Export HTML to PDF Service

This is a simple Docker container that runs a JSON API service that allows HTML to be converted to PDF or PNG/JPG images. This service accomplishes this by using a Chrome headless browser to ensure full rendering capabilities on par with Google Chrome.

Security Note: This is intended to run as a micro service - do not directly expose to the public internet

Usage

docker run -p 2305:2305 bedrockio/export-html

Or:

git clone git@github.com:bedrockio/export-html.git
cd export-html
yarn install
yarn start

Generating a PDF

curl \
-d '{"html": "<h1>Hello World</h1>"}' \
-H "Content-Type: application/json" \
--output hello.pdf \
-XPOST "http://localhost:2305/1/pdf"

Now open hello.pdf

The default format is "Letter" (US) but it can be set to other paper formats like so:

curl \
-d '{"html": "<h1>Hello World</h1>", "export": {"format": "A4"}}' \
-H "Content-Type: application/json" \
--output hello-a4.pdf \
-XPOST "http://localhost:2305/1/pdf"

Generating a PNG

curl \
-d '{"html": "<h1>Hello World</h1>", "export": {"type": "png"}}' \
-H "Content-Type: application/json" \
--output hello.png \
-XPOST "http://localhost:2305/1/screenshot"

Now open hello.png

Advanced Options

Each API call allows Puppeteer options via body.export

• POST /1/pdf
• POST /1/screenshot

Kubernetes Deployment Notes

This module runs a full browser and each request will open a virtual broeser tab. Many concurrent requests can increase memory usage signicantly.

Here's an example of a Kubernetes deployment that limits resources (this is used in production for generating invoices):

apiVersion: apps/v1
kind: Deployment
metadata:
name: export-html-deployment
spec:
replicas: 3
selector:
matchLabels:
name: export-html
template:
metadata:
labels:
name: export-html
spec:
affinity:
podAntiAffinity:
preferredDuringSchedulingIgnoredDuringExecution:
- weight: 100
podAffinityTerm:
labelSelector:
matchExpressions:
- key: name
operator: In
values:
- export-html
topologyKey: kubernetes.io/hostname
containers:
- image: bedrockio/export-html
imagePullPolicy: Always
name: export-html
resources:
requests:
memory: "1000Mi"
cpu: "500m"
limits:
memory: "3000Mi"
cpu: "2500m"
env:
- name: NODE_ENV
value: "production"
- name: ENV_NAME
value: "production"
ports:
- name: http-server
containerPort: 2305
volumeMounts:
- name: export-html-cache
mountPath: /workdir/data
volumes:
- name: export-html-cache
emptyDir: {}

Credits

This service is based on the excellent Puppeteer module