Dear Martin,

In List.thy it says

text ‹In the context of multisets, ‹count_list› is equivalent to
  \<^term>‹count ∘ mset› and it it advisable to use the latter.›

Thus I would rather not create a growing count_list library, if it can be avoided. Could you try to use the above hint and go via multisets? (I may add one or two of your lemmas anyway, as a compromise)

Tobias

On 28/02/2022 15:36, Martin Desharnais wrote:
Dear Isabelle developers,

I just needed a few simple lemmas about the count_list function and suggest adding them directly to HOL.List. A quick search on https://search.isabelle.in.tum.de revealed that some of them were duplicated in some AFP entries.



lemma count_list_append: "count_list (xs @ ys) x = count_list xs x + count_list ys x"
   by (induction xs) simp_all

See Groebner_Macaulay.Dube_Prelims.count_list_append, List_Update.TS.count_append, Signature_Groebner.Prelims.count_list_append, and Buildings.Prelim.count_list_append.



lemma count_list_eq_zero_conv: "count_list xs x = 0 ⟷ x ∉ set xs"
   by (induction xs) simp_all

See Groebner_Macaulay.Dube_Prelims.count_list_eq_0_iff, HOL.list.count_notin, and List_Update.TS.count_notin2.



lemma distinct_iff_count_list: "distinct xs ⟷ (∀x. count_list xs x = 0 ∨ count_list xs x = 1)"
   by (induction xs) (auto simp add: count_list_eq_zero_conv)

See Buildings.Prelim.distinct_count_list.



lemma count_list_filter:
   "P x ⟹ count_list (filter P xs) x = count_list xs x"
   "¬ P x ⟹ count_list (filter P xs) x = 0"
   by (induction xs) simp_all



Would it make sense to include them in the distribution?

Regards,
Martin

_______________________________________________
isabelle-dev mailing list
[email protected]
https://mailman46.in.tum.de/mailman/listinfo/isabelle-dev

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

_______________________________________________
isabelle-dev mailing list
[email protected]
https://mailman46.in.tum.de/mailman/listinfo/isabelle-dev

Reply via email to