cleaning

actes-princiers/src/actes_princiers/pipelines/xml_processing/nodes.py

@@ -23,15 +23,21 @@ def parse_xml_collection(datasets: Dict[str, EtreeXMLDataSet], param: str) -> Di
         # a manual load is required here, because
         # the dataset **is not** registered in kedro's catalog
         dataset._load()
+        descr = dataset._describe()
+        logger.info(f"dataset {descr} loaded")
+#        logger.info(str(dataset._describe()))
+#        logger.info(dataset.source_doc)
         output_source_doc = transform(dataset.get_source_doc(), param)
         # set dataset's output filepath
         output_filepath = dataset.filepath.replace("01_raw", "02_intermediate")
         output_xmldataset = EtreeXMLDataSet(output_filepath)
-        output_xmldataset.set_source_doc(output_source_doc)
-        output_datasets[dataset_filenamestem] = output_xmldataset
+
         # let's create subfolders now, if they don't exist
         output_filepath = Path(output_filepath)
         output_xmldataset_dir = output_filepath.parent
         output_xmldataset_dir.mkdir(parents=True, exist_ok=True)
+
+        output_xmldataset._save(output_source_doc)
+        output_datasets[dataset_filenamestem] = output_xmldataset
     return output_datasets
 

actes-princiers/src/actesdataset.py

@@ -83,14 +83,6 @@ class XMLDataSetCollection(AbstractDataSet):
         self._housename = housename
         self._folderpath = Path(folderpath)
             
-    def get_datasets(self) -> Dict[str, Any]:
-        "datasets mapper getter"
-        if hasattr(self, 'datasets'):
-            return self.datasets
-        else:
-            attr_error_msg = str(self._describe())
-            raise AttributeError(f"Object {attr_error_msg} has no attribute named : 'datasets'")
-            
     def _load(self) -> dict[str, EtreeXMLDataSet]:
         "kedro's API loader method"
         self.datasets = dict()
@@ -99,10 +91,14 @@ class XMLDataSetCollection(AbstractDataSet):
                 filepath=str(filepath))
         return self.datasets
 
-    def _save(self, datasets: dict[str, Any]) -> None:
-        "kedro's API saver method"
-        for stemfilename, dataset in datasets.items():
-            dataset._save(dataset.get_source_doc())
+    def _save(self, data) -> None:
+        """kedro's API saver method
+        
+         There is **nothing to save**, because  
+         this dataset collections is a *container* dataset.
+         this method is here only because kedro requires it.
+         """ 
+        pass
         
     def _describe(self) -> dict[str, Any]:
         "kedro's API repr()"