[7.x][ML] Reuse SourceDestValidator for data frame analytics (#50841) (#50850)

This commit removes validation logic of source and dest indices
for data frame analytics and replaces it with using the common
`SourceDestValidator` class which is already used by transforms.
This way the validations and their messages become consistent
while we reduce code.

This means that where these validations fail the error messages
will be slightly different for data frame analytics.

Backport of #50841

Author: dimitris-athanasiou

x-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/common/validation/SourceDestValidator.java

@@ -49,7 +49,6 @@ public final class SourceDestValidator {
 
     // messages
     public static final String SOURCE_INDEX_MISSING = "Source index [{0}] does not exist";
-    public static final String SOURCE_LOWERCASE = "Source index [{0}] must be lowercase";
     public static final String DEST_IN_SOURCE = "Destination index [{0}] is included in source expression [{1}]";
     public static final String DEST_LOWERCASE = "Destination index [{0}] must be lowercase";
     public static final String NEEDS_REMOTE_CLUSTER_SEARCH = "Source index is configured with a remote index pattern(s) [{0}]"
@@ -62,6 +61,7 @@ public final class SourceDestValidator {
         + "alias [{0}], at least a [{1}] license is required, found license [{2}]";
     public static final String REMOTE_CLUSTER_LICENSE_INACTIVE = "License check failed for remote cluster "
         + "alias [{0}], license is not active";
+    public static final String REMOTE_SOURCE_INDICES_NOT_SUPPORTED = "remote source indices are not supported";
 
     // workaround for 7.x: remoteClusterAliases does not throw
     private static final ClusterNameExpressionResolver clusterNameExpressionResolver = new ClusterNameExpressionResolver();
@@ -222,7 +222,7 @@ private void resolveLocalAndRemoteSource() {
         }
     }
 
-    interface SourceDestValidation {
+    public interface SourceDestValidation {
         void validate(Context context, ActionListener<Context> listener);
     }
 
@@ -234,18 +234,7 @@ interface SourceDestValidation {
     public static final SourceDestValidation REMOTE_SOURCE_VALIDATION = new RemoteSourceEnabledAndRemoteLicenseValidation();
     public static final SourceDestValidation DESTINATION_IN_SOURCE_VALIDATION = new DestinationInSourceValidation();
     public static final SourceDestValidation DESTINATION_SINGLE_INDEX_VALIDATION = new DestinationSingleIndexValidation();
-
-    // set of default validation collections, if you want to automatically benefit from new validators, use those
-    public static final List<SourceDestValidation> PREVIEW_VALIDATIONS = Arrays.asList(SOURCE_MISSING_VALIDATION, REMOTE_SOURCE_VALIDATION);
-
-    public static final List<SourceDestValidation> ALL_VALIDATIONS = Arrays.asList(
-        SOURCE_MISSING_VALIDATION,
-        REMOTE_SOURCE_VALIDATION,
-        DESTINATION_IN_SOURCE_VALIDATION,
-        DESTINATION_SINGLE_INDEX_VALIDATION
-    );
-
-    public static final List<SourceDestValidation> NON_DEFERABLE_VALIDATIONS = Arrays.asList(DESTINATION_SINGLE_INDEX_VALIDATION);
+    public static final SourceDestValidation REMOTE_SOURCE_NOT_SUPPORTED_VALIDATION = new RemoteSourceNotSupportedValidation();
 
     /**
      * Create a new Source Dest Validator
@@ -305,10 +294,11 @@ public void validate(
             }
         }, listener::onFailure);
 
+        // We traverse the validations in reverse order as we chain the listeners from back to front
         for (int i = validations.size() - 1; i >= 0; i--) {
-            final SourceDestValidation validation = validations.get(i);
+            SourceDestValidation validation = validations.get(i);
             final ActionListener<Context> previousValidationListener = validationListener;
-            validationListener = ActionListener.wrap(c -> { validation.validate(c, previousValidationListener); }, listener::onFailure);
+            validationListener = ActionListener.wrap(c -> validation.validate(c, previousValidationListener), listener::onFailure);
         }
 
         validationListener.onResponse(context);
@@ -433,13 +423,13 @@ public void validate(Context context, ActionListener<Context> listener) {
                 return;
             }
 
-            if (context.resolvedSource.contains(destIndex)) {
+            if (context.resolveSource().contains(destIndex)) {
                 context.addValidationError(DEST_IN_SOURCE, destIndex, Strings.arrayToCommaDelimitedString(context.getSource()));
                 listener.onResponse(context);
                 return;
             }
 
-            if (context.resolvedSource.contains(context.resolveDest())) {
+            if (context.resolveSource().contains(context.resolveDest())) {
                 context.addValidationError(
                     DEST_IN_SOURCE,
                     context.resolveDest(),
@@ -460,6 +450,17 @@ public void validate(Context context, ActionListener<Context> listener) {
         }
     }
 
+    static class RemoteSourceNotSupportedValidation implements SourceDestValidation {
+
+        @Override
+        public void validate(Context context, ActionListener<Context> listener) {
+            if (context.resolveRemoteSource().isEmpty() == false) {
+                context.addValidationError(REMOTE_SOURCE_INDICES_NOT_SUPPORTED);
+            }
+            listener.onResponse(context);
+        }
+    }
+
     private static String getMessage(String message, Object... args) {
         return new MessageFormat(message, Locale.ROOT).format(args);
     }

x-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/ml/action/PutDataFrameAnalyticsAction.java

@@ -18,9 +18,11 @@
 import org.elasticsearch.common.xcontent.ToXContentObject;
 import org.elasticsearch.common.xcontent.XContentBuilder;
 import org.elasticsearch.common.xcontent.XContentParser;
+import org.elasticsearch.xpack.core.common.validation.SourceDestValidator;
 import org.elasticsearch.xpack.core.ml.dataframe.DataFrameAnalyticsConfig;
 import org.elasticsearch.xpack.core.ml.dataframe.DataFrameAnalyticsSource;
 import org.elasticsearch.xpack.core.ml.job.messages.Messages;
+import org.elasticsearch.xpack.core.ml.utils.MlStrings;
 
 import java.io.IOException;
 import java.util.Objects;
@@ -90,14 +92,29 @@ public DataFrameAnalyticsConfig getConfig() {
         @Override
         public ActionRequestValidationException validate() {
             ActionRequestValidationException error = null;
+            error = checkConfigIdIsValid(config, error);
+            error = SourceDestValidator.validateRequest(error, config.getDest().getIndex());
             error = checkNoIncludedAnalyzedFieldsAreExcludedBySourceFiltering(config, error);
             return error;
         }
 
+        private ActionRequestValidationException checkConfigIdIsValid(DataFrameAnalyticsConfig config,
+                                                                      ActionRequestValidationException error) {
+            if (MlStrings.isValidId(config.getId()) == false) {
+                error = ValidateActions.addValidationError(Messages.getMessage(Messages.INVALID_ID, DataFrameAnalyticsConfig.ID,
+                    config.getId()), error);
+            }
+            if (!MlStrings.hasValidLengthForId(config.getId())) {
+                error = ValidateActions.addValidationError(Messages.getMessage(Messages.ID_TOO_LONG, DataFrameAnalyticsConfig.ID,
+                    config.getId(), MlStrings.ID_LENGTH_LIMIT), error);
+            }
+            return error;
+        }
+
         private ActionRequestValidationException checkNoIncludedAnalyzedFieldsAreExcludedBySourceFiltering(
                 DataFrameAnalyticsConfig config, ActionRequestValidationException error) {
             if (config.getAnalyzedFields() == null) {
-                return null;
+                return error;
             }
             for (String analyzedInclude : config.getAnalyzedFields().includes()) {
                 if (config.getSource().isFieldExcluded(analyzedInclude)) {
@@ -107,7 +124,7 @@ private ActionRequestValidationException checkNoIncludedAnalyzedFieldsAreExclude
                         + DataFrameAnalyticsSource._SOURCE.getPreferredName() + "]", error);
                 }
             }
-            return null;
+            return error;
         }
 
         @Override

x-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/ml/dataframe/DataFrameAnalyticsDest.java

@@ -13,15 +13,11 @@
 import org.elasticsearch.common.xcontent.ConstructingObjectParser;
 import org.elasticsearch.common.xcontent.ToXContentObject;
 import org.elasticsearch.common.xcontent.XContentBuilder;
-import org.elasticsearch.indices.InvalidIndexNameException;
 import org.elasticsearch.xpack.core.ml.utils.ExceptionsHelper;
 
 import java.io.IOException;
-import java.util.Locale;
 import java.util.Objects;
 
-import static org.elasticsearch.cluster.metadata.MetaDataCreateIndexService.validateIndexOrAliasName;
-
 public class DataFrameAnalyticsDest implements Writeable, ToXContentObject {
 
     public static final ParseField INDEX = new ParseField("index");
@@ -94,13 +90,4 @@ public String getIndex() {
     public String getResultsField() {
         return resultsField;
     }
-
-    public void validate() {
-        if (index != null) {
-            validateIndexOrAliasName(index, InvalidIndexNameException::new);
-            if (index.toLowerCase(Locale.ROOT).equals(index) == false) {
-                throw new InvalidIndexNameException(index, "dest.index must be lowercase");
-            }
-        }
-    }
 }