Closed albbas closed 16 years ago
Date: 2007-05-17 22:52:09 +0200
From: Trond Trosterud <
2-part compounds
It seems 2-part compounds with i and u shortening are not recognised: váldorádji váldouksa viessodihtor (it recognises viessudihtor) guollebusse (it recognises guollibusse)
analyser gives váldorádji váldu+N+SgNomCmp#rádji+N+Sg+Nom váldouksa váldu+N+SgNomCmp#uksa+N+Sg+Nom guollebusse guolli+N+SgNomCmp#busse+N+Sg+Nom
2-part compounds with other vowels are ok. giellaguolli
2-part compounds with consonants are ok duottarguolli
2-partcompounds with names are not recognised Oslobiila (ok: Oslo-biila) Harstadbiila (ok: Harstad-biila)
Date: 2007-05-18 00:06:34 +0200
From: Trond Trosterud <
... and also with cns-final 3syll: bárrastávvalsániin bárra+N+SgNomCmp#stávval+N+SgGenCmp#sátni+N+Pl+Loc says lookup, but speller says no.
Date: 2007-05-18 09:21:03 +0200
From: Thomas Omma <
here the PLX-konvertion (?) is to blame, the original-files workes as they should: guollebusse guolli+N+SgNomCmp#busse+N+Sg+Nom guollibusse guollibusse +?
the Oslo-biila thing seems excellent to me! Shouldnt it be this way?
Date: 2007-05-18 10:50:43 +0200
From: Trond Trosterud <
As for Oslobiila: I am not sure, actually. Tomi just opened up for it. Cf. my posting to Unison.
Date: 2007-05-21 09:25:47 +0200
From: Trond Trosterud <
The error is still in PBeta 1, fp: oahpponeavvuid, fn: oahppuneavvuid.
Date: 2007-06-07 15:28:33 +0200
From: Sjur Nørstebø Moshagen <
Oslo-biila is already discussed, and dismissed.
The other compounding problem is a known deficit in the released beta. Tomi is working on it.
Changed component and version correspondingly.
Date: 2007-06-08 12:06:44 +0200
From: Sjur Nørstebø Moshagen <
Updated the Summary to reflect what this is all about.
One more important category to check is words of the type:
suomasápmelaš
So far the speller has also accepted:
suopmasápmelaš
which is wrong.
Date: 2007-06-26 10:26:10 +0200
From: Sjur Nørstebø Moshagen <
The
suopmasápmelaš
error is a (slightly) different bug. I'll now close this one, since it is basically fixed, and create a new bug report for "suopmasápmelaš". Also other bugs found in the PLX conversion should get new bug reports, unless it really is a continuation of this one.
Date: 2007-10-05 07:32:10 +0200
From: Sjur Nørstebø Moshagen <
Davvisámi, public beta 2, 2007-10-03
There were two regressions in this bug:
bárrastávvalsániin duottarguolli
are no longer accepted. On the other hand:
suopmasápmelaš
is now rejected, which is a step forward. I still reopen this bug to investigate the two cases of regressions here.
Date: 2007-11-23 14:49:33 +0100
From: Sjur Nørstebø Moshagen <
Fixed a while ago.
Case closed.
This issue was created automatically with bugzilla2github
Bugzilla Bug 397
Date: 2007-05-17T22:52:09+02:00 From: Trond Trosterud <>
To: Tomi Pieski <>
CC: sjur.n.moshagen, thomas.omma
Last updated: 2007-11-23T14:49:33+01:00