HTML2PDF

A serverless function for generating PDF documents from URLs, Pug and HTML. Based on RelaxedJS & Puppeteer.
Provides quick setup & deployment to AWS Lambda & Google Cloud Functions.

Usage

AWS

Requirements

• AWS Lambda Nodejs runtime 10.x, 12.x
• Docker
• File-watching utility such as nodemon or entr

Function Payload

The function expects a JSON payload. Your payload must include the following values:

• type - The type of value provided. Allowed values are url and string.
• format - The format of the value if type is string. Accepted values are pug and html.
• value - The raw value can be a URL, HTML or Pug.

For example:

{
  "type": "url",
  "format": null,
  "value": "https://google.com"
}

Installing dependencies

docker run --rm --workdir=/app \
  -e PUPPETEER_SKIP_CHROMIUM_DOWNLOAD=true \
  -v "$PWD":/app \
  -v "$PWD"/nodejs/node_modules:/app/node_modules \
  lambci/lambda:build-nodejs12.x \
  bash -c "npm i -g yarn && yarn install && yarn autoclean --force"

Emulate AWS Lambda hosting the function

find ./ -name "*.js*" -not -path "*node_modules*" | \
  entr -cr docker run --rm \
  -e DOCKER_LAMBDA_STAY_OPEN=1 \
  -e AWS_LAMBDA_FUNCTION_NAME=html2pdf \
  -p 9001:9001 \
  -v "$PWD":/var/task:ro,delegated \
  -v "$PWD"/nodejs/node_modules:/opt/nodejs/node_modules:ro,delegated \
  lambci/lambda:nodejs12.x html2pdf.main

Invoke the function locally

aws lambda invoke \
  --endpoint http://localhost:9001 
  --no-sign-request \
  --function-name=html2pdf \
  --invocation-type=RequestResponse \
  --payload $(echo '{"type": "string", "format": "pug", "value": "h1 Hello World!\np My name is Zuko"}' | base64 ) \
  output.json