lucene-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Carrette, George" <>
Subject RE: Searching sub string
Date Sun, 27 Jan 2008 13:36:39 GMT
In answer to Moran Schemer's problem of not having access to the latest
Lucene features because he is using .NET:

I know this is heresy and anathema, but one way out of the Lucene .Net
ghetto problem is to use the mainline Lucene release Java source code
and compile it with a Java compiler that targets the .NET CLR.

That's why I did some testing and benchmarking of the idea, using the J#
compiler a couple years ago. is the result. I did
the work using Lucene 1.9.1 and also what was at the time the 2.0 trunc.
But the point is that I kept careful track of how much time it took and
how few lines of code needed to be modified, and the result was that you
can have a big payoff with very little effort, assuming that you are
comfortable with programming in Java.

So give it a try, it should take you at most a couple hours to compile
and test the Java source code for Lucene, (that's all it took me).

The heresy and anathema comes from the fact that Microsoft has announced
the retirement of their J# compiler and the end of all support in the
year 2015.
So you'll have 7 years to worry if anyone is going to complete the GCC
back-end that targets the CLR, or if any alternative Java compiler will
become available. But I'm not suggesting this as a long range plan, more
of a stop-gap measure that exploits the technology that currently

-----Original Message-----
From: mortic44 [] 
Sent: Saturday, January 26, 2008 4:40 AM
Subject: Re: Searching sub string

Thanks Daniel.

Unfortunately i'm using Lucene .Net and
"QueryParser.setAllowLeadingWildcard()" is available on ver 2.1 which
not released yet.


Daniel Naber-8 wrote:
> On Freitag, 25. Januar 2008, mortic44 wrote:
>> I'd like to perform a substring search (query like *foo*).
>> As you better know, it is not possible to use * as the first
> It is, at last in recent versions of Lucene Java, see 
> QueryParser.setAllowLeadingWildcard(). An alternative might be to
> an n-gram index.
> Regards
>  Daniel
> -- 

View this message in context:
Sent from the Lucene - General mailing list archive at

This message may contain privileged or otherwise confidential information. If you are not
the intended recipient, please immediately advise the sender by reply email and delete the
message and any attachments without using, copying or disclosing the contents. (FE1)

View raw message