Can you give a more detailed example, please? Including the schema bits. There is a bunch of assumptions in here that are hard to really make sense of. Solr works with tokens, but you are talking about letter repetitions. Also, if you want to sort by the string, why not just use sort parameter? It is not clear why your use case is harder than that
There is also a way to mark the field to not use frequency or length information (omitNorms, omitTermFreqAndPositions) but I hesitate to suggest those as you are talking about letters. Regards, Alex. On Mon, 30 Sep 2019 at 13:31, <yuri.glad...@swisscom.com> wrote: > > Hello > Is it possible to turn off the weighted search for Solr? > I mean the results have to be presented in a pure alphabetical order, not by > the default weighted order. So if a certain letter appears in a word 2 times, > this word shouldn' t be ranked higher. > I spent the whole day trying to find the solution in the internet, tried > various options with EdgeNGramFilterFactory, but it still doesn't work (the > results starting with "a" have to come up first and the results starting with > "z" come up last, currently I have "taa" as a first result, just because the > letter "a" appears there two times). > Thanks in advance > Yuri Gladkov >
