Machine Powered Refactoring: Leverage AST's to Push Your Legacy Code (& the Web) Forward

on YouTube ↗43m

" summary

Argues that regex-based refactors fail on dynamic JavaScript and that parsing code into an AST gives a predictable, queryable structure for safe, large-scale changes. Demonstrates AST Explorer and using parsers plus jscodeshift / Recast to find nodes, rewrite identifiers into function declarations, and regenerate code; mentions Babel plugins like preval to pre-evaluate at build time. Cites HTTP Archive numbers: average site ~400 KB and ~15s mobile load, and notes only ~25% of teams feel their codebase is healthy.

AI-written summary · about

" chapters (6)

1. 0:00Introduction and speaker background
2. 7:15State of JavaScript and tech debt
3. 15:33Introducing Abstract Syntax Trees
4. 24:03ASTs in tools like Babel
5. 33:10Code modifications with ASTs
6. 42:16Conclusion and resources

" related talks

• 47mComputer, refactor my code!Zbyszek Tenerowicz · GrillJS 2019
• 20mThey see me refactorin'. They hatin'Vit Gottwald · ReactiveConf 2018
• 34mMass Frontend Refactors: Less Stress; More SuccessWaller Goble · Connect.Tech 2021
• 40mHow To (Semi-)Automate JavaScript RefactoringGiles Bowkett · Fluent 2013
• 21mAutomating Front-End RefactoringAhmed El Gabri · AmsterdamJS Conference 2017
• 40mRefactoring large JavaScript codebasesMichele Riva · jsday 2022
• 29mAST: the Secret Weapon to Transform a CodebaseSophia Wang · EmberConf 2020
• 29mCodemods: Refactoring JavaScript using JavaScriptSaimon Sharif · DinosaurJS 2019