Hi Michael,
You checked out the latest code base. Look in the AssemblyInfo.cs file and
it should be "[assembly: AssemblyVersion("2.0.0.003")]" -- if so, then you
have the latest version.
In order for the NUnit test to pass the compression test, you have to setup
and provide your compression library. Try SharpZipLib
(http://www.icsharpcode.net/OpenSource/SharpZipLib/)
Regards,
-- George Aroush
-----Original Message-----
From: Michael Mitiaguin [mailto:[EMAIL PROTECTED]
Sent: Friday, February 23, 2007 12:09 AM
To: [email protected]
Subject: Re: Search using .Net code in Java based Index
Honestly , now I am not sure now what I checked out . I just selected head
trunk in tortoiseSVN screen . But just browsing the code it seems to be
strange to see : ( I thought there is no backward computability )
SupportClass.cs
/// <summary>
/// Use for .NET 1.1 Framework only.
/// </summary>
public class CompressionSupport
{
public interface ICompressionAdapter
{
byte[] Compress(byte[] input);
byte[] Uncompress(byte[] input);
}
private static ICompressionAdapter compressionAdapter;
public static byte[] Uncompress(byte[] input)
{
CheckCompressionSupport();
return compressionAdapter.Uncompress(input);
}
public static byte[] Compress(byte[] input)
{
CheckCompressionSupport();
return compressionAdapter.Compress(input);
}
private static void CheckCompressionSupport()
{
if (compressionAdapter == null)
{
System.String compressionLibClassName =
SupportClass.AppSettings.Get("Lucene.Net.CompressionLib.class", null);
if (compressionLibClassName == null)
throw new System.SystemException("Compression
support not configured");
Type compressionLibClass =
Type.GetType(compressionLibClassName, true);
System.Object compressionAdapterObj =
Activator.CreateInstance(compressionLibClass);
compressionAdapter = compressionAdapterObj as
ICompressionAdapter;
if (compressionAdapter == null)
throw new System.SystemException("Compression
adapter does not support the ICompressionAdapter interface");
}
}
}
}
Also I don't what version you are using . If you are in 1.9 try the
following , why there is no recognition for Net 2 , I don't know :
SharpZipLibAdapter.cs
/// To enable compression support in Lucene.Net 1.9 using .NET 1.1.
Framework, /// you will need to define 'SHARP_ZIP_LIB' and referance the
SharpLibZip /// library. The SharpLibZip library can be downloaded from:
/// http://www.icsharpcode.net/OpenSource/SharpZipLib/
///
/// You can use any other cmpression library you have by plugging it into
this /// code or by providing your own adapter as in this one.
#if SHARP_ZIP_LIB
...
On 2/23/07, Laxmilal Menaria <[EMAIL PROTECTED]> wrote:
> ok..thanks...
>
> but what should i do for now .. any other way.. ?
>
>
> On 2/23/07, Michael Mitiaguin <[EMAIL PROTECTED]> wrote:
> > Coincidentally I've just run Nunit tests against subversion version
> > and there were failed tests . One of them :
> > Lucene.Net.Document.TestBinaryDocument.TestBinaryFieldInIndex :
> > System.SystemException : Compression support not configured
> >
> >
> > Do we expect some tests to fail ?
> >
> > Michael
> >
> > On 2/23/07, Laxmilal Menaria <[EMAIL PROTECTED]> wrote:
> > > I have tried, Its gives me "Compression not configured"
> > > exception...while searching using .Net code..
> > >
> > >
> > >
> > > On 2/23/07, George Aroush <[EMAIL PROTECTED]> wrote:
> > > > Yes, the index generated by Java or C# Lucene are cross
> > > > compatible. If fact, if you share the lock file, you can have a
> > > > Java and a C# Lucene read/write to the same index concurrently.
> > > > Just make sure you are using the same release version if you do
this.
> > > >
> > > > -- George Aroush
> > > >
> > > >
> > > > -----Original Message-----
> > > > From: Dean Harding [mailto:[EMAIL PROTECTED]
> > > > Sent: Thursday, February 22, 2007 4:24 PM
> > > > To: [email protected]
> > > > Subject: RE: Search using .Net code in Java based Index
> > > >
> > > > > I am lucene user and I have created a Index using Java code
> > > > > and after that I wants to search using VB.Net code, so what should
i do ?
> > > >
> > > > The file format is the same, I believe, so it should "just work"
> > > > without anything special.
> > > >
> > > > Dean.
> > > >
> > > >
> > > >
> > >
> >
>
