From 209d670988c828c23008c6b59eb002a68fa4bf59 Mon Sep 17 00:00:00 2001 From: "russell@unturf.com" Date: Sun, 10 May 2026 09:22:39 -0400 Subject: [PATCH] =?UTF-8?q?Makefile:=20bench-fork-score=20CLI=20invocation?= =?UTF-8?q?=20v8=20=E2=86=92=20substrate?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Caught while sweeping for remaining "arborist v8" CLI invocations after bae5caf renamed the subcommand. Makefile target ``bench-fork-score`` (line 430) still called ``$(ARBORIST) v8 score`` directly — would have failed loudly on next ``make bench-fork-score`` run with an argparse error ("invalid choice: 'v8'"). Same kind of subtle breakage I caught in fork_score.py earlier today (85be5eb): a refactor's text-search misses a callsite that isn't picked up by Python imports. Lesson: after renaming a CLI subcommand, grep ``$(ARBORIST) `` AND ``arborist `` in Makefile, scripts/, and bench/ — argparse errors don't surface until someone runs the target. Hygiene ======= - ``grep -rn "arborist v8\\b" Makefile bench scripts docs`` → only the historical-note comment in cli.py:5131 remains. - ``make help | grep substrate`` → target descriptions accurate. - ``.venv/bin/arborist substrate score --help`` still 0. --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 3888ef3..1234e93 100644 --- a/Makefile +++ b/Makefile @@ -427,7 +427,7 @@ bench-fork-score: bootstrap ## #000012 Phase 1b — score current bench output v @mkdir -p bench/results PYTHONUNBUFFERED=1 $(PY) -m bench.batteries.runner --all \ --out $(FORK_CHILD) - $(ARBORIST) v8 score \ + $(ARBORIST) substrate score \ --parent $(FORK_PARENT) \ --child $(FORK_CHILD) \ --out $(FORK_REPORT)