Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

The Buchstab integral over an ordered box

Definition
LL_Decomp587h_buchstabIntegral

by sthubbar · Nov 18, 2026 · Mathlib 0df444a (Lean v4.33.1)

For ω:R→R\omega:\mathbb R\to\mathbb Rω:R→R, n∈Nn\in\mathbb Nn∈N, vectors ℓ,h∈Rn+2\ell,h\in\mathbb R^{n+2}ℓ,h∈Rn+2 with coordinates indexed by 0,…,n+10,\dots,n+10,…,n+1, and φ∈R\varphi\in\mathbb Rφ∈R, the real number ∫Dω((φ−∑jxj)/xn) ((∏jxj) xn)−1 dx\int_D \omega\bigl((\varphi-\sum_j x_j)/x_n\bigr)\,\bigl((\prod_j x_j)\,x_n\bigr)^{-1}\,dx∫D​ω((φ−∑j​xj​)/xn​)((∏j​xj​)xn​)−1dx, the integral with respect to Lebesgue measure on Rn+2\mathbb R^{n+2}Rn+2 over the set DDD of monotone xxx with ℓj≤xj≤hj\ell_j\le x_j\le h_jℓj​≤xj​≤hj​ for all jjj. Here xnx_nxn​ is the penultimate coordinate, so it appears squared in the denominator. As for every Bochner integral, the value is 000 if the integrand is not integrable on DDD.

Definition code
import Mathlib
import Definitions.Def_LL_Decomp587h_domain

open Real Set MeasureTheory
open scoped Classical

namespace LL_Decomp587h

noncomputable section

def buchstabIntegral (ω : ℝ → ℝ) (n : ℕ) (lo hi : Fin (n + 2) → ℝ) (φ : ℝ) : ℝ :=
  ∫ x in domain lo hi,
    ω ((φ - ∑ j, x j) / x ⟨n, by omega⟩) / ((∏ j, x j) * x ⟨n, by omega⟩)

end

end LL_Decomp587h
Read-back

What the Lean code literally says, in plain math · claude-haiku

Read-back of LL_Decomp587h.buchstabIntegral (namespace LL_Decomp587h, noncomputable definition).

Inputs: a function ω:R→R\omega:\mathbb{R}\to\mathbb{R}ω:R→R; a natural number nnn; two vectors lo,hi∈R n+2\mathrm{lo},\mathrm{hi}\in\mathbb{R}^{\,n+2}lo,hi∈Rn+2 indexed by {0,1,…,n+1}\{0,1,\dots,n+1\}{0,1,…,n+1}; and a real number φ\varphiφ. Output: a real number, defined as the Lebesgue integral (standard volume on Rn+2\mathbb{R}^{n+2}Rn+2)

∫D(lo,hi)ω ⁣(φ−∑j=0n+1xjxn)(∏j=0n+1xj)⋅xn  dx,\int_{D(\mathrm{lo},\mathrm{hi})} \frac{\omega\!\left(\dfrac{\varphi-\sum_{j=0}^{n+1} x_j}{x_n}\right)}{\Big(\prod_{j=0}^{n+1} x_j\Big)\cdot x_n}\;dx,∫D(lo,hi)​(∏j=0n+1​xj​)⋅xn​ω(xn​φ−∑j=0n+1​xj​​)​dx,

where x=(x0,…,xn+1)x=(x_0,\dots,x_{n+1})x=(x0​,…,xn+1​) ranges over the region D(lo,hi)D(\mathrm{lo},\mathrm{hi})D(lo,hi). The region DDD is the imported set domain lo hi from Definitions.Def_LL_Decomp587h_domain (not shown in this file), so its shape is not established here. The index nnn in the formula is the coordinate xnx_nxn​, the second-to-last of the n+2n+2n+2 coordinates (it is the Lean index ⟨n, _⟩, where the proof term shows n<n+2n<n+2n<n+2).

Concerns: (1) Lean convention for division by zero: a/0=0a/0 = 0a/0=0, so the integrand is set to 000 on the hyperplanes xn=0x_n=0xn​=0 and ∏jxj=0\prod_j x_j=0∏j​xj​=0; these are measure-zero, but they matter if the domain touches them. (2) Lean's Bochner integral returns 000 when the integrand is not integrable on DDD (or not measurable). So any bound of the form "buchstabIntegral ≤c\le c≤c" is trivially true whenever the integrand fails to be integrable, which would make the bounds in the r2j*_short theorems vacuous in those cases. Whether the integrand is integrable for the r2j* data is not decided by this file. (3) The denominator equals xn2∏j≠nxjx_n^2\prod_{j\ne n}x_jxn2​∏j=n​xj​, so the integrand carries a factor 1/(xn2∏j≠nxj)1/(x_n^2\prod_{j\ne n}x_j)1/(xn2​∏j=n​xj​) with 1/xn1/x_n1/xn​ also inside ω\omegaω's argument; the integrability question in (2) is therefore a real concern near xn=0x_n=0xn​=0 and near the other coordinate zeros. (4) The definition is a def, with no theorem attached.

Human review
  • Endorsed by Shuze Chen · Nov 18, 2026

    Confirmed by the moderator at approval.

  • Endorsed by sthubbar · Nov 18, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me