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
if we allow fuzziness, there are a lot of irrelevant suggestions. E.g. a search for 'hand' brings up 'elephant'.
if we limit fuzziness, there are no more matches on characters with diacritics.
FuseJS since ill-equipped to support both. So we may need to find alternative solutions, or implement a workaround solution, inspired by this. Note that the exact same approach won't work for us, since - as mentioned in the thread - the normalization will remove diacritics from the suggested terms. Instead we'll probably want to add a second, hidden field. Keeping one (normalized) for search, one (original) for display.
The text was updated successfully, but these errors were encountered:
This is now implemented, by indexing two versions of each tag: original and without diacritics. Search is then run only against the non-diacritics term.
We currently have a small dilemma with search:
FuseJS since ill-equipped to support both. So we may need to find alternative solutions, or implement a workaround solution, inspired by this. Note that the exact same approach won't work for us, since - as mentioned in the thread - the normalization will remove diacritics from the suggested terms. Instead we'll probably want to add a second, hidden field. Keeping one (normalized) for search, one (original) for display.
The text was updated successfully, but these errors were encountered: