Zulip Chat Archive

Stream: general

Topic: proving two zpowers are equal


Frederick Pu (Nov 13 2025 at 22:21):

Is there anything in mathlib proving the following statement about cyclic subgroups?

hy'x : y  Subgroup.zpowers x
this : orderOf x = orderOf y
 Subgroup.zpowers y = Subgroup.zpowers x

Kenny Lau (Nov 13 2025 at 22:25):

(x,yQ×)=(2,4)(x,y \in \Bbb Q^\times) = (2,4)

Frederick Pu (Nov 13 2025 at 22:26):

for finite groups i mean

Frederick Pu (Nov 13 2025 at 22:26):

so Subgroup.zpowers x and Subgroup.zpowers y are isomorphic

Frederick Pu (Nov 13 2025 at 22:26):

and you know that y must go to x or x^-1

Kenny Lau (Nov 13 2025 at 22:36):

Frederick Pu said:

and you know that y must go to x or x^-1

(x,y(Z/11)×)=(2,4)(x, y \in (\Bbb Z/11)^\times) = (2, 4)

Aaron Liu (Nov 13 2025 at 22:37):

I think the statement is still true (for finite groups) but it's definitely not that simple

Frederick Pu (Nov 13 2025 at 22:38):

i think it's just double inclusion

Aaron Liu (Nov 13 2025 at 22:38):

x could be any n-th power of y for n coprime to the order of x and y

Frederick Pu (Nov 13 2025 at 22:38):

since y \in zpowers x, zpowers y \subseteq zpowers x

Frederick Pu (Nov 13 2025 at 22:39):

so use the lemma that if H \subseteq K and |H| = |K| then H = K for finite groups

Kenny Lau (Nov 13 2025 at 22:40):

great, now you have a proof

Frederick Pu (Nov 13 2025 at 22:41):

but is it in mathlib tho

Frederick Pu (Nov 13 2025 at 22:41):

i feel like it would be convenient

Aaron Liu (Nov 13 2025 at 22:43):

I don't see it in my searches but you can use the proof you have and prove it yourself

Frederick Pu (Nov 13 2025 at 22:52):

:(

Frederick Pu (Nov 13 2025 at 22:52):

my assignments due in like 5 hours i think ill just accept that it's true lol

Aaron Liu (Nov 13 2025 at 23:06):

Surely it won't take five whole hours for such a simple proof?

Frederick Pu (Nov 13 2025 at 23:08):

i mean i have other parts of my main proof i havent finished yet

Aaron Liu (Nov 13 2025 at 23:13):

here you can have this

import Mathlib

example {G : Type*} [Group G] [Finite G] {x y : G}
    (hy'x : y  Subgroup.zpowers x)
    (this : orderOf x = orderOf y) :
    Subgroup.zpowers y = Subgroup.zpowers x := by
  refine SetLike.coe_injective (Set.eq_of_subset_of_ncard_le ?_ (le_of_eq ?_))
  · simpa using hy'x
  · rw [Set.ncard_eq_toFinset_card, Set.ncard_eq_toFinset_card]
    apply Finset.card_eq_of_equiv
    simpa using zpowersEquivZPowers this

Frederick Pu (Nov 13 2025 at 23:57):

also what about smth like this?

this : Subgroup.zpowers x < 
this : Subgroup.zpowers x  
  y, y  Subgroup.zpowers x

Ruben Van de Velde (Nov 14 2025 at 00:05):

docs#SetLike.exists_not_mem_of_ne_top ?

Frederick Pu (Nov 14 2025 at 00:10):

hmm im not 4.25 but i cant find that one

Ruben Van de Velde (Nov 14 2025 at 00:15):

Yeah, I believe it's pretty new. You can copy it, though

Ruben Van de Velde (Nov 14 2025 at 00:15):

docs#Set.ne_univ_iff_exists_notMem is the essential element

Aaron Liu (Nov 14 2025 at 00:25):

Ruben Van de Velde said:

docs#SetLike.exists_not_mem_of_ne_top ?

Didn't we switch over to calling it notMem a while ago


Last updated: Dec 20 2025 at 21:32 UTC