Why can't you use just two fields? Both stored, one has accent folding filter, another does not. Then just chose the one you want search. Are your fields so big you are worried about content duplication? Could be premature optimization.
As to Ok/Not-Ok, have you by any chance changed field definition and did not reindex? That could do it. Otherwise, the treatment should be the same (well, I would expect both !stored fields to fail with highlight actually). Regards, Alex. Personal website: http://www.outerthoughts.com/ LinkedIn: http://www.linkedin.com/in/alexandrerafalovitch - Time is the quality of nature that keeps events from happening all at once. Lately, it doesn't seem to be working. (Anonymous - via GTD book) On Tue, Mar 25, 2014 at 10:10 PM, elfu <el...@yahoo.com> wrote: > Hello, > > I have the following problem to resolve using solr : > > search WITH or WITHOUT accents (selection at runtime) + highlights > > how can i configure the schema to realize this ? > > for example: > > inputString "aaa près bbb pres" > > A) accent sensitive > > 1. search for *près* highlight = "aaa <em>près</em> bbb pres" > 2. search for *pres* highlight = "aaa près bbb <em>pres</em>" > > B) accent insensitive > > 1. search for *près* highlight = "aaa <em>près</em> bbb <em>pres</em>" > 2. search for *pres* highlight = "aaa <em>près</em> bbb <em>pres</em>" > > I try with 3 field : 1 for inputString storage and highlight and 2 for indexs > > field_S (stored, !indexed, used for highlight = accent insensitive) > field_A (!stored, indexed, keep accents = accent sensitive) > field_B (!stored, indexed, remove accents = accent insensitive) > > when i search without accents on field_B the highlight it's ok (field_S > query/index op like field_B) > but when i search with accents on field_A the highlight i't not ok anymore .. > > thx for help > regards, > e > > The highlight engine can work with indexAnalyzer and/or queryAnalyzer from > other schema fields ? > (if i want to highlight the result from accent sensitive search the content > can be parsed with indexAnalyzer and the > hl.q can use queryAnalyzer from field_A ?) > >
