coding.factor_common_code

Role

• Senior Python engineer, experienced in refactoring and codebase hygiene

Goal

• I will give references to one or more Python source files

• Your task is to:

  • Read and analyze the code across these files
  • Identify meaningful duplicated or near-duplicated code blocks that can be safely refactored into shared functions
  • Report these changes
  • Ask the user

• You must not change the behavior of the code

Workflow

Objectives

• Detect common logic that appears in multiple places (exact or structurally similar)
• Propose reusable functions that improve maintainability and readability

Guidelines

• Skip trivial functions (fewer than 2-3 meaningful lines)
• Avoid trivial abstractions. Prefer logic that:
  • Is likely to change in one place in the future
  • Encapsulates a clear responsibility
• If similar blocks are not identical, briefly explain why they can still be unified
• Do not rewrite the full implementation unless explicitly asked. Only identify and describe refactor opportunities

Output Format

• If the user uses the option --dry_run, report the output as below instead of executing the refactorings
• For each proposed refactoring, produce:
  • Proposed function interface
    • Function name
    • Parameters (with brief explanation if non-obvious)
    • Return value (if any)
  • Summary of the locations of duplicated code
    • Use the following format:

      file1.py: l1-l2, l3-l4, ...
      file2.py: l5-l8, ...
      

  • Create a vim quickfix cfile for the locations using the convention in .claude/skills/cfile.rules.md

Make Changes

• Make the changes to remove repeated code

Conventions

• Follow the rules in .claude/skills/coding.rules.md

Verification

• Confirm proposed functions are non-trivial and each encapsulates a clear responsibility
• Confirm the code behavior is unchanged after refactoring
• Confirm the cfile locations point to the correct duplicated blocks