Re: [Koha-devel] marc_word and searching

[Paul POULAIN]

Joshua Ferraro a écrit :

> > > that we want to use for searches (we don't need all of them as you pointed
> > > out; for instance, we will never use 300 for a search).  So we have three
> > > basic tasks:
> > >      
> > another idea, that would be better maybe :
> > replace ' by _.
> > Thus, o'brian searches o_brian, that will be stored in the DB.
> > The only limit is that a search on brian won't be successful. Tell me if 
> > it's a problem.
> >
> > Otherwise, we could add a 'index also 1 letter words', but, imho, ONLY 
> > with the 'do not index this subfield feature'.
> >    
> It seems to me like indexing on single-character words will lead to a more
> accurate search--though the marc_word table will be a bit bloated.  I suppose
> we should also think about other punctuation marks too--do we change them
> all to _ or do we leave them in the database?
>  
So... what do we decide here ? Joshua seems to prefer 1 letter indexing.
Is there anyone having an opinion on this topic ?

--
Paul POULAIN
Consultant indépendant en logiciels libres
responsable francophone de koha (SIGB libre http://www.koha-fr.org)