Handle local parameters (again).

Two aspects to the fix:
* Load in the 'param promoted' version of the SBML.
* Use the new 'separator' argument of validate_target_xpaths from biosimulators_utils to get IDs like "J0_n" instead of just "n".

.github/workflows/ci.yml

@@ -125,6 +125,7 @@ jobs:
           fi
 
           echo "version=$version" >> $GITHUB_OUTPUT
+          echo "version=$version"
 
       - id: get-docker-image-tag
         name: Determine Docker image tag

biosimulators_tellurium/core.py

@@ -431,7 +431,8 @@ def preprocess_sed_task(task, variables, config=None, simulator_config=None):
                                   warning_summary='Model `{}` may be invalid.'.format(model.id))
 
         # read model
-        road_runner = roadrunner.RoadRunner(model.source)
+        road_runner = roadrunner.RoadRunner()
+        road_runner = roadrunner.RoadRunner(road_runner.getParamPromotedSBML(model.source))
 
         # get algorithm to execute
         algorithm_substitution_policy = get_algorithm_substitution_policy(config=config)
@@ -583,7 +584,7 @@ def get_model_change_target_tellurium_change_map(model_etree, changes, alg_kisao
     Returns:
         :obj:`dict`: dictionary that maps the targets of changes to their corresponding tellurium identifiers
     """
-    change_targets_to_sbml_ids = validation.validate_target_xpaths(changes, model_etree, attr='id')
+    change_targets_to_sbml_ids = validation.validate_target_xpaths(changes, model_etree, attr='id', separator="_")
 
     species_ids = model.getFloatingSpeciesIds() + model.getBoundarySpeciesIds()
     component_ids = species_ids + model.getGlobalParameterIds() + model.getCompartmentIds()
@@ -889,3 +890,4 @@ def exec_sed_doc_with_tellurium(doc, working_dir, base_out_path, rel_out_path=No
 
     # Return a data structure with the results of the reports
     return report_results, log
+

requirements.txt

@@ -1,6 +1,6 @@
-biosimulators_utils[logging] >= 0.1.187
-kisao >= 2.33
-libroadrunner >= 2.5.0
+biosimulators_utils[logging] >= 0.2.3
+kisao >= 2.34
+libroadrunner >= 2.7.0
 lxml
 numpy
 pandas