Define a data model
Last updated
Was this helpful?
Last updated
Was this helpful?
Data modelling is the process of defining how data is used and stored within the system.
When configuring your data model on Extra Horizon, you will start by creating . In these schemas you describe how your data is structured.
When you've created a schema, you can add one or multiple documents to it. Each one of those documents need to conform to the schema you've defined. In developer terms, if you squint a little, you can look at schemas as being a class, and documents being the class instances of that class.
Depending on your data model, a single schema might not be enough. That's okay since you can create as many schema's as you want. Think in advance about how the documents of these schemas will be used & queried most efficiently in your application and design them accordingly.
For the purpose of the tutorial, we keep it simple. So we will create a schema called blood-pressure-measurement where each document represents a measurement.
In the tutorial repository, we'll be using the 1-data-model directory if you want to try this out.
We need to create a data model to support the creation of blood pressure measurements. Therefore we will create a Schema with the following properties:
systolic: A number defining the pressure during a heart beat
diastolic: A number defining the pressure in between heart beats
timestamp: The date and time the measurement was taken
For more detailed information regarding properties and their types please refer to the .
In the example below you see how we defined the properties for our blood pressure measurements. We also gave our data Schema a sensible name and a description. The name is important because that is the 'friendly name' which can be used to refer to the schema throughout your application implementation.
In the schemas directory of 1-data-model, you will find this model in blood-pressure.json.
We've added a single status created & made a creationTransition which just sets the initial status to created. Finally we list our properties under properties.
This is a global sync command available in the CLI, which will sync schemas, tasks, templates... you name it. It's a one-stop solution for syncing your entire project. By default, it makes a couple of assumptions about where these things are located. For example, it expects schemas to be in a schemas directory, tasks in a tasks directory, etc ...
For our schema, exh sync performs a couple of key operations:
Validation: It checks all JSON files within the schemas directory to verify that they are syntactically correct & consistent.
Synchronization: Once validated, it syncs the local schemas with the ExH data service. In this instance, a new schema titled blood-pressure-measurement will be created in the data service.
In the examples directory in the root of the repository, you can find the following scripts:
create-measurement.js: creates new measurements
get-measurement.js: retrieves a created measurement
delete-measurement.js: deletes an existing measurement
Please familiarize yourself with the contents of these script. They will show you how to use the SDK to create, read & delete blood pressure measurements.
Example: create a new measurement. The script will prompt to enter a blood pressure measurement.
And then retrieve the measurement you've just created:
Please refer to the documentation to learn more in detail about schemas.
We must now synchronize this schema to the ExH data service. This can be achieved by using the . In the 1-data-model directory, do:
This is fine for our purposes. If you want to customize this behaviour, your can create a repo-config.json file to change the paths where it will look for things. More information, see .