Hi again, thanks for this information. I've since run some tests and, of course, they reflect what you said, but it's good to have a confirmation.
Grazie, Danilo On Mon, Jan 30, 2017 at 3:42 PM, alessandro.benedetti <abenede...@apache.org > wrote: > You are welcome ! > Related your points : > > 1) until you replace chars in a 1 to 1 manner, you should see no problem. > If your mappings become different, you can always use an highlighter that > will take your analysis chain in consideration > > 2) term count would not be a problem as you will count the normalised > version of your term ( tempus -> tempvs , so tempvs will be in the index) . > Of course this means that you will treat effectively tempus and tempvs as > the same thing. > > Cheers > > > > -- > View this message in context: http://lucene.472066.n3. > nabble.com/allograph-matching-tp4317807p4317835.html > Sent from the Solr - User mailing list archive at Nabble.com. >
