When running divvun-cgspell (here on SME text, but the issue is language independent), the output includes weight tags like:
"<váikuhusaid>"
"váikkuhus" N Sem/Prod Pl Acc <spelled> <W:29.160135> <WA:99.16013> "váikkuhusaid"S
When running divvunspell suggest -v -A, the same correction shows:
váikkuhusaid váikkuhus N Sem/Prod Pl Acc 9.16014 29.16014 (lex: 9.16014, mut: 15.00000, rew: 0/5/0)
váikkuhusaid váikkuhus N Sem/Prod Pl Gen 9.16014 29.16014 (lex: 9.16014, mut: 15.00000, rew: 0/5/0)
While the total weight (the <W:nnn> tag in the first fragment, the second weight in the second fragment) is the same (and correct), the <WA:nnn> element in divvun-cgspell as used in divvun-runtime reports something out of the blue. I don't know whether we need the WA weight at all, but we do need weight breakdowns, as seen in the divvunspell example above.
A SME sentence to test with is:
Ráđđehus unnidii jahkkái 2020 bušeahta juste integrerejuvvon oahpahussii ja mii oadnit váikuhusaid das dál go Girona ja Staare gielddat leat mearridan ii fállat sámi integrerejuvvon oahpahusa dán jagi ee. go váilu ruhtadeapmi.
When running
divvun-cgspell(here on SME text, but the issue is language independent), the output includes weight tags like:When running
divvunspell suggest -v -A, the same correction shows:While the total weight (the
<W:nnn>tag in the first fragment, the second weight in the second fragment) is the same (and correct), the<WA:nnn>element indivvun-cgspellas used in divvun-runtime reports something out of the blue. I don't know whether we need theWAweight at all, but we do need weight breakdowns, as seen in thedivvunspellexample above.A SME sentence to test with is: