organicmaps/icu
13
Fork 0
mirror of https://github.com/unicode-org/icu.git synced 2025-04-07 06:25:30 +00:00
Code Issues 1 Projects Releases Wiki Activity Actions 2

ICU-1289 Add UScriptRun to ICU4J.

Browse source 
X-SVN-Rev: 8895
This commit is contained in:
Eric Mader 2002-06-14 19:08:57 +00:00
parent daa0a7488d
commit 05d2989deb
3 changed files with 683 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
icu4j/src/com/ibm/icu/dev/test/TestAll.java
View file

@ -5,8 +5,8 @@
*******************************************************************************
*
* $Source: /xsrl/Nsvn/icu/icu4j/src/com/ibm/icu/dev/test/TestAll.java,v $
* $Date: 2002/04/03 04:31:59 $
* $Revision: 1.25 $
* $Date: 2002/06/14 19:08:57 $
* $Revision: 1.26 $
*
*****************************************************************************************
*/
@ -126,6 +126,12 @@ public class TestAll extends TestFmwk {
});
}
public void TestUScriptRun() throws Exception {
run( new TestFmwk[] {
new com.ibm.icu.dev.test.lang.TestUScriptRun(),
});
}
public void TestNumberFormat() throws Exception {
run(new TestFmwk[] {
new com.ibm.icu.dev.test.format.IntlTestNumberFormat(),

272
icu4j/src/com/ibm/icu/dev/test/lang/TestUScriptRun.java Normal file
View file

@ -0,0 +1,272 @@
/**
*******************************************************************************
* Copyright (C) 1999-2002, International Business Machines Corporation and *
* others. All Rights Reserved. *
*******************************************************************************
*/
package com.ibm.icu.dev.test.lang;
import com.ibm.icu.lang.UScript;
import com.ibm.icu.lang.UScriptRun;
import com.ibm.icu.dev.test.TestFmwk;
public class TestUScriptRun extends TestFmwk
{
public TestUScriptRun()
{
// nothing
}
public static void main(String[] args) throws Exception {
new TestUScriptRun().run(args);
}
private static final class RunTestData
{
String runText;
int runScript;
public RunTestData(String theText, int theScriptCode)
{
runText = theText;
runScript = theScriptCode;
}
};
private static final RunTestData[] testData = {
new RunTestData("\u0020\u0946\u0939\u093F\u0928\u094D\u0926\u0940\u0020", UScript.DEVANAGARI),
new RunTestData("\u0627\u0644\u0639\u0631\u0628\u064A\u0629\u0020", UScript.ARABIC),
new RunTestData("\u0420\u0443\u0441\u0441\u043A\u0438\u0439\u0020", UScript.CYRILLIC),
new RunTestData("English (", UScript.LATIN),
new RunTestData("\u0E44\u0E17\u0E22", UScript.THAI),
new RunTestData(") ", UScript.LATIN),
new RunTestData("\u6F22\u5B75", UScript.HAN),
new RunTestData("\u3068\u3072\u3089\u304C\u306A\u3068", UScript.HIRAGANA),
new RunTestData("\u30AB\u30BF\u30AB\u30CA", UScript.KATAKANA),
new RunTestData("\uD801\uDC00\uD801\uDC01\uD801\uDC02\uD801\uDC03", UScript.DESERET)
};
private void CheckScriptRuns(UScriptRun scriptRun, int[] runStarts, RunTestData[] testData)
{
int run, runStart, runLimit;
int runScript;
/* iterate over all the runs */
run = 0;
while (scriptRun.next()) {
runStart = scriptRun.getScriptStart();
runLimit = scriptRun.getScriptLimit();
runScript = scriptRun.getScriptCode();
if (runStart != runStarts[run]) {
errln("Incorrect start offset for run " + run + ": expected " + runStarts[run] + ", got " + runStart);
}
if (runLimit != runStarts[run + 1]) {
errln("Incorrect limit offset for run " + run + ": expected " + runStarts[run + 1] + ", got " + runLimit);
}
if (runScript != testData[run].runScript) {
errln("Incorrect script for run " + run + ": expected \"" + UScript.getName(testData[run].runScript) + "\", got \"" + UScript.getName(runScript) + "\"");
}
run += 1;
/* stop when we've seen all the runs we expect to see */
if (run >= testData.length) {
break;
}
}
/* Complain if we didn't see then number of runs we expected */
if (run != testData.length) {
errln("Incorrect number of runs: expected " + testData.length + ", got " + run);
}
}
public void TestContstruction()
{
UScriptRun scriptRun = null;
char[] dummy = {'d', 'u', 'm', 'm', 'y'};
try {
scriptRun = new UScriptRun(null, 0, 100);
errln("new UScriptRun(null, 0, 100) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
try {
scriptRun = new UScriptRun(null, 100, 0);
errln("new UScriptRun(null, 100, 0) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
try {
scriptRun = new UScriptRun(null, 0, -100);
errln("new UScriptRun(null, 0, -100) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
try {
scriptRun = new UScriptRun(null, -100, 0);
errln("new UScriptRun(null, -100, 0) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
try {
scriptRun = new UScriptRun(dummy, 0, 6);
errln("new UScriptRun(dummy, 0, 6) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
try {
scriptRun = new UScriptRun(dummy, 6, 0);
errln("new UScriptRun(dummy, 6, 0) did not produce an IllegalArgumentException!");
}catch (IllegalArgumentException iae) {
}
try {
scriptRun = new UScriptRun(dummy, 0, -100);
errln("new UScriptRun(dummy, 0, -100) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
try {
scriptRun = new UScriptRun(dummy, -100, 0);
errln("new UScriptRun(dummy, -100, 0) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
}
public void TestReset()
{
UScriptRun scriptRun = null;
char[] dummy = {'d', 'u', 'm', 'm', 'y'};
try {
scriptRun = new UScriptRun();
} catch (IllegalArgumentException iae) {
errln("new UScriptRun() produced an IllegalArgumentException!");
}
try {
scriptRun.reset(0, 100);
errln("scriptRun.reset(0, 100) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
try {
scriptRun.reset(100, 0);
errln("scriptRun.reset(100, 0) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
try {
scriptRun.reset(0, -100);
errln("scriptRun.reset(0, -100) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
try {
scriptRun.reset(-100, 0);
errln("scriptRun.reset(-100, 0) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
try {
scriptRun.reset(dummy, 0, 6);
errln("scriptRun.reset(dummy, 0, 6) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
try {
scriptRun.reset(dummy, 6, 0);
errln("scriptRun.reset(dummy, 6, 0) did not produce an IllegalArgumentException!");
}catch (IllegalArgumentException iae) {
}
try {
scriptRun.reset(dummy, 0, -100);
errln("scriptRun.reset(dummy, 0, -100) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
try {
scriptRun.reset(dummy, -100, 0);
errln("scriptRun.reset(dummy, -100, 0) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
try {
scriptRun.reset(dummy, 0, dummy.length);
} catch (IllegalArgumentException iae) {
errln("scriptRun.reset(dummy, 0, dummy.length) produced an IllegalArgumentException!");
}
try {
scriptRun.reset(0, 6);
errln("scriptRun.reset(0, 6) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
try {
scriptRun.reset(6, 0);
errln("scriptRun.reset(6, 0) did not produce an IllegalArgumentException!");
} catch (IllegalArgumentException iae) {
}
}
public void TestRuns()
{
int stringLimit = 0;
int[] runStarts = new int[testData.length + 1];
String testString = "";
UScriptRun scriptRun = null;
/*
* Fill in the test string and the runStarts array.
*/
for (int run = 0; run < testData.length; run += 1) {
runStarts[run] = stringLimit;
stringLimit += testData[run].runText.length();
testString += testData[run].runText;
}
/* The limit of the last run */
runStarts[testData.length] = stringLimit;
try {
scriptRun = new UScriptRun(testString.toCharArray());
CheckScriptRuns(scriptRun, runStarts, testData);
} catch (IllegalArgumentException iae) {
errln("new UScriptRun(testString.toCharArray()) produced an IllegalArgumentException!");
}
try {
scriptRun.reset();
CheckScriptRuns(scriptRun, runStarts, testData);
} catch (IllegalArgumentException iae) {
errln("scriptRun.reset() on a valid UScriptRun produced an IllegalArgumentException!");
}
try {
scriptRun = new UScriptRun();
if (scriptRun.next()) {
errln("scriptRun.next() on an empty UScriptRun returned true!");
}
} catch (IllegalArgumentException iae) {
errln("new UScriptRun() produced an IllegalArgumentException!");
}
try {
scriptRun.reset(testString.toCharArray(), 0, testString.length());
CheckScriptRuns(scriptRun, runStarts, testData);
} catch (IllegalArgumentException iae) {
errln("scriptRun.reset(testString.toCharArray(), 0, testString.length) produced an IllegalArgumentException!");
}
}
}

403
icu4j/src/com/ibm/icu/lang/UScriptRun.java Normal file
View file

@ -0,0 +1,403 @@
/*
*******************************************************************************
*
* Copyright (C) 1999-2002, International Business Machines
* Corporation and others. All Rights Reserved.
*
*******************************************************************************
*/
package com.ibm.icu.lang;
/**
* <code>UScriptRun</code> is used to find runs of characters in
* the same script. It implements a simple iterator over an array
* of characters. The iterator will resolve script-neutral characters
* like punctuation into the script of the surrounding characters.
*
* The iterator will try to match paired punctuation. If it sees an
* opening punctuation character, it will remember the script that
* was assigned to that character, and assign the same script to the
* matching closing punctuation.
*
* Scripts are chosen based on the <code>UScript</code> class.
* No attempt is made to combine related scripts into a single run. In
* particular, Hiragana, Katakana, and Han characters will appear in seperate
* runs.
* Here is an example of how to iterate over script runs:
* <pre>
* void printScriptRuns(char []text)
* {
* UScriptRun scriptRun = new UScriptRun(text);
*
* while (scriptRun.next())) {
* int start = scriptRun.getScriptStart();
* int limit = scriptRun.getScriptLimit();
* int script = scriptRun.getScriptCode();
*
* System.out.println("Script \"" + UScript.getName(script) + "\" from " +
* start + " to " + limit + ".");
* }
* }
* </pre>
*/
public final class UScriptRun
{
/**
* Puts a copyright in the .class file
*/
private static final String copyrightNotice
= "Copyright \u00a91999-2002 IBM Corp. All rights reserved.";
/**
* Construct an empty <code>UScriptRun</code> object. The <code>next()</code>
* method will return <code>false</code> the first time it is called.
*/
public UScriptRun()
{
reset(null, 0, 0);
}
/**
* Construct a <code>UScriptRun</code> object which iterates over the given
* characetrs.
*
* @param chars the array of characters over which to iterate.
*/
public UScriptRun(char[] chars)
{
reset(chars, 0, chars.length);
}
/**
* Construct a <code>UScriptRun</code> object which iterates over a subrange
* of the given characetrs.
*
* @param chars the array of characters over which to iterate.
* @param start the index of the first character over which to iterate
* @param count the number of characters over which to iterate
*/
public UScriptRun(char[] chars, int start, int count)
{
reset(chars, start, count);
}
/**
* Reset the iterator to the start of the text.
*/
public final void reset() {
scriptStart = charStart;
scriptLimit = charStart;
scriptCode = UScript.COMMON;
}
/**
* Reset the iterator to iterate over the given range of the text. Throws
* IllegalArgumentException if the range is outside of the bounds of the
* character array.
*
* @param start the index of the new first character over which to iterate
* @param count the new number of characters over which to iterate.
* @exception IllegalArgumentException
*/
public final void reset(int start, int count)
throws IllegalArgumentException
{
int len = 0;
if (charArray != null) {
len = charArray.length;
}
if (start < 0 || count < 0 || start > len - count) {
throw new IllegalArgumentException();
}
charStart = start;
charLimit = start + count;
reset();
}
/**
* Reset the iterator to iterate over <code>count</code> characters
* in <code>chars</code> starting at <code>start</code>. This allows
* clients to reuse an iterator.
*
* @param chars the new array of characters over which to iterate.
* @param start the index of the first character over which to iterate.
* @param count the nuber of characters over which to iterate.
*/
public final void reset(char[] chars, int start, int count)
{
charArray = chars;
reset(start, count);
}
/**
* Get the starting index of the current script run.
*
* @returns the index of the first character in the current script run.
*/
public final int getScriptStart()
{
return scriptStart;
}
/**
* Get the index of the first character after the current script run.
*
* @return the index of the first character after the current script run.
*/
public final int getScriptLimit()
{
return scriptLimit;
}
/**
* Get the script code for the script of the current script run.
*
* @return the script code for the script of the current script run.
* @see com.ibm.icu.lang.UScript
*/
public final int getScriptCode()
{
return scriptCode;
}
/**
* Find the next script run. Returns <code>false</code> if there
* isn't another run, returns <code>true</code> if there is.
*
* @return <code>false</code> if there isn't another run, <code>true</code> if there is.
*/
public final boolean next()
{
int startSP = parenSP; // used to find the first new open character
// if we've fallen off the end of the text, we're done
if (scriptLimit >= charLimit) {
return false;
}
scriptCode = UScript.COMMON;
for (scriptStart = scriptLimit; scriptLimit < charLimit; scriptLimit += 1) {
int high = charArray[scriptLimit];
int ch = high;
// if the character is a high surrogate and it's not the last one
// in the text, see if it's followed by a low surrogate
if (high >= 0xD800 && high <= 0xDBFF && scriptLimit < charLimit - 1)
{
int low = charArray[scriptLimit + 1];
// if it is followed by a low surrogate,
// consume it and form the full character
if (low >= 0xDC00 && low <= 0xDFFF) {
ch = (high - 0xD800) * 0x0400 + low - 0xDC00 + 0x10000;
scriptLimit += 1;
}
}
int sc = UScript.getScript(ch);
int pairIndex = getPairIndex(ch);
// Paired character handling:
//
// if it's an open character, push it onto the stack.
// if it's a close character, find the matching open on the
// stack, and use that script code. Any non-matching open
// characters above it on the stack will be poped.
if (pairIndex >= 0) {
if ((pairIndex & 1) == 0) {
parenStack[++parenSP] = new ParenStackEntry(pairIndex, scriptCode);
} else if (parenSP >= 0) {
int pi = pairIndex & ~1;
while (parenSP >= 0 && parenStack[parenSP].pairIndex != pi) {
parenSP -= 1;
}
if (parenSP < startSP) {
startSP = parenSP;
}
if (parenSP >= 0) {
sc = parenStack[parenSP].scriptCode;
}
}
}
if (sameScript(scriptCode, sc)) {
if (scriptCode <= UScript.INHERITED && sc > UScript.INHERITED) {
scriptCode = sc;
// now that we have a final script code, fix any open
// characters we pushed before we knew the script code.
while (startSP < parenSP) {
parenStack[++startSP].scriptCode = scriptCode;
}
}
// if this character is a close paired character,
// pop it from the stack
if (pairIndex >= 0 && (pairIndex & 1) != 0 && parenSP >= 0) {
parenSP -= 1;
startSP -= 1;
}
} else {
// if the run broke on a surrogate pair,
// end it before the high surrogate
if (ch >= 0x10000) {
scriptLimit -= 1;
}
break;
}
}
return true;
}
/**
* Compare two script codes to see if they are in the same script. If one script is
* a strong script, and the other is INHERITED or COMMON, it will compare equal.
*
* @param scriptOne one of the script codes.
* @param scriptTwo the other script code.
* @return <code>true</code> if the two scripts are the same.
* @see com.ibm.icu.lang.UScript
*/
private static boolean sameScript(int scriptOne, int scriptTwo)
{
return scriptOne <= UScript.INHERITED || scriptTwo <= UScript.INHERITED || scriptOne == scriptTwo;
}
/*
* An internal class which holds entries on the paren stack.
*/
private static final class ParenStackEntry
{
int pairIndex;
int scriptCode;
public ParenStackEntry(int thePairIndex, int theScriptCode)
{
pairIndex = thePairIndex;
scriptCode = theScriptCode;
}
};
private int charStart;
private int charLimit;
private char charArray[];
private int scriptStart;
private int scriptLimit;
private int scriptCode;
private static ParenStackEntry parenStack[] = new ParenStackEntry[128];
private int parenSP;
/**
* Find the highest bit that's set in a word. Uses a binary search through
* the bits.
*
* @param n the word in which to find the highest bit that's set.
* @return the bit number (counting from the low order bit) of the highest bit.
*/
private static final byte highBit(int n)
{
if (n <= 0) {
return -32;
}
byte bit = 0;
if (n >= 1 << 16) {
n >>= 16;
bit += 16;
}
if (n >= 1 << 8) {
n >>= 8;
bit += 8;
}
if (n >= 1 << 4) {
n >>= 4;
bit += 4;
}
if (n >= 1 << 2) {
n >>= 2;
bit += 2;
}
if (n >= 1 << 1) {
n >>= 1;
bit += 1;
}
return bit;
}
/**
* Search the pairedChars array for the given character.
*
* @param ch the character for which to search.
* @return the index of the character in the table, or -1 if it's not there.
*/
private static int getPairIndex(int ch)
{
int probe = pairedCharPower;
int index = 0;
if (ch >= pairedChars[pairedCharExtra]) {
index = pairedCharExtra;
}
while (probe > (1 << 0)) {
probe >>= 1;
if (ch >= pairedChars[index + probe]) {
index += probe;
}
}
if (pairedChars[index] != ch) {
index = -1;
}
return index;
}
private static int pairedChars[] = {
0x0028, 0x0029, // ascii paired punctuation
0x003c, 0x003e,
0x005b, 0x005d,
0x007b, 0x007d,
0x00ab, 0x00bb, // guillemets
0x2018, 0x2019, // general punctuation
0x201c, 0x201d,
0x2039, 0x203a,
0x3008, 0x3009, // chinese paired punctuation
0x300a, 0x300b,
0x300c, 0x300d,
0x300e, 0x300f,
0x3010, 0x3011,
0x3014, 0x3015,
0x3016, 0x3017,
0x3018, 0x3019,
0x301a, 0x301b
};
private static int pairedCharPower = 1 << highBit(pairedChars.length);
private static int pairedCharExtra = pairedChars.length - pairedCharPower;
}