fix slice_dictionary implementation

Author: tybug

aws_lambda_powertools/shared/functions.py

@@ -140,8 +140,9 @@ def powertools_debug_is_set() -> bool:
 
 
 def slice_dictionary(data: dict, chunk_size: int) -> Generator[dict, None, None]:
-    for _ in range(0, len(data), chunk_size):
-        yield {dict_key: data[dict_key] for dict_key in itertools.islice(data, chunk_size)}
+    for i in range(0, len(data), chunk_size):
+        yield {key: data[key] for key in itertools.islice(data, i, i + chunk_size)}
+
 
 
 def extract_event_from_common_models(data: Any) -> dict | Any:

tests/unit/test_shared_functions.py

@@ -11,6 +11,7 @@
 from aws_lambda_powertools.shared import constants
 from aws_lambda_powertools.shared.functions import (
     abs_lambda_path,
+    slice_dictionary,
     extract_event_from_common_models,
     powertools_debug_is_set,
     powertools_dev_is_set,
@@ -202,3 +203,14 @@ def test_sanitize_xray_segment_name_with_no_special_characters():
     # THEN the sanitized name remains the same as the original name
     expected_name = valid_name
     assert sanitized_name == expected_name
+
+
+@pytest.mark.parametrize("chunk_size, expected", [
+    (1, [{"k0": 0}, {"k1": 1}, {"k2": 2}, {"k3": 3}]),
+    (2, [{"k0": 0, "k1": 1}, {"k2": 2, "k3": 3}]),
+    (3, [{"k0": 0, "k1": 1, "k2": 2}, {"k3": 3}]),
+])
+def test_slice_dictionary(chunk_size, expected):
+    data = {f"k{i}": i for i in range(4)}
+    chunks = list(slice_dictionary(data, chunk_size=chunk_size))
+    assert chunks == expected