wkhtmltopdf in a docker container as a web service.

This image is based on the wkhtmltopdf container.

Run the container with docker run and binding the ports to the host. The web service is exposed on port 80 in the container.

docker run -d -P openlabs/docker-wkhtmltopdf-aas

The container now runs as a daemon.

Find the port that the container is bound to:

docker port 071599a1373e 80

where 071599a1373e is the container SHA that docker assigned when docker run was executed in the previous command.

or

docker ps

To obtain some information about de container, which include de ip and the port where the container runs. Take a note of the public port number where docker binds to.

There are multiple ways to generate a PDF of HTML using the service.

This is a convenient way to use the service from command line utilities like curl.

curl -X POST -vv -F 'fichs=@path/to/local/file.html' http://<docker-host>:<port>/ -o path/to/output/file.pdf

where:

• docker-host is the hostname or address of the docker host running the container
• port is the public port to which the container is bound to.

To add multiple html files simply add them to the curl command:

curl -X POST -vv -F 'fichs=@file1.html' -F 'fichs=@file2.html' -F 'fichs=@file3.html' http://<docker-host>:<port>/ -o path/to/output/file.pdf

If you are planning on using this service in your application, it might be more convenient to use the JSON API that the service uses.

Here is an example using python requests:

import json import requests url = 'http://<docker_host>:<port>/' data = { 'fichs': { 'f1' : open('/file/to/file1.html').read().encode('base64') } } headers = { 'Content-Type': 'application/json', # This is important } response = requests.post(url, data=json.dumps(data), headers=headers) # Save the response contents to a file with open('/path/to/local/file.pdf', 'wb') as f: f.write(response.content.decode('base64'))

Here is another example in python, but this time we pass multiple files and options to wkhtmltopdf. When passing our settings we omit the double dash "--" at the start of the option. For documentation on what options are available, visit http://wkhtmltopdf.org/usage/wkhtmltopdf.txt

import json import requests url = 'http://<docker_host>:<port>/' data = { 'fichs': { 'f1' : open('/file/to/file1.html').read().encode('base64'), 'f2' : open('/file/to/file2.html').read().encode('base64'), 'f3' : open('/file/to/file3.html').read().encode('base64') }, 'options': { #Omitting the "--" at the start of the option 'margin-top': '6', 'margin-left': '6', 'margin-right': '6', 'margin-bottom': '6', 'page-width': '105mm', 'page-height': '40mm', 'enable-javascript': '' } } headers = { 'Content-Type': 'application/json', # This is important } response = requests.post(url, data=json.dumps(data), headers=headers) # Save the response contents to a file with open('/path/to/local/file.pdf', 'wb') as f: f.write(response.content.decode('base64'))

• Implement conversion of URLs to PDF
• Add documentation on passing options to the service
• Add curl example for JSON api
• Explain more gunicorn options

The development of the container takes place on Github. If you have a question or a bug report to file, you can report as a github issue.

This image was built at Openlabs.

This image is professionally supported by Openlabs. If you are looking for on-site teaching or consulting support, contact our sales and support teams.