markov-strings - npm Package Compare versions

Comparing version 2.1.0 to 3.0.0-beta.0

dist/index.d.ts

 export declare type MarkovGenerateOptions = {
     maxTries?: number;
-    prng?: any;
+    prng?: () => number;
     filter?: (result: MarkovResult) => boolean;
@@ -5,0 +5,0 @@ };

package.json

 {
   "name": "markov-strings",
-  "version": "2.1.0",
+  "version": "3.0.0-beta.0",
   "description": "A Markov string generator",
@@ -8,2 +8,3 @@ "main": "dist/index.js",
   "scripts": {
+    "start": "tsc --watch",
     "build": "tsc",
@@ -31,13 +32,13 @@ "test": "jest --coverage --coverageReporters=lcov --coverageReporters=html",
   "devDependencies": {
-    "@types/jest": "^24.0.15",
-    "@types/lodash": "^4.14.136",
-    "@types/node": "^12.6.2",
-    "coveralls": "^3.0.5",
-    "jest": "^24.8.0",
-    "ts-jest": "^24.0.2",
-    "typescript": "^3.5.3"
+    "@types/jest": "^25.1.3",
+    "@types/lodash": "^4.14.149",
+    "@types/node": "^13.7.7",
+    "coveralls": "^3.0.9",
+    "jest": "^25.1.0",
+    "ts-jest": "^25.2.1",
+    "typescript": "^3.8.3"
   },
   "dependencies": {
-    "lodash": "^4.17.14"
+    "lodash": "^4.17.15"
   }
 }

README.md

@@ -7,3 +7,3 @@ [![Build Status](https://travis-ci.org/scambier/markov-strings.svg?branch=master)](https://travis-ci.org/scambier/markov-strings)
 
-A simplistic Markov chain text generator.  
+A simplistic Markov chain text generator.
 Give it an array of strings, and it will output a randomly generated string.
@@ -13,12 +13,12 @@
 
-- [Markov-strings](#Markov-strings)
-  - [Prerequisites](#Prerequisites)
-  - [Installing](#Installing)
-  - [Usage](#Usage)
-  - [API](#API)
-    - [new Markov(data, [options])](#new-Markovdata-options)
-    - [.buildCorpus()](#buildCorpus)
+- [Markov-strings](#markov-strings)
+  - [Prerequisites](#prerequisites)
+  - [Installing](#installing)
+  - [Usage](#usage)
+  - [API](#api)
+    - [new Markov([options])](#new-markovoptions)
+    - [.addData(data)](#adddatadata)
     - [.generate([options])](#generateoptions)
-  - [Changelog](#Changelog)
-  - [Running the tests](#Running-the-tests)
+  - [Changelog](#changelog)
+  - [Running the tests](#running-the-tests)
 
@@ -43,5 +43,7 @@ ## Prerequisites
 // Build the Markov generator
-const markov = new Markov(data, { stateSize: 2 })
-markov.buildCorpus()
+const markov = new Markov({ stateSize: 2 })
 
+// Add data for the generator
+markov.addData(data)
+
 const options = {
@@ -72,6 +74,22 @@ maxTries: 20, // Give up if I don't have a sentence after 20 tries (default is 10)
 
-### new Markov(data, [options])
+### new Markov([options])
 
 Create a generator instance.
 
+#### options
+
+```js
+{
+  stateSize: number
+}
+```
+
+The `stateSize` is the number of words for each "link" of the generated sentence. `1` will output gibberish sentences without much sense. `2` is a sensible default for most cases. `3` and more can create good sentences if you have a corpus that allows it.
+
+### .addData(data)
+
+To function correctly, the Markov generator needs its internal data to be correctly structured. `.addData(data)` allows you add raw data, that is automatically formatted to fit the internal structure.
+
+You can call `.addData(data)` as often as you need, with new data each time.
+
 #### data
@@ -87,5 +105,7 @@
 
-`[ 'lorem ipsum', 'dolor sit amet' ]`  
+```js
+[ 'lorem ipsum', 'dolor sit amet' ]
+```
 
-or  
+or
 
@@ -99,19 +119,4 @@ ```js
 
-#### options
+Since `.addData(data)` can take some time (it loops for each word of each string), a non-blocking variant `.addDataAsync(data)` is conveniently available if you need it.
 
-```js
-{
-  stateSize: number
-}
-```
-
-The `stateSize` is the number of words for each "link" of the generated sentence. `1` will output gibberish sentences without much sense. `2` is a sensible default for most cases. `3` and more can create good sentences if you have a corpus that allows it.
-
-### .buildCorpus()
-
-This function **must** be called to build the corpus for Markov generation.
-It will iterate over all words from your `data` parameter to create an internal optimized structure.
-
-Since `.buildCorpus()` can take some time (it loops for each word of each string), a non-blocking variant `.buildCorpusAsync()` is conveniently available if you need it.
-
 ### .generate([options])
@@ -146,2 +151,6 @@
 
+#### 3.0.0
+
+Refactoring to facilitate iterative construction of the corpus (multiple `.addData()` instead of a one-time `buildCorpus()`)
+
 #### 2.1.0
@@ -148,0 +157,0 @@

src/index.ts

 import { assignIn, cloneDeep, flatten, includes, isEmpty, isString, slice, some, uniqBy } from 'lodash'
 
-function sampleWithPRNG (array: Array<any>, prng: any = Math.random) {
+function sampleWithPRNG<T>(array: T[], prng: () => number = Math.random): T | undefined {
   const length = array == null ? 0 : array.length
@@ -8,5 +8,7 @@ return length ? array[Math.floor(prng() * length)] : undefined
 
+export type MarkovInputData = Array<{ string: string }>
+
 export type MarkovGenerateOptions = {
   maxTries?: number,
-  prng?: any,
+  prng?: () => number,
   filter?: (result: MarkovResult) => boolean
@@ -22,3 +24,3 @@ }
   score: number,
-  refs: Array<{ string: string }>,
+  refs: MarkovInputData,
   tries: number
@@ -29,3 +31,3 @@ }
   words: string
-  refs: Array<{ string: string }>
+  refs: MarkovInputData
 }
@@ -36,3 +38,3 @@
 export default class Markov {
-  public data: Array<{ string: string }>
+  public data: MarkovInputData
   public startWords: MarkovFragment[] = []
@@ -50,15 +52,7 @@ public endWords: MarkovFragment[] = []
    *
-   * @param {(string[] | Array<{ string: string }>)} data An array of strings or objects.
-   * If 'data' is an array of objects, each object must have a 'string' attribute
    * @param {MarkovConstructorOptions} [options={}]
    * @memberof Markov
    */
-  constructor(data: string[] | Array<{ string: string }>, options: MarkovConstructorOptions = {}) {
-    // Format data if necessary
-    if (isString(data[0])) {
-      data = (data as string[]).map(s => ({ string: s }))
-    } else if (!data[0].hasOwnProperty('string')) {
-      throw new Error('Objects in your corpus must have a "string" property')
-    }
-    this.data = data as Array<{ string: string }>
+  constructor(options: MarkovConstructorOptions = {}) {
+    this.data = []
 
@@ -70,2 +64,17 @@ // Save options
 
+  public addData(rawData: MarkovInputData | string[]) {
+    // Format data if necessary
+    let input: MarkovInputData = []
+    if (isString(rawData[0])) {
+      input = (rawData as string[]).map(s => ({ string: s }))
+    }
+    else if (!rawData[0].hasOwnProperty('string')) {
+      throw new Error('Objects in your corpus must have a "string" property')
+    }
+
+    this.buildCorpus(input)
+
+    this.data = this.data.concat(input)
+  }
+
   /**
@@ -76,6 +85,6 @@ * Builds the corpus. You must call this before generating sentences.
    */
-  public buildCorpus(): void {
+  private buildCorpus(data: MarkovInputData): void {
     const options = this.options
 
-    this.data.forEach(item => {
+    data.forEach(item => {
       const line = item.string
@@ -137,6 +146,6 @@ const words = line.split(' ')
    */
-  public buildCorpusAsync(): Promise<void> {
+  public addDataAsync(data: MarkovInputData | string[]): Promise<void> {
     return new Promise((resolve, reject) => {
       try {
-        this.buildCorpus()
+        this.addData(data)
         resolve()
@@ -143,0 +152,0 @@ } catch (e) {

test/test.ts

@@ -17,11 +17,5 @@ import { map, some } from 'lodash'
   describe('Constructor', () => {
-    it('should throw an error if corpus is invalid', () => {
-      expect(() => {
-        // @ts-ignore
-        const markov = new Markov([{}])
-      }).toThrowError()
-    })
 
     it('should have a default stateSize', () => {
-      const markov = new Markov(data)
+      const markov = new Markov()
       expect(markov.options.stateSize).toBe(2)
@@ -31,3 +25,3 @@ })
     it('should save a different stateSize', () => {
-      const markov = new Markov(data, { stateSize: 3 })
+      const markov = new Markov({ stateSize: 3 })
       expect(markov.options.stateSize).toBe(3)
@@ -37,7 +31,7 @@ })
 
-  describe('Build the corpus', () => {
+  describe('Adding data', () => {
     it('should build synchronously', () => {
-      const markov = new Markov(data)
+      const markov = new Markov()
       expect(markov.corpus).toEqual({})
-      markov.buildCorpus()
+      markov.addData(data)
       expect(markov.corpus).not.toEqual({})
@@ -47,14 +41,23 @@ })
     it('should build asynchronously', async () => {
-      const markov = new Markov(data)
-      markov.buildCorpus = jest.fn()
-      await markov.buildCorpusAsync()
-      expect(markov.buildCorpus).toHaveBeenCalled()
+      const markov = new Markov()
+      markov.addData = jest.fn()
+      await markov.addDataAsync(data)
+      expect(markov.addData).toHaveBeenCalled()
     })
+
+    it('should throw an error if the data structure is invalid', () => {
+      const markov = new Markov()
+      expect(() => {
+        // @ts-ignore
+        markov.addData([{}])
+      }).toThrowError()
+    })
+
   })
 
-  describe('After building the corpus', () => {
+  describe('After adding data', () => {
     let markov: Markov
     beforeEach(() => {
-      markov = new Markov(data)
-      markov.buildCorpus()
+      markov = new Markov()
+      markov.addData(data)
     })
@@ -94,28 +97,22 @@
     })
-  })
 
-  describe('The corpus itself', () => {
-    let markov: Markov
-    beforeEach(() => {
-      markov = new Markov(data)
-      markov.buildCorpus()
+    describe('The corpus itself', () => {
+      it('should have the right values for the right keys', () => {
+        const corpus = markov.corpus
+        expect(some(corpus['Lorem ipsum'], { words: 'dolor sit' })).toBeTruthy()
+        expect(
+          some(corpus['Lorem ipsum'], { words: 'duplicate start' })
+        ).toBeTruthy()
+        expect(
+          some(corpus['tempor, erat'], { words: 'vel lacinia' })
+        ).toBeTruthy()
+      })
     })
-
-    it('should have the right values for the right keys', () => {
-      const corpus = markov.corpus
-      expect(some(corpus['Lorem ipsum'], { words: 'dolor sit' })).toBeTruthy()
-      expect(
-        some(corpus['Lorem ipsum'], { words: 'duplicate start' })
-      ).toBeTruthy()
-      expect(
-        some(corpus['tempor, erat'], { words: 'vel lacinia' })
-      ).toBeTruthy()
-    })
   })
 
-  describe('The sentence generation', () => {
+  describe('The sentence generator', () => {
     let markov: Markov
     beforeEach(() => {
-      markov = new Markov(data)
-      markov.buildCorpus()
+      markov = new Markov()
+      markov.addData(data)
     })
@@ -130,3 +127,3 @@
     it('should throw an error if the corpus is not built', () => {
-      markov = new Markov(data)
+      markov = new Markov()
       expect(() => {
@@ -133,0 +130,0 @@ markov.generate()

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

No v1

Quality

Package is not semver >=1. This means it is not stable and does not support ^ ranges.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

36168

increased by0.67%

Lines of code

707

increased by0.14%

Number of lines in readme file

199

increased by4.74%

Worsened metrics

Number of low quality alerts

2

increased by100%

Dependency changes

• Updatedlodash@^4.17.15