added new column to bpmn process definition to store the full bpmn process hash digest w/ burnettk

spiffworkflow-backend/migrations/versions/5d8e49f9c560_.py

@@ -0,0 +1,36 @@
+"""empty message
+
+Revision ID: 5d8e49f9c560
+Revises: 0b5dd14bfbac
+Create Date: 2023-04-17 11:28:39.714193
+
+"""
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import mysql
+
+# revision identifiers, used by Alembic.
+revision = '5d8e49f9c560'
+down_revision = '0b5dd14bfbac'
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    with op.batch_alter_table('bpmn_process_definition', schema=None) as batch_op:
+        batch_op.alter_column('hash', existing_type=sa.String(length=255), new_column_name='single_process_hash')
+        batch_op.add_column(sa.Column('full_process_model_hash', sa.String(length=255), nullable=True))
+        batch_op.create_unique_constraint(None, ['full_process_model_hash'])
+
+    # ### end Alembic commands ###
+
+
+def downgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    with op.batch_alter_table('bpmn_process_definition', schema=None) as batch_op:
+        batch_op.drop_constraint('full_process_model_hash', type_='unique')
+        batch_op.drop_column('full_process_model_hash')
+        batch_op.alter_column('single_process_hash', existing_type=sa.String(length=255), new_column_name='hash')
+
+    # ### end Alembic commands ###

spiffworkflow-backend/src/spiffworkflow_backend/models/bpmn_process_definition.py

@@ -21,7 +21,11 @@ class BpmnProcessDefinitionModel(SpiffworkflowBaseDBModel):
     # note that a call activity is its own row in this table, with its own hash,
     # and therefore it only gets stored once per version, and can be reused
     # by multiple calling processes.
-    hash: str = db.Column(db.String(255), nullable=False, unique=True)
+    single_process_hash: str = db.Column(db.String(255), nullable=False, unique=True)
+
+    # only the top level parent will have this set
+    # it includes all subprocesses and call activities
+    full_process_model_hash: str | None = db.Column(db.String(255), nullable=True, unique=True, default=None)
 
     bpmn_identifier: str = db.Column(db.String(255), nullable=False, index=True)
     bpmn_name: str = db.Column(db.String(255), nullable=True, index=True)

spiffworkflow-backend/src/spiffworkflow_backend/services/process_instance_processor.py

@@ -959,18 +959,31 @@ class ProcessInstanceProcessor:
         process_bpmn_properties: dict,
         bpmn_process_definition_parent: Optional[BpmnProcessDefinitionModel] = None,
         store_bpmn_definition_mappings: bool = False,
+        full_bpmn_spec_dict: Optional[dict] = None,
     ) -> BpmnProcessDefinitionModel:
         process_bpmn_identifier = process_bpmn_properties["name"]
         process_bpmn_name = process_bpmn_properties["description"]
-        new_hash_digest = sha256(json.dumps(process_bpmn_properties, sort_keys=True).encode("utf8")).hexdigest()
+
-        bpmn_process_definition: Optional[BpmnProcessDefinitionModel] = BpmnProcessDefinitionModel.query.filter_by(
+        bpmn_process_definition: Optional[BpmnProcessDefinitionModel] = None
-            hash=new_hash_digest
+        single_process_hash = sha256(json.dumps(process_bpmn_properties, sort_keys=True).encode("utf8")).hexdigest()
+        full_process_model_hash = None
+        if full_bpmn_spec_dict is not None:
+            full_process_model_hash = sha256(
+                json.dumps(full_bpmn_spec_dict, sort_keys=True).encode("utf8")
+            ).hexdigest()
+            bpmn_process_definition = BpmnProcessDefinitionModel.query.filter_by(
+                full_process_model_hash=full_process_model_hash
+            ).first()
+        else:
+            bpmn_process_definition = BpmnProcessDefinitionModel.query.filter_by(
+                single_process_hash=single_process_hash
             ).first()
 
         if bpmn_process_definition is None:
             task_specs = process_bpmn_properties.pop("task_specs")
             bpmn_process_definition = BpmnProcessDefinitionModel(
-                hash=new_hash_digest,
+                single_process_hash=single_process_hash,
+                full_process_model_hash=full_process_model_hash,
                 bpmn_identifier=process_bpmn_identifier,
                 bpmn_name=process_bpmn_name,
                 properties_json=process_bpmn_properties,
@@ -1050,6 +1063,7 @@ class ProcessInstanceProcessor:
         bpmn_process_definition_parent = self._store_bpmn_process_definition(
             bpmn_spec_dict["spec"],
             store_bpmn_definition_mappings=store_bpmn_definition_mappings,
+            full_bpmn_spec_dict=bpmn_spec_dict,
         )
         for process_bpmn_properties in bpmn_spec_dict["subprocess_specs"].values():
             self._store_bpmn_process_definition(