Query requests return with sample size defined in advanced setting

changelogs/fragments/9356.yml

@@ -0,0 +1,2 @@
+feat:
+- Remove sample size in default query string but use advanced setting ([#9356](https://github.com/opensearch-project/OpenSearch-Dashboards/pull/9356))

cypress/integration/core_opensearch_dashboards/opensearch_dashboards/apps/query_enhancements/queries.spec.js

@@ -94,7 +94,7 @@ const queriesTestSuite = () => {
         // Default SQL query should be set
         cy.waitForLoader(true);
         cy.getElementByTestId(`osdQueryEditor__multiLine`).contains(
-          `SELECT * FROM ${INDEX_WITH_TIME_1}* LIMIT 10`
+          `SELECT * FROM ${INDEX_WITH_TIME_1}`
         );
         cy.getElementByTestId(`queryResultCompleteMsg`).should('be.visible');
 
@@ -125,7 +125,7 @@ const queriesTestSuite = () => {
         // Default PPL query should be set
         cy.waitForLoader(true);
         cy.getElementByTestId(`osdQueryEditor__multiLine`).contains(
-          `source = ${INDEX_WITH_TIME_1}*`
+          `source = ${INDEX_WITH_TIME_1}* | head 500`
         );
         cy.waitForSearch();
         cy.getElementByTestId(`queryResultCompleteMsg`).should('be.visible');

cypress/utils/apps/query_enhancements/recent_queries.js

@@ -83,7 +83,7 @@ export const generateRecentQueriesTestConfiguration = (dataset, datasetType, lan
     PPL: 'OpenSearch SQL',
     'OpenSearch SQL': 'PPL',
   };
-  const defaultQuery = language.name === 'PPL' ? '' : ' LIMIT 10';
+  const defaultQuery = language.name === 'PPL' ? 'head 500' : '';
   const customDatasetType = RecentQueriesDataTypes[datasetType].name;
   return {
     dataset,

cypress/utils/apps/query_enhancements/shared.js

@@ -147,7 +147,7 @@ export const getDefaultQuery = (datasetName, language) => {
     case QueryLanguages.PPL.name:
       return `source = ${datasetName}`;
     case QueryLanguages.SQL.name:
-      return `SELECT * FROM ${datasetName} LIMIT 10`;
+      return `SELECT * FROM ${datasetName}`;
   }
 };
 

src/plugins/data/common/constants.ts

@@ -111,4 +111,5 @@ export const UI_SETTINGS = {
   SEARCH_QUERY_LANGUAGE_BLOCKLIST: 'search:queryLanguageBlocklist',
   NEW_HOME_PAGE: 'home:useNewHomePage',
   DATA_WITH_LONG_NUMERALS: 'data:withLongNumerals',
+  DISCOVER_SAMPLE_SIZE: 'discover:sampleSize',
 } as const;

src/plugins/data/public/query/query_string/dataset_service/lib/index_pattern_type.ts

@@ -86,7 +86,7 @@ export const indexPatternTypeConfig: DatasetTypeConfig = {
             title: i18n.translate('data.indexPatternType.sampleQuery.basicSQLQuery', {
               defaultMessage: 'Sample query for SQL',
             }),
-            query: `SELECT * FROM ${dataset.title} LIMIT 10`,
+            query: `SELECT * FROM ${dataset.title}`,
           },
         ];
     }

src/plugins/data/public/query/query_string/dataset_service/lib/index_type.ts

@@ -109,7 +109,7 @@ export const indexTypeConfig: DatasetTypeConfig = {
             title: i18n.translate('data.indexType.sampleQuery.basicSQLQuery', {
               defaultMessage: 'Sample query for SQL',
             }),
-            query: `SELECT * FROM ${dataset.title} LIMIT 10`,
+            query: `SELECT * FROM ${dataset.title}`,
           },
         ];
     }

src/plugins/data/public/query/query_string/dataset_service/types.ts

@@ -3,6 +3,7 @@
  * SPDX-License-Identifier: Apache-2.0
  */
 import { EuiIconProps } from '@elastic/eui';
+import { CoreStart } from 'opensearch-dashboards/public';
 import {
   Dataset,
   DatasetField,
@@ -117,5 +118,5 @@ export interface DatasetTypeConfig {
   /**
    * Returns the initial query that is added to the query editor when a dataset is selected.
    */
-  getInitialQueryString?: (query: Query) => string | void;
+  getInitialQueryString?: (query: Query, uiSettings?: CoreStart['uiSettings']) => string | void;
 }

src/plugins/data/public/query/query_string/query_string_manager.ts

@@ -74,7 +74,8 @@ export class QueryStringManager {
     }
 
     return (
-      typeConfig?.getInitialQueryString?.(query) ?? (languageConfig?.getQueryString(query) || '')
+      typeConfig?.getInitialQueryString?.(query, this.uiSettings) ??
+      (languageConfig?.getQueryString(query) || '')
     );
   }
 

src/plugins/query_enhancements/public/datasets/s3_type.ts

@@ -5,7 +5,7 @@
 
 import { i18n } from '@osd/i18n';
 import { trimEnd } from 'lodash';
-import { HttpSetup, SavedObjectsClientContract } from 'opensearch-dashboards/public';
+import { CoreStart, HttpSetup, SavedObjectsClientContract } from 'opensearch-dashboards/public';
 import semver from 'semver';
 import {
   DATA_STRUCTURE_META_TYPES,
@@ -15,6 +15,7 @@
   DataStructureCustomMeta,
   Dataset,
   DatasetField,
+  Query,
 } from '../../../data/common';
 import { DatasetTypeConfig, IDataPluginServices, OSD_FIELD_TYPES } from '../../../data/public';
 import {
@@ -131,7 +132,7 @@
             title: i18n.translate('queryEnhancements.s3Type.sampleQuery.basicPPLQuery', {
               defaultMessage: 'Sample query for PPL',
             }),
-            query: `source = ${dataset.title}`,
+            query: `source = ${dataset.title} | head 500`,
           },
         ];
       case 'SQL':
@@ -140,11 +141,24 @@
             title: i18n.translate('queryEnhancements.s3Type.sampleQuery.basicSQLQuery', {
               defaultMessage: 'Sample query for SQL',
             }),
-            query: `SELECT * FROM ${dataset.title} LIMIT 10`,
+            query: `SELECT * FROM ${dataset.title} LIMIT 500`,
           },
         ];
     }
   },
+
+  getInitialQueryString: (query: Query, uiSettings?: CoreStart['uiSettings']) => {
+    switch (query.language) {
+      case 'PPL':
+        return `source = ${query.dataset?.title} | head ${
+          uiSettings?.get('discover:sampleSize') ?? 500
+        }`;
+      case 'SQL':
+        return `SELECT * FROM ${query.dataset?.title} LIMIT ${
+          uiSettings?.get('discover:sampleSize') ?? 500
+        }`;
+    }
+  },
 };
 
 const fetch = async (

src/plugins/query_enhancements/public/plugin.tsx

@@ -5,7 +5,7 @@
 import { i18n } from '@osd/i18n';
 import { BehaviorSubject } from 'rxjs';
 import { CoreSetup, CoreStart, Plugin, PluginInitializerContext } from '../../../core/public';
-import { DataStorage } from '../../data/common';
+import { DataStorage, UI_SETTINGS } from '../../data/common';
 import {
   createEditor,
   DefaultInput,
@@ -65,7 +65,12 @@ export class QueryEnhancementsPlugin
         startServices: core.getStartServices(),
         usageCollector: data.search.usageCollector,
       }),
-      getQueryString: (currentQuery: Query) => `source = ${currentQuery.dataset?.title}`,
+      // TODO: we should remove sample size for default PPL query here
+      // once fetch_size is working: https://github.com/opensearch-project/OpenSearch-Dashboards/issues/9385
+      getQueryString: (currentQuery: Query) =>
+        `source = ${currentQuery.dataset?.title} | head ${core.uiSettings.get(
+          UI_SETTINGS.DISCOVER_SAMPLE_SIZE
+        )}`,
       fields: { sortable: false, filterable: false, visualizable: false },
       docLink: {
         title: i18n.translate('queryEnhancements.pplLanguage.docLink', {
@@ -130,8 +135,7 @@ export class QueryEnhancementsPlugin
         startServices: core.getStartServices(),
         usageCollector: data.search.usageCollector,
       }),
-      getQueryString: (currentQuery: Query) =>
-        `SELECT * FROM ${currentQuery.dataset?.title} LIMIT 10`,
+      getQueryString: (currentQuery: Query) => `SELECT * FROM ${currentQuery.dataset?.title}`,
       fields: { sortable: false, filterable: false, visualizable: false },
       docLink: {
         title: i18n.translate('queryEnhancements.sqlLanguage.docLink', {

src/plugins/query_enhancements/server/search/ppl_search_strategy.ts

@@ -35,6 +35,9 @@ export const pplSearchStrategyProvider = (
   return {
     search: async (context, request: any, options) => {
       try {
+        // TODO: PPL does not support fetch_size yet due to https://github.com/opensearch-project/sql/issues/3314
+        // We can add the below back for https://github.com/opensearch-project/OpenSearch-Dashboards/issues/9385
+        // request.body.fetch_size = await context.core.uiSettings.client.get('discover:sampleSize');
         const query: Query = request.body.query;
         const aggConfig: QueryAggConfig | undefined = request.body.aggConfig;
         const rawResponse: any = await pplFacet.describeQuery(context, request);

src/plugins/query_enhancements/server/search/sql_search_strategy.test.ts

@@ -26,7 +26,15 @@ describe('sqlSearchStrategyProvider', () => {
   let logger: Logger;
   let client: ILegacyClusterClient;
   let usage: SearchUsage;
-  const emptyRequestHandlerContext = ({} as unknown) as RequestHandlerContext;
+  const emptyRequestHandlerContext = ({
+    core: {
+      uiSettings: {
+        client: {
+          get: jest.fn(),
+        },
+      },
+    },
+  } as unknown) as RequestHandlerContext;
 
   beforeEach(() => {
     config$ = of({} as SharedGlobalConfig);
@@ -69,7 +77,6 @@ describe('sqlSearchStrategyProvider', () => {
       { name: 'field1', type: 'long' },
       { name: 'field2', type: 'text' },
     ]);
-
     const strategy = sqlSearchStrategyProvider(config$, logger, client, usage);
     const result = await strategy.search(
       emptyRequestHandlerContext,

src/plugins/query_enhancements/server/search/sql_search_strategy.ts

@@ -34,7 +34,12 @@ export const sqlSearchStrategyProvider = (
     search: async (context, request: any, options) => {
       try {
         const query: Query = request.body.query;
-        const rawResponse: any = await sqlFacet.describeQuery(context, request);
+        request.body.fetch_size = await context.core.uiSettings.client.get('discover:sampleSize');
+        const requestWithFetchSize = {
+          ...request,
+          body: { ...request.body, fetch_size: request.body.fetch_size },
+        };
+        const rawResponse: any = await sqlFacet.describeQuery(context, requestWithFetchSize);
 
         if (!rawResponse.success) throwFacetError(rawResponse);
 

src/plugins/query_enhancements/server/utils/facet.ts

@@ -45,6 +45,7 @@ export class Facet {
         body: {
           query: query.query,
           ...(meta?.name && { datasource: meta.name }),
+          ...(request.body?.fetch_size && { fetch_size: request.body.fetch_size }),
           ...(meta?.sessionId && {
             sessionId: meta.sessionId,
           }),