Zulip Chat Archive

Stream: Is there code for X?

Topic: Continuity at isolated points


Yongxi Lin (Aaron) (Dec 07 2025 at 15:08):

Do we have a theorem similar to the following one?

import Mathlib

open TopologicalSpace Topology

variable {X Y : Type*} (f : X  Y) [TopologicalSpace X] [TopologicalSpace Y] (x : X) (s : Set X)

lemma Continuous_at_isolated_point (h : 𝓝[s \ { x }] x = ) : ContinuousWithinAt f s x := by
  rw [ continuousWithinAt_diff_self]
  unfold ContinuousWithinAt
  simp [h]

I think this is useful and happy to make a PR.

Aaron Liu (Dec 07 2025 at 15:10):

this should be called something like ContinuousWithinAt.of_isolated

Yongxi Lin (Aaron) (Dec 07 2025 at 15:12):

Sorry for still not being familiar with Mathlib's naming conventions :smiling_face_with_tear:

Yongxi Lin (Aaron) (Dec 07 2025 at 15:14):

Are you suggesting that I should change the name of the theorem I posted (when I make a PR) or there is an existing theorem named ContinuousWithinAt.of_isolated?

Aaron Liu (Dec 07 2025 at 15:24):

I guess both?

Aaron Liu (Dec 07 2025 at 15:24):

if it exists already that would be great too

Etienne Marion (Dec 07 2025 at 15:49):

@loogle nhdsWithin, (⊥ : Filter _), "cont"

loogle (Dec 07 2025 at 15:49):

:shrug: nothing found

Sébastien Gouëzel (Dec 07 2025 at 16:45):

It's maybe better to phrase it in terms of AccPt as in docs#fderivWithin_zero_of_not_accPt

Yongxi Lin (Aaron) (Dec 19 2025 at 00:25):

#33065


Last updated: Dec 20 2025 at 21:32 UTC