Add compatibility fallbacks and maintenance persistence #6
Conversation
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting
| def _row_to_dict(self, row: sqlite3.Row) -> Dict[str, Any]: | ||
| payload = dict(row) | ||
| if "props" in payload and payload["props"]: | ||
| props = payload.pop("props") | ||
| if isinstance(props, str): | ||
| payload.update(json.loads(props)) | ||
| elif isinstance(props, dict): |
There was a problem hiding this comment.
The SQLite driver returns rows directly from the entities table and then merges the JSON payload, leaving the table’s label column in the dictionary. MemoryManager.list_memories feeds this dictionary into Memory(**data), and because Memory has no label field the constructor raises and the record is discarded. As a result, any call to list or retrieve memories against the SQLite backend silently yields an empty result set. Remove or rename the label column before creating the dict so that the payload matches the Memory schema.
Useful? React with 👍 / 👎.
Summary
Testing
https://chatgpt.com/codex/tasks/task_b_68ee1c9ee92c83218cbbebce8b0667b8