From 6deb8e547c75509d63b3f99feac6908455f3b375 Mon Sep 17 00:00:00 2001 From: Paul Trowbridge Date: Fri, 25 Sep 2026 00:16:50 -0400 Subject: [PATCH] Drop point-in-time counts and a temporary table name from CLAUDE.md Co-Authored-By: Claude Opus 5.5 (1M context) --- CLAUDE.md | 11 ++++------- 1 file changed, 4 insertions(+), 7 deletions(-) diff --git a/CLAUDE.md b/CLAUDE.md index 3c60247..8ec6da7 100644 --- a/CLAUDE.md +++ b/CLAUDE.md @@ -188,11 +188,9 @@ constant — read through a join on `pf.version`. That made the pivot the only reader that knew what bucket a row was in. The same CASE was copied into `/agg`, the `/data` cursor and `rlarp.load_osm_pf` in `/opt/forecast`, and any direct query had to copy it again. It also meant editing a version's fallback -name silently renamed every past adjustment. The 2026-09-24 backfill in -`setup_sql/01_schema.sql` stamped the existing entries with that rule; before it -ran, the pre-backfill `label`/`bucket` values were saved to -`pf.bak_log_names_20260924`. `pf.version.unlabeled_load` is read by nothing now; -the column remains. +name silently renamed every past adjustment. The backfill in +`setup_sql/01_schema.sql` stamped the existing entries with that rule. +`pf.version.unlabeled_load` is read by nothing now; the column remains. **What this replaced.** The prefix used to be computed client-side, as Perspective expression columns (`pf_bucket_ord`, `pf_segment_ord`) built from @@ -764,8 +762,7 @@ the negative rows stay where the business was — so a shifted recode is a timin move (pull forward, push out), and with a shift `set` may be empty. The moved rows' period dimensions are re-derived from `pf.dim_period` against the shifted date, as clone's are; a row whose date is null keeps the label it had, since a -calendar lookup on nothing would blank it (v34 has 1,033 baseline rows with no -`sdate_e`). Without a shift nothing is derived and recode copies as it always +calendar lookup on nothing would blank it. Without a shift nothing is derived and recode copies as it always did. Clone still blanks those labels on a null date; it has not been changed. `build_where()` validates every slice key against col_meta (only `role = dimension` allowed). Values are escaped but not parameterized — consistent with existing patterns, debuggable in pg logs.