group reduce: "distinct" (observablehq#427)

Fil · web-flow · commit a82e2fe3a6c1 · 2021-07-10T09:04:59.000+02:00
* group reduce: "distinct" closes observablehq#412 * missing test * document "distinct" reduce
diff --git a/README.md b/README.md
@@ -973,6 +973,7 @@ The following aggregation methods are supported:
 * *first* - the first value, in input order
 * *last* - the last value, in input order
 * *count* - the number of elements (frequency)
+* *distinct* - the number of distinct values
 * *sum* - the sum of values
 * *proportion* - the sum proportional to the overall total (weighted frequency)
 * *proportion-facet* - the sum proportional to the facet total
diff --git a/src/transforms/group.js b/src/transforms/group.js
@@ -1,4 +1,4 @@
-import {group as grouper, sort, sum, deviation, min, max, mean, median, variance} from "d3";
+import {group as grouper, sort, sum, deviation, min, max, mean, median, variance, InternSet} from "d3";
 import {firstof} from "../defined.js";
 import {valueof, maybeColor, maybeInput, maybeTransform, maybeTuple, maybeLazyChannel, lazyChannel, first, identity, take, labelof, range} from "../mark.js";
 
@@ -139,6 +139,7 @@ export function maybeReduce(reduce, value) {
     case "first": return reduceFirst;
     case "last": return reduceLast;
     case "count": return reduceCount;
+    case "distinct": return reduceDistinct;
     case "sum": return value == null ? reduceCount : reduceSum;
     case "proportion": return reduceProportion(value, "data");
     case "proportion-facet": return reduceProportion(value, "facet");
@@ -201,6 +202,15 @@ const reduceCount = {
   }
 };
 
+const reduceDistinct = {
+  label: "Distinct",
+  reduce: (I, X) => {
+    const s = new InternSet();
+    for (const i of I) s.add(X[i]);
+    return s.size;
+  }
+};
+
 const reduceSum = reduceAccessor(sum);
 
 function reduceProportion(value, scope) {
diff --git a/test/output/mobyDickLetterPairs.svg b/test/output/mobyDickLetterPairs.svg
diff --git a/test/plots/index.js b/test/plots/index.js
@@ -56,6 +56,7 @@ export {default as metroUnemploymentNormalize} from "./metro-unemployment-normal
 export {default as metroUnemploymentRidgeline} from "./metro-unemployment-ridgeline.js";
 export {default as mobyDickFaceted} from "./moby-dick-faceted.js";
 export {default as mobyDickLetterFrequency} from "./moby-dick-letter-frequency.js";
+export {default as mobyDickLetterPairs} from "./moby-dick-letter-pairs.js";
 export {default as mobyDickLetterPosition} from "./moby-dick-letter-position.js";
 export {default as mobyDickLetterRelativeFrequency} from "./moby-dick-letter-relative-frequency.js";
 export {default as morleyBoxplot} from "./morley-boxplot.js";
diff --git a/test/plots/moby-dick-letter-pairs.js b/test/plots/moby-dick-letter-pairs.js
@@ -0,0 +1,28 @@
+import * as Plot from "@observablehq/plot";
+import * as d3 from "d3";
+
+export default async function() {
+  const mobydick = await d3.text("data/moby-dick-chapter-1.txt");
+  const letters = [...mobydick].map(d => /\w/.test(d) ? d.toUpperCase() : "*");
+  const pairs = d3.pairs(letters).map(([letter, next]) => ({letter, next}));
+  return Plot.plot({
+    x: { axis: null },
+    y: { label: null },
+    marks: [
+      Plot.barX(pairs, Plot.groupY({x: "distinct"}, {x: "next", y: "letter"})),
+      Plot.text(pairs, Plot.stackX(
+        Plot.group({
+          text: "first",
+          y: "first",
+          x: "distinct"
+        }, {
+          y: "letter",
+          z: "next",
+          x: "next",
+          text: "next",
+          fill: "white"
+        })
+      ))
+    ]
+  });
+}
