/**
* This code is similar in style to code found in Solr:
* solr/core/src/java/org/apache/solr/analysis/BeiderMorseFilterFactory.java
*
* Making a JUnit test out of it to protect Solr from possible future
* regressions in Commons-Codec.
*/
private static String Encode(IDictionary <String, String> args, bool concat, String input)
{
LanguageSet languageSet;
PhoneticEngine engine;
// PhoneticEngine = NameType + RuleType + concat
// we use common-codec's defaults: GENERIC + APPROX + true
String nameTypeArg;
args.TryGetValue("nameType", out nameTypeArg);
NameType nameType = (nameTypeArg == null) ? NameType.GENERIC : (NameType)Enum.Parse(typeof(NameType), nameTypeArg, true);
String ruleTypeArg;
args.TryGetValue("ruleType", out ruleTypeArg);
RuleType ruleType = (ruleTypeArg == null) ? RuleType.APPROX : (RuleType)Enum.Parse(typeof(RuleType), ruleTypeArg, true);
engine = new PhoneticEngine(nameType, ruleType, concat);
// LanguageSet: defaults to automagic, otherwise a comma-separated list.
String languageSetArg;
args.TryGetValue("languageSet", out languageSetArg);
if (languageSetArg == null || languageSetArg.equals("auto"))
{
languageSet = null;
}
else
{
languageSet = LanguageSet.From(new HashSet <String>(Arrays.AsList(languageSetArg.Split(',').TrimEnd())));
}
/*
* org/apache/lucene/analysis/phonetic/BeiderMorseFilter.java (lines 96-98) does this:
*
* encoded = (languages == null)
* ? engine.encode(termAtt.toString())
* : engine.encode(termAtt.toString(), languages);
*
* Hence our approach, below:
*/
if (languageSet == null)
{
return(engine.Encode(input));
}
else
{
return(engine.Encode(input, languageSet));
}
}
public void TestEncode(String name, String phoneticExpected, NameType nameType,
RuleType ruleType, bool concat, int maxPhonemes)
{
PhoneticEngine engine = new PhoneticEngine(nameType, ruleType, concat, maxPhonemes);
String phoneticActual = engine.Encode(name);
//System.err.println("expecting: " + this.phoneticExpected);
//System.err.println("actual: " + phoneticActual);
Assert.AreEqual(phoneticExpected, phoneticActual, "phoneme incorrect");
if (concat)
{
String[] split = new Regex("\\|").Split(phoneticActual).TrimEnd();
Assert.True(split.Length <= maxPhonemes);
}
else
{
String[] words = phoneticActual.Split('-').TrimEnd();
foreach (String word in words)
{
String[] split = new Regex("\\|").Split(word).TrimEnd();
Assert.True(split.Length <= maxPhonemes);
}
}
}
public void Test()
{
PhoneticEngine engine = new PhoneticEngine(NameType.GENERIC, RuleType.APPROX, true);
String input = "Angelo";
long startMillis = DateTime.UtcNow.Ticks;
for (int i = 0; i < LOOP; i++)
{
engine.Encode(input);
}
long totalMillis = DateTime.UtcNow.Ticks - startMillis;
Console.WriteLine(String.Format("Time for encoding {0} times the input '{1}': {2} millis.", LOOP, input, totalMillis));
}
public void Test()
{
PhoneticEngine engine = new PhoneticEngine(NameType.GENERIC, RuleType.APPROX, true);
string input = "Angelo";
var sw = new Stopwatch();
sw.Start();
for (int i = 0; i < LOOP; i++)
{
engine.Encode(input);
}
sw.Stop();
long totalMillis = sw.ElapsedMilliseconds;
Console.WriteLine(string.Format("Time for encoding {0} times the input '{1}': {2} millis.", LOOP, input, totalMillis));
}