/// <summary>
/// Perform synonym expansion on a query.
/// </summary>
/// <param name="query">users query that is assumed to not have any "special" query syntax, thus it should be just normal words, so "big dog" makes sense, but a query like "title:foo^1.2" doesn't as this should presumably be passed directly to the default query parser </param>
/// <param name="syns">a opened to the Lucene index you previously created with <see cref="Syns2Index"/>. The searcher is not closed or otherwise altered. </param>
/// <param name="a">optional analyzer used to parse the users query else <see cref="StandardAnalyzer"/> is used </param>
/// <param name="field">optional field name to search in or null if you want the default of "contents" </param>
/// <param name="boost">optional boost applied to synonyms else no boost is applied </param>
/// <returns>the expanded Query </returns>
public static Query Expand(String query,
Searcher syns,
Analyzer a,
String field,
float boost)
{
already = new List <String>(); // avoid dups
var top = new List <String>(); // needs to be separately listed..
if (field == null)
{
field = "contents";
}
if (a == null)
{
a = new StandardAnalyzer(Lucene.Net.Util.Version.LUCENE_CURRENT);
}
// [1] Parse query into separate words so that when we expand we can avoid dups
var ts = a.TokenStream(field, new StringReader(query));
var termAtt = ts.AddAttribute <TermAttribute>();
while (ts.IncrementToken())
{
var word = termAtt.Term;
if (!already.Contains(word))
{
already.Add(word);
top.Add(word);
}
}
tmp = new BooleanQuery();
// [2] form query
System.Collections.IEnumerator it = top.GetEnumerator();
while (it.MoveNext())
{
// [2a] add to level words in
var word = (String)it.Current;
var tq = new TermQuery(new Term(field, word));
tmp.Add(tq, Occur.SHOULD);
var c = new CollectorImpl(field, boost);
syns.Search(new TermQuery(new Term(Syns2Index.F_WORD, word)), c);
}
return(tmp);
}
/// <summary>
/// Perform synonym expansion on a query.
/// </summary>
/// <param name="query">query</param>
/// <param name="syns">syns</param>
/// <param name="a">a</param>
/// <param name="field">field</param>
/// <param name="boost">boost</param>
public static Query Expand(String query,
Searcher syns,
Analyzer a,
String field,
float boost)
{
already = new List <String>(); // avoid dups
var top = new List <String>(); // needs to be separately listed..
var ts = a.TokenStream(field, new StringReader(query));
var termAtt = ts.AddAttribute <TermAttribute>();
while (ts.IncrementToken())
{
var word = termAtt.Term;
if (!already.Contains(word))
{
already.Add(word);
top.Add(word);
}
}
tmp = new BooleanQuery();
// [2] form query
System.Collections.IEnumerator it = top.GetEnumerator();
while (it.MoveNext())
{
// [2a] add to level words in
var word = (String)it.Current;
var tq = new TermQuery(new Term(field, word));
tmp.Add(tq, Occur.SHOULD);
var c = new CollectorImpl(field, boost);
syns.Search(new TermQuery(new Term(Syns2Index.F_WORD, word)), c);
}
return(tmp);
}
/// <summary>
/// Perform synonym expansion on a query.
/// </summary>
/// <param name="query">query</param>
/// <param name="syns">syns</param>
/// <param name="a">a</param>
/// <param name="field">field</param>
/// <param name="boost">boost</param>
public static Query Expand(String query,
Searcher syns,
Analyzer a,
String field,
float boost)
{
already = new List<String>(); // avoid dups
var top = new List<String>(); // needs to be separately listed..
var ts = a.TokenStream(field, new StringReader(query));
var termAtt = ts.AddAttribute<TermAttribute>();
while (ts.IncrementToken())
{
var word = termAtt.Term;
if (!already.Contains(word))
{
already.Add(word);
top.Add(word);
}
}
tmp = new BooleanQuery();
// [2] form query
System.Collections.IEnumerator it = top.GetEnumerator();
while (it.MoveNext())
{
// [2a] add to level words in
var word = (String)it.Current;
var tq = new TermQuery(new Term(field, word));
tmp.Add(tq, Occur.SHOULD);
var c = new CollectorImpl(field, boost);
syns.Search(new TermQuery(new Term(Syns2Index.F_WORD, word)), c);
}
return tmp;
}
