ontology to CSV converter

.python-version

@@ -0,0 +1 @@
+3.11.1/envs/dbo

BRANCH.md

@@ -0,0 +1,12 @@
+For managing our fork, there are primarily 3 branches we will be working with.
+
+1. upstream/master - This is the upstream package that we want to get all our public improvements in eventually
+2. main - This is our internal shadow of upstream/master, it should be 1:1. This is what we should make branches off of for changes we want to go upstream
+3. onboard-main - This is our final copy of digital buildings with all of the upstream changes, pending upstream changes, and any other changes that aren't compatible with the upstream/master
+
+Workflow for making improvements:
+1. Branch off of main
+2. make changes
+3. pull-request into upstream/master
+
+Upon acceptance and merge, we will update our internal main and subsequently update our onboard-main. If there are changes we NEED in onboard-main, we can accelerate part of the cycle and pull-request into onboard-main at the same time as we PR into upstream/master. When it does get accepted, it should result in a noop merge and everything will be healthy again. But complications will arise if changes get made to the PR after the PR lands in onboard-main, hence this not being the standard workflow.

tools/validators/instance_validator/ontology_dump.py

@@ -0,0 +1,61 @@
+#!/usr/bin/env python3
+import csv
+
+from validate import generate_universe
+
+universe = generate_universe.BuildUniverse()
+types_by_ns = universe.entity_type_universe.type_namespaces_map
+rows = []
+
+for namespace, entities in types_by_ns.items():
+  print(f"\n-- working on namespace '{namespace}'--\n")
+
+  # if namespace == 'HVAC':
+  #     __import__("IPython").embed()
+
+  for entity_name, entity in entities.valid_types_map.items():
+    print(f"Entity '{entity_name}'")
+
+    # if entity_name == 'AHU_BSPC_CO2M_DX4SC_ECON_EFSS_EFVSC_FDPM4X_HTSC_MOAFC_OAFC_SFSS_SFVSC_SSPC':
+    #     __import__("IPython").embed()
+
+    entity_row = {
+      'guid': entity.guid,
+      'namespace': namespace,
+      'name': entity_name,
+      'is_canonical': entity.is_canonical,
+      'is_abstract': entity.is_abstract,
+      'description': entity.description,
+      'parents': "|".join([p.typename for p in entity.parent_names.values()]),
+    }
+
+    # many canonical types don't define any new fields, they mix-in existing abstract
+    # types e.g. AHU_BSPC_CO2M_DX4SC_ECON_EFSS_EFVSC_FDPM4X_HTSC_MOAFC_OAFC_SFSS_SFVSC_SSPC
+    if not entity.local_field_names:
+      canonical_row = {
+        **entity_row,
+        'dbo.point_type': None,
+        'field_optional': None,
+      }
+      rows.append(canonical_row)
+
+    for field_name, field in entity.local_field_names.items():
+      increment = field.field.increment
+      field_row = {
+        **entity_row,
+        'dbo.point_type': field.field.field,
+        'dbo.point_type_increment': increment,
+        'field_optional': field.optional,
+      }
+      rows.append(field_row)
+
+if rows:
+  rows.sort(key=lambda r: r['guid'])
+
+  with open('flat_ontology.csv', 'w', newline='', encoding='utf-8') as f:
+    writer = csv.DictWriter(f, rows[0].keys(), delimiter=',', lineterminator='\n', quoting=csv.QUOTE_MINIMAL)
+    writer.writeheader()
+    writer.writerows(rows)
+else:
+  sys.stderr.write(
+    "No rows generated, ensure the validator is installed & updated (tools/validators/ontology_validator/README.md)")