From 0c552082dce50c9dd303a69f7515e9687f3bca40 Mon Sep 17 00:00:00 2001 From: mwiegand Date: Sat, 16 May 2026 13:01:25 +0200 Subject: [PATCH] spec(tz-aware-datetime): correct speculative l4d2host carve-out The previous version implied l4d2host has tz patterns to defer. An inventory grep showed it has no datetime usage at all (no `from datetime` import anywhere in the tree). Replace the bullet with the verified finding. Co-Authored-By: Claude Opus 4.7 (1M context) --- .../superpowers/specs/2026-05-16-tz-aware-datetime-design.md | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/docs/superpowers/specs/2026-05-16-tz-aware-datetime-design.md b/docs/superpowers/specs/2026-05-16-tz-aware-datetime-design.md index da9225b..f337a19 100644 --- a/docs/superpowers/specs/2026-05-16-tz-aware-datetime-design.md +++ b/docs/superpowers/specs/2026-05-16-tz-aware-datetime-design.md @@ -238,8 +238,9 @@ datetimes all along, just losing the tzinfo on the wire. - Postgres portability. `UtcDateTime` would adapt with `impl = DateTime(timezone=True)` if the dialect supports it, but switching dialects is its own project. -- Cleanup of any tz patterns in `l4d2host` (the dedicated-server side of - the repo). This work touches only `l4d2web`. +- `l4d2host` (the dedicated-server CLI side of the repo) has no datetime + usage — no `from datetime` import anywhere in the tree, verified by + grep. The scope boundary is documentary; no follow-up work exists. - Static enforcement (mypy plugin, ruff rule) for "naive datetime crossing into a `UtcDateTime` column." The runtime `TypeError` is the gate; a static check would be a nice-to-have.