Serverless HTML to PDF

This repo presents an example (and a base starting project template) of generating PDF files from HTML templates using:

• SST (https://sst.dev) for provisioning and deployment
• AWS Lambda (https://aws.amazon.com/lambda/) for serverless execution
• Puppeteer (https://pptr.dev) for final PDF generation
• Creating your own AWS Lambda Layer from @sparticuz/chromium release, that should be selected based on Puppeteer version used (https://github.com/Sparticuz/chromium/releases/) - for compatibility read https://pptr.dev/supported-browsers
• Nunjucks (https://mozilla.github.io/nunjucks/) for HTML templating

Prerequisites

• Node.JS v16.16.0+
• (Optional) NVM for easy Node version switch/management
• npm v8.15.0+
• AWS Toolkit extension for your IDE
• npm install in root to pull currently supported version of SST and related packages

Getting Started

• Run npm run deploy to deploy the stack to the default stage and provision all required infrastructure - this step is recommended whenever you're using AWS SQS (SST dev increases lambdas' timeouts which will make your queue provisioning throw an exception when it hits consumer)
• Run npm run dev to start local development and hot-reloading
• Have fun

Important notes

This example uses a pre-built release of @sparticuz/chromium to create a new Lambda Layer on demand. If you wish to lower your repo size you may decide to externally deploy the layer and later reference it in your stack. Remember to update the stacks/MyStack.ts:10 to use the new layer, then bump the Puppeteer version accordingly