Separated extraction methods for subsections of the file

force_bdss/io/workflow_reader.py

 import json
+import logging
 
 from traits.api import HasStrictTraits, Instance
 
@@ -26,35 +27,61 @@ class WorkflowReader(HasStrictTraits):
 
     def read(self, file):
         json_data = json.load(file)
-        registry = self.bundle_registry
 
         try:
             version = json_data["version"]
         except KeyError:
+            logging.error("File missing version information")
             raise InvalidFileException("Corrupted input file, no version"
                                        " specified")
 
         if version not in SUPPORTED_FILE_VERSIONS:
+            logging.error(
+                "File contains version {} that is not in the "
+                "list of supported versions {}".format(
+                version, SUPPORTED_FILE_VERSIONS)
+            )
             raise InvalidVersionException(
                 "File version {} not supported".format(json_data["version"]))
 
         wf = Workflow()
 
+        try:
+            wf.multi_criteria_optimizer = self._extract_mco(json_data)
+            wf.data_sources[:] = self._extract_data_sources(json_data)
+            wf.kpi_calculators[:] = self._extract_kpi_calculators(json_data)
+        except KeyError as e:
+            logging.exception("Could not read file")
+            raise InvalidFileException("Could not read file. {}".format(e))
+
+    def _extract_mco(self, json_data):
+        registry = self.bundle_registry
+
         mco_id = json_data["multi_criteria_optimizer"]["id"]
         mco_bundle = registry.mco_bundle_by_id(mco_id)
-        wf.multi_criteria_optimizer = mco_bundle.create_model(
+        return mco_bundle.create_model(
             json_data["multi_criteria_optimizer"]["model_data"])
 
+    def _extract_data_sources(self, json_data):
+        registry = self.bundle_registry
+
+        data_sources = []
         for ds_entry in json_data["data_sources"]:
             ds_id = ds_entry["id"]
             ds_bundle = registry.data_source_bundle_by_id(ds_id)
-            wf.data_sources.append(ds_bundle.create_model(
-                ds_entry["model_data"]))
+            data_sources.append(ds_bundle.create_model(ds_entry["model_data"]))
+
+        return data_sources
 
+    def _extract_kpi_calculators(self, json_data):
+        registry = self.bundle_registry
+
+        kpi_calculators = []
         for kpic_entry in json_data["kpi_calculators"]:
             kpic_id = kpic_entry["id"]
             kpic_bundle = registry.kpi_calculator_bundle_by_id(kpic_id)
-            wf.kpi_calculators.append(kpic_bundle.create_model(
-                kpic_entry["model_data"]))
 
-        return wf
+            kpi_calculators.append(
+                kpic_bundle.create_model(kpic_entry["model_data"]))
+
+        return kpi_calculators