From b4b7cf51550a0b4a6ab44062caea63d5ccd72c62 Mon Sep 17 00:00:00 2001 From: Dai Ha Date: Sat, 3 Oct 2026 19:59:07 +0200 Subject: [PATCH] fleetd #661: drop the validator counts from validateAll's javadoc FleetConfig declares eight public no-arg void validate* methods, at lines 2671, 2708, 2755, 2790, 2827, 2856, 2896 and 2958. The sweep's javadoc still named a count. The wording is now count-free, so it cannot drift again. --- .../java/dev/ltms/fleet/config/FleetConfig.java | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/fleetd/src/main/java/dev/ltms/fleet/config/FleetConfig.java b/fleetd/src/main/java/dev/ltms/fleet/config/FleetConfig.java index 97a5a96d..4559ded6 100644 --- a/fleetd/src/main/java/dev/ltms/fleet/config/FleetConfig.java +++ b/fleetd/src/main/java/dev/ltms/fleet/config/FleetConfig.java @@ -2977,11 +2977,11 @@ public record FleetConfig( * Runs every validator this class declares — found by reflection, not by name. * *

fleetd ticket "central allow-list of usable models", follow-up: mutation testing found - * that although each of the six validators above was well pinned on its own, nothing proved + * that although each validator above was well pinned on its own, nothing proved * either real caller ({@code Fleetd.main} and {@link ConfigRef#reload()}) still - * invoked it — deleting a call site left the full suite green. The fix is not a seventh test - * per caller; a hand-maintained list of six names here would have the exact same defect its - * own javadoc would warn against: the seventh validator someone adds next month has no reason + * invoked it — deleting a call site left the full suite green. The fix is not one more test + * per caller; a hand-maintained list of names here would have the exact same defect its + * own javadoc would warn against: the next validator someone adds has no reason * to be added to it. So this method does not name any validator. It sweeps {@link * #getClass()}'s own public, no-argument, {@code void} methods whose name starts with {@code * "validate"} (excluding itself) and invokes every one it finds, via {@link @@ -2990,7 +2990,7 @@ public record FleetConfig( * which it silently never runs. * *

{@code Fleetd.main} and {@link ConfigRef#reload()} each call this one method instead of - * the six individually — see the comments at those two call sites for why + * each validator individually — see the comments at those two call sites for why * each must run it. * *

Methods run in a fixed (alphabetical) order, so a config with more than one violation @@ -3007,9 +3007,9 @@ public record FleetConfig( /** * The reflective sweep behind {@link #validateAll()}, kept as its own method — taking any * {@code target}, not just {@code this} — so a test can prove the MECHANISM is generic (it - * would sweep a seventh {@code validateXxx()} method added to any class, not just something - * special-cased to today's six on {@link FleetConfig}) without needing to add a real, unwanted - * seventh validator to this class just to exercise that claim. See {@code + * would sweep any new {@code validateXxx()} method added to any class, not just something + * special-cased to the set {@link FleetConfig} declares today) without needing to add a real, + * unwanted extra validator to this class just to exercise that claim. See {@code * FleetConfigValidateAllTest} for that proof. * * @param target an object whose public, no-argument, {@code void} methods named {@code