[bug #49375] pkg name similarity measurement is off / too simplistic

Please use the bug tracker to post updates to a bug report.  The mailing list is intended as a read-only notification stream.  Info posted to this mailing list address won't appear in the tracker database where it is most useful.

Follow-up Comment #7, bug #49375 (group octave):

There is also a compiled oct function in statistics, named *editDistance*,
which computes Levenshtein distance and could be used to quickly return a
distance vector from an arbitrary string to the names of known packages.

With the current implementation (Octave 11.0.92), the similarity measurement
is happening in *get_pkg_info.m* using the following function

## Try a simplistic method to determine similar names.
    function d = fdist (x)

      len1 = numel (name);
      len2 = numel (x);
      lo = min (len1, len2);
      excess = max (len1, len2) - lo;
      d = sum (abs (lower (name(1:lo)) - lower (x(1:lo)))) + excess * 23;

    endfunction




    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?49375>

_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/

-----BEGIN PGP SIGNATURE-----

iHUEABYIAB0WIQQk97aszIMMAvLLwm6qLAuaBUf3TgUCaZHT1QAKCRCqLAuaBUf3
Tp2DAQCdyds2gCZjzr/M6goy4pBjLyAvYvIAk/n2BQJ5vpsZcgEAlwbc8E+Us1JK
DwQmAwf67HR7gkDHptrJVgzGMNePKA4=
=LhwD
-----END PGP SIGNATURE-----