public void TestReusableTokenStream()
{
Analyzer analyzer = new CzechAnalyzer(Version.LUCENE_CURRENT);
AssertAnalyzesToReuse(analyzer, "Pokud mluvime o volnem", new String[] { "mluvime", "volnem" });
AssertAnalyzesToReuse(analyzer, "Česká Republika", new String[] { "česká", "republika" });
}
public void TestInvalidStopWordFile()
{
CzechAnalyzer cz = new CzechAnalyzer(Version.LUCENE_CURRENT);
cz.LoadStopWords(new UnreliableInputStream(), Encoding.UTF8);
AssertAnalyzesTo(cz, "Pokud mluvime o volnem",
new String[] { "pokud", "mluvime", "o", "volnem" });
}
public void TestStopWordFileReuse()
{
CzechAnalyzer cz = new CzechAnalyzer(Version.LUCENE_CURRENT);
AssertAnalyzesToReuse(cz, "Česká Republika",
new String[] { "česká", "republika" });
Stream stopwords = new FileStream(customStopFile, FileMode.Open, FileAccess.Read);
cz.LoadStopWords(stopwords, Encoding.UTF8);
AssertAnalyzesToReuse(cz, "Česká Republika", new String[] { "česká" });
}
private Analyzer GuessAnalyzer(string filePath)
{
Analyzer ret = null;
switch (Path.GetFileName(filePath).Substring(0, 2).ToLowerInvariant())
{
case "zh":
ret = new ChineseAnalyzer();
break;
case "cs":
ret = new CzechAnalyzer();
break;
case "da":
ret = new SnowballAnalyzer("Danish");
break;
case "nl":
ret = new SnowballAnalyzer("Dutch");
break;
case "en":
ret = new SnowballAnalyzer("English");
break;
case "fi":
ret = new SnowballAnalyzer("Finnish");
break;
case "fr":
ret = new SnowballAnalyzer("French");
break;
case "de":
ret = new SnowballAnalyzer("German");
break;
case "it":
ret = new SnowballAnalyzer("Italian");
break;
case "ja":
ret = new CJKAnalyzer();
break;
case "ko":
ret = new CJKAnalyzer();
break;
case "no":
ret = new SnowballAnalyzer("Norwegian");
break;
case "pt":
ret = new SnowballAnalyzer("Portuguese");
break;
case "ru":
ret = new SnowballAnalyzer("Russian");
break;
case "es":
ret = new SnowballAnalyzer("Spanish");
break;
case "se":
ret = new SnowballAnalyzer("Swedish");
break;
default:
ret = new StandardAnalyzer();
break;
}
return(ret);
}
private Analyzer GuessAnalyzer(string filePath, out bool isRTL)
{
Analyzer ret = null;
isRTL = false;
switch (Path.GetFileName(filePath).Substring(0, 2).ToLowerInvariant())
{
case "zh":
ret = new ChineseAnalyzer();
break;
case "cs":
ret = new CzechAnalyzer();
break;
case "da":
ret = new SnowballAnalyzer("Danish");
break;
case "nl":
ret = new SnowballAnalyzer("Dutch");
break;
case "en":
ret = new SnowballAnalyzer("English");
break;
case "fi":
ret = new SnowballAnalyzer("Finnish");
break;
case "fr":
ret = new SnowballAnalyzer("French");
break;
case "de":
ret = new SnowballAnalyzer("German");
break;
case "it":
ret = new SnowballAnalyzer("Italian");
break;
case "ja":
ret = new CJKAnalyzer();
break;
case "ko":
ret = new CJKAnalyzer();
break;
case "no":
ret = new SnowballAnalyzer("Norwegian");
break;
case "pt":
ret = new SnowballAnalyzer("Portuguese");
break;
case "ru":
ret = new SnowballAnalyzer("Russian");
break;
case "es":
ret = new SnowballAnalyzer("Spanish");
break;
case "se":
ret = new SnowballAnalyzer("Swedish");
break;
case "ar":
isRTL = true;
// TODO: Lucene 2.9 has a light stemmer for Arabic providing good search results
ret = new StandardAnalyzer(Lucene.Net.Util.Version.LUCENE_29);
break;
case "he":
{
isRTL = true;
string hspellPath = System.Configuration.ConfigurationManager.AppSettings["hspellPath"];
if (!string.IsNullOrEmpty(hspellPath) && Directory.Exists(hspellPath))
{
try
{
ret = new Lucene.Net.Analysis.Hebrew.MorphAnalyzer(hspellPath);
break;
}
catch
{
}
}
ret = new Lucene.Net.Analysis.Hebrew.SimpleAnalyzer();
break;
}
default:
ret = new StandardAnalyzer(Lucene.Net.Util.Version.LUCENE_29);
break;
}
return(ret);
}
