Codehound: Helping Instructors Track Pedagogical Code Dependencies in Course Materials
summary
Codehound uses static analysis over an instructor-provided ordered YAML file of Jupyter notebooks to list which function calls students see first, implemented as a JupyterLab sidebar TypeScript extension that updates live when notebooks are saved. It automatically marks newly introduced methods, links each method to every file that uses it, and supports three workflows: creating content, refactoring (move a function and find all dependent files), and estimating change cost; current limits: notebook-only and simple call-level parsing.
AI-written summary · about
chapters (8)
- 0:00Codehound introduction
- 3:55Pedagogical dependencies
- 6:03Motivation for refactoring
- 7:28Sidebar features and updates
- 12:48Real-world usage scenarios
- 17:21Vision for courseware tools
- 21:31Future directions and limits
- 22:13Q&A
related talks
19mCodeopticon: Real-Time, One-To-Many Human Tutoring for Computer Programming
47mTreat Your Code as a Crime Scene
49mTreat Your Code as a Crime Scene
26mCodeProber: Live Compiler Exploration
38mCode reading
31mMaking sense of frontend code with forensic techniques
01h 01mCode Analysis++
34mCode the Docs: Interactive Document Environments