retrieve from catalog

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

@@ -12,16 +12,15 @@ logger = logging.getLogger(__name__)
 
 with KedroSession.create() as session:
     context = session.load_context()
-    catalog = context.get_catalog()
+#    catalog = context.get_catalog() # FIXME : ça porte à confusion de renvoyer un dict
+
 
 def parse_xml_collection(datasetcol: XMLDataSetCollection) -> XMLDataSetCollection:
     "node function entry point, performs batch processing"
     datasets = datasetcol.datasets
     housename = datasetcol._housename
-#    outputfolderpath = f"data/02_intermediate/houses/{housename}/xml"    
+    output_datasets = context.catalog.load(housename + '_xmlcontent')
-    output_catalog = catalog[housename + '_xmlcontent']
+    outputfolderpath = output_datasets._folderpath
-    outputfolderpath = output_catalog['folderpath']
-    output_datasets = XMLDataSetCollection(housename, str(outputfolderpath))
     for dataset_filenamestem, dataset in datasets.items():
         # a manual load is required here, because
         # the dataset **is not** registered in kedro's catalog
@@ -43,9 +42,8 @@ def make_json_collection(datasetcol: BsXMLDataSetCollection) -> JSONDataSetColle
     "node function entry point, performs batch processing"
     datasets = datasetcol.datasets
     housename = datasetcol._housename
-    output_catalog = catalog[housename + '_jsonoutput']
+    output_datasets = context.catalog.load(housename + '_jsonoutput')
-    outputfolderpath = output_catalog['folderpath']
+    outputfolderpath = output_datasets._folderpath
-    output_datasets = JSONDataSetCollection(housename, str(outputfolderpath))
     for dataset_filenamestem, dataset in datasets.items():
         # a manual load is required here, because
         # the dataset **is not** registered in kedro's catalog

actes-princiers/src/actesdataset.py

@@ -76,7 +76,6 @@ class BsXMLDataSet(XMLDataSet):
 
     def _load(self):
         "from the xml file, loads a internal xml repr (with bsoup)"
-        logger.info("------------------------- bsoup loader -------")
         with open(self._filepath, 'r', encoding="utf-8") as fhandle:
             self.soup = BeautifulSoup(fhandle, 'xml')
         ## xml.prettify() is the bsoup str(source_doc)
@@ -87,7 +86,6 @@ class BsXMLDataSet(XMLDataSet):
             json.dump(data, fp, sort_keys=True, indent=4)
 
     def transform(self):
-        logger.info("---------------- transform --------------")
         #soup = make_soup(os.path.join(folder, acte))
         # 1.1/ Get all data from XML (9). counter is the id (= numb_acte)
         numb = self.soup.TEI["xml:id"]  # /TEI[@xml:id] is always the acte's ID
@@ -169,7 +167,7 @@ class BsXMLDataSetCollection(DataSetCollection):
         return self
 
 
-class JSONDataSet: #(AbstractDataSet):
+class JSONDataSet:
     def __init__(self, filepath: str):
         self._filepath = filepath