On 19/02/2013 20:56, Paul Taylor wrote:
Strange test failure after converting code from Lucene 3.6 to Lucene 4.1
public void testIndexPuid() throws Exception {
addReleaseOne();
RAMDirectory ramDir = new RAMDirectory();
createIndex(ramDir);
IndexReader ir = IndexReader.open(ramDir);
Fields fields = MultiFields.getFields(ir);
Terms terms = fields.terms("puid");
TermsEnum termsEnum = terms.iterator(null);
termsEnum.next();
assertEquals("efd2ace2-b3b9-305f-8a53-9803595c0e38",
termsEnum.term());
}
returns:
Expected :efd2ace2-b3b9-305f-8a53-9803595c0e38
Actual :[65 66 64 32 61 63 65 32 2d 62 33 62 39 2d 33 30 35 66 2d 38
61 35 33 2d 39 38 30 33 35 39 35 63 30 65 33 38]
It seems to be adding the field as a binary field rather than a text
field, but I checked and the field is being added using the deprecated
new Field("puid", value, Field.Index.NOT_ANALYZED_NO_NORMS, new
KeywordAnalyzer())
so shouldn't that work the same way as before ?
Doh, my bad missing utf8ToString()
assertEquals("efd2ace2-b3b9-305f-8a53-9803595c0e38",
termsEnum.term().utf8ToString());
Paul
---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org
For additional commands, e-mail: java-user-h...@lucene.apache.org
