Merge pull request #399 from nodestream-proj/uppercase-normalizer

feat: add an uppercase normalizer to complement lowercase

Author: zprobst

nodestream/pipeline/normalizers/__init__.py

@@ -2,11 +2,13 @@
 from .normalizer import InvalidFlagError, Normalizer
 from .remove_trailing_dots import RemoveTrailingDots
 from .trim_whitespace import TrimWhitespace
+from .uppercase_strings import UppercaseStrings
 
 __all__ = (
-    "Normalizer",
+    "InvalidFlagError",
     "LowercaseStrings",
+    "Normalizer",
     "RemoveTrailingDots",
-    "InvalidFlagError",
     "TrimWhitespace",
+    "UppercaseStrings",
 )

nodestream/pipeline/normalizers/uppercase_strings.py

@@ -0,0 +1,8 @@
+from typing import Any
+
+from .normalizer import Normalizer
+
+
+class UppercaseStrings(Normalizer, alias="uppercase_strings"):
+    def normalize_value(self, value: Any) -> Any:
+        return value.upper() if isinstance(value, str) else value

tests/unit/pipeline/normalizers/test_uppercase_strings.py

@@ -0,0 +1,21 @@
+import pytest
+from hamcrest import assert_that, equal_to
+
+from nodestream.pipeline.normalizers import UppercaseStrings
+
+
+@pytest.mark.parametrize(
+    "input_value,expected_value",
+    [
+        ("SomERandomCaps", "SOMERANDOMCAPS"),
+        ("ALL_CAPS", "ALL_CAPS"),
+        ("all_lower", "ALL_LOWER"),
+        ("123456", "123456"),
+        ("", ""),
+        (None, None),
+        ([1, 2, 3], [1, 2, 3]),
+    ],
+)
+def test_uppercase_strings_normalization(input_value, expected_value):
+    subject = UppercaseStrings()
+    assert_that(subject.normalize(input_value), equal_to(expected_value))