Re: [bpfk] Re: Alternatives to {bo} in {i ... bo}?

[Gleki Arxokuna <gleki.is.my.name@gmail.com>]

2015-07-08 10:22 GMT+03:00 guskant <gusni.kantu@gmail.com>:

Le mercredi 8 juillet 2015 06:22:00 UTC, la durka a écrit :

iku'i e'uru'e 

I would rather simply add "jo'au zantufa xi pire", because this parser supports left-association of BO (though {ku} or {cu} become necessary after {la CMEVLA} because it supports also cmevla as tanru unit):

http://guskant.github.io/gerna_cipra/zantufa-0.2.html

([CU {broda VAU}] [i ca bo {CU <brode VAU>}] [i {te zu'e} bo {CU <brodo VAU>}])

 

--

Any reason you think it should change from right- to left-grouping? I mean, we just have to pick one and stick with it. Also, in your parse it looks like there's no grouping at all...

mu'o mi'e durkavore

First, left-grouping is the same as most of other grouping rules of Lojban.

For example, a string of tanru units is read as left-grouping:

((broda brode) brodi)

((broda ja brode) ja brodi)

I rather don't understand why only BO and CO should support right-grouping among many other connectives.

Secondly, just the same problem as la piier raised. When I first wrote a logical proof on "individuals" in Lojban, I should first mention "these BO-s are regarded as left-grouping":

broda

iseni'ibo

brode

iseni'ibo

brodi

....

Right grouping of BO is very annoying for me.

We see no grouping at all on the parsing tree. It's the same thing for JA-connection or simple string of tanru units:

http://mw.lojban.org/extensions/ilmentufa/camxes.html

([broda {ja brode} {ja brodi}] VAU) 

I guess it is because of PEG's characteristic (na'e birti). I heard about left-recursion problem of PEG. PEG first parses a string like

broda brode brodi brodo

 as 

(broda (brode (brodi brodo))).

We should therefore modify the grouping after parsing.

Grouping of JA-connectives is ugly. Not [only] that they group to the right but this

sumti_3 = sumti_4 (joik_ek stag? BO_clause free* sumti_3)?)

is what I completely disagree with. Node must couple either with another node of the same level or with newly created constructs. Thus sumti_4 mustn't be linked with its parent node sumti_3 as its sibling node.

Scoping, precedence and 'embedding clauses into clauses' are all smashed into one type of brackets in PEG which makes the real structure somewhat unclear.

A solution without providing any technical mechanism was proposed here: https://github.com/Ilmen-vodhr/ilmentufa/issues/116

 

--
You received this message because you are subscribed to the Google Groups "BPFK" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bpfk-list+unsubscribe@googlegroups.com.
To post to this group, send email to bpfk-list@googlegroups.com.
Visit this group at http://groups.google.com/group/bpfk-list.
For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups "BPFK" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bpfk-list+unsubscribe@googlegroups.com.
To post to this group, send email to bpfk-list@googlegroups.com.
Visit this group at http://groups.google.com/group/bpfk-list.
For more options, visit https://groups.google.com/d/optout.