.net lucene 实战搜索（三）----- 基本之搜索
详细请参阅：http://lucene.apache.org/java/docs/queryparsersyntax.html 这里不一一详细说明。
查询参数类别
public enum QueryItemType
{
And = 0,
Or = 1,
Range = 2,
Fuzzy = 3,
Wildcard = 4,
TimeRange = 5
}
查询条件参数
[Serializable]
public class QueryItem
{
private string fieldText = string.Empty;
private string fieldName = string.Empty;
private string value = string.Empty;
private string value_1 = string.Empty;
private QueryItemType type = QueryItemType.And;
string[] specialcharacters = new string[] { "+", "-", "&&", "||", "!", "(", ")", "{", "}", "[", "]", "^", "\"", "~", "*", "?", " :", "\\" };
private bool isToLower = true;
private bool isTextQuery = false;
public QueryItem()
{
}
public QueryItem(string sfieldName, string svalue, string svalue_1, QueryItemType stype)
{
fieldName = sfieldName;
Value = svalue;
Value_1 = svalue_1;
type = stype;
}
public QueryItem(string ftext, string sfieldName, string svalue, string svalue_1, QueryItemType stype)
{
this.fieldText = ftext;
fieldName = sfieldName;
Value = svalue;
Value_1 = svalue_1;
type = stype;
}
public string FieldName
{
get
{
return fieldName;
}
set
{
fieldName = value;
}
}
public string Value
{
get
{
return this.value;
}
set
{
string temp = value;
foreach (string str in specialcharacters)
{
temp.Replace(str, " ");
}
this.value = temp;
}
}
public string Value_1
{
get
{
return value_1;
}
set
{
string temp = value;
foreach (string str in specialcharacters)
{
temp.Replace(str, " ");
}
value_1 = temp;
}
}
public QueryItemType Type
{
get
{
return type;
}
set
{
type = value;
}
}
public string FieldText
{
get
{
return fieldText;
}
set
{
fieldText = value;
}
}
public bool IsToLower
{
get { return isToLower; }
set { isToLower = value; }
}
public bool IsTextQuery
{
get { return isTextQuery; }
set { isTextQuery = value; }
}
}
搜索参数
[Serializable]
public class SearchParameter
{
string[] stopwords = new string[] {"[","]","+","-","or","and","to",":","~","*",
"!","@","#","$","%","^","&","*","(",")","|","?","_","\\","//","'","\"",";","{","}","=","<",">"};
private int _pagesize = 100;
private int _pageindex = 1;
private List<QueryItem> queryitem = new List<QueryItem>();
private Hashtable _keywordfilter = new Hashtable();
private AnalyzerEnum _analyzertype;
private string _indexdir;
private List<SortItem> _sortItems;
public string IndexDir
{
get { return _indexdir; }
set { _indexdir = value; }
}
public int PageSize
{
get { return _pagesize; }
set { _pagesize = value; }
}
public int PageIndex
{
get { return _pageindex; }
set { _pageindex = value; }
}
public List<QueryItem> QueryItems
{
get { return queryitem; }
set
{
foreach (QueryItem item in value)
{
item.Value = ReplaceStr(item.Value);
}
queryitem = value;
}
}
public Hashtable KeywordFilter
{
get { return _keywordfilter; }
}
public void AddKeywordFilter(string QueryField, string QueryText)
{
_keywordfilter.Add(QueryField, QueryText);
}
public List<SortItem> SortItems
{
set
{
_sortItems = value;
}
}
public SortField[] SortFields
{
get
{
if (_sortItems == null) return null;
SortField[] sortfield = new SortField[_sortItems.Count];
for (int i = 0; i < _sortItems.Count; i++)
{
SortField sortitem = new SortField(ReplaceStr(_sortItems[i].FieldName.ToUpper()), _sortItems[i].ASC);
sortfield[i] = sortitem;
}
return sortfield;
}
}
public void RemoveKeywordFilter(string QueryField)
{
_keywordfilter.Remove(QueryField);
}
public AnalyzerEnum AnalyzerType
{
set { _analyzertype = value; }
get { return _analyzertype; }
}
public string GetQueryText()
{
if (queryitem.Count > 0)
{
foreach (QueryItem item in queryitem)
{
if (item.IsTextQuery)
{
return item.Value;
}
}
}
return string.Empty;
}
private string ReplaceStr(string source)
{
for (int i = 0; i < stopwords.Length; i++)
{
source = source.Replace(stopwords[i], " ");
}
return source;
}
}
[Serializable]
public class SortItem
{
//是否降序。true降序，false升序
public SortItem()
{
}
//是否降序。true降序，false升序
public SortItem(string fieldName, bool asc)
{
FieldName = fieldName;
ASC = asc;
}
public string FieldName = string.Empty;
public bool ASC = true;
}
以上代码说明：查询条件和数据库一样，有 or and not 等条件，QueryItemType就是用来设置条件的类型。
QueryItem是单个条件，可用来实现多个条件查询。SortItem是排序使用的，当然也可以多条件排序。注意默认的情况下luncene是以匹配最多的排序的。
这里只是对单索引进行查询，事实上，luncene是支持多索引分布式查询的。
//搜索器
public class Searcher
{
public DataTable SearchData(SearchParameter parm, string fields, out int count)
{
return SearchData(parm, fields, "", out count);
}
public DataTable SearchData(SearchParameter parm, string fields, string lightfields, out int count)
{
DataTable dt = new DataTable();
Hits hits = null;
IndexSearcher searcher = null;
fields = fields.ToUpper();
lightfields = lightfields.ToUpper();
count = 0;
try
{
searcher = new IndexSearcher(parm.IndexDir);
Query query = Formater.FormatSearchItem(parm);
if (parm.SortFields != null)
{
Sort sort = new Sort(parm.SortFields);
hits = searcher.Search(query, sort);
}
else
{
hits = searcher.Search(query);
}
if (hits == null || hits.Length() <= 0) return null;
int start = (parm.PageIndex - 1) * parm.PageSize;
int end = (parm.PageIndex) * parm.PageSize;
if (hits.Length() <= end)
{
end = hits.Length();
}
count = hits.Length();
DataColumn dc;
string[] fieldcolumns = fields.Split(',');
//创建表格
foreach (string field in fieldcolumns)
{
dc = new DataColumn();
dc.DataType = Type.GetType("System.String");
dc.ColumnName = field;
dt.Columns.Add(dc);
}
//将数据倒入表
if (lightfields.Length > 0)
{
string text = parm.GetQueryText();
List<string> tlist = new List<string>();
StringReader sr = new StringReader(text);
TokenStream ts = Formater.GetAnalyzer(parm.AnalyzerType).TokenStream("", sr);
Token token;
while ((token = ts.Next()) != null)
{
tlist.Add(token.TermText());
}
sr.Close();
for (int i = start; i < end; i++)
{
DataRow dr = dt.NewRow();
foreach (string field in fieldcolumns)
{
string temp = hits.Doc(i).Get(field);
if (text.Length > 0)
{
string[] lightfield = lightfields.Split(',');
foreach (string lfd in lightfield)
{
if (field == lfd)
{
foreach (string str in tlist)
{
temp = temp.Replace(str, string.Format("<font color='red'><b>{0}</b></font>", str));
}
}
}
}
dr[field] = temp;
}
dt.Rows.Add(dr);
}
}
else
{
for (int i = start; i < end; i++)
{
DataRow dr = dt.NewRow();
foreach (string field in fieldcolumns)
{
string temp = hits.Doc(i).Get(field);
dr[field] = temp;
}
dt.Rows.Add(dr);
}
}
}
catch (Exception err)
{
;
}
finally
{
if (searcher != null)
{
searcher.Close();
}
}
return dt;
}
public DataTable SearchDataAllField(SearchParameter parm, string lightfield, out int count)
{
DataTable dt = new DataTable();
Hits hits = null;
IndexSearcher searcher = null;
lightfield = lightfield.ToUpper();
count = 0;
try
{
searcher = new IndexSearcher(parm.IndexDir);
Query query = Formater.FormatSearchItem(parm);
if (parm.SortFields != null)
{
Sort sort = new Sort(parm.SortFields);
hits = searcher.Search(query, sort);
}
else
{
hits = searcher.Search(query);
}
if (hits == null || hits.Length() <= 0) return null;
int start = (parm.PageIndex - 1) * parm.PageSize;
int end = (parm.PageIndex) * parm.PageSize;
if (hits.Length() <= end)
{
end = hits.Length();
}
count = hits.Length();
DataColumn dc;
//创建表格
foreach (Field field in hits.Doc(0).Fields())
{
dc = new DataColumn();
dc.DataType = Type.GetType("System.String");
dc.ColumnName = field.Name();
dt.Columns.Add(dc);
}
//将数据倒入表
for (int i = start; i < end; i++)
{
DataRow dr = dt.NewRow();
foreach (Field field in hits.Doc(i).Fields())
{
string temp = field.StringValue();
foreach (QueryItem item in parm.QueryItems)
{
if (lightfield.Length > 0 && field.Name() == lightfield)
{
temp = temp.Replace(item.Value, string.Format("<font color=orangered><b>{0}</b></font>", item.Value));
break;
}
}
dr[field.Name()] = temp;
}
dt.Rows.Add(dr);
}
}
catch (Exception err)
{
;
}
finally
{
if (searcher != null)
{
searcher.Close();
}
}
return dt;
}
public DataTable SearchDataDistinct(SearchParameter parm, string fieldName, int count)
{
DataTable dt = new DataTable();
Hits hits = null;
IndexSearcher searcher = null;
fieldName = fieldName.ToUpper();
int ccount = 0;
try
{
searcher = new IndexSearcher(parm.IndexDir);
Query query = Formater.FormatSearchItem(parm);
if (parm.SortFields != null)
{
Sort sort = new Sort(parm.SortFields);
hits = searcher.Search(query, sort);
}
else
{
hits = searcher.Search(query);
}
if (hits == null || hits.Length() <= 0) return null;
DataColumn dc;
//创建表格
dc = new DataColumn();
dc.DataType = Type.GetType("System.String");
dc.ColumnName = fieldName;
dt.Columns.Add(dc);
//将数据倒入表
string temp = string.Empty;
for (int i = 0; i < hits.Length(); i++)
{
if (ccount >= count) break;
string currstring = hits.Doc(i).GetField(fieldName).StringValue();
if (currstring.Trim().Length <= 0) continue;
if (currstring != temp)
{
DataRow dr = dt.NewRow();
dr[fieldName] = hits.Doc(i).GetField(fieldName).StringValue();
dt.Rows.Add(dr);
ccount++;
temp = currstring;
}
}
}
catch (Exception err)
{
;
}
finally
{
if (searcher != null)
{
searcher.Close();
}
}
return dt;
}
public int SearchCount(SearchParameter parm)
{
Hits hits = null;
IndexSearcher searcher = null;
int count = 0;
try
{
searcher = new IndexSearcher(parm.IndexDir);
Query query = Formater.FormatSearchItem(parm);
hits = searcher.Search(query);
count = hits.Length();
return count;
}
catch
{
return 0;
}
finally
{
if (searcher != null)
{
searcher.Close();
}
}
}
public int GetDocID(SearchParameter parm)
{
Hits hits = null;
IndexSearcher searcher = new IndexSearcher(parm.IndexDir);
try
{
Query query = Formater.FormatSearchItem(parm);
if (parm.SortFields != null)
{
Sort sort = new Sort(parm.SortFields);
hits = searcher.Search(query, sort);
}
else
{
hits = searcher.Search(query);
}
if (hits == null || hits.Length() <= 0) return 0;
if (hits.Length() > 0)
{
return hits.Id(0);
}
return 0;
}
finally
{
if (searcher != null)
{
searcher.Close();
}
}
}
public List<int> GetDocIDs(SearchParameter parm)
{
Hits hits = null;
List<int> slist = new List<int>();
IndexSearcher searcher = new IndexSearcher(parm.IndexDir);
try
{
Query query = Formater.FormatSearchItem(parm);
if (parm.SortFields != null)
{
Sort sort = new Sort(parm.SortFields);
hits = searcher.Search(query, sort);
}
else
{
hits = searcher.Search(query);
}
if (hits == null || hits.Length() <= 0) return slist;
for (int i = 0; i < hits.Length(); i++)
{
slist.Add(hits.Id(i));
}
return slist;
}
finally
{
if (searcher != null)
{
searcher.Close();
}
}
}
}
以上是对luncene搜索的封装，可分页查询数据，针对单索引。
原来luncene有个高亮命中词的组建，不知道是不是我比较愚钝，一直没明白怎么用，而且跟版本捆绑太紧密，所以我将搜索结果通过分词替换掉。
下面是应用举例：
SearchParameter sp = new SearchParameter();
sp.AnalyzerType = AnalyzerEnum.SmartSegmentAnalyzer;
sp.IndexDir = UICommon.IndexDir;
sp.PageIndex = 1;
QueryItem item;
List<QueryItem> list;
Searcher search = new Searcher();
DataTable mdt;
item = new QueryItem();
item.FieldName = "artitype";
item.Value = ((int)SDataType.Course).ToString();
list.Add(item);
sp.QueryItems = list;
sp.PageSize = 3;
try
{
mdt = search.SearchData(sp, "courseid,coursename", out ccount);
}
catch
{
sp.IndexDir = UICommon.TempIndexDir;
mdt = new Searcher().SearchData(sp, "courseid,coursename", out ccount);
}
以表的形式返回，而lucene是以docuemnt的方式返回的。
以上就是luncene的基本使用，索引----〉搜索。
浙公网安备 33010602011771号