: But I always thought such code:
:
: Term term = new Term("text", str);
: TermQuery tq = new TermQuery(term);
: query.add(tq, Occur.SHOULD);
:
: would get query terms through analyzers - since they are specified under
the underlying lucene query classes (like TermQuery) don't know anything
baout analyzers -- definitely nothing about Solr specific field types.
if you think about it: it would be impossible for them to: if you've got a
quoted string like "foo bar" you need to analyze it in order to know
wether it should be a single TermQuery (because of KeywordTokenizer), a
PhraseQuery with two terms (because of whitespace tokenizer), a
PhraseQuery with a dozen terms (because of an ngram tokenizer) or a
MultiTermPrahseQuery (because of a synonym filter)
it's the SolrQueryParser that knows about the FieldTypes in the schema.xml
-Hoss
