gmane.comp.jakarta.lucene.clucene.devel

Re: Finding all the fields used in a query..

Itamar Syn-Hershko — 2012-05-03T08:49:17

It looks like you are trying to use Lucene as a database, a document
database to be specific, and it actually isn't really supported out of the
box

Take a look at MongoDB, CouchDB or RavenDB.

On Thu, May 3, 2012 at 10:56 AM, Mike Aubury <mike-236R7u3V+AEAvxtiuMwx3w< at >public.gmane.org> wrote:

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/_______________________________________________
CLucene-developers mailing list
CLucene-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/clucene-developers

Finding all the fields used in a query..

Mike Aubury — 2012-05-03T07:56:30

I'm writing some code at the minute and I have two main issues outstanding..

1) I want to aggregate data based on a user defined query - this means
searching through an index when I have a new record to see if a record
already exists
    If it does - I want to update the totals in there with some new value.
    If it does not exist - I want to create a new document containing my
totals to store in a secondary index - but only have the fields that are
used in the query..
    So - is there any way to traverse the Query to find out what fields are
used - so I can create a new document which only contains
    those fields ?

2) When I pull the data back at the end - I want to be able to order the
data by these totals - based on some query.
    So - for example - I might aggregate all the sales by salesman and
area, then select from that index for a particular area and have it ordered
by the total sales.
    If it makes it any clearer - in SQL it would be something like - select
salesman, area, sum(sales)  GROUP

[PATCH] Fix for CLMultimap::put and a trivial cleanup

Catalin Iacob — 2012-04-17T20:35:37

Hello,

Please find attached 2 patches. One is a trivial cleanup so that the
2nd patch can follow the same pattern as the other put functions in
that file without introducing a silly ;;

The second one probably solves a real bug. I discovered this while
compiling Libreoffice with Clang which warns that the v value is
unused in put. So I'm interested in committing this patch since
without it clients like Libreoffice can't compile with -Werror. I'm
not familiar with the code and I've never used CLucene before but I
suppose both patches are correct since they're quite trivial.

In case this is needed for accepting the contribution: patches are
under LGPL or Apache Version 2.0.

Catalin Iacob
------------------------------------------------------------------------------
Better than sec? Nothing is better than sec when it comes to
monitoring Big Data applications. Try Boundary one-second 
resolution app monitoring today. Free.
http://p.sf.net/sfu/Boundary-dev2dev_______________________________________________
CLu

Re: Bug in SegmentMerger and IndexWriter

Šplíchal Jiří — 2012-04-13T07:24:55

Hi Ben,

We are using slightly different version but I tested the issue with
sources from origin/master and also from origin/vc6-fixes-working
and the test crashes as described in all the cases - at least in debug version.
But we found the issue when using release version.
I am using VS2005 on Windows 7.

Jiri


From: Ben van Klinken [mailto:bvanklinken< at >gmail.com]
Sent: Friday, April 13, 2012 12:50 AM
To: clucene-developers< at >lists.sourceforge.net
Subject: Re: [CLucene-dev] Bug in SegmentMerger and IndexWriter

Hi Jiri,

What branch are you using? I just tried compiling and running that test and it ran without any issues

ben
2012/4/12 Šplíchal Jiří <splichal< at >tovek.cz<mailto:splichal< at >tovek.cz>>
Hello,

please find attached TestIndexWriter.cpp extended with one more test that
shows two different bugs:
1) If you leave the test as it is, than it crashes on line 680 when optimizing an index that
was previously empty (all documents were deleted) and then some documents has been added.

Re: Bug in SegmentMerger and IndexWriter

Ben van Klinken — 2012-04-12T22:50:19

Hi Jiri,

What branch are you using? I just tried compiling and running that test and
it ran without any issues

ben

2012/4/12 Šplíchal Jiří <splichal-jjj3dKdegmI< at >public.gmane.org>

------------------------------------------------------------------------------
For Developers, A Lot Can Happen In A Second.
Boundary is the first to Know...and Tell You.
Monitor Your Applications in Ultra-Fine Resolution. Try it FREE!
http://p.sf.net/sfu/Boundary-d2dvs2_______________________________________________
CLucene-developers mailing list
CLucene-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/clucene-developers

Bug in SegmentMerger and IndexWriter

Šplíchal Jiří — 2012-04-12T12:18:19

Hello, 

please find attached TestIndexWriter.cpp extended with one more test that
shows two different bugs:
1) If you leave the test as it is, than it crashes on line 680 when optimizing an index that 
was previously empty (all documents were deleted) and then some documents has been added.
We were able to fix this bug, the problem is in SegmentMerger.cpp 769 where 
          reader->norms(fi->name, normBuffer.values);
is called with normBuffer.values == NULL.
To prevent this it is sufficient to initialize the normBuffer at the beginning of the method
with length 1.

2) If you uncomment writer->optimize(); on line 660 in the test, than the following call to 
writer->close(); crashes. We have not found any solution for this issue where we could be 
sure about all consequences. But the problematic method in this case is
bool IndexWriter::flushDocStores()
which intializes a local variable files by calling:
const std::vector<std::string>& files = docWriter->files();
but the following calls have as side effect

Re: Does CLucene 32 bits version work on 64 bitswindows?

Veit Jahns — 2012-04-01T15:15:50

Hi Ahmed,

the same here. We also use CLucene on 64 bit Windows machines. Up to
now, we didn't face any problems with CLucene crashing.

Best regards,

Veit

2012/3/21 Ahmed <ci7nusha-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>:

------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure

MultiSearch How to?

Mohammad ismael — 2012-03-22T06:50:47

Hi,

I have a problem in searching multiple indexes.
We have two options to do this
*1- using MuliSearcher*
with this option i always get hits from the second index only however i use
MultiSearcher::subSearcher(int32_tn)  to be sure that the hit is from 
the index number (0)
here is a code sample
Searchable*s[3]={newIndexSearcher(firstIndexPath),newIndexSearcher(secondIndexPath()),NULL};
MultiSearcher * searcher  = new MultiSearcher(s);
Hits * hits = searcher->search(query);
for(int i = 0; i < hits->length(); i++){
cout<<searcher->subSearcher(i);
}

The Result is
0
0
1
1

So it means that i have hits from each index, but when i try to get the 
docs i always found that the docs came from the second index!

*2- Using MultiReader*
with this option i couldn't make the app to compile.
here is a code sample
         
lucene::util::Array<IndexReader*>*r=newlucene::util::Array<IndexReader*>(3); 


     r[0]  =  IndexReader::open(firstIndexPath);

     r[1]  =  IndexReader::open(secondIndexPath);

     r[2]  =  NULL;

Re: Does CLucene 32 bits version work on 64 bitswindows?

Ben van Klinken — 2012-03-21T23:02:48

I use 64 bit all the time, without any problems.

b


On 22 March 2012 08:31, Ahmed <ci7nusha-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org> wrote:

------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure_______________________________________________
CLucene-developers mailing list
CLucene-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/clucene-developers

Does CLucene 32 bits version work on 64 bits windows?

Ahmed — 2012-03-21T22:31:26

------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure_______________________________________________
CLucene-developers mailing list
CLucene-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/clucene-developers

Re: searchAfter / TopDocs Pagination

Veit Jahns — 2012-03-18T00:09:30

2012/2/28  <dulsi-BueaZOdMHbf4LOl6HQGc+WXnswh1EIUO< at >public.gmane.org>:

Sorry for coming back to your question so late. Somehow I thought, I
had already answered it. As far as I know Lucene++ is stable. For
CLucene we wanted it as fast as possible, i.e., reducing the usage of
the shared pointers in Lucene++.

Kind regards,
Veit

------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure

Re: help ! - using LanguageBasedAnalyzer/CJKTokenizer returns wrong result

Veit Jahns — 2012-03-11T21:20:05

Hi Vivien,

sorry, it took some time. Maybe this is an encoding issue. What
encoding do your source code files have? I got the same result, when
the encoding is UTF-8. But CLucene works internally with UTF-16 and in
this case handles the string and the query as an UTF-16 encoded
string. Thus, the created index is more or less useless.

But if I transcode the string from UTF-8 to UTF-16 before I put it
into the index, the index looks fine
and I get the correct number of results.

Kind regards,

Veit

------------------------------------------------------------------------------
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing 
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/

akaleigh

Yao Xu — 2012-03-05T14:50:37

http://cuduqjfj85.webs.com/index.htm?axipageID=30
Yeah, okay.Ill go myself, next time.Roberta, no, Henry said.

------------------------------------------------------------------------------
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2

vetabyrd

Yao Xu — 2012-03-05T14:48:25

http://jnagegum44.webs.com/index.htm?sufprof=55
 Hungry for hard information, people who were already on the run pulled off the highway to see the President speak. Thats enough, fella! Beaver cried.But the kid.

------------------------------------------------------------------------------
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2

Re: unicode type is not judged correctly on Macmachine

Vivien Meng — 2012-03-05T00:43:39

In case anyone else is having the same problem:

Problem is solved by setting the C++ compiler to LLVM GCC 4.2, rather than using Apple LLVM compiler 3.0.

-----Original Message-----
From: Vivien Meng [mailto:V.Meng-dF0/z2Fx1RMQrrorzV6ljw< at >public.gmane.org] 
Sent: Sunday, 4 March 2012 12:27 PM
To: clucene-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
Subject: [CLucene-dev] unicode type is not judged correctly on Mac machine

Hi:

I am using CLucene in Xcode on Mac.

While I debug into file gunichartables.cpp (belong to share), method bool cl_isletter(gunichar c), the TYPE that is returned by int t = TYPE(c) is not correct, I know that the particular character should belong to G_UNICODE_OTHER_LETTER (as it is a Chinese character), but TYPE does not return the correct type.

I know this works on Windows environment, I am wondering if anyone knows how to set some settings/switches on Mac to get the functions in gunichartables.cpp to work properly for multiple bytes characters on Mac.

It would

unicode type is not judged correctly on Mac machine

Vivien Meng — 2012-03-04T01:26:43

Hi:


I am using CLucene in Xcode on Mac.

While I debug into file gunichartables.cpp (belong to share), method bool cl_isletter(gunichar c), the TYPE that is returned by int t = TYPE(c) is not correct, I know that the particular character should belong to G_UNICODE_OTHER_LETTER (as it is a Chinese character), but TYPE does not return the correct type.

I know this works on Windows environment, I am wondering if anyone knows how to set some settings/switches on Mac to get the functions in gunichartables.cpp to work properly for multiple bytes characters on Mac.

It would be very much appreciated.

------------------------------------------------------------------------------
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing 
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/

Re: help ! - using LanguageBasedAnalyzer/CJKTokenizer returns wrong result

Ben van Klinken — 2012-02-28T21:47:27

I'd suggest using Luke to compare what javas behaviour is
On Feb 28, 2012 10:51 PM, "Veit Jahns" <nuncupatio-gM/Ye1E23mwN+BqQ9rBEUg< at >public.gmane.org> wrote:

------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d_______________________________________________
CLucene-developers mailing list
CLucene-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/clucene-developers

Re: searchAfter / TopDocs Pagination

2012-02-28T14:01:24

How stable is Lucene++? Does it just have some unimplemented functionality
from java lucene or is there other problems?


------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d

Re: help ! - using LanguageBasedAnalyzer/CJKTokenizer returns wrong result

Veit Jahns — 2012-02-28T12:51:17

Hi Vivien!

2012/2/27 Vivien Meng <V.Meng-dF0/z2Fx1RMQrrorzV6ljw< at >public.gmane.org>


No, I don't know anyone. Sorry.



Looks good. What will be interesting to know is, what q2->toString()
returns. Then you see what query is actually used for getting the hits. But
I will try your code myself also.

Best regards,

Veit
------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d_______________________________________________
CLucene-developers mailing list
CLucene-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/clucene-developers

Re: help ! - using LanguageBasedAnalyzer/CJKTokenizer returns wrong result

Vivien Meng — 2012-02-27T22:59:05

Hi Veit:


Thanks for your reply.

Do you know if there is anyone using CJKTokenizer?

Here is the code, no parsing error that I can see:

    Directory* store = NULL;
    store = _CLNEW RAMDirectory();
    const TCHAR* contentField = _T("contents");
    const TCHAR* test = _T("这里我们所论及的是有儿童参加的最初级足球协会red green");
    const TCHAR* cjk = _T("cjk");
    Document doc;

    IndexWriter* writer = NULL;
    IndexReader* reader = NULL;

    LanguageBasedAnalyzer* an = _CLNEW LanguageBasedAnalyzer(cjk, false);

    Query* q2 = NULL;
    Hits* h2 = NULL;
    const TCHAR* qry2 = _T("提");

    writer = _CLNEW IndexWriter(store, an, true);

    doc.add(*_CLNEW Field(contentField, test, Field::STORE_YES | Field::INDEX_TOKENIZED));

    writer->addDocument(&doc);
    writer->optimize();

    // Close and clean up
    writer->close();
    _CLLDELETE(writer);

    // verify the result
    reader =  IndexReader::open(store);
    IndexSearcher s

Re: searchAfter / TopDocs Pagination

Veit Jahns — 2012-02-27T14:38:59

Hi David,

the current CLucene version is a port of Lucene 2.3.2. But these methods
were introduced in a later version of Lucene (> 3, I think). So, there no
way to do paging out of the box.You have to implement it yourself.

Maybe Lucene++ [1] supports this. It is a port of a later Lucene version by
Alan and Ben. We tried once to make this the new CLucene version, but
havn't finished it yet.

Kind regards,

Veit

[1] https://github.com/luceneplusplus/LucenePlusPlus

2012/2/27 David Linde <davidlinde-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>

------------------------------------------------------------------------------
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2_______________________________________________
CLucene-developers mailing list
CLucene-developers