Rosie Pattern Language: Improving on 50-Year Old Regular Expression Technology
summary
Rosie Pattern Language replaces brittle regex with a PEG-based engine that guarantees linear time matching across large inputs, ships as a C library with Python and Go bindings, and supplies a CLI (rosie grep), REPL, unit tests, and a trace mode for deterministic debugging. It builds named, reusable libraries of patterns so things compose, runs feature extraction in IBM-scale pipelines, and benchmarks show 5× faster than grok on the same tasks while providing named composable patterns that can be maintained and updated.
AI-written summary · about
chapters (23)
- 0:00Introduction
- 1:41DevOps Tools
- 3:34The Problem
- 4:45Regular Expressions
- 7:43Troubleshooting Regular Expressions
- 8:16Improving Rosie
- 9:01Regular Expression Performance
- 11:13Log Stash
- 15:55Named Patterns
- 23:14Parsing Expressions
- 24:25Rosie Library
- 25:05Debugging
- 25:22Debugging Regular Expressions
- 26:49Read
- 28:28executable tests
- 28:50failure details
- 29:13non CLI use cases
- 30:21parsing source code
- 33:14how Rosie is used
- 35:13Rosie performance
- 36:21Conclusion
- 36:37Roadmap
- 37:45Join us
related talks
43mDesigning Cypher - a Graph Query Language
49mThe real power of algorithms on application performance
40mRAG Pipeline Evaluation
24mGraph-Powered Job Matcher: From SQL Tables to Neo4j RAG in Production
43mImplementing Row Pattern Recognition
26mRelational E-matching
17mBring Your Own Data Structures to Datalog
44mBuild RAG from Scratch