It's not a bug, because lucene work with strings, the sort is alphabetic, so 2 is greater than 1000000.
you must provide a alphabetic orderable string, is as easy as padding with 0 until the max length. If you work with negative numbers is a bit more complicated. On Fri, Oct 17, 2008 at 3:17 PM, Gaëtan Kesteloot (JIRA) <[EMAIL PROTECTED]> wrote: > RangeQuery > ---------- > > Key: LUCENENET-161 > URL: https://issues.apache.org/jira/browse/LUCENENET-161 > Project: Lucene.Net > Issue Type: Bug > Environment: Windows XP, Dotnet 2.0 > Reporter: Gaëtan Kesteloot > > > I create à range query on price. > > Datas: > Document1 => Field name=Price Value=90.01 > Document2 => Field name=Price Value=91.98 > Document3 => Field name=Price Value=105.00 > Document4 => Field name=Price Value=121.00 > Document5 => Field name=Price Value=126.00 > > When I create the range Query with name=Price LowerTerm=90.02 UpperTerm=120.00 > The result return no document > > IF i modify data and range Query > > > Document1 => Field name=Price Value=090.01 > Document2 => Field name=Price Value=091.98 > Document3 => Field name=Price Value=105.00 > Document4 => Field name=Price Value=121.00 > Document5 => Field name=Price Value=126.00 > > When I create the range Query with name=Price LowerTerm=090.02 > UpperTerm=120.00 > The result return Document2 and Document3 > > Why the range query class cannot accept lowerterm and upperterm with > different length > > -- > This message is automatically generated by JIRA. > - > You can reply to this email to add a comment to the issue online. > >
