sadpandajoe commented on code in PR #44550:
URL: https://github.com/apache/superset/pull/44550#discussion_r4319828225
##########
superset-frontend/plugins/plugin-chart-echarts/test/Timeseries/transformProps.test.ts:
##########
@@ -4303,6 +4303,59 @@ test('tooltip does not apply a metric currency format to
a grouped Percentage ti
expect(result).not.toContain('$ 0.25');
});
+test('tooltip does not apply a metric currency format to a Percentage
comparison labeled `<metric>, <offset>`', () => {
+ // A hidden sort-only metric makes renameOperator keep the metric prefix on
the
+ // comparison column, so the label_map entry leads with `<metric>, <offset>`.
+ const chartProps = createTestChartProps({
+ formData: {
+ metric: 'revenue',
+ metrics: ['revenue'],
+ groupby: ['region'],
+ richTooltip: true,
+ time_compare: ['1 year ago'],
Review Comment:
This fixture sets only `time_compare`, but the new derived-series branch in
`transformProps` reads the camelCase `timeCompare` from `formData`, and the
shared helper hands back the raw form data without normalizing it. So
`['revenue, 1 year ago', 'East']` is never registered as derived and the
assertions for `25.00%` / no `$ 0.25` should fail here. Could the fixture set
`timeCompare` (as the neighboring time-comparison tooltip tests do), and could
you confirm the test goes red when the new `endsWith` branch is removed?
##########
superset-frontend/plugins/plugin-chart-echarts/test/Timeseries/xAxisSortWithDimensions.test.ts:
##########
@@ -0,0 +1,774 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied. See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+import {
+ ChartDataResponseResult,
+ DataRecord,
+ QueryFormMetric,
+} from '@superset-ui/core';
+import { SortSeriesType } from '@superset-ui/chart-controls';
+import transformProps from '../../src/Timeseries/transformProps';
+import { DEFAULT_FORM_DATA } from '../../src/Timeseries/constants';
+import {
+ EchartsTimeseriesChartProps,
+ EchartsTimeseriesFormData,
+ EchartsTimeseriesSeriesType,
+} from '../../src/Timeseries/types';
+import { createEchartsTimeseriesTestChartProps } from '../helpers';
+
+// A bar chart of COUNT(*) per genre, split by platform, with the axis sorted
+// by a metric that is neither the value metric nor a dimension (#34352).
+// The rows are what the backend returns for that query: the value metric's
+// columns lose their metric label (`truncate_metric`), the sort-only metric
+// keeps it, and `label_map` records the structure behind each column.
+const naSales: QueryFormMetric = {
+ expressionType: 'SIMPLE',
+ aggregate: 'SUM',
+ column: { column_name: 'na_sales' },
+ label: 'SUM(na_sales)',
+};
+
+const rows: DataRecord[] = [
+ {
+ genre: 'Action',
+ PS4: 30,
+ XOne: 20,
+ 'SUM(na_sales), PS4': 5,
+ 'SUM(na_sales), XOne': 3,
+ },
+ {
+ genre: 'Puzzle',
+ PS4: 5,
+ XOne: 5,
+ 'SUM(na_sales), PS4': 20,
+ 'SUM(na_sales), XOne': 10,
+ },
+ {
+ genre: 'Sports',
+ PS4: 40,
+ XOne: 10,
+ 'SUM(na_sales), PS4': 1,
+ 'SUM(na_sales), XOne': 1,
+ },
+];
+
+const labelMap = {
+ genre: ['genre'],
+ PS4: ['PS4'],
+ XOne: ['XOne'],
+ 'SUM(na_sales), PS4': ['SUM(na_sales)', 'PS4'],
+ 'SUM(na_sales), XOne': ['SUM(na_sales)', 'XOne'],
+};
+
+const queriesData = [
+ {
+ annotation_data: null,
+ cache_key: null,
+ cache_timeout: null,
+ cached_dttm: null,
+ queried_dttm: null,
+ data: rows,
+ colnames: [],
+ coltypes: [],
+ error: null,
+ is_cached: false,
+ query: '',
+ rowcount: rows.length,
+ sql_rowcount: rows.length,
+ stacktrace: null,
+ status: 'success',
+ from_dttm: null,
+ to_dttm: null,
+ label_map: labelMap,
+ } as unknown as ChartDataResponseResult,
+];
+
+type Series = {
+ name: string;
+ data: [string, number][];
+ label: {
+ formatter: (params: {
+ value: [string, number];
+ dataIndex: number;
+ seriesIndex: number;
+ }) => string;
+ };
+};
+
+// transformProps reads camelCase keys from `formData` and snake_case keys
+// from `rawFormData`; the test helper hands the same object to both.
+function transform(
+ overrides: Record<string, unknown>,
+ verboseMap = {},
+ data = rows,
+): {
+ series: Series[];
+ seriesNames: string[];
+ axisOrder: string[];
+ stackedTotals: string[];
+} {
+ const chartProps = createEchartsTimeseriesTestChartProps<
+ EchartsTimeseriesFormData,
+ EchartsTimeseriesChartProps
+ >({
+ defaultFormData: DEFAULT_FORM_DATA,
+ defaultVizType: 'echarts_timeseries_bar',
+ formData: {
+ colorScheme: 'bnbColors',
+ seriesType: EchartsTimeseriesSeriesType.Bar,
+ x_axis: 'genre',
+ xAxis: 'genre',
+ metrics: ['count'],
+ groupby: ['platform'],
+ timeseries_limit_metric: naSales,
+ truncate_metric: true,
+ stack: true,
+ showValue: true,
+ onlyTotal: true,
+ ...overrides,
+ } as Partial<EchartsTimeseriesFormData>,
+ queriesData:
+ data === rows
+ ? queriesData
+ : [{ ...queriesData[0], data } as ChartDataResponseResult],
+ datasource: { verboseMap },
+ });
+ const transformed = transformProps(chartProps);
+ const series = transformed.echartOptions.series as Series[];
+ return {
+ series,
+ // Legend order is a separate control; only the set of series matters here.
+ seriesNames: series.map(({ name }) => name).sort(),
+ axisOrder: series[0].data.map(([genre]) => genre),
+ // With `onlyTotal` the top series' label carries the stacked total.
+ stackedTotals: series
+ .flatMap((entry, seriesIndex) =>
+ entry.data.map((value, dataIndex) =>
+ entry.label.formatter({ value, dataIndex, seriesIndex }),
+ ),
+ )
+ .filter(label => label !== ''),
+ };
+}
+
+test('sorts the axis by the sort-only metric across its pivoted columns', ()
=> {
+ const { axisOrder, seriesNames, stackedTotals } = transform({
+ x_axis_sort: 'SUM(na_sales)',
+ xAxisSort: 'SUM(na_sales)',
+ x_axis_sort_asc: false,
+ xAxisSortAsc: false,
+ });
+ // SUM(na_sales) totals: Puzzle 30, Action 8, Sports 2. Sorting by the value
+ // metric would put Action/Sports first, sorting by name would put Puzzle
+ // second.
+ expect(axisOrder).toEqual(['Puzzle', 'Action', 'Sports']);
+ expect(seriesNames).toEqual(['PS4', 'XOne']);
+ // COUNT(*) totals only, in the sorted order.
+ expect(stackedTotals).toEqual(['10', '50', '50']);
+});
+
+test('sorts the axis by the sort-only metric ascending', () => {
+ const { axisOrder } = transform({
+ x_axis_sort: 'SUM(na_sales)',
+ xAxisSort: 'SUM(na_sales)',
+ x_axis_sort_asc: true,
+ xAxisSortAsc: true,
+ });
+ expect(axisOrder).toEqual(['Sports', 'Action', 'Puzzle']);
+});
+
+test('sorts by the sort-only metric when it has a verbose name', () => {
+ const { axisOrder, seriesNames } = transform(
+ {
+ x_axis_sort: 'SUM(na_sales)',
+ xAxisSort: 'SUM(na_sales)',
+ x_axis_sort_asc: false,
+ xAxisSortAsc: false,
+ },
+ { 'SUM(na_sales)': 'NA Sales' },
+ );
+ expect(axisOrder).toEqual(['Puzzle', 'Action', 'Sports']);
+ expect(seriesNames).toEqual(['PS4', 'XOne']);
+});
+
+test('sorts the axis by the x-axis column with dimensions set', () => {
Review Comment:
The case-sensitive comparator is covered at the `sortRows` boundary, but
nothing at the `transformProps` level would catch the column resolver going
back to `SortSeriesType.Name`: the grouped fixtures only use `Action`,
`Puzzle`, and `Sports`, which order the same either way. Could a grouped case
with axis values `a` and `B`, sorted ascending by the actual x-axis column,
assert `['B', 'a']`?
##########
superset-frontend/plugins/plugin-chart-echarts/src/Timeseries/transformProps.ts:
##########
@@ -439,6 +442,15 @@ export default function transformProps(
) {
derivedComparisonSeries.add(entry[0]);
entry[1].shift();
+ } else if (
+ entry[1].length > groupBy.length &&
+ Array.isArray(timeCompare) &&
+ timeCompare.some(offset => entry[1][0].endsWith(`, ${offset}`))
Review Comment:
This suffix match also catches a hidden sort metric whose own label ends
with the offset. With a displayed `count`, dimension `region`, an actual-values
`1 year ago` comparison, and a Sort By metric labeled `sales, 1 year ago`, its
base entry `["sales, 1 year ago", "East"]` is recorded as derived.
`pivotedColumnsOf` then drops it, so sorting descending by it stays in query
order (A, B for sales 1/10 instead of B, A) and the hidden sales series is
drawn and added to the stacked totals. Could this check use the metric/offset
pairs `renameOperator` actually generates, rather than any trailing `,
<offset>`, with a regression for a metric labeled like that?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
