Raise an error when mapped column is not found (dfurtado#35)

Raise an error when mapped column is not found

Improve the error message when a mapped column is not found in the CSV
file. 
Added additional tests to reproduce two different scenarios, the first
when a mapped column is not found and the second and a non-mapped column
is not found.

Author: dfurtado

dataclass_csv/__init__.py

@@ -38,9 +38,4 @@
 from .exceptions import CsvValueError
 
 
-__all__ = [
-    'DataclassReader',
-    'dateformat',
-    'accept_whitespaces',
-    'CsvValueError',
-]
+__all__ = ['DataclassReader', 'dateformat', 'accept_whitespaces', 'CsvValueError']

dataclass_csv/dataclass_reader.py

@@ -32,7 +32,6 @@ def __init__(
         self.optional_fields = self._get_optional_fields()
         self.field_mapping = {}
 
-
         self.reader = csv.DictReader(
             f, fieldnames, restkey, restval, dialect, *args, **kwds
         )
@@ -60,19 +59,19 @@ def _get_default_value(self, field):
         )
 
     def _get_possible_keys(self, fieldname, row):
-        possible_keys = list(
-            filter(lambda x: x.strip() == fieldname, row.keys())
-        )
+        possible_keys = list(filter(lambda x: x.strip() == fieldname, row.keys()))
         if possible_keys:
             return possible_keys[0]
 
     def _get_value(self, row, field):
+        is_field_mapped = False
+
         try:
-            key = (
-                field.name
-                if field.name not in self.field_mapping.keys()
-                else self.field_mapping.get(field.name)
-            )
+            if field.name in self.field_mapping.keys():
+                is_field_mapped = True
+                key = self.field_mapping.get(field.name)
+            else:
+                key = field.name
 
             if key in row.keys():
                 value = row[key]
@@ -85,14 +84,15 @@ def _get_value(self, row, field):
             if field.name in self.optional_fields:
                 return self._get_default_value(field)
             else:
-                raise KeyError(
-                    f'The value `{field.name}` is missing in the CSV file.'
-                )
+                keyerror_message = f'The value for the column `{field.name}` is missing in the CSV file.'
+                if is_field_mapped:
+                    keyerror_message = f'The value for the mapped column `{key}` is missing in the CSV file'
+                raise KeyError(keyerror_message)
         else:
             if not value and field.name in self.optional_fields:
                 return self._get_default_value(field)
             elif not value and field.name not in self.optional_fields:
-                raise ValueError((f'The field `{field.name}` is required.'))
+                raise ValueError(f'The field `{field.name}` is required.')
             elif (
                 value
                 and field.type is str
@@ -144,9 +144,7 @@ def _process_row(self, row):
             try:
                 value = self._get_value(row, field)
             except ValueError as ex:
-                raise CsvValueError(
-                    ex, line_number=self.reader.line_num
-                ) from None
+                raise CsvValueError(ex, line_number=self.reader.line_num) from None
 
             if not value and field.default is None:
                 values.append(None)
@@ -161,19 +159,15 @@ def _process_row(self, row):
                 or '__origin__' in field_type.__dict__
                 and field_type.__origin__ is Union
             ):
-                real_types = [
-                    t for t in field_type.__args__ if t is not type(None)
-                ]
+                real_types = [t for t in field_type.__args__ if t is not type(None)]
                 if len(real_types) == 1:
                     field_type = real_types[0]
 
             if field_type is datetime:
                 try:
                     transformed_value = self._parse_date_value(field, value)
                 except ValueError as ex:
-                    raise CsvValueError(
-                        ex, line_number=self.reader.line_num
-                    ) from None
+                    raise CsvValueError(ex, line_number=self.reader.line_num) from None
                 else:
                     values.append(transformed_value)
                     continue
@@ -186,9 +180,7 @@ def _process_row(self, row):
                         else strtobool(str(value).strip()) == 1
                     )
                 except ValueError as ex:
-                    raise CsvValueError(
-                        ex, line_number=self.reader.line_num
-                    ) from None
+                    raise CsvValueError(ex, line_number=self.reader.line_num) from None
                 else:
                     values.append(transformed_value)
                     continue
@@ -219,7 +211,5 @@ def map(self, csv_fieldname):
         :param csv_fieldname: The name of the CSV field
         """
         return FieldMapper(
-            lambda property_name: self._add_to_mapping(
-                property_name, csv_fieldname
-            )
+            lambda property_name: self._add_to_mapping(property_name, csv_fieldname)
         )

tests/mocks.py

@@ -108,3 +108,10 @@ def __init__(self, val):
 class UserWithSSN:
     name: str
     ssn: SSN
+
+
+@dataclasses.dataclass
+class UserWithEmail:
+    name: str
+    email: str
+

tests/test_dataclass_reader.py

@@ -14,6 +14,7 @@
     UserWithDefaultDatetimeField,
     UserWithSSN,
     SSN,
+    UserWithEmail,
 )
 
 
@@ -173,7 +174,7 @@ def test_reader_with_optional_types(create_csv):
         list(reader)
 
 def test_reader_with_datetime_default_value(create_csv):
-    csv_file = create_csv({'name': 'User', 'bithday': ''})
+    csv_file = create_csv({'name': 'User', 'birthday': ''})
 
     with csv_file.open() as f:
         reader = DataclassReader(f, UserWithDefaultDatetimeField)
@@ -198,3 +199,23 @@ def test_should_parse_user_defined_types(create_csv):
 
         assert isinstance(items[1].ssn, SSN)
         assert items[1].ssn.val == '123-45-6789'
+
+
+def test_raise_error_when_mapped_column_not_found(create_csv):
+    csv_file = create_csv({'name': 'User1', 'e-mail': '[email protected]'})
+
+    with csv_file.open() as f:
+        with pytest.raises(KeyError, match='The value for the mapped column `e_mail` is missing in the CSV file'):
+            reader = DataclassReader(f, UserWithEmail)
+            reader.map('e_mail').to('email')
+            list(reader)
+
+
+def test_raise_error_when_field_not_found(create_csv):
+    csv_file = create_csv({'name': 'User1', 'e-mail': '[email protected]'})
+
+    with csv_file.open() as f:
+        with pytest.raises(KeyError, match='The value for the column `email` is missing in the CSV file.'):
+            reader = DataclassReader(f, UserWithEmail)
+            list(reader)
+