Ported for Lean Pool: imports relocated and code adapted to Lean/Mathlib v4.34.0.
Perron-family lower conclusion interfaces (CoreInterfaces) #
Part of the Perron existence development using the canonical Perron-family
upper side and lower-side conclusions from the bump argument. Split from
LowerInterfaces.lean; see the umbrella module docstring.
With the canonical Perron-family upper side, the upper envelope is admissible and therefore equals the pointwise Perron envelope on the PDE domain.
This is the maximality step used in the source proof before the lower bump is
applied directly to W.
Perron's method with the canonical Perron-family upper side and a supplied lower-envelope viscosity supersolution property.
Perron's method with the canonical Perron-family upper side and the lower subjet inequality supplied by the bump argument.
Perron's method with the canonical Perron-family upper side and the formal bump contradiction as the lower-side input.
Perron's method with the canonical Perron-family upper side and the localized improvement form of the lower bump lemma.
This is close to the payload of CIL Lemma 4.4: every failed lower-envelope subjet inequality produces a Dirichlet subsolution above the lower barrier that beats the Perron envelope at points arbitrarily near the failed contact.
Perron's method with the canonical Perron-family upper side and a direct
localized improvement over W^*.
The method layer first proves the source-proof maximality identity
W^* = W on C, then converts the direct improvement into the ordinary
Perron localized-improvement interface.
For the canonical Perron-family upper side, source-style strict patch data
supplies direct localized improvement over W^*.
The method layer first recovers the source maximality identity W^* = W on the
domain, and the bump layer then transfers the strict patch improvement from W
to W^*.
For the canonical Perron-family upper side, source outer-annulus lower-envelope
data supplies direct localized improvement over W^*.
This instantiates the direct bump bridge with old = W^*: upper stability and
comparison make W^* admissible, maximality gives W^* = W on C, and the
source annulus estimate patches over W^* without selecting Perron-family
branches.
Legacy source outer-annulus lower-envelope data still supplies the direct localized improvement after forgetting the branch lower-semicontinuity field.
Perron's method with the canonical Perron-family upper side and the neighborhood-form localized improvement input.
This exposes the source-document formulation directly: every failed contact has an improved Dirichlet subsolution that beats the Perron envelope in every relative neighborhood of that contact.
Perron's method with the canonical Perron-family upper side and the source-shaped strict patch form of the lower bump lemma.
Perron's method with the canonical Perron-family upper side and the max-patch form of the lower bump construction.
Perron's method with the canonical Perron-family upper side and the strict local-patch form of the lower bump construction.
Perron's method with the canonical Perron-family upper side and the source-shaped local quadratic max-patch form of the lower bump construction.
Perron's method with the canonical Perron-family upper side and the local-germ quadratic gluing form of the lower bump construction.
Perron's method with the canonical Perron-family upper side and the branch-local quadratic gluing form of the lower bump construction.
Perron's method with the canonical Perron-family upper side, where operator continuity supplies the lifted quadratic and the remaining input is the explicit patch-selection step.
Perron's method with the canonical Perron-family upper side, where operator continuity supplies the lifted quadratic and the remaining input is compact-inactive local max-patch data.
This endpoint permits the active quadratic branch to be certified only on local domains, which is the natural form produced by a localized bump.
Perron's method with the canonical Perron-family upper side, where operator continuity supplies the lifted quadratic, finite inactive selection supplies the old branch, and comparison with the upper barrier supplies the global upper bound for the patched subsolution.
Perron's method with the canonical Perron-family upper side, where the active compact-inactive patch input only has to certify the lifted quadratic itself as a local subsolution on relative-open domains.
Perron's method with the canonical Perron-family upper side, where the active compact-inactive input supplies only certified local domains for the lifted quadratic. Degenerate ellipticity and Hessian-symmetric invariance certify the quadratic active branch as a viscosity subsolution on those domains.
Perron's method with the canonical Perron-family upper side, using symmetrized certified active-domain compact-inactive data. This removes the need for failed subjet Hessians to be supplied as Hermitian.
Perron's method with the canonical Perron-family upper side, using boundary-safe compact-inactive data. The caller only proves the boundary subsolution inequality for the lifted quadratic; the max patch inherits it from the old Perron branch.
Perron's method with the canonical Perron-family upper side, using boundary-covered compact-inactive data. Covering the boundary by the inactive compact set lets finite Perron selection make the old branch dominate there, so no separate boundary inequality is required for the quadratic branch.
Perron's method with the canonical Perron-family upper side, using exterior
compact-inactive data. The active quadratic neighborhood is kept away from the
Dirichlet boundary, so the inactive compact covers the exterior in
C ∪ boundary and hence supplies the boundary-covered datum.
Perron's method with the canonical Perron-family upper side, using lower-barrier exterior compact-inactive data. On the inactive compact the lifted quadratic is locally below the lower barrier, which is already a Perron-family member.
Perron's method with the canonical Perron-family upper side, using the open-active lower-barrier exterior patch input.
Operator continuity supplies an open active strict-negativity neighborhood;
when C is open, the active local domain is C ∩ V. The remaining exterior
annulus data is stated only for open active neighborhoods.