Managing data models

<h1 id="title">Managing data models</h1>

<p>
  In Atomik, all data models for EHR and Demographic content are defined by openEHR Operational Templates (OPTs).
</p>

<p>
  There are two ways of uploading an OPT into Atomik. The first is via the Web Console by using the uploader
  functionality. The second is via the REST API.
</p>

<p>
  Uploading OPTs by using the Web Console allows to upload several versions of the same template. All versions
  are stored, and only the last one is active. The Web Console also allows to deleted templates that you don't
  need anymore. Though if you have data store for a template, that tempalte couldn't be deleted. The delete is
  more for the design and development phases than for a production environment. In production we prefer to
  inactivate the template, which is also done from the Web Console.
</p>

<p>
  When creating templates from the openEHR REST API, it doesn't allow to upload several versions of the same
  template. If the template exists, then a 409 Conflict response will be sent to the client. That is in compliance
  with the openEHR REST API specifications. So when working with templates during design and development, and using
  the REST API only, for deleting templates the only option is to do it from the Web Console, since there is not
  delete operation yet at the REST API level.
</p>

<p>
  Internally, Atomik uses the OPTs to:
  <ol>
    <li>validate committed data</li>
    <li>index data</li>
    <li>create and execute queries</li>
  </ol>
</p>

<p>
  So a well defined template is crucial to make the most out of all those functionalities.
</p>