Recover real packed event descriptor semantics

This commit is contained in:
Jan Petykiewicz 2026-04-15 09:50:58 -07:00
commit eb6c4833af
9 changed files with 719 additions and 120 deletions

View file

@ -77,11 +77,11 @@ The highest-value next passes are now:
avoid shell-first implementation bets
- keep using overlay imports as the context bridge when selectively executable packed rows still
need live company state that save slices do not persist
- treat normalized symbolic company targets as the active packed-event frontier now that
`selected_company`, `human_companies`, and `ai_companies` import and execute through the runtime
service path
- widen real packed-event executable coverage only after the compact-control, symbolic target, and
descriptor frontier is stable, not just after row framing is parsed
- treat broader real grouped-descriptor recovery as the active packed-event frontier now that
descriptor `2` `Company Cash` already parses, summarizes, and executes through the ordinary
runtime path when overlay context resolves its symbolic company scope
- widen real packed-event executable coverage descriptor by descriptor after identity, target mask,
and normalized effect semantics are all grounded, not just after row framing is parsed
- leave condition-relative company scopes explicit and blocked until condition evaluation has
grounded runtime semantics
- keep in mind that the current local `.gms` corpus still exports with no packed event collection,

View file

@ -28,10 +28,12 @@ Implemented today:
company-target model can execute `selected_company`, `human_companies`, and `ai_companies`
symbolic scopes through the ordinary runtime service path while keeping condition-relative
company scopes explicitly blocked
- real `0x4e9a` grouped rows now carry checked-in descriptor metadata, semantic family/preview
summaries, and one recovered executable family: descriptor `2` = `Company Cash`
That means the next implementation work is breadth, not bootstrap. The recommended next slice is
real grouped-descriptor semantic mapping on top of the now-stable compact-control and symbolic
target frontier, not another persistence scaffold pass.
broader real grouped-descriptor coverage beyond `Company Cash`, plus condition-relative execution
for the still-blocked symbolic scopes, not another persistence scaffold pass.
## Why This Boundary
@ -372,45 +374,44 @@ Checked-in fixture families already include:
## Next Slice
The recommended next implementation slice is real `0x4e9a` compact-control decode on top of the
existing real-row structural parse.
The recommended next implementation slice is broader real grouped-descriptor coverage on top of the
now-stable compact-control, symbolic-target, and first recovered real-family path.
Target behavior:
- decode the compact control block that sits above the real standalone-condition and grouped-effect
row families, carrying through raw grounded lanes such as mode byte `0x7ef`, primary selector
`0x7f0`, grouped mode `0x7f4`, one-shot header `0x7f5`, modifier bytes `0x7f9/0x7fa`, grouped
target-scope ordinals `0x7fb`, grouped scope checkboxes `0x7ff`, summary toggle `0x800`, and
grouped territory selectors `0x80f`
- keep real rows parity-only in runtime import, but replace the coarse `blocked_structural_only`
frontier with narrower outcomes such as `blocked_missing_compact_control` and
`blocked_unmapped_real_descriptor`
- keep the existing synthetic harness and overlay-backed executable import path working unchanged
- reserve the first real descriptor-to-effect mapping for a later slice once captured evidence is
tighter
- keep descriptor `2` `Company Cash` as the proof that real grouped rows can cross the whole path:
parse, semantic summary, overlay-backed import, and ordinary trigger execution
- recover more real descriptor identities from the checked-in effect table and expose their target
masks and conservative semantic previews without guessing unsupported behavior
- widen executable real import only when both descriptor identity and runtime effect semantics are
grounded enough to map into the normalized runtime path honestly
- keep condition-relative company scopes explicit until a real condition evaluator exists, instead
of silently degrading or inventing target resolution
Public-model additions for that slice:
Public-model expectations for that slice:
- compact-control summaries on packed-event records in both the save-side and runtime-side models
- runtime summary counts for compact-control-missing and unmapped-real-descriptor blockers
- trigger-kind and one-shot derivation only where the compact-control mapping is already grounded
- additional checked-in grouped-descriptor metadata entries keyed by recovered descriptor id
- more parity summaries with real descriptor labels, target masks, parameter families, and semantic
previews
- more selective real-row `decoded_actions` only where the descriptor-to-runtime mapping is
supported end to end
Fixture work for that slice:
- update the parity-heavy tracked sample so the real row includes compact-control state
- regression fixtures that keep synthetic executable import and overlay-backed company-context
upgrade behavior green
- state-fragment assertions that lock the new compact-control summary and narrower import blockers
- preserve the parity-heavy tracked sample as the condition-relative blocked frontier while it now
carries recovered `Company Cash` semantics
- add overlay-backed captured fixtures whenever a new real descriptor family becomes executable
- keep synthetic harness, save-slice, and overlay paths green as the real descriptor surface widens
Current local constraint:
- the local checked-in and on-disk `.gms` corpus currently exports with
`packed_event_collection_present = false`, so this slice must not depend on a newly captured real
packed-event-bearing save for acceptance
- the local checked-in and on-disk `.gms` corpus still does not provide a richer captured packed
event save set, so descriptor recovery must continue to rely on the grounded static tables and
tracked JSON artifacts until new captures exist
Do not mix this slice with:
- territory-access or selected-profile parity
- placed-structure batch placement parity
- shell queue/modal behavior
- broad speculative translation of real packed RT3 event rows into executable normalized effects
- territory-access or selected-profile parity
- broad condition evaluation without grounded runtime ownership
- speculative executable import for real rows whose descriptor meaning is still weak