Zulip Chat Archive

Stream: Infinity-Cosmos

Topic: preserves limits


Jon Eugster (Mar 14 2025 at 08:55):

Is this, or some modified similar version of this true?

import Mathlib

open CategoryTheory Limits

-- TODO: hyps are too strong and just copy-pasted over, some can be dropped, e.g. `V` & `EnrichedOrdinaryCategory`
theorem helper {J : Type u₁} [Category.{v₁, u₁} J]
    {K : Type u₂} [Category.{v₂, u₂} K]
    (V : Type u') [Category.{v', u'} V] [MonoidalCategory V]
    {C : Type u} [Category.{v, u} C] [EnrichedOrdinaryCategory V C]
    (F : J  C) (e : K  J) (G : C  V)
    (h : PreservesLimit F G) : PreservesLimit (e.functor  F) G := by
  constructor
  intro c hc
  apply Nonempty.intro
  sorry

Markus Himmel (Mar 14 2025 at 09:27):

Here is a quick-and-dirty proof:

import Mathlib

open CategoryTheory Limits

-- TODO: hyps are too strong and just copy-pasted over, some can be dropped, e.g. `V` & `EnrichedOrdinaryCategory`
theorem helper {J : Type u₁} [Category.{v₁, u₁} J]
    {K : Type u₂} [Category.{v₂, u₂} K]
    (V : Type u') [Category.{v', u'} V] [MonoidalCategory V]
    {C : Type u} [Category.{v, u} C] [EnrichedOrdinaryCategory V C]
    (F : J  C) (e : K  J) (G : C  V)
    (h : PreservesLimit F G) : PreservesLimit (e.functor  F) G := by
  constructor
  intro c hc
  apply Nonempty.intro
  have := hc.whiskerEquivalence e.symm
  let equ := e.invFunIdAssoc F
  have := (IsLimit.postcomposeHomEquiv equ _).symm this
  have := (isLimitOfPreserves G this)
  refine IsLimit.ofIsoLimit (this.whiskerEquivalence e) ?_
  refine Cones.ext (Iso.refl _) fun j => ?_
  simp [equ,  Functor.map_comp]
  have := c.w (e.unit.app j)
  simp at this
  simp [ this,  Functor.map_comp]

Jon Eugster (Mar 14 2025 at 09:30):

thank you :)


Last updated: May 02 2025 at 03:31 UTC