I am working on a B2B project. A lot of the customers already do know the unique identifier of the product and will do a search. The product identify is alphanumeric and if I only boost code in the commerceSearchTextPopulator with NonFuzzyFreeTextQueryBuilder, everything is fine. As soon as a add attribute name, it will add everything to the results and the product with the correct ID isn't even the first on the list. The unique key is NOT in the name fields, so I don't see why the other products are shown at all.
I had a look in the solr admin console and had the same result.
The example product UID looks like that A9B9999999999.
I have the bad feeling that Solr will recognize the numbers as separators and only search for A and B in the text. Does someone know if this is correct?