null\"]}]}");
   public static org.apache.avro.Schema getClassSchema() { return SCHEMA$; }
 
@@ -486,6 +488,40 @@ public OptionalGettersAllFieldsTest build() {
     READER$.read(this, SpecificData.getDecoder(in));
   }
 
+
+  @Override
+  public int hashCode() {
+    int result = 1;
+    result = 31 * result + (name == null ? 0 : name.hashCode());
+    result = 31 * result + (nullable_name == null ? 0 : nullable_name.hashCode());
+    result = 31 * result + (favorite_number == null ? 0 : favorite_number.hashCode());
+    result = 31 * result + (nullable_favorite_number == null ? 0 : nullable_favorite_number.hashCode());
+    return result;
+  }
+
+  @Override
+  public boolean equals(Object o) {
+    if (this == o) {
+      return true;
+    }
+    if (!(o instanceof OptionalGettersAllFieldsTest)) {
+      return false;
+    }
+    OptionalGettersAllFieldsTest other = (OptionalGettersAllFieldsTest) o;
+    if (Utf8.compareSequences(this.name, other.name) != 0) {
+      return false;
+    }
+    if (Utf8.compareSequences(this.nullable_name, other.nullable_name) != 0) {
+      return false;
+    }
+    if (!java.util.Objects.equals(this.favorite_number, other.favorite_number)) {
+      return false;
+    }
+    if (!java.util.Objects.equals(this.nullable_favorite_number, other.nullable_favorite_number)) {
+      return false;
+    }
+    return true;
+  }
 }