public virtual void TestSpanNearVersusPhrase()
{
Term t1 = RandomTerm();
Term t2 = RandomTerm();
SpanQuery[] subquery = new SpanQuery[] { new SpanTermQuery(t1), new SpanTermQuery(t2) };
SpanNearQuery q1 = new SpanNearQuery(subquery, 0, true);
PhraseQuery q2 = new PhraseQuery();
q2.Add(t1);
q2.Add(t2);
AssertSameSet(q1, q2);
}
static Query ExecuteAnalyzer(Analyzer analyzer, string field, string text)
{
TokenStream tokenStream = analyzer.TokenStream(field, new StringReader(text));
ITermAttribute termAttribute = tokenStream.AddAttribute <ITermAttribute>();
IPositionIncrementAttribute positionIncrementAttribute = tokenStream.AddAttribute <IPositionIncrementAttribute>();
List <List <Term> > terms = new List <List <Term> >();
List <Term> current = null;
while (tokenStream.IncrementToken())
{
if (positionIncrementAttribute.PositionIncrement > 0)
{
current = new List <Term>();
terms.Add(current);
}
if (current != null)
{
current.Add(new Term(field, termAttribute.Term));
}
}
if (terms.Count == 1 && terms[0].Count == 1)
{
return(new TermQuery(terms[0][0]));
}
else if (terms.Select(l => l.Count).Sum() == terms.Count)
{
PhraseQuery phraseQuery = new PhraseQuery();
foreach (var positionList in terms)
{
phraseQuery.Add(positionList[0]);
}
return(phraseQuery);
}
else
{
MultiPhraseQuery multiPhraseQuery = new MultiPhraseQuery();
foreach (var positionList in terms)
{
multiPhraseQuery.Add(positionList.ToArray());
}
return(multiPhraseQuery);
}
}
public override Query VisitPhraseQuery(PhraseQuery phraseq)
{
var terms = phraseq.GetTerms();
var field = terms[0].Field();
_text.Append(field);
_text.Append(":\"");
var positions = new int[terms.Length];
for (var i = 0; i < positions.Length; i++)
positions[i] = i;
var pieces = new string[terms.Length];
for (var i = 0; i < terms.Length; i++)
{
var pos = positions[i];
var s = pieces[pos];
if (s == null)
s = (terms[i]).Text();
else
s += "|" + (terms[i]).Text();
pieces[pos] = s;
}
for (var i = 0; i < pieces.Length; i++)
{
if (i > 0)
_text.Append(' ');
var s = pieces[i];
if (s == null)
_text.Append('?');
else
_text.Append(s);
}
_text.Append("\"");
var slop = phraseq.GetSlop();
if (slop != 0)
{
_text.Append("~");
_text.Append(slop);
}
_text.Append(BoostToString(phraseq.GetBoost()));
return base.VisitPhraseQuery(phraseq);
}
public virtual void TestORPhrase()
{
PhraseQuery phrase1 = new PhraseQuery();
phrase1.Add(new Term("field", "foo"));
phrase1.Add(new Term("field", "bar"));
PhraseQuery phrase2 = new PhraseQuery();
phrase2.Add(new Term("field", "star"));
phrase2.Add(new Term("field", "wars"));
BooleanQuery expected = new BooleanQuery();
expected.Add(phrase1, BooleanClause.Occur.SHOULD);
expected.Add(phrase2, BooleanClause.Occur.SHOULD);
assertEquals(expected, Parse("\"foo bar\"|\"star wars\""));
}
private QueryProvider <TLuceneEntity> _Phrase <TResult>(string value, Expression <Func <TLuceneEntity, TResult> > selector, Occur occur)
{
TrimTerm(ref value);
if (IsNull(ref value))
{
return(this);
}
var q = new PhraseQuery();
q.Add(new Term(selector.GetName(), value));
_current.Add(q, occur);
return(this);
}
public static Query BuildExactFieldValueClause(Index index, string fieldName, string fieldValue)
{
Assert.ArgumentNotNull(index, "Index");
if (string.IsNullOrEmpty(fieldName) || string.IsNullOrEmpty(fieldValue))
{
return(null);
}
fieldValue = IdHelper.ProcessGUIDs(fieldValue);
var phraseQuery = new PhraseQuery();
phraseQuery.Add(new Term(fieldName.ToLowerInvariant(), fieldValue.ToLowerInvariant()));
return(phraseQuery);
}
public virtual void TestANDPhrase()
{
PhraseQuery phrase1 = new PhraseQuery();
phrase1.Add(new Term("field", "foo"));
phrase1.Add(new Term("field", "bar"));
PhraseQuery phrase2 = new PhraseQuery();
phrase2.Add(new Term("field", "star"));
phrase2.Add(new Term("field", "wars"));
BooleanQuery expected = new BooleanQuery();
expected.Add(phrase1, Occur.MUST);
expected.Add(phrase2, Occur.MUST);
assertEquals(expected, Parse("\"foo bar\"+\"star wars\""));
}
private List <Models.ViewSearchContentModel> SearchBookContent()
{
var indexPath = @"C:\Users\Victor\Desktop\LuceneNetDir";
var kw = Common.WebCommon.GetPanGuWord(Request["txtContent"]);
FSDirectory directory = FSDirectory.Open(new DirectoryInfo(indexPath), new NoLockFactory());
IndexReader reader = IndexReader.Open(directory, true);
IndexSearcher searcher = new IndexSearcher(reader);
//搜索条件
PhraseQuery query = new PhraseQuery();
foreach (var word in kw)//先用空格，让用户去分词，空格分隔的就是词"计算机 专业"
{
query.Add(new Term("body", word));
}
//WITH_POSITIONS_OFFSETS -> 盘古分词在建立索引的时候已经将干扰词间距记录，lucene.net只要匹配就可以了
query.Slop = 100;//多个查询条件词之间的最大距离，在文章中相隔太远也就无意义（例如 “大学生”这个查询条件和“简历”这个查询条件之间如果间隔的词太多也就没有意义了。）
//TopScoreDocCollector是盛放查询结果的容器
TopScoreDocCollector collector = TopScoreDocCollector.Create(1000, true);
searcher.Search(query, null, collector); //根据query查询条件进行查询，查询结果放入cllector容器
ScoreDoc[] docs = collector.TopDocs(0, collector.TotalHits).ScoreDocs; //得到所有查询结果中的文档，collector.TotalHits:表示总条数；TopDocs(300,20):表示得到300(从300开始)，到320(结束）的文档内容
var list = new List <Models.ViewSearchContentModel>();
//可以用来实现分页内容
for (int i = 0; i < docs.Length; i++)
{
//搜索ScoreDoc[]只能获得文档的id，这样不会把查询结果的Document一次性加载到内存中。降低了内存的压力，需要获得文档的详细内容的时候通过searcher.Doc来根据文档id来获得文档的详细内容对象document
int docId = docs[i].Doc; //得到查询结果文档的id(Lucene内容分配的id)
Document doc = searcher.Doc(docId); //找到文档id对应的文档详细信息
list.Add(new Models.ViewSearchContentModel
{
Id = doc.Get("Id"),
Title = doc.Get("Title"),
Content = Common.WebCommon.CreateHighLight(Request["txtContent"], doc.Get("Content"))//搜索内容关键字高亮显示
});
}
SearchDetailsBll.AddEntity(new SearchDetails
{
Id = Guid.NewGuid(),
KeyWords = Request["txtContent"],
SearchDateTime = DateTime.Now,
});
return(list);
}
public Query CreateQuery(ILuceneQueryService builder, LuceneQueryContext context, string type, JObject query)
{
if (type != "match_phrase")
{
return(null);
}
var first = query.Properties().First();
var phraseQuery = new PhraseQuery();
JToken value;
switch (first.Value.Type)
{
case JTokenType.String:
value = first.Value;
break;
case JTokenType.Object:
var obj = (JObject)first.Value;
if (!obj.TryGetValue("value", out value))
{
throw new ArgumentException("Missing value in match phrase query");
}
// TODO: read "analyzer" property
if (obj.TryGetValue("slop", out var slop))
{
phraseQuery.Slop = slop.Value <int>();
}
break;
default: throw new ArgumentException("Invalid wildcard query");
}
foreach (var term in LuceneQueryService.Tokenize(first.Name, value.Value <string>(), context.DefaultAnalyzer))
{
phraseQuery.Add(new Term(first.Name, term));
}
return(phraseQuery);
}
/// <summary>
/// Create the quoted query.
/// </summary>
/// <param name="text">The quoted text.</param>
/// <param name="searchFieldName">The name of the field to search.</param>
/// <returns>The query result.</returns>
private PhraseQuery CreateQuotedQuery(string text, string searchFieldName)
{
// Create the query.
PhraseQuery query = new PhraseQuery();
query.Slop = 2;
// Quoted search exact phase.
string[] words = text.Words();
for (int i = 0; i < words.Length; i++)
{
// Add the query.
query.Add(new Term(searchFieldName, words[i].Trim().ToLower()));
}
// Return the query.
return(query);
}
/// <summary>
/// Match a multi-word phrase exactly. (This is like how QueryParser handles quoted phrases)
/// </summary>
/// <param name="field"></param>
/// <param name="phrase"></param>
/// <param name="slop"></param>
/// <returns></returns>
public QueryBuilder MatchPhrase(string field, string phrase, int slop = 0)
{
if (string.IsNullOrWhiteSpace(phrase))
{
return(this);
}
var query = new PhraseQuery();
foreach (var token in _analyzer.TokenListFromString(phrase))
{
query.Add(new Term(field, token));
}
query.SetSlop(slop);
return(AddSubQuery(query));
}
public void TestSmallerFragSizeThanPhraseQuery()
{
SimpleFragListBuilder sflb = new SimpleFragListBuilder();
PhraseQuery phraseQuery = new PhraseQuery();
phraseQuery.Add(new Term(F, "abcdefgh"));
phraseQuery.Add(new Term(F, "jklmnopqrs"));
FieldFragList ffl = sflb.CreateFieldFragList(fpl(phraseQuery, "abcdefgh jklmnopqrs"), sflb.minFragCharSize);
assertEquals(1, ffl.FragInfos.size());
if (VERBOSE)
{
Console.WriteLine(ffl.FragInfos[0].toString());
}
assertEquals("subInfos=(abcdefghjklmnopqrs((0,21)))/1.0(0,21)", ffl.FragInfos[0].toString());
}
public void TestSmallerFragSizeThanPhraseQuery()
{
SimpleFragListBuilder sflb = new SimpleFragListBuilder();
PhraseQuery phraseQuery = new PhraseQuery();
phraseQuery.Add(new Term(F, "abcdefgh"));
phraseQuery.Add(new Term(F, "jklmnopqrs"));
FieldFragList ffl = sflb.CreateFieldFragList(fpl(phraseQuery, "abcdefgh jklmnopqrs"), sflb.minFragCharSize);
assertEquals(1, ffl.FragInfos.size());
if (Verbose)
{
Console.WriteLine(ffl.FragInfos[0].ToString(CultureInfo.InvariantCulture)); // LUCENENET specific: use invariant culture, since we are culture-aware
}
assertEquals("subInfos=(abcdefghjklmnopqrs((0,21)))/1.0(0,21)", ffl.FragInfos[0].ToString(CultureInfo.InvariantCulture)); // LUCENENET specific: use invariant culture, since we are culture-aware
}
public virtual void TestDemo()
{
Analyzer analyzer = new MockAnalyzer(Random());
// Store the index in memory:
using (Directory directory = NewDirectory())
{
string longTerm = "longtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongterm";
string text = "this is the text to be indexed. " + longTerm;
// To store an index on disk, use this instead:
// Directory directory = FSDirectory.open(new File("/tmp/testindex"));
using (RandomIndexWriter iwriter = new RandomIndexWriter(Random(), directory, NewIndexWriterConfig(Random(), TEST_VERSION_CURRENT, analyzer)))
{
Documents.Document doc = new Documents.Document();
doc.Add(NewTextField("fieldname", text, Field.Store.YES));
iwriter.AddDocument(doc);
}
// Now search the index:
using (IndexReader ireader = DirectoryReader.Open(directory)) // read-only=true
{
IndexSearcher isearcher = NewSearcher(ireader);
Assert.AreEqual(1, isearcher.Search(new TermQuery(new Term("fieldname", longTerm)), 1).TotalHits);
Query query = new TermQuery(new Term("fieldname", "text"));
TopDocs hits = isearcher.Search(query, null, 1);
Assert.AreEqual(1, hits.TotalHits);
// Iterate through the results:
for (int i = 0; i < hits.ScoreDocs.Length; i++)
{
Documents.Document hitDoc = isearcher.Doc(hits.ScoreDocs[i].Doc);
Assert.AreEqual(text, hitDoc.Get("fieldname"));
}
// Test simple phrase query
PhraseQuery phraseQuery = new PhraseQuery();
phraseQuery.Add(new Term("fieldname", "to"));
phraseQuery.Add(new Term("fieldname", "be"));
Assert.AreEqual(1, isearcher.Search(phraseQuery, null, 1).TotalHits);
}
}
}
/// <summary>
/// 获取搜索结果
/// </summary>
protected void btnGetSearchResult_Click(object sender, EventArgs e)
{
string keyword = txtKeyWords.Text;
string indexPath = Context.Server.MapPath("~/Index"); // 索引文档保存位置
FSDirectory directory = FSDirectory.Open(new DirectoryInfo(indexPath), new NoLockFactory());
IndexReader reader = IndexReader.Open(directory, true);
IndexSearcher searcher = new IndexSearcher(reader);
// 查询条件
PhraseQuery query = new PhraseQuery();
// 等同于 where contains("msg",kw)
query.Add(new Term("msg", keyword));
// 两个词的距离大于100（经验值）就不放入搜索结果，因为距离太远相关度就不高了
query.SetSlop(100);
// TopScoreDocCollector:盛放查询结果的容器
TopScoreDocCollector collector = TopScoreDocCollector.create(1000, true);
// 使用query这个查询条件进行搜索，搜索结果放入collector
searcher.Search(query, null, collector);
// 从查询结果中取出第m条到第n条的数据
// collector.GetTotalHits()表示总的结果条数
ScoreDoc[] docs = collector.TopDocs(0, collector.GetTotalHits()).scoreDocs;
// 遍历查询结果
IList <SearchResult> resultList = new List <SearchResult>();
for (int i = 0; i < docs.Length; i++)
{
// 拿到文档的id，因为Document可能非常占内存（DataSet和DataReader的区别）
int docId = docs[i].doc;
// 所以查询结果中只有id，具体内容需要二次查询
// 根据id查询内容：放进去的是Document，查出来的还是Document
Document doc = searcher.Doc(docId);
SearchResult result = new SearchResult();
result.Id = Convert.ToInt32(doc.Get("id"));
result.Msg = HighlightHelper.HighLight(keyword, doc.Get("msg"));
resultList.Add(result);
}
// 绑定到Repeater
rptSearchResult.DataSource = resultList;
rptSearchResult.DataBind();
}
public static List <Record> SearchFromIndex(string searchKey)
{
string indexPath = HttpContext.Current.Server.MapPath("~/IndexData");
FSDirectory directory = FSDirectory.Open(new DirectoryInfo(indexPath), new NoLockFactory());
IndexReader reader = IndexReader.Open(directory, true);
IndexSearcher searcher = new IndexSearcher(reader);
//搜索条件
BooleanQuery bQuery = new BooleanQuery();
PhraseQuery tQuery = new PhraseQuery();
PhraseQuery cQuery = new PhraseQuery();
//把用户输入的关键字进行分词
foreach (string word in SplitWords(searchKey))
{
tQuery.Add(new Term("title", word));
cQuery.Add(new Term("content", word));
}
cQuery.Slop = 100; //指定关键词相隔最大距离
bQuery.Add(tQuery, Occur.SHOULD);
bQuery.Add(cQuery, Occur.SHOULD);
//TopScoreDocCollector盛放查询结果的容器
TopScoreDocCollector collector = TopScoreDocCollector.Create(1000, true);
searcher.Search(bQuery, null, collector); //根据query查询条件进行查询，查询结果放入collector容器
//TopDocs 指定0到GetTotalHits() 即所有查询结果中的文档 如果TopDocs(20,10)则意味着获取第20-30之间文档内容 达到分页的效果
ScoreDoc[] docs = collector.TopDocs(0, collector.TotalHits).ScoreDocs;
//展示数据实体对象集合
List <Record> records = new List <Record>();
for (int i = 0; i < docs.Length; i++)
{
int docId = docs[i].Doc; //得到查询结果文档的id（Lucene内部分配的id）
Document doc = searcher.Doc(docId); //根据文档id来获得文档对象Document
var record = new Record();
record.Title = doc.Get("title");
record.Summary = doc.Get("summary");
record.ID = Guid.Parse(doc.Get("id"));
records.Add(record);
}
return(records);
}
private IList <Query> BuildQueries()
{
IList <Query> queries = new List <Query>();
BooleanQuery booleanAB = new BooleanQuery();
booleanAB.Add(new TermQuery(new Term("contents", "a")), Occur.SHOULD);
booleanAB.Add(new TermQuery(new Term("contents", "b")), Occur.SHOULD);
queries.Add(booleanAB);
PhraseQuery phraseAB = new PhraseQuery();
phraseAB.Add(new Term("contents", "a"));
phraseAB.Add(new Term("contents", "b"));
queries.Add(phraseAB);
PhraseQuery phraseABC = new PhraseQuery();
phraseABC.Add(new Term("contents", "a"));
phraseABC.Add(new Term("contents", "b"));
phraseABC.Add(new Term("contents", "c"));
queries.Add(phraseABC);
BooleanQuery booleanAC = new BooleanQuery();
booleanAC.Add(new TermQuery(new Term("contents", "a")), Occur.SHOULD);
booleanAC.Add(new TermQuery(new Term("contents", "c")), Occur.SHOULD);
queries.Add(booleanAC);
PhraseQuery phraseAC = new PhraseQuery();
phraseAC.Add(new Term("contents", "a"));
phraseAC.Add(new Term("contents", "c"));
queries.Add(phraseAC);
PhraseQuery phraseACE = new PhraseQuery();
phraseACE.Add(new Term("contents", "a"));
phraseACE.Add(new Term("contents", "c"));
phraseACE.Add(new Term("contents", "e"));
queries.Add(phraseACE);
return(queries);
}
// This is a simplified query builder which works for single Terms and single Phrases
// Returns null, TermQuery, or PhraseQuery
public static Lucene.Net.Search.Query GetFieldQuery(Analyzer analyzer, string field, string queryText)
{
TokenStream stream = analyzer.TokenStream(field, new StringReader(queryText));
TokenFilter filter = new CachingTokenFilter(stream);
filter.Reset();
// This attribute way of getting token properties sucks, but it's the non-obsolete one.
var attr1 = (TermAttribute)filter.GetAttribute(typeof(TermAttribute));
var attr2 = (PositionIncrementAttribute)filter.GetAttribute(typeof(PositionIncrementAttribute));
Func <string> getText = () => attr1 != null?attr1.Term() : null;
Func <int> getPositionIncrement = () => attr2 != null?attr2.GetPositionIncrement() : 1;
// 0 tokens
if (!filter.IncrementToken())
{
return(new BooleanQuery());
}
// 1 token?
string token1 = getText();
int position = 0;
if (!filter.IncrementToken())
{
return(new TermQuery(new Term(field, token1)));
}
// many tokens - handle first token
PhraseQuery ret = new PhraseQuery();
ret.Add(new Term(field, token1));
do
{
// handle rest of tokens
string tokenNext = getText();
position += getPositionIncrement();
ret.Add(new Term(field, tokenNext), position);
}while (filter.IncrementToken());
return(ret);
}
/// <summary>
/// 从索引库中检索关键字
/// </summary>
private void SearchFromIndexData()
{
string indexPath = Context.Server.MapPath("~/IndexData");
FSDirectory directory = FSDirectory.Open(new DirectoryInfo(indexPath), new NoLockFactory());
IndexReader reader = IndexReader.Open(directory, true);
IndexSearcher searcher = new IndexSearcher(reader);
//搜索条件
PhraseQuery query = new PhraseQuery();
//把用户输入的关键字进行分词
foreach (string word in Common.SplitContent.SplitWords(Request.QueryString["SearchKey"]))
{
query.Add(new Term("content", word));
}
//query.Add(new Term("content", "C#"));//多个查询条件时 为且的关系
query.SetSlop(100); //指定关键词相隔最大距离
//TopScoreDocCollector盛放查询结果的容器
TopScoreDocCollector collector = TopScoreDocCollector.create(1000, true);
searcher.Search(query, null, collector);//根据query查询条件进行查询，查询结果放入collector容器
//TopDocs 指定0到GetTotalHits() 即所有查询结果中的文档 如果TopDocs(20,10)则意味着获取第20-30之间文档内容 达到分页的效果
ScoreDoc[] docs = collector.TopDocs(0, collector.GetTotalHits()).scoreDocs;
//展示数据实体对象集合
List <PZYM.Shop.Model.Books> bookResult = new List <PZYM.Shop.Model.Books>();
for (int i = 0; i < docs.Length; i++)
{
int docId = docs[i].doc; //得到查询结果文档的id（Lucene内部分配的id）
Document doc = searcher.Doc(docId); //根据文档id来获得文档对象Document
PZYM.Shop.Model.Books book = new PZYM.Shop.Model.Books();
book.Title = doc.Get("title");
//book.ContentDescription = doc.Get("content");//未使用高亮
//搜索关键字高亮显示 使用盘古提供高亮插件
book.ContentDescription = Common.SplitContent.HightLight(Request.QueryString["SearchKey"], doc.Get("content"));
book.Id = Convert.ToInt32(doc.Get("id"));
bookResult.Add(book);
}
Repeater1.DataSource = bookResult;
Repeater1.DataBind();
}
public static List <ViewModelContent> ShowSearchContent(HttpRequestBase Request, string msg)
{
string indexPath = ConfigurationManager.AppSettings["lucenedirPath"];
List <string> list = Common.WebCommon.PanGuSplitWord(msg);//对用户输入的搜索条件进行拆分。
FSDirectory directory = FSDirectory.Open(new DirectoryInfo(indexPath), new NoLockFactory());
IndexReader reader = IndexReader.Open(directory, true);
IndexSearcher searcher = new IndexSearcher(reader);
//搜索条件
PhraseQuery query = new PhraseQuery();
foreach (string word in list)//先用空格，让用户去分词，空格分隔的就是词“计算机 专业”
{
query.Add(new Term("Title", word));
}
//query.Add(new Term("body","语言"));--可以添加查询条件，两者是add关系.顺序没有关系.
// query.Add(new Term("body", "大学生"));
// query.Add(new Term("body", kw));//body中含有kw的文章
query.SetSlop(100);//多个查询条件的词之间的最大距离.在文章中相隔太远 也就无意义.（例如 “大学生”这个查询条件和"简历"这个查询条件之间如果间隔的词太多也就没有意义了。）
//TopScoreDocCollector是盛放查询结果的容器
TopScoreDocCollector collector = TopScoreDocCollector.create(1000, true);
searcher.Search(query, null, collector); //根据query查询条件进行查询，查询结果放入collector容器
ScoreDoc[] docs = collector.TopDocs(0, collector.GetTotalHits()).scoreDocs; //得到所有查询结果中的文档,GetTotalHits():表示总条数 TopDocs(300, 20);//表示得到300（从300开始），到320（结束）的文档内容.
//可以用来实现分页功能
List <Models.ViewModelContent> viewModelList = new List <Models.ViewModelContent>();
for (int i = 0; i < docs.Length; i++)
{
//
//搜索ScoreDoc[]只能获得文档的id,这样不会把查询结果的Document一次性加载到内存中。降低了内存压力，需要获得文档的详细内容的时候通过searcher.Doc来根据文档id来获得文档的详细内容对象Document.
Models.ViewModelContent viewModel = new Models.ViewModelContent();
int docId = docs[i].doc; //得到查询结果文档的id（Lucene内部分配的id）
Document doc = searcher.Doc(docId); //找到文档id对应的文档详细信息
viewModel.Id = Convert.ToInt32(doc.Get("Id")); // 取出放进字段的值
viewModel.Title = doc.Get("Title");
// viewModel.Content = Common.WebCommon.CreateHightLight(Request["txtSearch"], doc.Get("Content"));//将搜索的关键字高亮显示。
viewModelList.Add(viewModel);
}
//删除汇总表中所有的数据
//再将搜索的词插入到明细表中
return(viewModelList);
}
public Query ProcessQuery(string QueryString, bool PreProcess)
{
BooleanQuery FinalQuery = new BooleanQuery();
InitializeMultiFieldQueryParser(PreProcess);
if (PreProcess)
{
//Extract all phrases
List <string> PhraseList = _PosTaggerLexicalParser.FindPhrases(QueryString);
QueryString = _PosTaggerLexicalParser.Parse(QueryString);
foreach (string phrase in PhraseList)
{
PhraseQuery abstractPhraseQuery = new PhraseQuery();
PhraseQuery titlePhraseQuery = new PhraseQuery();
abstractPhraseQuery.Add(new Term(SEDocument.ABSTRACT_FN, phrase));
titlePhraseQuery.Add(new Term(SEDocument.TITLE_FN, phrase));
abstractPhraseQuery.Boost = 1.2F;
abstractPhraseQuery.Slop = 3;
FinalQuery.Add(abstractPhraseQuery, Occur.SHOULD);
titlePhraseQuery.Boost = 4.0F;
titlePhraseQuery.Slop = 3;
FinalQuery.Add(titlePhraseQuery, Occur.SHOULD);
}
}
else
{
FinalQuery.Add(_MultiFieldQueryParser.Parse(QueryString), Occur.SHOULD);
}
string[] tokens = _PosTaggerLexicalParser.TokeniseString(QueryString.Replace('\"', ' ').Replace('[', ' ').Replace(']', ' '));
foreach (string term in tokens)
{
FinalQuery.Add(_MultiFieldQueryParser.Parse(term.Replace("~", "") + "~"), Occur.SHOULD);
}
FinalQuery.MinimumNumberShouldMatch = 2;
return(FinalQuery);
}
public void SearchFromIndexData()
{
string indexPath = System.Web.HttpContext.Current.Server.MapPath("~/IndexData");
FSDirectory directory = FSDirectory.Open(new DirectoryInfo(indexPath), new NoLockFactory());
IndexReader reader = IndexReader.Open(directory, true);
IndexSearcher searcher = new IndexSearcher(reader);
//搜索条件
PhraseQuery query = new PhraseQuery();
//把用户输入的关键字进行分词
foreach (string word in WitKeyDu.Site.Web.SplitContent.SplitWords(Request.QueryString["SearchKey"]))
{
query.Add(new Term("ForumContent", word));
}
//query.Add(new Term("content", "C#"));//多个查询条件时 为且的关系
query.SetSlop(100); //指定关键词相隔最大距离
//TopScoreDocCollector盛放查询结果的容器
TopScoreDocCollector collector = TopScoreDocCollector.create(1000, true);
searcher.Search(query, null, collector);//根据query查询条件进行查询，查询结果放入collector容器
//TopDocs 指定0到GetTotalHits() 即所有查询结果中的文档 如果TopDocs(20,10)则意味着获取第20-30之间文档内容 达到分页的效果
ScoreDoc[] docs = collector.TopDocs(0, collector.GetTotalHits()).scoreDocs;
//展示数据实体对象集合
List <Forum> ForumResult = new List <Forum>();
for (int i = 0; i < docs.Length; i++)
{
int docId = docs[i].doc; //得到查询结果文档的id（Lucene内部分配的id）
Document doc = searcher.Doc(docId); //根据文档id来获得文档对象Document
Forum forum = new Forum();
forum.ForumName = doc.Get("ForumName");
//book.Title = doc.Get("title");
////book.ContentDescription = doc.Get("content");//未使用高亮
////搜索关键字高亮显示 使用盘古提供高亮插件
forum.ForumContent = WitKeyDu.Site.Web.SplitContent.HightLight(Request.QueryString["SearchKey"], doc.Get("ForumContent"));
forum.ForumTypeID = Convert.ToInt32(doc.Get("ID"));
ForumResult.Add(forum);
}
}
public virtual void TestDemo()
{
Analyzer analyzer = new MockAnalyzer(random());
// Store the index in memory:
Directory directory = newDirectory();
// To store an index on disk, use this instead:
// Directory directory = FSDirectory.open(new File("/tmp/testindex"));
RandomIndexWriter iwriter = new RandomIndexWriter(random(), directory, analyzer);
Document doc = new Document();
string longTerm = "longtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongtermlongterm";
string text = "this is the text to be indexed. " + longTerm;
doc.add(newTextField("fieldname", text, Field.Store.YES));
iwriter.addDocument(doc);
iwriter.close();
// Now search the index:
IndexReader ireader = DirectoryReader.open(directory); // read-only=true
IndexSearcher isearcher = newSearcher(ireader);
Assert.AreEqual(1, isearcher.search(new TermQuery(new Term("fieldname", longTerm)), 1).totalHits);
Query query = new TermQuery(new Term("fieldname", "text"));
TopDocs hits = isearcher.search(query, null, 1);
Assert.AreEqual(1, hits.totalHits);
// Iterate through the results:
for (int i = 0; i < hits.scoreDocs.length; i++)
{
Document hitDoc = isearcher.doc(hits.scoreDocs[i].doc);
Assert.AreEqual(text, hitDoc.get("fieldname"));
}
// Test simple phrase query
PhraseQuery phraseQuery = new PhraseQuery();
phraseQuery.add(new Term("fieldname", "to"));
phraseQuery.add(new Term("fieldname", "be"));
Assert.AreEqual(1, isearcher.search(phraseQuery, null, 1).totalHits);
ireader.close();
directory.close();
}
public override Query VisitPhraseQuery(PhraseQuery phraseq)
{
var terms = phraseq.GetTerms();
PhraseQuery newQuery = null;
int index = 0;
int count = terms.Length;
while (index < count)
{
var visitedTerm = VisitTerm(terms[index]);
if (newQuery != null)
{
if (visitedTerm != null)
{
newQuery.Add(visitedTerm);
}
}
else if (visitedTerm != terms[index])
{
newQuery = new PhraseQuery();
for (int i = 0; i < index; i++)
{
newQuery.Add(terms[i]);
}
if (visitedTerm != null)
{
newQuery.Add(visitedTerm);
}
}
index++;
}
if (newQuery != null)
{
if (newQuery.GetTerms().Length > 0)
{
return(newQuery);
}
return(null);
}
return(phraseq);
}
/// <summary>
/// 全文搜索
/// </summary>
/// <param name="keyword"></param>
/// <param name="startRowIndex"></param>
/// <param name="pageSize"></param>
/// <param name="totalCount"></param>
/// <returns></returns>
public static List <SearchResult> DoSearch(string keyword, int startRowIndex, int pageSize, out int totalCount)
{
FSDirectory directory = FSDirectory.Open(new DirectoryInfo(_indexPath), new NoLockFactory());
IndexReader reader = IndexReader.Open(directory, true);
//IndexSearcher是进行搜索的类
var searcher = new IndexSearcher(reader);
var query = new PhraseQuery();
foreach (string word in GetKeyWords(keyword))
{
query.Add(new Term("body", word));
}
query.SetSlop(100); //相聚100以内才算是查询到
TopScoreDocCollector collector = TopScoreDocCollector.create(1024, true); //最大1024条记录
searcher.Search(query, null, collector);
totalCount = collector.GetTotalHits(); //返回总条数
ScoreDoc[] docs = collector.TopDocs(startRowIndex, pageSize).scoreDocs; //分页,下标应该从0开始吧，0是第一条记录
var list = new List <SearchResult>();
for (int i = 0; i < docs.Length; i++)
{
int docId = docs[i].doc; //取文档的编号，这个是主键，lucene.net分配
//检索结果中只有文档的id，如果要取Document，则需要Doc再去取
//降低内容占用
Document doc = searcher.Doc(docId);
string number = doc.Get("number");
string title = doc.Get("title");
string fullPath = doc.Get("fullPath");
string body = doc.Get("body");
var searchResult = new SearchResult
{
Number = number,
Title = title,
FullPath = fullPath,
BodyPreview = Preview(body, keyword)
};
list.Add(searchResult);
}
return(list);
}
/// <summary>
/// 批量添加PhraseQuery
/// </summary>
/// <param name="phrase">待搜索的短语</param>
/// <param name="fieldNameAndBoosts">字段名称及权重集合</param>
/// <param name="occur">搜索条件间的关系</param>
/// <param name="asFilter">是否作为过滤器条件</param>
/// <returns></returns>
public LuceneSearchBuilder WithPhrases(Dictionary <string, BoostLevel> fieldNameAndBoosts, string phrase, BooleanClause.Occur occur, bool asFilter = false)
{
string filteredPhrase = ClauseScrubber.LuceneKeywordsScrub(phrase);
if (string.IsNullOrEmpty(filteredPhrase))
{
return(this);
}
string[] nameSegments = ClauseScrubber.SegmentForPhraseQuery(filteredPhrase);
if (nameSegments.Length == 1)
{
return(WithFields(fieldNameAndBoosts, nameSegments[0], false, occur, asFilter));
}
else
{
BooleanQuery query = new BooleanQuery();
foreach (var fieldNameAndBoost in fieldNameAndBoosts)
{
PhraseQuery phraseQuery = new PhraseQuery();
foreach (var nameSegment in nameSegments)
{
phraseQuery.Add(new Term(fieldNameAndBoost.Key, nameSegment));
}
phraseQuery.SetSlop(PhraseQuerySlop);
SetBoost(phraseQuery, fieldNameAndBoost.Value);
query.Add(phraseQuery, occur);
}
if (asFilter)
{
filters.Add(new BooleanClause(query, BooleanClause.Occur.MUST));
}
else
{
clauses.Add(new BooleanClause(query, BooleanClause.Occur.MUST));
}
return(this);
}
}
private void SearchFromIndexData(string searchkey)
{
string indexPath = Context.Server.MapPath("~/IndexData");
FSDirectory directory = FSDirectory.Open(new DirectoryInfo(indexPath), new NoLockFactory());
IndexReader reader = IndexReader.Open(directory, true);
IndexSearcher searcher = new IndexSearcher(reader);
//搜索条件
PhraseQuery query = new PhraseQuery();
//把用户输入的关键字进行分词
foreach (string word in SplitContent.SplitWords(searchkey))
{
query.Add(new Term("TITLE", word));
}
//query.Add(new Term("content", "C#"));//多个查询条件时 为且的关系
query.SetSlop(100); //指定关键词相隔最大距离
//TopScoreDocCollector盛放查询结果的容器
TopScoreDocCollector collector = TopScoreDocCollector.create(1000, true);
searcher.Search(query, null, collector); //根据query查询条件进行查询，查询结果放入collector容器
//TopDocs 指定0到GetTotalHits() 即所有查询结果中的文档 如果TopDocs(20,10)则意味着获取第20-30之间文档内容 达到分页的效果
ScoreDoc[] docs = collector.TopDocs(0, collector.GetTotalHits()).scoreDocs; //collector.GetTotalHits()
//展示数据实体对象集合
for (int i = 0; i < docs.Length; i++)
{
int docID = docs[i].doc; //得到查询结果文档的ID（Lucene内部分配的ID）
Document doc = searcher.Doc(docID); //根据文档ID来获得文档对象Document
SUC_NEWS mod = new SUC_NEWS();
mod.TITLE = SplitContent.HightLight(searchkey, doc.Get("TITLE"));
mod.TITLE = string.IsNullOrEmpty(mod.TITLE) ? doc.Get("TITLE") : mod.TITLE;
//book.ContentDESCRPTION = doc.Get("content");//未使用高亮
//搜索关键字高亮显示 使用盘古提供高亮插件
mod.CONTENT = SplitContent.HightLight(searchkey, doc.Get("CONTENT"));
mod.CONTENT = string.IsNullOrEmpty(mod.CONTENT) ? doc.Get("CONTENT") : mod.CONTENT;
mod.CONTENT = mod.CONTENT.Replace("<b>", "");
mod.ID = Convert.ToInt32(doc.Get("ID"));
mod.pandaWebUrl = doc.Get("URL");
modResult.Add(mod);
}
}
public void flatten(Query sourceQuery, Dictionary<Query,Query> flatQueries)
{
if (sourceQuery is BooleanQuery)
{
BooleanQuery bq = (BooleanQuery)sourceQuery;
foreach (BooleanClause clause in bq.GetClauses())
{
if (!clause.IsProhibited())
flatten(clause.GetQuery(), flatQueries);
}
}
else if (sourceQuery is DisjunctionMaxQuery)
{
DisjunctionMaxQuery dmq = (DisjunctionMaxQuery)sourceQuery;
System.Collections.IEnumerator en = dmq.Iterator();
while (en.MoveNext())
{
Query query = (Query)en.Current;
flatten(query, flatQueries);
}
}
else if (sourceQuery is TermQuery)
{
if (!flatQueries.ContainsKey(sourceQuery))
flatQueries.Add(sourceQuery, sourceQuery);
}
else if (sourceQuery is PhraseQuery)
{
if (!flatQueries.ContainsKey(sourceQuery))
{
PhraseQuery pq = (PhraseQuery)sourceQuery;
if (pq.GetTerms().Length > 1)
flatQueries.Add(pq, pq);
else if (pq.GetTerms().Length == 1)
{
Query q = new TermQuery(pq.GetTerms()[0]);
flatQueries.Add(q, q);
}
}
}
// else discard queries
}
//搜索包含关键词
public static List <JobSerach> SearchContent(string kw, int index, int skipCount)
{
//string indexPath = lucenePath;//最好将该项放在配置文件中。
kw = kw.ToLower();
FSDirectory directory = FSDirectory.Open(new DirectoryInfo(lucenePath), new NoLockFactory());
IndexReader reader = IndexReader.Open(directory, true);
IndexSearcher searcher = new IndexSearcher(reader);
//搜索条件
PhraseQuery query = new PhraseQuery();
foreach (string word in SplitWord(kw))//将用户输入的搜索内容进行了盘古分词、
{
query.Add(new Term("Title", word));
//query.Add(new Term("Content", word));
//query.Add(new Term("MaiDian", word));
}
query.SetSlop(100);//多个查询条件的词之间的最大距离.在文章中相隔太远 也就无意义.（例如 “大学生”这个查询条件和"简历"这个查询条件之间如果间隔的词太多也就没有意义了。）
//TopScoreDocCollector是盛放查询结果的容器
TopScoreDocCollector collector = TopScoreDocCollector.create(1000, true);
searcher.Search(query, null, collector); //根据query查询条件进行查询，查询结果放入collector容器
ScoreDoc[] docs = collector.TopDocs(index - 1, skipCount).scoreDocs; //得到所有查询结果中的文档,GetTotalHits():表示总条数 TopDocs(300, 20);//表示得到300（从300开始），到320（结束）的文档内容. //可以用来实现分页功能
List <JobSerach> list = new List <JobSerach>();
for (int i = 0; i < docs.Length; i++)
{
//
//搜索ScoreDoc[]只能获得文档的id,这样不会把查询结果的Document一次性加载到内存中。降低了内存压力，需要获得文档的详细内容的时候通过searcher.Doc来根据文档id来获得文档的详细内容对象Document.
int docId = docs[i].doc; //得到查询结果文档的id（Lucene内部分配的id）
Document doc = searcher.Doc(docId); //找到文档id对应的文档详细信息
JobSerach result = new JobSerach();
result.Title = Highlight(kw, doc.Get("Title"));
result.Id = Convert.ToInt32(doc.Get("Id"));
result.ImageAddress = doc.Get("ImageAddress");
result.MaiDian = doc.Get("MaiDian");
result.Price = double.Parse(doc.Get("Price"));
result.Content = doc.Get("Content");
list.Add(result);
}
return(list);
}
private Query clause(String field, float boost, params String[] terms)
{
Query q;
if (terms.Length == 1)
{
q = new TermQuery(new Term(field, terms[0]));
}
else
{
PhraseQuery pq = new PhraseQuery();
foreach (String term in terms)
{
pq.Add(new Term(field, term));
}
q = pq;
}
q.Boost = (boost);
return(q);
}
public void TestPhraseQuery()
{
SimpleFragListBuilder sflb = new SimpleFragListBuilder();
PhraseQuery phraseQuery = new PhraseQuery();
phraseQuery.Add(new Term(F, "a"));
phraseQuery.Add(new Term(F, "b"));
FieldFragList ffl = sflb.CreateFieldFragList(fpl(phraseQuery, "c d e"), 20);
assertEquals(0, ffl.FragInfos.size());
ffl = sflb.CreateFieldFragList(fpl(phraseQuery, "a c b"), 20);
assertEquals(0, ffl.FragInfos.size());
ffl = sflb.CreateFieldFragList(fpl(phraseQuery, "a b c"), 20);
assertEquals(1, ffl.FragInfos.size());
assertEquals("subInfos=(ab((0,3)))/1.0(0,20)", ffl.FragInfos[0].toString());
}
public virtual void TestWithPendingDeletes3()
{
// main directory
Directory dir = NewDirectory();
// auxiliary directory
Directory aux = NewDirectory();
SetUpDirs(dir, aux);
IndexWriter writer = NewWriter(dir, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random())).SetOpenMode(OpenMode_e.APPEND));
// Adds 10 docs, then replaces them with another 10
// docs, so 10 pending deletes:
for (int i = 0; i < 20; i++)
{
Document doc = new Document();
doc.Add(NewStringField("id", "" + (i % 10), Field.Store.NO));
doc.Add(NewTextField("content", "bbb " + i, Field.Store.NO));
writer.UpdateDocument(new Term("id", "" + (i % 10)), doc);
}
// Deletes one of the 10 added docs, leaving 9:
PhraseQuery q = new PhraseQuery();
q.Add(new Term("content", "bbb"));
q.Add(new Term("content", "14"));
writer.DeleteDocuments(q);
writer.AddIndexes(aux);
writer.ForceMerge(1);
writer.Commit();
VerifyNumDocs(dir, 1039);
VerifyTermDocs(dir, new Term("content", "aaa"), 1030);
VerifyTermDocs(dir, new Term("content", "bbb"), 9);
writer.Dispose();
dir.Dispose();
aux.Dispose();
}
public void TestGetBestFragmentsFilteredPhraseQuery()
{
var helper = new TestHighlightRunner();
helper.TestAction = () =>
{
numHighlights = 0;
var rf = new TermRangeFilter("contents", "john", "john", true, true);
var pq = new PhraseQuery();
pq.Add(new Term("contents", "john"));
pq.Add(new Term("contents", "kennedy"));
var fq = new FilteredQuery(pq, rf);
DoSearching(fq);
helper.DoStandardHighlights(analyzer, searcher, hits, query, this);
// Currently highlights "John" and "Kennedy" separately
Assert.IsTrue(numHighlights == 2,
"Failed to find correct number of highlights " + numHighlights +
" found");
};
helper.Start();
}
protected internal virtual void SmokeTestSearcher(IndexSearcher s)
{
RunQuery(s, new TermQuery(new Term("body", "united")));
RunQuery(s, new TermQuery(new Term("titleTokenized", "states")));
PhraseQuery pq = new PhraseQuery();
pq.Add(new Term("body", "united"));
pq.Add(new Term("body", "states"));
RunQuery(s, pq);
}
public virtual void TestPositionIncrementMultiFields()
{
Directory dir = NewDirectory();
RandomIndexWriter writer = new RandomIndexWriter(Random(), dir, Similarity, TimeZone);
writer.AddDocument(MakeDocumentWithFields());
IndexReader reader = writer.Reader;
IndexSearcher searcher = NewSearcher(reader);
PhraseQuery query = new PhraseQuery();
query.Add(new Term("indexed_not_tokenized", "test1"));
query.Add(new Term("indexed_not_tokenized", "test2"));
ScoreDoc[] hits = searcher.Search(query, null, 1000).ScoreDocs;
Assert.AreEqual(1, hits.Length);
DoAssert(searcher.Doc(hits[0].Doc), true);
writer.Dispose();
reader.Dispose();
dir.Dispose();
}
public virtual void TestCJKSloppyPhrase()
{
// individual CJK chars as terms
SimpleCJKAnalyzer analyzer = new SimpleCJKAnalyzer(this);
PhraseQuery expected = new PhraseQuery();
expected.Slop = 3;
expected.Add(new Term("field", "中"));
expected.Add(new Term("field", "国"));
QueryBuilder builder = new QueryBuilder(analyzer);
Assert.AreEqual(expected, builder.CreatePhraseQuery("field", "中国", 3));
}
public virtual void TestPhraseQueryPositionIncrements()
{
PhraseQuery expected = new PhraseQuery();
expected.Add(new Term("field", "1"));
expected.Add(new Term("field", "2"), 2);
CharacterRunAutomaton stopList = new CharacterRunAutomaton((new RegExp("[sS][tT][oO][pP]")).ToAutomaton());
Analyzer analyzer = new MockAnalyzer(Random(), MockTokenizer.WHITESPACE, false, stopList);
QueryBuilder builder = new QueryBuilder(analyzer);
Assert.AreEqual(expected, builder.CreatePhraseQuery("field", "1 stop 2"));
}
public virtual void TestPhrase()
{
PhraseQuery query = new PhraseQuery();
query.Add(new Term("field", "seventy"));
query.Add(new Term("field", "seven"));
CheckHits(query, new int[] { 77, 177, 277, 377, 477, 577, 677, 777, 877, 977, 1077, 1177, 1277, 1377, 1477, 1577, 1677, 1777, 1877, 1977 });
}
public virtual void TestPhrase2()
{
PhraseQuery query = new PhraseQuery();
query.Add(new Term("field", "seventish"));
query.Add(new Term("field", "sevenon"));
CheckHits(query, new int[] { });
}
/*
* This shows how to construct a phrase query containing shingles.
*/
//JAVA TO C# CONVERTER WARNING: Method 'throws' clauses are not available in .NET:
//ORIGINAL LINE: public void testShingleAnalyzerWrapperPhraseQuery() throws Exception
public virtual void testShingleAnalyzerWrapperPhraseQuery()
{
PhraseQuery q = new PhraseQuery();
TokenStream ts = analyzer.tokenStream("content", "this sentence");
try
{
int j = -1;
PositionIncrementAttribute posIncrAtt = ts.addAttribute(typeof(PositionIncrementAttribute));
CharTermAttribute termAtt = ts.addAttribute(typeof(CharTermAttribute));
ts.reset();
while (ts.incrementToken())
{
j += posIncrAtt.PositionIncrement;
string termText = termAtt.ToString();
q.add(new Term("content", termText), j);
}
ts.end();
}
finally
{
IOUtils.closeWhileHandlingException(ts);
}
ScoreDoc[] hits = searcher.search(q, null, 1000).scoreDocs;
int[] ranks = new int[] {0};
compareRanks(hits, ranks);
}
/// <summary>
/// Adds a standard type clause to this instance
/// </summary>
/// <param name="term">Term to add to this query.</param>
/// <param name="occurrence">Defines how the term is added to this query.</param>
/// <param name="slop">The amount of allowed slop in a phrase query.</param>
/// <remarks>
/// Slop is the amount of movement each word is allowed in a non-exact phrase query.
/// For instance if you search for "Adobe Systems Incorporated" and the slop is set to 0 then
/// only results with that term is allowed. If you set the slop to 2 then two movements can be
/// made, max, for each word. In the same example with slop set to 2 results would be returned
/// for "Adobe Systems Incorporated", "Adobe Incorporated Systems", "Systems Adobe Incorporated",
/// and "Systems Incorporated Adobe".
/// </remarks>
public void AddBooleanClause(SearchTerm term, ClauseOccurrence occurrence, int slop)
{
if (term == null)
throw new ArgumentNullException("term", "term cannot be null");
IncrementTotalClauses(1);
if (term.IsPhrase) {
PhraseQuery phraseQuery = new PhraseQuery();
phraseQuery.Add(term.GetLuceneTerm());
phraseQuery.SetSlop(slop);
phraseQuery.SetBoost(term.Boost);
this.luceneQuery.Add(phraseQuery, TypeConverter.ConvertToLuceneClauseOccurrence(occurrence));
phraseQuery = null;
}
else {
TermQuery termQuery = new TermQuery(term.GetLuceneTerm());
termQuery.SetBoost(term.Boost);
this.luceneQuery.Add(termQuery, TypeConverter.ConvertToLuceneClauseOccurrence(occurrence));
termQuery = null;
}
}