From 49a6bb2f26cca4e20b590bfec9eb936332e64392 Mon Sep 17 00:00:00 2001 From: Walter Rafelsberger Date: Tue, 14 Jul 2020 17:11:32 +0200 Subject: [PATCH] [ML] Kibana API endpoint for histogram chart data (#70976) (#71619) - Introduces dedicated Kibana API endpoints as part of ML and transform plugin API endpoints and moves the logic to query and transform the required data from client to server. - Adds support for sampling to retrieve the data for the field histograms. For now this is not configurable by the end user and is hard coded to 5000. This is to have a first iteration of this functionality in for 7.9 and protect users when querying large clusters. The button to enable the histogram charts now includes a tooltip that mentions the sampler. Co-authored-by: Elastic Machine --- .../ml/common/constants/field_histograms.ts | 8 + .../components/data_grid/data_grid.tsx | 41 ++- .../application/components/data_grid/index.ts | 2 +- .../data_grid/use_column_chart.test.ts | 18 ++ .../components/data_grid/use_column_chart.tsx | 186 +----------- .../hooks/use_index_data.ts | 24 +- .../use_exploration_results.ts | 28 +- .../outlier_exploration/use_outlier_data.ts | 30 +- .../index_based/common/index.ts | 2 +- .../index_based/common/request.ts | 7 + .../index_based/data_loader/data_loader.ts | 33 ++- .../datavisualizer/index_based/page.tsx | 8 +- .../services/ml_api_service/index.ts | 29 +- .../models/data_visualizer/data_visualizer.ts | 267 +++++++++++++++++- .../ml/server/models/data_visualizer/index.ts | 2 +- .../ml/server/routes/data_visualizer.ts | 61 +++- .../routes/schemas/data_visualizer_schema.ts | 9 + x-pack/plugins/ml/server/shared.ts | 1 + .../transform/public/app/hooks/use_api.ts | 26 ++ .../public/app/hooks/use_index_data.ts | 15 +- .../transform/public/shared_imports.ts | 2 +- .../server/routes/api/field_histograms.ts | 50 ++++ .../transform/server/routes/api/schema.ts | 18 ++ .../plugins/transform/server/routes/index.ts | 2 + .../transform/server/shared_imports.ts | 7 + .../data_visualizer/get_field_histograms.ts | 122 ++++++++ .../outlier_detection_creation.ts | 22 ++ .../ml/data_frame_analytics_creation.ts | 52 ++++ 28 files changed, 822 insertions(+), 250 deletions(-) create mode 100644 x-pack/plugins/ml/common/constants/field_histograms.ts create mode 100644 x-pack/plugins/ml/public/application/components/data_grid/use_column_chart.test.ts create mode 100644 x-pack/plugins/transform/server/routes/api/field_histograms.ts create mode 100644 x-pack/plugins/transform/server/shared_imports.ts create mode 100644 x-pack/test/api_integration/apis/ml/data_visualizer/get_field_histograms.ts diff --git a/x-pack/plugins/ml/common/constants/field_histograms.ts b/x-pack/plugins/ml/common/constants/field_histograms.ts new file mode 100644 index 00000000000000..5c86c00ac666f1 --- /dev/null +++ b/x-pack/plugins/ml/common/constants/field_histograms.ts @@ -0,0 +1,8 @@ +/* + * 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. + */ + +// Default sampler shard size used for field histograms +export const DEFAULT_SAMPLER_SHARD_SIZE = 5000; diff --git a/x-pack/plugins/ml/public/application/components/data_grid/data_grid.tsx b/x-pack/plugins/ml/public/application/components/data_grid/data_grid.tsx index 9af7a869e0e568..d4be2eab13d26b 100644 --- a/x-pack/plugins/ml/public/application/components/data_grid/data_grid.tsx +++ b/x-pack/plugins/ml/public/application/components/data_grid/data_grid.tsx @@ -20,10 +20,13 @@ import { EuiFlexItem, EuiSpacer, EuiTitle, + EuiToolTip, } from '@elastic/eui'; import { CoreSetup } from 'src/core/public'; +import { DEFAULT_SAMPLER_SHARD_SIZE } from '../../../../common/constants/field_histograms'; + import { INDEX_STATUS } from '../../data_frame_analytics/common'; import { euiDataGridStyle, euiDataGridToolbarSettings } from './common'; @@ -193,21 +196,31 @@ export const DataGrid: FC = memo( ...(chartsButtonVisible ? { additionalControls: ( - - {i18n.translate('xpack.ml.dataGrid.histogramButtonText', { - defaultMessage: 'Histogram charts', + + > + + {i18n.translate('xpack.ml.dataGrid.histogramButtonText', { + defaultMessage: 'Histogram charts', + })} + + ), } : {}), diff --git a/x-pack/plugins/ml/public/application/components/data_grid/index.ts b/x-pack/plugins/ml/public/application/components/data_grid/index.ts index 80bc6b861f7425..4bbd3595e5a7e4 100644 --- a/x-pack/plugins/ml/public/application/components/data_grid/index.ts +++ b/x-pack/plugins/ml/public/application/components/data_grid/index.ts @@ -12,7 +12,7 @@ export { showDataGridColumnChartErrorMessageToast, useRenderCellValue, } from './common'; -export { fetchChartsData, ChartData } from './use_column_chart'; +export { getFieldType, ChartData } from './use_column_chart'; export { useDataGrid } from './use_data_grid'; export { DataGrid } from './data_grid'; export { diff --git a/x-pack/plugins/ml/public/application/components/data_grid/use_column_chart.test.ts b/x-pack/plugins/ml/public/application/components/data_grid/use_column_chart.test.ts new file mode 100644 index 00000000000000..1b35ef238d09e7 --- /dev/null +++ b/x-pack/plugins/ml/public/application/components/data_grid/use_column_chart.test.ts @@ -0,0 +1,18 @@ +/* + * 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. + */ + +import { getFieldType } from './use_column_chart'; + +describe('getFieldType()', () => { + it('should return the Kibana field type for a given EUI data grid schema', () => { + expect(getFieldType('text')).toBe('string'); + expect(getFieldType('datetime')).toBe('date'); + expect(getFieldType('numeric')).toBe('number'); + expect(getFieldType('boolean')).toBe('boolean'); + expect(getFieldType('json')).toBe('object'); + expect(getFieldType('non-aggregatable')).toBe(undefined); + }); +}); diff --git a/x-pack/plugins/ml/public/application/components/data_grid/use_column_chart.tsx b/x-pack/plugins/ml/public/application/components/data_grid/use_column_chart.tsx index 6b207a999eb52a..a762c44e243bf0 100644 --- a/x-pack/plugins/ml/public/application/components/data_grid/use_column_chart.tsx +++ b/x-pack/plugins/ml/public/application/components/data_grid/use_column_chart.tsx @@ -16,8 +16,6 @@ import { i18n } from '@kbn/i18n'; import { KBN_FIELD_TYPES } from '../../../../../../../src/plugins/data/public'; -import { stringHash } from '../../../../common/util/string_utils'; - import { NON_AGGREGATABLE } from './common'; export const hoveredRow$ = new BehaviorSubject(null); @@ -40,7 +38,7 @@ const getXScaleType = (kbnFieldType: KBN_FIELD_TYPES | undefined): XScaleType => } }; -const getFieldType = (schema: EuiDataGridColumn['schema']): KBN_FIELD_TYPES | undefined => { +export const getFieldType = (schema: EuiDataGridColumn['schema']): KBN_FIELD_TYPES | undefined => { if (schema === NON_AGGREGATABLE) { return undefined; } @@ -67,188 +65,6 @@ const getFieldType = (schema: EuiDataGridColumn['schema']): KBN_FIELD_TYPES | un return fieldType; }; -interface NumericColumnStats { - interval: number; - min: number; - max: number; -} -type NumericColumnStatsMap = Record; -const getAggIntervals = async ( - indexPatternTitle: string, - esSearch: (payload: any) => Promise, - query: any, - columnTypes: EuiDataGridColumn[] -): Promise => { - const numericColumns = columnTypes.filter((cT) => { - const fieldType = getFieldType(cT.schema); - return fieldType === KBN_FIELD_TYPES.NUMBER || fieldType === KBN_FIELD_TYPES.DATE; - }); - - if (numericColumns.length === 0) { - return {}; - } - - const minMaxAggs = numericColumns.reduce((aggs, c) => { - const id = stringHash(c.id); - aggs[id] = { - stats: { - field: c.id, - }, - }; - return aggs; - }, {} as Record); - - const respStats = await esSearch({ - index: indexPatternTitle, - size: 0, - body: { - query, - aggs: minMaxAggs, - size: 0, - }, - }); - - return Object.keys(respStats.aggregations).reduce((p, aggName) => { - const stats = [respStats.aggregations[aggName].min, respStats.aggregations[aggName].max]; - if (!stats.includes(null)) { - const delta = respStats.aggregations[aggName].max - respStats.aggregations[aggName].min; - - let aggInterval = 1; - - if (delta > MAX_CHART_COLUMNS) { - aggInterval = Math.round(delta / MAX_CHART_COLUMNS); - } - - if (delta <= 1) { - aggInterval = delta / MAX_CHART_COLUMNS; - } - - p[aggName] = { interval: aggInterval, min: stats[0], max: stats[1] }; - } - - return p; - }, {} as NumericColumnStatsMap); -}; - -interface AggHistogram { - histogram: { - field: string; - interval: number; - }; -} - -interface AggCardinality { - cardinality: { - field: string; - }; -} - -interface AggTerms { - terms: { - field: string; - size: number; - }; -} - -type ChartRequestAgg = AggHistogram | AggCardinality | AggTerms; - -export const fetchChartsData = async ( - indexPatternTitle: string, - esSearch: (payload: any) => Promise, - query: any, - columnTypes: EuiDataGridColumn[] -): Promise => { - const aggIntervals = await getAggIntervals(indexPatternTitle, esSearch, query, columnTypes); - - const chartDataAggs = columnTypes.reduce((aggs, c) => { - const fieldType = getFieldType(c.schema); - const id = stringHash(c.id); - if (fieldType === KBN_FIELD_TYPES.NUMBER || fieldType === KBN_FIELD_TYPES.DATE) { - if (aggIntervals[id] !== undefined) { - aggs[`${id}_histogram`] = { - histogram: { - field: c.id, - interval: aggIntervals[id].interval !== 0 ? aggIntervals[id].interval : 1, - }, - }; - } - } else if (fieldType === KBN_FIELD_TYPES.STRING || fieldType === KBN_FIELD_TYPES.BOOLEAN) { - if (fieldType === KBN_FIELD_TYPES.STRING) { - aggs[`${id}_cardinality`] = { - cardinality: { - field: c.id, - }, - }; - } - aggs[`${id}_terms`] = { - terms: { - field: c.id, - size: MAX_CHART_COLUMNS, - }, - }; - } - return aggs; - }, {} as Record); - - if (Object.keys(chartDataAggs).length === 0) { - return []; - } - - const respChartsData = await esSearch({ - index: indexPatternTitle, - size: 0, - body: { - query, - aggs: chartDataAggs, - size: 0, - }, - }); - - const chartsData: ChartData[] = columnTypes.map( - (c): ChartData => { - const fieldType = getFieldType(c.schema); - const id = stringHash(c.id); - - if (fieldType === KBN_FIELD_TYPES.NUMBER || fieldType === KBN_FIELD_TYPES.DATE) { - if (aggIntervals[id] === undefined) { - return { - type: 'numeric', - data: [], - interval: 0, - stats: [0, 0], - id: c.id, - }; - } - - return { - data: respChartsData.aggregations[`${id}_histogram`].buckets, - interval: aggIntervals[id].interval, - stats: [aggIntervals[id].min, aggIntervals[id].max], - type: 'numeric', - id: c.id, - }; - } else if (fieldType === KBN_FIELD_TYPES.STRING || fieldType === KBN_FIELD_TYPES.BOOLEAN) { - return { - type: fieldType === KBN_FIELD_TYPES.STRING ? 'ordinal' : 'boolean', - cardinality: - fieldType === KBN_FIELD_TYPES.STRING - ? respChartsData.aggregations[`${id}_cardinality`].value - : 2, - data: respChartsData.aggregations[`${id}_terms`].buckets, - id: c.id, - }; - } - - return { - type: 'unsupported', - id: c.id, - }; - } - ); - - return chartsData; -}; - interface NumericDataItem { key: number; key_as_string?: string; diff --git a/x-pack/plugins/ml/public/application/data_frame_analytics/pages/analytics_creation/hooks/use_index_data.ts b/x-pack/plugins/ml/public/application/data_frame_analytics/pages/analytics_creation/hooks/use_index_data.ts index ee0e5c1955eadd..2cecffc9932570 100644 --- a/x-pack/plugins/ml/public/application/data_frame_analytics/pages/analytics_creation/hooks/use_index_data.ts +++ b/x-pack/plugins/ml/public/application/data_frame_analytics/pages/analytics_creation/hooks/use_index_data.ts @@ -4,15 +4,18 @@ * you may not use this file except in compliance with the Elastic License. */ -import { useEffect } from 'react'; +import { useEffect, useMemo } from 'react'; import { EuiDataGridColumn } from '@elastic/eui'; import { CoreSetup } from 'src/core/public'; import { IndexPattern } from '../../../../../../../../../src/plugins/data/public'; + +import { DataLoader } from '../../../../datavisualizer/index_based/data_loader'; + import { - fetchChartsData, + getFieldType, getDataGridSchemaFromKibanaFieldType, getFieldsFromKibanaIndexPattern, showDataGridColumnChartErrorMessageToast, @@ -103,13 +106,20 @@ export const useIndexData = ( // eslint-disable-next-line react-hooks/exhaustive-deps }, [indexPattern.title, JSON.stringify([query, pagination, sortingColumns])]); + const dataLoader = useMemo(() => new DataLoader(indexPattern, toastNotifications), [ + indexPattern, + ]); + const fetchColumnChartsData = async function () { try { - const columnChartsData = await fetchChartsData( - indexPattern.title, - ml.esSearch, - query, - columns.filter((cT) => dataGrid.visibleColumns.includes(cT.id)) + const columnChartsData = await dataLoader.loadFieldHistograms( + columns + .filter((cT) => dataGrid.visibleColumns.includes(cT.id)) + .map((cT) => ({ + fieldName: cT.id, + type: getFieldType(cT.schema), + })), + query ); dataGrid.setColumnCharts(columnChartsData); } catch (e) { diff --git a/x-pack/plugins/ml/public/application/data_frame_analytics/pages/analytics_exploration/components/exploration_results_table/use_exploration_results.ts b/x-pack/plugins/ml/public/application/data_frame_analytics/pages/analytics_exploration/components/exploration_results_table/use_exploration_results.ts index 796670f6a864df..98dd40986e32b6 100644 --- a/x-pack/plugins/ml/public/application/data_frame_analytics/pages/analytics_exploration/components/exploration_results_table/use_exploration_results.ts +++ b/x-pack/plugins/ml/public/application/data_frame_analytics/pages/analytics_exploration/components/exploration_results_table/use_exploration_results.ts @@ -4,7 +4,7 @@ * you may not use this file except in compliance with the Elastic License. */ -import { useEffect } from 'react'; +import { useEffect, useMemo } from 'react'; import { EuiDataGridColumn } from '@elastic/eui'; @@ -12,16 +12,17 @@ import { CoreSetup } from 'src/core/public'; import { IndexPattern } from '../../../../../../../../../../src/plugins/data/public'; +import { DataLoader } from '../../../../../datavisualizer/index_based/data_loader'; + import { - fetchChartsData, getDataGridSchemasFromFieldTypes, + getFieldType, showDataGridColumnChartErrorMessageToast, useDataGrid, useRenderCellValue, UseIndexDataReturnType, } from '../../../../../components/data_grid'; import { SavedSearchQuery } from '../../../../../contexts/ml'; -import { ml } from '../../../../../services/ml_api_service'; import { getIndexData, getIndexFields, DataFrameAnalyticsConfig } from '../../../../common'; import { @@ -72,14 +73,23 @@ export const useExplorationResults = ( // eslint-disable-next-line react-hooks/exhaustive-deps }, [jobConfig && jobConfig.id, dataGrid.pagination, searchQuery, dataGrid.sortingColumns]); + const dataLoader = useMemo( + () => + indexPattern !== undefined ? new DataLoader(indexPattern, toastNotifications) : undefined, + [indexPattern] + ); + const fetchColumnChartsData = async function () { try { - if (jobConfig !== undefined) { - const columnChartsData = await fetchChartsData( - jobConfig.dest.index, - ml.esSearch, - searchQuery, - columns.filter((cT) => dataGrid.visibleColumns.includes(cT.id)) + if (jobConfig !== undefined && dataLoader !== undefined) { + const columnChartsData = await dataLoader.loadFieldHistograms( + columns + .filter((cT) => dataGrid.visibleColumns.includes(cT.id)) + .map((cT) => ({ + fieldName: cT.id, + type: getFieldType(cT.schema), + })), + searchQuery ); dataGrid.setColumnCharts(columnChartsData); } diff --git a/x-pack/plugins/ml/public/application/data_frame_analytics/pages/analytics_exploration/components/outlier_exploration/use_outlier_data.ts b/x-pack/plugins/ml/public/application/data_frame_analytics/pages/analytics_exploration/components/outlier_exploration/use_outlier_data.ts index beb6836bf801fa..90294a09c0adc3 100644 --- a/x-pack/plugins/ml/public/application/data_frame_analytics/pages/analytics_exploration/components/outlier_exploration/use_outlier_data.ts +++ b/x-pack/plugins/ml/public/application/data_frame_analytics/pages/analytics_exploration/components/outlier_exploration/use_outlier_data.ts @@ -4,19 +4,21 @@ * you may not use this file except in compliance with the Elastic License. */ -import { useEffect } from 'react'; +import { useEffect, useMemo } from 'react'; import { EuiDataGridColumn } from '@elastic/eui'; import { IndexPattern } from '../../../../../../../../../../src/plugins/data/public'; +import { DataLoader } from '../../../../../datavisualizer/index_based/data_loader'; + import { useColorRange, COLOR_RANGE, COLOR_RANGE_SCALE, } from '../../../../../components/color_range_legend'; import { - fetchChartsData, + getFieldType, getDataGridSchemasFromFieldTypes, showDataGridColumnChartErrorMessageToast, useDataGrid, @@ -24,7 +26,6 @@ import { UseIndexDataReturnType, } from '../../../../../components/data_grid'; import { SavedSearchQuery } from '../../../../../contexts/ml'; -import { ml } from '../../../../../services/ml_api_service'; import { getToastNotifications } from '../../../../../util/dependency_cache'; import { getIndexData, getIndexFields, DataFrameAnalyticsConfig } from '../../../../common'; @@ -79,14 +80,25 @@ export const useOutlierData = ( // eslint-disable-next-line react-hooks/exhaustive-deps }, [jobConfig && jobConfig.id, dataGrid.pagination, searchQuery, dataGrid.sortingColumns]); + const dataLoader = useMemo( + () => + indexPattern !== undefined + ? new DataLoader(indexPattern, getToastNotifications()) + : undefined, + [indexPattern] + ); + const fetchColumnChartsData = async function () { try { - if (jobConfig !== undefined) { - const columnChartsData = await fetchChartsData( - jobConfig.dest.index, - ml.esSearch, - searchQuery, - columns.filter((cT) => dataGrid.visibleColumns.includes(cT.id)) + if (jobConfig !== undefined && dataLoader !== undefined) { + const columnChartsData = await dataLoader.loadFieldHistograms( + columns + .filter((cT) => dataGrid.visibleColumns.includes(cT.id)) + .map((cT) => ({ + fieldName: cT.id, + type: getFieldType(cT.schema), + })), + searchQuery ); dataGrid.setColumnCharts(columnChartsData); } diff --git a/x-pack/plugins/ml/public/application/datavisualizer/index_based/common/index.ts b/x-pack/plugins/ml/public/application/datavisualizer/index_based/common/index.ts index 5618f701e4c5fd..50278c300d1032 100644 --- a/x-pack/plugins/ml/public/application/datavisualizer/index_based/common/index.ts +++ b/x-pack/plugins/ml/public/application/datavisualizer/index_based/common/index.ts @@ -5,4 +5,4 @@ */ export { FieldVisConfig } from './field_vis_config'; -export { FieldRequestConfig } from './request'; +export { FieldHistogramRequestConfig, FieldRequestConfig } from './request'; diff --git a/x-pack/plugins/ml/public/application/datavisualizer/index_based/common/request.ts b/x-pack/plugins/ml/public/application/datavisualizer/index_based/common/request.ts index 9a886cbc899c24..fd4888b8729c18 100644 --- a/x-pack/plugins/ml/public/application/datavisualizer/index_based/common/request.ts +++ b/x-pack/plugins/ml/public/application/datavisualizer/index_based/common/request.ts @@ -4,6 +4,8 @@ * you may not use this file except in compliance with the Elastic License. */ +import { KBN_FIELD_TYPES } from '../../../../../../../../src/plugins/data/public'; + import { ML_JOB_FIELD_TYPES } from '../../../../../common/constants/field_types'; export interface FieldRequestConfig { @@ -11,3 +13,8 @@ export interface FieldRequestConfig { type: ML_JOB_FIELD_TYPES; cardinality: number; } + +export interface FieldHistogramRequestConfig { + fieldName: string; + type?: KBN_FIELD_TYPES; +} diff --git a/x-pack/plugins/ml/public/application/datavisualizer/index_based/data_loader/data_loader.ts b/x-pack/plugins/ml/public/application/datavisualizer/index_based/data_loader/data_loader.ts index a08821c65bfe79..34f86ffa187883 100644 --- a/x-pack/plugins/ml/public/application/datavisualizer/index_based/data_loader/data_loader.ts +++ b/x-pack/plugins/ml/public/application/datavisualizer/index_based/data_loader/data_loader.ts @@ -6,15 +6,17 @@ import { i18n } from '@kbn/i18n'; -import { getToastNotifications } from '../../../util/dependency_cache'; +import { CoreSetup } from 'src/core/public'; + import { IndexPattern } from '../../../../../../../../src/plugins/data/public'; import { SavedSearchQuery } from '../../../contexts/ml'; import { OMIT_FIELDS } from '../../../../../common/constants/field_types'; import { IndexPatternTitle } from '../../../../../common/types/kibana'; +import { DEFAULT_SAMPLER_SHARD_SIZE } from '../../../../../common/constants/field_histograms'; import { ml } from '../../../services/ml_api_service'; -import { FieldRequestConfig } from '../common'; +import { FieldHistogramRequestConfig, FieldRequestConfig } from '../common'; // Maximum number of examples to obtain for text type fields. const MAX_EXAMPLES_DEFAULT: number = 10; @@ -23,10 +25,15 @@ export class DataLoader { private _indexPattern: IndexPattern; private _indexPatternTitle: IndexPatternTitle = ''; private _maxExamples: number = MAX_EXAMPLES_DEFAULT; + private _toastNotifications: CoreSetup['notifications']['toasts']; - constructor(indexPattern: IndexPattern, kibanaConfig: any) { + constructor( + indexPattern: IndexPattern, + toastNotifications: CoreSetup['notifications']['toasts'] + ) { this._indexPattern = indexPattern; this._indexPatternTitle = indexPattern.title; + this._toastNotifications = toastNotifications; } async loadOverallData( @@ -90,10 +97,24 @@ export class DataLoader { return stats; } + async loadFieldHistograms( + fields: FieldHistogramRequestConfig[], + query: string | SavedSearchQuery, + samplerShardSize = DEFAULT_SAMPLER_SHARD_SIZE + ): Promise { + const stats = await ml.getVisualizerFieldHistograms({ + indexPatternTitle: this._indexPatternTitle, + query, + fields, + samplerShardSize, + }); + + return stats; + } + displayError(err: any) { - const toastNotifications = getToastNotifications(); if (err.statusCode === 500) { - toastNotifications.addDanger( + this._toastNotifications.addDanger( i18n.translate('xpack.ml.datavisualizer.dataLoader.internalServerErrorMessage', { defaultMessage: 'Error loading data in index {index}. {message}. ' + @@ -105,7 +126,7 @@ export class DataLoader { }) ); } else { - toastNotifications.addDanger( + this._toastNotifications.addDanger( i18n.translate('xpack.ml.datavisualizer.page.errorLoadingDataMessage', { defaultMessage: 'Error loading data in index {index}. {message}', values: { diff --git a/x-pack/plugins/ml/public/application/datavisualizer/index_based/page.tsx b/x-pack/plugins/ml/public/application/datavisualizer/index_based/page.tsx index 97b4043c9fd644..3c332d305d7e99 100644 --- a/x-pack/plugins/ml/public/application/datavisualizer/index_based/page.tsx +++ b/x-pack/plugins/ml/public/application/datavisualizer/index_based/page.tsx @@ -4,7 +4,7 @@ * you may not use this file except in compliance with the Elastic License. */ -import React, { FC, Fragment, useEffect, useState } from 'react'; +import React, { FC, Fragment, useEffect, useMemo, useState } from 'react'; import { merge } from 'rxjs'; import { i18n } from '@kbn/i18n'; @@ -43,6 +43,7 @@ import { kbnTypeToMLJobType } from '../../util/field_types_utils'; import { useTimefilter } from '../../contexts/kibana'; import { timeBasedIndexCheck, getQueryFromSavedSearch } from '../../util/index_utils'; import { getTimeBucketsFromCache } from '../../util/time_buckets'; +import { getToastNotifications } from '../../util/dependency_cache'; import { useUrlState } from '../../util/url_state'; import { FieldRequestConfig, FieldVisConfig } from './common'; import { ActionsPanel } from './components/actions_panel'; @@ -107,7 +108,10 @@ export const Page: FC = () => { autoRefreshSelector: true, }); - const dataLoader = new DataLoader(currentIndexPattern, kibanaConfig); + const dataLoader = useMemo(() => new DataLoader(currentIndexPattern, getToastNotifications()), [ + currentIndexPattern, + ]); + const [globalState, setGlobalState] = useUrlState('_g'); useEffect(() => { if (globalState?.time !== undefined) { diff --git a/x-pack/plugins/ml/public/application/services/ml_api_service/index.ts b/x-pack/plugins/ml/public/application/services/ml_api_service/index.ts index d1b6f95f32bed5..599e4d4bb8a10e 100644 --- a/x-pack/plugins/ml/public/application/services/ml_api_service/index.ts +++ b/x-pack/plugins/ml/public/application/services/ml_api_service/index.ts @@ -27,7 +27,10 @@ import { ModelSnapshot, } from '../../../../common/types/anomaly_detection_jobs'; import { ES_AGGREGATION } from '../../../../common/constants/aggregation_types'; -import { FieldRequestConfig } from '../../datavisualizer/index_based/common'; +import { + FieldHistogramRequestConfig, + FieldRequestConfig, +} from '../../datavisualizer/index_based/common'; import { DataRecognizerConfigResponse, Module } from '../../../../common/types/modules'; import { getHttp } from '../../util/dependency_cache'; @@ -494,6 +497,30 @@ export function mlApiServicesProvider(httpService: HttpService) { }); }, + getVisualizerFieldHistograms({ + indexPatternTitle, + query, + fields, + samplerShardSize, + }: { + indexPatternTitle: string; + query: any; + fields: FieldHistogramRequestConfig[]; + samplerShardSize?: number; + }) { + const body = JSON.stringify({ + query, + fields, + samplerShardSize, + }); + + return httpService.http({ + path: `${basePath()}/data_visualizer/get_field_histograms/${indexPatternTitle}`, + method: 'POST', + body, + }); + }, + getVisualizerOverallStats({ indexPatternTitle, query, diff --git a/x-pack/plugins/ml/server/models/data_visualizer/data_visualizer.ts b/x-pack/plugins/ml/server/models/data_visualizer/data_visualizer.ts index d58c797b446db6..d1a4a0b585fbb9 100644 --- a/x-pack/plugins/ml/server/models/data_visualizer/data_visualizer.ts +++ b/x-pack/plugins/ml/server/models/data_visualizer/data_visualizer.ts @@ -4,10 +4,12 @@ * you may not use this file except in compliance with the Elastic License. */ -import { LegacyCallAPIOptions, LegacyAPICaller } from 'kibana/server'; +import { LegacyAPICaller } from 'kibana/server'; import _ from 'lodash'; +import { KBN_FIELD_TYPES } from '../../../../../../src/plugins/data/server'; import { ML_JOB_FIELD_TYPES } from '../../../common/constants/field_types'; import { getSafeAggregationName } from '../../../common/util/job_utils'; +import { stringHash } from '../../../common/util/string_utils'; import { buildBaseFilterCriteria, buildSamplerAggregation, @@ -19,6 +21,8 @@ const SAMPLER_TOP_TERMS_SHARD_SIZE = 5000; const AGGREGATABLE_EXISTS_REQUEST_BATCH_SIZE = 200; const FIELDS_REQUEST_BATCH_SIZE = 10; +const MAX_CHART_COLUMNS = 20; + interface FieldData { fieldName: string; existsInDocs: boolean; @@ -35,6 +39,11 @@ export interface Field { cardinality: number; } +export interface HistogramField { + fieldName: string; + type: string; +} + interface Distribution { percentiles: any[]; minPercentile: number; @@ -98,6 +107,70 @@ interface FieldExamples { examples: any[]; } +interface NumericColumnStats { + interval: number; + min: number; + max: number; +} +type NumericColumnStatsMap = Record; + +interface AggHistogram { + histogram: { + field: string; + interval: number; + }; +} + +interface AggCardinality { + cardinality: { + field: string; + }; +} + +interface AggTerms { + terms: { + field: string; + size: number; + }; +} + +interface NumericDataItem { + key: number; + key_as_string?: string; + doc_count: number; +} + +interface NumericChartData { + data: NumericDataItem[]; + id: string; + interval: number; + stats: [number, number]; + type: 'numeric'; +} + +interface OrdinalDataItem { + key: string; + key_as_string?: string; + doc_count: number; +} + +interface OrdinalChartData { + type: 'ordinal' | 'boolean'; + cardinality: number; + data: OrdinalDataItem[]; + id: string; +} + +interface UnsupportedChartData { + id: string; + type: 'unsupported'; +} + +type ChartRequestAgg = AggHistogram | AggCardinality | AggTerms; + +// type ChartDataItem = NumericDataItem | OrdinalDataItem; +type ChartData = NumericChartData | OrdinalChartData | UnsupportedChartData; + type BatchStats = | NumericFieldStats | StringFieldStats @@ -106,12 +179,176 @@ type BatchStats = | DocumentCountStats | FieldExamples; +const getAggIntervals = async ( + callAsCurrentUser: LegacyAPICaller, + indexPatternTitle: string, + query: any, + fields: HistogramField[], + samplerShardSize: number +): Promise => { + const numericColumns = fields.filter((field) => { + return field.type === KBN_FIELD_TYPES.NUMBER || field.type === KBN_FIELD_TYPES.DATE; + }); + + if (numericColumns.length === 0) { + return {}; + } + + const minMaxAggs = numericColumns.reduce((aggs, c) => { + const id = stringHash(c.fieldName); + aggs[id] = { + stats: { + field: c.fieldName, + }, + }; + return aggs; + }, {} as Record); + + const respStats = await callAsCurrentUser('search', { + index: indexPatternTitle, + size: 0, + body: { + query, + aggs: buildSamplerAggregation(minMaxAggs, samplerShardSize), + size: 0, + }, + }); + + const aggsPath = getSamplerAggregationsResponsePath(samplerShardSize); + const aggregations = + aggsPath.length > 0 ? _.get(respStats.aggregations, aggsPath) : respStats.aggregations; + + return Object.keys(aggregations).reduce((p, aggName) => { + const stats = [aggregations[aggName].min, aggregations[aggName].max]; + if (!stats.includes(null)) { + const delta = aggregations[aggName].max - aggregations[aggName].min; + + let aggInterval = 1; + + if (delta > MAX_CHART_COLUMNS || delta <= 1) { + aggInterval = delta / (MAX_CHART_COLUMNS - 1); + } + + p[aggName] = { interval: aggInterval, min: stats[0], max: stats[1] }; + } + + return p; + }, {} as NumericColumnStatsMap); +}; + +// export for re-use by transforms plugin +export const getHistogramsForFields = async ( + callAsCurrentUser: LegacyAPICaller, + indexPatternTitle: string, + query: any, + fields: HistogramField[], + samplerShardSize: number +) => { + const aggIntervals = await getAggIntervals( + callAsCurrentUser, + indexPatternTitle, + query, + fields, + samplerShardSize + ); + + const chartDataAggs = fields.reduce((aggs, field) => { + const fieldName = field.fieldName; + const fieldType = field.type; + const id = stringHash(fieldName); + if (fieldType === KBN_FIELD_TYPES.NUMBER || fieldType === KBN_FIELD_TYPES.DATE) { + if (aggIntervals[id] !== undefined) { + aggs[`${id}_histogram`] = { + histogram: { + field: fieldName, + interval: aggIntervals[id].interval !== 0 ? aggIntervals[id].interval : 1, + }, + }; + } + } else if (fieldType === KBN_FIELD_TYPES.STRING || fieldType === KBN_FIELD_TYPES.BOOLEAN) { + if (fieldType === KBN_FIELD_TYPES.STRING) { + aggs[`${id}_cardinality`] = { + cardinality: { + field: fieldName, + }, + }; + } + aggs[`${id}_terms`] = { + terms: { + field: fieldName, + size: MAX_CHART_COLUMNS, + }, + }; + } + return aggs; + }, {} as Record); + + if (Object.keys(chartDataAggs).length === 0) { + return []; + } + + const respChartsData = await callAsCurrentUser('search', { + index: indexPatternTitle, + size: 0, + body: { + query, + aggs: buildSamplerAggregation(chartDataAggs, samplerShardSize), + size: 0, + }, + }); + + const aggsPath = getSamplerAggregationsResponsePath(samplerShardSize); + const aggregations = + aggsPath.length > 0 + ? _.get(respChartsData.aggregations, aggsPath) + : respChartsData.aggregations; + + const chartsData: ChartData[] = fields.map( + (field): ChartData => { + const fieldName = field.fieldName; + const fieldType = field.type; + const id = stringHash(field.fieldName); + + if (fieldType === KBN_FIELD_TYPES.NUMBER || fieldType === KBN_FIELD_TYPES.DATE) { + if (aggIntervals[id] === undefined) { + return { + type: 'numeric', + data: [], + interval: 0, + stats: [0, 0], + id: fieldName, + }; + } + + return { + data: aggregations[`${id}_histogram`].buckets, + interval: aggIntervals[id].interval, + stats: [aggIntervals[id].min, aggIntervals[id].max], + type: 'numeric', + id: fieldName, + }; + } else if (fieldType === KBN_FIELD_TYPES.STRING || fieldType === KBN_FIELD_TYPES.BOOLEAN) { + return { + type: fieldType === KBN_FIELD_TYPES.STRING ? 'ordinal' : 'boolean', + cardinality: + fieldType === KBN_FIELD_TYPES.STRING ? aggregations[`${id}_cardinality`].value : 2, + data: aggregations[`${id}_terms`].buckets, + id: fieldName, + }; + } + + return { + type: 'unsupported', + id: fieldName, + }; + } + ); + + return chartsData; +}; + export class DataVisualizer { - callAsCurrentUser: ( - endpoint: string, - clientParams: Record, - options?: LegacyCallAPIOptions - ) => Promise; + callAsCurrentUser: LegacyAPICaller; constructor(callAsCurrentUser: LegacyAPICaller) { this.callAsCurrentUser = callAsCurrentUser; @@ -200,6 +437,24 @@ export class DataVisualizer { return stats; } + // Obtains binned histograms for supplied list of fields. The statistics for each field in the + // returned array depend on the type of the field (keyword, number, date etc). + // Sampling will be used if supplied samplerShardSize > 0. + async getHistogramsForFields( + indexPatternTitle: string, + query: any, + fields: HistogramField[], + samplerShardSize: number + ): Promise { + return await getHistogramsForFields( + this.callAsCurrentUser, + indexPatternTitle, + query, + fields, + samplerShardSize + ); + } + // Obtains statistics for supplied list of fields. The statistics for each field in the // returned array depend on the type of the field (keyword, number, date etc). // Sampling will be used if supplied samplerShardSize > 0. diff --git a/x-pack/plugins/ml/server/models/data_visualizer/index.ts b/x-pack/plugins/ml/server/models/data_visualizer/index.ts index ed44e9b12e1d14..ca1df0fe8300c9 100644 --- a/x-pack/plugins/ml/server/models/data_visualizer/index.ts +++ b/x-pack/plugins/ml/server/models/data_visualizer/index.ts @@ -4,4 +4,4 @@ * you may not use this file except in compliance with the Elastic License. */ -export { DataVisualizer } from './data_visualizer'; +export { getHistogramsForFields, DataVisualizer } from './data_visualizer'; diff --git a/x-pack/plugins/ml/server/routes/data_visualizer.ts b/x-pack/plugins/ml/server/routes/data_visualizer.ts index 04008a896a1a22..9dd010e105b6e7 100644 --- a/x-pack/plugins/ml/server/routes/data_visualizer.ts +++ b/x-pack/plugins/ml/server/routes/data_visualizer.ts @@ -7,8 +7,9 @@ import { RequestHandlerContext } from 'kibana/server'; import { wrapError } from '../client/error_wrapper'; import { DataVisualizer } from '../models/data_visualizer'; -import { Field } from '../models/data_visualizer/data_visualizer'; +import { Field, HistogramField } from '../models/data_visualizer/data_visualizer'; import { + dataVisualizerFieldHistogramsSchema, dataVisualizerFieldStatsSchema, dataVisualizerOverallStatsSchema, indexPatternTitleSchema, @@ -65,10 +66,68 @@ function getStatsForFields( ); } +function getHistogramsForFields( + context: RequestHandlerContext, + indexPatternTitle: string, + query: any, + fields: HistogramField[], + samplerShardSize: number +) { + const dv = new DataVisualizer(context.ml!.mlClient.callAsCurrentUser); + return dv.getHistogramsForFields(indexPatternTitle, query, fields, samplerShardSize); +} + /** * Routes for the index data visualizer. */ export function dataVisualizerRoutes({ router, mlLicense }: RouteInitialization) { + /** + * @apiGroup DataVisualizer + * + * @api {post} /api/ml/data_visualizer/get_field_stats/:indexPatternTitle Get histograms for fields + * @apiName GetHistogramsForFields + * @apiDescription Returns the histograms on a list fields in the specified index pattern. + * + * @apiSchema (params) indexPatternTitleSchema + * @apiSchema (body) dataVisualizerFieldHistogramsSchema + * + * @apiSuccess {Object} fieldName histograms by field, keyed on the name of the field. + */ + router.post( + { + path: '/api/ml/data_visualizer/get_field_histograms/{indexPatternTitle}', + validate: { + params: indexPatternTitleSchema, + body: dataVisualizerFieldHistogramsSchema, + }, + options: { + tags: ['access:ml:canAccessML'], + }, + }, + mlLicense.basicLicenseAPIGuard(async (context, request, response) => { + try { + const { + params: { indexPatternTitle }, + body: { query, fields, samplerShardSize }, + } = request; + + const results = await getHistogramsForFields( + context, + indexPatternTitle, + query, + fields, + samplerShardSize + ); + + return response.ok({ + body: results, + }); + } catch (e) { + return response.customError(wrapError(e)); + } + }) + ); + /** * @apiGroup DataVisualizer * diff --git a/x-pack/plugins/ml/server/routes/schemas/data_visualizer_schema.ts b/x-pack/plugins/ml/server/routes/schemas/data_visualizer_schema.ts index b2d665954bd4dc..24e45514e1efce 100644 --- a/x-pack/plugins/ml/server/routes/schemas/data_visualizer_schema.ts +++ b/x-pack/plugins/ml/server/routes/schemas/data_visualizer_schema.ts @@ -11,6 +11,15 @@ export const indexPatternTitleSchema = schema.object({ indexPatternTitle: schema.string(), }); +export const dataVisualizerFieldHistogramsSchema = schema.object({ + /** Query to match documents in the index. */ + query: schema.any(), + /** The fields to return histogram data. */ + fields: schema.arrayOf(schema.any()), + /** Number of documents to be collected in the sample processed on each shard, or -1 for no sampling. */ + samplerShardSize: schema.number(), +}); + export const dataVisualizerFieldStatsSchema = schema.object({ /** Query to match documents in the index. */ query: schema.any(), diff --git a/x-pack/plugins/ml/server/shared.ts b/x-pack/plugins/ml/server/shared.ts index 3fca8ea1ba0478..100433b23f7d13 100644 --- a/x-pack/plugins/ml/server/shared.ts +++ b/x-pack/plugins/ml/server/shared.ts @@ -8,3 +8,4 @@ export * from '../common/types/anomalies'; export * from '../common/types/anomaly_detection_jobs'; export * from './lib/capabilities/errors'; export { ModuleSetupPayload } from './shared_services/providers/modules'; +export { getHistogramsForFields } from './models/data_visualizer/'; diff --git a/x-pack/plugins/transform/public/app/hooks/use_api.ts b/x-pack/plugins/transform/public/app/hooks/use_api.ts index 56528370a3ab9a..1d2752b9e939dc 100644 --- a/x-pack/plugins/transform/public/app/hooks/use_api.ts +++ b/x-pack/plugins/transform/public/app/hooks/use_api.ts @@ -5,6 +5,9 @@ */ import { useMemo } from 'react'; + +import { KBN_FIELD_TYPES } from '../../../../../../src/plugins/data/public'; + import { TransformId, TransformEndpointRequest, @@ -17,6 +20,15 @@ import { useAppDependencies } from '../app_dependencies'; import { GetTransformsResponse, PreviewRequestBody } from '../common'; import { EsIndex } from './use_api_types'; +import { SavedSearchQuery } from './use_search_items'; + +// Default sampler shard size used for field histograms +export const DEFAULT_SAMPLER_SHARD_SIZE = 5000; + +export interface FieldHistogramRequestConfig { + fieldName: string; + type?: KBN_FIELD_TYPES; +} export const useApi = () => { const { http } = useAppDependencies(); @@ -85,6 +97,20 @@ export const useApi = () => { getIndices(): Promise { return http.get(`/api/index_management/indices`); }, + getHistogramsForFields( + indexPatternTitle: string, + fields: FieldHistogramRequestConfig[], + query: string | SavedSearchQuery, + samplerShardSize = DEFAULT_SAMPLER_SHARD_SIZE + ) { + return http.post(`${API_BASE_PATH}field_histograms/${indexPatternTitle}`, { + body: JSON.stringify({ + query, + fields, + samplerShardSize, + }), + }); + }, }), [http] ); diff --git a/x-pack/plugins/transform/public/app/hooks/use_index_data.ts b/x-pack/plugins/transform/public/app/hooks/use_index_data.ts index c821c183ad370c..ad5850f26be2e2 100644 --- a/x-pack/plugins/transform/public/app/hooks/use_index_data.ts +++ b/x-pack/plugins/transform/public/app/hooks/use_index_data.ts @@ -9,7 +9,7 @@ import { useEffect } from 'react'; import { EuiDataGridColumn } from '@elastic/eui'; import { - fetchChartsData, + getFieldType, getDataGridSchemaFromKibanaFieldType, getFieldsFromKibanaIndexPattern, getErrorMessage, @@ -107,13 +107,16 @@ export const useIndexData = ( const fetchColumnChartsData = async function () { try { - const columnChartsData = await fetchChartsData( + const columnChartsData = await api.getHistogramsForFields( indexPattern.title, - api.esSearch, - isDefaultQuery(query) ? matchAllQuery : query, - columns.filter((cT) => dataGrid.visibleColumns.includes(cT.id)) + columns + .filter((cT) => dataGrid.visibleColumns.includes(cT.id)) + .map((cT) => ({ + fieldName: cT.id, + type: getFieldType(cT.schema), + })), + isDefaultQuery(query) ? matchAllQuery : query ); - setColumnCharts(columnChartsData); } catch (e) { showDataGridColumnChartErrorMessageToast(e, toastNotifications); diff --git a/x-pack/plugins/transform/public/shared_imports.ts b/x-pack/plugins/transform/public/shared_imports.ts index e0bbcd0b5d9db7..abbc39dd6c7287 100644 --- a/x-pack/plugins/transform/public/shared_imports.ts +++ b/x-pack/plugins/transform/public/shared_imports.ts @@ -14,7 +14,7 @@ export { } from '../../../../src/plugins/es_ui_shared/public'; export { - fetchChartsData, + getFieldType, getErrorMessage, extractErrorMessage, formatHumanReadableDateTimeSeconds, diff --git a/x-pack/plugins/transform/server/routes/api/field_histograms.ts b/x-pack/plugins/transform/server/routes/api/field_histograms.ts new file mode 100644 index 00000000000000..d602e49338846a --- /dev/null +++ b/x-pack/plugins/transform/server/routes/api/field_histograms.ts @@ -0,0 +1,50 @@ +/* + * 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. + */ +/* + * 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. + */ + +import { wrapEsError } from '../../../../../legacy/server/lib/create_router/error_wrappers'; + +import { getHistogramsForFields } from '../../shared_imports'; +import { RouteDependencies } from '../../types'; + +import { addBasePath } from '../index'; + +import { wrapError } from './error_utils'; +import { fieldHistogramsSchema, indexPatternTitleSchema, IndexPatternTitleSchema } from './schema'; + +export function registerFieldHistogramsRoutes({ router, license }: RouteDependencies) { + router.post( + { + path: addBasePath('field_histograms/{indexPatternTitle}'), + validate: { + params: indexPatternTitleSchema, + body: fieldHistogramsSchema, + }, + }, + license.guardApiRoute(async (ctx, req, res) => { + const { indexPatternTitle } = req.params as IndexPatternTitleSchema; + const { query, fields, samplerShardSize } = req.body; + + try { + const resp = await getHistogramsForFields( + ctx.transform!.dataClient.callAsCurrentUser, + indexPatternTitle, + query, + fields, + samplerShardSize + ); + + return res.ok({ body: resp }); + } catch (e) { + return res.customError(wrapError(wrapEsError(e))); + } + }) + ); +} diff --git a/x-pack/plugins/transform/server/routes/api/schema.ts b/x-pack/plugins/transform/server/routes/api/schema.ts index 7da3f1ccfe55e3..8aadef81b221b9 100644 --- a/x-pack/plugins/transform/server/routes/api/schema.ts +++ b/x-pack/plugins/transform/server/routes/api/schema.ts @@ -5,6 +5,24 @@ */ import { schema } from '@kbn/config-schema'; +export const fieldHistogramsSchema = schema.object({ + /** Query to match documents in the index. */ + query: schema.any(), + /** The fields to return histogram data. */ + fields: schema.arrayOf(schema.any()), + /** Number of documents to be collected in the sample processed on each shard, or -1 for no sampling. */ + samplerShardSize: schema.number(), +}); + +export const indexPatternTitleSchema = schema.object({ + /** Title of the index pattern for which to return stats. */ + indexPatternTitle: schema.string(), +}); + +export interface IndexPatternTitleSchema { + indexPatternTitle: string; +} + export const schemaTransformId = { params: schema.object({ transformId: schema.string(), diff --git a/x-pack/plugins/transform/server/routes/index.ts b/x-pack/plugins/transform/server/routes/index.ts index 07c21e58e64e44..4f35b094017a41 100644 --- a/x-pack/plugins/transform/server/routes/index.ts +++ b/x-pack/plugins/transform/server/routes/index.ts @@ -6,6 +6,7 @@ import { RouteDependencies } from '../types'; +import { registerFieldHistogramsRoutes } from './api/field_histograms'; import { registerPrivilegesRoute } from './api/privileges'; import { registerTransformsRoutes } from './api/transforms'; @@ -15,6 +16,7 @@ export const addBasePath = (uri: string): string => `${API_BASE_PATH}${uri}`; export class ApiRoutes { setup(dependencies: RouteDependencies) { + registerFieldHistogramsRoutes(dependencies); registerPrivilegesRoute(dependencies); registerTransformsRoutes(dependencies); } diff --git a/x-pack/plugins/transform/server/shared_imports.ts b/x-pack/plugins/transform/server/shared_imports.ts new file mode 100644 index 00000000000000..d1f86ac375721b --- /dev/null +++ b/x-pack/plugins/transform/server/shared_imports.ts @@ -0,0 +1,7 @@ +/* + * 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. + */ + +export { getHistogramsForFields } from '../../ml/server'; diff --git a/x-pack/test/api_integration/apis/ml/data_visualizer/get_field_histograms.ts b/x-pack/test/api_integration/apis/ml/data_visualizer/get_field_histograms.ts new file mode 100644 index 00000000000000..8b21c367d29f65 --- /dev/null +++ b/x-pack/test/api_integration/apis/ml/data_visualizer/get_field_histograms.ts @@ -0,0 +1,122 @@ +/* + * 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. + */ + +import expect from '@kbn/expect'; + +import { FtrProviderContext } from '../../../ftr_provider_context'; +import { USER } from '../../../../functional/services/ml/security_common'; +import { COMMON_REQUEST_HEADERS } from '../../../../functional/services/ml/common'; + +// eslint-disable-next-line import/no-default-export +export default ({ getService }: FtrProviderContext) => { + const esArchiver = getService('esArchiver'); + const supertest = getService('supertestWithoutAuth'); + const ml = getService('ml'); + + const fieldHistogramsTestData = { + testTitle: 'returns histogram data for fields', + index: 'ft_farequote', + user: USER.ML_POWERUSER, + requestBody: { + query: { bool: { should: [{ match_phrase: { airline: 'JZA' } }], minimum_should_match: 1 } }, + fields: [ + { fieldName: '@timestamp', type: 'date' }, + { fieldName: 'airline', type: 'string' }, + { fieldName: 'responsetime', type: 'number' }, + ], + samplerShardSize: -1, // No sampling, as otherwise counts could vary on each run. + }, + expected: { + responseCode: 200, + responseBody: [ + { + dataLength: 20, + type: 'numeric', + id: '@timestamp', + }, + { type: 'ordinal', dataLength: 1, id: 'airline' }, + { + dataLength: 20, + type: 'numeric', + id: 'responsetime', + }, + ], + }, + }; + + const errorTestData = { + testTitle: 'returns error for index which does not exist', + index: 'ft_farequote_not_exists', + user: USER.ML_POWERUSER, + requestBody: { + query: { bool: { must: [{ match_all: {} }] } }, + fields: [{ fieldName: 'responsetime', type: 'number' }], + samplerShardSize: -1, + }, + expected: { + responseCode: 404, + responseBody: { + statusCode: 404, + error: 'Not Found', + message: + '[index_not_found_exception] no such index [ft_farequote_not_exists], with { resource.type="index_or_alias" & resource.id="ft_farequote_not_exists" & index_uuid="_na_" & index="ft_farequote_not_exists" }', + }, + }, + }; + + async function runGetFieldHistogramsRequest( + index: string, + user: USER, + requestBody: object, + expectedResponsecode: number + ): Promise { + const { body } = await supertest + .post(`/api/ml/data_visualizer/get_field_histograms/${index}`) + .auth(user, ml.securityCommon.getPasswordForUser(user)) + .set(COMMON_REQUEST_HEADERS) + .send(requestBody) + .expect(expectedResponsecode); + + return body; + } + + describe('get_field_histograms', function () { + before(async () => { + await esArchiver.loadIfNeeded('ml/farequote'); + await ml.testResources.setKibanaTimeZoneToUTC(); + }); + + it(`${fieldHistogramsTestData.testTitle}`, async () => { + const body = await runGetFieldHistogramsRequest( + fieldHistogramsTestData.index, + fieldHistogramsTestData.user, + fieldHistogramsTestData.requestBody, + fieldHistogramsTestData.expected.responseCode + ); + + const expected = fieldHistogramsTestData.expected; + + const actual = body.map((b: any) => ({ + dataLength: b.data.length, + type: b.type, + id: b.id, + })); + expect(actual).to.eql(expected.responseBody); + }); + + it(`${errorTestData.testTitle}`, async () => { + const body = await runGetFieldHistogramsRequest( + errorTestData.index, + errorTestData.user, + errorTestData.requestBody, + errorTestData.expected.responseCode + ); + + expect(body.error).to.eql(errorTestData.expected.responseBody.error); + expect(body.message).to.eql(errorTestData.expected.responseBody.message); + }); + }); +}; diff --git a/x-pack/test/functional/apps/ml/data_frame_analytics/outlier_detection_creation.ts b/x-pack/test/functional/apps/ml/data_frame_analytics/outlier_detection_creation.ts index 6cdb9caa1e2db7..4ae93296f9be0a 100644 --- a/x-pack/test/functional/apps/ml/data_frame_analytics/outlier_detection_creation.ts +++ b/x-pack/test/functional/apps/ml/data_frame_analytics/outlier_detection_creation.ts @@ -37,6 +37,18 @@ export default function ({ getService }: FtrProviderContext) { modelMemory: '5mb', createIndexPattern: true, expected: { + histogramCharts: [ + { chartAvailable: true, id: '1stFlrSF', legend: '334 - 4692' }, + { chartAvailable: true, id: 'BsmtFinSF1', legend: '0 - 5644' }, + { chartAvailable: true, id: 'BsmtQual', legend: '0 - 5' }, + { chartAvailable: true, id: 'CentralAir', legend: '2 categories' }, + { chartAvailable: true, id: 'Condition2', legend: '2 categories' }, + { chartAvailable: true, id: 'Electrical', legend: '2 categories' }, + { chartAvailable: true, id: 'ExterQual', legend: '1 - 4' }, + { chartAvailable: true, id: 'Exterior1st', legend: '2 categories' }, + { chartAvailable: true, id: 'Exterior2nd', legend: '3 categories' }, + { chartAvailable: true, id: 'Fireplaces', legend: '0 - 3' }, + ], row: { type: 'outlier_detection', status: 'stopped', @@ -84,6 +96,16 @@ export default function ({ getService }: FtrProviderContext) { await ml.dataFrameAnalyticsCreation.assertSourceDataPreviewExists(); }); + it('enables the source data preview histogram charts', async () => { + await ml.dataFrameAnalyticsCreation.enableSourceDataPreviewHistogramCharts(); + }); + + it('displays the source data preview histogram charts', async () => { + await ml.dataFrameAnalyticsCreation.assertSourceDataPreviewHistogramCharts( + testData.expected.histogramCharts + ); + }); + it('displays the include fields selection', async () => { await ml.dataFrameAnalyticsCreation.assertIncludeFieldsSelectionExists(); }); diff --git a/x-pack/test/functional/services/ml/data_frame_analytics_creation.ts b/x-pack/test/functional/services/ml/data_frame_analytics_creation.ts index 1b756bbaca5d89..fc4aaa4fbf5fdd 100644 --- a/x-pack/test/functional/services/ml/data_frame_analytics_creation.ts +++ b/x-pack/test/functional/services/ml/data_frame_analytics_creation.ts @@ -128,6 +128,58 @@ export function MachineLearningDataFrameAnalyticsCreationProvider( await testSubjects.existOrFail('mlAnalyticsCreationDataGrid loaded', { timeout: 5000 }); }, + async assertIndexPreviewHistogramChartButtonExists() { + await testSubjects.existOrFail('mlAnalyticsCreationDataGridHistogramButton'); + }, + + async enableSourceDataPreviewHistogramCharts() { + await this.assertSourceDataPreviewHistogramChartButtonCheckState(false); + await testSubjects.click('mlAnalyticsCreationDataGridHistogramButton'); + await this.assertSourceDataPreviewHistogramChartButtonCheckState(true); + }, + + async assertSourceDataPreviewHistogramChartButtonCheckState(expectedCheckState: boolean) { + const actualCheckState = + (await testSubjects.getAttribute( + 'mlAnalyticsCreationDataGridHistogramButton', + 'aria-checked' + )) === 'true'; + expect(actualCheckState).to.eql( + expectedCheckState, + `Chart histogram button check state should be '${expectedCheckState}' (got '${actualCheckState}')` + ); + }, + + async assertSourceDataPreviewHistogramCharts( + expectedHistogramCharts: Array<{ chartAvailable: boolean; id: string; legend: string }> + ) { + // For each chart, get the content of each header cell and assert + // the legend text and column id and if the chart should be present or not. + await retry.tryForTime(5000, async () => { + for (const [index, expected] of expectedHistogramCharts.entries()) { + await testSubjects.existOrFail(`mlDataGridChart-${index}`); + + if (expected.chartAvailable) { + await testSubjects.existOrFail(`mlDataGridChart-${index}-histogram`); + } else { + await testSubjects.missingOrFail(`mlDataGridChart-${index}-histogram`); + } + + const actualLegend = await testSubjects.getVisibleText(`mlDataGridChart-${index}-legend`); + expect(actualLegend).to.eql( + expected.legend, + `Legend text for column '${index}' should be '${expected.legend}' (got '${actualLegend}')` + ); + + const actualId = await testSubjects.getVisibleText(`mlDataGridChart-${index}-id`); + expect(actualId).to.eql( + expected.id, + `Id text for column '${index}' should be '${expected.id}' (got '${actualId}')` + ); + } + }); + }, + async assertIncludeFieldsSelectionExists() { await testSubjects.existOrFail('mlAnalyticsCreateJobWizardIncludesSelect', { timeout: 5000 }); },