8276302: Locale.filterTags methods ignore actual weight when matching "*" (as if it is 1)

Reviewed-by: naoto
This commit is contained in:
Daniel Le 2021-12-23 14:11:11 +00:00 committed by Naoto Sato
parent 214f98f6b0
commit 87cc4e5009
2 changed files with 73 additions and 75 deletions
src/java.base/share/classes/sun/util/locale
test/jdk/java/util/Locale

@ -124,8 +124,18 @@ public final class LocaleMatcher {
for (LanguageRange lr : nonZeroRanges) {
String range = lr.getRange();
if (range.equals("*")) {
tags = removeTagsMatchingBasicZeroRange(zeroRanges, tags);
return new ArrayList<String>(tags);
for (String tag : tags) {
// change to lowercase for case-insensitive matching
String lowerCaseTag = tag.toLowerCase(Locale.ROOT);
if (!caseInsensitiveMatch(list, lowerCaseTag)
&& !shouldIgnoreFilterBasicMatch(zeroRanges, lowerCaseTag)) {
// preserving the case of the input tag
list.add(tag);
}
}
break;
} else {
for (String tag : tags) {
// change to lowercase for case-insensitive matching
@ -148,44 +158,6 @@ public final class LocaleMatcher {
return list;
}
/**
* Removes the tag(s) which are falling in the basic exclusion range(s) i.e
* range(s) with q=0 and returns the updated collection. If the basic
* language ranges contains '*' as one of its non zero range then instead of
* returning all the tags, remove those which are matching the range with
* quality weight q=0.
*/
private static Collection<String> removeTagsMatchingBasicZeroRange(
List<LanguageRange> zeroRange, Collection<String> tags) {
if (zeroRange.isEmpty()) {
tags = removeDuplicates(tags);
return tags;
}
List<String> matchingTags = new ArrayList<>();
for (String tag : tags) {
// change to lowercase for case-insensitive matching
String lowerCaseTag = tag.toLowerCase(Locale.ROOT);
if (!shouldIgnoreFilterBasicMatch(zeroRange, lowerCaseTag)
&& !caseInsensitiveMatch(matchingTags, lowerCaseTag)) {
matchingTags.add(tag); // preserving the case of the input tag
}
}
return matchingTags;
}
/**
* Remove duplicate tags from the given {@code tags} by
* ignoring case considerations.
*/
private static Collection<String> removeDuplicates(
Collection<String> tags) {
Set<String> distinctTags = new TreeSet<>(String.CASE_INSENSITIVE_ORDER);
return tags.stream().filter(x -> distinctTags.add(x))
.toList();
}
/**
* Returns true if the given {@code list} contains an element which matches
* with the given {@code tag} ignoring case considerations.
@ -240,8 +212,18 @@ public final class LocaleMatcher {
for (LanguageRange lr : nonZeroRanges) {
String range = lr.getRange();
if (range.equals("*")) {
tags = removeTagsMatchingExtendedZeroRange(zeroRanges, tags);
return new ArrayList<String>(tags);
for (String tag : tags) {
// change to lowercase for case-insensitive matching
String lowerCaseTag = tag.toLowerCase(Locale.ROOT);
if (!caseInsensitiveMatch(list, lowerCaseTag)
&& !shouldIgnoreFilterExtendedMatch(zeroRanges, lowerCaseTag)) {
// preserving the case of the input tag
list.add(tag);
}
}
break;
}
String[] rangeSubtags = range.split("-");
for (String tag : tags) {
@ -267,33 +249,6 @@ public final class LocaleMatcher {
return list;
}
/**
* Removes the tag(s) which are falling in the extended exclusion range(s)
* i.e range(s) with q=0 and returns the updated collection. If the extended
* language ranges contains '*' as one of its non zero range then instead of
* returning all the tags, remove those which are matching the range with
* quality weight q=0.
*/
private static Collection<String> removeTagsMatchingExtendedZeroRange(
List<LanguageRange> zeroRange, Collection<String> tags) {
if (zeroRange.isEmpty()) {
tags = removeDuplicates(tags);
return tags;
}
List<String> matchingTags = new ArrayList<>();
for (String tag : tags) {
// change to lowercase for case-insensitive matching
String lowerCaseTag = tag.toLowerCase(Locale.ROOT);
if (!shouldIgnoreFilterExtendedMatch(zeroRange, lowerCaseTag)
&& !caseInsensitiveMatch(matchingTags, lowerCaseTag)) {
matchingTags.add(tag); // preserve the case of the input tag
}
}
return matchingTags;
}
/**
* The tag which is falling in the extended exclusion range(s) should
* not be considered as the matching tag. Ignores the tag matching with the

@ -1,5 +1,5 @@
/*
* Copyright (c) 2014, 2020, Oracle and/or its affiliates. All rights reserved.
* Copyright (c) 2014, 2021, 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 7069824 8042360 8032842 8175539 8210443 8242010
* @bug 7069824 8042360 8032842 8175539 8210443 8242010 8276302
* @summary Verify implementation for Locale matching.
* @run testng/othervm Bug7069824
*/
@ -208,27 +208,61 @@ public class Bug7069824 {
Object[][] LFilterTagsData() {
return new Object[][] {
// Range, LanguageTags, FilteringMode, Expected language tags
{"fr-FR, fr-BG;q=0.8, *;q=0.5, en;q=0", "en-US, fr-FR, fr-CA, fr-BG",
null, "fr-FR, fr-BG, fr-CA"},
{"fr-FR, fr-*-BG;q=0.8, *;q=0.5, en;q=0", "en-US, fr-FR, fr-CA, fr-BG",
null, "fr-FR, fr-BG, fr-CA"},
{"en;q=0.2, *;q=0.6, ja", "de-DE, en, ja-JP-hepburn, fr-JP, he",
null, "de-DE, en, ja-JP-hepburn, fr-JP, he"},
null, "ja-JP-hepburn, de-DE, en, fr-JP, he"},
{"en;q=0.2, ja-JP, fr-JP", "de-DE, en, ja-JP-hepburn, fr, he",
null, "ja-JP-hepburn, en"},
{"en;q=0.2, ja-JP, fr-JP, iw", "de-DE, he, en, ja-JP-hepburn, fr, he-IL",
null, "ja-JP-hepburn, he, he-IL, en"},
{"en;q=0.2, ja-JP, fr-JP, he", "de-DE, en, ja-JP-hepburn, fr, iw-IL",
null, "ja-JP-hepburn, iw-IL, en"},
{"de-DE", "de-DE, de-de, de-Latn-DE, de-Latf-DE, de-DE-x-goethe, "
+ "de-Latn-DE-1996, de-Deva-DE, de, de-x-DE, de-Deva",
MAP_EXTENDED_RANGES, "de-DE, de-DE-x-goethe"},
null, "de-DE, de-DE-x-goethe"},
{"de-*-DE", "de-DE, de-de, de-Latn-DE, de-Latf-DE, de-DE-x-goethe, "
+ "de-Latn-DE-1996, de-Deva-DE, de, de-x-DE, de-Deva",
null,
"de-DE, de-Latn-DE, de-Latf-DE, de-DE-x-goethe, "
+ "de-Latn-DE-1996, de-Deva-DE"},
{"de-DE", "de-DE, de-de, de-Latn-DE, de-Latf-DE, de-DE-x-goethe, "
+ "de-Latn-DE-1996, de-Deva-DE, de, de-x-DE, de-Deva",
EXTENDED_FILTERING,
"de-DE, de-Latn-DE, de-Latf-DE, de-DE-x-goethe, "
+ "de-Latn-DE-1996, de-Deva-DE"},
+ "de-Latn-DE-1996, de-Deva-DE"},
{"de-*-DE", "de-DE, de-de, de-Latn-DE, de-Latf-DE, de-DE-x-goethe, "
+ "de-Latn-DE-1996, de-Deva-DE, de, de-x-DE, de-Deva",
EXTENDED_FILTERING,
"de-DE, de-Latn-DE, de-Latf-DE, de-DE-x-goethe, "
+ "de-Latn-DE-1996, de-Deva-DE"},
+ "de-Latn-DE-1996, de-Deva-DE"},
{"de-DE", "de-DE, de-de, de-Latn-DE, de-Latf-DE, de-DE-x-goethe, "
+ "de-Latn-DE-1996, de-Deva-DE, de, de-x-DE, de-Deva",
IGNORE_EXTENDED_RANGES,
"de-DE, de-DE-x-goethe"},
{"de-*-DE", "de-DE, de-de, de-Latn-DE, de-Latf-DE, de-DE-x-goethe, "
+ "de-Latn-DE-1996, de-Deva-DE, de, de-x-DE, de-Deva",
IGNORE_EXTENDED_RANGES,
""},
{"de-DE", "de-DE, de-de, de-Latn-DE, de-Latf-DE, de-DE-x-goethe, "
+ "de-Latn-DE-1996, de-Deva-DE, de, de-x-DE, de-Deva",
MAP_EXTENDED_RANGES, "de-DE, de-DE-x-goethe"},
{"de-*-DE", "de-DE, de-de, de-Latn-DE, de-Latf-DE, de-DE-x-goethe, "
+ "de-Latn-DE-1996, de-Deva-DE, de, de-x-DE, de-Deva",
MAP_EXTENDED_RANGES, "de-DE, de-DE-x-goethe"},
{"de-DE", "de-DE, de-de, de-Latn-DE, de-Latf-DE, de-DE-x-goethe, "
+ "de-Latn-DE-1996, de-Deva-DE, de, de-x-DE, de-Deva",
REJECT_EXTENDED_RANGES, "de-DE, de-DE-x-goethe"},
// The next test in this chain is in testLFilterTagsIAE.
};
}
@ -380,6 +414,15 @@ public class Bug7069824 {
ranges, tags, expectedTags, actualTags));
}
@Test(expectedExceptions = IllegalArgumentException.class)
public void testLFilterTagsIAE() {
String ranges = "de-*-DE";
String tags = "de-DE, de-de, de-Latn-DE, de-Latf-DE, de-DE-x-goethe, "
+ "de-Latn-DE-1996, de-Deva-DE, de, de-x-DE, de-Deva";
List<LanguageRange> priorityList = LanguageRange.parse(ranges);
showLanguageTags(Locale.filterTags(priorityList, generateLanguageTags(tags), REJECT_EXTENDED_RANGES));
}
@Test(dataProvider = "LLookupData")
public void testLLookup(String ranges, String tags, String expectedLocale) {
List<LanguageRange> priorityList = LanguageRange.parse(ranges);