(tika) branch main updated: TIKA-4875: leave H2 chunk retention at its default in tika-eval (#3150)

[tallison]

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

tballison pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/tika.git


The following commit(s) were added to refs/heads/main by this push:
     new ac83dc9e63 TIKA-4875: leave H2 chunk retention at its default in 
tika-eval (#3150)
ac83dc9e63 is described below

commit ac83dc9e63c0087673370ce230299226ca7254e8
Author: Tim Allison <talli...@apache.org>
AuthorDate: Thu Sep 10 06:37:13 2026 -0400

    TIKA-4875: leave H2 chunk retention at its default in tika-eval (#3150)
---
 CHANGES.txt                                                      | 9 ++++-----
 .../src/main/java/org/apache/tika/eval/app/db/JDBCUtil.java      | 8 ++++----
 .../src/test/java/org/apache/tika/eval/app/db/JDBCUtilTest.java  | 4 +++-
 3 files changed, 11 insertions(+), 10 deletions(-)

diff --git a/CHANGES.txt b/CHANGES.txt
index b1de57d4f2..e934fbc780 100644
--- a/CHANGES.txt
+++ b/CHANGES.txt
@@ -93,11 +93,10 @@ Release 4.1.0 - unreleased
      
    * tika-eval Profile/Compare speedups: single-pass URL/mail stripping
      replaces the bounded regexes in langdetect preprocessing (same output,
-     17-290x faster on web text), the default H2 db URL drops MVStore chunk
-     retention and sizes the page cache at a quarter of the heap clamped to
-     [64MB, 1GB] (override with -Dtika.eval.h2.cacheSizeKb=<kb>), and the
-     status log adds a last-interval docs-per-sec rate next to the cumulative
-     average (TIKA-4875).
+     17-290x faster on web text), the default H2 db URL sizes the page cache
+     at a quarter of the heap clamped to [64MB, 1GB] (override with
+     -Dtika.eval.h2.cacheSizeKb=<kb>), and the status log adds a last-interval
+     docs-per-sec rate next to the cumulative average (TIKA-4875).
      
    * New "content-enrichers" config list (TIKA-4872): select the OCR engine
      ("tesseract-ocr-parser", "tess4j-parser", "openai-vlm-parser", ...) by
diff --git 
a/tika-eval/tika-eval-app/src/main/java/org/apache/tika/eval/app/db/JDBCUtil.java
 
b/tika-eval/tika-eval-app/src/main/java/org/apache/tika/eval/app/db/JDBCUtil.java
index 43e439722d..45f355ec2e 100644
--- 
a/tika-eval/tika-eval-app/src/main/java/org/apache/tika/eval/app/db/JDBCUtil.java
+++ 
b/tika-eval/tika-eval-app/src/main/java/org/apache/tika/eval/app/db/JDBCUtil.java
@@ -88,16 +88,16 @@ public class JDBCUtil {
 
     /**
      * If dbPath is already a jdbc string, it is used as is; otherwise this 
builds the
-     * tika-eval h2 default: RETENTION_TIME=0 drops the 45s MVStore chunk 
retention
-     * (bloat + growing compaction cost) and CACHE_SIZE (KB) is sized by
-     * {@link #getH2CacheSizeKb()}.
+     * tika-eval h2 default: CACHE_SIZE (KB) sized by {@link 
#getH2CacheSizeKb()}.
+     * Chunk retention stays at H2's 45s default: RETENTION_TIME=0 let dead 
chunks be
+     * dropped under the background writer (assert in 
FileStore.serializeAndStore).
      */
     public static String getJdbcConnectionString(String dbPath) {
         if (dbPath.startsWith("jdbc:")) {
             return dbPath;
         }
         Path p = Paths.get(dbPath);
-        return "jdbc:h2:file:" + p.toAbsolutePath() + 
";RETENTION_TIME=0;CACHE_SIZE=" + getH2CacheSizeKb();
+        return "jdbc:h2:file:" + p.toAbsolutePath() + ";CACHE_SIZE=" + 
getH2CacheSizeKb();
     }
 
     /**
diff --git 
a/tika-eval/tika-eval-app/src/test/java/org/apache/tika/eval/app/db/JDBCUtilTest.java
 
b/tika-eval/tika-eval-app/src/test/java/org/apache/tika/eval/app/db/JDBCUtilTest.java
index 79df461426..ce92c10acc 100644
--- 
a/tika-eval/tika-eval-app/src/test/java/org/apache/tika/eval/app/db/JDBCUtilTest.java
+++ 
b/tika-eval/tika-eval-app/src/test/java/org/apache/tika/eval/app/db/JDBCUtilTest.java
@@ -17,6 +17,7 @@
 package org.apache.tika.eval.app.db;
 
 import static org.junit.jupiter.api.Assertions.assertEquals;
+import static org.junit.jupiter.api.Assertions.assertFalse;
 import static org.junit.jupiter.api.Assertions.assertTrue;
 
 import org.junit.jupiter.api.AfterEach;
@@ -54,7 +55,8 @@ public class JDBCUtilTest {
         assertTrue(cacheSizeKb >= 65_536L && cacheSizeKb <= 1_048_576L, 
"clamped to [64MB, 1GB]: " + cacheSizeKb);
         String connectionString = JDBCUtil.getJdbcConnectionString("mydb");
         assertTrue(connectionString.startsWith("jdbc:h2:file:"), 
connectionString);
-        assertTrue(connectionString.endsWith(";RETENTION_TIME=0;CACHE_SIZE=" + 
cacheSizeKb), connectionString);
+        assertTrue(connectionString.endsWith(";CACHE_SIZE=" + cacheSizeKb), 
connectionString);
+        assertFalse(connectionString.contains("RETENTION_TIME"), 
connectionString);
     }
 
     @Test