[FEATURE] Add support for file_encoding in sources

blackrock · May 28, 2024 · 277d1d6 · 277d1d6
1 parent 924dfa4
commit 277d1d6
Show file tree

Hide file tree

Showing 7 changed files with 52 additions and 14 deletions.
diff --git a/ingen/data_source/file_source.py b/ingen/data_source/file_source.py
@@ -46,7 +46,7 @@ def fetch_data(self, reader):
         """
         returns a DataFrame of data fetched from input FileSource.
         """
-        return reader.read(self._src)
+        return reader.read(self._src, encoding=self._src['file_encoding'])
 
     def fetch_validations(self):
         """
@@ -64,4 +64,6 @@ def format_file_path(self, source, params_map):
         path_parser = PathParser(run_date)
         if 'file_path' in source:
             source['file_path'] = path_parser.parse(source['file_path'])
+        if not source.get('file_encoding'):
+            source['file_encoding'] = 'utf-8'
         return source
diff --git a/ingen/reader/file_reader.py b/ingen/reader/file_reader.py
@@ -101,8 +101,8 @@ def read(self, src):
 
 
 def get_config(src):
-    header_size = src.get('skip_header_size' , 0)
-    trailer_size = src.get('skip_trailer_size' , 0)
+    header_size = src.get('skip_header_size', 0)
+    trailer_size = src.get('skip_trailer_size', 0)
     all_cols = src.get('columns')
     return {
         "header_size": header_size,
@@ -115,13 +115,13 @@ class ReaderFactory:
 
     @classmethod
     def get_reader(cls, src):
-        factory_types = {'delimited_file': CSVFileReader,
-                         'excel': ExcelFileReader,
-                         'xml': XMLFileReader,
-                         'json': JSONFileReader,
-                         "fixed_width": FixedWidthFileReader
-                         }
+        factory_types = {
+            'delimited_file': CSVFileReader,
+            'excel': ExcelFileReader,
+            'xml': XMLFileReader,
+            'json': JSONFileReader,
+            'fixed_width': FixedWidthFileReader
+        }
         reader_cls = factory_types.get(src.get('file_type'))
         if reader_cls:
             return reader_cls()
-
diff --git a/ingen/reader/xml_file_reader.py b/ingen/reader/xml_file_reader.py
@@ -13,7 +13,7 @@
 class XMLFileReader:
 
     def read(self, src):
-        xml_file = open(src['file_path'], 'r')
+        xml_file = open(src['file_path'], 'r', encoding=src['file_encoding'])
         try:
             data = xmltodict.parse(xml_file.read())
             tree = et.parse(src['file_path'])

diff --git a/test/data_source/test_file_source.py b/test/data_source/test_file_source.py
@@ -20,8 +20,7 @@ def setUp(self):
             'file_path': 'test',
             'skip_header_size': 1,
             'skip_trailer_size': 1,
-            'columns': ['col1', 'col2']
-
+            'columns': ['col1', 'col2'],
         }
         self.params_map = {'query_params': {'table_name': 'positions'}, 'infile': {}}
         self.source = FileSource(self._src, self.params_map)

diff --git a/test/input/test_utf_16.xml b/test/input/test_utf_16.xml
diff --git a/test/metadata/test_metadata.py b/test/metadata/test_metadata.py
@@ -131,7 +131,7 @@ def test_output_type(self):
         output = self.metadata.output
         self.assertEqual("file", output["type"])
 
-    def test_output_type(self):
+    def test_split_output_type(self):
         self.setUp({"use_split_file_config": True})
         output = self.metadata.output
         self.assertEqual("splitted_file", output["type"])
@@ -239,6 +239,20 @@ def test_sources_with_dynamic_data(self):
         self.assertIsInstance(metadata.sources[1], JsonSource)
         self.assertEqual("json_payload", metadata.sources[1]._id)
 
+    def test_default_file_encoding(self):
+        metadata = MetaData(
+            self.test_md_name, self.test_md_configurations, self.params_map
+        )
+        self.assertEqual("utf-8", metadata.sources[0]._src["file_encoding"])
+
+    def test_custom_file_encoding(self):
+        metadata_config = self.test_md_configurations.copy()
+        metadata_config["sources"][0]["file_encoding"] = "utf-16"
+        metadata = MetaData(
+            self.test_md_name, metadata_config, self.params_map
+        )
+        self.assertEqual("utf-16", metadata.sources[0]._src["file_encoding"])
+
 
 if __name__ == "__main__":
     unittest.main()
diff --git a/test/reader/test_xml_file_reader.py b/test/reader/test_xml_file_reader.py
@@ -77,6 +77,19 @@ def setUp(self):
             ],
             'root_tag': 'ORDER'
         }
+        self.xml_src_utf_16_encoding = {
+            'id': 'order_xml',
+            'type': 'file',
+            'file_type': 'xml',
+            'file_encoding': 'utf-16',
+            'file_path': f'{THIS_DIR.parent}/input/test_utf_16.xml',
+            'columns': [
+                'ORDER_ID',
+                'ORD_DETAIL_set.ORD_DETAIL.ITEM_NAME',
+                'ORD_DETAIL_set.ORD_DETAIL.QUANTITY'
+            ],
+            'root_tag': 'ORDER'
+        }
 
     def test_xml(self):
         source = self.xml_src
@@ -133,6 +146,16 @@ def test_empty_xml_file(self, mock_logging):
             error_msg = "XML file is empty"
             mock_logging.error.assert_called_with(error_msg)
 
+    def test_xml_utf_16_encoding(self):
+        source = self.xml_src_utf_16_encoding
+        reader = XMLFileReader()
+        data = reader.read(source)
+        keys = {}
+        expected_data = pd.DataFrame(
+            keys,
+            columns=source['columns'])
+        pd.testing.assert_frame_equal(expected_data, data)
+
 
 if __name__ == '__main__':
     unittest.main()