/// <summary>Fast version of <see cref="UScript.GetScript(int)"/>. Basic Latin is an array lookup.</summary>
private int GetScript(int codepoint)
{
if (0 <= codepoint && codepoint < basicLatin.Length)
{
return(basicLatin[codepoint]);
}
else
{
int script = UScript.GetScript(codepoint);
if (combineCJ)
{
if (script == UScript.Han || script == UScript.Hiragana || script == UScript.Katakana)
{
return(UScript.Japanese);
}
else if (codepoint >= 0xFF10 && codepoint <= 0xFF19)
{
// when using CJK dictionary breaking, don't let full width numbers go to it, otherwise
// they are treated as punctuation. we currently have no cleaner way to fix this!
return(UScript.Latin);
}
else
{
return(script);
}
}
else
{
return(script);
}
}
}
public void TestAllCodepoints()
{
int code;
//String oldId="";
//String oldAbbrId="";
for (int i = 0; i <= 0x10ffff; i++)
{
code = UScript.InvalidCode;
code = UScript.GetScript(i);
if (code == UScript.InvalidCode)
{
Errln("UScript.getScript for codepoint 0x" + Hex(i) + " failed");
}
String id = UScript.GetName(code);
if (id.IndexOf("INVALID", StringComparison.Ordinal) >= 0)
{
Errln("UScript.getScript for codepoint 0x" + Hex(i) + " failed");
}
String abbr = UScript.GetShortName(code);
if (abbr.IndexOf("INV", StringComparison.Ordinal) >= 0)
{
Errln("UScript.getScript for codepoint 0x" + Hex(i) + " failed");
}
}
}
static ScriptIterator()
{
for (int i = 0; i < basicLatin.Length; i++)
{
basicLatin[i] = UScript.GetScript(i);
}
}
private static int[] LoadBasicLatin() // LUCENENET: Avoid static constructors (see https://github.com/apache/lucenenet/pull/224#issuecomment-469284006)
{
var basicLatin = new int[128];
for (int i = 0; i < basicLatin.Length; i++)
{
basicLatin[i] = UScript.GetScript(i);
}
return(basicLatin);
}
/// <summary>
/// Returns TRUE if there are any more runs. TRUE is always
/// returned at least once. Upon return, the caller should
/// examine scriptCode, start, and limit.
/// </summary>
public virtual bool Next()
{
int ch;
int s;
ScriptCode = UScript.InvalidCode; // don't know script yet
Start = Limit;
// Are we done?
if (Start == textLimit)
{
return(false);
}
// Move start back to include adjacent <see cref="UScript.Common"/> / <see cref="UScript.Inherited"/>
// characters
while (Start > textStart)
{
ch = text.Char32At(Start - 1); // look back
s = UScript.GetScript(ch);
if (s == UScript.Common || s == UScript.Inherited)
{
--Start;
}
else
{
break;
}
}
// Move limit ahead to include COMMON, INHERITED, and characters
// of the current script.
while (Limit < textLimit)
{
ch = text.Char32At(Limit); // look ahead
s = UScript.GetScript(ch);
if (s != UScript.Common && s != UScript.Inherited)
{
if (ScriptCode == UScript.InvalidCode)
{
ScriptCode = s;
}
else if (s != ScriptCode)
{
break;
}
}
++Limit;
}
// Return TRUE even if the entire text is COMMON / INHERITED, in
// which case scriptCode will be UScript.InvalidCode.
return(true);
}
public void TestGetScriptOfCharsWithScriptExtensions()
{
/* test characters which have Script_Extensions */
if (!(
UScript.Common == UScript.GetScript(0x0640) &&
UScript.Inherited == UScript.GetScript(0x0650) &&
UScript.Arabic == UScript.GetScript(0xfdf2))
)
{
Errln("UScript.getScript(character with Script_Extensions) failed");
}
}
public void TestGetScript(int codepoint, int expected)
{
int code = UScript.InvalidCode;
code = UScript.GetScript(codepoint);
if (code != expected)
{
Errln("Error testing UScript.getScript(). Got: " + code + " Expected: " + expected
+ " for codepoint 0x + Hex(codepoint).");
}
}
public void TestScriptMetadataAPI()
{
/* API & code coverage. */
String sample = UScript.GetSampleString(UScript.Latin);
if (sample.Length != 1 || UScript.GetScript(sample[0]) != UScript.Latin)
{
Errln("UScript.getSampleString(Latn) failed");
}
sample = UScript.GetSampleString(UScript.InvalidCode);
if (sample.Length != 0)
{
Errln("UScript.getSampleString(invalid) failed");
}
if (UScript.GetUsage(UScript.Latin) != ScriptUsage.Recommended ||
// Unicode 10 gives up on "aspirational".
UScript.GetUsage(UScript.Yi) != ScriptUsage.LimitedUse ||
UScript.GetUsage(UScript.Cherokee) != ScriptUsage.LimitedUse ||
UScript.GetUsage(UScript.Coptic) != ScriptUsage.Excluded ||
UScript.GetUsage(UScript.Cirth) != ScriptUsage.NotEncoded ||
UScript.GetUsage(UScript.InvalidCode) != ScriptUsage.NotEncoded ||
UScript.GetUsage(UScript.CodeLimit) != ScriptUsage.NotEncoded)
{
Errln("UScript.getUsage() failed");
}
if (UScript.IsRightToLeft(UScript.Latin) ||
UScript.IsRightToLeft(UScript.Cirth) ||
!UScript.IsRightToLeft(UScript.Arabic) ||
!UScript.IsRightToLeft(UScript.Hebrew))
{
Errln("UScript.isRightToLeft() failed");
}
if (UScript.BreaksBetweenLetters(UScript.Latin) ||
UScript.BreaksBetweenLetters(UScript.Cirth) ||
!UScript.BreaksBetweenLetters(UScript.Han) ||
!UScript.BreaksBetweenLetters(UScript.Thai))
{
Errln("UScript.breaksBetweenLetters() failed");
}
if (UScript.IsCased(UScript.Cirth) ||
UScript.IsCased(UScript.Han) ||
!UScript.IsCased(UScript.Latin) ||
!UScript.IsCased(UScript.Greek))
{
Errln("UScript.isCased() failed");
}
}
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));
}
public void TestAllCodepointsUsingTry()
{
int code;
for (int i = 0; i <= 0x10ffff; i++)
{
code = UScript.GetScript(i);
if (code == UScript.InvalidCode)
{
Errln("UScript.GetScript for codepoint 0x" + Hex(i) + " failed");
}
if (!UScript.TryGetName(code, out string id) || id.IndexOf("INVALID", StringComparison.Ordinal) >= 0)
{
Errln("UScript.GetScript for codepoint 0x" + Hex(i) + " failed");
}
if (!UScript.TryGetShortName(code, out string abbr) || abbr.IndexOf("INV", StringComparison.Ordinal) >= 0)
{
Errln("UScript.GetScript for codepoint 0x" + Hex(i) + " failed");
}
}
}
