[01:31:11] *** Quits: proteusguy (~proteusgu@14.207.161.240) (Ping timeout: 248 seconds) [01:31:45] *** Joins: proteusguy (~proteusgu@14.207.48.222) [03:28:05] *** Quits: proteusguy (~proteusgu@14.207.48.222) (Ping timeout: 240 seconds) [03:30:03] *** Joins: proteusguy (~proteusgu@14.207.48.222) [03:50:05] *** Quits: proteusguy (~proteusgu@14.207.48.222) (Ping timeout: 240 seconds) [03:52:01] *** Joins: proteusguy (~proteusgu@14.207.48.222) [07:22:34] *** Quits: proteus-guy (~proteus-g@2405:9800:bc10:1ca:540d:72f4:515c:8c75) (Read error: Connection reset by peer) [09:21:53] *** Joins: proteus-guy (~proteus-g@180.183.114.68) [10:47:57] *** Quits: proteusguy (~proteusgu@14.207.48.222) (Ping timeout: 240 seconds) [10:59:27] *** Joins: proteusguy (~proteusgu@14.207.48.222) [11:23:21] *** Quits: koom_ (~koo5@79.98.72.195) (Ping timeout: 240 seconds) [11:43:25] *** Quits: drbean (~drbean@210.60.168.195) (*.net *.split) [11:45:34] *** Joins: jstar (~jstar@77.235.50.85) [11:49:04] *** Joins: drbean (~drbean@210.60.168.195) [12:37:34] inariksit: no, and i'm wondering if it is a bug; since generally nouns in french are preceded by some kind of article, but not in this particular situation. [12:45:04] it definitely sounds like a bug; the MassNP instance should create a NP without an article [12:46:43] did you check for other romance languages? [12:47:31] the comment on NounRomance makes it look like it's intended [12:47:31] s = table {Nom => artDef False g n Nom ++ cn.s ! n ; c => partitive g c ++ cn.s ! n} ; -- le vin est bon ; je bois du vin [12:49:19] you can look at CompCN from Romance [12:49:22] yes, that was what i found and started to think that it isn't a bug. [12:49:27] yeah [12:51:02] so MassNP is just a more pragmatic/semantic category and doesn't literally mean "don't put the article" [12:51:17] but CompCN would make a Comp out of a CN, which by definition has no article [12:51:54] OK, i'll try to understand CompCN then! [13:00:23] but even if CompCN does not have an article, UseComp will add one, so i don't see how to get a VP. [13:00:41] oh okay [13:01:44] really? Comp is just Comp = {s : Agr => Str ; cop : CopulaType} ; [13:02:06] it would need to match strings to know which article ... wait nevermind, there's the pre hack [13:04:20] okay maybe you can do some ugly hack like extract a string in nominative from CN, make an Adv of that string and then use mkComp for that adverb http://www.grammaticalframework.org/lib/doc/synopsis.html#Comp [13:05:14] or browse through Extensions or Extend to see if there's something less hacky there [13:06:28] thanks inariksit i'll look around (or go for the ugly hack) [13:09:28] oh yes, i' ve been there and thought that Extra was the way to go since we have [13:09:50] ComplCN : V2 -> CN -> VP ; -- j'ai soif [13:10:09] there [13:10:43] but it didn't work? [13:12:24] well, now i unfortunately don't remember what made me abandon that. it should work, shouldn't it? [13:19:10] probably because it seems unstraightforward to get make the copula into a V2 [13:21:37] aha right [13:21:48] you could maybe find one in DictionaryFre.gf [13:21:55] or an extension VasV2 or something like that [13:22:10] or maybe those coercion functions are only in our super secret development version we're doing with prasanth and aarne [13:22:13] so secret [13:29:49] interestingly (and confusingly) there is an etre in DictFre.gf [13:30:02] lin etre_2_V = mkV "ĂȘtre" ; [13:30:51] which if used together with ComplCN gives 'je ets professeur' [13:32:11] but then i can solve the problem somewhat inelegantly by copying in the construction of all the verb forms for etre [13:32:41] VasV2 is in ExtraMlt! [13:36:50] *** Quits: koom (~sirdancea@109.107.211.244) (Quit: Leaving) [13:37:52] *** Joins: koom (~sirdancea@109.107.211.244) [13:57:46] you have just a copula : Verb in DiffRomance, and in CatRomance, V2 = Verb ** {compl}, so you can totally just use that and make some dummy compl [13:58:35] or rather, Compl : Type = {s : Str ; c : Case ; isDir : Bool} ; insert the actual values [14:42:21] *** Quits: proteus-guy (~proteus-g@180.183.114.68) (Ping timeout: 246 seconds) [15:09:34] inariksit: yes! that works! thanks! [15:25:24] *** Joins: proteus-guy (~proteus-g@2405:9800:bc10:1ca:48ba:f481:5715:a731) [16:00:50] yay! [23:58:54] *** Joins: koom_ (~koo5@79.98.72.195)