Talk:Z10423

Original discussion

This discussion was moved from Suggest a Function --99of9 (talk) 01:47, 29 September 2023 (UTC)

• Hi. I don’t know where we are supposed to discuss function propositions. I just want to note that the notion of anagram is language dependent. For example, in French, the diacritics doesn’t count (“niché” is an anagram of “chien”). But in Esperanto, the letters with diacritics are considered different letters. So, “ico” is not an anagram of “ĉio”. Lepticed7 (talk) 10:40, 5 August 2023 (UTC)

  @Lepticed7: assuming that in the future a "remove diacritics" function would exist, anagram in this context would be strictly defined as string having exactly the same set of characters. 0xDeadbeef (talk) 16:17, 5 August 2023 (UTC)

  @0xDeadbeef Thus giving wrong results for Esperanto. Or at least results that do not match the common understanding of "anagrams" in Esperanto. Lepticed7 (talk) 16:34, 5 August 2023 (UTC)

  @Lepticed7: therefore you can't call it an anagram in Esperanto then. You have to translate what the function does. 0xDeadbeef (talk) 16:36, 5 August 2023 (UTC)

  @0xDeadbeef Well, what’s your definition of "anagram"? Lepticed7 (talk) 16:43, 5 August 2023 (UTC)

  @Lepticed7: For this function I am working on: are strings anagrams (Z10423) It is defined as, whether one string with its codepoints rearranged is equivalent to the other string. Actually I misread the discussion above I think. I'm just saying anyone can just create another function for anagrams in French that ignores diacritics, while anagrams can remain having a strict definition (sensitive to diacritics). 0xDeadbeef (talk) 16:47, 5 August 2023 (UTC)

  Ok. So maybe should we indicate in the function description that are strings anagrams (Z10423) is diacritics-sensitive? Lepticed7 (talk) 16:50, 5 August 2023 (UTC)

  Sure, but I hit the character limit for the short description, so maybe we could do that when long descriptions become a thing. 0xDeadbeef (talk) 16:51, 5 August 2023 (UTC)

  Well, if you start talking about codepoints, you have to be even more careful, as Unicode is tricky. Because now, your function says “QR̀” and “RQ̀“ are anagrams (while “AÈ” and “EÀ” are not)… [Note that I am not sure if some specification/design document of Wikifunctions even specifies that all strings are NFC?] --Mormegil (talk) 18:17, 19 August 2023 (UTC)

  Sorry, are you talking about unicode graphemes? I don't think codepoints are as tricky as graphemes. (since graphemes can be ligated and etc.) 0xDeadbeef (talk) 04:40, 20 August 2023 (UTC)

  I have done this and named it "is anagram (simple)" to avoid confusing it with a more linguistically aware version. Mtanti (talk) 12:56, 31 August 2023 (UTC)