Stars: 154
Forks: 16
Pull Requests: 48
Issues: 10
Watchers: 6
Last Updated: 2023-09-17 13:13:15
A Sidecar function to run Browsershot on Lambda.
License: MIT License
Languages: PHP, JavaScript, Shell
https://stefanzweifel.dev/posts/2022/06/21/introducing-sidecar-browsershot
This package allows you to run Browsershot on AWS Lambda through Sidecar.
You won't need to install Node, Puppeteer or Google Chrome on your server. The heavy lifting of booting a headless Google Chrome instance is happening on AWS Lambda.
This package requires that spatie/browsershot
and hammerstone/sidecar
have been installed in your Laravel application.
Follow their installation and configuration instructions. (You can skip the installation of puppeteer and Google Chrome for Browsershot though.)
You can install the package via composer:
composer require wnx/sidecar-browsershot
You can publish the config file with:
php artisan vendor:publish --tag="sidecar-browsershot-config"
Register the BrowsershotFunction::class
in your sidecar.php
config file.
/*
* All of your function classes that you'd like to deploy go here.
*/
'functions' => [
\Wnx\SidecarBrowsershot\Functions\BrowsershotFunction::class,
],
Deploy the Lambda function by running:
php artisan sidecar:deploy --activate
See Sidecar documentation for details.
You can use BrowsershotLambda
like the default Browsershot
-class coming from the Spatie package.
All you need to do is replace Browsershot
with BrowsershotLambda
.
use Wnx\SidecarBrowsershot\BrowsershotLambda;
// an image will be saved
BrowsershotLambda::url('https://example.com')->save($pathToImage);
// a pdf will be saved
BrowsershotLambda::url('https://example.com')->save('example.pdf');
// save your own HTML to a PDF
BrowsershotLambda::html('<h1>Hello world!!</h1>')->save('example.pdf');
// Get HTML of a URL and store it on a given disk
$html = BrowsershotLambda::url('https://example.com')->bodyHtml();
Storage::disk('s3')->put('example.html', $html);
sidecar-browsershot supports warming for faster execution.
To enable this feature set the SIDECAR_BROWSERSHOT_WARMING_INSTANCES
variable in your .env
to the desired number of instances Sidecar should warm for you.
SIDECAR_BROWSERSHOT_WARMING_INSTANCES=5
Alternatively you can publish the sidecar-browsershot.php
config file and change the warming
setting yourself.
You can store an HTML file on AWS S3 and pass the path to Lambda for it to create the PDF or image from. This is necessary for large source files in order to avoid restrictions on the size of Lambda requests.
use Wnx\SidecarBrowsershot\BrowsershotLambda;
// Use an HTML file from S3 to generate a PDF
BrowsershotLambda::readHtmlFromS3('html/example.html')->save('example.pdf');
// You can also pass a disk name if required (default: 's3')
BrowsershotLambda::readHtmlFromS3('html/example.html', 's3files')->save('example.pdf');
You can store your file directly on AWS S3 if you want to keep it there, or to avoid the size limit on Lambda responses.
You just need to pass a path and optional disk name (default: 's3') to the saveToS3
method.
use Wnx\SidecarBrowsershot\BrowsershotLambda;
// an image will be saved on S3
BrowsershotLambda::url('https://example.com')->saveToS3('example.jpg');
// a pdf will be saved on S3
BrowsershotLambda::url('https://example.com')->saveToS3('example.pdf');
// save your own html to a PDF on S3
BrowsershotLambda::html('<h1>Hello world!!</h1>')->saveToS3('example.pdf', 'example-store');
Like the original Browsershot package, you can manipulate the image size and format.
octicon-info mr-2" viewBox="0 0 16 16" version="1.1" width="16" height="16" aria-hidden="true">
If you're using fit()
in combination with saveToS3
, the image will be downloaded from S3 to your local disc, manipulated and then uploaded back to S3.
// Take screenshot at 1920x1080 and scale it down to fit 200x200
BrowsershotLambda::url('https://example.com')
->windowSize(1920, 1080)
->fit(Manipulations::FIT_CONTAIN, 200, 200)
->save('example.jpg');
// Take screenshot at 1920x1080 and scale it down to fit 200x200 and save it on S3
// Note: To do the image manipulation, BrowsershotLambda will download the image
// from S3 to the local disc of your app, manipulate it and then upload it back to S3.
BrowsershotLambda::url('https://example.com')
->windowSize(1920, 1080)
->fit(Manipulations::FIT_CONTAIN, 200, 200)
->saveToS3('example.jpg');
The testsuite makes connections to AWS and runs the deployed Lambda function. In order to run the testsuite, you will need an active AWS account.
We can use the native sidecar:configure
artisan command to create the necessary AWS credentials for Sidecar. First copy the testbench.example.yaml
file to testbench.yaml
.
Then run ./vendor/bin/testbench sidecar:configure
to start the Sidecar setup process. (You only have to do the setup once)
cp testbench.example.yaml testbench.yaml
cp .env.example .env
./vendor/bin/testbench sidecar:configure
After finishing the Sidecar setup process, you will have received a couple of SIDECAR_*
environment variables. Add these credentials to .env
.
Now we can deploy our local BrowsershotFunction
to AWS Lambda. Run the following command in your terminal, before executing the testsuite.
./vendor/bin/testbench sidecar-browsershot:setup
After the successful deployment, you can run the testsuite.
composer test
Please see CHANGELOG for more information on what has changed recently.
Please see CONTRIBUTING for details.
Please review our security policy on how to report security vulnerabilities.
The MIT License (MIT). Please see License File for more information.