docs(historian-gateway): note FU-3 alias handling (review fix) in follow-up plan

Claude-Session: https://claude.ai/code/session_012SDSQ3AcaXqPcBtDESBRii
2026-06-27 00:57:14 -04:00
parent 60695179ee
commit 10a6ac6f3e
1 changed files with 7 additions and 3 deletions
@@ -139,9 +139,13 @@ window anchored on the write time.
 ### FU-3 — Continuous-historization `HistorianTagname` override edge case — ✅ DONE (2026-06-27, commit `111adc92`)
 > **Outcome:** implemented the "carry both identifiers" fix below. A new `HistorizedTagRef(MuxRef,
 > HistorianName)` record threads through `IHistorizedTagSubscriptionSink` → the recorder; the recorder keeps
-> a muxRef→historianName map, registers/filters mux interest by `MuxRef` (= driver `FullName`) but writes
-> under `HistorianName` (override-or-FullName). The applier resolves both. Divergent-override + override-
-> rename-no-churn recorder tests added; applier feed tests assert the full pairs.
+> a **muxRef → SET-of-historian-names** map, registers/filters mux interest by `MuxRef` (= driver `FullName`)
+> but writes under every `HistorianName` (override-or-FullName) sharing that ref. The applier resolves both.
+> The set (not a single name) closes a code-review **Critical**: one driver ref can back several historized
+> equipment tags via aliasing (identical machines sharing a register), each with its own override — a single
+> fan must write ALL of them, not silently drop all but one. Tests: divergent-override, aliased-refs-each-
+> get-the-value, remove-one-alias-keeps-the-ref, override-rename updates the write target without mux churn;
+> applier feed tests assert the full pairs. Commits `111adc92` + `60695179` (review fix).

 The `ContinuousHistorizationRecorder` registers `DependencyMuxActor` interest **by the resolved
 historian name** (`HistorianTagname` override else `FullName`) — the same key the EnsureTags hook and