Scripting: Convert domainSplit function for ML to whitelist (#34426)

This commit moves the definition of domainSplit into java and exposes it
as a painless whitelist extension. The method also no longer needs
params, and version which ignores params is added and deprecated.

Author: rjernst

docs/reference/ml/transforms.asciidoc

@@ -490,10 +490,10 @@ PUT _xpack/ml/datafeeds/datafeed-test3
   },
   "script_fields":{
     "sub":{
-      "script":"return domainSplit(doc['query'].value, params).get(0);"
+      "script":"return domainSplit(doc['query'].value).get(0);"
     },
     "hrd":{
-      "script":"return domainSplit(doc['query'].value, params).get(1);"
+      "script":"return domainSplit(doc['query'].value).get(1);"
     }
   }
 }
@@ -511,10 +511,6 @@ registered domain. For example, the highest registered domain of
 `domainSplit()` function returns an array of two values: the first value is the
 subdomain; the second value is the highest registered domain.
 
-NOTE: The `domainSplit()` function takes two arguments. The first argument is
-the string you want to split. The second argument is always `params`. This is a
-technical implementation detail related to how Painless operates internally.
-
 The preview {dfeed} API returns the following results, which show that
 "www.ml.elastic.co" has been split into "elastic.co" and "www.ml":
 

x-pack/plugin/ml/build.gradle

@@ -9,9 +9,8 @@ esplugin {
     description 'Elasticsearch Expanded Pack Plugin - Machine Learning'
     classname 'org.elasticsearch.xpack.ml.MachineLearning'
     hasNativeController true
-    extendedPlugins = ['x-pack-core']
+    extendedPlugins = ['x-pack-core', 'lang-painless']
 }
-archivesBaseName = 'x-pack-ml'
 
 configurations {
     nativeBundle {
@@ -42,6 +41,7 @@ compileTestJava.options.compilerArgs << "-Xlint:-deprecation,-rawtypes,-serial,-
 dependencies {
     // "org.elasticsearch.plugin:x-pack-core:${version}" doesn't work with idea because the testArtifacts are also here
     compileOnly project(path: xpackModule('core'), configuration: 'default')
+    compileOnly "org.elasticsearch.plugin:elasticsearch-scripting-painless-spi:${versions.elasticsearch}"
     testCompile project(path: xpackModule('core'), configuration: 'testArtifacts')
     // This should not be here
     testCompile project(path: xpackModule('security'), configuration: 'testArtifacts')

x-pack/plugin/ml/qa/single-node-tests/src/test/java/org/elasticsearch/xpack/ml/transforms/PainlessDomainSplitIT.java

@@ -13,7 +13,6 @@
 import org.elasticsearch.common.settings.Settings;
 import org.elasticsearch.test.rest.ESRestTestCase;
 import org.elasticsearch.xpack.ml.MachineLearning;
-import org.elasticsearch.xpack.ml.utils.DomainSplitFunction;
 import org.joda.time.DateTime;
 
 import java.util.ArrayList;
@@ -190,8 +189,7 @@ public void testIsolated() throws Exception {
 
         Pattern pattern = Pattern.compile("domain_split\":\\[(.*?),(.*?)\\]");
 
-        Map<String, Object> params = new HashMap<>(DomainSplitFunction.params.size() + 1);
-        params.putAll(DomainSplitFunction.params);
+        Map<String, Object> params = new HashMap<>();
         for (TestConfiguration testConfig : tests) {
             params.put("host", testConfig.hostName);
             String mapAsJson = Strings.toString(jsonBuilder().map(params));
@@ -207,8 +205,8 @@ public void testIsolated() throws Exception {
                     "        \"domain_split\" : {\n" +
                     "            \"script\" : {\n" +
                     "                \"lang\": \"painless\",\n" +
-                    "                \"inline\": \"" + DomainSplitFunction.function +
-                    " return domainSplit(params['host'], params); \",\n" +
+                    "                \"inline\": \"" +
+                    " return domainSplit(params['host']); \",\n" +
                     "                \"params\": " + mapAsJson + "\n" +
                     "            }\n" +
                     "        }\n" +

x-pack/plugin/ml/src/main/java/org/elasticsearch/xpack/ml/MachineLearningPainlessExtension.java

@@ -0,0 +1,26 @@
+/*
+ * Copyright Elasticsearch B.V. and/or licensed to Elasticsearch B.V. under one
+ * or more contributor license agreements. Licensed under the Elastic License;
+ * you may not use this file except in compliance with the Elastic License.
+ */
+package org.elasticsearch.xpack.ml;
+
+import org.elasticsearch.painless.spi.PainlessExtension;
+import org.elasticsearch.painless.spi.Whitelist;
+import org.elasticsearch.painless.spi.WhitelistLoader;
+import org.elasticsearch.script.ScriptContext;
+import org.elasticsearch.script.SearchScript;
+
+import java.util.Collections;
+import java.util.List;
+import java.util.Map;
+
+public class MachineLearningPainlessExtension implements PainlessExtension {
+    private static final Whitelist WHITELIST =
+        WhitelistLoader.loadFromResourceFiles(MachineLearningPainlessExtension.class, "whitelist.txt");
+
+    @Override
+    public Map<ScriptContext<?>, List<Whitelist>> getContextWhitelists() {
+        return Collections.singletonMap(SearchScript.CONTEXT, Collections.singletonList(WHITELIST));
+    }
+}

x-pack/plugin/ml/src/main/java/org/elasticsearch/xpack/ml/datafeed/extractor/scroll/ScrollDataExtractor.java

@@ -17,22 +17,18 @@
 import org.elasticsearch.client.Client;
 import org.elasticsearch.common.logging.Loggers;
 import org.elasticsearch.common.unit.TimeValue;
-import org.elasticsearch.script.Script;
 import org.elasticsearch.search.SearchHit;
 import org.elasticsearch.search.fetch.StoredFieldsContext;
 import org.elasticsearch.search.fetch.subphase.DocValueFieldsContext;
 import org.elasticsearch.search.sort.SortOrder;
 import org.elasticsearch.xpack.core.ClientHelper;
 import org.elasticsearch.xpack.core.ml.datafeed.extractor.DataExtractor;
 import org.elasticsearch.xpack.core.ml.datafeed.extractor.ExtractorUtils;
-import org.elasticsearch.xpack.ml.utils.DomainSplitFunction;
 
 import java.io.ByteArrayInputStream;
 import java.io.ByteArrayOutputStream;
 import java.io.IOException;
 import java.io.InputStream;
-import java.util.HashMap;
-import java.util.Map;
 import java.util.NoSuchElementException;
 import java.util.Objects;
 import java.util.Optional;
@@ -130,26 +126,10 @@ private SearchRequestBuilder buildSearchRequest(long start) {
         } else {
             searchRequestBuilder.setFetchSource(sourceFields, null);
         }
-        context.scriptFields.forEach(f -> searchRequestBuilder.addScriptField(
-                f.fieldName(), injectDomainSplit(f.script())));
+        context.scriptFields.forEach(f -> searchRequestBuilder.addScriptField(f.fieldName(), f.script()));
         return searchRequestBuilder;
     }
 
-    private Script injectDomainSplit(Script script) {
-        String code = script.getIdOrCode();
-        if (code.contains("domainSplit(") && script.getLang().equals("painless")) {
-            String modifiedCode = DomainSplitFunction.function + code;
-            Map<String, Object> modifiedParams = new HashMap<>(script.getParams().size()
-                    + DomainSplitFunction.params.size());
-
-            modifiedParams.putAll(script.getParams());
-            modifiedParams.putAll(DomainSplitFunction.params);
-
-            return new Script(script.getType(), script.getLang(), modifiedCode, modifiedParams);
-        }
-        return script;
-    }
-
     private InputStream processSearchResponse(SearchResponse searchResponse) throws IOException {
 
         if (searchResponse.getFailedShards() > 0 && searchHasShardFailure == false) {