Look up the locale decimal point at conversion time, not lexer construction (#5597)

* Look up the locale decimal point at conversion time, not lexer construction

The lexer read localeconv()->decimal_point once in its constructor and wrote
that character into token_buffer in place of '.'. The strtod fallback then
used the locale current at conversion time, so an LC_NUMERIC change in
between (parser callback, SAX handler, another thread) truncated the value
in release builds and fired the endptr assertion in debug builds.

token_buffer now always holds '.'. Only the strtof/strtod/strtold fallback
depends on the locale: it looks up the decimal point right before the call,
restores '.' afterwards, and repeats the conversion if the locale changed in
between. As a side effect, std::from_chars and Clinger's fast path now also
apply under locales whose decimal point is not '.'.

Fixes #5198

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

* Stop the strtod retry loop when the decimal point is unchanged

convert_float_locale_aware() repeated the conversion until strtod
consumed the whole token, assuming an early stop can only mean a locale
change. Under a locale whose decimal point is not a single character
(e.g. the two-byte U+066B of ar_EG.UTF-8, ar_SA.UTF-8, or fa_IR.UTF-8,
all available on macOS), the in-place substitution can never succeed,
so parsing any float that reaches the strtod fallback (for example
3.14159265358979323846 at C++11) hung forever. Before this branch, the
same input was truncated.

Retry only if the decimal point changed since the previous attempt;
otherwise keep the value strtod parsed so far, as before. Add a test
that parses such numbers under a multi-byte decimal point locale; it
hangs without this change.

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

* Fix -Weffc++ errors in the #5198 locale test

GCC's -Weffc++ (an error in ci_test_gcc and ci_test_standards_gcc)
rejected LocaleSwitchingSax: it has a pointer data member but does not
declare its copy operations, and its vectors are not initialized in the
member initializer list. Store the locale name as a std::string and give
the vectors brace initializers, like SaxEventLogger in
unit-deserialization.cpp.

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

---------

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

include/nlohmann/detail/input/number_parse.hpp

@@ -118,14 +118,12 @@ std::strtod. The parser only activates for number_float_t == double; float and
 long double keep the std::strtof/std::strtold paths (see the templated overload
 below).
 
-@param[in]  first          pointer to the first character of the number
-@param[in]  last           pointer past the last character
-@param[in]  decimal_point  the (locale-dependent) decimal point character
-@param[out] out            the parsed value on success
+@param[in]  first  pointer to the first character of the number
+@param[in]  last   pointer past the last character
+@param[out] out    the parsed value on success
 @return true if the value was parsed exactly; false to fall back to strtod
 */
-template<typename DecimalPointType>
-bool parse_float_fast(const char* first, const char* last, DecimalPointType decimal_point, double& out) noexcept
+inline bool parse_float_fast(const char* first, const char* last, double& out) noexcept
 {
 #if defined(FLT_EVAL_METHOD) && FLT_EVAL_METHOD != 0
     // Clinger's fast path is only exact when double operations are evaluated in
@@ -136,7 +134,6 @@ bool parse_float_fast(const char* first, const char* last, DecimalPointType deci
     // std::from_chars / std::strtod path.
     static_cast<void>(first);
     static_cast<void>(last);
-    static_cast<void>(decimal_point);
     static_cast<void>(out);
     return false;
 #else
@@ -175,7 +172,7 @@ bool parse_float_fast(const char* first, const char* last, DecimalPointType deci
             ++num_digits;
             fractional_digits += static_cast<int>(seen_dot);
         }
-        else if (static_cast<DecimalPointType>(c) == decimal_point)
+        else if (c == '.')
         {
             if (JSON_HEDLEY_UNLIKELY(seen_dot))
             {
@@ -260,8 +257,8 @@ bool parse_float_fast(const char* first, const char* last, DecimalPointType deci
 }
 
 /// fast float path is only exact for `double`; decline for float/long double
-template<typename DecimalPointType, typename FloatType>
-bool parse_float_fast(const char* /*first*/, const char* /*last*/, DecimalPointType /*decimal_point*/, FloatType& /*out*/) noexcept
+template<typename FloatType>
+bool parse_float_fast(const char* /*first*/, const char* /*last*/, FloatType& /*out*/) noexcept
 {
     return false;
 }
@@ -273,9 +270,7 @@ std::from_chars is locale-independent, correctly rounded, and - via the
 Eisel-Lemire algorithm in modern standard libraries - much faster than strtod
 over the whole value range (not just the Clinger subset). It is used only when
 __cpp_lib_to_chars indicates full floating-point support and only when it
-consumes the entire token ([first, last)); a partial parse means the buffer
-uses a non-'.' locale decimal point, in which case the caller falls back to the
-locale-aware path. An under-/overflow (result_out_of_range) also declines, so
+consumes the entire token ([first, last)). An under-/overflow (result_out_of_range) also declines, so
 the caller's strtod fallback supplies the well-defined ±inf/0 result the parser
 expects (side-stepping the P4168 divergence between implementations).