Fix set-archive-status bare LID resolution and silent bundle cascade failure

CLAUDE.md

@@ -0,0 +1,111 @@
+# CLAUDE.md
+
+This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+
+## Project Overview
+
+Registry Loader is a NASA Planetary Data System (PDS) toolset for loading data into the PDS Registry (Elasticsearch/OpenSearch-based). It combines three Maven modules into a single multi-module project:
+
+- **common** (`registry-common`) - Shared library for Elasticsearch/OpenSearch connectivity, metadata extraction, and data dictionary operations
+- **harvest** - CLI tool that crawls file systems to discover PDS4 products and indexes metadata into the Registry
+- **manager** (`registry-manager`) - CLI tool for managing the Registry: creating indices, loading data dictionaries, setting archive status, and data operations
+
+## Build Commands
+
+```bash
+# Build all modules
+mvn package
+
+# Build specific module
+mvn -pl common package
+mvn -pl harvest package
+mvn -pl manager package
+
+# Run tests
+mvn test
+
+# Run single test class
+mvn -pl common test -Dtest=TestBulkResponseParser
+
+# Skip tests
+mvn package -DskipTests
+
+# Clean build
+mvn clean package
+
+# Install to local repository
+mvn install
+
+# Deploy release (requires GPG setup)
+mvn -P release clean deploy
+```
+
+## Running the Tools
+
+After building, executable JARs are in `{module}/target/`:
+
+```bash
+# Harvest - requires config file
+java -jar harvest/target/harvest-*.jar -c <config.xml>
+
+# Registry Manager - various subcommands
+java -jar manager/target/registry-manager-*.jar <command> <options>
+
+# Registry Manager commands:
+#   create-registry, delete-registry
+#   list-dd, load-dd, delete-dd, export-dd, upgrade-dd
+#   delete-data, export-file, set-archive-status, update-alt-ids
+```
+
+## Architecture
+
+### Module Dependencies
+```
+harvest ──────┐
+              ├──> common ──> Elasticsearch/OpenSearch
+manager ──────┘
+```
+
+### Key Packages
+
+**common** (`gov.nasa.pds.registry.common`):
+- `connection/` - Registry connection handling (AWS OpenSearch Serverless, direct ES/OS)
+- `connection/aws/` - AWS-specific implementations using OpenSearch SDK
+- `connection/es/` - Standard Elasticsearch REST client implementations
+- `es/dao/` - Data Access Objects for registry operations
+- `es/service/` - High-level services (schema updates, data loading)
+- `meta/` - Metadata extractors for PDS4 labels
+- `dd/` - Data dictionary parsing and loading
+
+**harvest** (`gov.nasa.pds.harvest`):
+- `HarvestCli` - CLI entry point
+- `cmd/` - Command implementations
+- `cfg/` - Configuration parsing
+
+**manager** (`gov.nasa.pds.registry.mgr`):
+- `RegistryManagerCli` - CLI entry point with command dispatcher
+- `cmd/reg/` - Registry management commands
+- `cmd/dd/` - Data dictionary commands
+- `cmd/data/` - Data manipulation commands
+
+### Connection Configuration
+
+Registry connections use XML configuration files. The `common` module provides:
+- Direct connections to Elasticsearch/OpenSearch
+- AWS OpenSearch Serverless with Cognito authentication
+- Connection factory pattern via `EstablishConnectionFactory`
+
+## Docker
+
+Build Docker image containing both tools:
+```bash
+docker image build -t nasapds/registry-loader -f docker/Dockerfile \
+  --build-arg harvest_package_path=harvest/target/harvest-*-bin.tar.gz \
+  --build-arg reg_manager_package_path=manager/target/registry-manager-*-bin.tar.gz .
+```
+
+## CI/CD
+
+- Releases triggered by pushing `release/*` tags
+- Uses NASA-PDS Roundup Action for automated releases
+- Docker images published to Docker Hub on stable releases

common/src/main/java/gov/nasa/pds/registry/common/es/service/ProductService.java

@@ -2,6 +2,7 @@
 
 import java.util.Arrays;
 import java.util.Collection;
+import java.util.Collections;
 import java.util.List;
 import java.util.Set;
 import java.util.TreeSet;
@@ -36,43 +37,72 @@
     }
     /**
      * Set archive status
-     * @param lidvid ID of a product to update. If it is a collection, 
-     * update primary references from collection inventory.
+     * @param lidvid LID or LIDVID of a product to update. If a bare LID is provided it is
+     * resolved to the latest LIDVID. If the product is a collection, primary references from
+     * collection inventory are also updated. If it is a bundle, all referenced collections and
+     * their products are updated.
      * @param status new status
      * @throws Exception an exception
      */
     public void updateArchiveStatus(String lidvid, String status) throws Exception
     {
         log.info("Setting product status and its references if bundle or collection. LIDVID = " + lidvid + ", status = " + status);
         int total = 1;
 
+        String resolvedLidvid = lidvid;
         String pClass = dao.getProductClass(lidvid);
-        if(pClass == null) 
+        if(pClass == null)
         {
-            log.warn("Unknown LIDVID: " + lidvid);
-            return;
+            // If the input has no version component it may be a bare LID; try to resolve it.
+            if(!lidvid.contains("::"))
+            {
+                List<String> resolved = dao.getLatestLidVids(Collections.singletonList(lidvid));
+                if(resolved != null && !resolved.isEmpty())
+                {
+                    resolvedLidvid = resolved.get(0);
+                    log.info("Resolved bare LID " + lidvid + " to LIDVID " + resolvedLidvid);
+                    pClass = dao.getProductClass(resolvedLidvid);
+                }
+            }
+
+            if(pClass == null)
+            {
+                throw new Exception("Unknown LID/LIDVID: " + lidvid
+                        + ". Verify that the identifier exists in the registry and that a full"
+                        + " LIDVID (e.g. urn:nasa:pds:bundle::1.0) is provided when multiple"
+                        + " versions are present.");
+            }
         }
-        
+
         // Update the product
-        dao.updateArchiveStatus(Arrays.asList(lidvid), status);
+        dao.updateArchiveStatus(Arrays.asList(resolvedLidvid), status);
 
         // Update collection inventory
         if("Product_Collection".equals(pClass))
         {
             log.info("Setting status of primary references from collection inventory");
-            total += updateCollectionInventory(lidvid, status);            
+            total += updateCollectionInventory(resolvedLidvid, status);            
         }
         else if("Product_Bundle".equals(pClass))
         {
             // Get collection IDs. There could be both LIDs and LIDVIDs at the same time.
-            LidvidSet collectionIds = dao.getCollectionIds(lidvid);
-            if(collectionIds == null) return;
-
-            Set<String> lidvids = new TreeSet<String>();            
-            if(collectionIds.lidvids != null) lidvids.addAll(collectionIds.lidvids);
-
-            List<String> tmp = dao.getLatestLidVids(collectionIds.lids);
-            if(tmp != null) lidvids.addAll(tmp);
+            LidvidSet collectionIds = dao.getCollectionIds(resolvedLidvid);
+
+            Set<String> lidvids = new TreeSet<String>();
+            if(collectionIds != null)
+            {
+                if(collectionIds.lidvids != null) lidvids.addAll(collectionIds.lidvids);
+
+                List<String> tmp = dao.getLatestLidVids(collectionIds.lids);
+                if(tmp != null) lidvids.addAll(tmp);
+            }
+
+            if(lidvids.isEmpty())
+            {
+                log.warn("No collection references found for bundle " + resolvedLidvid
+                        + ". Verify that the bundle document contains 'ref_lid_collection' or"
+                        + " 'ref_lidvid_collection' fields.");
+            }
 
             total += updateCollections(lidvids, status);
         }