Cache label identifier data to eliminate redundant parsing

CLAUDE.md

@@ -23,10 +23,10 @@ mvn package site
 # Build without running tests (useful when tests fail with extra logging)
 mvn clean package -DskipTests
 
-# Run tests only
+# Run full test suite (always use this to verify changes)
 mvn test
 
-# Run tests excluding ReferenceIntegrityTest with specific Cucumber tags
+# Run a targeted subset during development (NOT for verifying changes — excludes ReferenceIntegrityTest)
 mvn test -Dtest=\!ReferenceIntegrityTest* -Dcucumber.filter.tags='@v3.7.x'
 
 # Run a single Cucumber scenario by name (issue number and subtest, e.g. #1458-1)
@@ -172,7 +172,7 @@ Alternatively, use `--verbose 1` flag for INFO level output in reports (default
 - Source/Target: Java 17 (pom.xml:172-173)
 
 ### Main Dependencies
-- **pds4-jparser** (version 2.11.0) - PDS4 label parsing and validation
+- **pds4-jparser** (version 3.2.0-SNAPSHOT) - PDS4 label parsing and validation; **source is at `../pds4-jparser`** — check there first when tracing errors from pds4-jparser classes (e.g. `FieldValueValidator`, `ProblemType`, parser internals)
 - **pds3-product-tools** (version 4.4.2) - PDS3 validation
 - **Saxon-HE** (version 12.9) - XSLT and XPath processing
 - **registry-common** (version 2.1.0-SNAPSHOT) - Registry API integration

src/main/java/gov/nasa/pds/tools/util/LabelCacheEntry.java

@@ -0,0 +1,32 @@
+package gov.nasa.pds.tools.util;
+
+import java.util.ArrayList;
+
+/**
+ * Immutable cache of identifier data extracted from a parsed label, used to avoid re-parsing
+ * during the referential integrity phase.
+ */
+public class LabelCacheEntry {
+  private final ArrayList<String> logicalIdentifiers;
+  private final ArrayList<String> lidOrLidVidReferences;
+  private final ArrayList<String> contextAreaRefs;
+
+  public LabelCacheEntry(ArrayList<String> logicalIdentifiers,
+      ArrayList<String> lidOrLidVidReferences, ArrayList<String> contextAreaRefs) {
+    this.logicalIdentifiers = logicalIdentifiers;
+    this.lidOrLidVidReferences = lidOrLidVidReferences;
+    this.contextAreaRefs = contextAreaRefs;
+  }
+
+  public ArrayList<String> getLogicalIdentifiers() {
+    return logicalIdentifiers;
+  }
+
+  public ArrayList<String> getLidOrLidVidReferences() {
+    return lidOrLidVidReferences;
+  }
+
+  public ArrayList<String> getContextAreaRefs() {
+    return contextAreaRefs;
+  }
+}

src/main/java/gov/nasa/pds/tools/util/LabelUtil.java

@@ -365,14 +365,20 @@ public static synchronized ArrayList<String> getIdentifiersCommon(DOMSource sour
    * @return lidOrLidVidReference The LID or LIDVID referenced in this label.
    */
   public static ArrayList<String> getLidVidReferences(DOMSource source, URL context) {
+    return getLidVidReferences(source, context, true);
+  }
+
+  public static ArrayList<String> getLidVidReferences(DOMSource source, URL context,
+      boolean reportCarriageReturns) {
     LOG.debug("getLidVidReferences:MY_SOURCE[{}]", source);
 
     String[] tagsList = new String[2];
     tagsList[0] = LIDVID_REFERENCE;
     tagsList[1] = LID_REFERENCE;
 
     ArrayList<String> lidOrLidVidReferences =
-        LabelUtil.getIdentifiersCommon(source, context, tagsList, INTERNAL_REFERENCE_AREA);
+        LabelUtil.getIdentifiersCommon(source, context, tagsList, INTERNAL_REFERENCE_AREA,
+            reportCarriageReturns);
 
     LOG.debug("getLidVidReferences:context,lidOrLidVidReferences {},{}", context,
         lidOrLidVidReferences);
@@ -387,13 +393,19 @@ public static ArrayList<String> getLidVidReferences(DOMSource source, URL contex
    * @return logicalIdentifiers A list of logical identifiers in this label.
    */
   public static ArrayList<String> getLogicalIdentifiers(DOMSource source, URL context) {
+    return getLogicalIdentifiers(source, context, true);
+  }
+
+  public static ArrayList<String> getLogicalIdentifiers(DOMSource source, URL context,
+      boolean reportCarriageReturns) {
     LOG.debug("getLogicalIdentifiers:MY_SOURCE[{}]", source);
 
     String[] tagsList = new String[1];
     tagsList[0] = LOGICAL_IDENTIFIER_TAG;
 
     ArrayList<String> logicalIdentifiers =
-        LabelUtil.getIdentifiersCommon(source, context, tagsList, IDENTIFICATION_AREA);
+        LabelUtil.getIdentifiersCommon(source, context, tagsList, IDENTIFICATION_AREA,
+            reportCarriageReturns);
 
     LOG.debug("getLogicalIdentifiers:context,logicalIdentifiers {},{}", context,
         logicalIdentifiers);

src/main/java/gov/nasa/pds/tools/util/ReferentialIntegrityUtil.java

@@ -142,6 +142,7 @@ public class ReferentialIntegrityUtil {
   private static String[] tagsList = new String[2];
   private static HashSet<String> reportedErrorsReferenceSet = new HashSet<>();
   private static URL parentBundleURL = null;
+  private static HashMap<URL, LabelCacheEntry> labelIdentifierCache = new HashMap<>();
 
   /**
    * Initialize this class to ready for doing referential checks.
@@ -163,6 +164,14 @@ public static void initialize(String referenceType, URL target, ProblemListener
     ReferentialIntegrityUtil.tagsList[1] = LabelUtil.LID_REFERENCE;
   }
 
+  public static void cacheLabelIdentifiers(URL url, LabelCacheEntry entry) {
+    labelIdentifierCache.put(url, entry);
+  }
+
+  public static LabelCacheEntry getCachedLabelIdentifiers(URL url) {
+    return labelIdentifierCache.get(url);
+  }
+
   /**
    * Reset this class to its initial state in case running from regression tests.
    *
@@ -188,6 +197,7 @@ public static void reset() {
     ReferentialIntegrityUtil.urlsParsedCumulative.clear();
     ReferentialIntegrityUtil.reportedErrorsReferenceSet.clear();
     ReferentialIntegrityUtil.parentBundleURL = null;
+    ReferentialIntegrityUtil.labelIdentifierCache.clear();
   }
 
   /**
@@ -577,6 +587,30 @@ private static void addUniqueReferencesToMap(HashMap<String, HashSetReferenceInf
         ReferentialIntegrityUtil.getReferenceType(), parentId, url, numReferencesAdded);
   }
 
+  private static void collectAllContextReferences(ArrayList<String> allContextAreaRefs,
+      ArrayList<String> logicalIdentifiers, ArrayList<String> lidOrLidVidReferences,
+      boolean labelIsBundleFlag, boolean labelIsCollectionFlag, URL url) {
+    if ((logicalIdentifiers == null) || logicalIdentifiers.isEmpty()) {
+      return;
+    }
+    if (labelIsCollectionFlag || labelIsBundleFlag) {
+      if (labelIsBundleFlag) {
+        ReferentialIntegrityUtil.addUniqueReferencesToMap(ReferentialIntegrityUtil.bundleReferenceMap,
+            allContextAreaRefs, url, logicalIdentifiers.get(0));
+      } else {
+        ReferentialIntegrityUtil.addUniqueReferencesToMap(
+            ReferentialIntegrityUtil.collectionReferenceMap, allContextAreaRefs, url,
+            logicalIdentifiers.get(0));
+      }
+    } else {
+      ReferentialIntegrityUtil.addUniqueReferencesToMap(
+          ReferentialIntegrityUtil.contextReferencesCumulative, allContextAreaRefs, url,
+          logicalIdentifiers.get(0));
+    }
+    LOG.debug("collectAllContextReferences:url,contextReferencesCumulative {},{},{}", url,
+        contextReferencesCumulative, contextReferencesCumulative.size());
+  }
+
   private static void collectAllContextReferences(DOMSource domSource,
       ArrayList<String> logicalIdentifiers, ArrayList<String> lidOrLidVidReferences,
       boolean labelIsBundleFlag, boolean labelIsCollectionFlag, URL url) {
@@ -786,16 +820,23 @@ public static void additionalReferentialIntegrityChecks(URL crawlTarget, URL bun
           if (TargetExaminer.isTargetCollectionType (child.getUrl())) {
             labelIsCollectionFlag = true;
           }
-          xml = db.parse(url.openStream());
-          domSource = new DOMSource(xml);
-          // Note that the function getLidVidReferences() collects all Internal_Reference
-          // elements in the PDS4 core namespace (including those in Reference_List,
-          // Context_Area, discipline LDD areas, and any other location in the label).
-          // so the lidOrLidVidReferencesCumulative will be a cumulative collection of all
-          // references collected in lidOrLidVidReferences for each label.
-
-          ArrayList<String> lidOrLidVidReferences = LabelUtil.getLidVidReferences(domSource, url);
-          ArrayList<String> logicalIdentifiers = LabelUtil.getLogicalIdentifiers(domSource, url);
+          LabelCacheEntry cached = ReferentialIntegrityUtil.getCachedLabelIdentifiers(url);
+          ArrayList<String> lidOrLidVidReferences;
+          ArrayList<String> logicalIdentifiers;
+          if (cached != null) {
+            logicalIdentifiers = cached.getLogicalIdentifiers();
+            lidOrLidVidReferences = cached.getLidOrLidVidReferences();
+          } else {
+            xml = db.parse(url.openStream());
+            domSource = new DOMSource(xml);
+            // Note that the function getLidVidReferences() collects all Internal_Reference
+            // elements in the PDS4 core namespace (including those in Reference_List,
+            // Context_Area, discipline LDD areas, and any other location in the label).
+            // so the lidOrLidVidReferencesCumulative will be a cumulative collection of all
+            // references collected in lidOrLidVidReferences for each label.
+            lidOrLidVidReferences = LabelUtil.getLidVidReferences(domSource, url);
+            logicalIdentifiers = LabelUtil.getLogicalIdentifiers(domSource, url);
+          }
           LOG.debug("additionalReferentialIntegrityChecks:url,lidOrLidVidReferences {},{}", url,
               lidOrLidVidReferences.size());
           LOG.debug("additionalReferentialIntegrityChecks:url,logicalIdentifiers {},{}", url,
@@ -871,8 +912,14 @@ public static void additionalReferentialIntegrityChecks(URL crawlTarget, URL bun
           // Collect all the context references defined for each label under the
           // "Context_Area" tag.
           if (ReferentialIntegrityUtil.contextReferenceCheck) {
-            ReferentialIntegrityUtil.collectAllContextReferences(domSource, logicalIdentifiers,
-                lidOrLidVidReferences, labelIsBundleFlag, labelIsCollectionFlag, url);
+            if (cached != null) {
+              ReferentialIntegrityUtil.collectAllContextReferences(cached.getContextAreaRefs(),
+                  logicalIdentifiers, lidOrLidVidReferences, labelIsBundleFlag,
+                  labelIsCollectionFlag, url);
+            } else {
+              ReferentialIntegrityUtil.collectAllContextReferences(domSource, logicalIdentifiers,
+                  lidOrLidVidReferences, labelIsBundleFlag, labelIsCollectionFlag, url);
+            }
           }
 
         } else {

src/main/java/gov/nasa/pds/tools/validate/CrossLabelFileAreaReferenceChecker.java

@@ -3,6 +3,7 @@
 import java.io.IOException;
 import java.net.URISyntaxException;
 import java.nio.file.Path;
+import java.util.ArrayList;
 import java.util.Arrays;
 import java.util.HashMap;
 import java.util.List;
@@ -11,7 +12,9 @@
 import javax.xml.transform.dom.DOMSource;
 import org.w3c.dom.Document;
 import org.xml.sax.SAXException;
+import gov.nasa.pds.tools.util.LabelCacheEntry;
 import gov.nasa.pds.tools.util.LabelUtil;
+import gov.nasa.pds.tools.util.ReferentialIntegrityUtil;
 
 public class CrossLabelFileAreaReferenceChecker {
   final private static HashMap<String,Boolean> isObservational = new HashMap<String,Boolean>();
@@ -26,24 +29,25 @@ private static String resolve (String name, ValidationTarget target) throws URIS
    * @param name - the file being referenced by the file area
    * @param target - the label being validated
    * @return true if name and target are unique and only known references and false otherwise
-   * All of these exceptions should not happen because they are getting the LID from the target.
-   * Would not have made it this far if that could not have happened already, So, just pass them
-   * back to the called and let them handle it with their own generic exception handler/message.
-   * @throws IOException
-   * @throws ParserConfigurationException
-   * @throws SAXException
    * @throws URISyntaxException
    */
-  public static boolean add (String name, ValidationTarget target, boolean isObs)
+  public static boolean add(String name, ValidationTarget target, boolean isObs)
       throws IOException, ParserConfigurationException, SAXException, URISyntaxException {
     boolean success = false;
-    DocumentBuilderFactory dbf = DocumentBuilderFactory.newInstance();
-    Document xml = dbf.newDocumentBuilder().parse(target.getUrl().openStream());
-    DOMSource domSource = new DOMSource(xml);
     String full_name = resolve(name, target);
     isObservational.put(full_name,
         isObs || (isObservational.containsKey(full_name) ? isObservational.get(full_name) : false));
-    for (String lid : LabelUtil.getLogicalIdentifiers (domSource, target.getUrl())) {
+    ArrayList<String> logicalIdentifiers;
+    LabelCacheEntry cached = ReferentialIntegrityUtil.getCachedLabelIdentifiers(target.getUrl());
+    if (cached != null) {
+      logicalIdentifiers = cached.getLogicalIdentifiers();
+    } else {
+      DocumentBuilderFactory dbf = DocumentBuilderFactory.newInstance();
+      Document xml = dbf.newDocumentBuilder().parse(target.getUrl().openStream());
+      DOMSource domSource = new DOMSource(xml);
+      logicalIdentifiers = LabelUtil.getLogicalIdentifiers(domSource, target.getUrl());
+    }
+    for (String lid : logicalIdentifiers) {
       if (lid.contains("::")) {
         lid = lid.substring (0, lid.indexOf("::"));
       }
@@ -64,5 +68,6 @@ public static String getOtherFilename (String name, ValidationTarget target) thr
   }
   public static void reset() {
     knownRefs.clear();
+    isObservational.clear();
   }
 }

src/main/java/gov/nasa/pds/tools/validate/rule/pds4/FileReferenceValidationRule.java

@@ -331,7 +331,7 @@ private boolean validate(NodeInfo xml) {
                 ancestor = ancestor.getParent();
                 isObservational |= OBS_DATA_TAGS.contains(ancestor.getLocalPart());
               }
-              if (!CrossLabelFileAreaReferenceChecker.add (fullName, target, isObservational)) {
+              if (!CrossLabelFileAreaReferenceChecker.add(fullName, target, isObservational)) {
                 this.getListener().addProblem(
                     new ValidationProblem(
                         new ProblemDefinition(ExceptionType.ERROR, ProblemType.DUPLICATED_FILE_AREA_REFERENCE,

src/main/java/gov/nasa/pds/tools/validate/rule/pds4/LabelValidationRule.java

@@ -59,6 +59,10 @@
 import gov.nasa.pds.validate.constants.Constants;
 import net.sf.saxon.trans.XPathException;
 import net.sf.saxon.tree.tiny.TinyNodeImpl;
+import javax.xml.transform.dom.DOMSource;
+import gov.nasa.pds.tools.util.LabelCacheEntry;
+import gov.nasa.pds.tools.util.LabelUtil;
+import gov.nasa.pds.tools.util.ReferentialIntegrityUtil;
 
 /**
  * Implements a validation chain that validates PDS4 bundles. It is applicable if there is a bundle
@@ -318,6 +322,7 @@ public void validateLabel() {
       if (document != null) {
         getContext().put(PDS4Context.LABEL_DOCUMENT, document);
         labelIsValidFlag = true; // A non-null document signified that the label is valid.
+        cacheIdentifiers(document, getTarget());
       }
     } catch (SAXException | IOException | ParserConfigurationException | TransformerException
         | MissingLabelSchemaException e) {
@@ -337,6 +342,24 @@ public void validateLabel() {
     LOG.debug("validateLabel:target,labelIsValidFlag {},{}", target, labelIsValidFlag);
   }
 
+  private static void cacheIdentifiers(Document document, URL targetUrl) {
+    DOMSource domSource = new DOMSource(document);
+    String[] tagsArr = {LabelUtil.LIDVID_REFERENCE, LabelUtil.LID_REFERENCE};
+    // Report \n errors here (true), so referential integrity phase can use cache safely.
+    // Context area calls stay false — getLidVidReferences(true) above already covers those nodes.
+    ArrayList<String> logicalIds = LabelUtil.getLogicalIdentifiers(domSource, targetUrl, true);
+    ArrayList<String> lidVidRefs = LabelUtil.getLidVidReferences(domSource, targetUrl, true);
+    ArrayList<String> contextRefs = new ArrayList<>();
+    contextRefs.addAll(LabelUtil.getIdentifiersCommon(domSource, targetUrl, tagsArr,
+        LabelUtil.CONTEXT_AREA_INVESTIGATION_AREA_REFERENCE, false));
+    contextRefs.addAll(LabelUtil.getIdentifiersCommon(domSource, targetUrl, tagsArr,
+        LabelUtil.CONTEXT_AREA_OBSERVATION_SYSTEM_COMPONENT_REFERENCE, false));
+    contextRefs.addAll(LabelUtil.getIdentifiersCommon(domSource, targetUrl, tagsArr,
+        LabelUtil.CONTEXT_AREA_TARGET_IDENTIFICATION_REFERENCE, false));
+    ReferentialIntegrityUtil.cacheLabelIdentifiers(targetUrl,
+        new LabelCacheEntry(logicalIds, lidVidRefs, contextRefs));
+  }
+
   private boolean resolveSingleSchema(URL label, Map.Entry<String, URL> schemaLocation,
       URL schemaUrl, ProblemContainer container, ProblemContainer labelProblems,
       XMLCatalogResolver resolver) {

src/main/java/gov/nasa/pds/validate/ValidateLauncher.java

@@ -104,6 +104,7 @@
 import gov.nasa.pds.tools.util.ReferentialIntegrityUtil;
 import gov.nasa.pds.tools.util.XMLExtractor;
 import gov.nasa.pds.tools.validate.ContentProblem;
+import gov.nasa.pds.tools.validate.CrossLabelFileAreaReferenceChecker;
 import gov.nasa.pds.tools.validate.InMemoryRegistrar;
 import gov.nasa.pds.tools.validate.ProblemContainer;
 import gov.nasa.pds.tools.validate.ProblemDefinition;
@@ -1627,6 +1628,7 @@ public boolean doValidation(Map<URL, String> checksumManifest) throws Exception
     // Due to the util class ReferentialIntegrityUtil being static, it need to be
     // reset() if running a regression test.
     ReferentialIntegrityUtil.reset();
+    CrossLabelFileAreaReferenceChecker.reset();
 
     return success;
   }