[10:09:50] has anyone done a function with a list of grammatical gender? [10:09:51] like "Katze" + "feminine" = "Die Katze" or "chat" + "masculine" = "le chat" [10:09:53] if so, how do declare the input for gender (a list of strings ? or is/should there object for grammatical gender ?) [10:13:39] also, could an input be optional ? (like in French, to choose the article it depends also if the "h" is apirate or not : "le havre" but "l'heure" ; I could create a boolean input "is the h aspirate or not" but obviously it's useful only for words starting with a "h") [10:38:51] meanwhile, I did my 1000th edit on WF!! : https://tools-static.wmflabs.org/bridgebot/125e30eb/file_57249.jpg [16:00:45] This depends on whether it is Akkusative, Dative or Genetive. (re @Nicolas: has anyone done a function with a list of grammatical gender? [16:00:47] like "Katze" + "feminine" = "Die Katze" or "chat" + "masculine" = ...) [16:01:08] @vrandecic is more positioned to answer. I only have A1 in German. [16:17:39] the word changes on the case, yes [16:17:41] but not the selection of the article ;) (re @Csisc1994: This depends on whether it is Akkusativ, Dativ, or Genitiv.) [16:19:48] at the core, the question remains the same: for the gender/case/whatever, should it be a list of objects? [16:24:12] I'm not sure there's a right answer. there's various ways you could do it and there are probably pros and cons for each one [16:25:11] if we use a list of stings, I could see a lot of "cons" (not multilingual, prone to mispelling, inconsistency across functions, etc.) I guess objects is a bit more complex but better in the end, no ? (re @Nikki: I'm not sure there's a right answer. there's various ways you could do it and there are probably pros and cons for each one) [16:25:30] this seems to be veering into the area of things that will be easier when lexeme access is available [16:26:43] is it? [16:26:44] I mean the functions still need an input (and that part doesn't depends on Lexemes) (re @bgo_eiu: this seems to be veering into the area of things that will be easier when lexeme access is available) [16:30:34] in the example you gave of selecting a form to pair with Katze, if the function took a lexeme as an input, could it not also get the gender from that lexeme? (re @Nicolas: is it? [16:30:35] I mean the functions still need an input (and that part doesn't depends on Lexemes)) [16:30:42] for german, at least, you can't just stick a definite article before something. the noun needs to use the right case too and any adjectives need to change to the form they use with definite articles, so a generic function to just add an article wouldn't make much sense for german to me [16:31:20] yeah this is precisely how I'd do it in Udiron, at least (re @bgo_eiu: in the example you gave of selecting a form to pair with Katze, if the function took a lexeme as an input, could it not also get...) [16:32:33] hmm, I'm confused now... you generate a phrase without inputing at any point the grammatical trait (gender, case, etc.)? [16:34:25] well you can create a big and complex function directly but I prefer the several small functions approach (for instance : one for declension, one for article and one general that aggregate both) (re @Nikki: for german, at least, you can't just stick a definite article before something. the noun needs to use the right case too and any...) [16:35:44] and in any case, I'm guessing that at one point we will need input for gender/case, no? [16:36:28] let's get back to bases: is there any grammatical functions using list in input?