Lower scores also output with HANA Fuzzy Search

[former_member185171]

Hi,

I am developing in HANA and trying thiings out with the CONTAINS predicate.

In order to make all things work I defined a full text index on the requested column - this is a NVARCHAR in our secondary database.

If I execute the SQL below, I also get results with a score lower than 1.0. If I replace the FUZZY with EXACT, I get exactly the same effect.

select

  distinct score() as score,

  ident

from z_estri

where contains( ident,

  'ethylene' ,

  fuzzy( 1.0 , 'textsearch=compare' ) )

order by score desc,

  ident

We want the perfect matches as output, but " WHERE IDENT = 'ethylene' " does not work, because some records might contain uppercases and in this case the CONTAINS returns them all with a perfect score.

If I use FUZZY( 0.9 ) I also get the lower scores in my result.

Can anyone tell me how I can get the correct result.

Thanks a lot.

Kris

[Former Member]

Well, I can't offer an answer to your question, but perhaps a workaround for your problem.

If you need all occurrences of 'ethylene', no matter which case, the classical approach will be:

WHERE LOWER(IDENT) = 'ethylene'

regards