scope-pm4py 0.1.1

PM4Py loaders for Scope OCPT and OCPN JSON exports

scope-pm4py

Small Python adapter package for loading Scope OCPT and OCPN JSON exports into PM4Py runtime objects.

Install

python -m pip install scope-pm4py

For local development from this directory:

python -m pip install -e .

Usage

from scope_pm4py import load_ocpt, load_ocpn
import pm4py

tree = load_ocpt(r"C:\Users\Postb\Downloads\scope_ocpt_example_pm4py.json")
pm4py.view_process_tree(tree)

ocpn = load_ocpn(r"C:\Users\Postb\Downloads\scope_ocpn_example_pm4py.json")
pm4py.view_ocpn(ocpn, format="svg")

OCPT Object-Type Projection

Scope OCPT activity nodes store object-type metadata in PM4Py process-tree node properties. You can project a loaded OCPT tree to the behavior of a single object type:

from scope_pm4py import load_ocpt, load_ocpt_for_object_type, project_ocpt_to_object_type

orders_tree = load_ocpt_for_object_type(
    r"C:\Users\Postb\Downloads\scope_ocpt_example_pm4py.json",
    "Orders",
)

full_tree = load_ocpt(r"C:\Users\Postb\Downloads\scope_ocpt_example_pm4py.json")
orders_tree = project_ocpt_to_object_type(full_tree, "Orders")

The projection returns a new PM4Py ProcessTree and does not mutate the input tree. Activities are kept when the requested object type appears in node._properties["related_object_types"]; unrelated activities are projected to silent tau behavior and simplified where this is control-flow safe.

Expected JSON

The loader accepts files with a top-level payload object or files where the payload is the top-level object.

OCPT exports should contain:

{
  "schema": "scope.pm4py.ocpt",
  "schema_version": "1.0.0",
  "payload": {
    "tree": {
      "type": "operator",
      "operator": "sequence",
      "children": [
        { "type": "activity", "label": "Create Order" },
        { "type": "tau" }
      ]
    }
  }
}

OCPN exports should contain:

{
  "schema": "scope.pm4py.ocpn",
  "schema_version": "1.0.0",
  "payload": {
    "object_types": ["order"],
    "activities": ["Create Order"],
    "petri_nets": {
      "order": {
        "places": [
          { "id": "p1", "name": "source", "initial": true },
          { "id": "p2", "name": "sink", "final": true }
        ],
        "transitions": [
          { "id": "t1", "name": "Create Order", "label": "Create Order" }
        ],
        "arcs": [
          { "source": "p1", "target": "t1" },
          { "source": "t1", "target": "p2" }
        ]
      }
    }
  }
}

The OCPN loader reconstructs PM4Py PetriNet and Marking objects. A plain json.load() result is not enough for PM4Py visualization and analysis.

Build and Publish

Build the package:

python -m pip install build twine
python -m build
python -m twine check dist/*

Upload:

python -m twine upload dist/*