feat(validation): Handling

Author: niedch

src/main/java/ai/labs/eddi/configs/ingestion/IRestRagIngestionSourceStore.java

@@ -9,7 +9,6 @@
 import ai.labs.eddi.configs.ingestion.model.RagIngestionSource;
 import jakarta.annotation.security.RolesAllowed;
 import jakarta.ws.rs.*;
-import jakarta.ws.rs.core.MediaType;
 import jakarta.ws.rs.core.Response;
 import org.eclipse.microprofile.openapi.annotations.Operation;
 import org.eclipse.microprofile.openapi.annotations.parameters.Parameter;

src/main/java/ai/labs/eddi/configs/ingestion/model/RagIngestionSource.java

@@ -69,15 +69,11 @@ public record RagIngestionSource(
             type = "web";
         }
 
-        // Ensure sourceConfig is not null for web type
-        if (sourceConfig == null && "web".equals(type)) {
-            sourceConfig = new WebSourceConfig(null, null, null);
-        }
-
         // Set defaults for other nullable nested configs
         if (ingestionSettings == null) {
             ingestionSettings = new IngestionSettings();
         }
+
         if (schedule == null) {
             schedule = new Schedule();
         }

src/main/java/ai/labs/eddi/configs/ingestion/model/WebSourceConfig.java

@@ -4,6 +4,7 @@
  */
 package ai.labs.eddi.configs.ingestion.model;
 
+import ai.labs.eddi.modules.llm.tools.UrlValidationUtils;
 import java.util.ArrayList;
 import java.util.List;
 
@@ -40,6 +41,21 @@ public record WebSourceConfig(
         }
     }
 
+    /**
+     * Validates this configuration and throws {@link IllegalArgumentException} if
+     * any field is invalid. Call this before using the config for crawling.
+     */
+    public void validate() {
+        if (!UrlValidationUtils.isValidHttpUrl(startUrl)) {
+            throw new IllegalArgumentException(
+                    startUrl == null || startUrl.isBlank()
+                            ? "startUrl must not be null or blank"
+                            : "startUrl must be a valid http or https URL, got: " + startUrl);
+        }
+        scope.validate();
+        crawlSettings.validate();
+    }
+
     /**
      * Scope constraints for the web crawler. Controls how far and where the crawler
      * will follow links.
@@ -90,6 +106,15 @@ public record Scope(
         public Scope() {
             this(true, "/", 3, 200, new ArrayList<>());
         }
+
+        void validate() {
+            if (maxPages > 10_000) {
+                throw new IllegalArgumentException("maxPages must not exceed 10,000, got: " + maxPages);
+            }
+            if (maxDepth > 20) {
+                throw new IllegalArgumentException("maxDepth must not exceed 20, got: " + maxDepth);
+            }
+        }
     }
 
     /**
@@ -128,5 +153,14 @@ public record CrawlSettings(
         public CrawlSettings() {
             this(500, 15, "EDDI-Crawler/1.0");
         }
+
+        void validate() {
+            if (requestDelayMs > 30_000) {
+                throw new IllegalArgumentException("requestDelayMs must not exceed 30,000, got: " + requestDelayMs);
+            }
+            if (timeoutSeconds > 300) {
+                throw new IllegalArgumentException("timeoutSeconds must not exceed 300, got: " + timeoutSeconds);
+            }
+        }
     }
 }

src/main/java/ai/labs/eddi/configs/ingestion/rest/RestRagIngestionSourceStore.java

@@ -8,6 +8,7 @@
 import ai.labs.eddi.configs.ingestion.IRagIngestionSourceStore;
 import ai.labs.eddi.configs.ingestion.IRestRagIngestionSourceStore;
 import ai.labs.eddi.configs.ingestion.model.RagIngestionSource;
+import ai.labs.eddi.configs.ingestion.model.WebSourceConfig;
 import ai.labs.eddi.configs.rest.RestVersionInfo;
 import ai.labs.eddi.configs.schema.IJsonSchemaCreator;
 import ai.labs.eddi.configs.descriptors.model.DocumentDescriptor;
@@ -48,7 +49,6 @@ public class RestRagIngestionSourceStore implements IRestRagIngestionSourceStore
     private final IRagIngestionSourceStore sourceStore;
     private final IScheduleStore scheduleStore;
     private final IJsonSchemaCreator jsonSchemaCreator;
-    private final IDocumentDescriptorStore documentDescriptorStore;
     private final RagIngestionService ingestionService;
     private final IContentHashStore contentHashTracker;
     private final RestVersionInfo<RagIngestionSource> restVersionInfo;
@@ -64,7 +64,6 @@ public RestRagIngestionSourceStore(
         this.sourceStore = sourceStore;
         this.scheduleStore = scheduleStore;
         this.jsonSchemaCreator = jsonSchemaCreator;
-        this.documentDescriptorStore = documentDescriptorStore;
         this.ingestionService = ingestionService;
         this.contentHashTracker = contentHashTracker;
         this.restVersionInfo = new RestVersionInfo<>(resourceURI, sourceStore, documentDescriptorStore);
@@ -100,6 +99,8 @@ public RagIngestionSource readIngestionSource(String id, Integer version) {
 
     @Override
     public Response updateIngestionSource(String id, Integer version, RagIngestionSource source) {
+        validateSource(source);
+
         Response response = restVersionInfo.update(id, version, source);
 
         // Update associated schedule
@@ -116,6 +117,8 @@ public Response updateIngestionSource(String id, Integer version, RagIngestionSo
 
     @Override
     public Response createIngestionSource(RagIngestionSource source) {
+        validateSource(source);
+
         IResourceStore.IResourceId resourceId = restVersionInfo.createDocument(source);
 
         // Create associated schedule
@@ -194,6 +197,18 @@ public IResourceStore.IResourceId getCurrentResourceId(String id) throws IResour
         return sourceStore.getCurrentResourceId(id);
     }
 
+    // --- Source validation ---
+
+    private void validateSource(RagIngestionSource source) {
+        if ("web".equals(source.type())) {
+            if (source.sourceConfig() instanceof WebSourceConfig webConfig) {
+                webConfig.validate();
+            } else if (source.sourceConfig() == null) {
+                throw new IllegalArgumentException("sourceConfig is required for web-type ingestion sources");
+            }
+        }
+    }
+
     // --- Schedule management ---
 
     private void createScheduleForSource(String sourceId, RagIngestionSource source) throws Exception {

src/main/java/ai/labs/eddi/engine/exception/IllegalArgumentExceptionMapper.java

@@ -4,19 +4,23 @@
  */
 package ai.labs.eddi.engine.exception;
 
+import jakarta.ws.rs.core.MediaType;
 import jakarta.ws.rs.core.Response;
 import jakarta.ws.rs.ext.ExceptionMapper;
 import jakarta.ws.rs.ext.Provider;
 
+import java.util.Map;
+
 /**
  * @author ginccc
  */
 @Provider
 public class IllegalArgumentExceptionMapper implements ExceptionMapper<IllegalArgumentException> {
     @Override
     public Response toResponse(IllegalArgumentException exception) {
-        Response.ResponseBuilder response = Response.status(Response.Status.BAD_REQUEST);
-        response.entity(exception.getLocalizedMessage());
-        return response.build();
+        return Response.status(Response.Status.BAD_REQUEST)
+                .entity(Map.of("message", exception.getLocalizedMessage()))
+                .type(MediaType.APPLICATION_JSON)
+                .build();
     }
 }