Skip to content

A simple yet powerful library to effortlessly export any data into XLSX sheets, with the full power of type-safety and awesome DX

License

Notifications You must be signed in to change notification settings

ChronicStone/typed-xlsx

Repository files navigation

typed-xlsx

npm version npm downloads bundle JSDocs License

Export any data into xls/xlsx files effortlessly, while benefiting from great type-safety & developper experience.

Key Features :

  • 🛠 Type-safe Schema Builder: Design your spreadsheet schema with strong typing for enhanced reliability and developer experience.

  • 🔄 Type-safe Data Serialization & Transformation: Ensure data integrity through type-safe serialization and transformation functionalities.

  • 🔧 Shared Type-safe Custom Value Pre-processors: Utilize shared pre-processors for consistent value transformaiton

  • 🧮 Column Summary : Auto-insert computed column summaries for efficient data analysis and overview.

  • 🧩 Complex Row Structures with Auto-Merging: Implement advanced row layouts with sub-rows for automatic row merging / styling for seamless data organization and display.

  • 🎯 Easy Default Values Management: Manage default values effortlessly, ensuring your data is presented exactly as you intend.

  • 📊 Dynamic Column Selection: Selectively choose which columns of the schema to use when building a table

  • 🗺️ Dynamic Column Mapping with Type-safe Context: Map columns dynamically with a type-safe context, injected when building sheet

  • 🎨 Dynamic Cell Styling/Formatting: Customize cell styling and formatting dynamically per-row with ease

  • 📑 Multi-sheet Support: Create spreadsheets with multiple sheets

  • 🏗️ Multiple Tables Per Sheet Support: Include as many tables you need inside a same sheet

  • 🌐 Linear or Grid-like Layout for Sheets with Multiple Tables: Choose between linear or grid layouts for sheets

INSTALLATION

pnpm add @chronicstone/typed-xlsx

USAGE EXAMPLE

1. Define the type of exported data (Or infer it from a function / a db query, or wherever you want) :

interface Organization {
  id: number
  name: string
}

interface User {
  id: number
  firstName: string
  lastName: string
  email: string
  roles: string[]
  organizations: Organization[]
  results: {
    general: { overall: number }
    technical: { overall: number }
    interview?: { overall: number }
  }
}

2. Build a sheet schema :

import { ExcelSchemaBuilder } from '@chronicstone/typed-xlsx'

// OPTIONAL : DEFINE SHARED TRANSFORMERS THAT CAN BE USE TO TRANSFORM VALUE INSERTED INTO A CELL
const transformers = {
  boolean: (value: boolean) => value ? 'Yes' : 'No',
  list: (value: (string)[]) => value.join(', '),
  arrayLength: (value: any[]) => value.length,
} satisfies TransformersMap

// Use the schema builder to define your sheet schema
const userExportSchema = ExcelSchemaBuilder
  .create<User>()
  .withTransformers(transformers)
  .column('id', {
    key: 'id',
    summary: [{ value: () => 'TOTAL BEFORE VAT' }, { value: () => 'TOTAL' }],
  })
  .column('firstName', { key: 'firstName' })
  .column('lastName', { key: 'lastName' })
  .column('email', { key: 'email' })
  .column('roles', {
    key: 'roles',
    transform: 'list',
    cellStyle: data => ({ font: { color: { rgb: data.roles.includes('admin') ? 'd10808' : undefined } } }),
  })
  .column('balance', {
    key: 'balance',
    format: '"$"#,##0.00_);\\("$"#,##0.00\\)',
    summary: [
      {
        value: data => data.reduce((acc, user) => acc + user.balance, 0),
        format: '"$"#,##0.00_);\\("$"#,##0.00\\)',
      },
      {
        value: data => data.reduce((acc, user) => acc + user.balance, 0) * 1.2,
        format: '"$"#,##0.00_);\\("$"#,##0.00\\)',
      },
    ],
  })
  .column('nbOrgs', { key: 'organizations', transform: 'arrayLength' })
  .column('orgs', { key: 'organizations', transform: org => org.map(org => org.name).join(', ') })
  .column('generalScore', {
    key: 'results.general.overall',
    format: '# / 10',
    summary: [{
      value: data => data.reduce((acc, user) => acc + user.results.general.overall, 0) / data.length,
      format: '# / 10',
    }],
  })
  .column('technicalScore', {
    key: 'results.technical.overall',
    summary: [{
      value: data => data.reduce((acc, user) => acc + user.results.technical.overall, 0) / data.length,
    }],
  })
  .column('interviewScore', { key: 'results.interview.overall', default: 'N/A' })
  .column('createdAt', { key: 'createdAt', format: 'd mmm yyyy' })
  .group('group:org', (builder, context: Organization[]) => {
    for (const org of context) {
      builder
        .column(`orga-${org.id}`, {
          label: `User in ${org.name}`,
          key: 'organizations',
          transform: orgs => orgs.some(o => o.id === org.id) ? 'YES' : 'NO',
          cellStyle: data => ({
            font: {
              color: { rgb: data.organizations.some(o => o.id === org.id) ? '61eb34' : 'd10808' },
            },
          }),
        })
    }
  })
  .build()

3. Safely compose excel file from schemas

import { ExcelBuilder } from '@chronicstone/typed-xlsx'

const buffer = ExcelBuilder
  .create()
  .sheet('Users - full')
  .addTable({
    data: users,
    schema: assessmentExport,
    context: {
      'group:org': organizations,
    },
  })
  .sheet('Users - partial')
  .addTable({
    data: users,
    schema: assessmentExport,
    select: {
      firstName: true,
      lastName: true,
      email: true,
    },
  })
  .sheet('User - neg partial')
  .addTable({
    data: users,
    schema: assessmentExport,
    select: {
      firstName: false,
      lastName: false,
      email: false,
    },
    context: {
      'group:org': organizations,
    },
  })
  .sheet('User - Multiple tables')
  .sheet('Multi-tables-grid', { tablesPerRow: 2 })
  .addTable({
    title: 'Table 1',
    data: users.filter((_, i) => i < 5),
    schema: assessmentExport,
    select: { firstName: true, lastName: true, email: true, createdAt: true },
  })
  .addTable({
    title: 'Table 2',
    data: users.filter((_, i) => i < 5),
    schema: assessmentExport,
    select: { firstName: true, lastName: true, email: true, balance: true },
  })
  .addTable({
    title: 'Table 3',
    data: users.filter((_, i) => i < 5),
    schema: assessmentExport,
    select: { firstName: true, lastName: true, email: true, balance: true },
  })
  .addTable({
    title: 'Table 4',
    data: users.filter((_, i) => i < 5),
    schema: assessmentExport,
    select: { firstName: true, lastName: true, email: true, createdAt: true },
  })
  .build({ output: 'buffer' })

fs.writeFileSync('test.xlsx', arrayBuffer)

4. Have fun

Here's the generated file for the example from above

DOWNLOAD GENERATED EXAMPLE

OPEN EXAMPLE IN STACKBLITZ

License

MIT License © 2023-PRESENT Cyprien THAO