Merge pull request #260 from VisActor/feat/bin-count

feat: support `countField` option of `bin` and export `percentage`

Author: xile611

common/changes/@visactor/vdataset/feat-bin-count_2025-10-30-08-17.json

@@ -0,0 +1,11 @@
+{
+  "changes": [
+    {
+      "comment": "feat: support `countField` option of `bin` and export `percentage`\n\n",
+      "type": "none",
+      "packageName": "@visactor/vdataset"
+    }
+  ],
+  "packageName": "@visactor/vdataset",
+  "email": "[email protected]"
+}

packages/vdataset/__tests__/bin.test.ts

@@ -80,4 +80,40 @@ describe('bin transform', () => {
     const total = out.reduce((s: number, b: any) => s + b.cnt, 0);
     expect(total).toBe(3);
   });
+
+  test('countField is used as weights and percentage calculated correctly', () => {
+    const data = [
+      { v: 1, w: 2 }, // goes to first bin
+      { v: 2, w: 3 }, // goes to first bin
+      { v: 8, w: 5 } // goes to second bin
+    ];
+    // thresholds split at 5 -> two bins [0,5) and [5,10]
+    const out: any = bin(data, { field: 'v', thresholds: [0, 5, 10], countField: 'w' });
+    expect(out.length).toBe(2);
+    // first bin should have count 5 (2+3), second bin 5
+    expect(out[0].count).toBe(5);
+    expect(out[1].count).toBe(5);
+    // percentage should be 0.5 for both
+    expect(out[0].percentage).toBeCloseTo(0.5, 12);
+    expect(out[1].percentage).toBeCloseTo(0.5, 12);
+  });
+
+  test('renamed percentage field via outputNames is present and correct', () => {
+    const data = [
+      { v: 1, w: 1 },
+      { v: 2, w: 1 },
+      { v: 9, w: 2 }
+    ];
+    const out: any = bin(data, {
+      field: 'v',
+      thresholds: [0, 5, 10],
+      countField: 'w',
+      outputNames: { percentage: 'pct' }
+    });
+    expect(out.length).toBe(2);
+    // counts: first bin 2, second bin 2 -> percentages 0.5 each
+    expect(out[0].cnt === undefined).toBeTruthy(); // ensure default countName not renamed here
+    expect(out[0].pct).toBeCloseTo(0.5, 12);
+    expect(out[1].pct).toBeCloseTo(0.5, 12);
+  });
 });

packages/vdataset/src/transform/bin.ts

@@ -6,6 +6,10 @@ export interface IBinOptions {
    * numeric field to bin
    */
   field: string;
+  /**
+   * count of numeric field
+   */
+  countField?: string;
   /**
    *  number of bins (default 10)
    */
@@ -29,7 +33,7 @@ export interface IBinOptions {
   /**
    * the field name of output data
    */
-  outputNames?: { x0?: string; x1?: string; count?: string; values?: string };
+  outputNames?: { x0?: string; x1?: string; count?: string; values?: string; percentage?: string };
 }
 
 /**
@@ -41,13 +45,13 @@ export const bin: Transform = (data: Array<object>, options?: IBinOptions) => {
   if (!field) {
     return [];
   }
-
+  const countField = options.countField;
   const n = data.length;
   // compute data-driven extent
   let min = Infinity;
   let max = -Infinity;
 
-  if (options?.extent) {
+  if (options.extent) {
     min = options.extent[0];
     max = options.extent[1];
   } else {
@@ -74,11 +78,11 @@ export const bin: Transform = (data: Array<object>, options?: IBinOptions) => {
 
   // build thresholds
   let thresholds: number[] | undefined;
-  if (options && options.thresholds && options.thresholds.length) {
+  if (options.thresholds && options.thresholds.length) {
     // explicit thresholds provided by user
     thresholds = options.thresholds.slice();
     thresholds.sort((a, b) => a - b);
-  } else if (options && typeof options.step === 'number' && options.step > 0) {
+  } else if (typeof options.step === 'number' && options.step > 0) {
     // fixed bin width (step) provided: compute number of bins to cover [min, max]
     const stepSize = options.step;
     let startMin = min;
@@ -94,7 +98,7 @@ export const bin: Transform = (data: Array<object>, options?: IBinOptions) => {
     }
   } else {
     // fallback to bins count (default 10)
-    const bins = options?.bins && options.bins > 0 ? Math.floor(options.bins) : 10;
+    const bins = options.bins && options.bins > 0 ? Math.floor(options.bins) : 10;
     const stepSize = (max - min) / bins;
     thresholds = new Array(bins + 1);
     for (let i = 0; i <= bins; i++) {
@@ -111,14 +115,16 @@ export const bin: Transform = (data: Array<object>, options?: IBinOptions) => {
   const x1Name = options.outputNames?.x1 ?? 'x1';
   const countName = options.outputNames?.count ?? 'count';
   const valuesName = options.outputNames?.values ?? 'values';
+  const percentageName = options.outputNames?.percentage ?? 'percentage';
   const out: any[] = new Array(numBins);
   for (let i = 0; i < numBins; i++) {
     out[i] = { [x0Name]: thresholds[i], [x1Name]: thresholds[i + 1], [countName]: 0 };
-    if (options?.includeValues) {
+    if (options.includeValues) {
       out[i][valuesName] = [] as object[];
     }
   }
 
+  let totalCount = 0;
   // assign each datum to a bin (left-inclusive, right-exclusive except last bin includes max)
   for (let i = 0; i < n; i++) {
     const v: any = (data[i] as any)[field];
@@ -136,7 +142,9 @@ export const bin: Transform = (data: Array<object>, options?: IBinOptions) => {
       const right = out[j][x1Name];
       const isLast = j === numBins - 1;
       if ((num >= left && num < right) || (isLast && num <= right)) {
-        out[j][countName]++;
+        const count = (data[i] as any)[countField] ?? 1;
+        out[j][countName] += count;
+        totalCount += count;
         if (options && options.includeValues) {
           out[j][valuesName].push(data[i]);
         }
@@ -145,6 +153,10 @@ export const bin: Transform = (data: Array<object>, options?: IBinOptions) => {
     }
   }
 
+  for (let i = 0, len = out.length; i < len; i++) {
+    out[i][percentageName] = totalCount > 0 ? out[i][countName] / totalCount : 0;
+  }
+
   return out;
 };