https://bugs.kde.org/show_bug.cgi?id=391881

David Edmundson <[email protected]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[email protected]

--- Comment #2 from David Edmundson <[email protected]> ---
Architecturally it's not do-able for the general case.

In the example there, you have a static dictionary of known words. 

krunner doesn't know about all known things that could possibly be found; it
would be super memory expensive to do that and would be an API break for
runners which we can't feasibly do.

Making *just* the kservices runner support mispelled words (as per the original
report) would be maybe do-able.

You'd need a completely different algorithm to the one linked. Author there is
effectively finding things with a levenshtein distance of 1. but in krunner
we're almost always dealing with substrings in krunner i.e 'wri' for writer -
and even then in the service file, "writer" won't even be the whole string
either. It's is a completely separate (I don't think solved) problem.

-- 
You are receiving this mail because:
You are watching all bug changes.

Reply via email to