refactoring

actes-princiers/conf/base/parameters.yml

@@ -1 +1 @@
-xlststylesheet: static/xsl/actes_princiers.xsl
+xlststylesheet: templates/xsl/actes_princiers.xsl

actes-princiers/src/actes_princiers/customcontext.py

@@ -2,7 +2,6 @@ from pathlib import Path
 from typing import Dict
 
 from kedro.framework.context import KedroContext
-#from kedro.pipeline import Pipeline
 from kedro.pipeline import Pipeline, node, pipeline
 
 from actesdataset import XMLDataSet
@@ -14,36 +13,6 @@ def tree(directory, relative_to=None):
         trees[path.stem] = str(path.relative_to(relative_to))
     return trees
 
-def house_dataset_loader(catalog):
-    nodes_description = []
-    # FIXME : a custom DataSet Catalog that lists
-    # input_catalog = catalog.load(house_name)
-    # FIXME : set root path from config, not here 
-    # or make an autopath function helper
-    data_root_path = Path.cwd() / 'data' / '01_raw' / 'xml'
-    relative_to = Path.cwd()
-    for dataset_name, dataset_path in tree(data_root_path, relative_to=relative_to).items():
-        # dataset_path : data/01_raw/xml/bourbon/brb_ch_i_1437_08_18a.xml
-        catalog.add(data_set_name=dataset_name, 
-            data_set=XMLDataSet(filepath=dataset_path),
-            replace=True)
-        # adding an output catalog entry
-        output_dataset_name = dataset_name + "_output"
-        # FIXME : JE NE SUIS PAS SATISFAIT
-        output_dataset_path = Path(dataset_path.replace("01_raw", "02_intermediate"))
-        # let's create subfolders if they don't exist
-        output_dataset_dir = output_dataset_path.parent
-        output_dataset_dir.mkdir(parents=True, exist_ok=True)
-        catalog.add(data_set_name=output_dataset_name, 
-            data_set=XMLDataSet(filepath=output_dataset_path),
-            replace=True)
-        # prepare information for the next stage (the pipeline stage)
-        node_description = dict(
-            inputs=dataset_name,
-            outputs=output_dataset_name,
-            name=dataset_name)
-        nodes_description.append(node_description)
-    return nodes_description
 
 class ProjectContext(KedroContext):
     project_name = "actes princiers"
@@ -63,9 +32,37 @@ class ProjectContext(KedroContext):
         catalog = super()._get_catalog(*args, **kwargs)
         # kedro.io.data_catalog.DataCatalog
         # adding data sets
-        self.nodes_description = house_dataset_loader(catalog)
+        self.nodes_description = self._house_dataset_loader(catalog)
         return catalog 
 
+    def _house_dataset_loader(self, catalog):
+        nodes_description = []
+        # XXX : get root path from config, not here 
+        data_root_path = Path.cwd() / 'data' / '01_raw' / 'xml'
+        relative_to = Path.cwd()
+        for dataset_name, dataset_path in tree(data_root_path, relative_to=relative_to).items():
+            # dataset_path : data/01_raw/xml/bourbon/brb_ch_i_1437_08_18a.xml
+            catalog.add(data_set_name=dataset_name, 
+                data_set=XMLDataSet(filepath=dataset_path),
+                replace=True)
+            # adding an output catalog entry
+            output_dataset_name = dataset_name + "_output"
+            # XXX : make better
+            output_dataset_path = Path(dataset_path.replace("01_raw", "02_intermediate"))
+            # let's create subfolders if they don't exist
+            output_dataset_dir = output_dataset_path.parent
+            output_dataset_dir.mkdir(parents=True, exist_ok=True)
+            catalog.add(data_set_name=output_dataset_name, 
+                data_set=XMLDataSet(filepath=output_dataset_path),
+                replace=True)
+            # prepare information for the next stage (the pipeline stage)
+            node_description = dict(
+                inputs=dataset_name,
+                outputs=output_dataset_name,
+                name=dataset_name)
+            nodes_description.append(node_description)
+        return nodes_description
+
     def prepare_pipeline_creation(self):
         return self.nodes_description