Fix logic for indices without time field, fix translations

elastic · Aug 23, 2022 · 3a2fff6 · 3a2fff6
1 parent 757c4a5
commit 3a2fff6
Show file tree

Hide file tree

Showing 4 changed files with 39 additions and 37 deletions.
diff --git a/...r/public/application/index_data_visualizer/search_strategy/requests/get_document_stats.ts b/...r/public/application/index_data_visualizer/search_strategy/requests/get_document_stats.ts
@@ -5,7 +5,7 @@
  * 2.0.
  */
 
-import { each, get, sortedIndex } from 'lodash';
+import { each, get, sortedIndex, first } from 'lodash';
 import type * as estypes from '@elastic/elasticsearch/lib/api/typesWithBodyKey';
 import { isPopulatedObject } from '@kbn/ml-is-populated-object';
 import { DataPublicPluginStart, ISearchOptions } from '@kbn/data-plugin/public';
@@ -67,7 +67,7 @@ export const getDocumentCountStats = async (
   // so that we can calculate the next p value that's appropriate for the data set
   const initialDefaultProbability = probability ?? 1e-5;
 
-  const getAggsWithRandomSampling = (p: number) => ({
+  const getEventRateAggsWithRandomSampling = (p: number) => ({
     sampler: {
       aggs,
       random_sampler: {
@@ -83,7 +83,7 @@ export const getDocumentCountStats = async (
     index,
     body: {
       query,
-      ...(hasTimeField ? { aggs: aggregations } : {}),
+      ...(aggregations !== undefined ? { aggs: aggregations } : {}),
       ...(isPopulatedObject(runtimeFieldMap) ? { runtime_mappings: runtimeFieldMap } : {}),
     },
     track_total_hits: !hasTimeField,
@@ -93,7 +93,27 @@ export const getDocumentCountStats = async (
   const firstResp = await search
     .search(
       {
-        params: getSearchParams(getAggsWithRandomSampling(initialDefaultProbability)),
+        params: getSearchParams(
+          hasTimeField
+            ? getEventRateAggsWithRandomSampling(initialDefaultProbability)
+            : // If there's no time field, we just make a value_count aggregation on any field
+              // just so we can calculate what's the optimal probability from the sampled doc count
+              {
+                sampler: {
+                  random_sampler: {
+                    probability: initialDefaultProbability,
+                  },
+                  aggs: {
+                    price_percentiles: {
+                      value_count: {
+                        field:
+                          first(params.aggregatableFields) ?? first(params.nonAggregatableFields),
+                      },
+                    },
+                  },
+                },
+              }
+        ),
       },
       searchOptions
     )
@@ -102,23 +122,11 @@ export const getDocumentCountStats = async (
   if (firstResp === undefined) {
     throw Error(
       `An error occurred with the following query ${JSON.stringify(
-        getSearchParams(getAggsWithRandomSampling(initialDefaultProbability))
+        getSearchParams(getEventRateAggsWithRandomSampling(initialDefaultProbability))
       )}`
     );
   }
 
-  if (!hasTimeField && isDefined(firstResp.rawResponse)) {
-    return {
-      ...result,
-      took: firstResp.rawResponse.took,
-      totalCount:
-        (typeof firstResp.rawResponse.hits?.total === 'number'
-          ? firstResp.rawResponse.hits?.total
-          : firstResp.rawResponse.hits?.total?.value) ?? 0,
-      probability: 1,
-    };
-  }
-
   if (isDefined(probability)) {
     return {
       ...result,
@@ -136,13 +144,25 @@ export const getDocumentCountStats = async (
     const newProbability =
       (initialDefaultProbability * numDocs) / (numSampled - 2 * Math.sqrt(numSampled));
 
+    if (!hasTimeField) {
+      return {
+        ...result,
+        took: firstResp.rawResponse.took,
+        totalCount:
+          (typeof firstResp.rawResponse.hits?.total === 'number'
+            ? firstResp.rawResponse.hits?.total
+            : firstResp.rawResponse.hits?.total?.value) ?? 0,
+        probability: numSampled < 109 ? 1 : newProbability,
+      };
+    }
+
     // If the number of docs sampled is indicative of query with < 10 million docs
     // proceed to make a vanilla aggregation without any sampling
     if (numSampled === 0 || newProbability === Infinity || numSampled < 109) {
       const vanillaAggResp = await search
         .search(
           {
-            params: getSearchParams(getAggsWithRandomSampling(1)),
+            params: getSearchParams(getEventRateAggsWithRandomSampling(1)),
           },
           searchOptions
         )
@@ -162,7 +182,7 @@ export const getDocumentCountStats = async (
       const secondResp = await search
         .search(
           {
-            params: getSearchParams(getAggsWithRandomSampling(closestProbability)),
+            params: getSearchParams(getEventRateAggsWithRandomSampling(closestProbability)),
           },
           searchOptions
         )

diff --git a/x-pack/plugins/translations/translations/fr-FR.json b/x-pack/plugins/translations/translations/fr-FR.json
@@ -9745,8 +9745,6 @@
     "xpack.dataVisualizer.dataGrid.field.metricDistributionChart.tooltipValueBetweenLabel": "{percent} % des documents ont des valeurs comprises entre {minValFormatted} et {maxValFormatted}",
     "xpack.dataVisualizer.dataGrid.field.metricDistributionChart.tooltipValueEqualLabel": "{percent} % des documents ont une valeur de {valFormatted}",
     "xpack.dataVisualizer.dataGrid.field.removeFilterAriaLabel": "Exclure le {fieldName} : \"{value}\"",
-    "xpack.dataVisualizer.dataGrid.field.topValues.calculatedFromSampleDescription": "Calculé à partir d'un échantillon de {topValuesSamplerShardSize} documents par partition",
-    "xpack.dataVisualizer.dataGrid.fieldExpandedRow.choroplethMapTopValues.calculatedFromSampleDescription": "Calculé à partir d'un échantillon de {topValuesSamplerShardSize} documents par partition",
     "xpack.dataVisualizer.dataGrid.fieldExpandedRow.numberContent.displayingPercentilesLabel": "Affichage de {minPercent} - {maxPercent} centiles",
     "xpack.dataVisualizer.dataGrid.fieldText.fieldMayBePopulatedDescription": "Il peut être rempli, par exemple, à l'aide d'un paramètre {copyToParam} dans le mapping du document ou être réduit à partir du champ {sourceParam} après une indexation par l'utilisation des paramètres {includesParam} et {excludesParam}.",
     "xpack.dataVisualizer.dataGrid.fieldText.fieldNotPresentDescription": "Ce champ n'était pas présent dans le champ {sourceParam} des documents interrogés.",
@@ -9784,7 +9782,6 @@
     "xpack.dataVisualizer.nameCollisionMsg": "\"{name}\" existe déjà, veuillez fournir un nom unique",
     "xpack.dataVisualizer.randomSamplerSettingsPopUp.probabilityLabel": "Probabilité utilisée : {samplingProbability} %",
     "xpack.dataVisualizer.searchPanel.ofFieldsTotal": "sur un total de {totalCount}",
-    "xpack.dataVisualizer.searchPanel.sampleSizeOptionLabel": "Taille de l'échantillon (par partition) : {wrappedValue}",
     "xpack.dataVisualizer.searchPanel.totalDocCountLabel": "Total des documents : {prepend}{strongTotalCount}",
     "xpack.dataVisualizer.searchPanel.totalDocCountNumber": "{totalCount, plural, other {#}}",
     "xpack.dataVisualizer.addCombinedFieldsLabel": "Ajouter un champ combiné",
@@ -10063,13 +10060,10 @@
     "xpack.dataVisualizer.removeCombinedFieldsLabel": "Retirer le champ combiné",
     "xpack.dataVisualizer.samplingOptionsButton": "Options d’échantillonnage",
     "xpack.dataVisualizer.searchPanel.allFieldsLabel": "Tous les champs",
-    "xpack.dataVisualizer.searchPanel.allOptionLabel": "Tout rechercher",
     "xpack.dataVisualizer.searchPanel.invalidSyntax": "Syntaxe non valide",
     "xpack.dataVisualizer.searchPanel.numberFieldsLabel": "Champs de numéros",
-    "xpack.dataVisualizer.searchPanel.queryBarPlaceholder": "La sélection d'une taille d'échantillon plus petite réduira les temps d'exécution de la requête et la charge sur le cluster.",
     "xpack.dataVisualizer.searchPanel.queryBarPlaceholderText": "Rechercher… (par exemple, status:200 AND extension:\"PHP\")",
     "xpack.dataVisualizer.searchPanel.randomSamplerMessage": "Des valeurs approximatives sont indiquées dans le décompte de documents et le graphique, qui utilisent des agrégations par échantillonnage aléatoire.",
-    "xpack.dataVisualizer.searchPanel.sampleSizeAriaLabel": "Sélectionner le nombre de documents à échantillonner",
     "xpack.dataVisualizer.searchPanel.showEmptyFields": "Afficher les champs vides",
     "xpack.dataVisualizer.title": "Charger un fichier",
     "xpack.embeddableEnhanced.actions.panelNotifications.manyDrilldowns": "Le panneau comporte {count} recherches",

diff --git a/x-pack/plugins/translations/translations/ja-JP.json b/x-pack/plugins/translations/translations/ja-JP.json
@@ -9734,8 +9734,6 @@
     "xpack.dataVisualizer.dataGrid.field.metricDistributionChart.tooltipValueBetweenLabel": "{percent}% のドキュメントに {minValFormatted} から {maxValFormatted} の間の値があります",
     "xpack.dataVisualizer.dataGrid.field.metricDistributionChart.tooltipValueEqualLabel": "{percent}% のドキュメントに {valFormatted} の値があります",
     "xpack.dataVisualizer.dataGrid.field.removeFilterAriaLabel": "{fieldName}の除外：\"{value}\"",
-    "xpack.dataVisualizer.dataGrid.field.topValues.calculatedFromSampleDescription": "1 つのシャードにつき {topValuesSamplerShardSize} のドキュメントのサンプルで計算されています",
-    "xpack.dataVisualizer.dataGrid.fieldExpandedRow.choroplethMapTopValues.calculatedFromSampleDescription": "1 つのシャードにつき {topValuesSamplerShardSize} のドキュメントのサンプルで計算されています",
     "xpack.dataVisualizer.dataGrid.fieldExpandedRow.numberContent.displayingPercentilesLabel": "{minPercent} - {maxPercent} パーセンタイルを表示中",
     "xpack.dataVisualizer.dataGrid.fieldText.fieldMayBePopulatedDescription": "たとえば、ドキュメントマッピングで {copyToParam} パラメーターを使ったり、{includesParam} と {excludesParam} パラメーターを使用してインデックスした後に {sourceParam} フィールドから切り取ったりして入力される場合があります。",
     "xpack.dataVisualizer.dataGrid.fieldText.fieldNotPresentDescription": "このフィールドはクエリが実行されたドキュメントの {sourceParam} フィールドにありませんでした。",
@@ -9772,7 +9770,6 @@
     "xpack.dataVisualizer.nameCollisionMsg": "「{name}」はすでに存在します。一意の名前を入力してください。",
     "xpack.dataVisualizer.randomSamplerSettingsPopUp.probabilityLabel": "使用された確率：{samplingProbability}%",
     "xpack.dataVisualizer.searchPanel.ofFieldsTotal": "合計 {totalCount}",
-    "xpack.dataVisualizer.searchPanel.sampleSizeOptionLabel": "サンプルサイズ（シャード単位）：{wrappedValue}",
     "xpack.dataVisualizer.searchPanel.totalDocCountLabel": "合計ドキュメント数：{prepend}{strongTotalCount}",
     "xpack.dataVisualizer.searchPanel.totalDocCountNumber": "{totalCount, plural, other  {#}}",
     "xpack.dataVisualizer.addCombinedFieldsLabel": "結合されたフィールドを追加",
@@ -10051,13 +10048,10 @@
     "xpack.dataVisualizer.removeCombinedFieldsLabel": "結合されたフィールドを削除",
     "xpack.dataVisualizer.samplingOptionsButton": "抽出オプション",
     "xpack.dataVisualizer.searchPanel.allFieldsLabel": "すべてのフィールド",
-    "xpack.dataVisualizer.searchPanel.allOptionLabel": "すべて検索",
     "xpack.dataVisualizer.searchPanel.invalidSyntax": "無効な構文",
     "xpack.dataVisualizer.searchPanel.numberFieldsLabel": "数値フィールド",
-    "xpack.dataVisualizer.searchPanel.queryBarPlaceholder": "小さいサンプルサイズを選択することで、クエリの実行時間を短縮しクラスターへの負荷を軽減できます。",
     "xpack.dataVisualizer.searchPanel.queryBarPlaceholderText": "検索…（例：status:200 AND extension:\"PHP\"）",
     "xpack.dataVisualizer.searchPanel.randomSamplerMessage": "近似値は、ランダムサンプラーアグリゲーションを使用する、合計ドキュメント数およびグラフに表示されます。",
-    "xpack.dataVisualizer.searchPanel.sampleSizeAriaLabel": "サンプリングするドキュメント数を選択してください",
     "xpack.dataVisualizer.searchPanel.showEmptyFields": "空のフィールドを表示",
     "xpack.dataVisualizer.title": "ファイルをアップロード",
     "xpack.embeddableEnhanced.actions.panelNotifications.manyDrilldowns": "パネルには{count}個のドリルダウンがあります",

diff --git a/x-pack/plugins/translations/translations/zh-CN.json b/x-pack/plugins/translations/translations/zh-CN.json
@@ -9749,8 +9749,6 @@
     "xpack.dataVisualizer.dataGrid.field.metricDistributionChart.tooltipValueBetweenLabel": "{percent}% 的文档具有介于 {minValFormatted} 和 {maxValFormatted} 之间的值",
     "xpack.dataVisualizer.dataGrid.field.metricDistributionChart.tooltipValueEqualLabel": "{percent}% 的文档的值为 {valFormatted}",
     "xpack.dataVisualizer.dataGrid.field.removeFilterAriaLabel": "筛除 {fieldName}：“{value}”",
-    "xpack.dataVisualizer.dataGrid.field.topValues.calculatedFromSampleDescription": "基于每个分片的 {topValuesSamplerShardSize} 文档样例计算",
-    "xpack.dataVisualizer.dataGrid.fieldExpandedRow.choroplethMapTopValues.calculatedFromSampleDescription": "基于每个分片的 {topValuesSamplerShardSize} 文档样例计算",
     "xpack.dataVisualizer.dataGrid.fieldExpandedRow.numberContent.displayingPercentilesLabel": "正在显示 {minPercent} - {maxPercent} 百分位数",
     "xpack.dataVisualizer.dataGrid.fieldText.fieldMayBePopulatedDescription": "例如，可以使用文档映射中的 {copyToParam} 参数进行填充，也可以在索引后通过使用 {includesParam} 和 {excludesParam} 参数从 {sourceParam} 字段中修剪。",
     "xpack.dataVisualizer.dataGrid.fieldText.fieldNotPresentDescription": "查询的文档的 {sourceParam} 字段中不存在此字段。",
@@ -9788,7 +9786,6 @@
     "xpack.dataVisualizer.nameCollisionMsg": "“{name}”已存在，请提供唯一名称",
     "xpack.dataVisualizer.randomSamplerSettingsPopUp.probabilityLabel": "使用的概率：{samplingProbability}%",
     "xpack.dataVisualizer.searchPanel.ofFieldsTotal": "，共 {totalCount} 个",
-    "xpack.dataVisualizer.searchPanel.sampleSizeOptionLabel": "样本大小（每分片）：{wrappedValue}",
     "xpack.dataVisualizer.searchPanel.totalDocCountLabel": "文档总数：{prepend}{strongTotalCount}",
     "xpack.dataVisualizer.searchPanel.totalDocCountNumber": "{totalCount, plural, other {#}}",
     "xpack.dataVisualizer.addCombinedFieldsLabel": "添加组合字段",
@@ -10067,13 +10064,10 @@
     "xpack.dataVisualizer.removeCombinedFieldsLabel": "移除组合字段",
     "xpack.dataVisualizer.samplingOptionsButton": "采样选项",
     "xpack.dataVisualizer.searchPanel.allFieldsLabel": "所有字段",
-    "xpack.dataVisualizer.searchPanel.allOptionLabel": "搜索全部",
     "xpack.dataVisualizer.searchPanel.invalidSyntax": "语法无效",
     "xpack.dataVisualizer.searchPanel.numberFieldsLabel": "字段数目",
-    "xpack.dataVisualizer.searchPanel.queryBarPlaceholder": "选择较小的样例大小将减少查询运行时间和集群上的负载。",
     "xpack.dataVisualizer.searchPanel.queryBarPlaceholderText": "搜索……（例如，status:200 AND extension:\"PHP\"）",
     "xpack.dataVisualizer.searchPanel.randomSamplerMessage": "总文档计数和图表中将显示近似值，它们使用随机采样器聚合。",
-    "xpack.dataVisualizer.searchPanel.sampleSizeAriaLabel": "选择要采样的文档数目",
     "xpack.dataVisualizer.searchPanel.showEmptyFields": "显示空字段",
     "xpack.dataVisualizer.title": "上传文件",
     "xpack.embeddableEnhanced.actions.panelNotifications.manyDrilldowns": "面板有 {count} 个向下钻取",