Systems and repositories

The automation code lives in the company's git repositories. In Proton, each system (what is automated: a portal, an API, SAP) points to the repository with its code, and the repository uses a git integration so Proton can read the code. All three are in Settings.

The path, from scratch:

1. In Settings > Integrations, register the integration with GitHub, GitLab or Bitbucket, with the token. See Integrations.
2. In Settings > Git Repository, register the repository and the default branch.
3. In Settings > System, link the system to the repository.
4. Register the components in the system. When you start a run, choose the branch for each system, and the runner downloads the code from it.

Git Repository

The cards show the owner and the repository, the platform, the integration, the default branch and the BDD badge when the repository has Gherkin automations. The filters split them by GitHub, GitLab and Bitbucket.

Add opens Create Git Repository:

1. Select the platform: only the ones with a registered integration show up.
2. In Integration, choose the integration. The token comes from it.
3. Under General, fill in the owner (Owner on GitHub, the workspace on Bitbucket, the URL on GitLab) and the Repository.
4. Use Get Branches (or Validate Project, on GitLab) and choose the Default Branch from the list. On GitHub, the search already selects the repository's default branch.
5. Check This repository contains automations written in Gherkin if that's the case, and save.

With no git integration, the window warns you and takes you to Register Integration.

When you delete a repository, the systems linked to it are left without a repository.

How the runner uses the repository

During the run, the runner receives each system of the components with the repository and the branch chosen when the run was started, asks Proton for the integration token and downloads the code. The commit used is recorded as the code version of the run, in the run details. Proton only reads the repositories: it doesn't write to the company's git, except for the proton-bdd branch in repositories with BDD on GitHub.

System

The cards show the name, the platform and the repository with its default branch, or that the system has no repository. The With repository and Without repository filters separate the two cases.

Add opens Create System:

• System: the name, required, up to 50 characters and unique in the organization.
• Git Repository: Select Git Repository chooses the repository (the X unlinks it).

Several systems can use the same repository. Changing a system's repository changes where the next runs download the code from; check the branch of the recurring runs that use the system.

A system with components or recurring runs can't be deleted: move the components first.

Impact analysis

The card's Impact analysis icon shows what depends on the system before a change, in the Components, Processes, Automations and Groups tabs, with alerts for high or critical criticality processes and datasets that failed in their last run. What the team rule hides shows up only as a count.

Settings

Settings > Settings stores the organization's Configuration Parameter and Value pairs (up to 50 characters each, with no repeated parameter), with who saved them last and when. Clicking the parameter or the value copies it.

The values are stored in plain text: don't store passwords or tokens here. For the data the automation uses in each environment, such as addresses and users, use Environments.

Permissions and mobile

System, Git Repository, Settings and Integrations have their own permissions (view, create, edit and delete) in the profile's Configuration tab. Choosing a system's repository requires viewing repositories.

The screens are designed for desktop.