"китаб" instead of "китап" about apertium-tat

[jonorthwash]

I think the <err_orth> approach might make sense for archaic forms? Or should we do something different?

[ftyers]

I think err_orth would be ok, or maybe use_arch ?

[mansayk]

So, I make it this way in lexc file:
китаб:китаб N1 ; ! "use_arch"
or
китаб:китаб N1 ; ! "err_orth"
right?

[ftyers]

! Use/Arch

! Err/Orth

[mansayk]

I tried to use
китаб:китаб N1 ; ! "" ! Use/Arch
but it doesn't have any effect.
echo 'китаб' | apertium-destxt -n | lt-proc -z -w 'apertium-tat/tat.automorf.bin' | cg-proc -z 'apertium-tat/tat.rlx.bin' | cg-proc -z -w -1 'apertium-tat/dev/mansur.bin' | apertium-retxt ^китаб/китаб<n><sg><nom>$
There is no additional tag or any other difference.

[ftyers]

@mansayk what should the result be ?

[mansayk]

Let's suppose we have word "китабым". And as lemma I get "китаб" instead of correct "китап". And additional tag <err_orth> here won't help, because I don't get correct lemma. I need to remove these words (marked as Err/Orth or Use/Arch) from analyze at all.

[jonorthwash]

@ftyers, @mansayk, you seem not to be communicating well. Let me try to help...

@mansayk, I believe that the Use/Arch functionality does not yet exist, and @ftyers is offering to implement it. If it were to work as expected, I believe the correct entry would be the following:

китап:китаб N1 ; ! "book"  ! Use/Arch

@ftyers, I believe you'd want the output of analysis to be something like the following. @mansayk, could you confirm that this makes sense to you as well?

^китаб/китап<n><sg><nom><use_arch>$

[mansayk]

Yes,
^китаб/китап<n><sg><nom><use_arch>$
seems good for me. It gives correct lemma and has special tag 'use_arch'.