Re: [bpfk] extended rafsi and a tosmabru/slinku'i boondoggle

These didn't quite fix it. I think the problem has to do with stress, because I stumbled onto a change that is a half-fix (but I have no idea what else it breaks).

Before any of this thread's changes, we have this (output from souped-up vlatai.py in my camxes-py branch):

esporte'ysaske: valrtosmabru (e=cmavo/A sporte'ysaske=lujvo/sporte-saske)

After incorporating the change to rafsi-string and y-less-rafsi, there is no difference ✗.

However, if I further remove the &unstressed-syllabe lookahead from stressed-brivla-rafsi (so stressed-brivla-rafsi <- brivla-head stressed-syllable h y), then we do get some movement ✓:

esporte'yske: lujvo/esporte-ske
sporte'yske: nalvla
sporte: valslinku'i
esporte: fu'ivla/4
esporte'ysaske: valrtosmabru (e=cmavo/A sporte'ysaske=lujvo/sporte-saske)

So it only fixes some of the broken lujvo ✓✗. And it still accepts both "esporte'y" and "sporte'y" as brivla-rafsi, no matter how many redundant !slinkuhi lookaheads I add, so I don't understand that. But something is going on here WRT stress... unless it's a red herring.

mu'o mi'e selcfipu si la durkavore

On Fri, Nov 7, 2014 at 4:19 PM, Alex Burka <durka42@gmail.com> wrote:

I'm looking at the parse tree for "sporte'y" as a brivla_rafsi and I don't understand it, because it shows "sporte" as a brivla_head. But that should never happen because brivla_head has a !slinkuhi guard. What's going on?

OK, I think I figured it out. The problem is that because "orte'y" is a brivla rafsi, it thinks that "porte'y" is a slinku'i, and so "sporte'y" is fine. An additional problem with the slinku'i change is that it will also not accept "sprageti'y" as a rafsi because "prageti'y" is one, but "sprageti" is not a slinku'i!

So, we must go back to the original slinkuhi rule, and we add instead three final cases to the rafsi-string rule:

slinkuhi <- consonant rafsi-string  

rafsi-string <- y-less-rafsi* (gismu / CVV-final-rafsi / stressed-y-less-rafsi short-final-rafsi / y-rafsi / stressed-y-rafsi / stressed-y-less-rafsi? initial-pair y / long-rafsi vowel h y / CCV-rafsi h y / CVV-rafsi h y)

And I think we need to add the !h in this one as well:

y-less-rafsi <- !y-rafsi (CVC-rafsi !y / CCV-rafsi / CVV-rafsi) !h !any-extended-rafsi

Fingers crossed.

mu'o mi'e xorxes

--
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.