37151213cd
The existing architecture serializes all cards and revlog entries in the search range into a protobuf message, which the web frontend needs to decode and then process. The thinking at the time was that this would make it easier for add-ons to add extra graphs, but in the ~2.5 years since the new graphs were introduced, no add-ons appear to have taken advantage of it. The cards and revlog entries can grow quite large on large collections - on a collection I tested with approximately 2.5M reviews, the serialized data is about 110MB, which is a lot to have to deserialize in JavaScript. This commit shifts the preliminary processing of the data to the Rust end, which means the data is able to be processed faster, and less needs to be sent to the frontend. On the test collection above, this reduces the serialized data from about 110MB to about 160KB, resulting in a more than 2x performance improvement, and reducing frontend memory usage from about 400MB to about 40MB. This also makes #2043 more feasible - while it is still about 50-100% slower than protobufjs, with the much smaller message size, the difference is only about 10ms.
181 lines
5.3 KiB
TypeScript
181 lines
5.3 KiB
TypeScript
// Copyright: Ankitects Pty Ltd and contributors
|
|
// License: GNU AGPL, version 3 or later; http://www.gnu.org/licenses/agpl.html
|
|
|
|
/* eslint
|
|
@typescript-eslint/no-explicit-any: "off",
|
|
*/
|
|
|
|
import * as tr from "@tslib/ftl";
|
|
import { localizedNumber } from "@tslib/i18n";
|
|
import type { Stats } from "@tslib/proto";
|
|
import { timeSpan } from "@tslib/time";
|
|
import type { Bin } from "d3";
|
|
import { bin, extent, interpolateBlues, mean, quantile, scaleLinear, scaleSequential, sum } from "d3";
|
|
|
|
import type { SearchDispatch, TableDatum } from "./graph-helpers";
|
|
import { numericMap } from "./graph-helpers";
|
|
import type { HistogramData } from "./histogram-graph";
|
|
|
|
export interface IntervalGraphData {
|
|
intervals: number[];
|
|
}
|
|
|
|
export enum IntervalRange {
|
|
Month = 0,
|
|
Percentile50 = 1,
|
|
Percentile95 = 2,
|
|
All = 3,
|
|
}
|
|
|
|
export function gatherIntervalData(data: Stats.GraphsResponse): IntervalGraphData {
|
|
// This could be made more efficient - this graph currently expects a flat list of individual intervals which it
|
|
// uses to calculate a percentile and then converts into a histogram, and the percentile/histogram calculations
|
|
// in JS are relatively slow.
|
|
const map = numericMap(data.intervals!.intervals);
|
|
const totalCards = sum(map, ([_k, v]) => v);
|
|
const allIntervals: number[] = Array(totalCards);
|
|
let position = 0;
|
|
for (const entry of map.entries()) {
|
|
allIntervals.fill(entry[0], position, position + entry[1]);
|
|
position += entry[1];
|
|
}
|
|
allIntervals.sort((a, b) => a - b);
|
|
return { intervals: allIntervals };
|
|
}
|
|
|
|
export function intervalLabel(
|
|
daysStart: number,
|
|
daysEnd: number,
|
|
cards: number,
|
|
): string {
|
|
if (daysEnd - daysStart <= 1) {
|
|
// singular
|
|
return tr.statisticsIntervalsDaySingle({
|
|
day: daysStart,
|
|
cards,
|
|
});
|
|
} else {
|
|
// range
|
|
return tr.statisticsIntervalsDayRange({
|
|
daysStart,
|
|
daysEnd: daysEnd - 1,
|
|
cards,
|
|
});
|
|
}
|
|
}
|
|
|
|
function makeQuery(start: number, end: number): string {
|
|
if (start === end) {
|
|
return `"prop:ivl=${start}"`;
|
|
}
|
|
|
|
const fromQuery = `"prop:ivl>=${start}"`;
|
|
const tillQuery = `"prop:ivl<=${end}"`;
|
|
|
|
return `${fromQuery} AND ${tillQuery}`;
|
|
}
|
|
|
|
export function prepareIntervalData(
|
|
data: IntervalGraphData,
|
|
range: IntervalRange,
|
|
dispatch: SearchDispatch,
|
|
browserLinksSupported: boolean,
|
|
): [HistogramData | null, TableDatum[]] {
|
|
// get min/max
|
|
const allIntervals = data.intervals;
|
|
if (!allIntervals.length) {
|
|
return [null, []];
|
|
}
|
|
|
|
const xMin = 0;
|
|
let [, xMax] = extent(allIntervals);
|
|
let niceNecessary = false;
|
|
|
|
// cap max to selected range
|
|
switch (range) {
|
|
case IntervalRange.Month:
|
|
xMax = Math.min(xMax!, 30);
|
|
break;
|
|
case IntervalRange.Percentile50:
|
|
xMax = quantile(allIntervals, 0.5);
|
|
niceNecessary = true;
|
|
break;
|
|
case IntervalRange.Percentile95:
|
|
xMax = quantile(allIntervals, 0.95);
|
|
niceNecessary = true;
|
|
break;
|
|
case IntervalRange.All:
|
|
niceNecessary = true;
|
|
break;
|
|
}
|
|
|
|
xMax = xMax! + 1;
|
|
|
|
// do not show the zero interval
|
|
const increment = (x: number): number => x + 1;
|
|
|
|
const adjustTicks = (x: number, idx: number, ticks: number[]): number[] =>
|
|
idx === ticks.length - 1 ? [x - (ticks[0] - 1), x + 1] : [x - (ticks[0] - 1)];
|
|
|
|
// cap bars to available range
|
|
const desiredBars = Math.min(70, xMax! - xMin!);
|
|
|
|
const prescale = scaleLinear().domain([xMin!, xMax!]);
|
|
const scale = scaleLinear().domain(
|
|
(niceNecessary ? prescale.nice() : prescale).domain().map(increment),
|
|
);
|
|
|
|
const bins = bin()
|
|
.domain(scale.domain() as [number, number])
|
|
.thresholds(scale.ticks(desiredBars).flatMap(adjustTicks))(allIntervals);
|
|
|
|
// empty graph?
|
|
const totalInPeriod = sum(bins, (bin) => bin.length);
|
|
if (!totalInPeriod) {
|
|
return [null, []];
|
|
}
|
|
|
|
const adjustedRange = scaleLinear().range([0.7, 0.3]);
|
|
const colourScale = scaleSequential((n) => interpolateBlues(adjustedRange(n)!)).domain([xMax!, xMin!]);
|
|
|
|
function hoverText(
|
|
bin: Bin<number, number>,
|
|
_cumulative: number,
|
|
percent: number,
|
|
): string {
|
|
// const day = dayLabel(bin.x0!, bin.x1!);
|
|
const interval = intervalLabel(bin.x0!, bin.x1!, bin.length);
|
|
const total = tr.statisticsRunningTotal();
|
|
return `${interval}<br>${total}: \u200e${localizedNumber(percent, 1)}%`;
|
|
}
|
|
|
|
function onClick(bin: Bin<number, number>): void {
|
|
const start = bin.x0!;
|
|
const end = bin.x1! - 1;
|
|
const query = makeQuery(start, end);
|
|
dispatch("search", { query });
|
|
}
|
|
|
|
const meanInterval = Math.round(mean(allIntervals) ?? 0);
|
|
const meanIntervalString = timeSpan(meanInterval * 86400, false);
|
|
const tableData = [
|
|
{
|
|
label: tr.statisticsAverageInterval(),
|
|
value: meanIntervalString,
|
|
},
|
|
];
|
|
|
|
return [
|
|
{
|
|
scale,
|
|
bins,
|
|
total: totalInPeriod,
|
|
hoverText,
|
|
onClick: browserLinksSupported ? onClick : null,
|
|
colourScale,
|
|
showArea: true,
|
|
},
|
|
tableData,
|
|
];
|
|
}
|