(pulsar) branch master updated: [improve][misc] Optimize topic list hashing so that potentially large String allocation is avoided (#24525)

[lhotari]

This is an automated email from the ASF dual-hosted git repository.

lhotari pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/pulsar.git


The following commit(s) were added to refs/heads/master by this push:
     new 06f424c4823 [improve][misc] Optimize topic list hashing so that 
potentially large String allocation is avoided (#24525)
06f424c4823 is described below

commit 06f424c4823399ae66608ad9fd0e2133a1055b1a
Author: Lari Hotari <lhot...@users.noreply.github.com>
AuthorDate: Thu Jul 17 10:18:57 2025 +0300

    [improve][misc] Optimize topic list hashing so that potentially large 
String allocation is avoided (#24525)
---
 .../org/apache/pulsar/common/topics/TopicList.java   | 20 ++++++++++++++------
 .../apache/pulsar/common/topics/TopicListTest.java   |  7 +++++++
 2 files changed, 21 insertions(+), 6 deletions(-)

diff --git 
a/pulsar-common/src/main/java/org/apache/pulsar/common/topics/TopicList.java 
b/pulsar-common/src/main/java/org/apache/pulsar/common/topics/TopicList.java
index 98e4d086503..b3dcc7d09b7 100644
--- a/pulsar-common/src/main/java/org/apache/pulsar/common/topics/TopicList.java
+++ b/pulsar-common/src/main/java/org/apache/pulsar/common/topics/TopicList.java
@@ -18,9 +18,11 @@
  */
 package org.apache.pulsar.common.topics;
 
+import com.google.common.hash.Hasher;
 import com.google.common.hash.Hashing;
 import com.google.re2j.Pattern;
 import java.nio.charset.StandardCharsets;
+import java.util.Arrays;
 import java.util.Collection;
 import java.util.HashSet;
 import java.util.List;
@@ -46,6 +48,7 @@ public class TopicList {
         Pattern topicsPattern = Pattern.compile(regex);
         return filterTopics(original, topicsPattern);
     }
+
     public static List<String> filterTopics(List<String> original, Pattern 
topicsPattern) {
 
 
@@ -69,14 +72,19 @@ public class TopicList {
     }
 
     public static String calculateHash(List<String> topics) {
-        return Hashing.crc32c().hashBytes(topics.stream()
-                .sorted()
-                .collect(Collectors.joining(","))
-                .getBytes(StandardCharsets.UTF_8)).toString();
+        Hasher hasher = Hashing.crc32c().newHasher();
+        String[] sortedTopics = topics.toArray(new String[topics.size()]);
+        Arrays.sort(sortedTopics);
+        for (int i = 0; i < sortedTopics.length; i++) {
+            hasher.putString(sortedTopics[i], StandardCharsets.UTF_8);
+            // Skip the delimiter for the last item so that the hash format is 
compatible with previous versions
+            if (i < sortedTopics.length - 1) {
+                hasher.putByte((byte) ',');
+            }
+        }
+        return hasher.hash().toString();
     }
 
-
-
     // get topics, which are contained in list1, and not in list2
     public static Set<String> minus(Collection<String> list1, 
Collection<String> list2) {
         HashSet<String> s1 = new HashSet<>(list1);
diff --git 
a/pulsar-common/src/test/java/org/apache/pulsar/common/topics/TopicListTest.java
 
b/pulsar-common/src/test/java/org/apache/pulsar/common/topics/TopicListTest.java
index ba52a5ef1bc..a62aa012c7f 100644
--- 
a/pulsar-common/src/test/java/org/apache/pulsar/common/topics/TopicListTest.java
+++ 
b/pulsar-common/src/test/java/org/apache/pulsar/common/topics/TopicListTest.java
@@ -25,6 +25,7 @@ import static org.testng.Assert.fail;
 import com.google.common.collect.Lists;
 import com.google.re2j.Pattern;
 import java.util.Arrays;
+import java.util.Collections;
 import java.util.List;
 import java.util.Set;
 import java.util.stream.Stream;
@@ -100,10 +101,16 @@ public class TopicListTest {
         String hash1 = TopicList.calculateHash(Arrays.asList(topicName3, 
topicName2, topicName1));
         String hash2 = TopicList.calculateHash(Arrays.asList(topicName1, 
topicName3, topicName2));
         assertEquals(hash1, hash2, "Hash must not depend on order of topics in 
the list");
+        assertEquals(hash1, "90d4a04a", "Hash must be equal to the expected 
value");
 
         String hash3 = TopicList.calculateHash(Arrays.asList(topicName1, 
topicName2));
         assertNotEquals(hash1, hash3, "Different list must have different 
hashes");
 
+        String hash4 = TopicList.calculateHash(Arrays.asList(topicName1));
+        assertEquals(hash4, "0d0602ed", "Hash must be equal to the expected 
value");
+
+        String hash5 = TopicList.calculateHash(Collections.emptyList());
+        assertEquals(hash5, "00000000", "Hash of empty list must be 0");
     }
 
     @Test