Merge pull request #236 from blemale/bastien.lemale/perf_do_not_allocate_computing_metadata_size

perf: Do not allocate when computing metadata size

Author: vickenty

src/main/java/com/timgroup/statsd/NonBlockingDirectStatsDClient.java

@@ -64,10 +64,10 @@ public final boolean writeTo(StringBuilder builder, int capacity, String contain
         private int metadataSize(StringBuilder builder, String containerID) {
             if (metadataSize == -1) {
                 final int previousLength = builder.length();
-                final int previousEncodedLength = builder.toString().getBytes(UTF_8).length;
+                final int previousEncodedLength = Utf8.encodedLength(builder);
                 writeHeadMetadata(builder);
                 writeTailMetadata(builder, containerID);
-                metadataSize = builder.toString().getBytes(UTF_8).length - previousEncodedLength;
+                metadataSize = Utf8.encodedLength(builder) - previousEncodedLength;
                 builder.setLength(previousLength);
             }
             return metadataSize;

src/main/java/com/timgroup/statsd/Utf8.java

@@ -0,0 +1,96 @@
+/*
+ * Copyright (C) 2013 The Guava Authors
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except
+ * in compliance with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software distributed under the License
+ * is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express
+ * or implied. See the License for the specific language governing permissions and limitations under
+ * the License.
+ */
+
+package com.timgroup.statsd;
+
+import static java.lang.Character.MAX_SURROGATE;
+import static java.lang.Character.MIN_SURROGATE;
+
+import java.nio.charset.StandardCharsets;
+
+/**
+ * This class is a partial copy of the {@code com.google.common.base.Utf8}
+ * <a href="https://github.com/google/guava/blob/v33.0.0/guava/src/com/google/common/base/Utf8.java">class</a>
+ * from the Guava library.
+ * It is copied here to avoid a dependency on Guava.
+ */
+final class Utf8 {
+
+    private static final int UTF8_REPLACEMENT_LENGTH = StandardCharsets.UTF_8.newEncoder().replacement().length;
+
+    private Utf8() {
+    }
+
+    /**
+     * Returns the number of bytes in the UTF-8-encoded form of {@code sequence}. For a string, this
+     * method is equivalent to {@code string.getBytes(UTF_8).length}, but is more efficient in both
+     * time and space.
+     *
+     * @throws IllegalArgumentException if {@code sequence} contains ill-formed UTF-16 (unpaired
+     *                                  surrogates)
+     */
+    public static int encodedLength(CharSequence sequence) {
+        // Warning to maintainers: this implementation is highly optimized.
+        int utf16Length = sequence.length();
+        int utf8Length = utf16Length;
+        int index = 0;
+
+        // This loop optimizes for pure ASCII.
+        while (index < utf16Length && sequence.charAt(index) < 0x80) {
+            index++;
+        }
+
+        // This loop optimizes for chars less than 0x800.
+        for (; index < utf16Length; index++) {
+            char character = sequence.charAt(index);
+            if (character < 0x800) {
+                utf8Length += ((0x7f - character) >>> 31); // branch free!
+            } else {
+                utf8Length += encodedLengthGeneral(sequence, index);
+                break;
+            }
+        }
+
+        if (utf8Length < utf16Length) {
+            // Necessary and sufficient condition for overflow because of maximum 3x expansion
+            throw new IllegalArgumentException(
+                    "UTF-8 length does not fit in int: " + (utf8Length + (1L << 32)));
+        }
+        return utf8Length;
+    }
+
+    private static int encodedLengthGeneral(CharSequence sequence, int start) {
+        int utf16Length = sequence.length();
+        int utf8Length = 0;
+        for (int index = start; index < utf16Length; index++) {
+            char character = sequence.charAt(index);
+            if (character < 0x800) {
+                utf8Length += (0x7f - character) >>> 31; // branch free!
+            } else {
+                utf8Length += 2;
+                // jdk7+: if (Character.isSurrogate(character)) {
+                if (MIN_SURROGATE <= character && character <= MAX_SURROGATE) {
+                    // Check that we have a well-formed surrogate pair.
+                    if (Character.codePointAt(sequence, index) == character) {
+                        // Bad input so deduct char length and account for the replacement characters
+                        utf8Length += -2 + UTF8_REPLACEMENT_LENGTH - 1;
+                    } else {
+                        index++;
+                    }
+                }
+            }
+        }
+        return utf8Length;
+    }
+}

src/test/java/com/timgroup/statsd/Utf8Test.java

@@ -0,0 +1,32 @@
+package com.timgroup.statsd;
+
+import org.junit.Test;
+
+import java.nio.ByteBuffer;
+import java.nio.CharBuffer;
+import java.nio.charset.CharacterCodingException;
+import java.nio.charset.CharsetEncoder;
+import java.nio.charset.CodingErrorAction;
+import java.nio.charset.StandardCharsets;
+
+import static java.lang.Character.MIN_SURROGATE;
+import static org.hamcrest.MatcherAssert.assertThat;
+import static org.hamcrest.Matchers.equalTo;
+
+public class Utf8Test {
+
+    @Test
+    public void should_handle_malformed_inputs() throws CharacterCodingException {
+        shouldHandleMalformedInput("foo" + MIN_SURROGATE + "bar");
+        shouldHandleMalformedInput("🍻☀️😎🏖️" + MIN_SURROGATE + "🍻☀️😎🏖️");
+    }
+
+    private static void shouldHandleMalformedInput(String malformedInput) throws CharacterCodingException {
+        CharsetEncoder utf8Encoder = StandardCharsets.UTF_8.newEncoder()
+                .onMalformedInput(CodingErrorAction.REPLACE)
+                .onUnmappableCharacter(CodingErrorAction.REPLACE);
+        ByteBuffer encoded = utf8Encoder.encode(CharBuffer.wrap(malformedInput));
+
+        assertThat(Utf8.encodedLength(malformedInput), equalTo(encoded.limit()));
+    }
+}