假資料結構產生器

• 假資料結構產生器
  • Wrapping fakerjs Functions
  • 用途
    • Value
    • Selection
    • Object
    • Array
    • Tuple
    • BoundedSeries
  • 擴充

Wrapping fakerjs Functions

• Airline
• Animal
• Color
• Commerce
• Company
• Database
• Datatype
• Date(Datetime)
• Finance
• Food
• Git
• Hacker
• Image
• Internet
• Location
• Lorem
• Music
• Number
• Person
• Phone
• Science
• String
• System
• Vehicle

用途

使用 faker js 時是單一純值不能產生結構，需要自己手動組合結構，此專案利用撰寫設定檔的方式來產生一個特定的假資料函數，減少手動組合結構的麻煩。

import { StructConfig } from "struct-fakerator";

const scheme = StructConfig.object({
    name: StructConfig.value(() => "hello"),
    enum: StructConfig.select([1, 2, 3]),
    list: StructConfig.array(
        StructConfig.value(() => 10),
        5,
    ),
    tuple: StructConfig.tuple([
        StructConfig.value(() => "tuple1"),
        StructConfig.value(() => "100"),
    ]),
});

// or
import {
    createObjectConfig,
    createArrayConfig,
    createValueConfig,
    createTupleConfig,
    createSelectionConfig,
} from "struct-fakerator";

const scheme = createObjectConfig({
    name: createValueConfig(() => "hello"),
    enum: createSelectionConfig([1, 2, 3]),
    list: createArrayConfig(
        createValueConfig(() => 10),
        5,
    ),
    tuple: createTupleConfig([
        createValueConfig(() => "tuple1"),
        createValueConfig(() => 100),
        createValueConfig(() => false),
    ]),
});

// or
const scheme = {
    type: "obj",
    content: {
        name: {
            type: "value",
            generateFn: () => "hello",
        },
        enum: {
            type: "select",
            items: [1, 2, 3],
        },
        list: {
            type: "arr",
            item: {
                type: "value",
                generateFn: () => 10,
            },
            len: 5,
        },
        tuple: {
            type: "tuple",
            configItems: [
                {
                    type: "value",
                    generateFn: () => "tuple1",
                },
                {
                    type: "value",
                    generateFn: () => 100,
                },
                {
                    type: "value",
                    generateFn: () => false,
                },
            ],
        },
    },
};

import { StructGenerator } from "struct-fakerator";

const result = StructGenerator.genFn(scheme)();

console.log(result);
// {
//   name: 'hello',
//   enum: 1, // or 2 or 3
//   list: [10, 10, 10, 10, 10],
//   tuple: ['tuple1', 100, false]
// }

Value

const generateFn = StructGenerator.genFn(StructConfig.value(() => 10));

console.log(generateFn());
// 10

Selection

const generateFn = StructGenerator.genFn(StructConfig.select([1, 2, 3, 4, 5]));

console.log(generateFn());
// 1 or 2 or 3 or 4 or 5

Object

const generateFn = StructGenerator.genFn(
    StructConfig.object({
        name: StructConfig.value(() => "hello"),
        list: StructConfig.array(
            StructConfig.value(() => 10),
            5,
        ),
    }),
);

console.log(generateFn());
// {
//   name: 'hello',
//   list: [10, 10, 10, 10, 10]
// }

// with transformer function
const generateFn = StructGenerator.genFn(
    StructConfig.object(
        {
            name: StructConfig.value(() => "hello"),
            list: StructConfig.array(
                StructConfig.value(() => 10),
                5,
            ),
        },
        ({ name, list }) => list.map((item) => `${name} ${item}`),
    ),
);

console.log(generateFn());
// ["hello 10", "hello 10", "hello 10", "hello 10", "hello 10"]

Array

const generateFn = StructGenerator.genFn(
    StructConfig.array(
        StructConfig.value(() => 10),
        5,
    ),
);

console.log(generateFn());
// [10, 10, 10, 10, 10]

// with next function
const generateFn = StructGenerator.genFn(
    StructConfig.array(
        StructConfig.value(() => 10),
        5,
        (prev, current) => prev + 1 + current,
    ),
);

console.log(generateFn());
// [21, 32, 43, 54, 65]

Tuple

const generateFn = StructGenerator.genFn(
    StructConfig.tuple([
        StructConfig.value(() => "tuple1"),
        StructConfig.value(() => 100),
        StructConfig.value(() => false),
    ]),
);

console.log(generateFn());
// ['tuple1', 100, false]

BoundedSeries

const generateFn = StructGenerator.genFn(
    StructConfig.boundedSeries({
        upperLimit: 1.1,
        lowerLimit: 0.9,
        createInitValue: () => 100,
        count: 20,
    }),
);

console.log(generateFn());
// [100 * 0.9 <= num <= 100 * 1.1,
//  prev * 0.9 <= num <= prev * 1.1,
//  prev * 0.9 <= num <= prev * 1.1,
//  ...]

擴充

一切的值皆由 value 產生，可以自己創建各種不同亂數函數

flowchart TB
 value --> int
 value --> float
 value --> string
 value --> email
 value --> other[...]

Loading

但並不是所有人情況都能自己手動建立函數，有可能是開放給別人使用的服務，沒辦法在使用方建立函數，這時 createGeneratorByType 第二個可以讓製作服務的人帶入自己的擴充，這樣這個函數就能接受更多種型態。

const createIntValueConfig = (option) =>
    createValueConfig(() => faker.number.int(option));
const createEmailValueConfig = (option) =>
    createValueConfig(() => faker.internet.email(option));

const customTypeMatch = (config) => {
    if (config.type === "int") {
        return createIntValueConfig(config.option);
    }
    if (config.type === "email") {
        return createEmailValueConfig(config.option);
    }

    throw Error("error");
};

const config = {
    type: "obj",
    content: {
        name: { type: "value", generateFn: () => "John" },
        age: { type: "int" },
        email: { type: "email" },
    },
};

const result = createGeneratorByType(config, customTypeMatch)();

console.log(result);

/*
  {
    name: "John",
    age: 50,
    email: "xxx@example.com",
  }
*/