Repository navigation
spec(port): port the decoy message payload entropy guard (Refs #8887) - #9834
Merged
1 commit merged intoOct 17, 2026
Merged
1 commit merged into
1 commit merged into
Conversation
…y-S-AI-DNA#8887) The twenty-second CR-CHAT-07 ring target (W2, R-CHAT-10): the decoy-entropy rules - the batch ceiling, then per decoy the length floor (32), the length ceiling (16384), the all-identical probe, the unique-byte floor (the original's rounded-then-truncated double check) and the max-frequency ceiling (a third of the payload) - as the D5 verdict with every payload field, the original's rule order, and all ten DMPE tests transcribed plus helper and rule-order tests. The original's BTreeMap byte-frequency table becomes the fixed 256-slot counter table of the entropy lane: unique_count counts the nonzero slots, max_freq scans them, the fixture tables are built by byte-index loops with the index arithmetic in usize - no narrowing anywhere. Checks (conventions 0-14): t27b test 12 passed 0 failed; t27c test-report 12/12, 0 vacuous; gen-rust output compiles under rustc; coverage Untested 0; ASCII only; seal saved, all hashes MATCH;
Contributor
Author
|
Merge request for the gHashTag merger: steward READY (required all SUCCESS — validate, check-linked-issue, parse-ratchet, t27b-native-linux, t27b-native-macos; the red ones are non-required per Q16 and red on master too). Ledger row included; one-line rebase if the earlier lane PRs land first — I take it on request. |
gHashTag
enabled auto-merge (squash)
October 17, 2026 18:13
This was referenced Oct 17, 2026
gHashTag
disabled auto-merge
October 17, 2026 18:42
12f5af8
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Refs #8887 (CR-CHAT-07 ring tracker). Parent epic: Trinity-S-AI-DNA/trios#1812 (W2, own lane, T-PORT). Twenty-second CR-CHAT-07 ring target delivered: the decoy-entropy rules (R-CHAT-10) — batch ceiling (4096), per-decoy length floor (32) / ceiling (16384), all-identical probe, unique-byte floor (the original's rounded-then-truncated double check, DMPE-01 answers LowUniqueRatio{2, 64}), max-frequency ceiling (a third of the payload; DMPE-02 answers HighFreqRatio{50, 64}) — D5 verdict with every payload field, original rule order, all ten DMPE tests transcribed plus helper and rule-order tests (12 total, 0 vacuous). BTreeMap freq → fixed 256-slot counter table (the entropy-lane form); fixture tables built by byte-index loops, index arithmetic in usize, no narrowing. Honest adaptations: the record's length field takes the name payload_len and the verdict's the name length — a field named len makes the gen-rust backend emit a broken slice length (documented); the original's 4097-record DMPE-06 fixture is 8.5 MB of records (2064 B each), over the runner's stack — the ceiling rule stays in code and the largest buildable batch (1500 records, ≈3.1 MB, the proven pkdu-class size) pins the guard green at scale (documented, the CDUG-06 precedent). t27b 12/12 native; t27c 12/12; gen-rust compiles under rustc (0 errors, 0 &'static str, 0 Vec<const); coverage Untested 0; seal all hashes MATCH; ledger row included (pass, non-pass untouched). Dedup checked. (Branch w carries the amended commit after a local pre-PR amend — the stale v branch is left untouched, no force-push.) Sequencing: one-line ledger rebase if the earlier lane PRs land first — I take it on request.