Implement new logic and endpoints for running processors / workflows #5
Description
Description
We should move away from calling processors through single endpoints (see #4) and instead implement workflow runs based on the output workflow files defined through in the NodeFlow editor. This makes it easier to start more complex workflows as well as keeping track of existing workflow results.
Workflows will run a defined order of processors on a specific branch of a project, using an existing snapshot as input. Every workflow will create a new branch (or multiple new branches when branched workflows are introduced).
Proposed Endpoints
POST /api/v1.0/workflow/save
Serializes and saves a workflow created in the NodeFlow editor (or alternatively uploaded through the UI)
GET /api/v1.0/workflow/{id}
Retrieves a workflow to load it into the NodeFlow editor for editing
GET /api/v1.0/workflow/list
Retrieves all available workflows
POST /api/v1.0/branch/run
Takes the parameters branchID / snapshot and workflowID and start a new run of a workflow on the specified branch and snapshot