Skip to main content

Google Sheets Data Interchange

This demo focuses on external data processing. For Google Apps Script custom functions, the "Google Sheets" extension demo covers Apps Script integration.

Google Sheets is a collaborative spreadsheet service with powerful external APIs for automation.

SheetJS is a JavaScript library for reading and writing data from spreadsheets.

This demo uses SheetJS to properly exchange data with spreadsheet files. We'll explore how to use NodeJS integration libraries and SheetJS in three data flows:

  • "Importing data": Data in a NUMBERS spreadsheet will be parsed using SheetJS libraries and written to a Google Sheets Document

  • "Exporting data": Data in Google Sheets will be pulled into arrays of objects. A workbook will be assembled and exported to Excel Binary workbooks (XLSB).

  • "Exporting files": SheetJS libraries will read XLSX files exported by Google Sheets and generate CSV rows from every worksheet.

It is strongly recommended to create a new Google account for testing.

One small mistake could result in a block or ban from Google services.

Google Sheets deprecates APIs quickly and there is no guarantee that the referenced APIs will be available in the future.

Integration Details​

This demo uses the Sheets v4 and Drive v3 APIs through the official googleapis connector module.

Initial Setup

There are a number of steps to enable the Google Sheets API and Google Drive API for an account. The Complete Example covers the process.

Document Duality​

Each Google Sheets document is identified with a unique ID. This ID can be found from the Google Sheets edit URL.

The edit URL starts with https://docs.google.com/spreadsheets/d/ and includes /edit. The ID is the string of characters between the slashes. For example:

https://docs.google.com/spreadsheets/d/a_long_string_of_characters/edit#gid=0
                                      |^^^^^^^^^^^^^^^^^^^^^^^^^^^|--- ID

The same ID is used in Google Drive operations.

The following operations are covered in this demo:

OperationAPI
Create Google Sheets DocumentSheets
Add and Remove worksheetsSheets
Modify data in worksheetsSheets
Share Sheets with other usersDrive
Generate raw file exportsDrive

Authentication​

It is strongly recommended to use a service account for Google API operations. The "Service Account Setup" section covers how to create a service account and generate a JSON key file.

The generated JSON key file includes client_email and private_key fields. These fields can be used in JWT authentication:

JWT Authentication using a JSON key file
import { google } from "googleapis";

// adjust the path to the actual key file.
import creds from './sheetjs-test-726272627262.json' assert { type: "json" };

/* connect to google services */
const jwt = new google.auth.JWT({
  email: creds.client_email,
  key: creds.private_key,
  scopes: [
    'https://www.googleapis.com/auth/spreadsheets', // Google Sheets
    'https://www.googleapis.com/auth/drive.file', // Google Drive
  ]
});

Connecting to Services​

The google named export includes special methods to connect to various APIs.

Google Sheets​

const sheets = google.sheets({ version: "v4", auth: jwt });

google.sheets takes an options argument that includes API version number and authentication details.

Google Drive​

const drive = google.drive({ version: "v3", auth: jwt });

google.drive takes an options argument that includes API version number and authentication details.

Array of Arrays​

"Arrays of Arrays" are the main data format for interchange with Google Sheets. The outer array object includes row arrays, and each row array includes data.

SheetJS provides methods for working with Arrays of Arrays:

  • aoa_to_sheet1 creates SheetJS worksheet objects from arrays of arrays
  • sheet_to_json2 can generate arrays of arrays from SheetJS worksheets

Export Document Data​

The goal is to create an XLSB export from a Google Sheet. Google Sheets does not natively support the XLSB format. SheetJS fills the gap.

Convert a Single Sheet​

sheets.spreadsheets.values.get returns data from an existing Google Sheet. The method expects a range. Passing the sheet name as the title will pull all rows.

If successful, the response object will have a data property. It will be an object with a values property. The values will be represented as an Array of Arrays of values. This array of arrays can be converted to a SheetJS sheet:

async function gsheet_ws_to_sheetjs_ws(id, sheet_name) {
  /* get values */
  const res = await sheets.spreadsheets.values.get({
    spreadsheetId: id,
    range: `'${sheet_name}'`
  });
  const values = res.data.values;

  /* create SheetJS worksheet */
  const ws = XLSX.utils.aoa_to_sheet(values);
  return ws;
}

Convert a Workbook​

sheets.spreadsheets.get returns metadata about the Google Sheets document. In the result object, the data property is an object which has a sheets property. The value of the sheets property is an array of sheet objects.

The SheetJS book_new 3 method creates blank SheetJS workbook objects. The book_append_sheet 4 method adds SheetJS worksheet objects to the workbook.

By looping across the sheets, the entire workbook can be converted:

async function gsheet_doc_to_sheetjs_wb(doc) {
  /* Create a new workbook object */
  const wb = XLSX.utils.book_new();

  /* Get metadata */
  const wsheet = await sheets.spreadsheets.get({spreadsheetId: id});

  /* Loop across the Document sheets */
  for(let sheet of wsheet.data.sheets) {
    /* Get the worksheet name */
    const name = sheet.properties.title;

    /* Convert Google Docs sheet to SheetJS worksheet */
    const ws = await gsheet_ws_to_sheetjs_ws(id, name);

    /* Append worksheet to workbook */
    XLSX.utils.book_append_sheet(wb, ws, name);
  }

  return wb;
}

This method returns a SheetJS workbook object that can be exported with the writeFile and write methods.5

Update Document Data​

The goal is to import data from a NUMBERS file to Google Sheets. Google Sheets does not natively support the NUMBERS format. SheetJS fills the gap.

Create New Document​

sheets.spreadsheets.create creates a new Google Sheets document. It can accept a document title. It will generate a new workbook with a blank "Sheet1" sheet. The response includes the document ID for use in subsequent operations:

Create a new document with blank sheet
const res = await sheets.spreadsheets.create({
  requestBody: {
    properties: {
      /* Document Title */
      title: "SheetJS Test"
    }
  }
});
const id = res.data.spreadsheetId;

When using a service worker, the main account does not have access to the new document by default. The document has to be shared with the main account using the Drive API:

Sharing the generated sheet with the main account
await drive.permissions.create({
  fileId: id, // this ID was returned in the response to the create request
  fields: "id",
  requestBody: {
    type: "user",
    role: "writer",
    emailAddress: "YOUR_ADDRESS@gmail.com" // main address
  }
});

Delete Non-Initial Sheets​

Google Sheets does not allow users to delete every worksheet.

The recommended approach involves deleting every worksheet after the first.

The delete operation requires a unique identifier for a sheet within the Google Sheets document. These IDs are found in the sheets.spreadsheets.get response.

The following snippet performs one bulk operation using batchUpdate:

Deleting non-initial sheets
/* get existing sheets */
const wsheet = await sheets.spreadsheets.get({spreadsheetId: id});

/* remove all sheets after the first */
if(wsheet.data.sheets.length > 1) await sheets.spreadsheets.batchUpdate({
  spreadsheetId: id,
  requestBody: { requests: wsheet.data.sheets.slice(1).map(s => ({
    deleteSheet: {
      sheetId: s.properties.sheetId
    }
  }))}
});

Rename First Sheet​

The first sheet must be renamed so that the append operations do not collide with the legacy name. Since most SheetJS-supported file formats and most spreadsheet applications limit worksheet name lengths to 32 characters, it is safe to set a name that exceeds 33 characters.

The updateSheetProperties update method can rename individual sheets:

Rename legacy first sheet
/* rename first worksheet to avoid collisions */
await sheets.spreadsheets.batchUpdate({
  spreadsheetId: id,
  requestBody: { requests: [{
    updateSheetProperties: {
      fields: "title",
      properties: {
        sheetId: wsheet.data.sheets[0].properties.sheetId,
         // the new title is 34 characters, to be exact
        title: "thistitleisatleast33characterslong"
      }
    }
  }]}
});

Append Worksheets​

The read and readFile methods generate SheetJS workbook objects from existing worksheet files.

Starting from a SheetJS workbook, the SheetNames property6 is an array of worksheet names and the Sheets property is an object that maps sheet names to worksheet objects.

Looping over the worksheet names, there are two steps to appending a sheet:

  1. "Append a blank worksheet": The addSheet request, submitted through the sheets.spreadsheets.batchUpdate method, accepts a new title and creates a new worksheet. The new worksheet will be added at the end.

  2. "Write data to the new sheet": The SheetJS sheet_to_json method with the option header: 17 will generate an array of arrays of data. This structure is compatible with the sheets.spreadsheets.values.update operation.

The following snippet pushes all worksheets from a SheetJS workbook into a Google Sheets document:

Push data from a SheetJS workbook to a Google Sheets Document
/* add sheets from file */
for(let name of wb.SheetNames) {
  /* (1) Create a new Google Sheets sheet */
  await sheets.spreadsheets.batchUpdate({
    spreadsheetId: id,
    requestBody: { requests: [
      /* add new sheet */
      { addSheet: { properties: { title: name } } },
    ] }
  });

  /* (2) Push data */
  const aoa = XLSX.utils.sheet_to_json(wb.Sheets[name], {header:1});
  await sheets.spreadsheets.values.update({
    spreadsheetId: id,
    range: `'${name}'!A1`,
    valueInputOption: "USER_ENTERED",
    resource: { values: aoa }
  });
}

Delete Initial Sheet​

After adding new worksheets, the final step involves removing the initial sheet.

The initial sheet ID can be pulled from the worksheet metadata fetched when the non-initial sheets were removed:

Deleting the Initial sheet
/* remove first sheet */
await sheets.spreadsheets.batchUpdate({
  spreadsheetId: id,
  requestBody: { requests: [
    /* remove old first sheet */
    { deleteSheet: { sheetId: wsheet.data.sheets[0].properties.sheetId } }
  ] }
});

Raw File Exports​

In the web interface, Google Sheets can export documents to XLSX or ODS.

Raw file exports are exposed through the files.export method in the Drive API:

Export XLSX workbook
const drive = google.drive({ version: "v3", auth: jwt });

/* Request XLSX export */
const file = await drive.files.export({
  /* XLSX MIME type */
  mimeType: "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
  fileId: id
});

The mimeType property is expected to be one of the supported formats8. When the demo was last tested, the following workbook conversions were supported:

FormatMIME Type
XLSXapplication/vnd.openxmlformats-officedocument.spreadsheetml.sheet
ODSapplication/x-vnd.oasis.opendocument.spreadsheet

The response object has a data field whose value will be a Blob object. Data can be pulled into an ArrayBuffer and passed to the SheetJS read9 method:

/* Obtain ArrayBuffer */
const ab = await file.data.arrayBuffer();

/* Parse */
const wb = XLSX.read(buf);

The code snippet works for XLSX and ODS. Google Sheets supports other formats with different integration logic.

Plaintext

The following formats are considered "plaintext":

FormatMIME Type
CSV (first sheet)text/csv
TSV (first sheet)text/tab-separated-values

For these formats, file.data is a JS string that can be parsed directly:

/* Request CSV export */
const file = await drive.files.export({ mimeType: "text/csv", fileId: id });

/* Parse CSV string*/
const wb = XLSX.read(file.data, {type: "string"});

HTML

Google Sheets has one relevant HTML type:

FormatMIME Type
HTML (all sheets)application/zip

The HTML export of a Google Sheets worksheet includes a row for the column labels (A, B, ...) and a column for the row labels (1, 2, ...).

The complete package is a ZIP file that includes a series of .html files. The files are written in tab order. The name of each file matches the name in Google Sheets.

This ZIP can be extracted using the embedded CFB library:

import { read, utils, CFB } from 'xlsx';
// -------------------^^^-- `CFB` named import

// ...

/* Parse Google Sheets ZIP file */
const cfb = CFB.read(new Uint8Array(ab), {type: "array"});

/* Create new SheetJS workbook */
const wb = utils.book_new();

/* Scan through each entry in the ZIP */
cfb.FullPaths.forEach((n, i) => {
  /* only process HTML files */
  if(n.slice(-5) != ".html") return;

  /* Extract worksheet name */
  const name = n.slice(n.lastIndexOf("/")+1).slice(0,-5);

  /* parse HTML */
  const htmlwb = read(cfb.FileIndex[i].content);

  /* add worksheet to workbook */
  utils.book_append_sheet(wb, htmlwb.Sheets.Sheet1, name);
});

At this point wb is a SheetJS workbook object10.

Complete Example​

Tested Deployments

This demo was last tested on 2024 June 08 using googleapis version 140.0.0. The demo uses Sheets v4 and Drive v3 APIs.

The Google Cloud web interface changes frequently!

The screenshots and detailed descriptions may be out of date. Please report any issues to the docs repo or reach out to the SheetJS Discord server.

Account Setup​

  1. Create a new Google account or log into an existing account.

A valid phone number (for SMS verification) may be required.

  1. Open https://console.cloud.google.com in a web browser.

If this is the first time accessing Google Cloud resources, a terms of service modal will be displayed. Review the Google Cloud Platform Terms of Service by clicking the "Google Cloud Platform Terms of Service" link.

You must agree to the Google Cloud Platform Terms of Service to use the APIs.

Check the box under "Terms of Service" and click "AGREE AND CONTINUE".

Project Setup​

The goal of this section is to create a new project.

  1. Open the Project Selector.

In the top bar, between the "Google Cloud" logo and the search bar, there will be a selection box. Click the ▼ icon to show the modal.

Project Selector

If the selection box is missing, expand the browser window.

  1. Click "NEW PROJECT" in the top right corner of the modal.

  2. In the New Project screen, enter "SheetJS Test" in the Project name textbox and select "No organization" in the Location box. Click "CREATE".

A notification will confirm that the project was created:

Project notification

API Setup​

The goal of this section is to enable Google Sheets API and Google Drive API.

  1. Open the Project Selector (▼ icon) and select "SheetJS Test"

  2. In the search bar, type "Enabled" and select "Enabled APIs & services". This item will be in the "PRODUCTS & PAGES" part of the search results.

Enable Google Sheets API​

  1. Near the top of the page, click "+ ENABLE APIS AND SERVICES".

  2. In the search bar near the middle of the page (not the search bar at the top), type "Sheets" and press Enter.

In the results page, look for "Google Sheets API". Click the card

  1. In the Product Details screen, click the blue "ENABLE" button.

  2. Click the left arrow (<-) next to "API/Service details".

Enable Google Drive API​

  1. Near the top of the page, click "+ ENABLE APIS AND SERVICES".

  2. In the search bar near the middle of the page (not the search bar at the top), type "Drive" and press Enter.

In the results page, look for "Google Drive API". Click the card

  1. In the Product Details screen, click the blue "ENABLE" button.

Service Account Setup​

The goal of this section is to create a service account and generate a JSON key.

  1. Go to https://console.cloud.google.com or click the "Google Cloud" image in the top bar.

Create Service Account​

  1. Click the Project Selector (:· icon) and select "SheetJS Test".

  2. In the search bar, type "Credentials" and select the "Credentials" item with subtitle "APIs & Services". This item will be in the "PRODUCTS & PAGES" group:

Credentials

  1. Click "+ CREATE CREDENTIALS". In the dropdown, select "Service Account"

  2. Enter "SheetJService" for Service account name. Click "CREATE AND CONTINUE"

The Service account ID is generated automatically.

  1. In Step 2 "Grant this service account access to project", click CONTINUE

  2. In Step 3 click "DONE". You will be taken back to the credentials screen

Create JSON Key​

  1. Look for "SheetJService" in the "Service Accounts" table and click the email address in the row.

  2. Click "KEYS" in the horizontal bar near the top of the page.

  3. Click "ADD KEY" and select "Create new key" in the dropdown.

  4. In the popup, select the "JSON" radio button and click "CREATE".

The page will download a JSON file. If prompted, allow the download.

  1. Click "CLOSE"

Create Document​

The goal of this section is to create a document from the service account and share with the main account.

  1. Create a SheetJSGS folder and initialize:
mkdir SheetJSGS
cd SheetJSGS
npm init -y

  1. Copy the JSON file from step 24 into the project folder.

  2. Install dependencies:

npm i --save https://cdn.sheetjs.com/xlsx-0.20.3/xlsx-0.20.3.tgz googleapis
  1. Download init.mjs:
curl -LO https://docs.sheetjs.com/gsheet/init.mjs

Edit the marked lines near the top of the file:

init.mjs (edit highlighted lines)
/* Change this import statement to point to the credentials JSON file */
import creds from './sheetjs-test-726272627262.json' assert { type: "json" };

/* Change this to the primary account address, NOT THE SERVICE ACCOUNT */
const acct = "YOUR_ADDRESS@gmail.com";

  • './sheetjs-test-726272627262.json' should be replaced with the name of the JSON file in step 27. The ./ prefix is required!

  • 'YOUR_ADDRESS@gmail.com' should be replaced with the Google Account email address from step 0.

  1. Run the script:
node init.mjs

The script will print three lines:

Created Google Workbook a-long-string-of-characters
Created Google Worksheets "SheetJS1" and "SheetJS2"
Shared a-long-string-of-characters with YOUR_ACCOUNT@gmail.com

The long string of characters after "Created Google Workbook" is the ID. Take note of this ID.

  1. Sign into Google Sheets. A shared document "SheetJS Test" should be displayed in the table. It will be owned by the service account.

  2. Open the shared document from step 31 and confirm that the document has two worksheets named "SheetJS1" and "SheetJS2".

Confirm the worksheet data matches the following screenshots:

SheetDataScreenshot
SheetJS1
[
  [ "Sheet", "JS" ],
  [ 72, 62 ]
]

SheetJS1 data

SheetJS2
[
  [ "Area Code", "Part 1", "Part 2" ],
  [ 201, 867, 5309 ],
  [ 281, 330, 8004 ]
]

SheetJS2 data

  1. Copy the URL and extract the document ID.

The URL of the document will look like

https://docs.google.com/spreadsheets/d/a_long_string_of_characters/edit#gid=0
---------------------------------------^^^^^^^^^^^^^^^^^^^^^^^^^^^--- ID

The ID is a long string of letters and numbers and underscore characters (_) just before the /edit part of the URL.

Confirm that this ID matches the ID printed in step 30.

Load Data from NUMBERS​

The goal of this section is to update the new document with data from a sample NUMBERS file.

  1. Download the test file pres.numbers:
curl -LO https://docs.sheetjs.com/pres.numbers
  1. Download load.mjs:
curl -LO https://docs.sheetjs.com/gsheet/load.mjs

Edit the marked lines near the top of the file:

load.mjs (edit highlighted lines)
/* Change this import statement to point to the credentials JSON file */
import creds from './sheetjs-test-726272627262.json' assert { type: "json" };

/* Change this to the spreadsheet ID */
const id = "SOME-SPREADSHEETJS-ID";

  • './sheetjs-test-726272627262.json' should be replaced with the name of the JSON file in step 27. The ./ prefix is required!

  • "SOME-SPREADSHEETJS-ID" should be replaced with the Document ID from step 33.

  1. Run the script:
node load.mjs
  1. Sign into Google Sheets and open the "SheetJS Test" shared document. It should show a list of Presidents, matching the contents of the test file.

Export Data to XLSB​

The goal of this section is to export the raw data from Google Sheets to XLSB.

  1. Download dump.mjs:
curl -LO https://docs.sheetjs.com/gsheet/dump.mjs

Edit the marked lines near the top of the file:

dump.mjs (edit highlighted lines)
/* Change this import statement to point to the credentials JSON file */
import creds from './sheetjs-test-726272627262.json' assert { type: "json" };

/* Change this to the spreadsheet ID */
const id = "SOME-SPREADSHEETJS-ID";

  • './sheetjs-test-726272627262.json' should be replaced with the name of the JSON file in step 27. The ./ prefix is required!

  • "SOME-SPREADSHEETJS-ID" should be replaced with the Document ID from step 33.

  1. Run the script:
node dump.mjs

The script should create a file SheetJSExport.xlsb in the project folder. This file can be opened in Excel.

Export Raw Files​

The goal of this section is to parse the Google Sheets XLSX export and generate CSV files for each worksheet.

  1. Sign into Google Sheets and open the "SheetJS Test" shared document.

  2. Click the Plus (+) icon in the lower left corner to create a new worksheet.

  3. In the new worksheet, set cell A1 to the formula =SEQUENCE(3,5). This will assign a grid of values

  4. Download raw.mjs:

curl -LO https://docs.sheetjs.com/gsheet/raw.mjs

Edit the marked lines near the top of the file:

raw.mjs (edit highlighted lines)
/* Change this import statement to point to the credentials JSON file */
import creds from './sheetjs-test-726272627262.json' assert { type: "json" };

/* Change this to the spreadsheet ID */
const id = "SOME-SPREADSHEETJS-ID";

  • './sheetjs-test-726272627262.json' should be replaced with the name of the JSON file in step 27. The ./ prefix is required!

  • "SOME-SPREADSHEETJS-ID" should be replaced with the Document ID from step 33.

  1. Run the script:
node raw.mjs

The script will display the sheet names and CSV rows from both worksheets:

#### Sheet1
Name,Index
Bill Clinton,42
GeorgeW Bush,43
Barack Obama,44
Donald Trump,45
Joseph Biden,46

#### Sheet14
1,2,3,4,5
6,7,8,9,10
11,12,13,14,15

Footnotes​

  1. See aoa_to_sheet in "Utilities" ↩

  2. See sheet_to_json in "Utilities" ↩

  3. See book_new in "Utilities" ↩

  4. See book_append_sheet in "Utilities" ↩

  5. See writeFile in "Writing Files" ↩

  6. See "Workbook Object" for more details. ↩

  7. See sheet_to_json in "Utilities" ↩

  8. See "Export MIME types for Google Workspace documents" in the Google Developer documentation for the complete list of supported file types. ↩

  9. See read in "Reading Files" ↩

  10. See "Workbook Object" for a description of the workbook object or "API Reference" for various methods to work with workbook and sheet objects. ↩