Developing on Openhab stack

Hello Openhab Devs,
am starting to work on a small project, but I need some advise. Below is the idea.

The users take some measurements using a special sensor.
Via the providers application, the data is uploaded to the providers server (accessible via REST Api)

Usually each user installs an openhab-android-app and can provide username and password

  • This this information should be important for the backend (e.g: logging in)…

Backend:

  • The backend (made up of a single binding) shall perform these tasks
    • The binding needs to access the username of each user
    • Download his/her data from the REST api of the provider
    • Store the data of the user in the DB
    • Perform a data mining analysis and store the result in the database.
    • Shall present that result to the user, when the user starts the frontend app.

Question:

  • Do my Idea make sense especially in relation with openhab?

  • The “openhab stack” is an advice from my supervisor, but I dont know whether it would have been better to create my own application?

  • Roughly, how do i go about the backend?

  • The openhab android app allows the user to enter username and password
    How do I access this information on the backend?

I will be very greatifull for any form of advise.
Thanks!

Did your supervisor mean OpenStack, instead of openHAB?

hello yancym,
thanks for a quick response.
Yes, he means openHAB. Previously he thought we should use Eclipse Smarthome (ESH). But it was very difficult to install this, as it had many errors.
Openhab has been around for a while, and he thought, it should be more stable than ESH.

OH will allow you to have different users but that is only for logging in. There is no way to correlate the data sent to OH to a specific logged in user through OH. You will have to implement that yourself.

So you will have to add passing that information as part of your Binding’s API. OH provides no way to get the name of the user who sent it data.

Data mining and analysis is not supported by OH. You will need external tools to do that. This will drive your choice in persistence engines as you will want to use something that external tools can access such as MySQL or InfluxDB.

It seems to be a pretty reasonable use of OH, though I don’t know if OH provides as much out of the box as you think/hope it will.

Personally I think it is good advice. It would probably be a lot more work to write your own app from scratch.

Start here. for the architecture of OH and review the “How to Develop a Binding” on the right hand side for how to make your own binding. However, depending on the nature of the data you are sending and receiving you may not need to implement a binding at all. OH has a REST API that you can use to push data to so you may just need to configure some Items and persistence and you entier server side (minus the analysis) would be done.

You can’t. That information will need to be added to the information sent to OH, or managed some other way.

Just for clarity, ESH is the basis for openHAB 2.