The culprit here is text_general.  Your field is getting tokenized at index
time, and each token gets returned as a facet value.  Fields that you plan
to use for faceting (or sorting) should be string or some numeric type.
Common practice is to define a second field of type string and use a
copyField directive in your schema to copy the value of the first field
into the second.  Use the new field for sorting and faceting, and use the
original text_general field for full text search.

Reply via email to