Token Stream with Offsets (Token Sources class)

[vempap <ph...@emc.com>]

Hi,

  I've the following snippet code where I'm trying to extract weighted span
terms from the query (I do have term vectors enabled on the fields):

		File path = new File(
				"<path-to-index-dir>");
		FSDirectory directory = FSDirectory.open(path);
		IndexReader indexReader = DirectoryReader.open(directory);

		Map<String, WeightedSpanTerm> allWeightedSpanTerms = new HashMap<String,
WeightedSpanTerm>();

		WeightedSpanTermExtractor extractor = null;
		extractor = new WeightedSpanTermExtractor();
		TokenStream tokenStream = null;
		tokenStream = TokenSources.getTokenStreamWithOffsets(indexReader, 0,
"name");
		allWeightedSpanTerms.putAll(extractor.getWeightedSpanTerms(q,
tokenStream));

In the end, if I look at the map "allWeightedSpanTerms" - I don't have any
weighted span terms & when I tried to debug the code I found that when it is
trying to build the TermContext the statement "fields.terms(field);" is
returning "null" which I don't understand.

My query is : "Running Apple" (a phrase query)
my doc contents are :
name : Running Apple 60 GB iPod with Video Playback Black - Apple

Please let me know on what I'm doing anything wrong.

Thanks.
Phani.



--
View this message in context: http://lucene.472066.n3.nabble.com/Token-Stream-with-Offsets-Token-Sources-class-tp4054384.html
Sent from the Solr - User mailing list archive at Nabble.com.