Hi All,
NumberTools class is different in Lucene(Java) and Lucene.Net.
"RADIX" is 36 in lucene and 16 in Lucene.Net. Therefore Java-Index is not
compatible with .NET index.
I will propose a solution for this(I marked the changes as //DIGY
*******).
Best Regards,
DIGY
public class NumberTools
{
#region BASE36 OPS //DIGY ********** New
static string digits = "0123456789abcdefghijklmnopqrstuvwxyz";
static long[] powersOf36 =
{ 1L,
36L,
36L*36L,
36L*36L*36L,
36L*36L*36L*36L,
36L*36L*36L*36L*36L,
36L*36L*36L*36L*36L*36L,
36L*36L*36L*36L*36L*36L*36L,
36L*36L*36L*36L*36L*36L*36L*36L,
36L*36L*36L*36L*36L*36L*36L*36L*36L,
36L*36L*36L*36L*36L*36L*36L*36L*36L*36L,
36L*36L*36L*36L*36L*36L*36L*36L*36L*36L*36L,
36L*36L*36L*36L*36L*36L*36L*36L*36L*36L*36L*36L
};
public static string ToString(long lval)
{
int maxStrLen = powersOf36.Length;
long curval = lval;
char[] tb = new char[maxStrLen];
int outpos = 0;
for (int i = 0; i < maxStrLen; i++)
{
long pval = powersOf36[maxStrLen - i - 1];
int pos = (int)(curval / pval);
tb[outpos++] = digits.Substring(pos, 1).ToCharArray()[0];
curval = curval % pval;
}
if (outpos == 0) tb[outpos++] = '0';
return new string(tb, 0, outpos).TrimStart('0');
}
public static long ToLong(string t)
{
long ival = 0;
char[] tb = t.ToCharArray();
for (int i = 0; i < tb.Length; i++)
{
ival += powersOf36[i] * digits.IndexOf(tb[tb.Length - i -
1]);
}
return ival;
}
#endregion
private const int RADIX = 36; //DIGY ********** Changed
private const char NEGATIVE_PREFIX = '-';
// NB: NEGATIVE_PREFIX must be < POSITIVE_PREFIX
private const char POSITIVE_PREFIX = '0';
//NB: this must be less than
/// <summary> Equivalent to longToString(Long.MIN_VALUE)</summary>
public static readonly System.String MIN_STRING_VALUE =
NEGATIVE_PREFIX + "0000000000000";
/// <summary> Equivalent to longToString(Long.MAX_VALUE)</summary>
public static readonly System.String MAX_STRING_VALUE =
POSITIVE_PREFIX + "1y2p0ij32e8e7";
/// <summary> The length of (all) strings returned by [EMAIL PROTECTED]
#longToString}</summary>
public static readonly int STR_SIZE = MIN_STRING_VALUE.Length;
/// <summary> Converts a long to a String suitable for
indexing.</summary>
public static System.String LongToString(long l)
{
if (l == System.Int64.MinValue)
{
// special case, because long is not symetric around zero
return MIN_STRING_VALUE;
}
System.Text.StringBuilder buf = new
System.Text.StringBuilder(STR_SIZE);
if (l < 0)
{
buf.Append(NEGATIVE_PREFIX);
l = System.Int64.MaxValue + l + 1;
}
else
{
buf.Append(POSITIVE_PREFIX);
}
System.String num = ToString(l); //DIGY ********** Changed
int padLen = STR_SIZE - num.Length - buf.Length;
while (padLen-- > 0)
{
buf.Append('0');
}
buf.Append(num);
return buf.ToString();
}
/// <summary> Converts a String that was returned by [EMAIL PROTECTED]
#longToString} back to a
/// long.
///
/// </summary>
/// <throws> IllegalArgumentException </throws>
/// <summary> if the input is null
/// </summary>
/// <throws> NumberFormatException </throws>
/// <summary> if the input does not parse (it was not
a
String returned by
/// longToString()).
/// </summary>
public static long StringToLong(System.String str)
{
if (str == null)
{
throw new System.NullReferenceException("string cannot be
null");
}
if (str.Length != STR_SIZE)
{
throw new System.FormatException("string is the wrong
size");
}
if (str.Equals(MIN_STRING_VALUE))
{
return System.Int64.MinValue;
}
char prefix = str[0];
long l = ToLong(str.Substring(1)); //DIGY ********** Changed
if (prefix == POSITIVE_PREFIX)
{
// nop
}
else if (prefix == NEGATIVE_PREFIX)
{
l = l - System.Int64.MaxValue - 1;
}
else
{
throw new System.FormatException("string does not begin
with
the correct prefix");
}
return l;
}
}
-----Original Message-----
From: George Aroush [mailto:[EMAIL PROTECTED]
Sent: Wednesday, April 18, 2007 4:14 AM
To: [email protected]
Subject: RE: Exception while search in Lucene.Net and Index prepared by
LuceneJava
Hi Laxmilal,
I don't have the free cycles to look into this right now. However, now
that
you narrowed this down to one field, "CellNo", can you narrow it down even
further to one document or does it not matter? Also, can you try
Field.Store.NO instead of Field.Store.YES ?
Regards,
-- George
> -----Original Message-----
> From: Laxmilal Menaria [mailto:[EMAIL PROTECTED]
> Sent: Monday, April 16, 2007 5:26 AM
> To: [email protected]
> Subject: Re: Exception while search in Lucene.Net and Index
> prepared by LuceneJava
>
> I have reindex the same, and got exception while
> reading..also posted sample java program for indexing and
> vb.net for reading..
>
> Thanks
> --LM
>
> On 4/15/07, Laxmilal Menaria <[EMAIL PROTECTED]> wrote:
> >
> >
> >
> > On 4/13/07, Laxmilal Menaria <[EMAIL PROTECTED]> wrote:
> > >
> > > JAVA CODE :
> > >
> > > import java.io.*;
> > > import java.util.Date;
> > >
> > > import org.apache.lucene.analysis.standard.*;
> > > import org.apache.lucene.document.*; import
> > > org.apache.lucene.index.*;
> > >
> > > public class Indexer {
> > > public static void main(String[] args) {
> > > try
> > > {
> > >
> > > File indexDir = new File("C:/LuceneApp/Testindex");
> > > long start = new Date().getTime();
> > > int numIndexed = index(indexDir);
> > > long end = new Date().getTime();
> > > System.out.println("Indexing " + numIndexed +
> " docs took "
> > > + (end - start) + " milliseconds");
> > >
> > > }catch(Exception e)
> > > {
> > > System.out.println("Exception by
> main=>"+e.toString());
> > > }
> > >
> > > }
> > > // open an index and start file directory traversal
> > > public static int index(File indexDir) throws IOException
> > > {
> > > System.out.println("Entering into Index method");
> > > IndexWriter writer = new IndexWriter(indexDir,new
> > > StandardAnalyzer(), true);
> > > writer.setUseCompoundFile(false);
> > >
> > > indexFile(writer);
> > > int numIndexed = writer.docCount();
> > > writer.optimize();
> > > writer.close();
> > > return numIndexed;
> > >
> > > }
> > > private static void indexFile(IndexWriter writer)
> > > {
> > > try
> > > {
> > > Document doc = new Document();
> > > doc.add(new Field("CellNo", NumberTools.longToString
> > > (12345678),Field.Store.YES,Field.Index.TOKENIZED));
> > > writer.addDocument(doc);
> > >
> > > doc = new Document();
> > > doc.add(new Field("CellNo", NumberTools.longToString
> > > (234569455),Field.Store.YES,Field.Index.TOKENIZED));
> > > writer.addDocument(doc);
> > >
> > > doc = new Document();
> > > doc.add(new Field("CellNo", NumberTools.longToString
> > > (567895000),Field.Store.YES,Field.Index.TOKENIZED));
> > > writer.addDocument(doc);
> > >
> > > doc = new Document();
> > > doc.add(new Field("CellNo", NumberTools.longToString
> > > (789511111),Field.Store.YES,Field.Index.TOKENIZED));
> > > writer.addDocument(doc);
> > >
> > > doc = new Document();
> > > doc.add(new Field("CellNo", NumberTools.longToString
> > > (455656448),Field.Store.YES,Field.Index.TOKENIZED));
> > > writer.addDocument(doc);
> > > }
> > > catch(Exception e)
> > > {
> > > System.out.println("Exception =>"+ e.toString());
> > > }
> > >
> > >
> > > }
> > > }
> > >
> > > VB.NET CODE:
> > >
> > > Dim objIndexReader As Lucene.Net.Index.IndexReader =
> > > Lucene.Net.Index.IndexReader.Open("C:\LuceneApp\Testindex")
> > > Dim sResults() As String
> > > Dim iTotalDocs As Integer = objIndexReader.NumDocs
> > > Dim iCounter As Integer
> > > While iCounter < iTotalDocs
> > > ReDim Preserve sResults(iCounter)
> > > sResults(iCounter) =
> > >
> Lucene.Net.Documents.NumberTools.StringToLong(objIndexReader.Documen
> > > t
> > > (iCounter).Get("CellNo"))
> > > iCounter += 1
> > > End While
> > >
> > > On 4/13/07, Laxmilal Menaria <[EMAIL PROTECTED] > wrote:
> > > >
> > > > ok..thanks, I will post sample of both ..
> > > >
> > > > -LM
> > > >
> > > >
> > > > On 4/13/07, George Aroush <[EMAIL PROTECTED] > wrote:
> > > > >
> > > > > Hi Laxmilal,
> > > > >
> > > > > Can you re-index with one field at a time and see which field
> > > > > maybe the cause? I'm betting it's the one you are storing
> > > > > NumberTools.StringToLong.
> > > > >
> > > > > Also, can you post the sample Java code?
> > > > >
> > > > > Thanks!
> > > > >
> > > > > -- George
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: Laxmilal Menaria [mailto: [EMAIL PROTECTED]
> > > > > > Sent: Friday, April 13, 2007 7:09 AM
> > > > > > To: [email protected]
> > > > > > Subject: Exception while search in Lucene.Net and Index
> > > > > > prepared by LuceneJava
> > > > > >
> > > > > > Hello everyone,
> > > > > >
> > > > > > I have created Index using MyJava program..but when I have
> > > > > > search using .Net code, it gives exception
> "Run-time exception
> > > > > > thrown :
> > > > > > System.FormatException- string is the wrong size "..
> > > > > >
> > > > > > MyIndexCollection have 3 Fileds: Name, Address and Cell No.
> > > > > > and I am index Cell No as NumberTools.LongtoString,
> and while
> > > > > > reading convert to NumberTools.StringToLong, its working in
> > > > > > java, but when created same with Lucene.net, it gives above
> > > > > > exception..
> > > > > >
> > > > > > What is the problem? Please suggest me...
> > > > > >
> > > > > > My lucene version in both Java and .Net is 2.0.
> > > > > >
> > > > > > Thanks in advance,
> > > > > > Laxmilal
> > > > > >
> > > > >
> > > > >
> > > >
> > >
> >
>
