Sync pyrog mapping with a github repo ?

[nriss]

Some times, we have two environments :

• production, with real world data
• test, with fake data.

The test environment for example is outside the hospital and the prod environment is inside the hospital.

When we want to develop a web app, it is easier to work outside the hospital, but the data are more consistant in the prod environment.

So, the mapping will be done on the prod env and we want to update our test env with the same mapping.

One possibility would be to bind pyrog to a github repository to pull / push mapping versions without having to export / import it.

[MiskoG]

This is the situation more or less as I understand it. "P" stands for Pyrog

I have a few thoughts / questions

• How do you handle this situation today ?
• "When we want to develop a web app, it is easier to work outside the hospital" => Could you be more precise about "the work" happening ? Does it have to be done right in the env (dev or prod) or could it be done from anywhere, with an access token or something ?

[nriss]

1. Today, we have to choose a source of truth, it can be the prod or test environment. We develop the mapping on it. Then, regularly, we have to export the mapping and push it to github.

If the source of truth is prod for example, to update test mapping, we have to delete the source, reset the database and reimport the mapping

2. The access to the VM / data from croissant env for example:

• The VM is accessible using windows (internet explorer is mandatory)
• Oqapi access depends on the VPN connection, Alexandre doesn't have access to oqapi for example with its connection.

The easier is to work on another environment with test data, on which we have 100% control (like on scaleway)