Parser Frontend Performance & API
6 sections
Overview
Maximize Ori's parser frontend throughput and API quality across the full pipeline — lexer, parser, and Salsa integration — by amplifying existing architectural strengths (parallel tag arrays, static binding power tables, Elm-style progress tracking) and closing identified gaps (file hygiene blockers, missing #[inline] annotations, inactive incremental parsing, Salsa query granularity). Target: measurable throughput improvement on existing benchmarks with zero regressions.
Planned
6 sections
Performance Baselines
Establish reproducible performance baselines for lexer, parser, and Salsa query throughput using Criterion benchmarks.
File Hygiene
Split all ori_parse files exceeding the 500-line limit into focused submodules with zero behavioral changes.
Lexer Optimizations
Measurable throughput improvement in ori_lexer benchmarks via targeted inline annotations, arena pre-allocation, and cooker fast-path expansion.
Parser Optimizations
Measurable throughput improvement in ori_parse benchmarks via inline audit, arena pre-allocation, snapshot enhancement, and expression parsing tuning.
Salsa Integration
Reduce Salsa query overhead and activate incremental parsing infrastructure for sub-linear re-parse on small edits.
Verification
Prove measurable improvement across all parser frontend benchmarks with zero regressions.