Infra: Fix jvm coverage report

Signed-off-by: Arthur Chan <[email protected]>

Author: arthurscchan

infra/base-images/base-runner/coverage

@@ -457,6 +457,18 @@ elif [[ $FUZZING_LANGUAGE == "jvm" ]]; then
     then
       tempdir=$(mktemp -d)
       cd $tempdir && jar xvf $jar_file && cd -
+
+      # Remove class files that are not compiled from the java files found in $OUT/$SRC before copying
+      find $tempdir -type f -name "*.class" | while read -r class_file; do
+        relative_path=${class_file#$tempdir/}
+        java_file_name=${relative_path%.class}.java
+        if ! find $OUT$SRC -type f -name $(basename $java_file_name) | grep -q $java_file_name
+        then
+          echo "Removing: $class_file"
+          rm -f "$class_file"
+        fi
+      done
+
       cp -r $tempdir/* $classes_dir/
       rm -r $tempdir
     fi
@@ -473,6 +485,8 @@ elif [[ $FUZZING_LANGUAGE == "jvm" ]]; then
   # Remove all class files which have a dot in their file name which are normally
   # used to name a duplication of the legitimate class file.
   find $classes_dir/*/ -type f -name "*.*.class" | xargs rm -f
+  # Clean up empty package directories
+#  find $classes_dir -type d -empty -exec rmdir -p {} + 2>/dev/null
 
   # Heuristically determine source directories based on Maven structure.
   # Always include the $SRC root as it likely contains the fuzzer sources.