lucenenet-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Popov Egor (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (LUCENENET-556) Search on several terms returns invalid data
Date Tue, 17 Nov 2015 07:27:11 GMT

     [ https://issues.apache.org/jira/browse/LUCENENET-556?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Popov Egor updated LUCENENET-556:
---------------------------------
    Description: 
We have Index on 3 fields:
map.Key(p => p.ChildId);
map.Property(p => p.ChildId).NotAnalyzed().Stored();
map.Property(p => p.Organization).AnalyzeWith(new KeywordAnalyzer()).Stored();
map.Property(p => p.YearOfRest).AnalyzeWith(new KeywordAnalyzer()).Stored();

And we make simple query:
Lucene.Net.Search.IndexSearcher searcher = new Lucene.Net.Search.IndexSearcher(dir);
BooleanQuery booleanQuery = new BooleanQuery();
var analyzer = new StandardAnalyzer();
var query = new QueryParser(Lucene.Net.Util.Version.LUCENE_30, "Organization", analyzer).Parse("199");

var query2 = new QueryParser(Lucene.Net.Util.Version.LUCENE_30, "YearOfRest", analyzer).Parse("1");

booleanQuery.Add(query , Occur.MUST);
booleanQuery.Add(query2 , Occur.MUST);

var doc = searcher.Search(booleanQuery, 20);

doc has invalid data(docs with organization = 0), we checked query in Luke program, and everything
work fine. Sometimes, program crashes with IndexOutOfRangeException somewhere in scorer doc
has Id over bigger than actual in index.
Querying on one field works fine.
Can attach or send index file in 50 Mb

  was:
We have Index on 3 fields:
map.Key(p => p.ChildId);
map.Property(p => p.ChildId).NotAnalyzed().Stored();
map.Property(p => p.Organization).AnalyzeWith(new KeywordAnalyzer()).Stored();
map.Property(p => p.YearOfRest).AnalyzeWith(new KeywordAnalyzer()).Stored();

And we make simple query:
Lucene.Net.Search.IndexSearcher searcher = new Lucene.Net.Search.IndexSearcher(dir);
BooleanQuery booleanQuery = new BooleanQuery();
var analyzer = new StandardAnalyzer();
var query = new QueryParser(Lucene.Net.Util.Version.LUCENE_30, "Organization", analyzer).Parse("199");

var query2 = new QueryParser(Lucene.Net.Util.Version.LUCENE_30, "YearOfRest", analyzer).Parse("1");

booleanQuery.Add(analyzer, Occur.MUST);
booleanQuery.Add(analyzer2, Occur.MUST);

var doc = searcher.Search(booleanQuery, 20);

doc has invalid data(docs with organization = 0), we checked query in Luke program, and everything
work fine. Sometimes, program crashes with IndexOutOfRangeException somewhere in scorer doc
has Id over bigger than actual in index.
Querying on one field works fine.
Can attach or send index file in 50 Mb


> Search on several terms returns invalid data
> --------------------------------------------
>
>                 Key: LUCENENET-556
>                 URL: https://issues.apache.org/jira/browse/LUCENENET-556
>             Project: Lucene.Net
>          Issue Type: Bug
>    Affects Versions: Lucene.Net 3.0.3
>         Environment: Windows 10 x64, .NET 4.5.2
>            Reporter: Popov Egor
>            Priority: Critical
>         Attachments: RestChild.7z
>
>
> We have Index on 3 fields:
> map.Key(p => p.ChildId);
> map.Property(p => p.ChildId).NotAnalyzed().Stored();
> map.Property(p => p.Organization).AnalyzeWith(new KeywordAnalyzer()).Stored();
> map.Property(p => p.YearOfRest).AnalyzeWith(new KeywordAnalyzer()).Stored();
> And we make simple query:
> Lucene.Net.Search.IndexSearcher searcher = new Lucene.Net.Search.IndexSearcher(dir);
> BooleanQuery booleanQuery = new BooleanQuery();
> var analyzer = new StandardAnalyzer();
> var query = new QueryParser(Lucene.Net.Util.Version.LUCENE_30, "Organization", analyzer).Parse("199");
> var query2 = new QueryParser(Lucene.Net.Util.Version.LUCENE_30, "YearOfRest", analyzer).Parse("1");
> booleanQuery.Add(query , Occur.MUST);
> booleanQuery.Add(query2 , Occur.MUST);
> var doc = searcher.Search(booleanQuery, 20);
> doc has invalid data(docs with organization = 0), we checked query in Luke program, and
everything work fine. Sometimes, program crashes with IndexOutOfRangeException somewhere in
scorer doc has Id over bigger than actual in index.
> Querying on one field works fine.
> Can attach or send index file in 50 Mb



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message