@matthewfritz/node-html-to-image

v1.2.0

Published

5 months ago

A Node.js library that generates images from HTML, supporting concurrency and Puppeteer clustering control

0High
0Medium
0Low

NOTE: I forked this from frinyvonnick/node-html-to-image - Release v5.0.0 (original tag commit | original tag release | up-to-date diff) for the express purpose of refactoring the concurrency and Puppeteer clustering functionality.

My fork is intended to allow the calling logic / service to have greater control over the Puppeteer operations and not auto-terminate clusters and processes to promote reuse in distributed environments. It is also intended to have drop-in parity with the way the original package functions by default even if the concurrency updates are not being leveraged.

The original usage license has also remained unmodified to ensure there is no sudden change in criteria when using this fork versus the original package.

Much of the documentation has also remained the same with some reorganization and additive changes also made to describe new functionality.

-- @matthewfritz

A Node.js library that generates images from HTML

🏠 Homepage

Description

This module exposes a function that generates images (png, jpeg) from HTML. It uses puppeteer in headless mode to achieve it. Additionally, it embarks Handlebars to provide a way to add logic in your HTML.

Install

npm install @matthewfritz/node-html-to-image
# or
yarn add @matthewfritz/node-html-to-image

Note: When you install Puppeteer, it downloads a recent version of Chromium (~170MB Mac, ~282MB Linux, ~280MB Win) that is guaranteed to work with the API.

Usage

TypeScript support

The library is written in Typescript so it is available out of the box:

import nodeHtmlToImage from 'node-html-to-image'

Options

List of all available original options:

| option | description | type | required | |-------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------|-------------| | output | The output path for generated image | string | optional | | html | The html used to generate image content | string | required | | type | The type of the generated image | jpeg or png (default: png) | optional | | quality | The quality of the generated image (only applicable to jpg) | number (default: 80) | optional | | content | If provided html property is considered an handlebars template and use content value to fill it | object or Array | optional | | waitUntil | Define when to consider markup succeded. Learn more. | string or Array (default: networkidle0) | optional | | puppeteer | The puppeteer property let you use a different puppeteer library (like puppeteer-core or puppeteer-extra). | object (default: puppeteer) | optional | | puppeteerArgs | The puppeteerArgs property let you pass down custom configuration to puppeteer. Learn more. | object | optional | | beforeScreenshot | An async function that will execute just before screenshot is taken. Gives access to puppeteer page element. | Function | optional | | transparent | The transparent property lets you generate images with transparent background (for png type). | boolean | optional | | encoding | The encoding property of the image. Options are binary (default) or base64. | string | optional | | selector | The selector property lets you target a specific element to perform the screenshot on. (default body) | string | optional | | handlebarsHelpers | The handlebarsHelpers property lets add custom logic to the templates using Handlebars sub-expressions. Learn more. | object | optional | | timeout | Timeout for a puppeteer-cluster (in ms). Defaults to 30000 (30 seconds). | number | optional |

List of all available added options with set defaults to allow drop-in parity with original package:

| option | description | type | required | |-------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------|-------------| | cluster | A running puppeteer-cluster instance to use instead of spawning a new one by default within the nodeHtmlToImage() call | Cluster | optional | | clusterOptions | An object containing configuration options for the default puppeteer-cluster instance spawned when not supplying the cluster argument | object | optional | | triggerClusterIdleAfterScreenshots | Whether to call the idle() method on the cluster after the screenshot process has finished successfully | boolean (default: true) | optional | | triggerClusterCloseAfterScreenshots | Whether to call the close() method on the cluster after the screenshot process has finished successfully | boolean (default: true) | optional | | triggerClusterCloseOnError | Whether to call the close() method on the cluster after an error has occurred during the screenshot operation | boolean (default: true) | optional | | triggerProcessExitOnError | Whether to call process.exit(1) to terminate the process after an error has occurred during the screenshot operation | boolean (default: true) | optional | | errorLogLinePrefix | A string prefix to write within the console line logged when an error has occurred during the screenshot operation | string | optional | | errorLogLineAdditionalData | Any additional data to write within the console line logged when an error has occurred during the screenshot operation | any | optional |

Examples

Simple example

const nodeHtmlToImage = require('node-html-to-image')

nodeHtmlToImage({
  output: './image.png',
  html: '<html><body>Hello world!</body></html>'
})
  .then(() => console.log('The image was created successfully!'))

Setting output image resolution

node-html-to-image takes a screenshot of the body tag's content. If you want to set output image's resolution you need to set its dimension using CSS like in the following example.

const nodeHtmlToImage = require('node-html-to-image')

nodeHtmlToImage({
  output: './image.png',
  html: `<html>
    <head>
      <style>
        body {
          width: 2480px;
          height: 3508px;
        }
      </style>
    </head>
    <body>Hello world!</body>
  </html>
  `
})
  .then(() => console.log('The image was created successfully!'))

Example with Handlebars

Handlerbars is a templating language. It generates HTML from a template and an input object. In the following example we provide a template to node-html-to-image and a content object to fill the template.

const nodeHtmlToImage = require('node-html-to-image')

nodeHtmlToImage({
  output: './image.png',
  html: '<html><body>Hello {{name}}!</body></html>',
  content: { name: 'you' }
})
  .then(() => console.log('The image was created successfully!'))

Handlebars provides a lot of expressions to handle common use cases like conditions or loops.

Using Handlebars helpers

Handlerbars sub-expressions can be used to add custom logic to the templates. To do this, you must pass a handlebarsHelpers object with functions defined within.

For example, if you had a variable and wanted to do some conditional rendering depending on its value, you could do this:

const nodeHtmlToImage = require('node-html-to-image')

nodeHtmlToImage({
  output: './image.png',
  content: { myVar: 'foo' },
  handlebarsHelpers: {
    equals: (a, b) => a === b,
  },
  html: `
    <html>
      <body>
        {{#if (equals myVar 'foo')}}<div>Foo</div>{{/if}}
        {{#if (equals myVar 'bar')}}<div>Bar</div>{{/if}}
      </body>
    </html>`
  
})

Dealing with images

If you want to display an image which is stored remotely do it as usual. In case your image is stored locally I recommend having your image in base64. Then you need to pass it to the template with the content property. Here is an example:

const nodeHtmlToImage = require('node-html-to-image')
const fs = require('fs');

const image = fs.readFileSync('./image.jpg');
const base64Image = new Buffer.from(image).toString('base64');
const dataURI = 'data:image/jpeg;base64,' + base64Image

nodeHtmlToImage({
  output: './image.png',
  html: '<html><body><img src="{{{imageSource}}}" /></body></html>',
  content: { imageSource: dataURI }
})

Dealing with fonts

If you want to apply fonts, you need to synchronize your parts loading of your website. One way doing it is to convert your font to base64 and add it to your style in your html. For example:

const font2base64 = require('node-font2base64')

const _data = font2base64.encodeToDataUrlSync('../my/awesome/font.ttf')

const html = `
<html>
  <head>
    <style>
      @font-face {
        font-family: 'testFont';
        src: url("{{{_data}}}") format('woff2'); // don't forget the format!
      }
    </style>
  </head>
...

Using the buffer instead of saving to disk

If you don't want to save the image to disk and would rather do something with it immediately, you can use the returned value instead! The example below shows how you can generate an image and send it back to a client via using express.

const express = require('express');
const router = express.Router();
const nodeHtmlToImage = require('node-html-to-image');

router.get(`/api/tweet/render`, async function(req, res) {
  const image = await nodeHtmlToImage({
    html: '<html><body><div>Check out what I just did! #cool</div></body></html>'
  });
  res.writeHead(200, { 'Content-Type': 'image/png' });
  res.end(image, 'binary');
});

Generating multiple images

If you want to generate multiple images in one call you must provide an array to the content property.

Saving to disk

To save on the disk you must provide the output property on each object in the content property.

nodeHtmlToImage({
  html: '<html><body>Hello {{name}}!</body></html>',
  content: [{ name: 'Pierre', output: './image1.png' }, { name: 'Paul', output: './image2.png' }, { name: 'Jacques', output: './image3.png' }]
})
  .then(() => console.log('The images were created successfully!'))

Using buffers

If you don't want to save the images to disk you can use the returned value instead. It returns an array of Buffer objects.

const images = await nodeHtmlToImage({
  html: '<html><body>Hello {{name}}!</body></html>',
  content: [{ name: 'Pierre' }, { name: 'Paul' }, { name: 'Jacques' }]
})

Using different puppeteer libraries

If you want to use different puppeteer library you must provide the puppeteer property.

const chrome = require('chrome-aws-lambda');
const nodeHtmlToImage = require('node-html-to-image')
const puppeteerCore = require('puppeteer-core');

const image = await nodeHtmlToImage({
  html: '<html><body><div>Hello</div></body></html>',
  puppeteer: puppeteerCore,
  puppeteerArgs: {
      args: chromium.args,
      executablePath: await chrome.executablePath,
  }
})

Handling errors instead of terminating the process automatically

You may want to handle errors directly instead of terminating the process by default when a problem has occurred.

For that, we use the triggerProcessExitOnError boolean option and then wrap the call in a try...catch statement:

const nodeHtmlToImage = require('node-html-to-image')

try {
  const image = await nodeHtmlToImage({
    html: '<html><body><div>Hello</div></body></html>',

    // suppress automatic process.exit(1) behavior on error
    triggerProcessExitOnError: false,
  });
} catch (err) {
  // handle the error without terminating the process automatically
}

Using your own puppteer-cluster instance

Sometimes you don't want to spawn the default puppeteer-cluster instance during the nodeHtmlToImage() call but rather use your own.

The way we handle this is to supply the cluster option as well set the three triggerCluster booleans to false.

This means you can have complete control over the concurrent behavior and configuration of the cluster itself:

const { Cluster } = require('puppeteer-cluster');
const nodeHtmlToImage = require('node-html-to-image');

(async () => {
  // start the cluster manually before generating the screenshot
  // https://github.com/thomasdondorf/puppeteer-cluster#clusterlaunchoptions
  const screenshotCluster = await Cluster.launch({
    concurrency: Cluster.CONCURRENCY_CONTEXT,
    maxConcurrency: 2,
  });

  // take the screenshot with our custom cluster
  const image = await nodeHtmlToImage({
    html: '<html><body><div>Hello</div></body></html>',
    cluster: screenshotCluster,

    // suppress automatic cluster idling behavior on success
    triggerClusterIdleAfterScreenshots: false,

    // suppress automatic cluster closing behavior on success
    triggerClusterCloseAfterScreenshots: false,

    // suppress automatic cluster closing behavior on error
    triggerClusterCloseOnError: false,
  });

  // idle and close the cluster manually
  await cluster.idle();
  await cluster.close();
})();

You can of course opt to leave out the booleans and allow nodeHtmlToImage() to provide that default functionality. However, the caveat is that you will not be able to re-use the cluster and therefore not be able to support more in-depth distributed operations as a result, as you'll need to start a fresh cluster and therefore a fresh Puppeteer instance each time you want to call the function.

You can also combine this with handling errors directly instead of exiting the process to gain additional control when the screenshot operation has failed:

const { Cluster } = require('puppeteer-cluster');
const nodeHtmlToImage = require('node-html-to-image');

(async () => {
  // start the cluster manually before generating the screenshot
  // https://github.com/thomasdondorf/puppeteer-cluster#clusterlaunchoptions
  const screenshotCluster = await Cluster.launch({
    concurrency: Cluster.CONCURRENCY_CONTEXT,
    maxConcurrency: 2,
  });

  try {
    // attempt to take the screenshot with our custom cluster
    const image = await nodeHtmlToImage({
      html: '<html><body><div>Hello</div></body></html>',
      cluster: screenshotCluster,

      // suppress automatic cluster idling behavior on success
      triggerClusterIdleAfterScreenshots: false,

      // suppress automatic cluster closing behavior on success
      triggerClusterCloseAfterScreenshots: false,

      // suppress automatic cluster closing behavior on error
      triggerClusterCloseOnError: false,

      // suppress automatic process.exit(1) behavior on error
      triggerProcessExitOnError: false,
    });
  } catch (err) {
    // handle the error gracefully
  }

  // idle and close the cluster manually
  await cluster.idle();
  await cluster.close();
})();

Run tests

yarn test

Libraries

node-html-to-image-cli - CLI for this module

Articles

Author

Custom Concurrency + Cluster Control

👤 Matthew Fritz [email protected]

GitHub: @matthewfritz

Original Package

👤 FRIN Yvonnick [email protected]

Show your support

Give a ⭐️ if this project helped you!

📝 License

The original version of this README was generated with ❤️ by readme-md-generator