[Intl] Validate extension keys

- Fix ParseBCP47 to return just the extension. The second argument to
  substr function is actually the length, not the end position of the
  substring :')
- Remove extension keys that aren't part of the relevant extension keys
- Use 'ca' as an relevant extension key for DateTimeFormat
- Use the canonicalized locale tag after create the ICU locale is
  created, rather than the input locale tag.

Bug: chromium:895942, v8:5751
Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng
Change-Id: I189709714d847e684b04409b734a60ff04ed7dd2
Reviewed-on: https://chromium-review.googlesource.com/c/1291076Reviewed-by: Frank Tang <ftang@chromium.org>
Commit-Queue: Sathya Gunasekaran <gsathya@chromium.org>
Cr-Commit-Position: refs/heads/master@{#57274}

src/objects/intl-objects.cc

@@ -1445,8 +1445,8 @@ ParsedLocale ParseBCP47Locale(const std::string& locale) {
 
   const std::string end = locale.substr(unicode_extension_end);
   parsed_locale.no_extensions_locale = beginning + end;
-  parsed_locale.extension =
-      locale.substr(unicode_extension_start, unicode_extension_end);
+  parsed_locale.extension = locale.substr(
+      unicode_extension_start, unicode_extension_end - unicode_extension_start);
   return parsed_locale;
 }
 
@@ -1630,13 +1630,13 @@ MaybeHandle<JSObject> Intl::SupportedLocalesOf(
 
 namespace {
 
-std::map<std::string, std::string> LookupUnicodeExtensions(
-    const icu::Locale& icu_locale, const std::set<std::string>& relevant_keys) {
+std::map<std::string, std::string> LookupAndValidateUnicodeExtensions(
+    icu::Locale* icu_locale, const std::set<std::string>& relevant_keys) {
   std::map<std::string, std::string> extensions;
 
   UErrorCode status = U_ZERO_ERROR;
   std::unique_ptr<icu::StringEnumeration> keywords(
-      icu_locale.createKeywords(status));
+      icu_locale->createKeywords(status));
   if (U_FAILURE(status)) return extensions;
 
   if (!keywords) return extensions;
@@ -1654,7 +1654,7 @@ std::map<std::string, std::string> LookupUnicodeExtensions(
       continue;
     }
 
-    icu_locale.getKeywordValue(keyword, value, ULOC_FULLNAME_CAPACITY, status);
+    icu_locale->getKeywordValue(keyword, value, ULOC_FULLNAME_CAPACITY, status);
 
     // Ignore failures in ICU and skip to the next keyword.
     //
@@ -1666,11 +1666,14 @@ std::map<std::string, std::string> LookupUnicodeExtensions(
 
     const char* bcp47_key = uloc_toUnicodeLocaleKey(keyword);
 
-    // Ignore keywords that we don't recognize - spec allows that.
     if (bcp47_key && (relevant_keys.find(bcp47_key) != relevant_keys.end())) {
       const char* bcp47_value = uloc_toUnicodeLocaleType(bcp47_key, value);
       extensions.insert(
           std::pair<std::string, std::string>(bcp47_key, bcp47_value));
+    } else {
+      status = U_ZERO_ERROR;
+      icu_locale->setKeywordValue(keyword, nullptr, status);
+      CHECK(U_SUCCESS(status));
     }
   }
 
@@ -1751,11 +1754,17 @@ Intl::ResolvedLocale Intl::ResolveLocale(
 
   icu::Locale icu_locale = CreateICULocale(locale);
   std::map<std::string, std::string> extensions =
-      LookupUnicodeExtensions(icu_locale, relevant_extension_keys);
+      LookupAndValidateUnicodeExtensions(&icu_locale, relevant_extension_keys);
+
+  char canonicalized_locale[ULOC_FULLNAME_CAPACITY];
+  UErrorCode status = U_ZERO_ERROR;
+  uloc_toLanguageTag(icu_locale.getName(), canonicalized_locale,
+                     ULOC_FULLNAME_CAPACITY, true, &status);
+  CHECK(U_SUCCESS(status));
 
   // TODO(gsathya): Remove privateuse subtags from extensions.
 
-  return Intl::ResolvedLocale{locale, icu_locale, extensions};
+  return Intl::ResolvedLocale{canonicalized_locale, icu_locale, extensions};
 }
 
 Managed<icu::UnicodeString>* Intl::SetTextToBreakIterator(

src/objects/js-date-time-format.cc

@@ -798,13 +798,22 @@ MaybeHandle<JSDateTimeFormat> JSDateTimeFormat::Initialize(
   // 9. Set opt.[[hc]] to hourCycle.
   // TODO(ftang): change behavior based on hour_cycle.
 
+  // ecma402/#sec-intl.datetimeformat-internal-slots
+  // The value of the [[RelevantExtensionKeys]] internal slot is
+  // « "ca", "nu", "hc" ».
+  //
+  // TODO(ftang): Add "hc" to this list of keys:
+  // https://bugs.chromium.org/p/v8/issues/detail?id=7482
+  std::set<std::string> relevant_extension_keys = {"nu", "ca"};
+
   // 10. Let localeData be %DateTimeFormat%.[[LocaleData]].
   // 11. Let r be ResolveLocale( %DateTimeFormat%.[[AvailableLocales]],
   //     requestedLocales, opt, %DateTimeFormat%.[[RelevantExtensionKeys]],
   //     localeData).
-  Intl::ResolvedLocale r =
-      Intl::ResolveLocale(isolate, JSDateTimeFormat::GetAvailableLocales(),
-                          requested_locales, locale_matcher, {"nu"});
+  //
+  Intl::ResolvedLocale r = Intl::ResolveLocale(
+      isolate, JSDateTimeFormat::GetAvailableLocales(), requested_locales,
+      locale_matcher, relevant_extension_keys);
 
   // TODO(ftang): Make sure that "nu" key doesn't have "native",
   // "traditio" or "finance" values.

test/intl/regress-895942.js

+// Copyright 2018 the V8 project authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file.
+
+x = new Intl.DateTimeFormat("en-u-foo-x-u");
+assertEquals('en', x.resolvedOptions().locale);