feat: add safe repr implementations for watsonx and litellm backends by cptnm3 · Pull Request #1359 · generative-computing/mellea

cptnm3 · 2026-06-30T15:31:59Z

Pull Request

Issue

Fixes #1317

Description

Added __repr__ and __str__ to show useful information with masked api_key

Testing

Tests added to the respective file if code was changed
New code has 100% coverage if code was added
Ensure existing tests and github automation passes (a maintainer will kick off the github automation when the rest of the PR is populated)

Attribution

AI coding assistants used

Adding a new component, requirement, sampling strategy, or tool?

If your PR adds or modifies one of the types below, check the matching box. A checklist of type-specific review items will be posted as a comment.

Component
Requirement
Sampling Strategy
Tool

NOTE: Please ensure you have an issue that has been acknowledged by a core contributor and routed you to open a pull request against this repository. Otherwise, please open an issue before continuing with this pull request.

Added __repr__ and __str__ to show useful information with masked api_key Closes generative-computing#1317 Signed-off-by: Vishal V <VishalV@ibm.com>

psschwei · 2026-06-30T15:46:42Z

@cptnm3 thanks for the PR! If you don't mind, could you leave a comment on the issue so we can assign it to you?

cptnm3 · 2026-06-30T15:50:04Z

Done @psschwei . Sorry, in the contribution guidelines I did not see that I have to be assigned to an issue before working on it. Will keep in mind next time. Thanks!

psschwei · 2026-06-30T16:19:14Z

no worries!

AngeloDanducci

Thanks for the contribution, @cptnm3! The implementation looks good overall and follows the existing OpenAIBackend pattern nicely.

I've left a few inline comments with some requested changes — mostly moving and mocking some of the tests, and a couple of small test improvements. Once those are addressed this should be good to go.

AngeloDanducci · 2026-06-30T21:00:38Z

+    def __repr__(self) -> str:
+        """Return a useful string representation for debugging."""
+        return (
+            f"{self.__class__.__name__}("
+            f"model_id={self._model_id!r}, "
+            f"base_url={self._base_url!r})"
+        )


Suggested change

def __repr__(self) -> str:

"""Return a useful string representation for debugging."""

return (

f"{self.__class__.__name__}("

f"model_id={self._model_id!r}, "

f"base_url={self._base_url!r})"

)

def __repr__(self) -> str:

"""Return a useful string representation for debugging."""

base_url_repr = self._base_url if self._explicit_base_url else None

return (

f"{self.__class__.__name__}("

f"model_id={self._model_id!r}, "

f"base_url={base_url_repr!r})"

)

self._base_url always holds the fallback value http://localhost:11434/v1 when base_url is not passed by the caller.

Recommend adding this to catch that case.

If you do the above also recommend adding another test case to verify

def test_repr_no_base_url_shows_none(): backend = LiteLLMBackend(model_id="ollama_chat/test-model") r = repr(backend) assert "base_url=None" in r assert "http://localhost:11434" not in r

AngeloDanducci · 2026-06-30T21:05:01Z

    session.reset()


+def test_repr_shows_model_id_and_base_url():


Recommend adding a test_str_* requivalent.

def test_str_shows_model_id_and_base_url(): backend = LiteLLMBackend( model_id="ollama_chat/test-model", base_url="http://localhost:11434/v1" ) assert "ollama_chat/test-model" in str(backend) assert "object at 0x" not in str(backend)

or similar.

AngeloDanducci · 2026-06-30T21:13:38Z

 from mellea.stdlib.context import ChatContext, SimpleContext


+def test_repr_masks_api_key():


Recommend moving test_repr_masks_api_key, test_repr_no_key_shows_none, and test_str_masks_api_key to a separate file (e.g. test/backends/test_watsonx_repr.py) that uses pytest.importorskip only for the import itself at the top but carries no require_api_key or e2e marker — making them plain unit tests that run by default.

Same with the test_litellm_ollama.py addition.

Moved to mellea/backends/test_litellm_repr.py and test/backends/test_watsonx_repr.py

AngeloDanducci · 2026-06-30T21:29:19Z

+    assert "test-project" not in r
+
+
+def test_repr_no_key_shows_none(monkeypatch: pytest.MonkeyPatch):


Inverted assertions here — the test passes api_key=None with WATSONX_API_KEY set in the environment. The constructor resolves None → "env-key" before storing it (watsonx.py lines 132–133), so self._api_key is "env-key" and repr shows '***'. The assertions assert "***" not in r and assert "_api_key=None" in r are both inverted and will always fail.

Corrected this. Please take a look

AngeloDanducci · 2026-06-30T21:37:41Z

+    assert "test-project" not in r
+
+
+def test_str_masks_api_key():


When moving these tests over, recommend using a helper function

def clear_watsonx_env(monkeypatch: pytest.MonkeyPatch): monkeypatch.delenv("WATSONX_API_KEY", raising=False) monkeypatch.delenv("WATSONX_URL", raising=False) monkeypatch.delenv("WATSONX_PROJECT_ID", raising=False)

and then also using monkeypatch for some of the setup

def test_str_masks_api_key(clear_watsonx_env): with patch("mellea.backends.watsonx.APIClient"), patch( "mellea.backends.watsonx.ModelInference" ):

This avoids the issue of the init validating the url with fake credentials before the assertions.

AngeloDanducci · 2026-06-30T21:38:53Z

+    assert "***" in str(backend)
+    assert "test-project" not in str(backend)
+
+


Recommend adding an additional test test_repr_env_key_is_masked or similar.

Watsonx repr tests moved to test_watsonx_repr.py — plain unit tests, no e2e/require_api_key markers. Fix inverted assertions in test_repr_no_key_shows_none; add test_repr_env_key_is_masked for env-resolved key path. Use _make_backend helper + mock Credentials to avoid real SDK init. LiteLLM repr test moved to test_litellm_repr.py. Add test_repr_omits_default_base_url and test_str_shows_model_id_and_base_url. Fix LiteLLMBackend.__repr__ to show base_url=None when caller omitted it. Signed-off-by: Vishal V <VishalV@ibm.com>

cptnm3 · 2026-07-02T20:14:22Z

Hi @AngeloDanducci. thanks for the review. I have made the suggested changes. Please take a look and let me know in case I have missed anything else

feat: add safe repr implementations for watsonx and litellm backends

d7e23c4

Added __repr__ and __str__ to show useful information with masked api_key Closes generative-computing#1317 Signed-off-by: Vishal V <VishalV@ibm.com>

cptnm3 requested a review from a team as a code owner June 30, 2026 15:32

cptnm3 requested review from AngeloDanducci, akihikokuroda and markstur June 30, 2026 15:32

github-actions Bot added the enhancement New feature or request label Jun 30, 2026

AngeloDanducci requested changes Jun 30, 2026

View reviewed changes

cptnm3 requested a review from AngeloDanducci July 3, 2026 06:13

		from mellea.stdlib.context import ChatContext, SimpleContext


		def test_repr_masks_api_key():

		assert "test-project" not in r


		def test_repr_no_key_shows_none(monkeypatch: pytest.MonkeyPatch):

		assert "***" in str(backend)
		assert "test-project" not in str(backend)

Uh oh!

Conversation

cptnm3 commented Jun 30, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Pull Request

Issue

Description

Testing

Attribution

Adding a new component, requirement, sampling strategy, or tool?

Uh oh!

psschwei commented Jun 30, 2026

Uh oh!

cptnm3 commented Jun 30, 2026

Uh oh!

psschwei commented Jun 30, 2026

Uh oh!

AngeloDanducci left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

cptnm3 commented Jul 2, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

cptnm3 commented Jun 30, 2026 •

edited

Loading