├── .prettierrc ├── src ├── index.ts ├── tools.ts ├── regression-plugin.ts ├── MetaData.ts ├── types.ts └── MetaSection.ts ├── tslint.json ├── tsconfig.json ├── bin └── setversion.js ├── .gitignore ├── LICENSE ├── webpack.config.js ├── demo ├── demo.css ├── index.html ├── demoHelpers.js ├── demo.js └── demoData.js ├── package.json ├── README.md └── dist ├── chartjs-plugin-regression-0.2.1.js ├── chartjs-plugin-regression-0.1.1.js └── chartjs-plugin-regression-0.2.0.js /.prettierrc: -------------------------------------------------------------------------------- 1 | { 2 | "printWidth": 80, 3 | "trailingComma": "none", 4 | "singleQuote": true, 5 | "arrowParens": "avoid" 6 | } -------------------------------------------------------------------------------- /src/index.ts: -------------------------------------------------------------------------------- 1 | export * from './types'; 2 | export * from './MetaData'; 3 | export * from './MetaSection'; 4 | export * from './regression-plugin'; 5 | -------------------------------------------------------------------------------- /tslint.json: -------------------------------------------------------------------------------- 1 | { 2 | "extends": ["tslint:recommended", "tslint-config-prettier"], 3 | "rules": { 4 | "triple-equals": false, 5 | "one-variable-per-declaration": false, 6 | "no-bitwise": false 7 | } 8 | } -------------------------------------------------------------------------------- /tsconfig.json: -------------------------------------------------------------------------------- 1 | { 2 | "compilerOptions": { 3 | "target": "es5", 4 | "lib": ["ES2016", "DOM"], 5 | "module": "commonjs", 6 | "declaration": true, 7 | "outDir": "./lib", 8 | "strict": true, 9 | "sourceMap": true 10 | }, 11 | "include": ["src"], 12 | "exclude": ["node_modules", "**/__tests__/*"] 13 | } -------------------------------------------------------------------------------- /bin/setversion.js: -------------------------------------------------------------------------------- 1 | #!/bin/env node 2 | 3 | const 4 | pck = require('../package.json'), 5 | fs = require('fs'), 6 | files = ['./demo/index.html', './README.md']; 7 | 8 | files.forEach(replaceVersion); 9 | 10 | function replaceVersion(file) { 11 | var index = fs.readFileSync(file, 'utf8'); 12 | index = index.replace(/(dist\/chartjs-plugin-regression-).*?\.js/, `$1${pck.version}.js`); 13 | fs.writeFileSync(file, index, 'utf8'); 14 | } -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- 1 | # Logs 2 | logs 3 | *.log 4 | npm-debug.log* 5 | 6 | # Maps 7 | *.map 8 | 9 | # Diagnostic reports (https://nodejs.org/api/report.html) 10 | report.[0-9]*.[0-9]*.[0-9]*.[0-9]*.json 11 | 12 | # Runtime data 13 | pids 14 | *.pid 15 | *.seed 16 | *.pid.lock 17 | 18 | # Directory for instrumented libs generated by jscoverage/JSCover 19 | lib-cov 20 | 21 | # Coverage directory used by tools like istanbul 22 | coverage 23 | *.lcov 24 | 25 | # Dependency directories 26 | node_modules/ 27 | 28 | # TypeScript v1 declaration files 29 | typings/ 30 | 31 | # TypeScript cache 32 | *.tsbuildinfo 33 | 34 | # Optional npm cache directory 35 | .npm 36 | 37 | # Optional eslint cache 38 | .eslintcache 39 | 40 | # Output of 'npm pack' 41 | *.tgz 42 | 43 | # build / generate output 44 | /lib 45 | -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | ISC License 2 | 3 | Copyright (c) 2020, Pomgui Informatica Ltda. 4 | 5 | Permission to use, copy, modify, and/or distribute this software for any 6 | purpose with or without fee is hereby granted, provided that the above 7 | copyright notice and this permission notice appear in all copies. 8 | 9 | THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES 10 | WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF 11 | MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR 12 | ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES 13 | WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN 14 | ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF 15 | OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. -------------------------------------------------------------------------------- /webpack.config.js: -------------------------------------------------------------------------------- 1 | const path = require('path'); 2 | const pck = require('./package.json'); 3 | 4 | module.exports = (env, argv) => { 5 | const config = { 6 | entry: './lib/index.js', 7 | mode: 'development', 8 | devtool: 'source-map', 9 | output: { 10 | path: path.resolve(__dirname, 'dist'), 11 | filename: 'chartjs-plugin-regression-' + pck.version + '.js' 12 | } 13 | }; 14 | 15 | if (argv.mode == 'production') { 16 | Object.assign(config, { 17 | mode: 'production', 18 | devtool: undefined, 19 | module: { 20 | rules: [ 21 | { 22 | test: /\.js$/, 23 | exclude: /node_modules/, 24 | use: ['source-map-loader'], 25 | enforce: 'pre' 26 | } 27 | ] 28 | } 29 | }); 30 | } 31 | 32 | return config; 33 | }; 34 | -------------------------------------------------------------------------------- /src/tools.ts: -------------------------------------------------------------------------------- 1 | /** 2 | * Makes a deep copy of the object including a copy of all the children and so on. 3 | * @param source source object. It can be any type 4 | * @param fields list of root fields that should be copied. If not sent, it will copy all the fields. 5 | * Note: the fields only filters the children of the source not the grandchildren. 6 | */ 7 | export function deepCopy(source: T, fields?: string[]): T { 8 | let target: T = source; 9 | if (source && (typeof source === 'object' || Array.isArray(source))) { 10 | if (source instanceof Date) 11 | return (new Date(source.getTime()) as unknown) as T; 12 | target = (Array.isArray(source) ? [] : {}) as any; 13 | if (!fields) 14 | for (const field in source) target[field] = deepCopy(source[field]); 15 | else 16 | fields.forEach( 17 | field => ((target as any)[field] = deepCopy((source as any)[field])) 18 | ); 19 | } 20 | return target; 21 | } 22 | -------------------------------------------------------------------------------- /demo/demo.css: -------------------------------------------------------------------------------- 1 | body * { 2 | box-sizing: border-box; 3 | } 4 | 5 | h1, 6 | h2, 7 | h3 { 8 | margin-bottom: 0; 9 | } 10 | 11 | .sample>.row { 12 | height: 300px; 13 | } 14 | 15 | .row::after { 16 | content: ''; 17 | display: table; 18 | clear : both; 19 | } 20 | 21 | .cntr { 22 | position: relative; 23 | padding : 10px; 24 | height : 100%; 25 | } 26 | 27 | .cntr>* { 28 | position: absolute; 29 | left : 0; 30 | top : 0; 31 | right : 0; 32 | bottom : 0; 33 | overflow: auto; 34 | border : 1px solid #aaa; 35 | padding : 5px; 36 | margin : 5px; 37 | } 38 | 39 | .cntr.chart { 40 | width: 70%; 41 | float: left; 42 | } 43 | 44 | .cntr.params { 45 | width: 30%; 46 | float: right; 47 | } 48 | 49 | pi-chart canvas { 50 | width : 100%; 51 | height: 100%; 52 | } 53 | 54 | pi-chart>.info { 55 | font : 16px Arial, helvetica, sans-serif; 56 | position: absolute; 57 | left : 52px; 58 | top : 48px; 59 | } 60 | 61 | .info, 62 | .info td { 63 | border : 1px solid #aaa; 64 | background-color: #fff8; 65 | } 66 | 67 | button { 68 | position: absolute; 69 | right : 0; 70 | top : 0; 71 | } -------------------------------------------------------------------------------- /package.json: -------------------------------------------------------------------------------- 1 | { 2 | "name": "chartjs-plugin-regression", 3 | "version": "0.2.1", 4 | "description": "Plugin to draw a regression line", 5 | "main": "lib/index.js", 6 | "scripts": { 7 | "build": "tsc && webpack --mode development", 8 | "format": "prettier --write \"src/**/*.ts\" \"demo/**/*.js\"", 9 | "lint": "tslint -p tsconfig.json", 10 | "test": "echo \"Warning: no test specified\"", 11 | "preversion": "npm test", 12 | "version": "node bin/setversion.js && npm run prepare && git add -A", 13 | "postversion": "git push && git push --tags", 14 | "prepare": "find . -name '*.map' -delete; tsc && webpack --mode production" 15 | }, 16 | "repository": { 17 | "type": "git", 18 | "url": "git+https://github.com/pomgui/chartjs-plugin-regression.git" 19 | }, 20 | "keywords": [ 21 | "charts", 22 | "chartjs", 23 | "regression", 24 | "data", 25 | "fiting", 26 | "analysis" 27 | ], 28 | "bugs": { 29 | "url": "https://github.com/pomgui/chartjs-plugin-regression/issues" 30 | }, 31 | "homepage": "https://github.com/pomgui/chartjs-plugin-regression#readme", 32 | "author": "Wilfredo Pomier (wpomier@pomgui.com)", 33 | "license": "ISC", 34 | "devDependencies": { 35 | "@types/chart.js": "^2.9.22", 36 | "@types/regression": "^2.0.0", 37 | "prettier": "^2.0.5", 38 | "source-map-loader": "^1.0.1", 39 | "tslint": "^6.1.2", 40 | "tslint-config-prettier": "^1.18.0", 41 | "typescript": "^3.9.6" 42 | }, 43 | "dependencies": { 44 | "regression": "^2.0.1" 45 | }, 46 | "peerDependencies": { 47 | "chart.js": "^2.6.0" 48 | }, 49 | "files": [ 50 | "lib/**/*" 51 | ] 52 | } 53 | -------------------------------------------------------------------------------- /demo/index.html: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | chartjs-plugin-regression 7 | 9 | 10 | 12 | 14 | 15 | 16 | 17 | 18 | 19 | 20 |
21 |
22 |

23 |
24 |

25 |
26 |
27 | 28 | 29 |
30 |
31 |
32 |
33 |

Global Options Config

34 | 

36 |                             

37 |                             

38 |                                 
Dataset Config

39 |                                 

41 |                             

42 |                         

43 |                     

44 |                 

45 |             

46 |         

47 |     

48 | 
49 | 
50 | 


--------------------------------------------------------------------------------
/demo/demoHelpers.js:
--------------------------------------------------------------------------------
 1 | var helpers = new (function () {
 2 |   this.generateRandomData = function (numElems, type, prediction) {
 3 |     var types = { linear, logarithmic, power, exponential };
 4 |     var fn = (types[type] || types.exponential)();
 5 |     var delta = 0;
 6 |     return new Array(numElems).fill(0).map(function (v, i) {
 7 |       return !i || (prediction == 'last' && i >= demo.NUM_NORMAL_ELEMS)
 8 |         ? null
 9 |         : fn(i) + helpers.rnd(delta);
10 |     });
11 |   };
12 | 
13 |   function linear() {
14 |     var A = helpers.rnd(10) + 10,
15 |       B = helpers.rnd(100),
16 |       rnd = f(demo.NUM_ELEMS_4);
17 |     function f(x) {
18 |       return A * x + B;
19 |     }
20 |     return function (x) {
21 |       return f(x) + Math.random() * rnd;
22 |     };
23 |   }
24 |   function exponential() {
25 |     var A = Math.random() * 10,
26 |       B = Math.random() * 0.2 + 1e-2,
27 |       rnd = f(demo.NUM_ELEMS_2);
28 |     function f(x) {
29 |       return A * Math.exp(x * B);
30 |     }
31 |     return function (x) {
32 |       return f(x) + Math.random() * rnd;
33 |     };
34 |   }
35 |   function logarithmic() {
36 |     var A = helpers.rnd(200),
37 |       B = helpers.rnd(50) + 1,
38 |       rnd = f(0.5);
39 |     function f(x) {
40 |       return A + B * Math.log(x);
41 |     }
42 |     return function (x) {
43 |       return f(x) + Math.random() * rnd;
44 |     };
45 |   }
46 |   function power() {
47 |     var A = Math.random() * 2 + 0.2,
48 |       B = Math.random() * 2 + 0.2,
49 |       rnd = f(demo.NUM_ELEMS_4);
50 |     function f(x) {
51 |       return A * Math.pow(x, B);
52 |     }
53 |     return function (x) {
54 |       return f(x) + Math.random() * rnd;
55 |     };
56 |   }
57 | 
58 |   this.generateColors = function (numElems, id, alpha) {
59 |     var normal = this.color(id, alpha);
60 |     var predicted = 'rgba(136,136,136,' + alpha + ')';
61 |     var colors = new Array(numElems);
62 |     for (var i = 0; i < numElems; i++)
63 |       colors[i] = i < demo.NUM_NORMAL_ELEMS ? normal : predicted;
64 |     return colors;
65 |   };
66 | 
67 |   this.rnd = function (max) {
68 |     return Math.trunc(max * Math.random());
69 |   };
70 | 
71 |   this.formatJson = function (cfg) {
72 |     var s = JSON.stringify(cfg, jsonReplacer, 2)
73 |       .replace(/"(\w+)":/g, '$1:')
74 |       .replace(/\{[^{}}]+\}/gm, strReplacer)
75 |       .replace(/\[[^\[\]]+\]/gm, strReplacer);
76 |     return s;
77 |     function jsonReplacer(k, v) {
78 |       return k == '$$hashKey' ? undefined : v;
79 |     }
80 |     function strReplacer(g) {
81 |       return g.length < 80 ? g.replace(/\s+/g, ' ') : g;
82 |     }
83 |   };
84 | 
85 |   this.color = function (id, alpha) {
86 |     const c = new Color({ h: id * 16, s: 100, l: 50 });
87 |     c.alpha(alpha);
88 |     return c.rgbaString();
89 |   }
90 | })();
91 | 


--------------------------------------------------------------------------------
/src/regression-plugin.ts:
--------------------------------------------------------------------------------
 1 | import { MetaDataSet } from './MetaData';
 2 | import {
 3 |   PluginServiceGlobalRegistration,
 4 |   PluginServiceRegistrationOptions,
 5 |   Easing
 6 | } from 'chart.js';
 7 | import { MetaSection } from './MetaSection';
 8 | import { ChartDataSetsEx, OptionsConfig } from './types';
 9 | 
10 | // Cache for all plugins' metadata
11 | var _metadataMap: any = {};
12 | var _chartId = 0;
13 | 
14 | interface ChartEx extends Chart {
15 |   $$id: number;
16 | }
17 | 
18 | class Plugin
19 |   implements PluginServiceGlobalRegistration, PluginServiceRegistrationOptions {
20 |   id = 'regressions';
21 | 
22 |   beforeInit(chart: any) {
23 |     chart.$$id = ++_chartId;
24 |   }
25 | 
26 |   /**
27 |    * Called after update (when the chart is created and when chart.update() is called)
28 |    * @param chart
29 |    */
30 |   beforeUpdate?(chart: Chart, options?: any): void {
31 |     let o, p, r: OptionsConfig;
32 |     const onComplete =
33 |       (o = chart.config.options) &&
34 |       (p = o.plugins) &&
35 |       (r = p.regressions) &&
36 |       r.onCompleteCalculation;
37 | 
38 |     forEach(chart, (ds, meta, datasetIndex) => {
39 |       meta = new MetaDataSet(chart, ds);
40 |       const id = (chart as ChartEx).$$id * 1000 + datasetIndex;
41 |       _metadataMap[id] = meta;
42 |     });
43 |     if (onComplete) onComplete(chart);
44 |   }
45 | 
46 |   /**
47 |    * It's called once before all the drawing
48 |    * @param chart
49 |    */
50 |   beforeRender(chart: Chart, options?: any): void {
51 |     forEach(chart, (ds, meta) => meta.adjustScales());
52 |   }
53 | 
54 |   /** Draws the vertical lines before the datasets are drawn */
55 |   beforeDatasetsDraw(chart: Chart, easing: Easing, options?: any): void {
56 |     forEach(chart, (ds, meta) => meta.drawRightBorders());
57 |   }
58 | 
59 |   /** Draws the regression lines */
60 |   afterDatasetsDraw(chart: Chart, easing: Easing, options?: any): void {
61 |     forEach(chart, (ds, meta) => meta.drawRegressions());
62 |   }
63 | 
64 |   destroy(chart: Chart): void {
65 |     Object.keys(_metadataMap)
66 |       .filter((k: any) => (k / 1000) >> 0 == (chart as ChartEx).$$id)
67 |       .forEach(k => delete _metadataMap[k]);
68 |   }
69 | 
70 |   /** Get dataset's meta data */
71 |   getDataset(chart: ChartEx, datasetIndex: number): MetaDataSet {
72 |     const id = chart.$$id * 1000 + datasetIndex;
73 |     return _metadataMap[id];
74 |   }
75 | 
76 |   /** Get dataset's meta sections */
77 |   getSections(chart: ChartEx, datasetIndex: number): MetaSection[] {
78 |     const ds = this.getDataset(chart, datasetIndex);
79 |     return ds && ds.sections;
80 |   }
81 | }
82 | 
83 | function forEach(
84 |   chart: any,
85 |   fn: (ds: ChartDataSetsEx, meta: MetaDataSet, datasetIndex: number) => void
86 | ) {
87 |   chart.data.datasets.forEach((ds: ChartDataSetsEx, i: number) => {
88 |     if (ds.regressions && chart.isDatasetVisible(i)) {
89 |       const meta = ChartRegressions.getDataset(chart, i);
90 |       fn(ds, meta, i);
91 |     }
92 |   });
93 | }
94 | 
95 | export const ChartRegressions = new Plugin();
96 | 
97 | declare var window: any;
98 | window.ChartRegressions = ChartRegressions;
99 | 


--------------------------------------------------------------------------------
/src/MetaData.ts:
--------------------------------------------------------------------------------
  1 | import { Options, DataPoint } from 'regression';
  2 | import { Section, LineOptions, ChartDataSetsEx, DatasetConfig } from './types';
  3 | import { Point } from 'chart.js';
  4 | import { MetaSection } from './MetaSection';
  5 | 
  6 | type GetXY = (x: number, y: number) => Point;
  7 | 
  8 | export class MetaDataSet {
  9 |   chart: Chart;
 10 |   dataset: ChartDataSetsEx;
 11 | 
 12 |   /** Sections */
 13 |   sections: MetaSection[];
 14 | 
 15 |   /** Scales wil be initialized in beforeDraw hook */
 16 |   getXY: GetXY = undefined as any;
 17 |   topY?: number;
 18 |   bottomY?: number;
 19 | 
 20 |   /** Normalized dataset's data */
 21 |   normalizedData: DataPoint[];
 22 |   /** Is the dataset's data an array of {x,y}? */
 23 |   isXY = false;
 24 | 
 25 |   constructor(chart: Chart, ds: ChartDataSetsEx) {
 26 |     const cfg = ds.regressions;
 27 |     this.chart = chart;
 28 |     this.dataset = ds;
 29 |     this.normalizedData = this._normalizeData(ds.data!);
 30 |     this.sections = this._createMetaSections(cfg);
 31 |     this._calculate();
 32 |   }
 33 | 
 34 |   /**
 35 |    * Normalize data to DataPoint[]
 36 |    * Only supports number[] and {x:number,y:number}
 37 |    */
 38 |   _normalizeData(data: any[]): DataPoint[] {
 39 |     return data.map((value: any, index: number) => {
 40 |       let p: DataPoint;
 41 |       if (typeof value == 'number' || value == null || value === undefined) {
 42 |         p = [index, value];
 43 |       } else {
 44 |         this.isXY = true;
 45 |         p = [value.x as number, value.y as number];
 46 |       }
 47 |       return p;
 48 |     });
 49 |   }
 50 | 
 51 |   /** @private */
 52 |   _createMetaSections(cfg: DatasetConfig): MetaSection[] {
 53 |     const source = cfg.sections || [
 54 |       { startIndex: 0, endIndex: this.dataset.data!.length - 1 }
 55 |     ];
 56 |     return source.map((s: Section) => new MetaSection(s, this));
 57 |   }
 58 | 
 59 |   /** @private */
 60 |   _calculate() {
 61 |     this.sections.forEach(section => section.calculate()); // Calculate Section Results
 62 |   }
 63 | 
 64 |   adjustScales() {
 65 |     if (this.topY !== undefined) return;
 66 |     let xScale: any;
 67 |     let yScale: any;
 68 |     const scales = (this.chart as any).scales;
 69 |     Object.keys(scales).forEach(
 70 |       k => (k[0] == 'x' && (xScale = scales[k])) || (yScale = scales[k])
 71 |     );
 72 |     this.topY = yScale.top;
 73 |     this.bottomY = yScale.bottom;
 74 |     this.getXY = (x: number, y: number) => ({
 75 |       x: xScale.getPixelForValue(x, undefined, undefined, true),
 76 |       y: yScale.getPixelForValue(y)
 77 |     });
 78 |   }
 79 | 
 80 |   drawRegressions() {
 81 |     const ctx: CanvasRenderingContext2D = (this.chart as any).chart.ctx;
 82 |     ctx.save();
 83 |     try {
 84 |       this.sections.forEach(section => section.drawRegressions(ctx));
 85 |     } finally {
 86 |       ctx.restore();
 87 |     }
 88 |   }
 89 | 
 90 |   drawRightBorders() {
 91 |     const ctx: CanvasRenderingContext2D = (this.chart as any).chart.ctx;
 92 |     ctx.save();
 93 |     try {
 94 |       for (let i = 0; i < this.sections.length - 1; i++)
 95 |         this.sections[i].drawRightBorder(ctx);
 96 |     } finally {
 97 |       ctx.restore();
 98 |     }
 99 |   }
100 | }
101 | 


--------------------------------------------------------------------------------
/src/types.ts:
--------------------------------------------------------------------------------
 1 | import { ChartDataSets } from 'chart.js';
 2 | import * as regression from 'regression';
 3 | 
 4 | export type Type =
 5 |   | 'copy'
 6 |   | 'linear'
 7 |   | 'exponential'
 8 |   | 'power'
 9 |   | 'polynomial'
10 |   | 'polynomial3'
11 |   | 'polynomial4'
12 |   | 'logarithmic';
13 | export type OverwritingType = 'none' | 'all' | 'empty' | 'last';
14 | 
15 | export interface LineOptions {
16 |   width?: number;
17 |   color?: string;
18 |   dash?: number[];
19 | }
20 | 
21 | export type CalculationOptions = regression.Options;
22 | 
23 | export interface CopyOptions {
24 |   /**
25 |    * none - No data will be overwriten (default)
26 |    * all  - All data will be overwriten.
27 |    * empty- Only zero, undefined, or null data will be overwriten.
28 |    * last - Only the last item (data[endIndex]) will be overwriten.
29 |    */
30 |   overwriteData?: OverwritingType;
31 |   /** Minimum value that the predicted value can be written into the data */
32 |   minValue?: number;
33 |   /** Maximum value that the predicted value can be written into the data */
34 |   maxValue?: number;
35 | }
36 | 
37 | export interface BasicOptions {
38 |   /** Type of regression to be calculated for all the sections unless they define their own type */
39 |   type?: Type | Type[];
40 |   /** Line configuration for all the sections unless they define their own line */
41 |   line?: LineOptions;
42 |   /** Precision and polynomial order of the values returned by the regression calculations */
43 |   calculation?: CalculationOptions;
44 |   /** Previous sections predictions for the current section will be drawed as dashed lines */
45 |   extendPredictions?: boolean;
46 |   /** Only if type=='copy' */
47 |   copy?: CopyOptions;
48 | }
49 | 
50 | export interface CopyOptionsEx extends CopyOptions {
51 |   /** Copy the predictions calculated by the section with index fromSectionIndex */
52 |   fromSectionIndex?: number;
53 | }
54 | 
55 | export interface Section extends BasicOptions {
56 |   /** Start index on dataset's data. Default: 0 */
57 |   startIndex?: number;
58 |   /** End index on dataset's data. Default: data.length-1 */
59 |   endIndex?: number;
60 |   /** If type=='copy' the section can be configured with this extended options for copy */
61 |   copy?: CopyOptionsEx;
62 |   /** Label that will be drawn in the top of the right border line. Default: xaxis' label */
63 |   label?: string;
64 | }
65 | 
66 | export interface ChartDataSetsEx extends ChartDataSets {
67 |   regressions: DatasetConfig;
68 | }
69 | 
70 | export interface DatasetConfig extends BasicOptions {
71 |   /** Sections of the data that shall draw a regression. If not specified it's assumed {start:0,end:data.length-1} */
72 |   sections?: Section[];
73 | }
74 | 
75 | /**
76 |  * The options.plugin.regressions have a global configuration that can be override by
77 |  * the Dataset or Section configuration.
78 |  * Example: If In options is configured a precision, all the datasets and sections will
79 |  * use it, unless they set their own precision.
80 |  * This hierarchical configuration avoids repetitive configurations
81 |  */
82 | export interface OptionsConfig extends BasicOptions {
83 |   /**
84 |    * Callback called when the regressions for all the datasets in
85 |    * a chart have been calculated
86 |    */
87 |   onCompleteCalculation?: (chart: Chart) => void;
88 | }
89 | 


--------------------------------------------------------------------------------
/demo/demo.js:
--------------------------------------------------------------------------------
  1 | var app = angular.module('demoApp', []);
  2 | app.controller('DemoController', function () {
  3 |   this.groups = demo.groups;
  4 |   this.format = helpers.formatJson;
  5 | });
  6 | 
  7 | app.directive('piChart', function () {
  8 |   var me = this;
  9 |   return {
 10 |     restrict: 'E',
 11 |     template:
 12 |       '' +
 13 |       '' +
 14 |       '' +
 15 |       '  ' +
 16 |       '    ' +
 17 |       '    ' +
 18 |       '    ' +
 19 |       '  ' +
 20 |       '
{{s.type}}R² = {{s.r2}}{{s.string}}
',
 21 |     scope: {
 22 |       sample: '<'
 23 |     },
 24 |     link: piChart_LinkFn
 25 |   };
 26 | });
 27 | 
 28 | function piChart_LinkFn($scope, element) {
 29 |   $scope.shuffle = shuffle;
 30 |   var chart, chartType, isBarChart;
 31 |   var numElems = $scope.sample.prediction
 32 |     ? demo.NUM_ELEMS_PREDICT
 33 |     : demo.NUM_NORMAL_ELEMS;
 34 |   var chartConfig = createChartConfiguration();
 35 |   shuffle();
 36 |   return;
 37 | 
 38 |   function createChartConfiguration() {
 39 |     chartType = $scope.sample.chartType;
 40 |     chartType = chartType || ['bar', 'line'][helpers.rnd(2)];
 41 |     isBarChart = chartType == 'bar';
 42 | 
 43 |     // Chart.js configuration:
 44 |     return {
 45 |       type: chartType,
 46 |       data: {
 47 |         labels: new Array(numElems).fill().map((v, i) => i),
 48 |         // helpers.generateLabels(numElems, $scope.sample.prediction),
 49 |         datasets: generateDatasets(
 50 |           $scope.sample.numDatasets || 1,
 51 |           $scope.sample.datasetCfg
 52 |         )
 53 |       },
 54 |       plugins: [ChartRegressions],
 55 |       options: {
 56 |         plugins: {
 57 |           // Global configuration of the plugin for all the datasets
 58 |           regressions: Object.assign({
 59 |             onCompleteCalculation: showRegressionResults
 60 |           },
 61 |             $scope.sample.optionsCfg
 62 |           )
 63 |         },
 64 |         responsive: true,
 65 |         maintainAspectRatio: false
 66 |       }
 67 |     };
 68 |   }
 69 | 
 70 |   /** Change the data, so the regression results may change as well */
 71 |   function shuffle() {
 72 |     $scope.btnShuffle = 'wait...';
 73 |     var s = $scope.sample;
 74 |     var dsType = s.datasetCfg.type;
 75 |     var opType = s.optionsCfg && s.optionsCfg.type;
 76 |     chartConfig.data.datasets.forEach(function (ds) {
 77 |       ds.data = helpers.generateRandomData(
 78 |         numElems,
 79 |         dsType || opType,
 80 |         s.prediction
 81 |       );
 82 |     });
 83 | 
 84 |     // Update or create the chart
 85 |     if (chart) chart.update();
 86 |     else {
 87 |       var canvas = element[0].firstChild;
 88 |       chart = new Chart(canvas, chartConfig);
 89 |     }
 90 |   }
 91 | 
 92 |   function generateDatasets(numDatasets, datasetCfg) {
 93 |     var ds = [];
 94 |     for (var i = 0; i < numDatasets; i++) {
 95 |       var color = helpers.rnd(23);
 96 |       ds.push({
 97 |         label: 'data#' + (i + 1),
 98 |         data: undefined, // settled in shuffle() function
 99 |         borderColor: helpers.generateColors(numElems, color, 1),
100 |         backgroundColor: !isBarChart
101 |           ? helpers.color(color, 0.5)
102 |           : helpers.generateColors(numElems, color, 0.5),
103 |         showLine: isBarChart,
104 |         // Configuration of the plugin for the dataset:
105 |         regressions: datasetCfg
106 |       });
107 |     }
108 |     return ds;
109 |   }
110 | 
111 |   function showRegressionResults(chart2) {
112 |     var results = [];
113 |     for (var i = 0; i < ($scope.sample.numDatasets || 1); i++) {
114 |       var sections = ChartRegressions.getSections(chart2, i);
115 |       sections.forEach(function (s) {
116 |         if (s.result.r2) {
117 |           s = Object.assign({ line: s.line }, s.result);
118 |           s.r2 = Math.round(s.r2 * 1000) / 10 + '%';
119 |           results.push(s);
120 |         }
121 |       });
122 |     }
123 |     $scope.results = results;
124 |     $scope.btnShuffle = 'shuffle';
125 |   }
126 | }
127 | 


--------------------------------------------------------------------------------
/src/MetaSection.ts:
--------------------------------------------------------------------------------
  1 | import * as regression from 'regression';
  2 | import {
  3 |   Section,
  4 |   Type,
  5 |   LineOptions,
  6 |   BasicOptions,
  7 |   CalculationOptions,
  8 |   CopyOptionsEx
  9 | } from './types';
 10 | import { MetaDataSet } from './MetaData';
 11 | 
 12 | const defaultConfig: BasicOptions = {
 13 |   type: 'linear',
 14 |   calculation: {
 15 |     precision: 2,
 16 |     order: 2
 17 |   },
 18 |   line: {
 19 |     width: 2,
 20 |     color: '#000',
 21 |     dash: []
 22 |   },
 23 |   extendPredictions: false,
 24 |   copy: {
 25 |     overwriteData: 'none'
 26 |   }
 27 | };
 28 | 
 29 | export interface Result extends regression.Result {
 30 |   type: Type;
 31 | }
 32 | 
 33 | export class MetaSection implements Section, BasicOptions {
 34 |   type: Type[];
 35 |   startIndex: number;
 36 |   endIndex: number;
 37 |   line: LineOptions;
 38 |   extendPredictions: boolean;
 39 |   result?: regression.Result;
 40 |   copy: CopyOptionsEx;
 41 |   calculation: CalculationOptions;
 42 |   label: string;
 43 | 
 44 |   constructor(sec: Section, private _meta: MetaDataSet) {
 45 |     const chart = _meta.chart;
 46 |     const ds = _meta.dataset;
 47 |     const cfg = getConfig([
 48 |       'type',
 49 |       'calculation',
 50 |       'line',
 51 |       'extendPredictions',
 52 |       'copy'
 53 |     ]);
 54 |     this.startIndex = sec.startIndex || 0;
 55 |     this.endIndex = sec.endIndex || ds.data!.length - 1;
 56 |     this.type = Array.isArray(cfg.type) ? cfg.type : [cfg.type];
 57 |     this.line = cfg.line;
 58 |     this.calculation = cfg.calculation;
 59 |     this.extendPredictions = cfg.extendPredictions;
 60 |     this.copy = cfg.copy;
 61 |     this.label =
 62 |       sec.label || (this._meta.chart.data.labels![this.endIndex] as string);
 63 |     this._validateType();
 64 | 
 65 |     // --- constructor helpers
 66 | 
 67 |     /**
 68 |      * Calculate the inherited configuration from defaultConfig, globalConfig,
 69 |      * dataset config, and section config (in that order)
 70 |      */
 71 |     function getConfig(fields: string[]): any {
 72 |       let o, p;
 73 |       const globalConfig =
 74 |         ((o = chart.config.options) && (p = o.plugins) && p.regressions) || {};
 75 |       return configMerge(
 76 |         fields,
 77 |         defaultConfig,
 78 |         globalConfig,
 79 |         ds.regressions,
 80 |         sec
 81 |       );
 82 | 
 83 |       /** merge the config objects */
 84 |       function configMerge(fields: string[], ...cfgList: any[]): any {
 85 |         const dstConfig: any = {};
 86 |         fields.forEach(f => {
 87 |           cfgList.forEach(srcConfig => {
 88 |             const o = srcConfig[f];
 89 |             const t = typeof o;
 90 |             if (t != 'undefined') {
 91 |               if (Array.isArray(o) || t != 'object' || o == null)
 92 |                 dstConfig[f] = o;
 93 |               else
 94 |                 dstConfig[f] = Object.assign(
 95 |                   {},
 96 |                   dstConfig[f],
 97 |                   configMerge(Object.keys(o), o)
 98 |                 );
 99 |             }
100 |           });
101 |         });
102 |         return dstConfig;
103 |       }
104 |     }
105 |   }
106 |   /** Validates the type to avoid inconsistences */
107 |   _validateType() {
108 |     if (this.type.length > 1 && this.type.includes('copy'))
109 |       throw Error(
110 |         'Invalid regression type:' +
111 |           this.type +
112 |           '. "none" cannot be combined with other type!'
113 |       );
114 |   }
115 | 
116 |   /** Calculates the regression(s) and sets the result objects */
117 |   calculate() {
118 |     const sectionData = this._meta.normalizedData.slice(
119 |       this.startIndex,
120 |       this.endIndex + 1
121 |     );
122 |     if (this.type[0] == 'copy') this._calculateCopySection(sectionData);
123 |     else this._calculateBestR2(sectionData);
124 |   }
125 | 
126 |   private _calculateBestR2(sectionData: regression.DataPoint[]) {
127 |     this.result = this.type.reduce((max: any, type) => {
128 |       let calculation = Object.assign({}, this.calculation);
129 |       let realType = type;
130 |       if (/polynomial[34]$/.test(type)) {
131 |         calculation.order = parseInt(type.substr(10));
132 |         realType = type.substr(0, 10) as Type;
133 |       }
134 |       const r: Result = (regression as any)[realType](sectionData, calculation);
135 |       r.type = type;
136 |       return !max || max.r2 < r.r2 ? r : max;
137 |     }, null);
138 |   }
139 | 
140 |   private _calculateCopySection(sectionData: regression.DataPoint[]) {
141 |     const from = this._meta.sections[this.copy.fromSectionIndex!],
142 |       r = (this.result = Object.assign({}, from.result)),
143 |       overwrite = this.copy.overwriteData,
144 |       data = this._meta.normalizedData;
145 |     r.points = sectionData.map(p => r.predict(p[0])) as any;
146 |     delete r.r2;
147 |     if (overwrite != 'none') {
148 |       const dsdata = this._meta.dataset.data!,
149 |         isXY = this._meta.isXY;
150 |       r.points.forEach(([x, y], i) => {
151 |         const index = i + this.startIndex;
152 |         if (
153 |           (index < from.startIndex || index > from.endIndex) &&
154 |           (overwrite == 'all' ||
155 |             (overwrite == 'last' && index == this.endIndex) ||
156 |             (overwrite == 'empty' && !data[index]))
157 |         ) {
158 |           if (this.copy.maxValue) y = Math.min(this.copy.maxValue, y);
159 |           if (this.copy.minValue !== undefined)
160 |             y = Math.max(this.copy.minValue, y);
161 |           dsdata[index] = isXY ? { x, y } : y;
162 |         }
163 |       });
164 |     }
165 |   }
166 | 
167 |   drawRightBorder(ctx: CanvasRenderingContext2D): void {
168 |     ctx.beginPath();
169 |     this._setLineAttrs(ctx);
170 |     ctx.setLineDash([10, 2]);
171 |     ctx.lineWidth = 2;
172 |     // Print vertical line
173 |     const p = this._meta.getXY(this.endIndex, 0);
174 |     ctx.moveTo(p.x, this._meta.topY!);
175 |     ctx.lineTo(p.x, this._meta.bottomY!);
176 |     ctx.fillStyle = this.line.color!;
177 |     ctx.fillText(this.label, p.x, this._meta.topY!);
178 |     ctx.stroke();
179 |   }
180 | 
181 |   drawRegressions(ctx: CanvasRenderingContext2D): void {
182 |     for (let i = 0, len = this._meta.sections.length; i < len; i++) {
183 |       const section = this._meta.sections[i];
184 |       const isMe = section == this;
185 |       if (
186 |         (isMe && this.type[0] != 'copy') ||
187 |         (!isMe && this.extendPredictions)
188 |       ) {
189 |         section.drawRange(ctx, this.startIndex, this.endIndex, !isMe);
190 |       }
191 |       if (isMe) break;
192 |     }
193 |   }
194 | 
195 |   drawRange(
196 |     ctx: CanvasRenderingContext2D,
197 |     startIndex: number,
198 |     endIndex: number,
199 |     forceDash: boolean
200 |   ): void {
201 |     ctx.beginPath();
202 |     this._setLineAttrs(ctx);
203 |     if (forceDash) ctx.setLineDash([5, 5]);
204 |     const predict = this.result!.predict;
205 |     const f = (x: number) => this._meta.getXY(x, predict(x)[1]);
206 |     let p = f(startIndex);
207 |     ctx.moveTo(p.x, p.y);
208 |     for (let x = startIndex + 1; x <= endIndex; x++) {
209 |       p = f(x);
210 |       ctx.lineTo(p.x, p.y);
211 |     }
212 |     ctx.stroke();
213 |   }
214 | 
215 |   private _setLineAttrs(ctx: CanvasRenderingContext2D) {
216 |     if (this.line.width) ctx.lineWidth = this.line.width;
217 |     if (this.line.color) ctx.strokeStyle = this.line.color;
218 |     if (this.line.dash) ctx.setLineDash(this.line.dash);
219 |   }
220 | }
221 | 


--------------------------------------------------------------------------------
/demo/demoData.js:
--------------------------------------------------------------------------------
  1 | 'use strict';
  2 | 
  3 | var demo = new (function () {
  4 |   this.NUM_NORMAL_ELEMS = 50;
  5 |   this.NUM_ELEMS_2 = (this.NUM_NORMAL_ELEMS / 2) | 0;
  6 |   this.NUM_ELEMS_4 = (this.NUM_NORMAL_ELEMS / 4) | 0;
  7 |   this.NUM_ELEMS_34 = ((3 * this.NUM_NORMAL_ELEMS) / 4) | 0;
  8 |   this.NUM_ELEMS_PREDICT = this.NUM_NORMAL_ELEMS + 10;
  9 | 
 10 |   this.groups = [
 11 |     {
 12 |       title: 'Single Regression Type',
 13 |       samples: [
 14 |         {
 15 |           subtitle: 'Linear regression',
 16 |           // optionsCfg: {},
 17 |           datasetCfg: { type: 'linear', line: { color: 'red' } }
 18 |         },
 19 |         {
 20 |           subtitle: 'Exponential regression',
 21 |           // optionsCfg: {},
 22 |           datasetCfg: {
 23 |             type: 'exponential',
 24 |             line: { color: '#0000ff', width: 3 }
 25 |           }
 26 |         },
 27 |         {
 28 |           subtitle: 'Polynomial regression',
 29 |           // optionsCfg: {},
 30 |           datasetCfg: {
 31 |             type: 'polynomial',
 32 |             line: { color: '#0000ff', width: 3 },
 33 |             calculation: { precision: 10, order: 4 }
 34 |           }
 35 |         },
 36 |         {
 37 |           subtitle: 'Power regression',
 38 |           // optionsCfg: {},
 39 |           datasetCfg: { type: 'power', line: { color: '#0000ff', width: 3 } }
 40 |         },
 41 |         {
 42 |           subtitle: 'Logarithmic',
 43 |           // optionsCfg: {},
 44 |           datasetCfg: {
 45 |             type: 'logarithmic',
 46 |             calculation: { precision: 10 },
 47 |             line: { color: '#0000ff', width: 3 }
 48 |           }
 49 |         }
 50 |       ]
 51 |     },
 52 |     {
 53 |       title: 'Drawing the best regression type',
 54 |       samples: [
 55 |         {
 56 |           subtitle: 'Best R² between Exponential and Polynomial',
 57 |           // optionsCfg: {},
 58 |           datasetCfg: {
 59 |             type: ['exponential', 'polynomial'],
 60 |             line: { color: '#0000ff', width: 3 },
 61 |             calculation: { order: 3 }
 62 |           }
 63 |         }
 64 |       ]
 65 |     },
 66 |     {
 67 |       title: 'Sections',
 68 |       samples: [
 69 |         {
 70 |           subtitle: 'Single section of the data',
 71 |           optionsCfg: {
 72 |             line: { color: 'red', width: 3 }
 73 |           },
 74 |           datasetCfg: {
 75 |             type: ['linear', 'exponential', 'polynomial'],
 76 |             sections: [
 77 |               { startIndex: this.NUM_ELEMS_4, endIndex: this.NUM_ELEMS_34 }
 78 |             ]
 79 |           }
 80 |         },
 81 |         {
 82 |           subtitle: 'Multiple sections extending previous predictions',
 83 |           optionsCfg: {
 84 |             line: { color: 'blue', width: 3 }
 85 |           },
 86 |           datasetCfg: {
 87 |             type: ['linear', 'exponential', 'polynomial'],
 88 |             extendPredictions: true,
 89 |             sections: [
 90 |               { endIndex: this.NUM_ELEMS_2, line: { color: 'red' } },
 91 |               {
 92 |                 startIndex: this.NUM_ELEMS_2,
 93 |                 endIndex: this.NUM_NORMAL_ELEMS - 1
 94 |               }
 95 |             ]
 96 |           }
 97 |         },
 98 |         {
 99 |           subtitle: 'Multiple sections with custom label on right border',
100 |           optionsCfg: {
101 |             line: { color: 'blue', width: 3 }
102 |           },
103 |           datasetCfg: {
104 |             type: ['linear', 'exponential', 'polynomial'],
105 |             extendPredictions: true,
106 |             sections: [
107 |               {
108 |                 endIndex: this.NUM_ELEMS_2,
109 |                 line: { color: 'red' },
110 |                 label: 'custom text'
111 |               },
112 |               {
113 |                 startIndex: this.NUM_ELEMS_2,
114 |                 endIndex: this.NUM_NORMAL_ELEMS - 1
115 |               }
116 |             ]
117 |           }
118 |         }
119 |       ]
120 |     },
121 |     {
122 |       title: 'Multiple regressions',
123 |       samples: [
124 |         {
125 |           subtitle: 'One dataset with 3 different regressions',
126 |           chartType: 'line',
127 |           // optionsCfg: {},
128 |           datasetCfg: {
129 |             line: { width: 3, color: 'blue' },
130 |             sections: [
131 |               { type: 'linear', line: { width: 1.5 } },
132 |               { type: 'polynomial', line: { color: 'orange', dash: [8, 2] } },
133 |               { type: 'exponential', line: { color: 'red' } }
134 |             ]
135 |           }
136 |         },
137 |         {
138 |           subtitle: 'One dataset with 3 polynomial with different order',
139 |           chartType: 'line',
140 |           optionsCfg: { type: 'polynomial', line: { width: 3, color: 'blue' } },
141 |           datasetCfg: {
142 |             sections: [
143 |               { calculation: { order: 2 }, line: { width: 1.5 } },
144 |               {
145 |                 calculation: { order: 3 },
146 |                 line: { color: 'orange', dash: [8, 2] }
147 |               },
148 |               { calculation: { order: 4 }, line: { color: 'red' } }
149 |             ]
150 |           }
151 |         }
152 |       ]
153 |     },
154 |     {
155 |       title: "Drawing predictions using other section's regression",
156 |       samples: [
157 |         {
158 |           prediction: 'none',
159 |           subtitle: "Without overwriting the last section's data",
160 |           chartType: 'bar',
161 |           // optionsCfg: {},
162 |           datasetCfg: {
163 |             type: ['linear', 'exponential', 'polynomial'],
164 |             line: { color: 'blue', width: 3 },
165 |             extendPredictions: true,
166 |             sections: [
167 |               { endIndex: this.NUM_ELEMS_2 },
168 |               {
169 |                 startIndex: this.NUM_ELEMS_2,
170 |                 endIndex: this.NUM_NORMAL_ELEMS - 1,
171 |                 line: { color: 'red' }
172 |               },
173 |               {
174 |                 type: 'copy',
175 |                 copy: { fromSectionIndex: 1, overwriteData: 'none' },
176 |                 startIndex: this.NUM_NORMAL_ELEMS - 1
177 |               }
178 |             ]
179 |           }
180 |         },
181 |         {
182 |           prediction: 'all',
183 |           subtitle: "Overwrites the last section's data",
184 |           chartType: 'bar',
185 |           // optionsCfg: {},
186 |           datasetCfg: {
187 |             type: ['linear', 'exponential', 'polynomial'],
188 |             line: { color: 'blue', width: 3 },
189 |             extendPredictions: true,
190 |             sections: [
191 |               { endIndex: this.NUM_ELEMS_2 },
192 |               {
193 |                 startIndex: this.NUM_ELEMS_2,
194 |                 endIndex: this.NUM_NORMAL_ELEMS - 1,
195 |                 line: { color: 'red' }
196 |               },
197 |               {
198 |                 type: 'copy',
199 |                 copy: { fromSectionIndex: 1, overwriteData: 'all' },
200 |                 startIndex: this.NUM_NORMAL_ELEMS - 1
201 |               }
202 |             ]
203 |           }
204 |         },
205 |         {
206 |           prediction: 'last',
207 |           subtitle: 'Overwrites only the last data item',
208 |           chartType: 'bar',
209 |           // optionsCfg: {},
210 |           datasetCfg: {
211 |             type: ['linear', 'exponential', 'polynomial'],
212 |             line: { color: 'blue', width: 3 },
213 |             extendPredictions: true,
214 |             sections: [
215 |               { endIndex: this.NUM_ELEMS_2 },
216 |               {
217 |                 startIndex: this.NUM_ELEMS_2,
218 |                 endIndex: this.NUM_NORMAL_ELEMS - 1,
219 |                 line: { color: 'red' }
220 |               },
221 |               {
222 |                 type: 'copy',
223 |                 copy: { fromSectionIndex: 1, overwriteData: 'last' },
224 |                 startIndex: this.NUM_NORMAL_ELEMS - 1
225 |               }
226 |             ]
227 |           }
228 |         }
229 |       ]
230 |     },
231 |     {
232 |       title: 'Configuration inheritance',
233 |       samples: [
234 |         {
235 |           subtitle: 'Inherits calculation: precision',
236 |           optionsCfg: { calculation: { precision: 5 } },
237 |           datasetCfg: { type: 'linear' }
238 |         },
239 |         {
240 |           subtitle: 'Inherits all in two different datasets',
241 |           numDatasets: 2,
242 |           chartType: 'line',
243 |           optionsCfg: { type: 'linear', calculation: { precision: 4 } },
244 |           datasetCfg: {}
245 |         },
246 |         {
247 |           subtitle:
248 |             'Inherits line and polynomial order, but overrides type and precision',
249 |           optionsCfg: {
250 |             type: 'linear',
251 |             line: { width: 3, dash: [2, 2], color: '#f00' },
252 |             calculation: { precision: 5, order: 4 }
253 |           },
254 |           datasetCfg: { type: 'polynomial', calculation: { precision: 3 } }
255 |         }
256 |       ]
257 |     }
258 |   ];
259 | })();
260 | 


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
  1 | [![NPM version](https://img.shields.io/npm/v/chartjs-plugin-regression.svg)](https://npmjs.org/package/chartjs-plugin-regression) 
  2 | [![License](http://img.shields.io/badge/license-ISC-blue.svg)](https://github.com/pomgui/chartjs-plugin-regression/blob/master/LICENSE)
  3 | [![NPM downloads](https://img.shields.io/npm/dm/chartjs-plugin-regression.svg)](https://npmjs.org/package/chartjs-plugin-regression)
  4 | 
  5 | # chartjs-plugin-regression
  6 | Chart.js plugin to calculate and draw statistical linear, exponential, power, 
  7 | logarithmic, and polynomial regressions using chart datasets data.
  8 | 
  9 | The plugin, at the current version, uses the [regression](https://www.npmjs.com/package/regression)
 10 | npm package as its calculation engine.
 11 | 
 12 | ### Important
 13 | - Only `bar`, `line`, and `scatter` chart types are supported.
 14 | - The plugin works just fine with chart.js@2.5.0, however that version may have some problems handling certain color configuration (not related with the plugin). No problems have been found with chart.js@^2.6.0.
 15 | - The plugin does not work with chart.js@3.x
 16 | 
 17 | ## Demo
 18 | For a better understanding of the capabilities of this plugin, please see this 
 19 | [Live Demo](https://pomgui.github.io/chartjs-plugin-regression/demo/).
 20 | 
 21 | ## Download
 22 | The [compressed](https://pomgui.github.io/chartjs-plugin-regression/dist/chartjs-plugin-regression-0.2.1.js)
 23 | version includes the regression package.
 24 | 
 25 | ## Installation
 26 | 
 27 |     npm install --save chartjs-plugin-regression
 28 | 
 29 | ## Usage
 30 | 
 31 | For a single chart, it needs to be listed in plugins section. 
 32 | 
 33 | ### Example:
 34 | 
 35 | ```javascript
 36 | new Chart(ctx, {
 37 |   type: 'bar',
 38 |   plugins: [
 39 |     // This chart will use the plugin
 40 |     ChartRegressions
 41 |   ],
 42 |   data: {
 43 |     ...
 44 |     datasets: [
 45 |       {
 46 |         ...
 47 |         // Configuration of the plugin per dataset (only will be drawn the datasets with this property) 
 48 |         regressions: {
 49 |           type: 'linear',
 50 |           line: { color: 'red', width: 3},
 51 |           ...
 52 |         }
 53 |       }
 54 |     ]
 55 |     ...
 56 |   }
 57 | });
 58 | ```
 59 | 
 60 | Also, it's possible to register the plugin for all the charts:
 61 | 
 62 | ```javascript
 63 | Chart.plugins.register(ChartRegressions);
 64 | ```
 65 | 
 66 | ## Configuration
 67 | 
 68 | The plugin has three levels of configuration:
 69 | 
 70 | - global (for all the datasets in a chart)
 71 | - Per dataset
 72 | - Per section
 73 | 
 74 | There are common properties that the three levels share, and the priority of them are: section, dataset, and global.
 75 | 
 76 | ### Common properties
 77 | 
 78 | Common to the three levels of configuration.
 79 | 
 80 | | Property | Description |
 81 | |---|---|
 82 | | type | Type of regression to be calculated. It can be 'copy', 'linear', 'exponential', 'power', 'polynomial', 'polynomial3', 'polynomial4', or 'logarithmic'. It also can be an array with a combination of these types, in which case the regression type with the best [_R²_](https://en.wikipedia.org/wiki/Coefficient_of_determination) will be drawn. |
 83 | | line | Line configuration for drawing the regression. It has the following properties: `{width, color, dash}` |
 84 | | calculation | Precision and polynomial order of the values returned by the regression calculations |
 85 | | extendPredictions | Previous sections predictions for the current section will be drawed as dashed lines |
 86 | | copy | Only if type=='copy'. Behavior of sections that copy other section's calculation |
 87 | 
 88 | Some considerations:
 89 | 
 90 | - `type`: polynomial3 and polynomial4 are pseudo-types added for convenience, they allow combinationss where the plugin will draw the regression with bigger R². Example:
 91 | 
 92 | ```javascript
 93 | {
 94 |   type: ['polynomial', 'polynomial3', 'polynomial4'],
 95 |   calculation: {order 2}
 96 | }
 97 | ```
 98 | 
 99 | - `calculation` has the following properties:
100 | 
101 | | Property | Description |
102 | |---|---|
103 | | `precision` | Determines how many decimals will have the results (default: 2). |
104 | | `order` | Only for `polynomial` regression type, i.e. `polynomial3` and `polynomial4` are not affected by this property. Example: _ax² + bx + c_ has order 2.  |
105 | 
106 | - `copy` has the following properties:
107 | 
108 | | Property | Description |
109 | |---|---|
110 | | `overwriteData` | Possible values: 'none', 'all', 'empty', or 'last'. Default: 'none'. It determines how the dataset's data will be overwritten in this section (empty: Only zero, undefined, or null data will be overwriten). **Obs.** the plugin is only prepared to overwrite numerical data arrays, e.g. `[1,2,3,...]`, scatter charts use xy data arrays, e.g. `[{x:1,y:1}, {x:2,y:2},...]`, with them the behavior is undetermined. In these cases it's better use `overwriteData: 'none'`. |
111 | | `minValue` | Minimum value that the predicted value can be written into the data. |
112 | | `maxValue` | Maximum value that the predicted value can be written into the data. |
113 | 
114 | ### Global
115 | 
116 | The global configuration affects all the regressions calculated for all the datasets in the chart. It contains all the common properties and the following properties:
117 | 
118 | | Property | Description |
119 | |---|---|
120 | | onCompleteCalculation | Callback called when the regressions for all the datasets in a chart have been calculated |
121 | 
122 | Example:
123 | 
124 | ```javascript
125 | options: {
126 |   plugins: {
127 |     regressions: {
128 |       type: ['linear', 'polynomial'],
129 |       line: { color: 'blue', width: 3 },
130 |       onCompleteCalculation: function callback(chart){ ... }
131 |     }
132 |   }
133 | }
134 | ```
135 | 
136 | ### Per Dataset
137 | 
138 | It's possible to configure the regressions per dataset. The configuration will contain all the common properties and the following properties:
139 | 
140 | | Property | Description |
141 | |---|---|
142 | | sections | Array of sections of the data that shall be drawn. If not specified it's assumed `[{start:0,end:data.length-1}]` |
143 | 
144 | Example:
145 | 
146 | ```javascript
147 | datasets: [
148 |   {
149 |     ...
150 |     // Configuration of the plugin per dataset (only will be drawn the datasets with this property) 
151 |     regressions: {
152 |       type: ['linear','exponential'],
153 |       line: { color: '#ff0', width: 3},
154 |       calculation: { precision: 5 },
155 |       sections: [{startIndex: 10, endIndex: 50}],
156 |       ...
157 |     }
158 |   }
159 | ]
160 | ```
161 | 
162 | ### Per Section
163 | 
164 | Each section can be configured independently using all the common properties and the following properties:
165 | 
166 | | Property | Description |
167 | |---|---|
168 | | startIndex | Start index on dataset's data. Default: 0 |
169 | | endIndex | End index on dataset's data. Default: data.length-1 |
170 | | label | Label that will be drawn in the top of the right border line. Default: xaxis' label |  
171 | | copy.fromSectionIndex | Copy the predictions calculated by other section (the one with index fromSectionIndex) |
172 | 
173 | Example:
174 | 
175 | ```javascript
176 | datasets: [
177 |   {
178 |     ...
179 |     // Configuration of the plugin per dataset (only will be drawn the datasets with this property) 
180 |     regressions: {
181 |       line: { width: 3 },
182 |       calculation: { precision: 5 },
183 |       sections: [
184 |         {
185 |           type: ['linear','exponential'],
186 |           line: { color: 'red' },
187 |           startIndex: 10, 
188 |           endIndex: 50
189 |         },
190 |         {
191 |           type: 'polynomial',
192 |           line: { color: 'green' },
193 |           startIndex: 50, 
194 |           endIndex: 80,
195 |           calculation: { order: 4 }
196 |         },
197 |       ]
198 |     }
199 |   }
200 |   ...
201 | ]
202 | ```
203 | 
204 | ## API
205 | 
206 | ### .getDataset(chart, datasetIndex)
207 | 
208 | Returns the metadata associated to one dataset used internally by the plugin to work.
209 | 
210 | ```javascript
211 | var meta = ChartRegressions.getDataset(chart, datasetIndex);
212 | ```
213 | 
214 | This object provides the following information:
215 | 
216 | | Property | Description |
217 | |---|---|
218 | | `sections` | array of sections for each dataset (it will contain at least 1 section) |
219 | | `getXY(x, y)` | Returns the canvas coordinates {x,y} for the data point `x, y`. |
220 | | `topY` | Minimum y coordinate in the canvas. |
221 | | `bottomY` | Maximum y coordinate in the canvas. |
222 | 
223 | ### .getSections(chart, datasetIndex)
224 | 
225 | Returns the sections with all the properties calculated (some with default values, or inherited from dataset's plugin configuration or the global configuration in options).
226 | 
227 | This object provides the following information:
228 | 
229 | | Property | Description |
230 | |---|---|
231 | | `type` | array of regression types used to calculate and draw the section. |
232 | | `startIndex` | Index of the dataset's data. |
233 | | `endIndex` | Index of the dataset's data. |
234 | | `line` | Configuration used to draw the lines {color, width, dash}. |
235 | | `result` | Regression calculation result (see [demo](https://pomgui.github.io/chartjs-plugin-regression/demo/)) to see how to use this information. |
236 | 
237 | ## Events
238 | 
239 | ### onCompleteCalculation(chart)
240 | 
241 | The plugin provides one single event to inform when the calculation of all the regresions for a chart have been conmpleted. 
242 | 
243 | This callback should be configured in the chart options.
244 | 
245 | Example:
246 | 
247 | ```javascript
248 | options: {
249 |   plugins: {
250 |     regressions: {
251 |       onCompleteCalculation: function callback(chart){ ... }
252 |     }
253 |   }
254 | }
255 | ```
256 | 
257 | ## License
258 | The project is released under the [ISC license](https://github.com/pomgui/chartjs-plugin-regression/blob/master/LICENSE).
259 | 


--------------------------------------------------------------------------------
/dist/chartjs-plugin-regression-0.2.1.js:
--------------------------------------------------------------------------------
1 | !function(t){var e={};function n(r){if(e[r])return e[r].exports;var i=e[r]={i:r,l:!1,exports:{}};return t[r].call(i.exports,i,i.exports,n),i.l=!0,i.exports}n.m=t,n.c=e,n.d=function(t,e,r){n.o(t,e)||Object.defineProperty(t,e,{enumerable:!0,get:r})},n.r=function(t){"undefined"!=typeof Symbol&&Symbol.toStringTag&&Object.defineProperty(t,Symbol.toStringTag,{value:"Module"}),Object.defineProperty(t,"__esModule",{value:!0})},n.t=function(t,e){if(1&e&&(t=n(t)),8&e)return t;if(4&e&&"object"==typeof t&&t&&t.__esModule)return t;var r=Object.create(null);if(n.r(r),Object.defineProperty(r,"default",{enumerable:!0,value:t}),2&e&&"string"!=typeof t)for(var i in t)n.d(r,i,function(e){return t[e]}.bind(null,i));return r},n.n=function(t){var e=t&&t.__esModule?function(){return t.default}:function(){return t};return n.d(e,"a",e),e},n.o=function(t,e){return Object.prototype.hasOwnProperty.call(t,e)},n.p="",n(n.s=2)}([function(t,e,n){"use strict";Object.defineProperty(e,"__esModule",{value:!0}),e.MetaDataSet=void 0;var r=n(1),i=function(){function t(t,e){this.getXY=void 0,this.isXY=!1;var n=e.regressions;this.chart=t,this.dataset=e,this.normalizedData=this._normalizeData(e.data),this.sections=this._createMetaSections(n),this._calculate()}return t.prototype._normalizeData=function(t){var e=this;return t.map((function(t,n){var r;return"number"==typeof t||null==t||void 0===t?r=[n,t]:(e.isXY=!0,r=[t.x,t.y]),r}))},t.prototype._createMetaSections=function(t){var e=this;return(t.sections||[{startIndex:0,endIndex:this.dataset.data.length-1}]).map((function(t){return new r.MetaSection(t,e)}))},t.prototype._calculate=function(){this.sections.forEach((function(t){return t.calculate()}))},t.prototype.adjustScales=function(){if(void 0===this.topY){var t,e,n=this.chart.scales;Object.keys(n).forEach((function(r){return"x"==r[0]&&(t=n[r])||(e=n[r])})),this.topY=e.top,this.bottomY=e.bottom,this.getXY=function(n,r){return{x:t.getPixelForValue(n,void 0,void 0,!0),y:e.getPixelForValue(r)}}}},t.prototype.drawRegressions=function(){var t=this.chart.chart.ctx;t.save();try{this.sections.forEach((function(e){return e.drawRegressions(t)}))}finally{t.restore()}},t.prototype.drawRightBorders=function(){var t=this.chart.chart.ctx;t.save();try{for(var e=0;e1&&this.type.includes("copy"))throw Error("Invalid regression type:"+this.type+'. "none" cannot be combined with other type!')},t.prototype.calculate=function(){var t=this._meta.normalizedData.slice(this.startIndex,this.endIndex+1);"copy"==this.type[0]?this._calculateCopySection(t):this._calculateBestR2(t)},t.prototype._calculateBestR2=function(t){var e=this;this.result=this.type.reduce((function(n,i){var o=Object.assign({},e.calculation),a=i;/polynomial[34]$/.test(i)&&(o.order=parseInt(i.substr(10)),a=i.substr(0,10));var s=r[a](t,o);return s.type=i,!n||n.r2n.endIndex)&&("all"==i||"last"==i&&l==e.endIndex||"empty"==i&&!o[l])&&(e.copy.maxValue&&(u=Math.min(e.copy.maxValue,u)),void 0!==e.copy.minValue&&(u=Math.max(e.copy.minValue,u)),a[l]=s?{x:c,y:u}:u)}))}},t.prototype.drawRightBorder=function(t){t.beginPath(),this._setLineAttrs(t),t.setLineDash([10,2]),t.lineWidth=2;var e=this._meta.getXY(this.endIndex,0);t.moveTo(e.x,this._meta.topY),t.lineTo(e.x,this._meta.bottomY),t.fillStyle=this.line.color,t.fillText(this.label,e.x,this._meta.topY),t.stroke()},t.prototype.drawRegressions=function(t){for(var e=0,n=this._meta.sections.length;eMath.abs(n[o][a])&&(a=s);for(var c=o;c=o;p--)n[p][l]-=n[p][o]*n[o][l]/n[o][o]}for(var h=r-1;h>=0;h--){for(var f=0,d=h+1;d=0;m--)b+=m>1?v[m]+"x^"+m+" + ":1===m?v[m]+"x + ":v[m];return{string:b,points:x,predict:g,equation:[].concat(n(v)).reverse(),r2:o(i(t,x),e.precision)}}};t.exports=Object.keys(a).reduce((function(t,n){return e({_round:o},t,(c=function(t,i){return a[n](t,e({},r,i))},(s=n)in(i={})?Object.defineProperty(i,s,{value:c,enumerable:!0,configurable:!0,writable:!0}):i[s]=c,i));var i,s,c}),{})})?r.apply(e,i):r)||(t.exports=o)},function(t,e,n){"use strict";Object.defineProperty(e,"__esModule",{value:!0}),e.ChartRegressions=void 0;var r=n(0),i={},o=0,a=function(){function t(){this.id="regressions"}return t.prototype.beforeInit=function(t){t.$$id=++o},t.prototype.beforeUpdate=function(t,e){var n,o,a,c=(n=t.config.options)&&(o=n.plugins)&&(a=o.regressions)&&a.onCompleteCalculation;s(t,(function(e,n,o){n=new r.MetaDataSet(t,e);var a=1e3*t.$$id+o;i[a]=n})),c&&c(t)},t.prototype.beforeRender=function(t,e){s(t,(function(t,e){return e.adjustScales()}))},t.prototype.beforeDatasetsDraw=function(t,e,n){s(t,(function(t,e){return e.drawRightBorders()}))},t.prototype.afterDatasetsDraw=function(t,e,n){s(t,(function(t,e){return e.drawRegressions()}))},t.prototype.destroy=function(t){Object.keys(i).filter((function(e){return e/1e3>>0==t.$$id})).forEach((function(t){return delete i[t]}))},t.prototype.getDataset=function(t,e){var n=1e3*t.$$id+e;return i[n]},t.prototype.getSections=function(t,e){var n=this.getDataset(t,e);return n&&n.sections},t}();function s(t,n){t.data.datasets.forEach((function(r,i){if(r.regressions&&t.isDatasetVisible(i)){var o=e.ChartRegressions.getDataset(t,i);n(r,o,i)}}))}e.ChartRegressions=new a,window.ChartRegressions=e.ChartRegressions}]);


--------------------------------------------------------------------------------
/dist/chartjs-plugin-regression-0.1.1.js:
--------------------------------------------------------------------------------
  1 | /******/ (function(modules) { // webpackBootstrap
  2 | /******/ 	// The module cache
  3 | /******/ 	var installedModules = {};
  4 | /******/
  5 | /******/ 	// The require function
  6 | /******/ 	function __webpack_require__(moduleId) {
  7 | /******/
  8 | /******/ 		// Check if module is in cache
  9 | /******/ 		if(installedModules[moduleId]) {
 10 | /******/ 			return installedModules[moduleId].exports;
 11 | /******/ 		}
 12 | /******/ 		// Create a new module (and put it into the cache)
 13 | /******/ 		var module = installedModules[moduleId] = {
 14 | /******/ 			i: moduleId,
 15 | /******/ 			l: false,
 16 | /******/ 			exports: {}
 17 | /******/ 		};
 18 | /******/
 19 | /******/ 		// Execute the module function
 20 | /******/ 		modules[moduleId].call(module.exports, module, module.exports, __webpack_require__);
 21 | /******/
 22 | /******/ 		// Flag the module as loaded
 23 | /******/ 		module.l = true;
 24 | /******/
 25 | /******/ 		// Return the exports of the module
 26 | /******/ 		return module.exports;
 27 | /******/ 	}
 28 | /******/
 29 | /******/
 30 | /******/ 	// expose the modules object (__webpack_modules__)
 31 | /******/ 	__webpack_require__.m = modules;
 32 | /******/
 33 | /******/ 	// expose the module cache
 34 | /******/ 	__webpack_require__.c = installedModules;
 35 | /******/
 36 | /******/ 	// define getter function for harmony exports
 37 | /******/ 	__webpack_require__.d = function(exports, name, getter) {
 38 | /******/ 		if(!__webpack_require__.o(exports, name)) {
 39 | /******/ 			Object.defineProperty(exports, name, { enumerable: true, get: getter });
 40 | /******/ 		}
 41 | /******/ 	};
 42 | /******/
 43 | /******/ 	// define __esModule on exports
 44 | /******/ 	__webpack_require__.r = function(exports) {
 45 | /******/ 		if(typeof Symbol !== 'undefined' && Symbol.toStringTag) {
 46 | /******/ 			Object.defineProperty(exports, Symbol.toStringTag, { value: 'Module' });
 47 | /******/ 		}
 48 | /******/ 		Object.defineProperty(exports, '__esModule', { value: true });
 49 | /******/ 	};
 50 | /******/
 51 | /******/ 	// create a fake namespace object
 52 | /******/ 	// mode & 1: value is a module id, require it
 53 | /******/ 	// mode & 2: merge all properties of value into the ns
 54 | /******/ 	// mode & 4: return value when already ns object
 55 | /******/ 	// mode & 8|1: behave like require
 56 | /******/ 	__webpack_require__.t = function(value, mode) {
 57 | /******/ 		if(mode & 1) value = __webpack_require__(value);
 58 | /******/ 		if(mode & 8) return value;
 59 | /******/ 		if((mode & 4) && typeof value === 'object' && value && value.__esModule) return value;
 60 | /******/ 		var ns = Object.create(null);
 61 | /******/ 		__webpack_require__.r(ns);
 62 | /******/ 		Object.defineProperty(ns, 'default', { enumerable: true, value: value });
 63 | /******/ 		if(mode & 2 && typeof value != 'string') for(var key in value) __webpack_require__.d(ns, key, function(key) { return value[key]; }.bind(null, key));
 64 | /******/ 		return ns;
 65 | /******/ 	};
 66 | /******/
 67 | /******/ 	// getDefaultExport function for compatibility with non-harmony modules
 68 | /******/ 	__webpack_require__.n = function(module) {
 69 | /******/ 		var getter = module && module.__esModule ?
 70 | /******/ 			function getDefault() { return module['default']; } :
 71 | /******/ 			function getModuleExports() { return module; };
 72 | /******/ 		__webpack_require__.d(getter, 'a', getter);
 73 | /******/ 		return getter;
 74 | /******/ 	};
 75 | /******/
 76 | /******/ 	// Object.prototype.hasOwnProperty.call
 77 | /******/ 	__webpack_require__.o = function(object, property) { return Object.prototype.hasOwnProperty.call(object, property); };
 78 | /******/
 79 | /******/ 	// __webpack_public_path__
 80 | /******/ 	__webpack_require__.p = "";
 81 | /******/
 82 | /******/
 83 | /******/ 	// Load entry module and return exports
 84 | /******/ 	return __webpack_require__(__webpack_require__.s = "./lib/index.js");
 85 | /******/ })
 86 | /************************************************************************/
 87 | /******/ ({
 88 | 
 89 | /***/ "./lib/MetaData.js":
 90 | /*!*************************!*\
 91 |   !*** ./lib/MetaData.js ***!
 92 |   \*************************/
 93 | /*! no static exports found */
 94 | /***/ (function(module, exports, __webpack_require__) {
 95 | 
 96 | "use strict";
 97 | 
 98 | Object.defineProperty(exports, "__esModule", { value: true });
 99 | exports.MetaDataSet = void 0;
100 | var MetaSection_1 = __webpack_require__(/*! ./MetaSection */ "./lib/MetaSection.js");
101 | var MetaDataSet = /** @class */ (function () {
102 |     function MetaDataSet(chart, ds) {
103 |         /** Scales wil be initialized in beforeDraw hook */
104 |         this.getXY = undefined;
105 |         /** Is the dataset's data an array of {x,y}? */
106 |         this.isXY = false;
107 |         var cfg = ds.regressions;
108 |         this.chart = chart;
109 |         this.dataset = ds;
110 |         this.normalizedData = this._normalizeData(ds.data);
111 |         this.sections = this._createMetaSections(cfg);
112 |         this._calculate();
113 |     }
114 |     /**
115 |      * Normalize data to DataPoint[]
116 |      * Only supports number[] and {x:number,y:number}
117 |      */
118 |     MetaDataSet.prototype._normalizeData = function (data) {
119 |         var _this = this;
120 |         return data.map(function (value, index) {
121 |             var p;
122 |             if (typeof value == 'number' || value == null || value === undefined) {
123 |                 p = [index, value];
124 |             }
125 |             else {
126 |                 _this.isXY = true;
127 |                 p = [value.x, value.y];
128 |             }
129 |             return p;
130 |         });
131 |     };
132 |     /** @private */
133 |     MetaDataSet.prototype._createMetaSections = function (cfg) {
134 |         var _this = this;
135 |         var source = cfg.sections || [{ startIndex: 0, endIndex: this.dataset.data.length - 1 }];
136 |         return source.map(function (s) { return new MetaSection_1.MetaSection(s, _this); });
137 |     };
138 |     /** @private */
139 |     MetaDataSet.prototype._calculate = function () {
140 |         this.sections.forEach(function (section) { return section.calculate(); }); // Calculate Section Results
141 |     };
142 |     MetaDataSet.prototype.adjustScales = function () {
143 |         if (this.topY !== undefined)
144 |             return;
145 |         var xScale;
146 |         var yScale;
147 |         var scales = this.chart.scales;
148 |         Object.keys(scales).forEach(function (k) { return k[0] == 'x' && (xScale = scales[k]) || (yScale = scales[k]); });
149 |         this.topY = yScale.top;
150 |         this.bottomY = yScale.bottom;
151 |         this.getXY =
152 |             function (x, y) {
153 |                 return ({
154 |                     x: xScale.getPixelForValue(x, undefined, undefined, true),
155 |                     y: yScale.getPixelForValue(y)
156 |                 });
157 |             };
158 |     };
159 |     MetaDataSet.prototype.draw = function () {
160 |         var ctx = this.chart.chart.ctx;
161 |         ctx.save();
162 |         try {
163 |             this.sections.forEach(function (section) { return section.drawRegressions(ctx); });
164 |         }
165 |         finally {
166 |             ctx.restore();
167 |         }
168 |     };
169 |     return MetaDataSet;
170 | }());
171 | exports.MetaDataSet = MetaDataSet;
172 | //# sourceMappingURL=MetaData.js.map
173 | 
174 | /***/ }),
175 | 
176 | /***/ "./lib/MetaSection.js":
177 | /*!****************************!*\
178 |   !*** ./lib/MetaSection.js ***!
179 |   \****************************/
180 | /*! no static exports found */
181 | /***/ (function(module, exports, __webpack_require__) {
182 | 
183 | "use strict";
184 | 
185 | Object.defineProperty(exports, "__esModule", { value: true });
186 | exports.MetaSection = void 0;
187 | var regression = __webpack_require__(/*! regression */ "./node_modules/regression/dist/regression.js");
188 | var defaultConfig = {
189 |     type: "linear",
190 |     calculation: {
191 |         precision: 2,
192 |         order: 2
193 |     },
194 |     line: {
195 |         width: 2,
196 |         color: '#000',
197 |         dash: []
198 |     },
199 |     extendPredictions: false,
200 |     copy: {
201 |         overwriteData: 'none'
202 |     }
203 | };
204 | var MetaSection = /** @class */ (function () {
205 |     function MetaSection(sec, _meta) {
206 |         this._meta = _meta;
207 |         var chart = _meta.chart;
208 |         var ds = _meta.dataset;
209 |         var cfg = getConfig(['type', 'calculation', 'line', 'extendPredictions', 'copy']);
210 |         this.startIndex = sec.startIndex || 0;
211 |         this.endIndex = sec.endIndex || ds.data.length - 1;
212 |         this.type = Array.isArray(cfg.type) ? cfg.type : [cfg.type];
213 |         this.line = cfg.line;
214 |         this.calculation = cfg.calculation;
215 |         this.extendPredictions = cfg.extendPredictions;
216 |         this.copy = cfg.copy;
217 |         this.label = sec.label || this._meta.chart.data.labels[this.endIndex];
218 |         this._validateType();
219 |         // --- constructor helpers
220 |         /**
221 |          * Calculate the inherited configuration from defaultConfig, globalConfig,
222 |          * dataset config, and section config (in that order)
223 |          */
224 |         function getConfig(fields) {
225 |             var o, p;
226 |             var globalConfig = (o = chart.config.options) && (p = o.plugins) && p.regressions || {};
227 |             return configMerge(fields, defaultConfig, globalConfig, ds.regressions, sec);
228 |             /** merge the config objects */
229 |             function configMerge(fields) {
230 |                 var cfgList = [];
231 |                 for (var _i = 1; _i < arguments.length; _i++) {
232 |                     cfgList[_i - 1] = arguments[_i];
233 |                 }
234 |                 var dstConfig = {};
235 |                 fields.forEach(function (f) {
236 |                     cfgList.forEach(function (srcConfig) {
237 |                         var o = srcConfig[f];
238 |                         var t = typeof o;
239 |                         if (t != 'undefined') {
240 |                             if (Array.isArray(o) || t != 'object' || o == null)
241 |                                 dstConfig[f] = o;
242 |                             else
243 |                                 dstConfig[f] = Object.assign({}, dstConfig[f], configMerge(Object.keys(o), o));
244 |                         }
245 |                     });
246 |                 });
247 |                 return dstConfig;
248 |             }
249 |         }
250 |     }
251 |     /** Validates the type to avoid inconsistences */
252 |     MetaSection.prototype._validateType = function () {
253 |         if (this.type.length > 1 && this.type.includes('copy'))
254 |             throw Error('Invalid regression type:' + this.type + '. "none" cannot be combined with other type!');
255 |     };
256 |     /** Calculates the regression(s) and sets the result objects */
257 |     MetaSection.prototype.calculate = function () {
258 |         var sectionData = this._meta.normalizedData.slice(this.startIndex, this.endIndex + 1);
259 |         if (this.type[0] == 'copy')
260 |             this._calculateCopySection(sectionData);
261 |         else
262 |             this._calculateBestR2(sectionData);
263 |     };
264 |     MetaSection.prototype._calculateBestR2 = function (sectionData) {
265 |         var _this = this;
266 |         this.result = this.type.reduce(function (max, type) {
267 |             var calculation = Object.assign({}, _this.calculation);
268 |             var realType = type;
269 |             if (/polynomial[34]$/.test(type)) {
270 |                 calculation.order = parseInt(type.substr(10));
271 |                 realType = type.substr(0, 10);
272 |             }
273 |             var r = regression[realType](sectionData, calculation);
274 |             r.type = type;
275 |             return (!max || max.r2 < r.r2) ? r : max;
276 |         }, null);
277 |     };
278 |     MetaSection.prototype._calculateCopySection = function (sectionData) {
279 |         var _this = this;
280 |         var from = this._meta.sections[this.copy.fromSectionIndex], r = this.result = Object.assign({}, from.result), overwrite = this.copy.overwriteData, data = this._meta.normalizedData;
281 |         r.points = sectionData.map(function (p) { return r.predict(p[0]); });
282 |         delete r.r2;
283 |         if (overwrite != 'none') {
284 |             var dsdata_1 = this._meta.dataset.data, isXY_1 = this._meta.isXY;
285 |             r.points.forEach(function (_a, i) {
286 |                 var x = _a[0], y = _a[1];
287 |                 var index = i + _this.startIndex;
288 |                 if ((index < from.startIndex || index > from.endIndex) &&
289 |                     (overwrite == 'all' ||
290 |                         overwrite == 'last' && index == _this.endIndex ||
291 |                         overwrite == 'empty' && !data[index])) {
292 |                     if (_this.copy.maxValue)
293 |                         y = Math.min(_this.copy.maxValue, y);
294 |                     if (_this.copy.minValue !== undefined)
295 |                         y = Math.max(_this.copy.minValue, y);
296 |                     dsdata_1[index] = isXY_1 ? { x: x, y: y } : y;
297 |                 }
298 |             });
299 |         }
300 |     };
301 |     MetaSection.prototype.drawRightBorder = function (ctx) {
302 |         ctx.beginPath();
303 |         this._setLineAttrs(ctx);
304 |         ctx.setLineDash([10, 2]);
305 |         ctx.lineWidth = 2;
306 |         // Print vertical line
307 |         var p = this._meta.getXY(this.endIndex, 0);
308 |         ctx.moveTo(p.x, this._meta.topY);
309 |         ctx.lineTo(p.x, this._meta.bottomY);
310 |         ctx.fillStyle = this.line.color;
311 |         ctx.fillText(this.label, p.x, this._meta.topY);
312 |         ctx.stroke();
313 |     };
314 |     MetaSection.prototype.drawRegressions = function (ctx) {
315 |         for (var i = 0, len = this._meta.sections.length; i < len; i++) {
316 |             var section = this._meta.sections[i];
317 |             var isMe = section == this;
318 |             if (isMe && this.type[0] != 'copy' || !isMe && this.extendPredictions) {
319 |                 section.drawRange(ctx, this.startIndex, this.endIndex, !isMe);
320 |             }
321 |             if (isMe)
322 |                 break;
323 |         }
324 |     };
325 |     MetaSection.prototype.drawRange = function (ctx, startIndex, endIndex, forceDash) {
326 |         var _this = this;
327 |         ctx.beginPath();
328 |         this._setLineAttrs(ctx);
329 |         if (forceDash)
330 |             ctx.setLineDash([5, 5]);
331 |         var predict = this.result.predict;
332 |         var f = function (x) { return _this._meta.getXY(x, predict(x)[1]); };
333 |         var p = f(startIndex);
334 |         ctx.moveTo(p.x, p.y);
335 |         for (var x = startIndex + 1; x <= endIndex; x++) {
336 |             p = f(x);
337 |             ctx.lineTo(p.x, p.y);
338 |         }
339 |         ctx.stroke();
340 |     };
341 |     MetaSection.prototype._setLineAttrs = function (ctx) {
342 |         if (this.line.width)
343 |             ctx.lineWidth = this.line.width;
344 |         if (this.line.color)
345 |             ctx.strokeStyle = this.line.color;
346 |         if (this.line.dash)
347 |             ctx.setLineDash(this.line.dash);
348 |     };
349 |     return MetaSection;
350 | }());
351 | exports.MetaSection = MetaSection;
352 | //# sourceMappingURL=MetaSection.js.map
353 | 
354 | /***/ }),
355 | 
356 | /***/ "./lib/index.js":
357 | /*!**********************!*\
358 |   !*** ./lib/index.js ***!
359 |   \**********************/
360 | /*! no static exports found */
361 | /***/ (function(module, exports, __webpack_require__) {
362 | 
363 | "use strict";
364 | 
365 | var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
366 |     if (k2 === undefined) k2 = k;
367 |     Object.defineProperty(o, k2, { enumerable: true, get: function() { return m[k]; } });
368 | }) : (function(o, m, k, k2) {
369 |     if (k2 === undefined) k2 = k;
370 |     o[k2] = m[k];
371 | }));
372 | var __exportStar = (this && this.__exportStar) || function(m, exports) {
373 |     for (var p in m) if (p !== "default" && !exports.hasOwnProperty(p)) __createBinding(exports, m, p);
374 | };
375 | Object.defineProperty(exports, "__esModule", { value: true });
376 | __exportStar(__webpack_require__(/*! ./types */ "./lib/types.js"), exports);
377 | __exportStar(__webpack_require__(/*! ./MetaData */ "./lib/MetaData.js"), exports);
378 | __exportStar(__webpack_require__(/*! ./MetaSection */ "./lib/MetaSection.js"), exports);
379 | __exportStar(__webpack_require__(/*! ./regression-plugin */ "./lib/regression-plugin.js"), exports);
380 | //# sourceMappingURL=index.js.map
381 | 
382 | /***/ }),
383 | 
384 | /***/ "./lib/regression-plugin.js":
385 | /*!**********************************!*\
386 |   !*** ./lib/regression-plugin.js ***!
387 |   \**********************************/
388 | /*! no static exports found */
389 | /***/ (function(module, exports, __webpack_require__) {
390 | 
391 | "use strict";
392 | 
393 | Object.defineProperty(exports, "__esModule", { value: true });
394 | exports.ChartRegressions = void 0;
395 | var MetaData_1 = __webpack_require__(/*! ./MetaData */ "./lib/MetaData.js");
396 | // Cache for all plugins' metadata
397 | var _metadataMap = {};
398 | var _chartId = 0;
399 | var Plugin = /** @class */ (function () {
400 |     function Plugin() {
401 |         this.id = 'regressions';
402 |     }
403 |     Plugin.prototype.beforeInit = function (chart) {
404 |         chart.$$id = ++_chartId;
405 |     };
406 |     /**
407 |      * Called after update (when the chart is created and when chart.update() is called)
408 |      * @param chart
409 |      */
410 |     Plugin.prototype.beforeUpdate = function (chart, options) {
411 |         var o, p, r;
412 |         var onComplete = (o = chart.config.options) && (p = o.plugins)
413 |             && (r = p.regressions) && r.onCompleteCalculation;
414 |         forEach(chart, function (ds, meta, datasetIndex) {
415 |             meta = new MetaData_1.MetaDataSet(chart, ds);
416 |             var id = chart.$$id * 1000 + datasetIndex;
417 |             _metadataMap[id] = meta;
418 |         });
419 |         if (onComplete)
420 |             onComplete(chart);
421 |     };
422 |     /**
423 |      * It's called once before all the drawing
424 |      * @param chart
425 |      */
426 |     Plugin.prototype.beforeRender = function (chart, options) {
427 |         forEach(chart, function (ds, meta) { return meta.adjustScales(); });
428 |     };
429 |     /** Draws the vertical lines before the datasets are drawn */
430 |     Plugin.prototype.beforeDatasetsDraw = function (chart, easing, options) {
431 |         forEach(chart, function (ds, meta) {
432 |             var ctx = chart.ctx;
433 |             ctx.save();
434 |             try {
435 |                 for (var i = 0; i < meta.sections.length - 1; i++)
436 |                     meta.sections[i].drawRightBorder(ctx);
437 |             }
438 |             finally {
439 |                 ctx.restore();
440 |             }
441 |             ;
442 |         });
443 |     };
444 |     /** Draws the regression lines */
445 |     Plugin.prototype.afterDatasetsDraw = function (chart, easing, options) {
446 |         forEach(chart, function (ds, meta) { return meta.draw(); });
447 |     };
448 |     Plugin.prototype.destroy = function (chart) {
449 |         Object.keys(_metadataMap)
450 |             .filter(function (k) { return (k / 1000) >> 0 == chart.$$id; })
451 |             .forEach(function (k) { return delete _metadataMap[k]; });
452 |     };
453 |     /** Get dataset's meta data */
454 |     Plugin.prototype.getDataset = function (chart, datasetIndex) {
455 |         var id = chart.$$id * 1000 + datasetIndex;
456 |         return _metadataMap[id];
457 |     };
458 |     /** Get dataset's meta sections */
459 |     Plugin.prototype.getSections = function (chart, datasetIndex) {
460 |         var ds = this.getDataset(chart, datasetIndex);
461 |         return ds && ds.sections;
462 |     };
463 |     return Plugin;
464 | }());
465 | function forEach(chart, fn) {
466 |     chart.data.datasets.forEach(function (ds, i) {
467 |         if (ds.regressions && chart.isDatasetVisible(i)) {
468 |             var meta = exports.ChartRegressions.getDataset(chart, i);
469 |             fn(ds, meta, i);
470 |         }
471 |     });
472 | }
473 | exports.ChartRegressions = new Plugin();
474 | window.ChartRegressions = exports.ChartRegressions;
475 | //# sourceMappingURL=regression-plugin.js.map
476 | 
477 | /***/ }),
478 | 
479 | /***/ "./lib/types.js":
480 | /*!**********************!*\
481 |   !*** ./lib/types.js ***!
482 |   \**********************/
483 | /*! no static exports found */
484 | /***/ (function(module, exports, __webpack_require__) {
485 | 
486 | "use strict";
487 | 
488 | Object.defineProperty(exports, "__esModule", { value: true });
489 | //# sourceMappingURL=types.js.map
490 | 
491 | /***/ }),
492 | 
493 | /***/ "./node_modules/regression/dist/regression.js":
494 | /*!****************************************************!*\
495 |   !*** ./node_modules/regression/dist/regression.js ***!
496 |   \****************************************************/
497 | /*! no static exports found */
498 | /***/ (function(module, exports, __webpack_require__) {
499 | 
500 | var __WEBPACK_AMD_DEFINE_FACTORY__, __WEBPACK_AMD_DEFINE_ARRAY__, __WEBPACK_AMD_DEFINE_RESULT__;(function (global, factory) {
501 |   if (true) {
502 |     !(__WEBPACK_AMD_DEFINE_ARRAY__ = [module], __WEBPACK_AMD_DEFINE_FACTORY__ = (factory),
503 | 				__WEBPACK_AMD_DEFINE_RESULT__ = (typeof __WEBPACK_AMD_DEFINE_FACTORY__ === 'function' ?
504 | 				(__WEBPACK_AMD_DEFINE_FACTORY__.apply(exports, __WEBPACK_AMD_DEFINE_ARRAY__)) : __WEBPACK_AMD_DEFINE_FACTORY__),
505 | 				__WEBPACK_AMD_DEFINE_RESULT__ !== undefined && (module.exports = __WEBPACK_AMD_DEFINE_RESULT__));
506 |   } else { var mod; }
507 | })(this, function (module) {
508 |   'use strict';
509 | 
510 |   function _defineProperty(obj, key, value) {
511 |     if (key in obj) {
512 |       Object.defineProperty(obj, key, {
513 |         value: value,
514 |         enumerable: true,
515 |         configurable: true,
516 |         writable: true
517 |       });
518 |     } else {
519 |       obj[key] = value;
520 |     }
521 | 
522 |     return obj;
523 |   }
524 | 
525 |   var _extends = Object.assign || function (target) {
526 |     for (var i = 1; i < arguments.length; i++) {
527 |       var source = arguments[i];
528 | 
529 |       for (var key in source) {
530 |         if (Object.prototype.hasOwnProperty.call(source, key)) {
531 |           target[key] = source[key];
532 |         }
533 |       }
534 |     }
535 | 
536 |     return target;
537 |   };
538 | 
539 |   function _toConsumableArray(arr) {
540 |     if (Array.isArray(arr)) {
541 |       for (var i = 0, arr2 = Array(arr.length); i < arr.length; i++) {
542 |         arr2[i] = arr[i];
543 |       }
544 | 
545 |       return arr2;
546 |     } else {
547 |       return Array.from(arr);
548 |     }
549 |   }
550 | 
551 |   var DEFAULT_OPTIONS = { order: 2, precision: 2, period: null };
552 | 
553 |   /**
554 |   * Determine the coefficient of determination (r^2) of a fit from the observations
555 |   * and predictions.
556 |   *
557 |   * @param {Array>} data - Pairs of observed x-y values
558 |   * @param {Array>} results - Pairs of observed predicted x-y values
559 |   *
560 |   * @return {number} - The r^2 value, or NaN if one cannot be calculated.
561 |   */
562 |   function determinationCoefficient(data, results) {
563 |     var predictions = [];
564 |     var observations = [];
565 | 
566 |     data.forEach(function (d, i) {
567 |       if (d[1] !== null) {
568 |         observations.push(d);
569 |         predictions.push(results[i]);
570 |       }
571 |     });
572 | 
573 |     var sum = observations.reduce(function (a, observation) {
574 |       return a + observation[1];
575 |     }, 0);
576 |     var mean = sum / observations.length;
577 | 
578 |     var ssyy = observations.reduce(function (a, observation) {
579 |       var difference = observation[1] - mean;
580 |       return a + difference * difference;
581 |     }, 0);
582 | 
583 |     var sse = observations.reduce(function (accum, observation, index) {
584 |       var prediction = predictions[index];
585 |       var residual = observation[1] - prediction[1];
586 |       return accum + residual * residual;
587 |     }, 0);
588 | 
589 |     return 1 - sse / ssyy;
590 |   }
591 | 
592 |   /**
593 |   * Determine the solution of a system of linear equations A * x = b using
594 |   * Gaussian elimination.
595 |   *
596 |   * @param {Array>} input - A 2-d matrix of data in row-major form [ A | b ]
597 |   * @param {number} order - How many degrees to solve for
598 |   *
599 |   * @return {Array} - Vector of normalized solution coefficients matrix (x)
600 |   */
601 |   function gaussianElimination(input, order) {
602 |     var matrix = input;
603 |     var n = input.length - 1;
604 |     var coefficients = [order];
605 | 
606 |     for (var i = 0; i < n; i++) {
607 |       var maxrow = i;
608 |       for (var j = i + 1; j < n; j++) {
609 |         if (Math.abs(matrix[i][j]) > Math.abs(matrix[i][maxrow])) {
610 |           maxrow = j;
611 |         }
612 |       }
613 | 
614 |       for (var k = i; k < n + 1; k++) {
615 |         var tmp = matrix[k][i];
616 |         matrix[k][i] = matrix[k][maxrow];
617 |         matrix[k][maxrow] = tmp;
618 |       }
619 | 
620 |       for (var _j = i + 1; _j < n; _j++) {
621 |         for (var _k = n; _k >= i; _k--) {
622 |           matrix[_k][_j] -= matrix[_k][i] * matrix[i][_j] / matrix[i][i];
623 |         }
624 |       }
625 |     }
626 | 
627 |     for (var _j2 = n - 1; _j2 >= 0; _j2--) {
628 |       var total = 0;
629 |       for (var _k2 = _j2 + 1; _k2 < n; _k2++) {
630 |         total += matrix[_k2][_j2] * coefficients[_k2];
631 |       }
632 | 
633 |       coefficients[_j2] = (matrix[n][_j2] - total) / matrix[_j2][_j2];
634 |     }
635 | 
636 |     return coefficients;
637 |   }
638 | 
639 |   /**
640 |   * Round a number to a precision, specificed in number of decimal places
641 |   *
642 |   * @param {number} number - The number to round
643 |   * @param {number} precision - The number of decimal places to round to:
644 |   *                             > 0 means decimals, < 0 means powers of 10
645 |   *
646 |   *
647 |   * @return {numbr} - The number, rounded
648 |   */
649 |   function round(number, precision) {
650 |     var factor = Math.pow(10, precision);
651 |     return Math.round(number * factor) / factor;
652 |   }
653 | 
654 |   /**
655 |   * The set of all fitting methods
656 |   *
657 |   * @namespace
658 |   */
659 |   var methods = {
660 |     linear: function linear(data, options) {
661 |       var sum = [0, 0, 0, 0, 0];
662 |       var len = 0;
663 | 
664 |       for (var n = 0; n < data.length; n++) {
665 |         if (data[n][1] !== null) {
666 |           len++;
667 |           sum[0] += data[n][0];
668 |           sum[1] += data[n][1];
669 |           sum[2] += data[n][0] * data[n][0];
670 |           sum[3] += data[n][0] * data[n][1];
671 |           sum[4] += data[n][1] * data[n][1];
672 |         }
673 |       }
674 | 
675 |       var run = len * sum[2] - sum[0] * sum[0];
676 |       var rise = len * sum[3] - sum[0] * sum[1];
677 |       var gradient = run === 0 ? 0 : round(rise / run, options.precision);
678 |       var intercept = round(sum[1] / len - gradient * sum[0] / len, options.precision);
679 | 
680 |       var predict = function predict(x) {
681 |         return [round(x, options.precision), round(gradient * x + intercept, options.precision)];
682 |       };
683 | 
684 |       var points = data.map(function (point) {
685 |         return predict(point[0]);
686 |       });
687 | 
688 |       return {
689 |         points: points,
690 |         predict: predict,
691 |         equation: [gradient, intercept],
692 |         r2: round(determinationCoefficient(data, points), options.precision),
693 |         string: intercept === 0 ? 'y = ' + gradient + 'x' : 'y = ' + gradient + 'x + ' + intercept
694 |       };
695 |     },
696 |     exponential: function exponential(data, options) {
697 |       var sum = [0, 0, 0, 0, 0, 0];
698 | 
699 |       for (var n = 0; n < data.length; n++) {
700 |         if (data[n][1] !== null) {
701 |           sum[0] += data[n][0];
702 |           sum[1] += data[n][1];
703 |           sum[2] += data[n][0] * data[n][0] * data[n][1];
704 |           sum[3] += data[n][1] * Math.log(data[n][1]);
705 |           sum[4] += data[n][0] * data[n][1] * Math.log(data[n][1]);
706 |           sum[5] += data[n][0] * data[n][1];
707 |         }
708 |       }
709 | 
710 |       var denominator = sum[1] * sum[2] - sum[5] * sum[5];
711 |       var a = Math.exp((sum[2] * sum[3] - sum[5] * sum[4]) / denominator);
712 |       var b = (sum[1] * sum[4] - sum[5] * sum[3]) / denominator;
713 |       var coeffA = round(a, options.precision);
714 |       var coeffB = round(b, options.precision);
715 |       var predict = function predict(x) {
716 |         return [round(x, options.precision), round(coeffA * Math.exp(coeffB * x), options.precision)];
717 |       };
718 | 
719 |       var points = data.map(function (point) {
720 |         return predict(point[0]);
721 |       });
722 | 
723 |       return {
724 |         points: points,
725 |         predict: predict,
726 |         equation: [coeffA, coeffB],
727 |         string: 'y = ' + coeffA + 'e^(' + coeffB + 'x)',
728 |         r2: round(determinationCoefficient(data, points), options.precision)
729 |       };
730 |     },
731 |     logarithmic: function logarithmic(data, options) {
732 |       var sum = [0, 0, 0, 0];
733 |       var len = data.length;
734 | 
735 |       for (var n = 0; n < len; n++) {
736 |         if (data[n][1] !== null) {
737 |           sum[0] += Math.log(data[n][0]);
738 |           sum[1] += data[n][1] * Math.log(data[n][0]);
739 |           sum[2] += data[n][1];
740 |           sum[3] += Math.pow(Math.log(data[n][0]), 2);
741 |         }
742 |       }
743 | 
744 |       var a = (len * sum[1] - sum[2] * sum[0]) / (len * sum[3] - sum[0] * sum[0]);
745 |       var coeffB = round(a, options.precision);
746 |       var coeffA = round((sum[2] - coeffB * sum[0]) / len, options.precision);
747 | 
748 |       var predict = function predict(x) {
749 |         return [round(x, options.precision), round(round(coeffA + coeffB * Math.log(x), options.precision), options.precision)];
750 |       };
751 | 
752 |       var points = data.map(function (point) {
753 |         return predict(point[0]);
754 |       });
755 | 
756 |       return {
757 |         points: points,
758 |         predict: predict,
759 |         equation: [coeffA, coeffB],
760 |         string: 'y = ' + coeffA + ' + ' + coeffB + ' ln(x)',
761 |         r2: round(determinationCoefficient(data, points), options.precision)
762 |       };
763 |     },
764 |     power: function power(data, options) {
765 |       var sum = [0, 0, 0, 0, 0];
766 |       var len = data.length;
767 | 
768 |       for (var n = 0; n < len; n++) {
769 |         if (data[n][1] !== null) {
770 |           sum[0] += Math.log(data[n][0]);
771 |           sum[1] += Math.log(data[n][1]) * Math.log(data[n][0]);
772 |           sum[2] += Math.log(data[n][1]);
773 |           sum[3] += Math.pow(Math.log(data[n][0]), 2);
774 |         }
775 |       }
776 | 
777 |       var b = (len * sum[1] - sum[0] * sum[2]) / (len * sum[3] - Math.pow(sum[0], 2));
778 |       var a = (sum[2] - b * sum[0]) / len;
779 |       var coeffA = round(Math.exp(a), options.precision);
780 |       var coeffB = round(b, options.precision);
781 | 
782 |       var predict = function predict(x) {
783 |         return [round(x, options.precision), round(round(coeffA * Math.pow(x, coeffB), options.precision), options.precision)];
784 |       };
785 | 
786 |       var points = data.map(function (point) {
787 |         return predict(point[0]);
788 |       });
789 | 
790 |       return {
791 |         points: points,
792 |         predict: predict,
793 |         equation: [coeffA, coeffB],
794 |         string: 'y = ' + coeffA + 'x^' + coeffB,
795 |         r2: round(determinationCoefficient(data, points), options.precision)
796 |       };
797 |     },
798 |     polynomial: function polynomial(data, options) {
799 |       var lhs = [];
800 |       var rhs = [];
801 |       var a = 0;
802 |       var b = 0;
803 |       var len = data.length;
804 |       var k = options.order + 1;
805 | 
806 |       for (var i = 0; i < k; i++) {
807 |         for (var l = 0; l < len; l++) {
808 |           if (data[l][1] !== null) {
809 |             a += Math.pow(data[l][0], i) * data[l][1];
810 |           }
811 |         }
812 | 
813 |         lhs.push(a);
814 |         a = 0;
815 | 
816 |         var c = [];
817 |         for (var j = 0; j < k; j++) {
818 |           for (var _l = 0; _l < len; _l++) {
819 |             if (data[_l][1] !== null) {
820 |               b += Math.pow(data[_l][0], i + j);
821 |             }
822 |           }
823 |           c.push(b);
824 |           b = 0;
825 |         }
826 |         rhs.push(c);
827 |       }
828 |       rhs.push(lhs);
829 | 
830 |       var coefficients = gaussianElimination(rhs, k).map(function (v) {
831 |         return round(v, options.precision);
832 |       });
833 | 
834 |       var predict = function predict(x) {
835 |         return [round(x, options.precision), round(coefficients.reduce(function (sum, coeff, power) {
836 |           return sum + coeff * Math.pow(x, power);
837 |         }, 0), options.precision)];
838 |       };
839 | 
840 |       var points = data.map(function (point) {
841 |         return predict(point[0]);
842 |       });
843 | 
844 |       var string = 'y = ';
845 |       for (var _i = coefficients.length - 1; _i >= 0; _i--) {
846 |         if (_i > 1) {
847 |           string += coefficients[_i] + 'x^' + _i + ' + ';
848 |         } else if (_i === 1) {
849 |           string += coefficients[_i] + 'x + ';
850 |         } else {
851 |           string += coefficients[_i];
852 |         }
853 |       }
854 | 
855 |       return {
856 |         string: string,
857 |         points: points,
858 |         predict: predict,
859 |         equation: [].concat(_toConsumableArray(coefficients)).reverse(),
860 |         r2: round(determinationCoefficient(data, points), options.precision)
861 |       };
862 |     }
863 |   };
864 | 
865 |   function createWrapper() {
866 |     var reduce = function reduce(accumulator, name) {
867 |       return _extends({
868 |         _round: round
869 |       }, accumulator, _defineProperty({}, name, function (data, supplied) {
870 |         return methods[name](data, _extends({}, DEFAULT_OPTIONS, supplied));
871 |       }));
872 |     };
873 | 
874 |     return Object.keys(methods).reduce(reduce, {});
875 |   }
876 | 
877 |   module.exports = createWrapper();
878 | });
879 | 
880 | 
881 | /***/ })
882 | 
883 | /******/ });
884 | //# sourceMappingURL=chartjs-plugin-regression-0.1.1.js.map


--------------------------------------------------------------------------------
/dist/chartjs-plugin-regression-0.2.0.js:
--------------------------------------------------------------------------------
  1 | /******/ (function(modules) { // webpackBootstrap
  2 | /******/ 	// The module cache
  3 | /******/ 	var installedModules = {};
  4 | /******/
  5 | /******/ 	// The require function
  6 | /******/ 	function __webpack_require__(moduleId) {
  7 | /******/
  8 | /******/ 		// Check if module is in cache
  9 | /******/ 		if(installedModules[moduleId]) {
 10 | /******/ 			return installedModules[moduleId].exports;
 11 | /******/ 		}
 12 | /******/ 		// Create a new module (and put it into the cache)
 13 | /******/ 		var module = installedModules[moduleId] = {
 14 | /******/ 			i: moduleId,
 15 | /******/ 			l: false,
 16 | /******/ 			exports: {}
 17 | /******/ 		};
 18 | /******/
 19 | /******/ 		// Execute the module function
 20 | /******/ 		modules[moduleId].call(module.exports, module, module.exports, __webpack_require__);
 21 | /******/
 22 | /******/ 		// Flag the module as loaded
 23 | /******/ 		module.l = true;
 24 | /******/
 25 | /******/ 		// Return the exports of the module
 26 | /******/ 		return module.exports;
 27 | /******/ 	}
 28 | /******/
 29 | /******/
 30 | /******/ 	// expose the modules object (__webpack_modules__)
 31 | /******/ 	__webpack_require__.m = modules;
 32 | /******/
 33 | /******/ 	// expose the module cache
 34 | /******/ 	__webpack_require__.c = installedModules;
 35 | /******/
 36 | /******/ 	// define getter function for harmony exports
 37 | /******/ 	__webpack_require__.d = function(exports, name, getter) {
 38 | /******/ 		if(!__webpack_require__.o(exports, name)) {
 39 | /******/ 			Object.defineProperty(exports, name, { enumerable: true, get: getter });
 40 | /******/ 		}
 41 | /******/ 	};
 42 | /******/
 43 | /******/ 	// define __esModule on exports
 44 | /******/ 	__webpack_require__.r = function(exports) {
 45 | /******/ 		if(typeof Symbol !== 'undefined' && Symbol.toStringTag) {
 46 | /******/ 			Object.defineProperty(exports, Symbol.toStringTag, { value: 'Module' });
 47 | /******/ 		}
 48 | /******/ 		Object.defineProperty(exports, '__esModule', { value: true });
 49 | /******/ 	};
 50 | /******/
 51 | /******/ 	// create a fake namespace object
 52 | /******/ 	// mode & 1: value is a module id, require it
 53 | /******/ 	// mode & 2: merge all properties of value into the ns
 54 | /******/ 	// mode & 4: return value when already ns object
 55 | /******/ 	// mode & 8|1: behave like require
 56 | /******/ 	__webpack_require__.t = function(value, mode) {
 57 | /******/ 		if(mode & 1) value = __webpack_require__(value);
 58 | /******/ 		if(mode & 8) return value;
 59 | /******/ 		if((mode & 4) && typeof value === 'object' && value && value.__esModule) return value;
 60 | /******/ 		var ns = Object.create(null);
 61 | /******/ 		__webpack_require__.r(ns);
 62 | /******/ 		Object.defineProperty(ns, 'default', { enumerable: true, value: value });
 63 | /******/ 		if(mode & 2 && typeof value != 'string') for(var key in value) __webpack_require__.d(ns, key, function(key) { return value[key]; }.bind(null, key));
 64 | /******/ 		return ns;
 65 | /******/ 	};
 66 | /******/
 67 | /******/ 	// getDefaultExport function for compatibility with non-harmony modules
 68 | /******/ 	__webpack_require__.n = function(module) {
 69 | /******/ 		var getter = module && module.__esModule ?
 70 | /******/ 			function getDefault() { return module['default']; } :
 71 | /******/ 			function getModuleExports() { return module; };
 72 | /******/ 		__webpack_require__.d(getter, 'a', getter);
 73 | /******/ 		return getter;
 74 | /******/ 	};
 75 | /******/
 76 | /******/ 	// Object.prototype.hasOwnProperty.call
 77 | /******/ 	__webpack_require__.o = function(object, property) { return Object.prototype.hasOwnProperty.call(object, property); };
 78 | /******/
 79 | /******/ 	// __webpack_public_path__
 80 | /******/ 	__webpack_require__.p = "";
 81 | /******/
 82 | /******/
 83 | /******/ 	// Load entry module and return exports
 84 | /******/ 	return __webpack_require__(__webpack_require__.s = "./lib/index.js");
 85 | /******/ })
 86 | /************************************************************************/
 87 | /******/ ({
 88 | 
 89 | /***/ "./lib/MetaData.js":
 90 | /*!*************************!*\
 91 |   !*** ./lib/MetaData.js ***!
 92 |   \*************************/
 93 | /*! no static exports found */
 94 | /***/ (function(module, exports, __webpack_require__) {
 95 | 
 96 | "use strict";
 97 | 
 98 | Object.defineProperty(exports, "__esModule", { value: true });
 99 | exports.MetaDataSet = void 0;
100 | var MetaSection_1 = __webpack_require__(/*! ./MetaSection */ "./lib/MetaSection.js");
101 | var MetaDataSet = /** @class */ (function () {
102 |     function MetaDataSet(chart, ds) {
103 |         /** Scales wil be initialized in beforeDraw hook */
104 |         this.getXY = undefined;
105 |         /** Is the dataset's data an array of {x,y}? */
106 |         this.isXY = false;
107 |         var cfg = ds.regressions;
108 |         this.chart = chart;
109 |         this.dataset = ds;
110 |         this.normalizedData = this._normalizeData(ds.data);
111 |         this.sections = this._createMetaSections(cfg);
112 |         this._calculate();
113 |     }
114 |     /**
115 |      * Normalize data to DataPoint[]
116 |      * Only supports number[] and {x:number,y:number}
117 |      */
118 |     MetaDataSet.prototype._normalizeData = function (data) {
119 |         var _this = this;
120 |         return data.map(function (value, index) {
121 |             var p;
122 |             if (typeof value == 'number' || value == null || value === undefined) {
123 |                 p = [index, value];
124 |             }
125 |             else {
126 |                 _this.isXY = true;
127 |                 p = [value.x, value.y];
128 |             }
129 |             return p;
130 |         });
131 |     };
132 |     /** @private */
133 |     MetaDataSet.prototype._createMetaSections = function (cfg) {
134 |         var _this = this;
135 |         var source = cfg.sections || [
136 |             { startIndex: 0, endIndex: this.dataset.data.length - 1 }
137 |         ];
138 |         return source.map(function (s) { return new MetaSection_1.MetaSection(s, _this); });
139 |     };
140 |     /** @private */
141 |     MetaDataSet.prototype._calculate = function () {
142 |         this.sections.forEach(function (section) { return section.calculate(); }); // Calculate Section Results
143 |     };
144 |     MetaDataSet.prototype.adjustScales = function () {
145 |         if (this.topY !== undefined)
146 |             return;
147 |         var xScale;
148 |         var yScale;
149 |         var scales = this.chart.scales;
150 |         Object.keys(scales).forEach(function (k) { return (k[0] == 'x' && (xScale = scales[k])) || (yScale = scales[k]); });
151 |         this.topY = yScale.top;
152 |         this.bottomY = yScale.bottom;
153 |         this.getXY = function (x, y) { return ({
154 |             x: xScale.getPixelForValue(x, undefined, undefined, true),
155 |             y: yScale.getPixelForValue(y)
156 |         }); };
157 |     };
158 |     MetaDataSet.prototype.drawRegressions = function () {
159 |         var ctx = this.chart.chart.ctx;
160 |         ctx.save();
161 |         try {
162 |             this.sections.forEach(function (section) { return section.drawRegressions(ctx); });
163 |         }
164 |         finally {
165 |             ctx.restore();
166 |         }
167 |     };
168 |     MetaDataSet.prototype.drawRightBorders = function () {
169 |         var ctx = this.chart.chart.ctx;
170 |         ctx.save();
171 |         try {
172 |             for (var i = 0; i < this.sections.length - 1; i++)
173 |                 this.sections[i].drawRightBorder(ctx);
174 |         }
175 |         finally {
176 |             ctx.restore();
177 |         }
178 |     };
179 |     return MetaDataSet;
180 | }());
181 | exports.MetaDataSet = MetaDataSet;
182 | //# sourceMappingURL=MetaData.js.map
183 | 
184 | /***/ }),
185 | 
186 | /***/ "./lib/MetaSection.js":
187 | /*!****************************!*\
188 |   !*** ./lib/MetaSection.js ***!
189 |   \****************************/
190 | /*! no static exports found */
191 | /***/ (function(module, exports, __webpack_require__) {
192 | 
193 | "use strict";
194 | 
195 | Object.defineProperty(exports, "__esModule", { value: true });
196 | exports.MetaSection = void 0;
197 | var regression = __webpack_require__(/*! regression */ "./node_modules/regression/dist/regression.js");
198 | var defaultConfig = {
199 |     type: 'linear',
200 |     calculation: {
201 |         precision: 2,
202 |         order: 2
203 |     },
204 |     line: {
205 |         width: 2,
206 |         color: '#000',
207 |         dash: []
208 |     },
209 |     extendPredictions: false,
210 |     copy: {
211 |         overwriteData: 'none'
212 |     }
213 | };
214 | var MetaSection = /** @class */ (function () {
215 |     function MetaSection(sec, _meta) {
216 |         this._meta = _meta;
217 |         var chart = _meta.chart;
218 |         var ds = _meta.dataset;
219 |         var cfg = getConfig([
220 |             'type',
221 |             'calculation',
222 |             'line',
223 |             'extendPredictions',
224 |             'copy'
225 |         ]);
226 |         this.startIndex = sec.startIndex || 0;
227 |         this.endIndex = sec.endIndex || ds.data.length - 1;
228 |         this.type = Array.isArray(cfg.type) ? cfg.type : [cfg.type];
229 |         this.line = cfg.line;
230 |         this.calculation = cfg.calculation;
231 |         this.extendPredictions = cfg.extendPredictions;
232 |         this.copy = cfg.copy;
233 |         this.label =
234 |             sec.label || this._meta.chart.data.labels[this.endIndex];
235 |         this._validateType();
236 |         // --- constructor helpers
237 |         /**
238 |          * Calculate the inherited configuration from defaultConfig, globalConfig,
239 |          * dataset config, and section config (in that order)
240 |          */
241 |         function getConfig(fields) {
242 |             var o, p;
243 |             var globalConfig = ((o = chart.config.options) && (p = o.plugins) && p.regressions) || {};
244 |             return configMerge(fields, defaultConfig, globalConfig, ds.regressions, sec);
245 |             /** merge the config objects */
246 |             function configMerge(fields) {
247 |                 var cfgList = [];
248 |                 for (var _i = 1; _i < arguments.length; _i++) {
249 |                     cfgList[_i - 1] = arguments[_i];
250 |                 }
251 |                 var dstConfig = {};
252 |                 fields.forEach(function (f) {
253 |                     cfgList.forEach(function (srcConfig) {
254 |                         var o = srcConfig[f];
255 |                         var t = typeof o;
256 |                         if (t != 'undefined') {
257 |                             if (Array.isArray(o) || t != 'object' || o == null)
258 |                                 dstConfig[f] = o;
259 |                             else
260 |                                 dstConfig[f] = Object.assign({}, dstConfig[f], configMerge(Object.keys(o), o));
261 |                         }
262 |                     });
263 |                 });
264 |                 return dstConfig;
265 |             }
266 |         }
267 |     }
268 |     /** Validates the type to avoid inconsistences */
269 |     MetaSection.prototype._validateType = function () {
270 |         if (this.type.length > 1 && this.type.includes('copy'))
271 |             throw Error('Invalid regression type:' +
272 |                 this.type +
273 |                 '. "none" cannot be combined with other type!');
274 |     };
275 |     /** Calculates the regression(s) and sets the result objects */
276 |     MetaSection.prototype.calculate = function () {
277 |         var sectionData = this._meta.normalizedData.slice(this.startIndex, this.endIndex + 1);
278 |         if (this.type[0] == 'copy')
279 |             this._calculateCopySection(sectionData);
280 |         else
281 |             this._calculateBestR2(sectionData);
282 |     };
283 |     MetaSection.prototype._calculateBestR2 = function (sectionData) {
284 |         var _this = this;
285 |         this.result = this.type.reduce(function (max, type) {
286 |             var calculation = Object.assign({}, _this.calculation);
287 |             var realType = type;
288 |             if (/polynomial[34]$/.test(type)) {
289 |                 calculation.order = parseInt(type.substr(10));
290 |                 realType = type.substr(0, 10);
291 |             }
292 |             var r = regression[realType](sectionData, calculation);
293 |             r.type = type;
294 |             return !max || max.r2 < r.r2 ? r : max;
295 |         }, null);
296 |     };
297 |     MetaSection.prototype._calculateCopySection = function (sectionData) {
298 |         var _this = this;
299 |         var from = this._meta.sections[this.copy.fromSectionIndex], r = (this.result = Object.assign({}, from.result)), overwrite = this.copy.overwriteData, data = this._meta.normalizedData;
300 |         r.points = sectionData.map(function (p) { return r.predict(p[0]); });
301 |         delete r.r2;
302 |         if (overwrite != 'none') {
303 |             var dsdata_1 = this._meta.dataset.data, isXY_1 = this._meta.isXY;
304 |             r.points.forEach(function (_a, i) {
305 |                 var x = _a[0], y = _a[1];
306 |                 var index = i + _this.startIndex;
307 |                 if ((index < from.startIndex || index > from.endIndex) &&
308 |                     (overwrite == 'all' ||
309 |                         (overwrite == 'last' && index == _this.endIndex) ||
310 |                         (overwrite == 'empty' && !data[index]))) {
311 |                     if (_this.copy.maxValue)
312 |                         y = Math.min(_this.copy.maxValue, y);
313 |                     if (_this.copy.minValue !== undefined)
314 |                         y = Math.max(_this.copy.minValue, y);
315 |                     dsdata_1[index] = isXY_1 ? { x: x, y: y } : y;
316 |                 }
317 |             });
318 |         }
319 |     };
320 |     MetaSection.prototype.drawRightBorder = function (ctx) {
321 |         ctx.beginPath();
322 |         this._setLineAttrs(ctx);
323 |         ctx.setLineDash([10, 2]);
324 |         ctx.lineWidth = 2;
325 |         // Print vertical line
326 |         var p = this._meta.getXY(this.endIndex, 0);
327 |         ctx.moveTo(p.x, this._meta.topY);
328 |         ctx.lineTo(p.x, this._meta.bottomY);
329 |         ctx.fillStyle = this.line.color;
330 |         ctx.fillText(this.label, p.x, this._meta.topY);
331 |         ctx.stroke();
332 |     };
333 |     MetaSection.prototype.drawRegressions = function (ctx) {
334 |         for (var i = 0, len = this._meta.sections.length; i < len; i++) {
335 |             var section = this._meta.sections[i];
336 |             var isMe = section == this;
337 |             if ((isMe && this.type[0] != 'copy') ||
338 |                 (!isMe && this.extendPredictions)) {
339 |                 section.drawRange(ctx, this.startIndex, this.endIndex, !isMe);
340 |             }
341 |             if (isMe)
342 |                 break;
343 |         }
344 |     };
345 |     MetaSection.prototype.drawRange = function (ctx, startIndex, endIndex, forceDash) {
346 |         var _this = this;
347 |         ctx.beginPath();
348 |         this._setLineAttrs(ctx);
349 |         if (forceDash)
350 |             ctx.setLineDash([5, 5]);
351 |         var predict = this.result.predict;
352 |         var f = function (x) { return _this._meta.getXY(x, predict(x)[1]); };
353 |         var p = f(startIndex);
354 |         ctx.moveTo(p.x, p.y);
355 |         for (var x = startIndex + 1; x <= endIndex; x++) {
356 |             p = f(x);
357 |             ctx.lineTo(p.x, p.y);
358 |         }
359 |         ctx.stroke();
360 |     };
361 |     MetaSection.prototype._setLineAttrs = function (ctx) {
362 |         if (this.line.width)
363 |             ctx.lineWidth = this.line.width;
364 |         if (this.line.color)
365 |             ctx.strokeStyle = this.line.color;
366 |         if (this.line.dash)
367 |             ctx.setLineDash(this.line.dash);
368 |     };
369 |     return MetaSection;
370 | }());
371 | exports.MetaSection = MetaSection;
372 | //# sourceMappingURL=MetaSection.js.map
373 | 
374 | /***/ }),
375 | 
376 | /***/ "./lib/index.js":
377 | /*!**********************!*\
378 |   !*** ./lib/index.js ***!
379 |   \**********************/
380 | /*! no static exports found */
381 | /***/ (function(module, exports, __webpack_require__) {
382 | 
383 | "use strict";
384 | 
385 | var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
386 |     if (k2 === undefined) k2 = k;
387 |     Object.defineProperty(o, k2, { enumerable: true, get: function() { return m[k]; } });
388 | }) : (function(o, m, k, k2) {
389 |     if (k2 === undefined) k2 = k;
390 |     o[k2] = m[k];
391 | }));
392 | var __exportStar = (this && this.__exportStar) || function(m, exports) {
393 |     for (var p in m) if (p !== "default" && !exports.hasOwnProperty(p)) __createBinding(exports, m, p);
394 | };
395 | Object.defineProperty(exports, "__esModule", { value: true });
396 | __exportStar(__webpack_require__(/*! ./types */ "./lib/types.js"), exports);
397 | __exportStar(__webpack_require__(/*! ./MetaData */ "./lib/MetaData.js"), exports);
398 | __exportStar(__webpack_require__(/*! ./MetaSection */ "./lib/MetaSection.js"), exports);
399 | __exportStar(__webpack_require__(/*! ./regression-plugin */ "./lib/regression-plugin.js"), exports);
400 | //# sourceMappingURL=index.js.map
401 | 
402 | /***/ }),
403 | 
404 | /***/ "./lib/regression-plugin.js":
405 | /*!**********************************!*\
406 |   !*** ./lib/regression-plugin.js ***!
407 |   \**********************************/
408 | /*! no static exports found */
409 | /***/ (function(module, exports, __webpack_require__) {
410 | 
411 | "use strict";
412 | 
413 | Object.defineProperty(exports, "__esModule", { value: true });
414 | exports.ChartRegressions = void 0;
415 | var MetaData_1 = __webpack_require__(/*! ./MetaData */ "./lib/MetaData.js");
416 | // Cache for all plugins' metadata
417 | var _metadataMap = {};
418 | var _chartId = 0;
419 | var Plugin = /** @class */ (function () {
420 |     function Plugin() {
421 |         this.id = 'regressions';
422 |     }
423 |     Plugin.prototype.beforeInit = function (chart) {
424 |         chart.$$id = ++_chartId;
425 |     };
426 |     /**
427 |      * Called after update (when the chart is created and when chart.update() is called)
428 |      * @param chart
429 |      */
430 |     Plugin.prototype.beforeUpdate = function (chart, options) {
431 |         var o, p, r;
432 |         var onComplete = (o = chart.config.options) &&
433 |             (p = o.plugins) &&
434 |             (r = p.regressions) &&
435 |             r.onCompleteCalculation;
436 |         forEach(chart, function (ds, meta, datasetIndex) {
437 |             meta = new MetaData_1.MetaDataSet(chart, ds);
438 |             var id = chart.$$id * 1000 + datasetIndex;
439 |             _metadataMap[id] = meta;
440 |         });
441 |         if (onComplete)
442 |             onComplete(chart);
443 |     };
444 |     /**
445 |      * It's called once before all the drawing
446 |      * @param chart
447 |      */
448 |     Plugin.prototype.beforeRender = function (chart, options) {
449 |         forEach(chart, function (ds, meta) { return meta.adjustScales(); });
450 |     };
451 |     /** Draws the vertical lines before the datasets are drawn */
452 |     Plugin.prototype.beforeDatasetsDraw = function (chart, easing, options) {
453 |         forEach(chart, function (ds, meta) { return meta.drawRightBorders(); });
454 |     };
455 |     /** Draws the regression lines */
456 |     Plugin.prototype.afterDatasetsDraw = function (chart, easing, options) {
457 |         forEach(chart, function (ds, meta) { return meta.drawRegressions(); });
458 |     };
459 |     Plugin.prototype.destroy = function (chart) {
460 |         Object.keys(_metadataMap)
461 |             .filter(function (k) { return (k / 1000) >> 0 == chart.$$id; })
462 |             .forEach(function (k) { return delete _metadataMap[k]; });
463 |     };
464 |     /** Get dataset's meta data */
465 |     Plugin.prototype.getDataset = function (chart, datasetIndex) {
466 |         var id = chart.$$id * 1000 + datasetIndex;
467 |         return _metadataMap[id];
468 |     };
469 |     /** Get dataset's meta sections */
470 |     Plugin.prototype.getSections = function (chart, datasetIndex) {
471 |         var ds = this.getDataset(chart, datasetIndex);
472 |         return ds && ds.sections;
473 |     };
474 |     return Plugin;
475 | }());
476 | function forEach(chart, fn) {
477 |     chart.data.datasets.forEach(function (ds, i) {
478 |         if (ds.regressions && chart.isDatasetVisible(i)) {
479 |             var meta = exports.ChartRegressions.getDataset(chart, i);
480 |             fn(ds, meta, i);
481 |         }
482 |     });
483 | }
484 | exports.ChartRegressions = new Plugin();
485 | window.ChartRegressions = exports.ChartRegressions;
486 | //# sourceMappingURL=regression-plugin.js.map
487 | 
488 | /***/ }),
489 | 
490 | /***/ "./lib/types.js":
491 | /*!**********************!*\
492 |   !*** ./lib/types.js ***!
493 |   \**********************/
494 | /*! no static exports found */
495 | /***/ (function(module, exports, __webpack_require__) {
496 | 
497 | "use strict";
498 | 
499 | Object.defineProperty(exports, "__esModule", { value: true });
500 | //# sourceMappingURL=types.js.map
501 | 
502 | /***/ }),
503 | 
504 | /***/ "./node_modules/regression/dist/regression.js":
505 | /*!****************************************************!*\
506 |   !*** ./node_modules/regression/dist/regression.js ***!
507 |   \****************************************************/
508 | /*! no static exports found */
509 | /***/ (function(module, exports, __webpack_require__) {
510 | 
511 | var __WEBPACK_AMD_DEFINE_FACTORY__, __WEBPACK_AMD_DEFINE_ARRAY__, __WEBPACK_AMD_DEFINE_RESULT__;(function (global, factory) {
512 |   if (true) {
513 |     !(__WEBPACK_AMD_DEFINE_ARRAY__ = [module], __WEBPACK_AMD_DEFINE_FACTORY__ = (factory),
514 | 				__WEBPACK_AMD_DEFINE_RESULT__ = (typeof __WEBPACK_AMD_DEFINE_FACTORY__ === 'function' ?
515 | 				(__WEBPACK_AMD_DEFINE_FACTORY__.apply(exports, __WEBPACK_AMD_DEFINE_ARRAY__)) : __WEBPACK_AMD_DEFINE_FACTORY__),
516 | 				__WEBPACK_AMD_DEFINE_RESULT__ !== undefined && (module.exports = __WEBPACK_AMD_DEFINE_RESULT__));
517 |   } else { var mod; }
518 | })(this, function (module) {
519 |   'use strict';
520 | 
521 |   function _defineProperty(obj, key, value) {
522 |     if (key in obj) {
523 |       Object.defineProperty(obj, key, {
524 |         value: value,
525 |         enumerable: true,
526 |         configurable: true,
527 |         writable: true
528 |       });
529 |     } else {
530 |       obj[key] = value;
531 |     }
532 | 
533 |     return obj;
534 |   }
535 | 
536 |   var _extends = Object.assign || function (target) {
537 |     for (var i = 1; i < arguments.length; i++) {
538 |       var source = arguments[i];
539 | 
540 |       for (var key in source) {
541 |         if (Object.prototype.hasOwnProperty.call(source, key)) {
542 |           target[key] = source[key];
543 |         }
544 |       }
545 |     }
546 | 
547 |     return target;
548 |   };
549 | 
550 |   function _toConsumableArray(arr) {
551 |     if (Array.isArray(arr)) {
552 |       for (var i = 0, arr2 = Array(arr.length); i < arr.length; i++) {
553 |         arr2[i] = arr[i];
554 |       }
555 | 
556 |       return arr2;
557 |     } else {
558 |       return Array.from(arr);
559 |     }
560 |   }
561 | 
562 |   var DEFAULT_OPTIONS = { order: 2, precision: 2, period: null };
563 | 
564 |   /**
565 |   * Determine the coefficient of determination (r^2) of a fit from the observations
566 |   * and predictions.
567 |   *
568 |   * @param {Array>} data - Pairs of observed x-y values
569 |   * @param {Array>} results - Pairs of observed predicted x-y values
570 |   *
571 |   * @return {number} - The r^2 value, or NaN if one cannot be calculated.
572 |   */
573 |   function determinationCoefficient(data, results) {
574 |     var predictions = [];
575 |     var observations = [];
576 | 
577 |     data.forEach(function (d, i) {
578 |       if (d[1] !== null) {
579 |         observations.push(d);
580 |         predictions.push(results[i]);
581 |       }
582 |     });
583 | 
584 |     var sum = observations.reduce(function (a, observation) {
585 |       return a + observation[1];
586 |     }, 0);
587 |     var mean = sum / observations.length;
588 | 
589 |     var ssyy = observations.reduce(function (a, observation) {
590 |       var difference = observation[1] - mean;
591 |       return a + difference * difference;
592 |     }, 0);
593 | 
594 |     var sse = observations.reduce(function (accum, observation, index) {
595 |       var prediction = predictions[index];
596 |       var residual = observation[1] - prediction[1];
597 |       return accum + residual * residual;
598 |     }, 0);
599 | 
600 |     return 1 - sse / ssyy;
601 |   }
602 | 
603 |   /**
604 |   * Determine the solution of a system of linear equations A * x = b using
605 |   * Gaussian elimination.
606 |   *
607 |   * @param {Array>} input - A 2-d matrix of data in row-major form [ A | b ]
608 |   * @param {number} order - How many degrees to solve for
609 |   *
610 |   * @return {Array} - Vector of normalized solution coefficients matrix (x)
611 |   */
612 |   function gaussianElimination(input, order) {
613 |     var matrix = input;
614 |     var n = input.length - 1;
615 |     var coefficients = [order];
616 | 
617 |     for (var i = 0; i < n; i++) {
618 |       var maxrow = i;
619 |       for (var j = i + 1; j < n; j++) {
620 |         if (Math.abs(matrix[i][j]) > Math.abs(matrix[i][maxrow])) {
621 |           maxrow = j;
622 |         }
623 |       }
624 | 
625 |       for (var k = i; k < n + 1; k++) {
626 |         var tmp = matrix[k][i];
627 |         matrix[k][i] = matrix[k][maxrow];
628 |         matrix[k][maxrow] = tmp;
629 |       }
630 | 
631 |       for (var _j = i + 1; _j < n; _j++) {
632 |         for (var _k = n; _k >= i; _k--) {
633 |           matrix[_k][_j] -= matrix[_k][i] * matrix[i][_j] / matrix[i][i];
634 |         }
635 |       }
636 |     }
637 | 
638 |     for (var _j2 = n - 1; _j2 >= 0; _j2--) {
639 |       var total = 0;
640 |       for (var _k2 = _j2 + 1; _k2 < n; _k2++) {
641 |         total += matrix[_k2][_j2] * coefficients[_k2];
642 |       }
643 | 
644 |       coefficients[_j2] = (matrix[n][_j2] - total) / matrix[_j2][_j2];
645 |     }
646 | 
647 |     return coefficients;
648 |   }
649 | 
650 |   /**
651 |   * Round a number to a precision, specificed in number of decimal places
652 |   *
653 |   * @param {number} number - The number to round
654 |   * @param {number} precision - The number of decimal places to round to:
655 |   *                             > 0 means decimals, < 0 means powers of 10
656 |   *
657 |   *
658 |   * @return {numbr} - The number, rounded
659 |   */
660 |   function round(number, precision) {
661 |     var factor = Math.pow(10, precision);
662 |     return Math.round(number * factor) / factor;
663 |   }
664 | 
665 |   /**
666 |   * The set of all fitting methods
667 |   *
668 |   * @namespace
669 |   */
670 |   var methods = {
671 |     linear: function linear(data, options) {
672 |       var sum = [0, 0, 0, 0, 0];
673 |       var len = 0;
674 | 
675 |       for (var n = 0; n < data.length; n++) {
676 |         if (data[n][1] !== null) {
677 |           len++;
678 |           sum[0] += data[n][0];
679 |           sum[1] += data[n][1];
680 |           sum[2] += data[n][0] * data[n][0];
681 |           sum[3] += data[n][0] * data[n][1];
682 |           sum[4] += data[n][1] * data[n][1];
683 |         }
684 |       }
685 | 
686 |       var run = len * sum[2] - sum[0] * sum[0];
687 |       var rise = len * sum[3] - sum[0] * sum[1];
688 |       var gradient = run === 0 ? 0 : round(rise / run, options.precision);
689 |       var intercept = round(sum[1] / len - gradient * sum[0] / len, options.precision);
690 | 
691 |       var predict = function predict(x) {
692 |         return [round(x, options.precision), round(gradient * x + intercept, options.precision)];
693 |       };
694 | 
695 |       var points = data.map(function (point) {
696 |         return predict(point[0]);
697 |       });
698 | 
699 |       return {
700 |         points: points,
701 |         predict: predict,
702 |         equation: [gradient, intercept],
703 |         r2: round(determinationCoefficient(data, points), options.precision),
704 |         string: intercept === 0 ? 'y = ' + gradient + 'x' : 'y = ' + gradient + 'x + ' + intercept
705 |       };
706 |     },
707 |     exponential: function exponential(data, options) {
708 |       var sum = [0, 0, 0, 0, 0, 0];
709 | 
710 |       for (var n = 0; n < data.length; n++) {
711 |         if (data[n][1] !== null) {
712 |           sum[0] += data[n][0];
713 |           sum[1] += data[n][1];
714 |           sum[2] += data[n][0] * data[n][0] * data[n][1];
715 |           sum[3] += data[n][1] * Math.log(data[n][1]);
716 |           sum[4] += data[n][0] * data[n][1] * Math.log(data[n][1]);
717 |           sum[5] += data[n][0] * data[n][1];
718 |         }
719 |       }
720 | 
721 |       var denominator = sum[1] * sum[2] - sum[5] * sum[5];
722 |       var a = Math.exp((sum[2] * sum[3] - sum[5] * sum[4]) / denominator);
723 |       var b = (sum[1] * sum[4] - sum[5] * sum[3]) / denominator;
724 |       var coeffA = round(a, options.precision);
725 |       var coeffB = round(b, options.precision);
726 |       var predict = function predict(x) {
727 |         return [round(x, options.precision), round(coeffA * Math.exp(coeffB * x), options.precision)];
728 |       };
729 | 
730 |       var points = data.map(function (point) {
731 |         return predict(point[0]);
732 |       });
733 | 
734 |       return {
735 |         points: points,
736 |         predict: predict,
737 |         equation: [coeffA, coeffB],
738 |         string: 'y = ' + coeffA + 'e^(' + coeffB + 'x)',
739 |         r2: round(determinationCoefficient(data, points), options.precision)
740 |       };
741 |     },
742 |     logarithmic: function logarithmic(data, options) {
743 |       var sum = [0, 0, 0, 0];
744 |       var len = data.length;
745 | 
746 |       for (var n = 0; n < len; n++) {
747 |         if (data[n][1] !== null) {
748 |           sum[0] += Math.log(data[n][0]);
749 |           sum[1] += data[n][1] * Math.log(data[n][0]);
750 |           sum[2] += data[n][1];
751 |           sum[3] += Math.pow(Math.log(data[n][0]), 2);
752 |         }
753 |       }
754 | 
755 |       var a = (len * sum[1] - sum[2] * sum[0]) / (len * sum[3] - sum[0] * sum[0]);
756 |       var coeffB = round(a, options.precision);
757 |       var coeffA = round((sum[2] - coeffB * sum[0]) / len, options.precision);
758 | 
759 |       var predict = function predict(x) {
760 |         return [round(x, options.precision), round(round(coeffA + coeffB * Math.log(x), options.precision), options.precision)];
761 |       };
762 | 
763 |       var points = data.map(function (point) {
764 |         return predict(point[0]);
765 |       });
766 | 
767 |       return {
768 |         points: points,
769 |         predict: predict,
770 |         equation: [coeffA, coeffB],
771 |         string: 'y = ' + coeffA + ' + ' + coeffB + ' ln(x)',
772 |         r2: round(determinationCoefficient(data, points), options.precision)
773 |       };
774 |     },
775 |     power: function power(data, options) {
776 |       var sum = [0, 0, 0, 0, 0];
777 |       var len = data.length;
778 | 
779 |       for (var n = 0; n < len; n++) {
780 |         if (data[n][1] !== null) {
781 |           sum[0] += Math.log(data[n][0]);
782 |           sum[1] += Math.log(data[n][1]) * Math.log(data[n][0]);
783 |           sum[2] += Math.log(data[n][1]);
784 |           sum[3] += Math.pow(Math.log(data[n][0]), 2);
785 |         }
786 |       }
787 | 
788 |       var b = (len * sum[1] - sum[0] * sum[2]) / (len * sum[3] - Math.pow(sum[0], 2));
789 |       var a = (sum[2] - b * sum[0]) / len;
790 |       var coeffA = round(Math.exp(a), options.precision);
791 |       var coeffB = round(b, options.precision);
792 | 
793 |       var predict = function predict(x) {
794 |         return [round(x, options.precision), round(round(coeffA * Math.pow(x, coeffB), options.precision), options.precision)];
795 |       };
796 | 
797 |       var points = data.map(function (point) {
798 |         return predict(point[0]);
799 |       });
800 | 
801 |       return {
802 |         points: points,
803 |         predict: predict,
804 |         equation: [coeffA, coeffB],
805 |         string: 'y = ' + coeffA + 'x^' + coeffB,
806 |         r2: round(determinationCoefficient(data, points), options.precision)
807 |       };
808 |     },
809 |     polynomial: function polynomial(data, options) {
810 |       var lhs = [];
811 |       var rhs = [];
812 |       var a = 0;
813 |       var b = 0;
814 |       var len = data.length;
815 |       var k = options.order + 1;
816 | 
817 |       for (var i = 0; i < k; i++) {
818 |         for (var l = 0; l < len; l++) {
819 |           if (data[l][1] !== null) {
820 |             a += Math.pow(data[l][0], i) * data[l][1];
821 |           }
822 |         }
823 | 
824 |         lhs.push(a);
825 |         a = 0;
826 | 
827 |         var c = [];
828 |         for (var j = 0; j < k; j++) {
829 |           for (var _l = 0; _l < len; _l++) {
830 |             if (data[_l][1] !== null) {
831 |               b += Math.pow(data[_l][0], i + j);
832 |             }
833 |           }
834 |           c.push(b);
835 |           b = 0;
836 |         }
837 |         rhs.push(c);
838 |       }
839 |       rhs.push(lhs);
840 | 
841 |       var coefficients = gaussianElimination(rhs, k).map(function (v) {
842 |         return round(v, options.precision);
843 |       });
844 | 
845 |       var predict = function predict(x) {
846 |         return [round(x, options.precision), round(coefficients.reduce(function (sum, coeff, power) {
847 |           return sum + coeff * Math.pow(x, power);
848 |         }, 0), options.precision)];
849 |       };
850 | 
851 |       var points = data.map(function (point) {
852 |         return predict(point[0]);
853 |       });
854 | 
855 |       var string = 'y = ';
856 |       for (var _i = coefficients.length - 1; _i >= 0; _i--) {
857 |         if (_i > 1) {
858 |           string += coefficients[_i] + 'x^' + _i + ' + ';
859 |         } else if (_i === 1) {
860 |           string += coefficients[_i] + 'x + ';
861 |         } else {
862 |           string += coefficients[_i];
863 |         }
864 |       }
865 | 
866 |       return {
867 |         string: string,
868 |         points: points,
869 |         predict: predict,
870 |         equation: [].concat(_toConsumableArray(coefficients)).reverse(),
871 |         r2: round(determinationCoefficient(data, points), options.precision)
872 |       };
873 |     }
874 |   };
875 | 
876 |   function createWrapper() {
877 |     var reduce = function reduce(accumulator, name) {
878 |       return _extends({
879 |         _round: round
880 |       }, accumulator, _defineProperty({}, name, function (data, supplied) {
881 |         return methods[name](data, _extends({}, DEFAULT_OPTIONS, supplied));
882 |       }));
883 |     };
884 | 
885 |     return Object.keys(methods).reduce(reduce, {});
886 |   }
887 | 
888 |   module.exports = createWrapper();
889 | });
890 | 
891 | 
892 | /***/ })
893 | 
894 | /******/ });
895 | //# sourceMappingURL=chartjs-plugin-regression-0.2.0.js.map


--------------------------------------------------------------------------------