ogar-class-view: register 0xC6 + typed_field mints; ogar-r2il: execute-never-convert ruling - #286
Conversation
…n the execute-never-convert ruling Two fixes in one arc: 1. ogar-class-view was red on main: typed_field (0x080A) and the four 0xC6 MMIO concepts landed in ogar-vocab's CODEBOOK + class_ids::ALL without registry entries here, tripping both drift gates (known_class_ids_iterates_in_stable_codebook_order and every_codebook_id_appears_in_class_ids_all — the gates did their job). Register all five; 12/12 green. 2. ogar-r2il doc: record the operator ruling that R2IL is EXECUTED in realtime by the r2sleigh interpreter — never pre-converted through ruff_r2il's V4→V3 SPO projection, which is a lossy static shadow of semantics the runtime holds first-class. This crate is addressing glue, not a converter.
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars. ⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Plus Run ID: Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Bugbot couldn't run - usage limit reachedBugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit. A user or team admin can review and increase usage limits in the Cursor dashboard. (requestId: serverGenReqId_27a2bcb4-b8e7-4193-b25b-435517bb7d65) |
… R2IL ops is bit-identical to the native hop
The §7.8 zipper-isomorphism falsifier (lance-graph
r2il-machine-semantic-contract-v1.md): the same 3-level rail path over a
197-row facet slab, answered two ways — one fused u128 prefix compare
(the 'compiled' closure shape) vs an op-at-a-time interpretation of a
straight-line Load/IntEqual/BoolAnd program dispatched through the real
R2ILFn table. Bit-identical masks, both directions non-trivial.
Zero-copy end to end: rows are FacetCascade::ref_from_bytes reinterpret
views into one slab; Load reads the facet register in place;
intermediates live in unique-space scratch; both sides answer with
masks. No serialization anywhere past the fixture's construction.
The fence is executable: Store is refused by the pure-fragment
interpreter ('behavior riding the address'), and a core byte below the
domain floor is refused by the table itself.
Five falsifiers, five disable runs red-then-green (lo-byte-only
IntEqual / AND-as-OR / skipped last level / dispatch without
R2ILFn::ordinal / floor-sized tail word), logged in the module doc.
lance-graph-contract added as a DEV-dependency only — the lib keeps its
ogar-loco-only surface.
Co-Authored-By: Claude <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DCfrD5y19cvFc4AoyydXYv
Bugbot couldn't run - usage limit reachedBugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit. A user or team admin can review and increase usage limits in the Cursor dashboard. (requestId: serverGenReqId_f320db25-6ff5-4055-9714-a9808cff7c26) |
2 participants
Two fixes:
ogar-class-view was red on main (not ogar-r2il: R2IL as an ogar-loco Vocabulary + the masked lane projection #285's diff):
typed_field(0x080A) and the four 0xC6 MMIO concepts landed in ogar-vocab'sCODEBOOK+class_ids::ALLwithout registry entries here, tripping both drift gates (known_class_ids_iterates_in_stable_codebook_order,every_codebook_id_appears_in_class_ids_all) — the gates did exactly their documented job ("guards against a CODEBOOK promotion landing in OGAR without being added here"). Registered all five;ogar-class-view12/12 green, clippy-D warnings+ fmt clean.ogar-r2il doc — operator ruling pinned: R2IL is EXECUTED in realtime by the r2sleigh interpreter, never pre-converted through
ruff_r2il's V4→V3 SPO projection (a lossy static shadow of semantics the runtime holds first-class). This crate is addressing glue, not a converter.🤖 Generated with Claude Code
https://claude.ai/code/session_01DCfrD5y19cvFc4AoyydXYv
Generated by Claude Code