You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on May 6, 2021. It is now read-only.
Looks like two of these (âtayôhkâna and êkawiya) are Cree words on in the English glosses, that the Stanford English parser has analyzed as "English" (likely using contextual heuristics to guess their POS). These should not be English search keys, but might require some special coding to postprocess the English parse results (and probably some manual scrutiny). The circumflexes in these words could be a give-away on them actually being Cree.
And then the French preposition à is often recognized by English parsers who the sake of dealing with French names used in English text. That might be acceptable.
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Labels
bugSomething isn't working🌭nakkiLittle tasks thrown at somebody to be Finnished.
Ryan Pineo found the following "English" words in
engcrk.xml
:The text was updated successfully, but these errors were encountered: