Repository navigation
Construction: reversible integer-lifting cleanup for the P104 collector - #3025
Merged
Mister-Meeseeks merged 1 commit intoOct 16, 2026
Merged
Conversation
This was referenced Oct 17, 2026
This was referenced Oct 18, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
PR2999 shows that independently nearest-rounded native calls can lose input information and leave a nonzero private residue in the first P104 collector. This construction changes the numerical map: integer-lifting butterflies use a literal paired inverse, restoring all five dirty sources and the private field exactly without retaining rounding residues or per-coordinate history.
Both completed calls meet relative Hilbert error at most
2^-bon the paired execution. The query remains approximate on its retained working grid; the absolute outbound error is at most112*2^-F. The inverse schedule and grid must be preserved. Standard reversible lifting is credited; the contribution is its binding to PR2586 events 2025–2039, the paired accuracy/cleanup proof, and a fully charged scalar-reference recurrence.Evidence:
Exact base:
d8e0f83f00191df62174b641e9fff055c357f641, priority revision 6. Native source: ywz65/PR25867953c9a8f788a8de5cf2ced63d9a61b67c22c6f1; collision predecessor: Arham Tahir/PR2999ac6025235b6f1233291000d35a067198628569e9. All consumed bytes are pinned in SOURCES.json. Package proof, result and remaining questions. Canonical claim.Scope: a local numerical construction. It is nonlinear and evaluates every butterfly; it does not inherit a cheap recursive child tariff. A complete numerical consumer, physical hosting/routing, all-size admission and a favorable supplier moment remain unresolved. No new kappa, selected-frontier change, full selected replay or formal/human review is claimed. Latest PR3019's relaxed matching support and PR3024's packing cap address distinct obligations. Fresh complete overlap pagination failed; the retained historical corpus, recent PRs, canonical thread and relevant bodies were checked, with limits recorded in TASK.json.
Credit Arham Tahir/PR2999/2891/2665, ywz65/PR2586/2730/2555, Nick Petalas/PR2754/2771, Chafik Boukhalfa/PR628 and the retained native/geometry/compiler lineage. Prepared by Tekalign with substantial OpenAI Codex assistance. Apache-2.0.
The change is eight additive research-package files. Frozen sources, shared checkers, upstream/ and the selected frontier are unchanged.