public void TestScriptMetadata()
{
UnicodeSet rtl = new UnicodeSet("[[:bc=R:][:bc=AL:]-[:Cn:]-[:sc=Common:]]");
// So far, sample characters are uppercase.
// Georgian is special.
UnicodeSet cased = new UnicodeSet("[[:Lu:]-[:sc=Common:]-[:sc=Geor:]]");
for (int sc = 0; sc < UScript.CodeLimit; ++sc)
{
String sn = UScript.GetShortName(sc);
ScriptUsage usage = UScript.GetUsage(sc);
String sample = UScript.GetSampleString(sc);
UnicodeSet scriptSet = new UnicodeSet();
scriptSet.ApplyInt32PropertyValue(UProperty.Script, sc);
if (usage == ScriptUsage.NotEncoded)
{
assertTrue(sn + " not encoded, no sample", sample.Length == 0); // Java 6: sample.isEmpty()
assertFalse(sn + " not encoded, not RTL", UScript.IsRightToLeft(sc));
assertFalse(sn + " not encoded, not LB letters", UScript.BreaksBetweenLetters(sc));
assertFalse(sn + " not encoded, not cased", UScript.IsCased(sc));
assertTrue(sn + " not encoded, no characters", scriptSet.IsEmpty);
}
else
{
assertFalse(sn + " encoded, has a sample character", sample.Length == 0); // Java 6: sample.isEmpty()
int firstChar = sample.CodePointAt(0);
int charScript = GetCharScript(sc);
assertEquals(sn + " script(sample(script))",
charScript, UScript.GetScript(firstChar));
assertEquals(sn + " RTL vs. set", rtl.Contains(firstChar), UScript.IsRightToLeft(sc));
assertEquals(sn + " cased vs. set", cased.Contains(firstChar), UScript.IsCased(sc));
assertEquals(sn + " encoded, has characters", sc == charScript, !scriptSet.IsEmpty);
if (UScript.IsRightToLeft(sc))
{
rtl.RemoveAll(scriptSet);
}
if (UScript.IsCased(sc))
{
cased.RemoveAll(scriptSet);
}
}
}
assertEquals("no remaining RTL characters", "[]", rtl.ToPattern(true));
assertEquals("no remaining cased characters", "[]", cased.ToPattern(true));
assertTrue("Hani breaks between letters", UScript.BreaksBetweenLetters(UScript.Han));
assertTrue("Thai breaks between letters", UScript.BreaksBetweenLetters(UScript.Thai));
assertFalse("Latn does not break between letters", UScript.BreaksBetweenLetters(UScript.Latin));
}