This is an automated email from the ASF dual-hosted git repository.
rmaucher pushed a commit to branch 9.0.x
in repository https://gitbox.apache.org/repos/asf/tomcat.git
The following commit(s) were added to refs/heads/9.0.x by this push:
new c8d9d52170 Add dedicated AcceptLanguage2 test
c8d9d52170 is described below
commit c8d9d52170fa5bceab5272ea5a1fbdec23556337
Author: remm <[email protected]>
AuthorDate: Mon Oct 26 13:53:44 2026 +0100
Add dedicated AcceptLanguage2 test
Co authored with OpenCode.
---
.../util/http/parser/TestAcceptLanguage2.java | 276 +++++++++++++++++++++
1 file changed, 276 insertions(+)
diff --git a/test/org/apache/tomcat/util/http/parser/TestAcceptLanguage2.java
b/test/org/apache/tomcat/util/http/parser/TestAcceptLanguage2.java
new file mode 100644
index 0000000000..b59a86a78c
--- /dev/null
+++ b/test/org/apache/tomcat/util/http/parser/TestAcceptLanguage2.java
@@ -0,0 +1,276 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package org.apache.tomcat.util.http.parser;
+
+import java.io.StringReader;
+import java.util.List;
+import java.util.Locale;
+
+import org.junit.Assert;
+import org.junit.Test;
+
+/*
+ * Tests for AcceptLanguage2, the Accept-Language parser that creates Locales
+ * with the deprecated Locale constructor.
+ * <p>
+ * The Locales match those that AcceptLanguage (which uses
+ * Locale.forLanguageTag()) produces for the language, region and variant
+ * components of a tag. Script subtags are the documented exception: the
+ * deprecated constructor cannot represent them, so they are dropped and the
+ * result is the Locale for the tag with the script subtag removed.
+ */
+public class TestAcceptLanguage2 {
+
+ private static final Locale L_EN = Locale.forLanguageTag("en");
+ private static final Locale L_EN_GB = Locale.forLanguageTag("en-gb");
+ private static final Locale L_FR = Locale.forLanguageTag("fr");
+ private static final double Q1_000 = 1;
+ private static final double Q0_500 = 0.5;
+
+ private static List<AcceptLanguage> parse(String value) throws Exception {
+ return AcceptLanguage2.parse(new StringReader(value), -1);
+ }
+
+ private static Locale locale(String value) throws Exception {
+ List<AcceptLanguage> actual = parse(value);
+ Assert.assertEquals(1, actual.size());
+ return actual.get(0).getLocale();
+ }
+
+ /*
+ * Removes the script subtag of a tag, which is what this parser is
expected
+ * to do with it.
+ */
+ private static String withoutScript(String tag) {
+ int start = tag.indexOf('-');
+ if (start < 0) {
+ return tag;
+ }
+ int end = tag.indexOf('-', start + 1);
+ int subEnd = end < 0 ? tag.length() : end;
+ if (subEnd - start - 1 != 4) {
+ return tag;
+ }
+ for (int i = start + 1; i < subEnd; i++) {
+ char c = tag.charAt(i);
+ if ((c < 'a' || c > 'z') && (c < 'A' || c > 'Z')) {
+ return tag;
+ }
+ }
+ return tag.substring(0, start) + (end < 0 ? "" : tag.substring(end));
+ }
+
+ /*
+ * The reported case: the script subtag of zh-Hans-CN must no longer end up
+ * in the Locale variant, which produced the language tag
+ * zh-CN-x-lvariant-HANS rather than the tag the client sent.
+ */
+ @Test
+ public void testScriptIsNotMappedToTheVariant() throws Exception {
+ List<AcceptLanguage> actual = parse("zh-Hans-CN;q=1.0");
+
+ Assert.assertEquals(1, actual.size());
+ Locale locale = actual.get(0).getLocale();
+ Assert.assertEquals("zh", locale.getLanguage());
+ Assert.assertEquals("", locale.getScript());
+ Assert.assertEquals("CN", locale.getCountry());
+ Assert.assertEquals("", locale.getVariant());
+ Assert.assertFalse(locale.toLanguageTag(),
locale.toLanguageTag().contains("x-lvariant"));
+ Assert.assertEquals("zh-CN", locale.toLanguageTag());
+ Assert.assertEquals(Locale.forLanguageTag("zh-CN"), locale);
+ Assert.assertEquals(Q1_000, actual.get(0).getQuality(), 0.0001);
+ }
+
+ /*
+ * Script subtags, in the various cases a header may use, are dropped. The
+ * result is the Locale for the same tag without the script subtag, and the
+ * region that follows the script is still used.
+ */
+ @Test
+ public void testScriptIsDropped() throws Exception {
+ String[] tags = { "zh-Hans-CN", "zh-Hant-TW", "zh-hans-cn",
"ZH-HANS-CN", "zh-Hant-HK", "sr-Latn-RS",
+ "sr-Cyrl-ME", "yue-Hant-HK", "az-Latn-AZ", "uz-Arab",
"mn-Mong-CN", "hy-npoy", "pa-Pahk-PK",
+ "bs-Cyrl-BA", "zh-Hani", "sr-latn", "zh-Latn-419" };
+
+ for (String tag : tags) {
+ Locale actual = locale(tag);
+ Assert.assertEquals(tag,
Locale.forLanguageTag(withoutScript(tag)), actual);
+ Assert.assertEquals(tag, "", actual.getScript());
+ Assert.assertEquals(tag,
Locale.forLanguageTag(withoutScript(tag)).toLanguageTag(),
+ actual.toLanguageTag());
+ }
+ }
+
+ /*
+ * Script subtags that are not in the IANA registry, and script subtags in
+ * any case, are dropped like the registered ones. Locale.forLanguageTag()
+ * accepts them, so they must not make the tag invalid here.
+ */
+ @Test
+ public void testUnregisteredScript() throws Exception {
+ String[] tags = { "qaa-Qaak-CN", "zh-Qaaa-CN", "en-Zzzz-US",
"abcd-Hans-US" };
+
+ for (String tag : tags) {
+ Assert.assertEquals(tag,
Locale.forLanguageTag(withoutScript(tag)), locale(tag));
+ }
+ }
+
+ /*
+ * Dropping the script does not affect the variant, which is kept
unchanged,
+ * nor the region.
+ */
+ @Test
+ public void testVariantWithScript() throws Exception {
+ String[] tags = { "zh-Hans-CN-1996", "zh-Hant-1996",
"de-Latn-DE-POSIX", "sr-Latn-RS-1996",
+ "en-Latn-US-oxendict", "zh-Hans-US-VALAMY" };
+
+ for (String tag : tags) {
+ Locale expected = Locale.forLanguageTag(withoutScript(tag));
+ Locale actual = locale(tag);
+ Assert.assertEquals(tag, expected, actual);
+ Assert.assertEquals(tag, expected.getVariant(),
actual.getVariant());
+ Assert.assertEquals(tag, expected.toLanguageTag(),
actual.toLanguageTag());
+ }
+ }
+
+ /*
+ * Only the first four letter subtag before the region is a script. A
second
+ * one is not a valid variant subtag and drops the remainder of the tag,
+ * while a script subtag after the region is ignored rather than replacing
+ * the region.
+ */
+ @Test
+ public void testRepeatedScript() throws Exception {
+ Assert.assertEquals(Locale.forLanguageTag("zh"),
locale("zh-Hans-Hans"));
+ Assert.assertEquals(Locale.forLanguageTag("en-US"),
locale("en-Latn-US-Latn"));
+ Assert.assertEquals(Locale.forLanguageTag("zh-CN"),
locale("zh-CN-Hans"));
+ }
+
+ /*
+ * Same as the bug56848 test for AcceptLanguage. Note that the two entries
+ * of that header produce two different Locales with AcceptLanguage, while
+ * this parser drops the scripts and cannot tell zh-Hant-CN and zh-Hans-CN
+ * apart. That is the accepted cost of using the deprecated constructor.
+ */
+ @Test
+ public void bug56848() throws Exception {
+ List<AcceptLanguage> actual =
parse("zh-hant-CN;q=0.5,zh-hans-TW;q=0.05");
+
+ Assert.assertEquals(2, actual.size());
+
+ Assert.assertEquals(Locale.forLanguageTag("zh-CN"),
actual.get(0).getLocale());
+ Assert.assertEquals(Q0_500, actual.get(0).getQuality(), 0.0001);
+ Assert.assertEquals(Locale.forLanguageTag("zh-TW"),
actual.get(1).getLocale());
+ Assert.assertEquals(Q1_000 * 0.05, actual.get(1).getQuality(), 0.0001);
+ }
+
+ /*
+ * Tags without a script subtag produce the same Locale as
+ * Locale.forLanguageTag().
+ */
+ @Test
+ public void testNoScript() throws Exception {
+ String[] tags = { "en", "en-US", "en-GB", "fr-FR", "es-419", "zh-CN",
"ja", "de-DE-1901", "de-DE-POSIX",
+ "en-US-oxendict", "en-US-1996", "nl-NL", "pt-BR" };
+
+ for (String tag : tags) {
+ Assert.assertEquals(tag, Locale.forLanguageTag(tag), locale(tag));
+ }
+ }
+
+ /*
+ * Header parsing itself is shared with AcceptLanguage, so the basic cases
+ * behave the same way.
+ */
+ @Test
+ public void testMultiple() throws Exception {
+ List<AcceptLanguage> actual = parse("en; q= 0.05,fr;q=0.5,zh-Hans-CN");
+
+ Assert.assertEquals(3, actual.size());
+ Assert.assertEquals(L_EN, actual.get(0).getLocale());
+ Assert.assertEquals(0.05, actual.get(0).getQuality(), 0.0001);
+ Assert.assertEquals(L_FR, actual.get(1).getLocale());
+ Assert.assertEquals(Q0_500, actual.get(1).getQuality(), 0.0001);
+ Assert.assertEquals(Locale.forLanguageTag("zh-CN"),
actual.get(2).getLocale());
+ Assert.assertEquals(Q1_000, actual.get(2).getQuality(), 0.0001);
+ }
+
+
+ @Test
+ public void testSingle() throws Exception {
+ List<AcceptLanguage> actual = AcceptLanguage2.parse(new
StringReader("en-gb;q=0.500"), -1);
+
+ Assert.assertEquals(1, actual.size());
+ Assert.assertEquals(L_EN_GB, actual.get(0).getLocale());
+ Assert.assertEquals(Q0_500, actual.get(0).getQuality(), 0.0001);
+ }
+
+ /*
+ * Malformed tags are either reduced to what could be parsed or dropped, as
+ * documented for this parser.
+ */
+ @Test
+ public void testMalformed() throws Exception {
+ // Invalid primary subtags give Locale.ROOT
+ Assert.assertEquals(Locale.ROOT, locale("a"));
+ Assert.assertEquals(Locale.ROOT, locale("abcdefghi-US"));
+
+ // A tag that is invalid after the language subtag keeps what was
parsed
+ Locale actual = locale("zh-Hans-");
+ Assert.assertEquals("zh", actual.getLanguage());
+ Assert.assertEquals("", actual.getScript());
+ Assert.assertEquals("", actual.getCountry());
+ Assert.assertEquals("en", locale("en-US!").getLanguage());
+ }
+
+ /*
+ * Accepted degradation: extensions are dropped, unlike
+ * Locale.forLanguageTag() which records them on the Locale. The script is
+ * dropped as well and the region is still parsed.
+ */
+ @Test
+ public void testExtensionsAreDropped() throws Exception {
+ Locale actual = locale("zh-Hans-CN-u-ca-buddhist");
+
+ Assert.assertEquals("", actual.getScript());
+ Assert.assertEquals("CN", actual.getCountry());
+ Assert.assertNull(actual.getExtension('u'));
+ Assert.assertEquals(Locale.forLanguageTag("zh-CN"), actual);
+
+ actual = locale("en-Latn-GB-x-priv");
+
+ Assert.assertEquals("", actual.getScript());
+ Assert.assertEquals("GB", actual.getCountry());
+ Assert.assertNull(actual.getExtension('x'));
+ }
+
+ /*
+ * The limit on the number of entries is enforced while the header is
+ * parsed.
+ */
+ @Test
+ public void testLimited() throws Exception {
+ List<AcceptLanguage> actual = AcceptLanguage2.parse(new
StringReader("zh-Hans-CN,en,fr,de"), 2);
+
+ Assert.assertEquals(2, actual.size());
+ Assert.assertEquals(Locale.forLanguageTag("zh-CN"),
actual.get(0).getLocale());
+ Assert.assertEquals(L_EN, actual.get(1).getLocale());
+
+ Assert.assertEquals(0, AcceptLanguage2.parse(new
StringReader("zh-Hans-CN,en"), 0).size());
+ }
+
+}
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
