Skip to content

Liu: state x-subtraction, quartic boundary and root-order lemmas directly (part 2) - #1100

Merged
PerAlexandersson merged 5 commits into
mainfrom
chore/wrappers-liu-2
Oct 5, 2026
Merged

PerAlexandersson merged 5 commits into
mainfrom
chore/wrappers-liu-2

Conversation

@PerAlexandersson

Copy link
Copy Markdown
Owner

Part 2 of the Liu cluster wrapper cleanup. It is stacked on #1096; retarget to main once that merges. After this PR, the only …Statement left in RealRooted/LiuOppositeSigns/** is the refuted published forward direction, kept beside its negation.

Deleted (63 declarations)

  • 20 …Statement defs: xSub{CubicCubic,CubicQuadratic,QuadraticCubic,QuadraticQuadratic,QuarticCubic,LinearQuadratic}Splits, the two discriminant certificates, quarticSubQuadraticSplits, the seven quartic/cubic boundary packages (and the now-empty module XSub/QuarticCubicBoundary/Statements), CompatibleCubic{Linear,Pair}RootOrder, and positiveSplit{Same,Succ}DegreeRootCountAboveNonRoot.
  • 37 conditional lemmas:
    • 30 …_of_monic / …_of_cubicDiscrim / …_of_cubicLinearRootOrder / …_of_cubicPairRootOrder / …_of_packages lemmas. Each body was merged into the existing unconditional lemma of the same name without the suffix, so signatures and call sites are unchanged.
    • The redundant alternative quartic assembly routes.
    • The (3,2)∨(2,3) composition package.
  • Conditional routes to results that are already proved unconditionally. They took the open CommonInterleaver succ-degree statements:
    • compatiblePairHasCommonInterleaver_of_{positiveSplitRootCountAboveNonRoot,sameDegreeAnalytic_and_succCommonLeftInterleaver} and chudnovskySeymour_fourWay_of_…, superseded by chudnovskySeymour_compatiblePairHasCommonInterleaver;
    • RootCountCompatible.of_compatible_of_succDegreeRootCountAboveNonRoot and its helper, superseded by RootCountCompatible.of_compatible;
    • positiveSplitSuccDegreeRootCountAboveNonRoot_of_rootCountAboveNonRoot.

Restated with explicit binders (binder order and implicitness follow the old statement, so applied call sites are unchanged)

  • xSubCubicCubicSplits, xSubCubicQuadraticSplits, xSubQuadraticCubicSplits, xSubQuadraticQuadraticSplits, xSubQuarticCubicSplits, xSubLinearQuadraticSplits (merged with its discriminant reduction), xSubLinearQuadraticDiscrimNonneg, xSubQuadraticLinearCubicDiscrimNonneg, quarticSubQuadraticSplits.
  • xSubQuarticCubic{StrictLeftRepeatedRight,RepeatedLeft,RepeatedRight,LeftOnlyEndpointZero,RightOnlyEndpointZero,EndpointZero,SideBoundary}BoundaryCases.
  • compatibleCubicLinearRootOrder and compatibleCubicPairRootOrder. The (3,2)/(2,3) forward cases move next to the latter as theorem21RootCountBranches_of_compatible_natDegree_{three_two,two_three}.
  • positiveSplitSameDegreeRootCountAboveNonRoot, instantiated with the proved analytic same-degree count.
  • Renamed: splits_X_mul_sub_C_mul_of_positiveSplit_natDegree_two_one_of_cubicDiscrim is now …_two_one. Its hypothesis is proved.

Kept

  • theorem21CompatibleToRootCountBranchesNonconstantStatement, with its checked negation (see PROOF_STATUS.md).
  • No proofs-repo, tactic or catalog names were touched.

🤖 Generated with Claude Code

PerAlexandersson and others added 4 commits October 2, 2026 16:07
Restate the proved Liu results with plain hypotheses and replace the
predicate/relation-parameterized factor-return route by a direct proof:
the three degree cases of a left deletion branch give the translated
right-pencil splitting, which restores compatibility; the right branch
follows by symmetry.  The common-interleaver hypothesis of the old
factor-return statements was never used and is gone.

Delete the theorem-shaped statement defs of Interfaces, Theorem21Assembly,
Corollary22, DeletionBranches, the FactorReturn* layers and the
positive-split x-subtraction leaves, all wrappers taking them as
hypotheses (including every theorem conditional on the refuted
branch-only forward direction), and the superseded low-degree packagings.
Only the refuted published forward statement stays, beside its negation.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…directly

Restate every remaining proved Liu statement with explicit binders:
the normalized x-subtraction splittings and discriminant bounds, the
quartic/cubic boundary cases, the cubic/linear and cubic/quadratic
root-order obstructions, and the same-degree positive-split pair lemma.
Their `_of_monic`/`_of_cubicDiscrim`/`_of_…RootOrder` consumers are
merged into the unconditional versions (signatures unchanged), and the
alternative quartic assembly routes are dropped.

Delete the conditional succ-degree root-count routes, which only
re-derive results already proved unconditionally
(`RootCountCompatible.of_compatible`,
`chudnovskySeymour_compatiblePairHasCommonInterleaver`), and the empty
`XSub.QuarticCubicBoundary.Statements` module.  The only `…Statement`
left in the Liu cluster is the refuted published forward direction,
kept beside its negation.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@PerAlexandersson
PerAlexandersson changed the base branch from chore/wrappers-liu to main October 5, 2026 08:41
@PerAlexandersson
PerAlexandersson merged commit 52895d0 into main Oct 5, 2026
4 checks passed
@PerAlexandersson
PerAlexandersson deleted the chore/wrappers-liu-2 branch October 5, 2026 08:54
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant