[libcxx] [test] Fix the thousands_sep expectation for fr_FR locales on Windows

[mstorsjo]

Summary

Windows uses \u00A0 (No-Break Space) as thousands separator in the
fr_FR locale.

[Mordante]

LGTM, the CI error seems unrelated.

[• Quuxplusone]

LGTM%comment!

libcxx/test/std/localization/locale.categories/category.monetary/locale.money.get/locale.money.get.members/get_long_double_fr_FR.pass.cpp

55–58

The original commenter seems to have mixed up U+202F NARROW NO-BREAK SPACE with U+2027 HYPHENATION POINT. I don't imagine that any platform would use U+2027 HYPHENATION POINT as its thousands separator.

libcxx/test/std/localization/locale.categories/category.monetary/locale.money.put/locale.money.put.members/put_long_double_fr_FR.pass.cpp

55–58

Same deal with this comment, please.

libcxx/test/std/localization/locale.categories/category.monetary/locale.moneypunct.byname/thousands_sep.pass.cpp

140

Pre-existing: I find it unlikely that we have to "work around" the use of U+002E FULL STOP as a decimal point character. ;) That's... normal. :)

libcxx/test/std/localization/locale.categories/category.monetary/locale.money.get/locale.money.get.members/get_long_double_fr_FR.pass.cpp

 //===----------------------------------------------------------------------===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
 // XFAIL: darwin
 
 // NetBSD does not support LC_MONETARY at the moment
 // XFAIL: netbsd
 
-// XFAIL: LIBCXX-WINDOWS-FIXME
 // XFAIL: LIBCXX-AIX-FIXME
 
 // REQUIRES: locale.fr_FR.UTF-8
 
 // <locale>
 
 // class money_get<charT, InputIterator>
 

 class my_facetw
     : public Fw
 {
 public:
     explicit my_facetw(std::size_t refs = 0)
         : Fw(refs) {}
 };
 
 // GLIBC 2.27 and newer use U2027 (narrow non-breaking space) as a thousands sep.
 // this function converts the spaces in string inputs to that character if need
 // be. FreeBSD's locale data also uses U2027 since 2018.
+// Windows uses U00A0.
 static std::wstring convert_thousands_sep(std::wstring const& in) {
-#if defined(_CS_GNU_LIBC_VERSION) || defined(__FreeBSD__)
+#if defined(_CS_GNU_LIBC_VERSION) || defined(__FreeBSD__) || defined(_WIN32)
 #if defined(_CS_GNU_LIBC_VERSION)
   if (glibc_version_less_than("2.27"))
     return in;
 #endif
   std::wstring out;
   unsigned I = 0;
   bool seen_decimal = false;
   for (; I < in.size(); ++I) {
     if (seen_decimal || in[I] != L' ') {
       seen_decimal |= in[I] == L',';
       out.push_back(in[I]);
       continue;
     }
     assert(in[I] == L' ');
+#if defined(_WIN32)
+    out.push_back(L'\u00A0');
+#else
     out.push_back(L'\u202F');
+#endif
   }
   return out;
 #else
   return in;
 #endif
 }
 #endif // TEST_HAS_NO_WIDE_CHARACTERS
 

libcxx/test/std/localization/locale.categories/category.monetary/locale.money.put/locale.money.put.members/put_long_double_fr_FR.pass.cpp

 //===----------------------------------------------------------------------===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
 // XFAIL: darwin
 
 // NetBSD does not support LC_MONETARY at the moment
 // XFAIL: netbsd
 
-// XFAIL: LIBCXX-WINDOWS-FIXME
 // XFAIL: LIBCXX-AIX-FIXME
 
 // REQUIRES: locale.fr_FR.UTF-8
 
 // <locale>
 
 // class money_put<charT, OutputIterator>
 

 class my_facetw
     : public Fw
 {
 public:
     explicit my_facetw(std::size_t refs = 0)
         : Fw(refs) {}
 };
 
 // GLIBC 2.27 and newer use U2027 (narrow non-breaking space) as a thousands sep.
 // this function converts the spaces in string inputs to that character if need
 // be. FreeBSD's locale data also uses U2027 since 2018.
+// Windows uses U00A0.
 static std::wstring convert_thousands_sep(std::wstring const& in) {
-#if defined(_CS_GNU_LIBC_VERSION) || defined(__FreeBSD__)
+#if defined(_CS_GNU_LIBC_VERSION) || defined(__FreeBSD__) || defined(_WIN32)
 #if defined(_CS_GNU_LIBC_VERSION)
   if (glibc_version_less_than("2.27"))
     return in;
 #endif
   std::wstring out;
   unsigned I = 0;
   bool seen_num_start = false;
   bool seen_decimal = false;
   for (; I < in.size(); ++I) {
     seen_decimal |= in[I] == L',';
     seen_num_start |= in[I] == '-' || std::iswdigit(in[I]);
     if (seen_decimal || !seen_num_start || in[I] != L' ') {
       out.push_back(in[I]);
       continue;
     }
     assert(in[I] == L' ');
+#if defined(_WIN32)
+    out.push_back(L'\u00A0');
+#else
     out.push_back(L'\u202F');
+#endif
   }
   return out;
 #else
   return in;
 #endif
 }
 #endif // TEST_HAS_NO_WIDE_CHARACTERS
 

libcxx/test/std/localization/locale.categories/category.monetary/locale.moneypunct.byname/thousands_sep.pass.cpp

 //===----------------------------------------------------------------------===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
 // NetBSD does not support LC_MONETARY at the moment
 // XFAIL: netbsd
 
-// XFAIL: LIBCXX-WINDOWS-FIXME
-
 // REQUIRES: locale.en_US.UTF-8
 // REQUIRES: locale.fr_FR.UTF-8
 // REQUIRES: locale.ru_RU.UTF-8
 // REQUIRES: locale.zh_CN.UTF-8
 
 // <locale>
 
 // class moneypunct_byname<charT, International>
#endif
     {
         Fnt f(LOCALE_fr_FR_UTF_8, 1);
         assert(f.thousands_sep() == ' ');
     }
     // The below tests work around GLIBC's use of U202F as mon_thousands_sep.
 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
 #if defined(_CS_GNU_LIBC_VERSION)
     const wchar_t fr_sep = glibc_version_less_than("2.27") ? L' ' : L'\u202F';
+#elif defined(_WIN32)
+    const wchar_t fr_sep = L'\u00A0';
 #else
     const wchar_t fr_sep = L' ';
 #endif
     {
         Fwf f(LOCALE_fr_FR_UTF_8, 1);
         assert(f.thousands_sep() == fr_sep);
     }
     {
         Fwt f(LOCALE_fr_FR_UTF_8, 1);
         assert(f.thousands_sep() == fr_sep);
     }
 #endif // TEST_HAS_NO_WIDE_CHARACTERS
     const char sep = ' ';
     {
         Fnf f(LOCALE_ru_RU_UTF_8, 1);
         assert(f.thousands_sep() == sep);
     }
     {
         Fnt f(LOCALE_ru_RU_UTF_8, 1);
         assert(f.thousands_sep() == sep);
     }
 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
     // The below tests work around GLIBC's use of U00A0 as mon_thousands_sep
     // and U002E as mon_decimal_point.
     // TODO: Fix thousands_sep for 'char'.
     // related to https://gcc.gnu.org/bugzilla/show_bug.cgi?id=16006
 #   if defined(_CS_GNU_LIBC_VERSION)
     // FIXME libc++ specifically works around \u00A0 by translating it into
     // a regular space.
     const wchar_t wsep = glibc_version_less_than("2.27") ? L'\u00A0' : L'\u202F';
+#   elif defined(_WIN32)
+    const wchar_t wsep = L'\u00A0';
 #   else
     const wchar_t wsep = L' ';
 #   endif
     {
         Fwf f(LOCALE_ru_RU_UTF_8, 1);
         assert(f.thousands_sep() == wsep);
     }
     {

libcxx/test/std/localization/locale.categories/facet.numpunct/locale.numpunct.byname/thousands_sep.pass.cpp

 //===----------------------------------------------------------------------===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
 // NetBSD does not support LC_NUMERIC at the moment
 // XFAIL: netbsd
 
-// XFAIL: LIBCXX-WINDOWS-FIXME
 // XFAIL: LIBCXX-AIX-FIXME
 
 // REQUIRES: locale.en_US.UTF-8
 // REQUIRES: locale.fr_FR.UTF-8
 
 // <locale>
 
 // template <class charT> class numpunct_byname;
#ifndef TEST_HAS_NO_WIDE_CHARACTERS
             assert(np.thousands_sep() == L',');
         }
 #endif
     }
     {
         // The below tests work around GLIBC's use of U202F as LC_NUMERIC thousands_sep.
         std::locale l(LOCALE_fr_FR_UTF_8);
         {
-#if defined(_CS_GNU_LIBC_VERSION)
+#if defined(_CS_GNU_LIBC_VERSION) || defined(_WIN32)
             const char sep = ' ';
 #else
             const char sep = ',';
 #endif
             typedef char C;
             const std::numpunct<C>& np = std::use_facet<std::numpunct<C> >(l);
             assert(np.thousands_sep() == sep);
         }
 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
         {
 #if defined(_CS_GNU_LIBC_VERSION)
             const wchar_t wsep = glibc_version_less_than("2.27") ? L' ' : L'\u202f';
+#elif defined(_WIN32)
+            const wchar_t wsep = L'\u00A0';
 #else
             const wchar_t wsep = L',';
 #endif
             typedef wchar_t C;
             const std::numpunct<C>& np = std::use_facet<std::numpunct<C> >(l);
             assert(np.thousands_sep() == wsep);
         }
 #endif // TEST_HAS_NO_WIDE_CHARACTERS
     }
 
     return 0;
 }