Set of microservices to autodeploy posts at cactus blog. Autodeployment is meant to be done by sending email message to a chosen email address.
In order to implement autocactus service I will use the following software stack. First of all, Python will be used. In the first stage of architecture development, the Celery task engine will be used. All received posts by email will be stored at MongoDB documents database.
In this section autocactus architecture will be presented. In the drawing below the sketch of architecture is presented. First of all received email will be filtered against a whiteliste of allowed senders. Any other email will be deleted immediately. Next checker microservice will parse email and create new object in the MongoDB database. For now I think that object will have following fields: title, author, content, tags, publish date, creation date. In this stage of work I am not considering attachments. The role of the first microservice ends here. Next, every hour publisher microservice will be awaken to check if there is any post to publish. If there are some new posts, they will be transformed into files under GIT repository and will be published with Cactus.