Skip to content

Feat/preprocessor benchmarks#44

Merged
evanbrobertson merged 3 commits into
masterfrom
feat/preprocessor-benchmarks
Apr 16, 2026
Merged

Feat/preprocessor benchmarks#44
evanbrobertson merged 3 commits into
masterfrom
feat/preprocessor-benchmarks

Conversation

@evanbrobertson
Copy link
Copy Markdown
Contributor

No description provided.

Brainstorm + plan for the preprocessor benchmark suite shipped in the
follow-up commit, plus an independent brainstorm capturing the v1
semantic-layer scope decisions for later planning.
Establish a CodSpeed baseline for the preprocessor so future tuning
(include cache, hot-path work) has a measurable signal. Three new bench
groups across two crates:

- oxabl_preprocessor/benches/preprocessor_bench.rs: isolation bench plus
  micro-benches for &SCOPED-DEFINE, {&var} substitution, and &IF
  evaluation, all driven by InMemoryFileSystem so the measurement is
  preprocessor work alone.
- oxabl_parser/benches/parser_bench.rs gains pipeline_with_includes:
  tokenize -> preprocess -> parse against an include-heavy fixture with
  real disk resolution via RealFileSystem, mirroring the corpus shape
  that drops throughput from ~1k to ~70 files/sec.

Synthetic .abl + .i fixtures in resources/ keep the suite hermetic and
reproducible. CodSpeed CI auto-discovers the new [[bench]] target; no
workflow changes.
@evanbrobertson evanbrobertson merged commit 5157434 into master Apr 16, 2026
6 checks passed
@evanbrobertson evanbrobertson deleted the feat/preprocessor-benchmarks branch April 16, 2026 20:02
evanbrobertson added a commit that referenced this pull request Apr 17, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants