Agent/agent taskstate 1.1.0#1
Conversation
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 07842ef133
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 "@codex address that feedback".
| "updated_at": connection.updated_at, | ||
| } | ||
| ) |
There was a problem hiding this comment.
Register tracker adapters before CLI fetches
When a user runs tracker fetch after adding a GitHub/Jira connection, this constructs a fresh TrackerBridgeService and immediately calls fetch_issue without registering an adapter from the saved provider/config/secret-env data, so every configured provider follows the No adapter registered path and returns issue: null. The same construction pattern is used by the outbound tracker commands, so the new explicit tracker CLI cannot actually talk to configured trackers.
Useful? React with 👍 / 👎.
| def _columns(conn: sqlite3.Connection, table: str) -> set[str]: | ||
| return {row["name"] for row in conn.execute(f"PRAGMA table_info({table})")} |
There was a problem hiding this comment.
Make migrations work with default sqlite rows
If run_migrations() is used directly with a normal sqlite3.connect(...) connection, PRAGMA table_info returns tuples, not sqlite3.Row objects, so row["name"] raises TypeError before migration 002 runs. CLI init masks this by setting row_factory, but the public migration helper should not fail for standalone callers migrating an existing plural database.
Useful? React with 👍 / 👎.
No description provided.