From d3652c5e01478cf7df8ed79145ae892980cdb2dc Mon Sep 17 00:00:00 2001
From: Luca Foppiano <luca@foppiano.org>
Date: Tue, 24 Feb 2026 22:50:33 +0100
Subject: [PATCH 1/5] fix: tests from org.commoncrawl not being executed

---
 build.xml | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/build.xml b/build.xml
index a4530c40f1..29e9896662 100644
--- a/build.xml
+++ b/build.xml
@@ -503,7 +503,24 @@
           <sysproperty key="javax.xml.parsers.DocumentBuilderFactory" value="com.sun.org.apache.xerces.internal.jaxp.DocumentBuilderFactoryImpl"/>
         </fork>
         <fileset dir="${test.build.classes}">
-          <include name="**/Test*.class"/>
+          <include name="org/apache/nutch/**/Test*.class"/>
+          <exclude name="**/${test.exclude}.class"/>
+        </fileset>
+      </testclasses>
+      <!-- org.commoncrawl tests run in their own forked JVM to isolate them
+           from any System.exit() or JVM crashes in the org.apache.nutch batch -->
+      <testclasses outputDir="${test.build.dir}" unless="testcase">
+        <listener type="legacy-plain" sendSysOut="true" sendSysErr="true"/>
+        <listener type="legacy-xml" sendSysOut="true" sendSysErr="true"/>
+        <fork>
+          <jvmarg value="-Xmx1000m"/>
+          <sysproperty key="test.build.data" value="${test.build.data}"/>
+          <sysproperty key="test.src.dir" value="${test.src.dir}"/>
+          <sysproperty key="test.include.slow" value="${test.include.slow}"/>
+          <sysproperty key="javax.xml.parsers.DocumentBuilderFactory" value="com.sun.org.apache.xerces.internal.jaxp.DocumentBuilderFactoryImpl"/>
+        </fork>
+        <fileset dir="${test.build.classes}">
+          <include name="org/commoncrawl/**/Test*.class"/>
           <exclude name="**/${test.exclude}.class"/>
         </fileset>
       </testclasses>

From fd8521e7c6c0e9ca5d16276c51fddd5573e97466 Mon Sep 17 00:00:00 2001
From: Luca Foppiano <luca@foppiano.org>
Date: Wed, 25 Feb 2026 13:38:53 +0100
Subject: [PATCH 2/5] feat: add test summaries

---
 .github/workflows/cc-build.yml | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/.github/workflows/cc-build.yml b/.github/workflows/cc-build.yml
index e382c8771a..5c55a4612e 100644
--- a/.github/workflows/cc-build.yml
+++ b/.github/workflows/cc-build.yml
@@ -55,3 +55,21 @@ jobs:
           curl https://publicsuffix.org/list/public_suffix_list.dat -o conf/effective_tld_names.dat
       - name: Test
         run: ant clean test -buildfile build.xml
+      - name: Upload Test Results
+        if: always()
+        uses: actions/upload-artifact@v4
+        with:
+          name: test-results
+          path: |
+            ./build/test/TEST-*.xml
+            ./build/**/test/TEST-*.xml
+          retention-days: 1
+      - name: Publish Test Results
+        if: always()
+        uses: EnricoMi/publish-unit-test-result-action@v2
+        with:
+          files: |
+            ./build/test/TEST-*.xml
+            ./build/**/test/TEST-*.xml
+          comment_mode: always
+          fail_on: "test failures"

From df5a0addd806cd92953e7c78a61b53f138f3db28 Mon Sep 17 00:00:00 2001
From: Luca Foppiano <luca@foppiano.org>
Date: Wed, 25 Feb 2026 14:59:26 +0100
Subject: [PATCH 3/5] feat: remove System.exit() inside the code

---
 src/java/org/apache/nutch/tools/CommonCrawlDataDumper.java | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/java/org/apache/nutch/tools/CommonCrawlDataDumper.java b/src/java/org/apache/nutch/tools/CommonCrawlDataDumper.java
index d5d5035e89..32b3b1f369 100644
--- a/src/java/org/apache/nutch/tools/CommonCrawlDataDumper.java
+++ b/src/java/org/apache/nutch/tools/CommonCrawlDataDumper.java
@@ -274,7 +274,7 @@ public void dump(File outputDir, File segmentRootDir, File linkdb, boolean gzip,
     if (parts == null || parts.size() == 0) {
       LOG.error( "No segment directories found in {} ",
           segmentRootDir.getAbsolutePath());
-      System.exit(1);
+      return;
     }
     LOG.info("Found {} segment parts", parts.size());
     if (gzip && !warc) {

From 75601ba81fe815c462b7f48770b54a48876c78fd Mon Sep 17 00:00:00 2001
From: Luca Foppiano <luca@foppiano.org>
Date: Wed, 25 Feb 2026 15:00:50 +0100
Subject: [PATCH 4/5] fix: support tests that end with Test

---
 build.xml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/build.xml b/build.xml
index 29e9896662..c117ac38c3 100644
--- a/build.xml
+++ b/build.xml
@@ -504,6 +504,7 @@
         </fork>
         <fileset dir="${test.build.classes}">
           <include name="org/apache/nutch/**/Test*.class"/>
+          <include name="org/apache/nutch/**/*Test.class"/>
           <exclude name="**/${test.exclude}.class"/>
         </fileset>
       </testclasses>

From 3bcac7983f5a1a8d81da90709de263276b1a9155 Mon Sep 17 00:00:00 2001
From: Luca Foppiano <luca@foppiano.org>
Date: Wed, 25 Feb 2026 17:05:51 +0100
Subject: [PATCH 5/5] fix: enable fork per test

---
 build.xml | 22 ++--------------------
 1 file changed, 2 insertions(+), 20 deletions(-)

diff --git a/build.xml b/build.xml
index c117ac38c3..60b3ebde79 100644
--- a/build.xml
+++ b/build.xml
@@ -495,25 +495,7 @@
       <testclasses outputDir="${test.build.dir}" unless="testcase">
         <listener type="legacy-plain" sendSysOut="true" sendSysErr="true"/>
         <listener type="legacy-xml" sendSysOut="true" sendSysErr="true"/>
-        <fork>
-          <jvmarg value="-Xmx1000m"/>
-          <sysproperty key="test.build.data" value="${test.build.data}"/>
-          <sysproperty key="test.src.dir" value="${test.src.dir}"/>
-          <sysproperty key="test.include.slow" value="${test.include.slow}"/>
-          <sysproperty key="javax.xml.parsers.DocumentBuilderFactory" value="com.sun.org.apache.xerces.internal.jaxp.DocumentBuilderFactoryImpl"/>
-        </fork>
-        <fileset dir="${test.build.classes}">
-          <include name="org/apache/nutch/**/Test*.class"/>
-          <include name="org/apache/nutch/**/*Test.class"/>
-          <exclude name="**/${test.exclude}.class"/>
-        </fileset>
-      </testclasses>
-      <!-- org.commoncrawl tests run in their own forked JVM to isolate them
-           from any System.exit() or JVM crashes in the org.apache.nutch batch -->
-      <testclasses outputDir="${test.build.dir}" unless="testcase">
-        <listener type="legacy-plain" sendSysOut="true" sendSysErr="true"/>
-        <listener type="legacy-xml" sendSysOut="true" sendSysErr="true"/>
-        <fork>
+        <fork forkMode="perTestClass">
           <jvmarg value="-Xmx1000m"/>
           <sysproperty key="test.build.data" value="${test.build.data}"/>
           <sysproperty key="test.src.dir" value="${test.src.dir}"/>
@@ -521,7 +503,7 @@
           <sysproperty key="javax.xml.parsers.DocumentBuilderFactory" value="com.sun.org.apache.xerces.internal.jaxp.DocumentBuilderFactoryImpl"/>
         </fork>
         <fileset dir="${test.build.classes}">
-          <include name="org/commoncrawl/**/Test*.class"/>
+          <include name="**/Test*.class"/>
           <exclude name="**/${test.exclude}.class"/>
         </fileset>
       </testclasses>