Overview

In a nutshell, hugo-mx-gateway provides a RESTful POST endpoint for static contact/demo request pages. It's a simple, yet a powerful tool built for this only-designated purpose.

This project is at its early open source life, but with a production deployment. The documentation is still being refining (notably for wording and typos). Contributions is any form (code/docs/ideas...) are welcomed.

Table of contents

• Overview
• Table of contents
• Why hugo-mx-gateway
  • How does it work
  • Prerequisites
• Deployment
  • Deployment on Google Appp Engine
  • Deployment on a Kubernetes cluster
  • Deployment on Docker
• Use the built-in form for Hugo
• Configuration variables
• License & Copyrights
• Support & Contributions

Why hugo-mx-gateway

Did you ever experience building a static website (e.g. using Hugo or whatever alternative), then did get stuck when coming the time to add a contact/demo request form?

You're at the right place.

This project, namely hugo-mx-gateway, is meant to provide a RESTful API that adds the dynamicity required to handle contact/demo requests for static web sites. It's a simple, yet a powerful tool built for this only-designated purpose.

How does it work

hugo-mx-gateway is built upon a simple request handling workflow:

• Create an HTML form with a POST action pointing towards the hugo-mx-gateway service. This service is a RESTful POST endpoint backed by an application easily deployable on Google App Engine, Kubernetes, or Docker.
• For each user request, hugo-mx-gateway automatically retrieves information submitted by the user (email, subject, message details...), then generates and sends a templated email to the user-provided email address, while bccing a copy of that email to an address that you can define for internal tracking and follow up.
• Once a request is processed (upon success or failure), hugo-mx-gateway handles the reply back towards the calling static page by redirecting the browser to the origin page with additional URL parameters describing the completion status of the processing (e.g. /contact.html?status=success&message=request%20submitted). The parameters can then be easily retrieved and shown to the user, e.g. with a few lines of Javascript within the static page.
• The project is shipped with a sample HTML form intending to cover a basic use case of contact and demo requests. That said, this is a open source software, so you're free to adapt it for your specific use cases.

Prerequisites

hugo-mx-gateway is easily deployable in minutes, subject to fulfil the following requirements:

• Deployment platform (at your choice): Google App Engine, Kubernetes, or Docker.
• An SMTP account with the server supporting TLS.

These are basic requirements in the today's Internet world.

While it can be tempting to use a personal email account, it's highly recommended for security and privacy reasons not do that for an internet-hosted service. Therefore, according to your situation, you may need to ask for an SMTP account to your corporate's IT staff. If you're free to take the decision, you can opt for a cloud-based SMTP service like SendGrid, Mailgun, and Mailjet--they all offer free tier for small usage.

Deployment

Consider one of the following resources according to the selected installation platform.

Deployment on Google Appp Engine

This requires to have an active GCP account and Google Cloud SDK (gcloud) installed on your work station.

• Create/select a GCP project to deploy hugo-mx-gateway.

  Note that each GCP project can hold only a single App Engine instance. Several applications can be co-hosted as services for the root App Engine instance. In this case, a new application has to be declared as service in the app.yaml file.

• Create the Google App Engine configuration file

  cp app.yaml.sample app.yaml
  

• Open the app.yaml file with your favorite editor.

• Edit the configuration variables as described here.

• Start the deployment

  make deploy-gcp
  

• Check that hugo-mx-gateway is up and functioning

  curl https://PROJECT-ID.REGION.r.appspot.com/healthz 
  

  Replace PROJECT-ID with the GCP project ID, and REGION with the deployment region.

Deployment on a Kubernetes cluster

There is a Helm chart to ease the deployment on Kubernetes clusters.

Important: The chart is validated with Helm 3 and the pod is run in an unprivileged mode within a Security Context.

Proceed with the deployment as follows:

• First edit the values.yaml file to set configuration values appropriately.

• Choose a deployment namespace. In the sample commands provided hereafter, it's assumed that the target namespace is hugo-mx-gateway. If you opt for another namespace, do consider to adapt the commands accordingly.

• Apply the deployment with Helm

  helm upgrade \
    --namespace hugo-mx-gateway \
    --install hugo-mx-gateway \
    helm/
  

• Check that the application is up and running.

  kubectl -n hugo-mx-gateway port-forward service/hugo-mx-gateway 8080:80
  curl https://127.0.0.1:8080/healthz 
  

Deployment on Docker

An instance of hugo-mx-gateway can be quickly started on any machine running Docker.

• Please first review the configuration variables and adapt them appropriately.

• Then apply the following command:

  $ docker run -d \
    --publish 8080:8080 \
    --name '`hugo-mx-gateway`' \
    -e SMTP_SERVER_ADDR="smtp.example.com:465" \
    -e SMTP_VERITY_CERT=true \
    -e SMTP_CLIENT_USERNAME="postmaster@example.com" \
    -e SMTP_CLIENT_PASSWORD="postmasterSecretPassWord" \
    -e CONTACT_REPLY_EMAIL="noreply@example.com" \
    -e CONTACT_REPLY_BCC_EMAIL="contact@example.com" \
    -e DEMO_URL="https://demo.example.com/" \
    -e ALLOWED_ORIGINS="127.0.0.1,example.com" \
    rchakode/`hugo-mx-gateway`
  

• Check that the container is up and functionning.

  curl https://127.0.0.1:8080/healthz 
  

Use the built-in form for Hugo

The file ./samples/hugo-partial-contact-form.html contains a sample HTML form for Hugo. It can be used for both contact and demo requests.

Open the file in a your favorite editor and review it.

Notice that the form is configured to be rendered specifically according to a Hugo parameter named tags, which is actually a list of tags. If the parameter holds a tag named contact then, the form will be rendered as a contact form. Otherwise, it'll be rendered as a demo form.

The integration works as follows:

• Copy the HTML form content in your target Hugo HTML template.
• Modify the <form> tag to make the action point to the URL of the sendmail backend deployed previously.
  • On Google App Engine, the endpoint shall be: https://PROJECT-ID.REGION.r.appspot.com/sendmail. Replace PROJECT-ID and REGION, repectively, with the GCP's project ID and the deployment region.
  • On Kubernetes, the in-cluster endpoint shall be: http://hugo-mx-gateway.hugo-mx-gateway.svc.cluster.local/sendmail
  • On Docker, the endpoint shall be: http://DOCKER-HOST:8080/sendmail. Replace DOCKER-HOST with the IP adress or the hostname of the Docker machine.
• Edit the Hugo Markdown content of the target contact/demo page to ensure that the tags parameter holds a appropriate value (i.e. contact for a contact form, or demo for a demo request form).

Here is an example of header for a Hugo Markdown page. The tags parameter holds a tag named contact) meaning that the page will be rendered as a contact request form.

---
title: "Contact Us"
description: "Contact request page"
date: 2020-04-25T00:20:27+02:00
tags: [contact]
---

Configuration variables

Regardless of the deployment platform (Google App Engine, Kubernetes, Docker), the following configuration parameters must be provided as environment variables:

• SMTP_SERVER_ADDR: Set the address of the SMTP server in the form of host:port. It's required that the SMTP server being supporting TLS.
• SMTP_VERITY_CERT: Tell whether the SMTP certificate should be validated against top level authorities. If you're using a self-signed certificate on the SMTP server, this value must be set to false.
• SMTP_CLIENT_USERNAME: Set the username to connect to the SMTP server.
• SMTP_CLIENT_PASSWORD: Set the password to connect to the SMTP server.
• CONTACT_REPLY_EMAIL: Set an email address for the reply email. It's not necessary a valid email address, for example if don't want the user to reply you can use something like noreply@example.com.
• CONTACT_REPLY_BCC_EMAIL: Sets an email address for bcc copy of the email sent to the user. This is useful for tracking and follow up.
• DEMO_URL: Specific for demo forms, it can be used to set the URL of the demo site that will be included to the user reply email (e.g. https://demo.example.com/).
• ALLOWED_ORIGINS: Set a list of comma-separated domains that the hugo-mx-gateway App shoudl trust. This is for security reason to filter requests. Only requests with an Origin header belonging to the defined origins will be accepted, through it's only required that the request has a valid Referer header. It's expected in the future to these request filtering and admission rules.
• TEMPLATE_DEMO_REQUEST_REPLY (optional): Specify the path of the template to reply a demo request. The default templare is templates/template_reply_demo_request.html
• TEMPLATE_CONTACT_REQUEST_REPLY (optional): Specify the path of the template to reply a contact request. The default templare is templates/template_reply_contact_request.html.

License & Copyrights

This tool (code and documentation) is licensed under the terms of Apache 2.0 License. Read the LICENSE file for more details on the license terms.

The tool may inlcude third-party libraries provided with their owns licenses and copyrights, but always compatible with the Apache 2.0 License terms of use.

Support & Contributions

We encourage feedback and do make our best to handle any troubles you may encounter when using this tool.

Here is the link to submit issues: https://github.com/rchakode/hugo-mx-gateway/issues.

New ideas are welcomed, please open an issue to submit your idea if you have any one.

Contributions are accepted subject that the code and documentation be released under the terms of Apache 2.0 License.

To contribute bug patches or new features, please use the Github Pull Request model.