LUCENENET specific class to patch the behavior of the ICU BreakIterator. Corrects the breaking of words by finding transitions between Thai and non-Thai characters. This logic assumes that the Java BreakIterator also breaks up Thai numerals from Arabic numerals (1, 2, 3, etc.). That is, it assumes the first test below passes and the second test fails in Lucene (not attempted). ThaiAnalyzer analyzer = new ThaiAnalyzer(TEST_VERSION_CURRENT, CharArraySet.EMPTY_SET); AssertAnalyzesTo(analyzer, "๑๒๓456", new string[] { "๑๒๓", "456" }); AssertAnalyzesTo(analyzer, "๑๒๓456", new string[] { "๑๒๓456" });