Skip to content

plan.md

Latest commit

 

History

History
78 lines (52 loc) · 2.06 KB

plan.md

File metadata and controls

78 lines (52 loc) · 2.06 KB

Clippy Refactoring Plan

High-priority fixes and low-hanging fruit based on code review.

Status: ✅ All tasks completed (1092 tests passing)

Critical: Fix Immediately

1. Lock vs RLock Bug ✅

File: src/clippy/models.py:404

Problem: Code uses threading.Lock() but makes reentrant calls.

Solution: Changed to threading.RLock()

High Priority

2. Narrow Exception Handling ✅

Files:

  • src/clippy/agent/core.py:304 - save_conversation() → catches (OSError, IOError, TypeError)
  • src/clippy/agent/core.py:347 - load_conversation() → catches (OSError, IOError, json.JSONDecodeError)
  • src/clippy/mcp/config.py - catches (OSError, IOError, json.JSONDecodeError, ValueError)

3. Consolidate Duplicated _is_reasoner_model() ✅

Solution: Moved to src/clippy/llm/utils.py to avoid circular imports.

  • providers.py imports from llm.utils
  • llm/openai.py imports from .utils

Medium Priority (High Value)

4. Create ToolResult Dataclass ✅

File: src/clippy/tools/result.py

Includes helper methods success_result() and failure_result().

5. Replace If-Elif Chain with Dispatch Table ✅

File: src/clippy/executor.py

Individual handler functions (_handle_read_file, _handle_write_file, etc.) with _TOOL_HANDLERS dispatch table.

6. Create AgentLoopConfig Dataclass ✅

File: src/clippy/agent/loop.py

Updated callers:

  • src/clippy/agent/core.py - _run_agent_loop()
  • src/clippy/agent/subagent.py - _run_agent_loop()

Updated tests:

  • tests/agent/test_loop.py (21 tests)
  • tests/agent/test_multi_tool_sequences.py (8 tests)
  • tests/agent/test_subagent.py
  • tests/integration/test_subagent_workflow.py
  • tests/agent/test_conversation_persistence.py (exception types)

Tracking

Task Status
1. Lock → RLock ✅
2. Narrow exceptions ✅
3. Consolidate _is_reasoner_model ✅
4. ToolResult dataclass ✅
5. Dispatch table ✅
6. AgentLoopConfig dataclass ✅