stan/jdk-24
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

8041791: String.toLowerCase regression - violates Unicode standard

Browse Source 
Reviewed-by: peytoia
This commit is contained in:
Naoto Sato 2014-05-14 10:52:51 -07:00
parent 3b959b9adf
commit d0dadf23c5
3 changed files with 25 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
jdk/src/share/classes/java/lang/ConditionalSpecialCasing.java
View File

@ -62,6 +62,7 @@ final class ConditionalSpecialCasing {
//# Conditional mappings
//# ================================================================================
new Entry(0x03A3, new char[]{0x03C2}, new char[]{0x03A3}, null, FINAL_CASED), // # GREEK CAPITAL LETTER SIGMA
new Entry(0x0130, new char[]{0x0069, 0x0307}, new char[]{0x0130}, null, 0), // # LATIN CAPITAL LETTER I WITH DOT ABOVE
//# ================================================================================
//# Locale-sensitive mappings
@ -77,8 +78,8 @@ final class ConditionalSpecialCasing {
//# ================================================================================
//# Turkish and Azeri
// new Entry(0x0130, new char[]{0x0069}, new char[]{0x0130}, "tr", 0), // # LATIN CAPITAL LETTER I WITH DOT ABOVE
// new Entry(0x0130, new char[]{0x0069}, new char[]{0x0130}, "az", 0), // # LATIN CAPITAL LETTER I WITH DOT ABOVE
new Entry(0x0130, new char[]{0x0069}, new char[]{0x0130}, "tr", 0), // # LATIN CAPITAL LETTER I WITH DOT ABOVE
new Entry(0x0130, new char[]{0x0069}, new char[]{0x0130}, "az", 0), // # LATIN CAPITAL LETTER I WITH DOT ABOVE
new Entry(0x0307, new char[]{}, new char[]{0x0307}, "tr", AFTER_I), // # COMBINING DOT ABOVE
new Entry(0x0307, new char[]{}, new char[]{0x0307}, "az", AFTER_I), // # COMBINING DOT ABOVE
new Entry(0x0049, new char[]{0x0131}, new char[]{0x0049}, "tr", NOT_BEFORE_DOT), // # LATIN CAPITAL LETTER I
@ -147,21 +148,25 @@ final class ConditionalSpecialCasing {
private static char[] lookUpTable(String src, int index, Locale locale, boolean bLowerCasing) {
HashSet<Entry> set = entryTable.get(new Integer(src.codePointAt(index)));
char[] ret = null;
if (set != null) {
Iterator<Entry> iter = set.iterator();
String currentLang = locale.getLanguage();
while (iter.hasNext()) {
Entry entry = iter.next();
String conditionLang= entry.getLanguage();
String conditionLang = entry.getLanguage();
if (((conditionLang == null) || (conditionLang.equals(currentLang))) &&
isConditionMet(src, index, locale, entry.getCondition())) {
return (bLowerCasing ? entry.getLowerCase() : entry.getUpperCase());
ret = bLowerCasing ? entry.getLowerCase() : entry.getUpperCase();
if (conditionLang != null) {
break;
}
}
}
}
return null;
return ret;
}
private static boolean isConditionMet(String src, int index, Locale locale, int condition) {

3
jdk/src/share/classes/java/lang/String.java
View File

@ -2583,6 +2583,9 @@ public final class String
if (cp == '\u03A3') { // GREEK CAPITAL LETTER SIGMA
return toLowerCaseEx(result, i, locale, false);
}
if (cp == '\u0130') { // LATIN CAPITAL LETTER I WITH DOT ABOVE
return toLowerCaseEx(result, i, locale, true);
}
cp = Character.toLowerCase(cp);
if (!Character.isBmpCodePoint(cp)) {
return toLowerCaseEx(result, i, locale, false);

16
jdk/test/java/lang/String/ToLowerCase.java
View File

@ -1,5 +1,5 @@
/*
* Copyright (c) 2003, 2013, Oracle and/or its affiliates. All rights reserved.
* Copyright (c) 2003, 2014, Oracle and/or its affiliates. All rights reserved.
* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
*
* This code is free software; you can redistribute it and/or modify it
@ -23,7 +23,7 @@
/*
@test
@bug 4217441 4533872 4900935 8020037 8032012
@bug 4217441 4533872 4900935 8020037 8032012 8041791
@summary toLowerCase should lower-case Greek Sigma correctly depending
on the context (final/non-final). Also it should handle
Locale specific (lt, tr, and az) lowercasings and supplementary
@ -72,8 +72,10 @@ public class ToLowerCase {
// I-dot tests
test("\u0130", turkish, "i");
test("\u0130", az, "i");
test("\u0130", lt, "i");
test("\u0130", Locale.US, "i");
test("\u0130", lt, "\u0069\u0307");
test("\u0130", Locale.US, "\u0069\u0307");
test("\u0130", Locale.JAPAN, "\u0069\u0307");
test("\u0130", Locale.ROOT, "\u0069\u0307");
// Remove dot_above in the sequence I + dot_above (Turkish and Azeri)
test("I\u0307", turkish, "i");
@ -111,6 +113,12 @@ public class ToLowerCase {
if (cp >= Character.MIN_HIGH_SURROGATE && cp <= Character.MAX_HIGH_SURROGATE) {
continue;
}
if (cp == 0x0130) {
// Although UnicodeData.txt has the lower case char as \u0069, it should be
// handled with the rules in SpecialCasing.txt, i.e., \u0069\u0307 in
// non Turkic locales.
continue;
}
int lowerCase = Character.toLowerCase(cp);
if (lowerCase == -1) { //Character.ERROR
continue;