Zulip Chat Archive

Stream: new members

Topic: Closed subsets of closed subspaces


Joseph Cooper (Jul 22 2024 at 12:30):

import Mathlib

example: @IsClosed ((Metric.closedBall (0:) 2)) instTopologicalSpaceSubtype ({z| z  Metric.closedBall (0:) (1)}):= by
  sorry

In order to prove this statement, is there a lemma in Mathlib similar to this?
Or is there an alternative approach to this problem, that may be easier, given lemmas/theorems in Mathlib?

Yaël Dillies (Jul 22 2024 at 12:33):

docs#isClosed_ball ?

Joseph Cooper (Jul 22 2024 at 13:30):

Yaël Dillies said:

docs#isClosed_ball ?

Applying Metric.isClosed_ball results in "busily processing" for a long time. Do results like this figure out the context of the subspace topology?

Jireh Loreaux (Jul 22 2024 at 13:51):

Did you get the Mathlib cache?

Jireh Loreaux (Jul 22 2024 at 13:52):

oh, I see, you're working with the subtype.

Luigi Massacci (Jul 22 2024 at 14:00):

Can I ask what you want to do? (the statement is not very natural in Lean) This seems like an #xy problem kind of situation

Luigi Massacci (Jul 22 2024 at 14:22):

Though here is a bad proof if you care for it:

example: @IsClosed (Metric.closedBall (0:) 2) instTopologicalSpaceSubtype ({z| z  Metric.closedBall (0:) (1)}):= by
  simp only [Metric.mem_closedBall, dist_zero_right, Complex.norm_eq_abs]
  exact isClosed_le (Continuous.comp Complex.continuous_abs continuous_subtype_val) continuous_const

Last updated: May 02 2025 at 03:31 UTC