[PhaeezAli]

Description

WalEntry.from_dict checked that key and value were present but not that they were strings. A hand-edited or corrupted log line such as {"op": "set", "key": null, "value": "d"} was replayed into the in-memory dict as a non-string key, which the engine never expects.

This raises TypeError for a non-string key or value, so replay() handles the line the same way it already handles other malformed records: it logs a warning and truncates the log to the last good entry.

The duplicate-field case in the issue is standard JSON last-value-wins and is left unchanged.

Closes #322

Type of Change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update

Key Changes

• from_dict raises TypeError when key is not a str, or when value is not a str for set entries.
• Docstring Raises section updated to match.
• Three new cases in test_replay_recovers_from_malformed_entry: null key, integer key, list value.

Validation

• uv run pytest -q → 187 passed
• uv run ruff check . and uv run ruff format --check . → clean
• uv run ty check → clean
• uv run --group docs scripts/docs-reference.py --check → up to date
• Confirmed the three new test cases fail on main without the wal.py change and pass with it.

Checklist

• I reviewed the diff and kept it focused on the described change.
• Python changes pass uv run ruff check . and uv run ruff format --check ., or this PR does not change Python.
• Changed behavior is covered by appropriate tests, including a regression test for bug fixes; any limitations are explained under Validation.
• Validation lists the checks or manual flows I ran and their results.
• Relevant documentation is updated, or no documentation change is needed.

Note

AI tooling has been used for: explaining the existing code path, reviewing my diff, and running the check commands. The fix and tests were written and validated by me.
Assisted by: GPTgo