On 27/09/2016 11:19, selpahi wrote:
Hmm, under Zantufa, {ge ti gi ta gi tu mi} parses as {ge ti gi ta gi tu GIhI mi VAU}, so there when there's only one term in the fist connectand, it does seem to expect one term in each other connectand. However things like {ge ti gi naku ta gi tu mi} parses as {ge ti gi naku ta gi tu mi GIhI VAU}, and {ge ti gi ke ta tu (ke'e) gi do mi} parsefails. Weird.On 27.09.2016 11:13, Ilmen wrote:This parser doesn't implement n-ary forethough connectives terminated with {gi'i} though, but the Zantufa parser does implement it: • http://guskant.github.io/gerna_cipra/zantufa-1.html (However I seem to understand that Solpahi proposes that each {gi} would introduce only one new term, with the benefit of requiring the connection terminator {gi'i} much less often, whereas in Zantufa, any number of bridi terms can appear after each {gi}, so they behave like termset connectives, and I suppose {gi'i} is required more often there. Please correct me if I misunderstood.){ge mi gi do ti gi ti ta} does not parse in zantufa, but {ge mi ti gi ti ta tu gi ti ta} does. When you start with a gek-termset then each connected unit allows an n-ary termset (of arbitrary length) and then {gi'i} is not elidible if any other terms follow, but if you start with a unary connectand, then each connected unit also must be unary. I think.~~~mi'e la solpa'i
Maybe there's a bug in there. —Ilmen. -- You received this message because you are subscribed to the Google Groups "lojban" group. To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com. To post to this group, send email to lojban@googlegroups.com. Visit this group at https://groups.google.com/group/lojban. For more options, visit https://groups.google.com/d/optout.