[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lojban] le vlatai be lo cmene



On Thu, Dec 06, 2001 at 05:25:43PM -0500, John Cowan wrote:
> Pierre Abbat wrote:
> 
> > I'm reading the Lessons and I saw a name {antonis.routc.}. This looks odd to 
> > me, as {ou} is not a valid diphthong. vlatai calls {routc} valid, but since 
> > {ou} is not a Lojban diphthong, it must be implicitly inserting a comma.
> 
> 
> No two words differ solely because of the presence or absence of commas,
> so "routc" and "ro,utc" are the same word.

This 'ruling' was incorporated into the lexing engine of vlatai and
jbofihe at the 0.38 release.

> 
> 
> > 
> > I then tried {ryytc}, {raatc}, {ryitc}, and {riytc}. {ryytc} is rejected as 
> > "Bad cmene - doesn't break up", but {ry,ytc} is accepted. {ryitc} is rejected 
> > with or without the comma. What is going on?
> 

I'm afraid I can't reproduce this with version 0.38.  The message "bad cmene
doesn't break up" is generated by words like {bylaplas}, which have the
illegal 'la' inside, but couldn't break up in the normal way because the
preceding {by} would then have to have a pause after it.  The word(s)
being tested do not have this form.

> 
> Those look like bugs.
> 

I've got vlatai to 'advise' on all the forms like this that I could
think of.  Here are its findings:

raatc                     : cmene :  raatc
raetc                     : cmene :  raetc
raitc                     : cmene :  raitc
raotc                     : cmene :  raotc
rautc                     : cmene :  rautc
raytc                     : UNMATCHED : raytc
reatc                     : cmene :  reatc
reetc                     : cmene :  reetc
reitc                     : cmene :  reitc
reotc                     : cmene :  reotc
reutc                     : cmene :  reutc
reytc                     : UNMATCHED : reytc
riatc                     : cmene :  riatc
rietc                     : cmene :  rietc
riitc                     : cmene :  riitc
riotc                     : cmene :  riotc
riutc                     : cmene :  riutc
riytc                     : cmene :  riytc
roatc                     : cmene :  roatc
roetc                     : cmene :  roetc
roitc                     : cmene :  roitc
rootc                     : cmene :  rootc
routc                     : cmene :  routc
roytc                     : UNMATCHED : roytc
ruatc                     : cmene :  ruatc
ruetc                     : cmene :  ruetc
ruitc                     : cmene :  ruitc
ruotc                     : cmene :  ruotc
ruutc                     : cmene :  ruutc
ruytc                     : cmene :  ruytc
ryatc                     : UNMATCHED : ryatc
ryetc                     : UNMATCHED : ryetc
ryitc                     : UNMATCHED : ryitc
ryotc                     : UNMATCHED : ryotc
ryutc                     : UNMATCHED : ryutc
ryytc                     : cmene :  ryytc
ra,atc                    : cmene :  ra,atc
ra,etc                    : cmene :  ra,etc
ra,itc                    : cmene :  ra,itc
ra,otc                    : cmene :  ra,otc
ra,utc                    : cmene :  ra,utc
ra,ytc                    : UNMATCHED : ra,ytc
re,atc                    : cmene :  re,atc
re,etc                    : cmene :  re,etc
re,itc                    : cmene :  re,itc
re,otc                    : cmene :  re,otc
re,utc                    : cmene :  re,utc
re,ytc                    : UNMATCHED : re,ytc
ri,atc                    : cmene :  ri,atc
ri,etc                    : cmene :  ri,etc
ri,itc                    : cmene :  ri,itc
ri,otc                    : cmene :  ri,otc
ri,utc                    : cmene :  ri,utc
ri,ytc                    : cmene :  ri,ytc
ro,atc                    : cmene :  ro,atc
ro,etc                    : cmene :  ro,etc
ro,itc                    : cmene :  ro,itc
ro,otc                    : cmene :  ro,otc
ro,utc                    : cmene :  ro,utc
ro,ytc                    : UNMATCHED : ro,ytc
ru,atc                    : cmene :  ru,atc
ru,etc                    : cmene :  ru,etc
ru,itc                    : cmene :  ru,itc
ru,otc                    : cmene :  ru,otc
ru,utc                    : cmene :  ru,utc
ru,ytc                    : cmene :  ru,ytc
ry,atc                    : UNMATCHED : ry,atc
ry,etc                    : UNMATCHED : ry,etc
ry,itc                    : UNMATCHED : ry,itc
ry,otc                    : UNMATCHED : ry,otc
ry,utc                    : UNMATCHED : ry,utc
ry,ytc                    : cmene :  ry,ytc
ra'ytc                    : cmene :  ra'ytc
re'ytc                    : cmene :  re'ytc
ri'ytc                    : cmene :  ri'ytc
ro'ytc                    : cmene :  ro'ytc
ru'ytc                    : cmene :  ru'ytc
ry'ytc                    : cmene :  ry'ytc
ry'atc                    : cmene :  ry'atc
ry'etc                    : cmene :  ry'etc
ry'itc                    : cmene :  ry'itc
ry'otc                    : cmene :  ry'otc
ry'utc                    : cmene :  ry'utc

(The other 25 forms {[aeiou]'[aeiou]} with apostrophes in them will all
be 'approved' as cmene and aren't interesting to show.)  I think the
'special' cases above are that iy and uy are allowed in cmene, but other
pairs involving {y} are not allowed.  That was my reading of the book
anyway.  I think we established some time back that any cluster involving
just a,e,i,o or u was allowed in cmene and fu'ivla (although for fu'ivla
the consonant structure must not match that of a lujvo or gismu.)

Are there any systematic faults in this analysis?  If there are, please
let me know!

-- 
Richard.
---------------------------------+-------------------------------------
Richard P. Curnow                | C++: n., An octopus made by
Weston-super-Mare, UK            |  nailing extra legs on a cat.
http://www.rrbcurnow.freeuk.com/ |