Zulip Chat Archive
Stream: Is there code for X?
Topic: ContinuousOn Compacts implies Continuous
David Ledvinka (May 23 2025 at 02:07):
I would like something like
import Mathlib.Topology.Defs.Filter
example {X Y : Type*} [TopologicalSpace X] [LocallyCompactSpace X] [TopologicalSpace Y]
{f : X → Y} (h : ∀ K : Set X, IsCompact K → ContinuousOn f K) : Continuous f := by sorry
More generally I see that there is some glueing lemmas in mathlib but as far as I could tell none are statement in a form where the hypothesis is ContinuousOn (for example it suffices to check continuity on a basis of the topology).
Aaron Liu (May 23 2025 at 02:22):
import Mathlib.Topology.ContinuousOn
example {X Y : Type*} [TopologicalSpace X] [WeaklyLocallyCompactSpace X] [TopologicalSpace Y]
{f : X → Y} (h : ∀ K : Set X, IsCompact K → ContinuousOn f K) : Continuous f := by
rw [continuous_iff_continuousAt]
intro x
obtain ⟨K, hK, hx⟩ := exists_compact_mem_nhds x
exact (h K hK).continuousAt hx
David Ledvinka (May 23 2025 at 02:25):
Do you think this should go in mathlib or just be inlined?
Aaron Liu (May 23 2025 at 02:25):
I have no idea
Kevin Buzzard (May 23 2025 at 04:09):
If you have a use for it and it's not already there, I think that's an argument for adding it.
Anatole Dedecker (May 26 2025 at 09:05):
This "exists" as a combination of docs#Topology.IsCoherentWith.continuous_iff and docs#Topology.IsCoherentWith.isCompact_of_weaklyLocallyCompact. But the true lemma will come out of #21237 (which I'm having a look at right now).
Last updated: Dec 20 2025 at 21:32 UTC