public void TestRoundTrip()
{
String[] HANGUL = { "\uAC03\uC2F8",
"\uC544\uC5B4" };
Transliterator latinJamo = Transliterator.GetInstance("Latin-Jamo");
Transliterator jamoLatin = latinJamo.GetInverse();
Transliterator jamoHangul = Transliterator.GetInstance("NFC");
Transliterator hangulJamo = Transliterator.GetInstance("NFD");
StringBuffer buf = new StringBuffer();
for (int i = 0; i < HANGUL.Length; ++i)
{
String hangul = HANGUL[i];
String jamo = hangulJamo.Transliterate(hangul);
String latin = jamoLatin.Transliterate(jamo);
String jamo2 = latinJamo.Transliterate(latin);
String hangul2 = jamoHangul.Transliterate(jamo2);
buf.Length = (0);
buf.Append(hangul + " => " +
jamoToName(jamo) + " => " +
latin + " => " + jamoToName(jamo2)
+ " => " + hangul2
);
if (!hangul.Equals(hangul2))
{
Errln("FAIL: " + Utility.Escape(buf.ToString()));
}
else
{
Logln(Utility.Escape(buf.ToString()));
}
}
}
public virtual Transliterator GetInstance()
{
List <Transliterator> transliterators = new List <Transliterator>();
int passNumber = 1;
int limit = Math.Max(idBlockVector.Count, dataVector.Count);
for (int i = 0; i < limit; i++)
{
if (i < idBlockVector.Count)
{
string idBlock = idBlockVector[i];
if (idBlock.Length > 0)
{
transliterators.Add(Transliterator.GetInstance(idBlock));
}
}
if (i < dataVector.Count)
{
#pragma warning disable 612, 618
Data data = dataVector[i];
transliterators.Add(new RuleBasedTransliterator("%Pass" + passNumber++, data, null));
#pragma warning restore 612, 618
}
}
Transliterator t = new CompoundTransliterator(transliterators, passNumber - 1);
t.ID = id;
if (compoundFilter != null)
{
t.Filter = compoundFilter;
}
return(t);
}
public void TestForWidth()
{
Transliterator widen = Transliterator.GetInstance("halfwidth-fullwidth");
Transliterator narrow = Transliterator.GetInstance("fullwidth-halfwidth");
UnicodeSet ASCII = new UnicodeSet("[:ascii:]");
string lettersAndSpace = "abc def";
string punctOnly = "( )";
String wideLettersAndSpace = widen.Transform(lettersAndSpace);
String widePunctOnly = widen.Transform(punctOnly);
assertContainsNone("Should be wide", ASCII, wideLettersAndSpace);
assertContainsNone("Should be wide", ASCII, widePunctOnly);
String back;
back = narrow.Transform(wideLettersAndSpace);
assertEquals("Should be narrow", lettersAndSpace, back);
back = narrow.Transform(widePunctOnly);
assertEquals("Should be narrow", punctOnly, back);
Transliterator latin = Transliterator.GetInstance("any-Latn");
back = latin.Transform(wideLettersAndSpace);
assertEquals("Should be ascii", lettersAndSpace, back);
back = latin.Transform(widePunctOnly);
assertEquals("Should be ascii", punctOnly, back);
// Han-Latin is now forward-only per CLDR ticket #5630
//Transliterator t2 = Transliterator.GetInstance("any-Han");
//back = t2.transform(widePunctOnly);
//assertEquals("Should be same", widePunctOnly, back);
}
public void TestAnyTranslit()
{
Transliterator tx = Transliterator.GetInstance("Any-Latin");
List <Thread> threads = new List <Thread>();
for (int i = 0; i < 8; i++)
{
threads.Add(new Thread(() => { tx.Transliterate("διαφορετικούς"); }));
}
foreach (Thread th in threads)
{
th.Start();
}
foreach (Thread th in threads)
{
#if !NETCOREAPP1_0
try
{
#endif
th.Join();
#if !NETCOREAPP1_0
}
catch (ThreadInterruptedException e)
{
Errln("Uexpected exception: " + e);
}
#endif
}
}
public void TestIncrementalProgress(string lang, string text)
{
var targets = Transliterator.GetAvailableTargets(lang);
foreach (string target in targets)
{
var variants = Transliterator.GetAvailableVariants(lang, target);
foreach (var variant in variants)
{
String id = lang + "-" + target + "/" + variant;
Logln("id: " + id);
Transliterator t = Transliterator.GetInstance(id);
CheckIncrementalAux(t, text);
String rev = t.Transliterate(text);
// Special treatment: This transliterator has no registered inverse, skip for now.
if (id.Equals("Devanagari-Arabic/"))
{
continue;
}
Transliterator inv = t.GetInverse();
CheckIncrementalAux(inv, rev);
}
}
}
public void TestRandomStrings()
{
Transliterator transform = Transliterator.GetInstance("Any-Latin");
using Analyzer a = Analyzer.NewAnonymous(createComponents: (fieldName, reader) =>
{
Tokenizer tokenizer = new MockTokenizer(reader, MockTokenizer.WHITESPACE, false);
return(new TokenStreamComponents(tokenizer, new ICUTransformFilter(tokenizer, transform)));
});
CheckRandomData(Random, a, 1000 * RandomMultiplier);
}
public override void Run()
{
Transliterator tx = Transliterator.GetInstance("Latin-Thai");
for (int loop = 0; loop < outerInstance.iterationCount; loop++)
{
foreach (String s in WORDS)
{
count += tx.Transliterate(s).Length;
}
}
}
// TODO: add support for custom rules
/// <summary>Creates a new <see cref="ICUTransformFilterFactory"/>.</summary>
public ICUTransformFilterFactory(IDictionary <string, string> args)
: base(args)
{
string id = Require(args, "id");
string direction = Get(args, "direction", new string[] { "forward", "reverse" }, "forward", false);
TransliterationDirection dir = "forward".Equals(direction, StringComparison.Ordinal) ? Transliterator.Forward : Transliterator.Reverse;
transliterator = Transliterator.GetInstance(id, dir);
if (args.Count > 0)
{
throw new ArgumentException(string.Format(J2N.Text.StringFormatter.CurrentCulture, "Unknown parameters: {0}", args));
}
}
// TODO: add support for custom rules
/// <summary>Creates a new <see cref="ICUTransformFilterFactory"/>.</summary>
public ICUTransformFilterFactory(IDictionary <string, string> args)
: base(args)
{
string id = Require(args, "id");
string direction = Get(args, "direction", new string[] { "forward", "reverse" }, "forward", false);
TransliterationDirection dir = "forward".Equals(direction) ? Transliterator.Forward : Transliterator.Reverse;
transliterator = Transliterator.GetInstance(id, dir);
if (args.Count != 0)
{
throw new ArgumentException("Unknown parameters: " + args);
}
}
public void TestBasicFunctionality()
{
CheckToken(Transliterator.GetInstance("Traditional-Simplified"),
"簡化字", "简化字");
CheckToken(Transliterator.GetInstance("Katakana-Hiragana"),
"ヒラガナ", "ひらがな");
CheckToken(Transliterator.GetInstance("Fullwidth-Halfwidth"),
"アルアノリウ", "ｱﾙｱﾉﾘｳ");
CheckToken(Transliterator.GetInstance("Any-Latin"),
"Αλφαβητικός Κατάλογος", "Alphabētikós Katálogos");
CheckToken(Transliterator.GetInstance("NFD; [:Nonspacing Mark:] Remove"),
"Alphabētikós Katálogos", "Alphabetikos Katalogos");
CheckToken(Transliterator.GetInstance("Han-Latin"),
"中国", "zhōng guó");
}
public void TestCommonDigits()
{
UnicodeSet westernDigitSet = new UnicodeSet("[0-9]");
UnicodeSet westernDigitSetAndMarks = new UnicodeSet("[[0-9][:Mn:]]");
UnicodeSet arabicDigitSet = new UnicodeSet("[[:Nd:]&[:block=Arabic:]]");
Transliterator latin = Transliterator.GetInstance("Any-Latn");
Transliterator arabic = Transliterator.GetInstance("Any-Arabic");
String westernDigits = getList(westernDigitSet);
String arabicDigits = getList(arabicDigitSet);
String fromArabic = latin.Transform(arabicDigits);
assertContainsAll("Any-Latin transforms Arabic digits", westernDigitSetAndMarks, fromArabic);
if (false)
{ // we don't require conversion to Arabic digits
String fromLatin = arabic.Transform(westernDigits);
assertContainsAll("Any-Arabic transforms Western digits", arabicDigitSet, fromLatin);
}
}
public void TestGetTransliterator()
{
Logln("Testing the getTransliterator() API of CompoundTransliterator");
String ID = "Latin-Greek;Greek-Latin;Latin-Devanagari;Devanagari-Latin;Latin-Cyrillic;Cyrillic-Latin;Any-Hex;Hex-Any";
Transliterator ct1 = null;
try
{
//ct1=new CompoundTransliterator(ID);
ct1 = Transliterator.GetInstance(ID);
}
catch (ArgumentException iae)
{
Errln("CompoundTransliterator construction failed for ID=" + ID);
throw iae;
}
//int count=ct1.getCount();
Transliterator[] elems = ct1.GetElements();
int count = elems.Length;
String[] array = split(ID, ';');
if (count != array.Length)
{
Errln("Error: getCount() failed. Expected:" + array.Length + " got:" + count);
}
for (int i = 0; i < count; i++)
{
//String child= ct1.getTransliterator(i).getID();
String child = elems[i].ID;
if (!child.Equals(array[i]))
{
Errln("Error getTransliterator() failed: Expected->" + array[i] + " Got->" + child);
}
else
{
Logln("OK: getTransliterator() passed: Expected->" + array[i] + " Got->" + child);
}
}
}
public void TestRBTErrors()
{
String rules = "ab>y";
String id = "MyRandom-YReverse";
String goodPattern = "[[:L:]&[\\u0000-\\uFFFF]]"; /* all BMP letters */
UnicodeSet set = null;
try
{
set = new UnicodeSet(goodPattern);
try
{
Transliterator t =
Transliterator.CreateFromRules(id, rules, Transliterator.REVERSE);
t.Filter = (set);
Transliterator.RegisterClass(id, t.GetType(), null);
Transliterator.Unregister(id);
try
{
Transliterator.GetInstance(id, Transliterator.REVERSE);
Errln("FAIL: construction of unregistered ID should have failed.");
}
catch (ArgumentException e)
{
}
}
catch (ArgumentException e)
{
Errln("FAIL: Was not able to create a good RBT to test registration.");
}
}
catch (ArgumentException e)
{
Errln("FAIL: Was not able to create a good UnicodeSet based on valid patterns.");
return;
}
}
public void TestPiecemeal()
{
String hangul = "\uBC0F";
String jamo = nameToJamo("(Mi)(I)(Cf)");
String latin = "mic";
String latin2 = "mich";
Transliterator t = null;
t = Transliterator.GetInstance("NFD"); // was Hangul-Jamo
TransliteratorTest.Expect(t, hangul, jamo);
t = Transliterator.GetInstance("NFC"); // was Jamo-Hangul
TransliteratorTest.Expect(t, jamo, hangul);
t = Transliterator.GetInstance("Latin-Jamo");
TransliteratorTest.Expect(t, latin, jamo);
t = Transliterator.GetInstance("Jamo-Latin");
TransliteratorTest.Expect(t, jamo, latin2);
t = Transliterator.GetInstance("Hangul-Latin");
TransliteratorTest.Expect(t, hangul, latin2);
t = Transliterator.GetInstance("Latin-Hangul");
TransliteratorTest.Expect(t, latin, hangul);
t = Transliterator.GetInstance("Hangul-Latin; Latin-Jamo");
TransliteratorTest.Expect(t, hangul, jamo);
t = Transliterator.GetInstance("Jamo-Latin; Latin-Hangul");
TransliteratorTest.Expect(t, jamo, hangul);
t = Transliterator.GetInstance("Hangul-Latin; Latin-Hangul");
TransliteratorTest.Expect(t, hangul, hangul);
}
internal void Check(String transliteratorName, String test, String shouldProduceStyles)
{
TestReplaceable tr = new TestReplaceable(test, null);
String original = tr.ToString();
Transliterator t;
if (transliteratorName.StartsWith("*", StringComparison.Ordinal))
{
transliteratorName = transliteratorName.Substring(1);
t = Transliterator.CreateFromRules("test", transliteratorName,
Transliterator.FORWARD);
}
else
{
t = Transliterator.GetInstance(transliteratorName);
}
t.Transliterate(tr);
String newStyles = tr.GetStyles();
if (!newStyles.Equals(shouldProduceStyles))
{
Errln("FAIL Styles: " + transliteratorName + " ( "
+ original + " ) => " + tr.ToString() + "; should be {" + shouldProduceStyles + "}!");
}
else
{
Logln("OK: " + transliteratorName + " ( " + original + " ) => " + tr.ToString());
}
if (!tr.HasMetaData || tr.Chars.HasMetaData ||
tr.Styles.HasMetaData)
{
Errln("Fail hasMetaData()");
}
}
public void TestScripts()
{
// get a couple of characters of each script for testing
StringBuffer testBuffer = new StringBuffer();
for (int script = 0; script < UScript.CodeLimit; ++script)
{
UnicodeSet test = new UnicodeSet().ApplyPropertyAlias("script", UScript.GetName(script));
int count = Math.Min(20, test.Count);
for (int i = 0; i < count; ++i)
{
testBuffer.Append(UTF16.ValueOf(test[i]));
}
}
{
String test = testBuffer.ToString();
Logln("Test line: " + test);
int inclusion = TestFmwk.GetExhaustiveness();
bool testedUnavailableScript = false;
for (int script = 0; script < UScript.CodeLimit; ++script)
{
if (script == UScript.Common || script == UScript.Inherited)
{
continue;
}
// if the inclusion rate is not 10, skip all but a small number of items.
// Make sure, however, that we test at least one unavailable script
if (inclusion < 10 && script != UScript.Latin &&
script != UScript.Han &&
script != UScript.Hiragana &&
testedUnavailableScript
)
{
continue;
}
String scriptName = UScript.GetName(script); // long name
UCultureInfo locale = new UCultureInfo(scriptName);
if (locale.Language.Equals("new") || locale.Language.Equals("pau"))
{
if (logKnownIssue("11171",
"long script name loosely looks like a locale ID with a known likely script"))
{
continue;
}
}
Transliterator t;
try
{
t = Transliterator.GetInstance("any-" + scriptName);
}
catch (Exception e)
{
testedUnavailableScript = true;
Logln("Skipping unavailable: " + scriptName);
continue; // we don't handle all scripts
}
Logln("Checking: " + scriptName);
if (t != null)
{
t.Transform(test); // just verify we don't crash
}
String shortScriptName = UScript.GetShortName(script); // 4-letter script code
try
{
t = Transliterator.GetInstance("any-" + shortScriptName);
}
catch (Exception e)
{
Errln("Transliterator.GetInstance() worked for \"any-" + scriptName +
"\" but not for \"any-" + shortScriptName + '\"');
}
t.Transform(test); // just verify we don't crash
}
}
}
public void TestEmptyTerm()
{
using (Analyzer a = Analyzer.NewAnonymous(createComponents: (fieldName, reader) =>
{
Tokenizer tokenizer = new KeywordTokenizer(reader);
return(new TokenStreamComponents(tokenizer, new ICUTransformFilter(tokenizer, Transliterator.GetInstance("Any-Latin"))));
}))
{
CheckOneTerm(a, "", "");
}
}
public void TestConstruction()
{
Logln("Testing the construction of the compound Transliterator");
String[] names = { "Greek-Latin", "Latin-Devanagari", "Devanagari-Latin", "Latin-Greek" };
try
{
Transliterator.GetInstance(names[0]);
Transliterator.GetInstance(names[1]);
Transliterator.GetInstance(names[2]);
Transliterator.GetInstance(names[3]);
}
catch (ArgumentException ex)
{
Errln("FAIL: Transliterator construction failed" + ex.ToString());
throw ex;
}
string[] IDs =
{
names[0],
names[0] + ";" + names[3],
names[3] + ";" + names[1] + ";" + names[2],
names[0] + ";" + names[1] + ";" + names[2] + ";" + names[3]
};
for (int i = 0; i < 4; i++)
{
try
{
Transliterator.GetInstance(IDs[i]);
}
catch (ArgumentException ex1)
{
Errln("FAIL: construction using CompoundTransliterator(String ID) failed for " + IDs[i]);
throw ex1;
}
try
{
Transliterator.GetInstance(IDs[i], Transliterator.Forward);
}
catch (ArgumentException ex2)
{
Errln("FAIL: construction using CompoundTransliterator(String ID, int direction=FORWARD) failed for " + IDs[i]);
throw ex2;
}
try
{
Transliterator.GetInstance(IDs[i], Transliterator.Reverse);
}
catch (ArgumentException ex3)
{
Errln("FAIL: construction using CompoundTransliterator(String ID, int direction=REVERSE) failed for " + IDs[i]);
throw ex3;
}
// try{
// CompoundTransliterator cpdtrans=new CompoundTransliterator(IDs[i], Transliterator.FORWARD, null);
// cpdtrans = null;
// }catch(IllegalArgumentException ex4) {
// Errln("FAIL: construction using CompoundTransliterator(String ID, int direction=FORWARD," +
// "UnicodeFilter adoptedFilter=0) failed for " + IDs[i]);
// throw ex4;
// }
//
//
// try{
// CompoundTransliterator cpdtrans2=new CompoundTransliterator(transarray[i], null);
// cpdtrans2 = null;
// }catch(IllegalArgumentException ex5) {
// Errln("FAIL: Construction using CompoundTransliterator(Transliterator transliterators[]," +
// "UnicodeFilter adoptedFilter = 0) failed");
// throw ex5;
// }
}
}
public void TestTransliterate()
{
Logln("Testing the handleTransliterate() API of CompoundTransliterator");
Transliterator ct1 = null;
try
{
ct1 = Transliterator.GetInstance("Any-Hex;Hex-Any");
}
catch (ArgumentException iae)
{
Errln("FAIL: construction using CompoundTransliterator(String ID) failed for " + "Any-Hex;Hex-Any");
throw iae;
}
String s = "abcabc";
expect(ct1, s, s);
TransliterationPosition index = new TransliterationPosition();
ReplaceableString rsource2 = new ReplaceableString(s);
String expectedResult = s;
ct1.Transliterate(rsource2, index);
ct1.FinishTransliteration(rsource2, index);
String result = rsource2.ToString();
expectAux(ct1.ID + ":ReplaceableString, index(0,0,0,0)", s + "->" + rsource2, result.Equals(expectedResult), expectedResult);
TransliterationPosition index2 = new TransliterationPosition(1, 3, 2, 3);
ReplaceableString rsource3 = new ReplaceableString(s);
ct1.Transliterate(rsource3, index2);
ct1.FinishTransliteration(rsource3, index2);
result = rsource3.ToString();
expectAux(ct1.ID + ":String, index2(1,2,2,3)", s + "->" + rsource3, result.Equals(expectedResult), expectedResult);
String[] Data =
{
//ID, input string, transliterated string
"Any-Hex;Hex-Any;Any-Hex", "hello", "\\u0068\\u0065\\u006C\\u006C\\u006F",
"Any-Hex;Hex-Any", "hello! How are you?", "hello! How are you?",
"Devanagari-Latin;Latin-Devanagari", "\u092D\u0948'\u0930'\u0935", "\u092D\u0948\u0930\u0935", // quotes lost
"Latin-Cyrillic;Cyrillic-Latin", "a'b'k'd'e'f'g'h'i'j'Shch'shch'zh'h", "a'b'k'd'e'f'g'h'i'j'Shch'shch'zh'h",
"Latin-Greek;Greek-Latin", "ABGabgAKLMN", "ABGabgAKLMN",
//"Latin-Arabic;Arabic-Latin", "Ad'r'a'b'i'k'dh'dd'gh", "Adrabikdhddgh",
"Hiragana-Katakana", "\u3041\u308f\u3099\u306e\u304b\u3092\u3099",
"\u30A1\u30f7\u30ce\u30ab\u30fa",
"Hiragana-Katakana;Katakana-Hiragana", "\u3041\u308f\u3099\u306e\u304b\u3051",
"\u3041\u308f\u3099\u306e\u304b\u3051",
"Katakana-Hiragana;Hiragana-Katakana", "\u30A1\u30f7\u30ce\u30f5\u30f6",
"\u30A1\u30f7\u30ce\u30ab\u30b1",
"Latin-Katakana;Katakana-Latin", "vavivuvevohuzizuzoninunasesuzezu",
"vavivuvevohuzizuzoninunasesuzezu",
};
Transliterator ct2 = null;
for (int i = 0; i < Data.Length; i += 3)
{
try
{
ct2 = Transliterator.GetInstance(Data[i + 0]);
}
catch (ArgumentException iae2)
{
Errln("FAIL: CompoundTransliterator construction failed for " + Data[i + 0]);
throw iae2;
}
expect(ct2, Data[i + 1], Data[i + 2]);
}
}
public void TestJamo()
{
Transliterator latinJamo = Transliterator.GetInstance("Latin-Jamo");
Transliterator jamoLatin = latinJamo.GetInverse();
String[] CASE =
{
// Column 1 is the latin text L1 to be fed to Latin-Jamo
// to yield output J.
// Column 2 is expected value of J. J is fed to
// Jamo-Latin to yield output L2.
// Column 3 is expected value of L2. If the expected
// value of L2 is L1, then L2 is null.
// add tests for the update to fix problems where it didn't follow the standard
// see also http://www.unicode.org/cldr/data/charts/transforms/Latin-Hangul.html
"gach", "(Gi)(A)(Cf)", null,
"geumhui", "(Gi)(EU)(Mf)(Hi)(YI)", null,
"choe", "(Ci)(OE)", null,
"wo", "(IEUNG)(WEO)", null,
"Wonpil", "(IEUNG)(WEO)(Nf)(Pi)(I)(L)", "wonpil",
"GIPPEUM", "(Gi)(I)(BB)(EU)(Mf)", "gippeum",
"EUTTEUM", "(IEUNG)(EU)(DD)(EU)(Mf)", "eutteum",
"KKOTNAE", "(GGi)(O)(Tf)(Ni)(AE)", "kkotnae",
"gaga", "(Gi)(A)(Gi)(A)", null,
"gag-a", "(Gi)(A)(Gf)(IEUNG)(A)", null,
"gak-ka", "(Gi)(A)(Kf)(Ki)(A)", null,
"gakka", "(Gi)(A)(GGi)(A)", null,
"gakk-a", "(Gi)(A)(GGf)(IEUNG)(A)", null,
"gakkka", "(Gi)(A)(GGf)(Ki)(A)", null,
"gak-kka", "(Gi)(A)(Kf)(GGi)(A)", null,
"bab", "(Bi)(A)(Bf)", null,
"babb", "(Bi)(A)(Bf)(Bi)(EU)", "babbeu",
"babbba", "(Bi)(A)(Bf)(Bi)(EU)(Bi)(A)", "babbeuba",
"bagg", "(Bi)(A)(Gf)(Gi)(EU)", "baggeu",
"baggga", "(Bi)(A)(Gf)(Gi)(EU)(Gi)(A)", "baggeuga",
//"bag"+SEP+"gga", "(Bi)(A)(Gf)"+SEP+"(Gi)(EU)(Gi)(A)", "bag"+SEP+"geuga",
"kabsa", "(Ki)(A)(Bf)(Si)(A)", null,
"kabska", "(Ki)(A)(BS)(Ki)(A)", null,
"gabsbka", "(Gi)(A)(BS)(Bi)(EU)(Ki)(A)", "gabsbeuka", // not (Kf)
"gga", "(Gi)(EU)(Gi)(A)", "geuga",
"bsa", "(Bi)(EU)(Si)(A)", "beusa",
"agg", "(IEUNG)(A)(Gf)(Gi)(EU)", "aggeu",
"agga", "(IEUNG)(A)(Gf)(Gi)(A)", null,
"la", "(R)(A)", null,
"bs", "(Bi)(EU)(Sf)", "beus",
"kalgga", "(Ki)(A)(L)(Gi)(EU)(Gi)(A)", "kalgeuga",
// 'r' in a final position is treated like 'l'
"karka", "(Ki)(A)(L)(Ki)(A)", "kalka",
};
for (int i = 0; i < CASE.Length; i += 3)
{
String jamo = nameToJamo(CASE[i + 1]);
if (CASE[i + 2] == null)
{
TransliteratorTest.Expect(latinJamo, CASE[i], jamo, jamoLatin);
}
else
{
// Handle case where round-trip is expected to fail
TransliteratorTest.Expect(latinJamo, CASE[i], jamo);
TransliteratorTest.Expect(jamoLatin, jamo, CASE[i + 2]);
}
}
}
/// <summary>
/// Returns a transliterator from the given source to our target or
/// target/variant. Returns NULL if the source is the same as our
/// target script, or if the source is <see cref="UScript.InvalidCode"/>.
/// Caches the result and returns the same transliterator the next
/// time. The caller does NOT own the result and must not delete
/// it.
/// </summary>
private Transliterator GetTransliterator(int source)
{
if (source == targetScript || source == UScript.InvalidCode)
{
if (IsWide(targetScript))
{
return(null);
}
else
{
return(widthFix);
}
}
int key = (int)source;
Transliterator t = cache.Get(key);
if (!cache.TryGetValue(key, out t) || t == null)
{
string sourceName = UScript.GetName(source);
string id = sourceName + TARGET_SEP + target;
try
{
t = Transliterator.GetInstance(id, FORWARD);
}
catch (Exception e) { }
if (t == null)
{
// Try to pivot around Latin, our most common script
id = sourceName + LATIN_PIVOT + target;
try
{
t = Transliterator.GetInstance(id, FORWARD);
}
catch (Exception e) { }
}
if (t != null)
{
if (!IsWide(targetScript))
{
IList <Transliterator> v = new List <Transliterator>();
v.Add(widthFix);
v.Add(t);
t = new CompoundTransliterator(v);
}
//Transliterator prevCachedT = cache.putIfAbsent(key, t);
Transliterator prevCachedT;
// ICU4N: This is to simulate putIfAbsent
// ICU4N TODO: If this works, make it into a PutIfAbsent extension method so we can go back to using ConcurrentDictionary elsewhere
if (!cache.TryGetValue(key, out prevCachedT))
{
// If another thread beat us here, set the prevCachedT
// value to NullTransliterator to indicate it already exists
if (!cache.TryAdd(key, t))
{
prevCachedT = new NullTransliterator();
}
}
if (prevCachedT != null)
{
t = prevCachedT;
}
}
else if (!IsWide(targetScript))
{
return(widthFix);
}
}
return(t);
}
public void TestOptimizer2()
{
CheckToken(Transliterator.GetInstance("Traditional-Simplified; CaseFold"),
"ABCDE", "abcde");
}
public void TestRealText()
{
Transliterator latinJamo = Transliterator.GetInstance("Latin-Jamo");
Transliterator jamoLatin = latinJamo.GetInverse();
Transliterator jamoHangul = Transliterator.GetInstance("NFC");
Transliterator hangulJamo = Transliterator.GetInstance("NFD");
//Transliterator rt = new CompoundTransliterator(new Transliterator[] {
// hangulJamo, jamoLatin, latinJamo, jamoHangul });
Transliterator rt = Transliterator.GetInstance("NFD;Jamo-Latin;Latin-Jamo;NFC");
int pos = 0;
StringBuffer buf = new StringBuffer();
int total = 0;
int errors = 0;
while (pos < WHAT_IS_UNICODE.Length)
{
int space = WHAT_IS_UNICODE.IndexOf(' ', pos + 1);
if (space < 0)
{
space = WHAT_IS_UNICODE.Length;
}
if (pos < space)
{
++total;
String hangul = WHAT_IS_UNICODE.Substring(pos, space - pos); // ICU4N: Corrected 2nd parameter
String hangulX = rt.Transliterate(hangul);
if (!hangul.Equals(hangulX))
{
++errors;
String jamo = hangulJamo.Transliterate(hangul);
String latin = jamoLatin.Transliterate(jamo);
String jamo2 = latinJamo.Transliterate(latin);
String hangul2 = jamoHangul.Transliterate(jamo2);
if (hangul.Equals(hangul2))
{
buf.Length = (0);
buf.Append("FAIL (Compound transliterator problem): ");
buf.Append(hangul + " => " +
jamoToName(jamo) + " => " +
latin + " => " + jamoToName(jamo2)
+ " => " + hangul2 +
"; but " + hangul + " =cpd=> " + jamoToName(hangulX)
);
Errln(Utility.Escape(buf.ToString()));
}
else if (jamo.Equals(jamo2))
{
buf.Length = (0);
buf.Append("FAIL (Jamo<>Hangul problem): ");
if (!hangul2.Equals(hangulX))
{
buf.Append("(Weird: " + hangulX + " != " + hangul2 + ")");
}
buf.Append(hangul + " => " +
jamoToName(jamo) + " => " +
latin + " => " + jamoToName(jamo2)
+ " => " + hangul2
);
Errln(Utility.Escape(buf.ToString()));
}
else
{
buf.Length = (0);
buf.Append("FAIL: ");
if (!hangul2.Equals(hangulX))
{
buf.Append("(Weird: " + hangulX + " != " + hangul2 + ")");
}
// The Hangul-Jamo conversion is not usually the
// bug here, so we hide it from display.
// Uncomment lines to see the Hangul.
buf.Append(//hangul + " => " +
jamoToName(jamo) + " => " +
latin + " => " + jamoToName(jamo2)
//+ " => " + hangul2
);
Errln(Utility.Escape(buf.ToString()));
}
}
}
pos = space + 1;
}
if (errors != 0)
{
Errln("Test word failures: " + errors + " out of " + total);
}
else
{
Logln("All " + total + " test words passed");
}
}
public void TestTransliteratorErrors()
{
String trans = "Latin-Greek";
String bogusID = "LATINGREEK-GREEKLATIN";
String newID = "Bogus-Latin";
String newIDRules = "zzz > Z; f <> ph";
String bogusRules = "a } [b-g m-p ";
ReplaceableString testString =
new ReplaceableString("A quick fox jumped over the lazy dog.");
String insertString = "cats and dogs";
int stoppedAt = 0, len;
Transliterator.Position pos = new Transliterator.Position();
Transliterator t =
Transliterator.GetInstance(trans, Transliterator.FORWARD);
if (t == null)
{
Errln("FAIL: construction of Latin-Greek");
return;
}
len = testString.Length;
stoppedAt = t.Transliterate(testString, 0, 100);
if (stoppedAt != -1)
{
Errln("FAIL: Out of bounds check failed (1).");
}
else if (testString.Length != len)
{
testString =
new ReplaceableString("A quick fox jumped over the lazy dog.");
Errln("FAIL: Transliterate fails and the target string was modified.");
}
stoppedAt = t.Transliterate(testString, 100, testString.Length - 1);
if (stoppedAt != -1)
{
Errln("FAIL: Out of bounds check failed (2).");
}
else if (testString.Length != len)
{
testString =
new ReplaceableString("A quick fox jumped over the lazy dog.");
Errln("FAIL: Transliterate fails and the target string was modified.");
}
pos.Start = 100;
pos.Limit = testString.Length;
try
{
t.Transliterate(testString, pos);
Errln("FAIL: Start offset is out of bounds, error not reported.");
}
catch (ArgumentException e)
{
Logln("Start offset is out of bounds and detected.");
}
pos.Limit = 100;
pos.Start = 0;
try
{
t.Transliterate(testString, pos);
Errln("FAIL: Limit offset is out of bounds, error not reported.\n");
}
catch (ArgumentException e)
{
Logln("Start offset is out of bounds and detected.");
}
len = pos.ContextLimit = testString.Length;
pos.ContextStart = 0;
pos.Limit = len - 1;
pos.Start = 5;
try
{
t.Transliterate(testString, pos, insertString);
if (len == pos.Limit)
{
Errln("FAIL: Test insertion with string: the transliteration position limit didn't change as expected.");
}
}
catch (ArgumentException e)
{
Errln("Insertion test with string failed for some reason.");
}
pos.ContextStart = 0;
pos.ContextLimit = testString.Length;
pos.Limit = testString.Length - 1;
pos.Start = 5;
try
{
t.Transliterate(testString, pos, 0x0061);
if (len == pos.Limit)
{
Errln("FAIL: Test insertion with character: the transliteration position limit didn't change as expected.");
}
}
catch (ArgumentException e)
{
Errln("FAIL: Insertion test with UTF-16 code point failed for some reason.");
}
len = pos.Limit = testString.Length;
pos.ContextStart = 0;
pos.ContextLimit = testString.Length - 1;
pos.Start = 5;
try
{
t.Transliterate(testString, pos, insertString);
Errln("FAIL: Out of bounds check failed (3).");
if (testString.Length != len)
{
Errln("FAIL: The input string was modified though the offsets were out of bounds.");
}
}
catch (ArgumentException e)
{
Logln("Insertion test with out of bounds indexes.");
}
Transliterator t1 = null;
try
{
t1 = Transliterator.GetInstance(bogusID, Transliterator.FORWARD);
if (t1 != null)
{
Errln("FAIL: construction of bogus ID \"LATINGREEK-GREEKLATIN\"");
}
}
catch (ArgumentException e)
{
}
//try { // unneeded - Exception cannot be thrown
Transliterator t2 =
Transliterator.CreateFromRules(
newID,
newIDRules,
Transliterator.FORWARD);
try
{
Transliterator t3 = t2.GetInverse();
Errln("FAIL: The newID transliterator was not registered so createInverse should fail.");
if (t3 != null)
{
Errln("FAIL: The newID transliterator was not registered so createInverse should fail.");
}
}
catch (Exception e)
{
}
//} catch (Exception e) { }
try
{
Transliterator t4 =
Transliterator.CreateFromRules(
newID,
bogusRules,
Transliterator.FORWARD);
if (t4 != null)
{
Errln("FAIL: The rules is malformed but error was not reported.");
}
}
catch (Exception e)
{
}
}
