I'm currently using LUCENE 3.0 for the Web site at Electronic Code of Federal Regulations.

I can see how to assign an analyzer to index the entire document and then specify that each field is either ANALYZED or NOT_ANALYZED.

But is it possible to specify different analyzers for different fields.

I would like to index most of the document using the StandardAnalyzer, but there is one field that I want to index using the KeywordAnalyzer. Is this possible? How do you do that?

I have a field that holds a node for the document that looks something like this: sp40.7.60.aa_lc_a. The standard analyzer turns the underscores to spaces and makes the final "a" a stop word. I would like to search this field. I guess the other option would be to make it NOT_ANALYZED, but then I can't see how to use the MultiFieldQueryParser with a not analyzed field.

Any help would be greatly appreciated.