On 3-Mar-08, at 4:59 PM, Karsten Sperling wrote:
Is there any work going on or planned on a Similarity sub-class that allows the lenghtNorm() behaviour to be changed via schema.xml on a field-by-field basis? At the moment I'm really only after omitting length norms for certain fields, but using a custom similarity that checks for hard coded field names seems like a rather inflexible kludge. Ideally the lengthNorm() call would be delegated to the FieldType instance of the field in question where a specific behaviour could be plugged in by specifying the class to use (similar to analyzer and tokenizer). Any comments on what others have done in this direction would be appreciated.
I'm using this "kludge". I only have 3-4 distinct field type classes, so it doesn't feel too bad.
-Mike
