Automated extraction of grammar optimization rule configurations in a metamodel-gr...
summary
Automated extraction produces configuration files that let a grammar optimizer replay manual syntax fixes after a metamodel change, implemented as a grammar Optimizer plus a configure generator that diffing-generated and target grammars. Evaluated on 6 real cases: full automation succeeds for 3, the other 3 reach agreement rates between 88% and 67%. The tool reduces manual effort but misses changes when line order or attribute grouping matters and when optimizer capabilities lack coverage.
AI-written summary · about
chapters (13)
- 0:00Welcome
- 1:08Introducing paper and team
- 3:23Motivation for grammar generation
- 5:28Manual optimization challenges
- 6:51Grammar Optimizer tool
- 8:18Co-evolution overview
- 10:28Configuration generator development
- 12:15Configuration extraction
- 14:03Example extraction process
- 15:34Evaluation results
- 17:07Limitations and missing
- 18:56Conclusion and future work
- 20:20Q&A
related talks
15mAn Interactive Feedback System for Grammar Development (Tool Paper)
16mAutomatically Deriving Control-Flow Graph Generators from Operational Semantics
12mAutomatic Grammar Repair
15mDetecting and explaining (in-)equivalence of context-free grammars
21mVerifying term graph optimizations using Isabelle/HOL
21mSynthesizing Formal Semantics from Executable Interpreters
47mHow an Optimizing Compiler Works
20mAutomating Pruning in Top-Down Enumeration for Program Synthesis Problems with(…)