Bumped kafka-python to 2.3.0 and test dependencies. Added unit tests for JSON deserialization logic.

Author: starichkov

requirements-dev.txt

@@ -1,4 +1,4 @@
-pytest==8.4.1
-pytest-cov==6.2.1
-coverage==7.10.6
-testcontainers[kafka]==4.12.0
+pytest==8.4.2
+pytest-cov==6.3.0
+coverage==7.10.7
+testcontainers[kafka]==4.13.3

requirements.txt

@@ -1 +1 @@
-kafka-python==2.2.15
+kafka-python==2.2.18

tests/test_serialization_json_format.py

@@ -0,0 +1,47 @@
+import json
+
+
+def test_json_deserializer_valid():
+    from serialization.json_format import json_deserializer
+
+    data = {"a": 1, "b": "text"}
+    raw = json.dumps(data).encode("utf-8")
+
+    result = json_deserializer(raw)
+
+    assert result == data
+
+
+def test_json_or_text_deserializer_json_ok():
+    from serialization.json_format import json_or_text_deserializer
+
+    payload = {"x": 123}
+    raw = json.dumps(payload).encode("utf-8")
+
+    out = json_or_text_deserializer(raw)
+
+    assert out == payload
+    assert isinstance(out, dict)
+
+
+def test_json_or_text_deserializer_invalid_json_falls_back_to_text():
+    from serialization.json_format import json_or_text_deserializer
+
+    raw = b"this is not json"
+
+    out = json_or_text_deserializer(raw)
+
+    assert out == "this is not json"
+    assert isinstance(out, str)
+
+
+def test_json_or_text_deserializer_invalid_utf8_replaced():
+    from serialization.json_format import json_or_text_deserializer
+
+    raw = b"\xff\xfe\xfd"  # invalid UTF-8
+
+    out = json_or_text_deserializer(raw)
+
+    # Should decode with replacement characters via plain_text_deserializer
+    assert isinstance(out, str)
+    assert "\ufffd" in out or "�" in out