Slashing by an upper-triangular representative preserves behaviour at i∞ #
Mathlib's UpperHalfPlane.IsBoundedAtImInfty.slash and IsZeroAtImInfty.slash carry the
hypothesis g 1 0 = 0, so they apply when g is upper triangular. (The hypothesis is sufficient,
not necessary — the zero function stays bounded and vanishing after slashing by any matrix.)
SlashActionRat.lean states their rational forms; this file discharges the hypothesis for this
repository's coset representatives upperTriRep and assembles the operator-level statement for
heckeSlashUpperTri.
That operator is the merged ∑ b < p, f ∣[k] !![1, b; 0, p] of UpperTri/Sum.lean, which had no
statement about its behaviour at i∞. The cusp-level results for the general double-coset sum
live in HeckeSlash/Cusps.lean and are about a different operator; neither implies the other.
Main results #
HeckeRing.GL2.isBoundedAtImInfty_slash_upperTriRep,HeckeRing.GL2.isZeroAtImInfty_slash_upperTriRep: slashing by a representative preserves boundedness, resp. vanishing, ati∞.HeckeRing.GL2.isBoundedAtImInfty_heckeSlashUpperTri,HeckeRing.GL2.isZeroAtImInfty_heckeSlashUpperTri: the same for the whole sumheckeSlashUpperTri.
Provenance #
No code is transcribed. The step corresponds to the analytic core of the AINTLIB
LeanModularForms project (Chris Birkbeck, Apache-2.0),
LeanModularForms/HeckeRIngs/GL2/AdjointTheory.lean at commit
2baa76f742bdb4fb8ee323fabba41203bd390e08, whose heckeT_p_ut_zero_at_cusps (lines 62-70) needs
exactly this fact about each summand. Here it is stated for a single representative and for
upperTriRep, this repository's general-n family at n = 2, rather than for a transcribed
matrix.
References #
Slashing by a representative preserves boundedness at i∞.
Slashing by a representative preserves vanishing at i∞.
The upper-triangular sum is bounded at i∞ when the function is. Each summand is
handled by isBoundedAtImInfty_slash_upperTriRep, and the sum by Filter.BoundedAtFilter.sum.
The upper-triangular sum vanishes at i∞ when the function does.