NUTCH-3191 Fix SonarCloud PR analysis: missing JaCoCo XML, test binaries path, and source encoding warnings

.github/workflows/junit-report.yml

@@ -55,13 +55,15 @@ jobs:
             echo "::error::Download path $root is missing."
             exit 1
           fi
-          files=("$root"/build/**/TEST-*.xml)
+          # upload-artifact strips the common 'build/' prefix from the uploaded
+          # paths, so reports land as $root/test/... and $root/<plugin>/test/...
+          files=("$root"/**/TEST-*.xml)
           if [ ${#files[@]} -eq 0 ] || [ ! -e "${files[0]}" ]; then
-            echo "::error::No TEST-*.xml under $root/build/ (artifact missing, wrong layout, or download failed)."
+            echo "::error::No TEST-*.xml under $root/ (artifact missing, wrong layout, or download failed)."
             find "$root" -maxdepth 5 -type d -print 2>/dev/null | head -80 || true
             exit 1
           fi
-          echo "Found ${#files[@]} JUnit report file(s) under $root/build/."
+          echo "Found ${#files[@]} JUnit report file(s) under $root/."
       - name: Resolve PR number
         id: pr
         run: |
@@ -82,7 +84,7 @@ jobs:
       - name: Publish Test Report
         uses: mikepenz/action-junit-report@v6
         with:
-          report_paths: ./junit-ubuntu-jdk17/build/**/TEST-*.xml
+          report_paths: ./junit-ubuntu-jdk17/**/TEST-*.xml
           check_name: JUnit Test Report
           commit: ${{ github.event.workflow_run.head_sha }}
           fail_on_failure: false

.github/workflows/master-build.yml

@@ -280,6 +280,8 @@ jobs:
             ./build/classes/**/*.class
             # plugin and test class files
             ./build/**/classes/**/*.class
+            # plugin test class files (compiled into build/<plugin>/test/)
+            ./build/*/test/**/*.class
             # dependencies
             ./build/lib/*.jar
             ./build/test/lib/*.jar
@@ -298,6 +300,8 @@ jobs:
         if: always() && matrix.os == 'ubuntu-latest'
         with:
           name: coverage-data-ubuntu-jdk${{ matrix.java }}
-          path: ./build/coverage/*.exec
+          path: |
+            ./build/coverage/*.exec
+            ./build/coverage/jacoco.xml
           retention-days: 1
           if-no-files-found: ignore

.github/workflows/sonarcloud.yml

@@ -79,9 +79,9 @@ jobs:
           name: junit-test-results-ubuntu-latest-jdk17
           workflow: master-build.yml
           run_id: ${{ github.event.workflow_run.id }}
-          path: |
-            ./build/test/TEST-*.xml
-            ./build/**/test/TEST-*.xml
+          # Artifact has build/ stripped (upload-artifact LCA); re-root under build/
+          # so paths match build/test/... and build/<plugin>/test/...
+          path: ./build/
         continue-on-error: true
 
       - name: Flatten test reports (JDK 17 only)

sonar-project.properties

@@ -27,7 +27,7 @@ sonar.links.ci=https://github.com/apache/nutch/actions
 sonar.sources=src/java,src/plugin
 sonar.tests=src/test,src/plugin
 sonar.test.inclusions=**/src/test/**/*.java,**/Test*.java,**/*IT.java
-sonar.exclusions=**/build.xml,**/build-ivy.xml,**/build-plugin.xml,**/ivy.xml,**/plugin.xml
+sonar.exclusions=**/build.xml,**/build-ivy.xml,**/build-plugin.xml,**/ivy.xml,**/plugin.xml,**/sample/**,**/data/**,**/logs/**
 sonar.source.encoding=UTF-8
 sonar.java.source=17
 

src/java/org/apache/nutch/util/StringUtil.java

@@ -151,12 +151,12 @@ public static boolean isEmpty(String str) {
   }
 
   /**
-   * Simple character substitution which cleans/removes all � chars from a given String.
+   * Simple character substitution which cleans/removes all \uFFFD chars from a given String.
    * @param value the String to clean
    * @return substituted cleaned string
    */
   public static String cleanField(String value) {
-    return value.replaceAll("�", "");
+    return value.replaceAll("\uFFFD", "");
   }
 
   /**

src/java/org/apache/nutch/util/TableUtil.java

@@ -146,7 +146,7 @@ public static String unreverseHost(String reversedHostName) {
   }
 
   /**
-   * Convert given Utf8 instance to String and and cleans out any offending "�"
+   * Convert given Utf8 instance to String and and cleans out any offending "\uFFFD"
    * from the String.
    * 
    *