Index: src/locale.cpp =================================================================== --- src/locale.cpp +++ src/locale.cpp @@ -1,6262 +1,6109 @@ -//===------------------------- locale.cpp ---------------------------------===// -// -// The LLVM Compiler Infrastructure -// -// This file is dual licensed under the MIT and the University of Illinois Open -// Source Licenses. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - -// On Solaris, we need to define something to make the C99 parts of localeconv -// visible. -#ifdef __sun__ -#define _LCONV_C99 -#endif - -#include "string" -#include "locale" -#include "codecvt" -#include "vector" -#include "algorithm" -#include "typeinfo" -#ifndef _LIBCPP_NO_EXCEPTIONS -# include "type_traits" -#endif -#include "clocale" -#include "cstring" -#include "cwctype" -#include "__sso_allocator" -#if defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) -#include -#elif !defined(__ANDROID__) -#include -#endif -#include -#include - -// On Linux, wint_t and wchar_t have different signed-ness, and this causes -// lots of noise in the build log, but no bugs that I know of. -#if defined(__clang__) -#pragma clang diagnostic ignored "-Wsign-conversion" -#endif - -_LIBCPP_BEGIN_NAMESPACE_STD - -#ifdef __cloc_defined -locale_t __cloc() { - // In theory this could create a race condition. In practice - // the race condition is non-fatal since it will just create - // a little resource leak. Better approach would be appreciated. - static locale_t result = newlocale(LC_ALL_MASK, "C", 0); - return result; -} -#endif // __cloc_defined - -namespace { - -struct release -{ - void operator()(locale::facet* p) {p->__release_shared();} -}; - -template -inline -T& -make(A0 a0) -{ - static typename aligned_storage::type buf; - ::new (&buf) T(a0); - return *reinterpret_cast(&buf); -} - -template -inline -T& -make(A0 a0, A1 a1) -{ - static typename aligned_storage::type buf; - ::new (&buf) T(a0, a1); - return *reinterpret_cast(&buf); -} - -template -inline -T& -make(A0 a0, A1 a1, A2 a2) -{ - static typename aligned_storage::type buf; - ::new (&buf) T(a0, a1, a2); - return *reinterpret_cast(&buf); -} - -template -inline -_LIBCPP_CONSTEXPR -size_t -countof(const T (&)[N]) -{ - return N; -} - -template -inline -_LIBCPP_CONSTEXPR -size_t -countof(const T * const begin, const T * const end) -{ - return static_cast(end - begin); -} - -} - -#if defined(_AIX) -// Set priority to INT_MIN + 256 + 150 -# pragma priority ( -2147483242 ) -#endif - -const locale::category locale::none; -const locale::category locale::collate; -const locale::category locale::ctype; -const locale::category locale::monetary; -const locale::category locale::numeric; -const locale::category locale::time; -const locale::category locale::messages; -const locale::category locale::all; - -#if defined(__clang__) -#pragma clang diagnostic push -#pragma clang diagnostic ignored "-Wpadded" -#endif - -class _LIBCPP_HIDDEN locale::__imp - : public facet -{ - enum {N = 28}; -#if defined(_LIBCPP_MSVC) -// FIXME: MSVC doesn't support aligned parameters by value. -// I can't get the __sso_allocator to work here -// for MSVC I think for this reason. - vector facets_; -#else - vector > facets_; -#endif - string name_; -public: - explicit __imp(size_t refs = 0); - explicit __imp(const string& name, size_t refs = 0); - __imp(const __imp&); - __imp(const __imp&, const string&, locale::category c); - __imp(const __imp& other, const __imp& one, locale::category c); - __imp(const __imp&, facet* f, long id); - ~__imp(); - - const string& name() const {return name_;} - bool has_facet(long id) const - {return static_cast(id) < facets_.size() && facets_[static_cast(id)];} - const locale::facet* use_facet(long id) const; - - static const locale& make_classic(); - static locale& make_global(); -private: - void install(facet* f, long id); - template void install(F* f) {install(f, f->id.__get());} - template void install_from(const __imp& other); -}; - -#if defined(__clang__) -#pragma clang diagnostic pop -#endif - -locale::__imp::__imp(size_t refs) - : facet(refs), - facets_(N), - name_("C") -{ - facets_.clear(); - install(&make<_VSTD::collate >(1u)); - install(&make<_VSTD::collate >(1u)); - install(&make<_VSTD::ctype >(nullptr, false, 1u)); - install(&make<_VSTD::ctype >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make >(1u)); - install(&make<_VSTD::messages >(1u)); - install(&make<_VSTD::messages >(1u)); -} - -locale::__imp::__imp(const string& name, size_t refs) - : facet(refs), - facets_(N), - name_(name) -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - try - { -#endif // _LIBCPP_NO_EXCEPTIONS - facets_ = locale::classic().__locale_->facets_; - for (unsigned i = 0; i < facets_.size(); ++i) - if (facets_[i]) - facets_[i]->__add_shared(); - install(new collate_byname(name_)); - install(new collate_byname(name_)); - install(new ctype_byname(name_)); - install(new ctype_byname(name_)); - install(new codecvt_byname(name_)); - install(new codecvt_byname(name_)); - install(new codecvt_byname(name_)); - install(new codecvt_byname(name_)); - install(new numpunct_byname(name_)); - install(new numpunct_byname(name_)); - install(new moneypunct_byname(name_)); - install(new moneypunct_byname(name_)); - install(new moneypunct_byname(name_)); - install(new moneypunct_byname(name_)); - install(new time_get_byname(name_)); - install(new time_get_byname(name_)); - install(new time_put_byname(name_)); - install(new time_put_byname(name_)); - install(new messages_byname(name_)); - install(new messages_byname(name_)); -#ifndef _LIBCPP_NO_EXCEPTIONS - } - catch (...) - { - for (unsigned i = 0; i < facets_.size(); ++i) - if (facets_[i]) - facets_[i]->__release_shared(); - throw; - } -#endif // _LIBCPP_NO_EXCEPTIONS -} - -// NOTE avoid the `base class should be explicitly initialized in the -// copy constructor` warning emitted by GCC -#if defined(__clang__) || _GNUC_VER >= 406 -#pragma GCC diagnostic push -#pragma GCC diagnostic ignored "-Wextra" -#endif - -locale::__imp::__imp(const __imp& other) - : facets_(max(N, other.facets_.size())), - name_(other.name_) -{ - facets_ = other.facets_; - for (unsigned i = 0; i < facets_.size(); ++i) - if (facets_[i]) - facets_[i]->__add_shared(); -} - -#if defined(__clang__) || _GNUC_VER >= 406 -#pragma GCC diagnostic pop -#endif - -locale::__imp::__imp(const __imp& other, const string& name, locale::category c) - : facets_(N), - name_("*") -{ - facets_ = other.facets_; - for (unsigned i = 0; i < facets_.size(); ++i) - if (facets_[i]) - facets_[i]->__add_shared(); -#ifndef _LIBCPP_NO_EXCEPTIONS - try - { -#endif // _LIBCPP_NO_EXCEPTIONS - if (c & locale::collate) - { - install(new collate_byname(name)); - install(new collate_byname(name)); - } - if (c & locale::ctype) - { - install(new ctype_byname(name)); - install(new ctype_byname(name)); - install(new codecvt_byname(name)); - install(new codecvt_byname(name)); - install(new codecvt_byname(name)); - install(new codecvt_byname(name)); - } - if (c & locale::monetary) - { - install(new moneypunct_byname(name)); - install(new moneypunct_byname(name)); - install(new moneypunct_byname(name)); - install(new moneypunct_byname(name)); - } - if (c & locale::numeric) - { - install(new numpunct_byname(name)); - install(new numpunct_byname(name)); - } - if (c & locale::time) - { - install(new time_get_byname(name)); - install(new time_get_byname(name)); - install(new time_put_byname(name)); - install(new time_put_byname(name)); - } - if (c & locale::messages) - { - install(new messages_byname(name)); - install(new messages_byname(name)); - } -#ifndef _LIBCPP_NO_EXCEPTIONS - } - catch (...) - { - for (unsigned i = 0; i < facets_.size(); ++i) - if (facets_[i]) - facets_[i]->__release_shared(); - throw; - } -#endif // _LIBCPP_NO_EXCEPTIONS -} - -template -inline -void -locale::__imp::install_from(const locale::__imp& one) -{ - long id = F::id.__get(); - install(const_cast(static_cast(one.use_facet(id))), id); -} - -locale::__imp::__imp(const __imp& other, const __imp& one, locale::category c) - : facets_(N), - name_("*") -{ - facets_ = other.facets_; - for (unsigned i = 0; i < facets_.size(); ++i) - if (facets_[i]) - facets_[i]->__add_shared(); -#ifndef _LIBCPP_NO_EXCEPTIONS - try - { -#endif // _LIBCPP_NO_EXCEPTIONS - if (c & locale::collate) - { - install_from<_VSTD::collate >(one); - install_from<_VSTD::collate >(one); - } - if (c & locale::ctype) - { - install_from<_VSTD::ctype >(one); - install_from<_VSTD::ctype >(one); - install_from<_VSTD::codecvt >(one); - install_from<_VSTD::codecvt >(one); - install_from<_VSTD::codecvt >(one); - install_from<_VSTD::codecvt >(one); - } - if (c & locale::monetary) - { - install_from >(one); - install_from >(one); - install_from >(one); - install_from >(one); - install_from >(one); - install_from >(one); - install_from >(one); - install_from >(one); - } - if (c & locale::numeric) - { - install_from >(one); - install_from >(one); - install_from >(one); - install_from >(one); - install_from >(one); - install_from >(one); - } - if (c & locale::time) - { - install_from >(one); - install_from >(one); - install_from >(one); - install_from >(one); - } - if (c & locale::messages) - { - install_from<_VSTD::messages >(one); - install_from<_VSTD::messages >(one); - } -#ifndef _LIBCPP_NO_EXCEPTIONS - } - catch (...) - { - for (unsigned i = 0; i < facets_.size(); ++i) - if (facets_[i]) - facets_[i]->__release_shared(); - throw; - } -#endif // _LIBCPP_NO_EXCEPTIONS -} - -locale::__imp::__imp(const __imp& other, facet* f, long id) - : facets_(max(N, other.facets_.size()+1)), - name_("*") -{ - f->__add_shared(); - unique_ptr hold(f); - facets_ = other.facets_; - for (unsigned i = 0; i < other.facets_.size(); ++i) - if (facets_[i]) - facets_[i]->__add_shared(); - install(hold.get(), id); -} - -locale::__imp::~__imp() -{ - for (unsigned i = 0; i < facets_.size(); ++i) - if (facets_[i]) - facets_[i]->__release_shared(); -} - -void -locale::__imp::install(facet* f, long id) -{ - f->__add_shared(); - unique_ptr hold(f); - if (static_cast(id) >= facets_.size()) - facets_.resize(static_cast(id+1)); - if (facets_[static_cast(id)]) - facets_[static_cast(id)]->__release_shared(); - facets_[static_cast(id)] = hold.release(); -} - -const locale::facet* -locale::__imp::use_facet(long id) const -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - if (!has_facet(id)) - throw bad_cast(); -#endif // _LIBCPP_NO_EXCEPTIONS - return facets_[static_cast(id)]; -} - -// locale - -const locale& -locale::__imp::make_classic() -{ - // only one thread can get in here and it only gets in once - static aligned_storage::type buf; - locale* c = reinterpret_cast(&buf); - c->__locale_ = &make<__imp>(1u); - return *c; -} - -const locale& -locale::classic() -{ - static const locale& c = __imp::make_classic(); - return c; -} - -locale& -locale::__imp::make_global() -{ - // only one thread can get in here and it only gets in once - static aligned_storage::type buf; - ::new (&buf) locale(locale::classic()); - return *reinterpret_cast(&buf); -} - -locale& -locale::__global() -{ - static locale& g = __imp::make_global(); - return g; -} - -locale::locale() _NOEXCEPT - : __locale_(__global().__locale_) -{ - __locale_->__add_shared(); -} - -locale::locale(const locale& l) _NOEXCEPT - : __locale_(l.__locale_) -{ - __locale_->__add_shared(); -} - -locale::~locale() -{ - __locale_->__release_shared(); -} - -const locale& -locale::operator=(const locale& other) _NOEXCEPT -{ - other.__locale_->__add_shared(); - __locale_->__release_shared(); - __locale_ = other.__locale_; - return *this; -} - -locale::locale(const char* name) -#ifndef _LIBCPP_NO_EXCEPTIONS - : __locale_(name ? new __imp(name) - : throw runtime_error("locale constructed with null")) -#else // _LIBCPP_NO_EXCEPTIONS - : __locale_(new __imp(name)) -#endif -{ - __locale_->__add_shared(); -} - -locale::locale(const string& name) - : __locale_(new __imp(name)) -{ - __locale_->__add_shared(); -} - -locale::locale(const locale& other, const char* name, category c) -#ifndef _LIBCPP_NO_EXCEPTIONS - : __locale_(name ? new __imp(*other.__locale_, name, c) - : throw runtime_error("locale constructed with null")) -#else // _LIBCPP_NO_EXCEPTIONS - : __locale_(new __imp(*other.__locale_, name, c)) -#endif -{ - __locale_->__add_shared(); -} - -locale::locale(const locale& other, const string& name, category c) - : __locale_(new __imp(*other.__locale_, name, c)) -{ - __locale_->__add_shared(); -} - -locale::locale(const locale& other, const locale& one, category c) - : __locale_(new __imp(*other.__locale_, *one.__locale_, c)) -{ - __locale_->__add_shared(); -} - -string -locale::name() const -{ - return __locale_->name(); -} - -void -locale::__install_ctor(const locale& other, facet* f, long id) -{ - if (f) - __locale_ = new __imp(*other.__locale_, f, id); - else - __locale_ = other.__locale_; - __locale_->__add_shared(); -} - -locale -locale::global(const locale& loc) -{ - locale& g = __global(); - locale r = g; - g = loc; - if (g.name() != "*") - setlocale(LC_ALL, g.name().c_str()); - return r; -} - -bool -locale::has_facet(id& x) const -{ - return __locale_->has_facet(x.__get()); -} - -const locale::facet* -locale::use_facet(id& x) const -{ - return __locale_->use_facet(x.__get()); -} - -bool -locale::operator==(const locale& y) const -{ - return (__locale_ == y.__locale_) - || (__locale_->name() != "*" && __locale_->name() == y.__locale_->name()); -} - -// locale::facet - -locale::facet::~facet() -{ -} - -void -locale::facet::__on_zero_shared() _NOEXCEPT -{ - delete this; -} - -// locale::id - -int32_t locale::id::__next_id = 0; - -namespace -{ - -class __fake_bind -{ - locale::id* id_; - void (locale::id::* pmf_)(); -public: - __fake_bind(void (locale::id::* pmf)(), locale::id* id) - : id_(id), pmf_(pmf) {} - - void operator()() const - { - (id_->*pmf_)(); - } -}; - -} - -long -locale::id::__get() -{ - call_once(__flag_, __fake_bind(&locale::id::__init, this)); - return __id_ - 1; -} - -void -locale::id::__init() -{ - __id_ = __sync_add_and_fetch(&__next_id, 1); -} - -// template <> class collate_byname - -collate_byname::collate_byname(const char* n, size_t refs) - : collate(refs), - __l(newlocale(LC_ALL_MASK, n, 0)) -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - if (__l == 0) - throw runtime_error("collate_byname::collate_byname" - " failed to construct for " + string(n)); -#endif // _LIBCPP_NO_EXCEPTIONS -} - -collate_byname::collate_byname(const string& name, size_t refs) - : collate(refs), - __l(newlocale(LC_ALL_MASK, name.c_str(), 0)) -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - if (__l == 0) - throw runtime_error("collate_byname::collate_byname" - " failed to construct for " + name); -#endif // _LIBCPP_NO_EXCEPTIONS -} - -collate_byname::~collate_byname() -{ - freelocale(__l); -} - -int -collate_byname::do_compare(const char_type* __lo1, const char_type* __hi1, - const char_type* __lo2, const char_type* __hi2) const -{ - string_type lhs(__lo1, __hi1); - string_type rhs(__lo2, __hi2); - int r = strcoll_l(lhs.c_str(), rhs.c_str(), __l); - if (r < 0) - return -1; - if (r > 0) - return 1; - return r; -} - -collate_byname::string_type -collate_byname::do_transform(const char_type* lo, const char_type* hi) const -{ - const string_type in(lo, hi); - string_type out(strxfrm_l(0, in.c_str(), 0, __l), char()); - strxfrm_l(const_cast(out.c_str()), in.c_str(), out.size()+1, __l); - return out; -} - -// template <> class collate_byname - -collate_byname::collate_byname(const char* n, size_t refs) - : collate(refs), - __l(newlocale(LC_ALL_MASK, n, 0)) -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - if (__l == 0) - throw runtime_error("collate_byname::collate_byname(size_t refs)" - " failed to construct for " + string(n)); -#endif // _LIBCPP_NO_EXCEPTIONS -} - -collate_byname::collate_byname(const string& name, size_t refs) - : collate(refs), - __l(newlocale(LC_ALL_MASK, name.c_str(), 0)) -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - if (__l == 0) - throw runtime_error("collate_byname::collate_byname(size_t refs)" - " failed to construct for " + name); -#endif // _LIBCPP_NO_EXCEPTIONS -} - -collate_byname::~collate_byname() -{ - freelocale(__l); -} - -int -collate_byname::do_compare(const char_type* __lo1, const char_type* __hi1, - const char_type* __lo2, const char_type* __hi2) const -{ - string_type lhs(__lo1, __hi1); - string_type rhs(__lo2, __hi2); - int r = wcscoll_l(lhs.c_str(), rhs.c_str(), __l); - if (r < 0) - return -1; - if (r > 0) - return 1; - return r; -} - -collate_byname::string_type -collate_byname::do_transform(const char_type* lo, const char_type* hi) const -{ - const string_type in(lo, hi); - string_type out(wcsxfrm_l(0, in.c_str(), 0, __l), wchar_t()); - wcsxfrm_l(const_cast(out.c_str()), in.c_str(), out.size()+1, __l); - return out; -} - -// template <> class ctype; - -const ctype_base::mask ctype_base::space; -const ctype_base::mask ctype_base::print; -const ctype_base::mask ctype_base::cntrl; -const ctype_base::mask ctype_base::upper; -const ctype_base::mask ctype_base::lower; -const ctype_base::mask ctype_base::alpha; -const ctype_base::mask ctype_base::digit; -const ctype_base::mask ctype_base::punct; -const ctype_base::mask ctype_base::xdigit; -const ctype_base::mask ctype_base::blank; -const ctype_base::mask ctype_base::alnum; -const ctype_base::mask ctype_base::graph; - -locale::id ctype::id; - -ctype::~ctype() -{ -} - -bool -ctype::do_is(mask m, char_type c) const -{ - return isascii(c) ? (ctype::classic_table()[c] & m) != 0 : false; -} - -const wchar_t* -ctype::do_is(const char_type* low, const char_type* high, mask* vec) const -{ - for (; low != high; ++low, ++vec) - *vec = static_cast(isascii(*low) ? - ctype::classic_table()[*low] : 0); - return low; -} - -const wchar_t* -ctype::do_scan_is(mask m, const char_type* low, const char_type* high) const -{ - for (; low != high; ++low) - if (isascii(*low) && (ctype::classic_table()[*low] & m)) - break; - return low; -} - -const wchar_t* -ctype::do_scan_not(mask m, const char_type* low, const char_type* high) const -{ - for (; low != high; ++low) - if (!(isascii(*low) && (ctype::classic_table()[*low] & m))) - break; - return low; -} - -wchar_t -ctype::do_toupper(char_type c) const -{ -#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE - return isascii(c) ? _DefaultRuneLocale.__mapupper[c] : c; -#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__NetBSD__) - return isascii(c) ? ctype::__classic_upper_table()[c] : c; -#else - return (isascii(c) && iswlower_l(c, __cloc())) ? c-L'a'+L'A' : c; -#endif -} - -const wchar_t* -ctype::do_toupper(char_type* low, const char_type* high) const -{ - for (; low != high; ++low) -#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE - *low = isascii(*low) ? _DefaultRuneLocale.__mapupper[*low] : *low; -#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__NetBSD__) - *low = isascii(*low) ? ctype::__classic_upper_table()[*low] - : *low; -#else - *low = (isascii(*low) && islower_l(*low, __cloc())) ? (*low-L'a'+L'A') : *low; -#endif - return low; -} - -wchar_t -ctype::do_tolower(char_type c) const -{ -#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE - return isascii(c) ? _DefaultRuneLocale.__maplower[c] : c; -#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__NetBSD__) - return isascii(c) ? ctype::__classic_lower_table()[c] : c; -#else - return (isascii(c) && isupper_l(c, __cloc())) ? c-L'A'+'a' : c; -#endif -} - -const wchar_t* -ctype::do_tolower(char_type* low, const char_type* high) const -{ - for (; low != high; ++low) -#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE - *low = isascii(*low) ? _DefaultRuneLocale.__maplower[*low] : *low; -#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__NetBSD__) - *low = isascii(*low) ? ctype::__classic_lower_table()[*low] - : *low; -#else - *low = (isascii(*low) && isupper_l(*low, __cloc())) ? *low-L'A'+L'a' : *low; -#endif - return low; -} - -wchar_t -ctype::do_widen(char c) const -{ - return c; -} - -const char* -ctype::do_widen(const char* low, const char* high, char_type* dest) const -{ - for (; low != high; ++low, ++dest) - *dest = *low; - return low; -} - -char -ctype::do_narrow(char_type c, char dfault) const -{ - if (isascii(c)) - return static_cast(c); - return dfault; -} - -const wchar_t* -ctype::do_narrow(const char_type* low, const char_type* high, char dfault, char* dest) const -{ - for (; low != high; ++low, ++dest) - if (isascii(*low)) - *dest = static_cast(*low); - else - *dest = dfault; - return low; -} - -// template <> class ctype; - -locale::id ctype::id; - -ctype::ctype(const mask* tab, bool del, size_t refs) - : locale::facet(refs), - __tab_(tab), - __del_(del) -{ - if (__tab_ == 0) - __tab_ = classic_table(); -} - -ctype::~ctype() -{ - if (__tab_ && __del_) - delete [] __tab_; -} - -char -ctype::do_toupper(char_type c) const -{ -#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE - return isascii(c) ? - static_cast(_DefaultRuneLocale.__mapupper[static_cast(c)]) : c; -#elif defined(__NetBSD__) - return static_cast(__classic_upper_table()[static_cast(c)]); -#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) - return isascii(c) ? - static_cast(__classic_upper_table()[static_cast(c)]) : c; -#else - return (isascii(c) && islower_l(c, __cloc())) ? c-'a'+'A' : c; -#endif -} - -const char* -ctype::do_toupper(char_type* low, const char_type* high) const -{ - for (; low != high; ++low) -#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE - *low = isascii(*low) ? - static_cast(_DefaultRuneLocale.__mapupper[static_cast(*low)]) : *low; -#elif defined(__NetBSD__) - *low = static_cast(__classic_upper_table()[static_cast(*low)]); -#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) - *low = isascii(*low) ? - static_cast(__classic_upper_table()[static_cast(*low)]) : *low; -#else - *low = (isascii(*low) && islower_l(*low, __cloc())) ? *low-'a'+'A' : *low; -#endif - return low; -} - -char -ctype::do_tolower(char_type c) const -{ -#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE - return isascii(c) ? - static_cast(_DefaultRuneLocale.__maplower[static_cast(c)]) : c; -#elif defined(__NetBSD__) - return static_cast(__classic_lower_table()[static_cast(c)]); -#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__NetBSD__) - return isascii(c) ? - static_cast(__classic_lower_table()[static_cast(c)]) : c; -#else - return (isascii(c) && isupper_l(c, __cloc())) ? c-'A'+'a' : c; -#endif -} - -const char* -ctype::do_tolower(char_type* low, const char_type* high) const -{ - for (; low != high; ++low) -#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE - *low = isascii(*low) ? static_cast(_DefaultRuneLocale.__maplower[static_cast(*low)]) : *low; -#elif defined(__NetBSD__) - *low = static_cast(__classic_lower_table()[static_cast(*low)]); -#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) - *low = isascii(*low) ? static_cast(__classic_lower_table()[static_cast(*low)]) : *low; -#else - *low = (isascii(*low) && isupper_l(*low, __cloc())) ? *low-'A'+'a' : *low; -#endif - return low; -} - -char -ctype::do_widen(char c) const -{ - return c; -} - -const char* -ctype::do_widen(const char* low, const char* high, char_type* dest) const -{ - for (; low != high; ++low, ++dest) - *dest = *low; - return low; -} - -char -ctype::do_narrow(char_type c, char dfault) const -{ - if (isascii(c)) - return static_cast(c); - return dfault; -} - -const char* -ctype::do_narrow(const char_type* low, const char_type* high, char dfault, char* dest) const -{ - for (; low != high; ++low, ++dest) - if (isascii(*low)) - *dest = *low; - else - *dest = dfault; - return low; -} - -#ifdef __EMSCRIPTEN__ -extern "C" const unsigned short ** __ctype_b_loc(); -extern "C" const int ** __ctype_tolower_loc(); -extern "C" const int ** __ctype_toupper_loc(); -#endif - -#ifdef _LIBCPP_PROVIDES_DEFAULT_RUNE_TABLE -const ctype::mask* -ctype::classic_table() _NOEXCEPT -{ - static _LIBCPP_CONSTEXPR const ctype::mask builtin_table[table_size] = { - cntrl, cntrl, - cntrl, cntrl, - cntrl, cntrl, - cntrl, cntrl, - cntrl, cntrl | space | blank, - cntrl | space, cntrl | space, - cntrl | space, cntrl | space, - cntrl, cntrl, - cntrl, cntrl, - cntrl, cntrl, - cntrl, cntrl, - cntrl, cntrl, - cntrl, cntrl, - cntrl, cntrl, - cntrl, cntrl, - cntrl, cntrl, - space | blank | print, punct | print, - punct | print, punct | print, - punct | print, punct | print, - punct | print, punct | print, - punct | print, punct | print, - punct | print, punct | print, - punct | print, punct | print, - punct | print, punct | print, - digit | print | xdigit, digit | print | xdigit, - digit | print | xdigit, digit | print | xdigit, - digit | print | xdigit, digit | print | xdigit, - digit | print | xdigit, digit | print | xdigit, - digit | print | xdigit, digit | print | xdigit, - punct | print, punct | print, - punct | print, punct | print, - punct | print, punct | print, - punct | print, upper | xdigit | print | alpha, - upper | xdigit | print | alpha, upper | xdigit | print | alpha, - upper | xdigit | print | alpha, upper | xdigit | print | alpha, - upper | xdigit | print | alpha, upper | print | alpha, - upper | print | alpha, upper | print | alpha, - upper | print | alpha, upper | print | alpha, - upper | print | alpha, upper | print | alpha, - upper | print | alpha, upper | print | alpha, - upper | print | alpha, upper | print | alpha, - upper | print | alpha, upper | print | alpha, - upper | print | alpha, upper | print | alpha, - upper | print | alpha, upper | print | alpha, - upper | print | alpha, upper | print | alpha, - upper | print | alpha, punct | print, - punct | print, punct | print, - punct | print, punct | print, - punct | print, lower | xdigit | print | alpha, - lower | xdigit | print | alpha, lower | xdigit | print | alpha, - lower | xdigit | print | alpha, lower | xdigit | print | alpha, - lower | xdigit | print | alpha, lower | print | alpha, - lower | print | alpha, lower | print | alpha, - lower | print | alpha, lower | print | alpha, - lower | print | alpha, lower | print | alpha, - lower | print | alpha, lower | print | alpha, - lower | print | alpha, lower | print | alpha, - lower | print | alpha, lower | print | alpha, - lower | print | alpha, lower | print | alpha, - lower | print | alpha, lower | print | alpha, - lower | print | alpha, lower | print | alpha, - lower | print | alpha, punct | print, - punct | print, punct | print, - punct | print, cntrl, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 - }; - return builtin_table; -} -#else -const ctype::mask* -ctype::classic_table() _NOEXCEPT -{ -#if defined(__APPLE__) || defined(__FreeBSD__) - return _DefaultRuneLocale.__runetype; -#elif defined(__NetBSD__) - return _C_ctype_tab_ + 1; -#elif defined(__GLIBC__) - return __cloc()->__ctype_b; -#elif __sun__ - return __ctype_mask; -#elif defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) - return _ctype+1; // internal ctype mask table defined in msvcrt.dll -// This is assumed to be safe, which is a nonsense assumption because we're -// going to end up dereferencing it later... -#elif defined(__EMSCRIPTEN__) - return *__ctype_b_loc(); -#elif defined(_NEWLIB_VERSION) - // Newlib has a 257-entry table in ctype_.c, where (char)0 starts at [1]. - return _ctype_ + 1; -#elif defined(_AIX) - return (const unsigned int *)__lc_ctype_ptr->obj->mask; -#else - // Platform not supported: abort so the person doing the port knows what to - // fix -# warning ctype::classic_table() is not implemented - printf("ctype::classic_table() is not implemented\n"); - abort(); - return NULL; -#endif -} -#endif - -#if defined(__GLIBC__) -const int* -ctype::__classic_lower_table() _NOEXCEPT -{ - return __cloc()->__ctype_tolower; -} - -const int* -ctype::__classic_upper_table() _NOEXCEPT -{ - return __cloc()->__ctype_toupper; -} -#elif __NetBSD__ -const short* -ctype::__classic_lower_table() _NOEXCEPT -{ - return _C_tolower_tab_ + 1; -} - -const short* -ctype::__classic_upper_table() _NOEXCEPT -{ - return _C_toupper_tab_ + 1; -} - -#elif defined(__EMSCRIPTEN__) -const int* -ctype::__classic_lower_table() _NOEXCEPT -{ - return *__ctype_tolower_loc(); -} - -const int* -ctype::__classic_upper_table() _NOEXCEPT -{ - return *__ctype_toupper_loc(); -} -#endif // __GLIBC__ || __EMSCRIPTEN__ || __NETBSD__ - -// template <> class ctype_byname - -ctype_byname::ctype_byname(const char* name, size_t refs) - : ctype(0, false, refs), - __l(newlocale(LC_ALL_MASK, name, 0)) -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - if (__l == 0) - throw runtime_error("ctype_byname::ctype_byname" - " failed to construct for " + string(name)); -#endif // _LIBCPP_NO_EXCEPTIONS -} - -ctype_byname::ctype_byname(const string& name, size_t refs) - : ctype(0, false, refs), - __l(newlocale(LC_ALL_MASK, name.c_str(), 0)) -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - if (__l == 0) - throw runtime_error("ctype_byname::ctype_byname" - " failed to construct for " + name); -#endif // _LIBCPP_NO_EXCEPTIONS -} - -ctype_byname::~ctype_byname() -{ - freelocale(__l); -} - -char -ctype_byname::do_toupper(char_type c) const -{ - return static_cast(toupper_l(static_cast(c), __l)); -} - -const char* -ctype_byname::do_toupper(char_type* low, const char_type* high) const -{ - for (; low != high; ++low) - *low = static_cast(toupper_l(static_cast(*low), __l)); - return low; -} - -char -ctype_byname::do_tolower(char_type c) const -{ - return static_cast(tolower_l(static_cast(c), __l)); -} - -const char* -ctype_byname::do_tolower(char_type* low, const char_type* high) const -{ - for (; low != high; ++low) - *low = static_cast(tolower_l(static_cast(*low), __l)); - return low; -} - -// template <> class ctype_byname - -ctype_byname::ctype_byname(const char* name, size_t refs) - : ctype(refs), - __l(newlocale(LC_ALL_MASK, name, 0)) -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - if (__l == 0) - throw runtime_error("ctype_byname::ctype_byname" - " failed to construct for " + string(name)); -#endif // _LIBCPP_NO_EXCEPTIONS -} - -ctype_byname::ctype_byname(const string& name, size_t refs) - : ctype(refs), - __l(newlocale(LC_ALL_MASK, name.c_str(), 0)) -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - if (__l == 0) - throw runtime_error("ctype_byname::ctype_byname" - " failed to construct for " + name); -#endif // _LIBCPP_NO_EXCEPTIONS -} - -ctype_byname::~ctype_byname() -{ - freelocale(__l); -} - -bool -ctype_byname::do_is(mask m, char_type c) const -{ -#ifdef _LIBCPP_WCTYPE_IS_MASK - return static_cast(iswctype_l(c, m, __l)); -#else - bool result = false; - wint_t ch = static_cast(c); - if ((m & space) == space) result |= (iswspace_l(ch, __l) != 0); - if ((m & print) == print) result |= (iswprint_l(ch, __l) != 0); - if ((m & cntrl) == cntrl) result |= (iswcntrl_l(ch, __l) != 0); - if ((m & upper) == upper) result |= (iswupper_l(ch, __l) != 0); - if ((m & lower) == lower) result |= (iswlower_l(ch, __l) != 0); - if ((m & alpha) == alpha) result |= (iswalpha_l(ch, __l) != 0); - if ((m & digit) == digit) result |= (iswdigit_l(ch, __l) != 0); - if ((m & punct) == punct) result |= (iswpunct_l(ch, __l) != 0); - if ((m & xdigit) == xdigit) result |= (iswxdigit_l(ch, __l) != 0); - if ((m & blank) == blank) result |= (iswblank_l(ch, __l) != 0); - return result; -#endif -} - -const wchar_t* -ctype_byname::do_is(const char_type* low, const char_type* high, mask* vec) const -{ - for (; low != high; ++low, ++vec) - { - if (isascii(*low)) - *vec = static_cast(ctype::classic_table()[*low]); - else - { - *vec = 0; - wint_t ch = static_cast(*low); - if (iswspace_l(ch, __l)) - *vec |= space; -#ifndef _LIBCPP_CTYPE_MASK_IS_COMPOSITE_PRINT - if (iswprint_l(ch, __l)) - *vec |= print; -#endif - if (iswcntrl_l(ch, __l)) - *vec |= cntrl; - if (iswupper_l(ch, __l)) - *vec |= upper; - if (iswlower_l(ch, __l)) - *vec |= lower; -#ifndef _LIBCPP_CTYPE_MASK_IS_COMPOSITE_ALPHA - if (iswalpha_l(ch, __l)) - *vec |= alpha; -#endif - if (iswdigit_l(ch, __l)) - *vec |= digit; - if (iswpunct_l(ch, __l)) - *vec |= punct; -#ifndef _LIBCPP_CTYPE_MASK_IS_COMPOSITE_XDIGIT - if (iswxdigit_l(ch, __l)) - *vec |= xdigit; -#endif -#if !defined(__sun__) - if (iswblank_l(ch, __l)) - *vec |= blank; -#endif - } - } - return low; -} - -const wchar_t* -ctype_byname::do_scan_is(mask m, const char_type* low, const char_type* high) const -{ - for (; low != high; ++low) - { -#ifdef _LIBCPP_WCTYPE_IS_MASK - if (iswctype_l(*low, m, __l)) - break; -#else - wint_t ch = static_cast(*low); - if ((m & space) == space && iswspace_l(ch, __l)) break; - if ((m & print) == print && iswprint_l(ch, __l)) break; - if ((m & cntrl) == cntrl && iswcntrl_l(ch, __l)) break; - if ((m & upper) == upper && iswupper_l(ch, __l)) break; - if ((m & lower) == lower && iswlower_l(ch, __l)) break; - if ((m & alpha) == alpha && iswalpha_l(ch, __l)) break; - if ((m & digit) == digit && iswdigit_l(ch, __l)) break; - if ((m & punct) == punct && iswpunct_l(ch, __l)) break; - if ((m & xdigit) == xdigit && iswxdigit_l(ch, __l)) break; - if ((m & blank) == blank && iswblank_l(ch, __l)) break; -#endif - } - return low; -} - -const wchar_t* -ctype_byname::do_scan_not(mask m, const char_type* low, const char_type* high) const -{ - for (; low != high; ++low) - { -#ifdef _LIBCPP_WCTYPE_IS_MASK - if (!iswctype_l(*low, m, __l)) - break; -#else - wint_t ch = static_cast(*low); - if ((m & space) == space && iswspace_l(ch, __l)) continue; - if ((m & print) == print && iswprint_l(ch, __l)) continue; - if ((m & cntrl) == cntrl && iswcntrl_l(ch, __l)) continue; - if ((m & upper) == upper && iswupper_l(ch, __l)) continue; - if ((m & lower) == lower && iswlower_l(ch, __l)) continue; - if ((m & alpha) == alpha && iswalpha_l(ch, __l)) continue; - if ((m & digit) == digit && iswdigit_l(ch, __l)) continue; - if ((m & punct) == punct && iswpunct_l(ch, __l)) continue; - if ((m & xdigit) == xdigit && iswxdigit_l(ch, __l)) continue; - if ((m & blank) == blank && iswblank_l(ch, __l)) continue; - break; -#endif - } - return low; -} - -wchar_t -ctype_byname::do_toupper(char_type c) const -{ - return towupper_l(c, __l); -} - -const wchar_t* -ctype_byname::do_toupper(char_type* low, const char_type* high) const -{ - for (; low != high; ++low) - *low = towupper_l(*low, __l); - return low; -} - -wchar_t -ctype_byname::do_tolower(char_type c) const -{ - return towlower_l(c, __l); -} - -const wchar_t* -ctype_byname::do_tolower(char_type* low, const char_type* high) const -{ - for (; low != high; ++low) - *low = towlower_l(*low, __l); - return low; -} - -wchar_t -ctype_byname::do_widen(char c) const -{ -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - return btowc_l(c, __l); -#else - return __btowc_l(c, __l); -#endif -} - -const char* -ctype_byname::do_widen(const char* low, const char* high, char_type* dest) const -{ - for (; low != high; ++low, ++dest) -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - *dest = btowc_l(*low, __l); -#else - *dest = __btowc_l(*low, __l); -#endif - return low; -} - -char -ctype_byname::do_narrow(char_type c, char dfault) const -{ -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - int r = wctob_l(c, __l); -#else - int r = __wctob_l(c, __l); -#endif - return r != static_cast(WEOF) ? static_cast(r) : dfault; -} - -const wchar_t* -ctype_byname::do_narrow(const char_type* low, const char_type* high, char dfault, char* dest) const -{ - for (; low != high; ++low, ++dest) - { -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - int r = wctob_l(*low, __l); -#else - int r = __wctob_l(*low, __l); -#endif - *dest = r != static_cast(WEOF) ? static_cast(r) : dfault; - } - return low; -} - -// template <> class codecvt - -locale::id codecvt::id; - -codecvt::~codecvt() -{ -} - -codecvt::result -codecvt::do_out(state_type&, - const intern_type* frm, const intern_type*, const intern_type*& frm_nxt, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - frm_nxt = frm; - to_nxt = to; - return noconv; -} - -codecvt::result -codecvt::do_in(state_type&, - const extern_type* frm, const extern_type*, const extern_type*& frm_nxt, - intern_type* to, intern_type*, intern_type*& to_nxt) const -{ - frm_nxt = frm; - to_nxt = to; - return noconv; -} - -codecvt::result -codecvt::do_unshift(state_type&, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - to_nxt = to; - return noconv; -} - -int -codecvt::do_encoding() const _NOEXCEPT -{ - return 1; -} - -bool -codecvt::do_always_noconv() const _NOEXCEPT -{ - return true; -} - -int -codecvt::do_length(state_type&, - const extern_type* frm, const extern_type* end, size_t mx) const -{ - return static_cast(min(mx, static_cast(end-frm))); -} - -int -codecvt::do_max_length() const _NOEXCEPT -{ - return 1; -} - -// template <> class codecvt - -locale::id codecvt::id; - -codecvt::codecvt(size_t refs) - : locale::facet(refs), - __l(_LIBCPP_GET_C_LOCALE) -{ -} - -codecvt::codecvt(const char* nm, size_t refs) - : locale::facet(refs), - __l(newlocale(LC_ALL_MASK, nm, 0)) -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - if (__l == 0) - throw runtime_error("codecvt_byname::codecvt_byname" - " failed to construct for " + string(nm)); -#endif // _LIBCPP_NO_EXCEPTIONS -} - -codecvt::~codecvt() -{ - if (__l != _LIBCPP_GET_C_LOCALE) - freelocale(__l); -} - -codecvt::result -codecvt::do_out(state_type& st, - const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ - // look for first internal null in frm - const intern_type* fend = frm; - for (; fend != frm_end; ++fend) - if (*fend == 0) - break; - // loop over all null-terminated sequences in frm - to_nxt = to; - for (frm_nxt = frm; frm != frm_end && to != to_end; frm = frm_nxt, to = to_nxt) - { - // save state in case it is needed to recover to_nxt on error - mbstate_t save_state = st; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - size_t n = wcsnrtombs_l(to, &frm_nxt, static_cast(fend-frm), - static_cast(to_end-to), &st, __l); -#else - size_t n = __wcsnrtombs_l(to, &frm_nxt, fend-frm, to_end-to, &st, __l); -#endif - if (n == size_t(-1)) - { - // need to recover to_nxt - for (to_nxt = to; frm != frm_nxt; ++frm) - { -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - n = wcrtomb_l(to_nxt, *frm, &save_state, __l); -#else - n = __wcrtomb_l(to_nxt, *frm, &save_state, __l); -#endif - if (n == size_t(-1)) - break; - to_nxt += n; - } - frm_nxt = frm; - return error; - } - if (n == 0) - return partial; - to_nxt += n; - if (to_nxt == to_end) - break; - if (fend != frm_end) // set up next null terminated sequence - { - // Try to write the terminating null - extern_type tmp[MB_LEN_MAX]; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - n = wcrtomb_l(tmp, intern_type(), &st, __l); -#else - n = __wcrtomb_l(tmp, intern_type(), &st, __l); -#endif - if (n == size_t(-1)) // on error - return error; - if (n > static_cast(to_end-to_nxt)) // is there room? - return partial; - for (extern_type* p = tmp; n; --n) // write it - *to_nxt++ = *p++; - ++frm_nxt; - // look for next null in frm - for (fend = frm_nxt; fend != frm_end; ++fend) - if (*fend == 0) - break; - } - } - return frm_nxt == frm_end ? ok : partial; -} - -codecvt::result -codecvt::do_in(state_type& st, - const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, - intern_type* to, intern_type* to_end, intern_type*& to_nxt) const -{ - // look for first internal null in frm - const extern_type* fend = frm; - for (; fend != frm_end; ++fend) - if (*fend == 0) - break; - // loop over all null-terminated sequences in frm - to_nxt = to; - for (frm_nxt = frm; frm != frm_end && to != to_end; frm = frm_nxt, to = to_nxt) - { - // save state in case it is needed to recover to_nxt on error - mbstate_t save_state = st; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - size_t n = mbsnrtowcs_l(to, &frm_nxt, static_cast(fend-frm), - static_cast(to_end-to), &st, __l); -#else - size_t n = __mbsnrtowcs_l(to, &frm_nxt, fend-frm, to_end-to, &st, __l); -#endif - if (n == size_t(-1)) - { - // need to recover to_nxt - for (to_nxt = to; frm != frm_nxt; ++to_nxt) - { -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - n = mbrtowc_l(to_nxt, frm, static_cast(fend-frm), - &save_state, __l); -#else - n = __mbrtowc_l(to_nxt, frm, fend-frm, &save_state, __l); -#endif - switch (n) - { - case 0: - ++frm; - break; - case size_t(-1): - frm_nxt = frm; - return error; - case size_t(-2): - frm_nxt = frm; - return partial; - default: - frm += n; - break; - } - } - frm_nxt = frm; - return frm_nxt == frm_end ? ok : partial; - } - if (n == (size_t)-1) - return error; - to_nxt += n; - if (to_nxt == to_end) - break; - if (fend != frm_end) // set up next null terminated sequence - { - // Try to write the terminating null -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - n = mbrtowc_l(to_nxt, frm_nxt, 1, &st, __l); -#else - n = __mbrtowc_l(to_nxt, frm_nxt, 1, &st, __l); -#endif - if (n != 0) // on error - return error; - ++to_nxt; - ++frm_nxt; - // look for next null in frm - for (fend = frm_nxt; fend != frm_end; ++fend) - if (*fend == 0) - break; - } - } - return frm_nxt == frm_end ? ok : partial; -} - -codecvt::result -codecvt::do_unshift(state_type& st, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ - to_nxt = to; - extern_type tmp[MB_LEN_MAX]; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - size_t n = wcrtomb_l(tmp, intern_type(), &st, __l); -#else - size_t n = __wcrtomb_l(tmp, intern_type(), &st, __l); -#endif - if (n == size_t(-1) || n == 0) // on error - return error; - --n; - if (n > static_cast(to_end-to_nxt)) // is there room? - return partial; - for (extern_type* p = tmp; n; --n) // write it - *to_nxt++ = *p++; - return ok; -} - -int -codecvt::do_encoding() const _NOEXCEPT -{ -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - if (mbtowc_l(nullptr, nullptr, MB_LEN_MAX, __l) == 0) -#else - if (__mbtowc_l(nullptr, nullptr, MB_LEN_MAX, __l) == 0) -#endif - { - // stateless encoding -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - if (__l == 0 || MB_CUR_MAX_L(__l) == 1) // there are no known constant length encodings -#else - if (__l == 0 || __mb_cur_max_l(__l) == 1) // there are no known constant length encodings -#endif - return 1; // which take more than 1 char to form a wchar_t - return 0; - } - return -1; -} - -bool -codecvt::do_always_noconv() const _NOEXCEPT -{ - return false; -} - -int -codecvt::do_length(state_type& st, - const extern_type* frm, const extern_type* frm_end, size_t mx) const -{ - int nbytes = 0; - for (size_t nwchar_t = 0; nwchar_t < mx && frm != frm_end; ++nwchar_t) - { -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - size_t n = mbrlen_l(frm, static_cast(frm_end-frm), &st, __l); -#else - size_t n = __mbrlen_l(frm, frm_end-frm, &st, __l); -#endif - switch (n) - { - case 0: - ++nbytes; - ++frm; - break; - case size_t(-1): - case size_t(-2): - return nbytes; - default: - nbytes += n; - frm += n; - break; - } - } - return nbytes; -} - -int -codecvt::do_max_length() const _NOEXCEPT -{ -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - return __l == 0 ? 1 : static_cast( MB_CUR_MAX_L(__l)); -#else - return __l == 0 ? 1 : static_cast(__mb_cur_max_l(__l)); -#endif -} - -// Valid UTF ranges -// UTF-32 UTF-16 UTF-8 # of code points -// first second first second third fourth -// 000000 - 00007F 0000 - 007F 00 - 7F 127 -// 000080 - 0007FF 0080 - 07FF C2 - DF, 80 - BF 1920 -// 000800 - 000FFF 0800 - 0FFF E0 - E0, A0 - BF, 80 - BF 2048 -// 001000 - 00CFFF 1000 - CFFF E1 - EC, 80 - BF, 80 - BF 49152 -// 00D000 - 00D7FF D000 - D7FF ED - ED, 80 - 9F, 80 - BF 2048 -// 00D800 - 00DFFF invalid -// 00E000 - 00FFFF E000 - FFFF EE - EF, 80 - BF, 80 - BF 8192 -// 010000 - 03FFFF D800 - D8BF, DC00 - DFFF F0 - F0, 90 - BF, 80 - BF, 80 - BF 196608 -// 040000 - 0FFFFF D8C0 - DBBF, DC00 - DFFF F1 - F3, 80 - BF, 80 - BF, 80 - BF 786432 -// 100000 - 10FFFF DBC0 - DBFF, DC00 - DFFF F4 - F4, 80 - 8F, 80 - BF, 80 - BF 65536 - -static -codecvt_base::result -utf16_to_utf8(const uint16_t* frm, const uint16_t* frm_end, const uint16_t*& frm_nxt, - uint8_t* to, uint8_t* to_end, uint8_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & generate_header) - { - if (to_end-to_nxt < 3) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xEF); - *to_nxt++ = static_cast(0xBB); - *to_nxt++ = static_cast(0xBF); - } - for (; frm_nxt < frm_end; ++frm_nxt) - { - uint16_t wc1 = *frm_nxt; - if (wc1 > Maxcode) - return codecvt_base::error; - if (wc1 < 0x0080) - { - if (to_end-to_nxt < 1) - return codecvt_base::partial; - *to_nxt++ = static_cast(wc1); - } - else if (wc1 < 0x0800) - { - if (to_end-to_nxt < 2) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xC0 | (wc1 >> 6)); - *to_nxt++ = static_cast(0x80 | (wc1 & 0x03F)); - } - else if (wc1 < 0xD800) - { - if (to_end-to_nxt < 3) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xE0 | (wc1 >> 12)); - *to_nxt++ = static_cast(0x80 | ((wc1 & 0x0FC0) >> 6)); - *to_nxt++ = static_cast(0x80 | (wc1 & 0x003F)); - } - else if (wc1 < 0xDC00) - { - if (frm_end-frm_nxt < 2) - return codecvt_base::partial; - uint16_t wc2 = frm_nxt[1]; - if ((wc2 & 0xFC00) != 0xDC00) - return codecvt_base::error; - if (to_end-to_nxt < 4) - return codecvt_base::partial; - if (((((wc1 & 0x03C0UL) >> 6) + 1) << 16) + - ((wc1 & 0x003FUL) << 10) + (wc2 & 0x03FF) > Maxcode) - return codecvt_base::error; - ++frm_nxt; - uint8_t z = ((wc1 & 0x03C0) >> 6) + 1; - *to_nxt++ = static_cast(0xF0 | (z >> 2)); - *to_nxt++ = static_cast(0x80 | ((z & 0x03) << 4) | ((wc1 & 0x003C) >> 2)); - *to_nxt++ = static_cast(0x80 | ((wc1 & 0x0003) << 4) | ((wc2 & 0x03C0) >> 6)); - *to_nxt++ = static_cast(0x80 | (wc2 & 0x003F)); - } - else if (wc1 < 0xE000) - { - return codecvt_base::error; - } - else - { - if (to_end-to_nxt < 3) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xE0 | (wc1 >> 12)); - *to_nxt++ = static_cast(0x80 | ((wc1 & 0x0FC0) >> 6)); - *to_nxt++ = static_cast(0x80 | (wc1 & 0x003F)); - } - } - return codecvt_base::ok; -} - -static -codecvt_base::result -utf16_to_utf8(const uint32_t* frm, const uint32_t* frm_end, const uint32_t*& frm_nxt, - uint8_t* to, uint8_t* to_end, uint8_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & generate_header) - { - if (to_end-to_nxt < 3) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xEF); - *to_nxt++ = static_cast(0xBB); - *to_nxt++ = static_cast(0xBF); - } - for (; frm_nxt < frm_end; ++frm_nxt) - { - uint16_t wc1 = static_cast(*frm_nxt); - if (wc1 > Maxcode) - return codecvt_base::error; - if (wc1 < 0x0080) - { - if (to_end-to_nxt < 1) - return codecvt_base::partial; - *to_nxt++ = static_cast(wc1); - } - else if (wc1 < 0x0800) - { - if (to_end-to_nxt < 2) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xC0 | (wc1 >> 6)); - *to_nxt++ = static_cast(0x80 | (wc1 & 0x03F)); - } - else if (wc1 < 0xD800) - { - if (to_end-to_nxt < 3) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xE0 | (wc1 >> 12)); - *to_nxt++ = static_cast(0x80 | ((wc1 & 0x0FC0) >> 6)); - *to_nxt++ = static_cast(0x80 | (wc1 & 0x003F)); - } - else if (wc1 < 0xDC00) - { - if (frm_end-frm_nxt < 2) - return codecvt_base::partial; - uint16_t wc2 = static_cast(frm_nxt[1]); - if ((wc2 & 0xFC00) != 0xDC00) - return codecvt_base::error; - if (to_end-to_nxt < 4) - return codecvt_base::partial; - if (((((wc1 & 0x03C0UL) >> 6) + 1) << 16) + - ((wc1 & 0x003FUL) << 10) + (wc2 & 0x03FF) > Maxcode) - return codecvt_base::error; - ++frm_nxt; - uint8_t z = ((wc1 & 0x03C0) >> 6) + 1; - *to_nxt++ = static_cast(0xF0 | (z >> 2)); - *to_nxt++ = static_cast(0x80 | ((z & 0x03) << 4) | ((wc1 & 0x003C) >> 2)); - *to_nxt++ = static_cast(0x80 | ((wc1 & 0x0003) << 4) | ((wc2 & 0x03C0) >> 6)); - *to_nxt++ = static_cast(0x80 | (wc2 & 0x003F)); - } - else if (wc1 < 0xE000) - { - return codecvt_base::error; - } - else - { - if (to_end-to_nxt < 3) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xE0 | (wc1 >> 12)); - *to_nxt++ = static_cast(0x80 | ((wc1 & 0x0FC0) >> 6)); - *to_nxt++ = static_cast(0x80 | (wc1 & 0x003F)); - } - } - return codecvt_base::ok; -} - -static -codecvt_base::result -utf8_to_utf16(const uint8_t* frm, const uint8_t* frm_end, const uint8_t*& frm_nxt, - uint16_t* to, uint16_t* to_end, uint16_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & consume_header) - { - if (frm_end-frm_nxt >= 3 && frm_nxt[0] == 0xEF && frm_nxt[1] == 0xBB && - frm_nxt[2] == 0xBF) - frm_nxt += 3; - } - for (; frm_nxt < frm_end && to_nxt < to_end; ++to_nxt) - { - uint8_t c1 = *frm_nxt; - if (c1 > Maxcode) - return codecvt_base::error; - if (c1 < 0x80) - { - *to_nxt = static_cast(c1); - ++frm_nxt; - } - else if (c1 < 0xC2) - { - return codecvt_base::error; - } - else if (c1 < 0xE0) - { - if (frm_end-frm_nxt < 2) - return codecvt_base::partial; - uint8_t c2 = frm_nxt[1]; - if ((c2 & 0xC0) != 0x80) - return codecvt_base::error; - uint16_t t = static_cast(((c1 & 0x1F) << 6) | (c2 & 0x3F)); - if (t > Maxcode) - return codecvt_base::error; - *to_nxt = t; - frm_nxt += 2; - } - else if (c1 < 0xF0) - { - if (frm_end-frm_nxt < 3) - return codecvt_base::partial; - uint8_t c2 = frm_nxt[1]; - uint8_t c3 = frm_nxt[2]; - switch (c1) - { - case 0xE0: - if ((c2 & 0xE0) != 0xA0) - return codecvt_base::error; - break; - case 0xED: - if ((c2 & 0xE0) != 0x80) - return codecvt_base::error; - break; - default: - if ((c2 & 0xC0) != 0x80) - return codecvt_base::error; - break; - } - if ((c3 & 0xC0) != 0x80) - return codecvt_base::error; - uint16_t t = static_cast(((c1 & 0x0F) << 12) - | ((c2 & 0x3F) << 6) - | (c3 & 0x3F)); - if (t > Maxcode) - return codecvt_base::error; - *to_nxt = t; - frm_nxt += 3; - } - else if (c1 < 0xF5) - { - if (frm_end-frm_nxt < 4) - return codecvt_base::partial; - uint8_t c2 = frm_nxt[1]; - uint8_t c3 = frm_nxt[2]; - uint8_t c4 = frm_nxt[3]; - switch (c1) - { - case 0xF0: - if (!(0x90 <= c2 && c2 <= 0xBF)) - return codecvt_base::error; - break; - case 0xF4: - if ((c2 & 0xF0) != 0x80) - return codecvt_base::error; - break; - default: - if ((c2 & 0xC0) != 0x80) - return codecvt_base::error; - break; - } - if ((c3 & 0xC0) != 0x80 || (c4 & 0xC0) != 0x80) - return codecvt_base::error; - if (to_end-to_nxt < 2) - return codecvt_base::partial; - if ((((c1 & 7UL) << 18) + - ((c2 & 0x3FUL) << 12) + - ((c3 & 0x3FUL) << 6) + (c4 & 0x3F)) > Maxcode) - return codecvt_base::error; - *to_nxt = static_cast( - 0xD800 - | (((((c1 & 0x07) << 2) | ((c2 & 0x30) >> 4)) - 1) << 6) - | ((c2 & 0x0F) << 2) - | ((c3 & 0x30) >> 4)); - *++to_nxt = static_cast( - 0xDC00 - | ((c3 & 0x0F) << 6) - | (c4 & 0x3F)); - frm_nxt += 4; - } - else - { - return codecvt_base::error; - } - } - return frm_nxt < frm_end ? codecvt_base::partial : codecvt_base::ok; -} - -static -codecvt_base::result -utf8_to_utf16(const uint8_t* frm, const uint8_t* frm_end, const uint8_t*& frm_nxt, - uint32_t* to, uint32_t* to_end, uint32_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & consume_header) - { - if (frm_end-frm_nxt >= 3 && frm_nxt[0] == 0xEF && frm_nxt[1] == 0xBB && - frm_nxt[2] == 0xBF) - frm_nxt += 3; - } - for (; frm_nxt < frm_end && to_nxt < to_end; ++to_nxt) - { - uint8_t c1 = *frm_nxt; - if (c1 > Maxcode) - return codecvt_base::error; - if (c1 < 0x80) - { - *to_nxt = static_cast(c1); - ++frm_nxt; - } - else if (c1 < 0xC2) - { - return codecvt_base::error; - } - else if (c1 < 0xE0) - { - if (frm_end-frm_nxt < 2) - return codecvt_base::partial; - uint8_t c2 = frm_nxt[1]; - if ((c2 & 0xC0) != 0x80) - return codecvt_base::error; - uint16_t t = static_cast(((c1 & 0x1F) << 6) | (c2 & 0x3F)); - if (t > Maxcode) - return codecvt_base::error; - *to_nxt = static_cast(t); - frm_nxt += 2; - } - else if (c1 < 0xF0) - { - if (frm_end-frm_nxt < 3) - return codecvt_base::partial; - uint8_t c2 = frm_nxt[1]; - uint8_t c3 = frm_nxt[2]; - switch (c1) - { - case 0xE0: - if ((c2 & 0xE0) != 0xA0) - return codecvt_base::error; - break; - case 0xED: - if ((c2 & 0xE0) != 0x80) - return codecvt_base::error; - break; - default: - if ((c2 & 0xC0) != 0x80) - return codecvt_base::error; - break; - } - if ((c3 & 0xC0) != 0x80) - return codecvt_base::error; - uint16_t t = static_cast(((c1 & 0x0F) << 12) - | ((c2 & 0x3F) << 6) - | (c3 & 0x3F)); - if (t > Maxcode) - return codecvt_base::error; - *to_nxt = static_cast(t); - frm_nxt += 3; - } - else if (c1 < 0xF5) - { - if (frm_end-frm_nxt < 4) - return codecvt_base::partial; - uint8_t c2 = frm_nxt[1]; - uint8_t c3 = frm_nxt[2]; - uint8_t c4 = frm_nxt[3]; - switch (c1) - { - case 0xF0: - if (!(0x90 <= c2 && c2 <= 0xBF)) - return codecvt_base::error; - break; - case 0xF4: - if ((c2 & 0xF0) != 0x80) - return codecvt_base::error; - break; - default: - if ((c2 & 0xC0) != 0x80) - return codecvt_base::error; - break; - } - if ((c3 & 0xC0) != 0x80 || (c4 & 0xC0) != 0x80) - return codecvt_base::error; - if (to_end-to_nxt < 2) - return codecvt_base::partial; - if ((((c1 & 7UL) << 18) + - ((c2 & 0x3FUL) << 12) + - ((c3 & 0x3FUL) << 6) + (c4 & 0x3F)) > Maxcode) - return codecvt_base::error; - *to_nxt = static_cast( - 0xD800 - | (((((c1 & 0x07) << 2) | ((c2 & 0x30) >> 4)) - 1) << 6) - | ((c2 & 0x0F) << 2) - | ((c3 & 0x30) >> 4)); - *++to_nxt = static_cast( - 0xDC00 - | ((c3 & 0x0F) << 6) - | (c4 & 0x3F)); - frm_nxt += 4; - } - else - { - return codecvt_base::error; - } - } - return frm_nxt < frm_end ? codecvt_base::partial : codecvt_base::ok; -} - -static -int -utf8_to_utf16_length(const uint8_t* frm, const uint8_t* frm_end, - size_t mx, unsigned long Maxcode = 0x10FFFF, - codecvt_mode mode = codecvt_mode(0)) -{ - const uint8_t* frm_nxt = frm; - if (mode & consume_header) - { - if (frm_end-frm_nxt >= 3 && frm_nxt[0] == 0xEF && frm_nxt[1] == 0xBB && - frm_nxt[2] == 0xBF) - frm_nxt += 3; - } - for (size_t nchar16_t = 0; frm_nxt < frm_end && nchar16_t < mx; ++nchar16_t) - { - uint8_t c1 = *frm_nxt; - if (c1 > Maxcode) - break; - if (c1 < 0x80) - { - ++frm_nxt; - } - else if (c1 < 0xC2) - { - break; - } - else if (c1 < 0xE0) - { - if ((frm_end-frm_nxt < 2) || (frm_nxt[1] & 0xC0) != 0x80) - break; - uint16_t t = static_cast(((c1 & 0x1F) << 6) | (frm_nxt[1] & 0x3F)); - if (t > Maxcode) - break; - frm_nxt += 2; - } - else if (c1 < 0xF0) - { - if (frm_end-frm_nxt < 3) - break; - uint8_t c2 = frm_nxt[1]; - uint8_t c3 = frm_nxt[2]; - switch (c1) - { - case 0xE0: - if ((c2 & 0xE0) != 0xA0) - return static_cast(frm_nxt - frm); - break; - case 0xED: - if ((c2 & 0xE0) != 0x80) - return static_cast(frm_nxt - frm); - break; - default: - if ((c2 & 0xC0) != 0x80) - return static_cast(frm_nxt - frm); - break; - } - if ((c3 & 0xC0) != 0x80) - break; - if ((((c1 & 0x0Fu) << 12) | ((c2 & 0x3Fu) << 6) | (c3 & 0x3Fu)) > Maxcode) - break; - frm_nxt += 3; - } - else if (c1 < 0xF5) - { - if (frm_end-frm_nxt < 4 || mx-nchar16_t < 2) - break; - uint8_t c2 = frm_nxt[1]; - uint8_t c3 = frm_nxt[2]; - uint8_t c4 = frm_nxt[3]; - switch (c1) - { - case 0xF0: - if (!(0x90 <= c2 && c2 <= 0xBF)) - return static_cast(frm_nxt - frm); - break; - case 0xF4: - if ((c2 & 0xF0) != 0x80) - return static_cast(frm_nxt - frm); - break; - default: - if ((c2 & 0xC0) != 0x80) - return static_cast(frm_nxt - frm); - break; - } - if ((c3 & 0xC0) != 0x80 || (c4 & 0xC0) != 0x80) - break; - if ((((c1 & 7UL) << 18) + - ((c2 & 0x3FUL) << 12) + - ((c3 & 0x3FUL) << 6) + (c4 & 0x3F)) > Maxcode) - break; - ++nchar16_t; - frm_nxt += 4; - } - else - { - break; - } - } - return static_cast(frm_nxt - frm); -} - -static -codecvt_base::result -ucs4_to_utf8(const uint32_t* frm, const uint32_t* frm_end, const uint32_t*& frm_nxt, - uint8_t* to, uint8_t* to_end, uint8_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & generate_header) - { - if (to_end-to_nxt < 3) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xEF); - *to_nxt++ = static_cast(0xBB); - *to_nxt++ = static_cast(0xBF); - } - for (; frm_nxt < frm_end; ++frm_nxt) - { - uint32_t wc = *frm_nxt; - if ((wc & 0xFFFFF800) == 0x00D800 || wc > Maxcode) - return codecvt_base::error; - if (wc < 0x000080) - { - if (to_end-to_nxt < 1) - return codecvt_base::partial; - *to_nxt++ = static_cast(wc); - } - else if (wc < 0x000800) - { - if (to_end-to_nxt < 2) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xC0 | (wc >> 6)); - *to_nxt++ = static_cast(0x80 | (wc & 0x03F)); - } - else if (wc < 0x010000) - { - if (to_end-to_nxt < 3) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xE0 | (wc >> 12)); - *to_nxt++ = static_cast(0x80 | ((wc & 0x0FC0) >> 6)); - *to_nxt++ = static_cast(0x80 | (wc & 0x003F)); - } - else // if (wc < 0x110000) - { - if (to_end-to_nxt < 4) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xF0 | (wc >> 18)); - *to_nxt++ = static_cast(0x80 | ((wc & 0x03F000) >> 12)); - *to_nxt++ = static_cast(0x80 | ((wc & 0x000FC0) >> 6)); - *to_nxt++ = static_cast(0x80 | (wc & 0x00003F)); - } - } - return codecvt_base::ok; -} - -static -codecvt_base::result -utf8_to_ucs4(const uint8_t* frm, const uint8_t* frm_end, const uint8_t*& frm_nxt, - uint32_t* to, uint32_t* to_end, uint32_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & consume_header) - { - if (frm_end-frm_nxt >= 3 && frm_nxt[0] == 0xEF && frm_nxt[1] == 0xBB && - frm_nxt[2] == 0xBF) - frm_nxt += 3; - } - for (; frm_nxt < frm_end && to_nxt < to_end; ++to_nxt) - { - uint8_t c1 = static_cast(*frm_nxt); - if (c1 < 0x80) - { - if (c1 > Maxcode) - return codecvt_base::error; - *to_nxt = static_cast(c1); - ++frm_nxt; - } - else if (c1 < 0xC2) - { - return codecvt_base::error; - } - else if (c1 < 0xE0) - { - if (frm_end-frm_nxt < 2) - return codecvt_base::partial; - uint8_t c2 = frm_nxt[1]; - if ((c2 & 0xC0) != 0x80) - return codecvt_base::error; - uint32_t t = static_cast(((c1 & 0x1F) << 6) - | (c2 & 0x3F)); - if (t > Maxcode) - return codecvt_base::error; - *to_nxt = t; - frm_nxt += 2; - } - else if (c1 < 0xF0) - { - if (frm_end-frm_nxt < 3) - return codecvt_base::partial; - uint8_t c2 = frm_nxt[1]; - uint8_t c3 = frm_nxt[2]; - switch (c1) - { - case 0xE0: - if ((c2 & 0xE0) != 0xA0) - return codecvt_base::error; - break; - case 0xED: - if ((c2 & 0xE0) != 0x80) - return codecvt_base::error; - break; - default: - if ((c2 & 0xC0) != 0x80) - return codecvt_base::error; - break; - } - if ((c3 & 0xC0) != 0x80) - return codecvt_base::error; - uint32_t t = static_cast(((c1 & 0x0F) << 12) - | ((c2 & 0x3F) << 6) - | (c3 & 0x3F)); - if (t > Maxcode) - return codecvt_base::error; - *to_nxt = t; - frm_nxt += 3; - } - else if (c1 < 0xF5) - { - if (frm_end-frm_nxt < 4) - return codecvt_base::partial; - uint8_t c2 = frm_nxt[1]; - uint8_t c3 = frm_nxt[2]; - uint8_t c4 = frm_nxt[3]; - switch (c1) - { - case 0xF0: - if (!(0x90 <= c2 && c2 <= 0xBF)) - return codecvt_base::error; - break; - case 0xF4: - if ((c2 & 0xF0) != 0x80) - return codecvt_base::error; - break; - default: - if ((c2 & 0xC0) != 0x80) - return codecvt_base::error; - break; - } - if ((c3 & 0xC0) != 0x80 || (c4 & 0xC0) != 0x80) - return codecvt_base::error; - uint32_t t = static_cast(((c1 & 0x07) << 18) - | ((c2 & 0x3F) << 12) - | ((c3 & 0x3F) << 6) - | (c4 & 0x3F)); - if (t > Maxcode) - return codecvt_base::error; - *to_nxt = t; - frm_nxt += 4; - } - else - { - return codecvt_base::error; - } - } - return frm_nxt < frm_end ? codecvt_base::partial : codecvt_base::ok; -} - -static -int -utf8_to_ucs4_length(const uint8_t* frm, const uint8_t* frm_end, - size_t mx, unsigned long Maxcode = 0x10FFFF, - codecvt_mode mode = codecvt_mode(0)) -{ - const uint8_t* frm_nxt = frm; - if (mode & consume_header) - { - if (frm_end-frm_nxt >= 3 && frm_nxt[0] == 0xEF && frm_nxt[1] == 0xBB && - frm_nxt[2] == 0xBF) - frm_nxt += 3; - } - for (size_t nchar32_t = 0; frm_nxt < frm_end && nchar32_t < mx; ++nchar32_t) - { - uint8_t c1 = static_cast(*frm_nxt); - if (c1 < 0x80) - { - if (c1 > Maxcode) - break; - ++frm_nxt; - } - else if (c1 < 0xC2) - { - break; - } - else if (c1 < 0xE0) - { - if ((frm_end-frm_nxt < 2) || ((frm_nxt[1] & 0xC0) != 0x80)) - break; - if ((((c1 & 0x1Fu) << 6) | (frm_nxt[1] & 0x3Fu)) > Maxcode) - break; - frm_nxt += 2; - } - else if (c1 < 0xF0) - { - if (frm_end-frm_nxt < 3) - break; - uint8_t c2 = frm_nxt[1]; - uint8_t c3 = frm_nxt[2]; - switch (c1) - { - case 0xE0: - if ((c2 & 0xE0) != 0xA0) - return static_cast(frm_nxt - frm); - break; - case 0xED: - if ((c2 & 0xE0) != 0x80) - return static_cast(frm_nxt - frm); - break; - default: - if ((c2 & 0xC0) != 0x80) - return static_cast(frm_nxt - frm); - break; - } - if ((c3 & 0xC0) != 0x80) - break; - if ((((c1 & 0x0Fu) << 12) | ((c2 & 0x3Fu) << 6) | (c3 & 0x3Fu)) > Maxcode) - break; - frm_nxt += 3; - } - else if (c1 < 0xF5) - { - if (frm_end-frm_nxt < 4) - break; - uint8_t c2 = frm_nxt[1]; - uint8_t c3 = frm_nxt[2]; - uint8_t c4 = frm_nxt[3]; - switch (c1) - { - case 0xF0: - if (!(0x90 <= c2 && c2 <= 0xBF)) - return static_cast(frm_nxt - frm); - break; - case 0xF4: - if ((c2 & 0xF0) != 0x80) - return static_cast(frm_nxt - frm); - break; - default: - if ((c2 & 0xC0) != 0x80) - return static_cast(frm_nxt - frm); - break; - } - if ((c3 & 0xC0) != 0x80 || (c4 & 0xC0) != 0x80) - break; - if ((((c1 & 0x07u) << 18) | ((c2 & 0x3Fu) << 12) | - ((c3 & 0x3Fu) << 6) | (c4 & 0x3Fu)) > Maxcode) - break; - frm_nxt += 4; - } - else - { - break; - } - } - return static_cast(frm_nxt - frm); -} - -static -codecvt_base::result -ucs2_to_utf8(const uint16_t* frm, const uint16_t* frm_end, const uint16_t*& frm_nxt, - uint8_t* to, uint8_t* to_end, uint8_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & generate_header) - { - if (to_end-to_nxt < 3) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xEF); - *to_nxt++ = static_cast(0xBB); - *to_nxt++ = static_cast(0xBF); - } - for (; frm_nxt < frm_end; ++frm_nxt) - { - uint16_t wc = *frm_nxt; - if ((wc & 0xF800) == 0xD800 || wc > Maxcode) - return codecvt_base::error; - if (wc < 0x0080) - { - if (to_end-to_nxt < 1) - return codecvt_base::partial; - *to_nxt++ = static_cast(wc); - } - else if (wc < 0x0800) - { - if (to_end-to_nxt < 2) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xC0 | (wc >> 6)); - *to_nxt++ = static_cast(0x80 | (wc & 0x03F)); - } - else // if (wc <= 0xFFFF) - { - if (to_end-to_nxt < 3) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xE0 | (wc >> 12)); - *to_nxt++ = static_cast(0x80 | ((wc & 0x0FC0) >> 6)); - *to_nxt++ = static_cast(0x80 | (wc & 0x003F)); - } - } - return codecvt_base::ok; -} - -static -codecvt_base::result -utf8_to_ucs2(const uint8_t* frm, const uint8_t* frm_end, const uint8_t*& frm_nxt, - uint16_t* to, uint16_t* to_end, uint16_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & consume_header) - { - if (frm_end-frm_nxt >= 3 && frm_nxt[0] == 0xEF && frm_nxt[1] == 0xBB && - frm_nxt[2] == 0xBF) - frm_nxt += 3; - } - for (; frm_nxt < frm_end && to_nxt < to_end; ++to_nxt) - { - uint8_t c1 = static_cast(*frm_nxt); - if (c1 < 0x80) - { - if (c1 > Maxcode) - return codecvt_base::error; - *to_nxt = static_cast(c1); - ++frm_nxt; - } - else if (c1 < 0xC2) - { - return codecvt_base::error; - } - else if (c1 < 0xE0) - { - if (frm_end-frm_nxt < 2) - return codecvt_base::partial; - uint8_t c2 = frm_nxt[1]; - if ((c2 & 0xC0) != 0x80) - return codecvt_base::error; - uint16_t t = static_cast(((c1 & 0x1F) << 6) - | (c2 & 0x3F)); - if (t > Maxcode) - return codecvt_base::error; - *to_nxt = t; - frm_nxt += 2; - } - else if (c1 < 0xF0) - { - if (frm_end-frm_nxt < 3) - return codecvt_base::partial; - uint8_t c2 = frm_nxt[1]; - uint8_t c3 = frm_nxt[2]; - switch (c1) - { - case 0xE0: - if ((c2 & 0xE0) != 0xA0) - return codecvt_base::error; - break; - case 0xED: - if ((c2 & 0xE0) != 0x80) - return codecvt_base::error; - break; - default: - if ((c2 & 0xC0) != 0x80) - return codecvt_base::error; - break; - } - if ((c3 & 0xC0) != 0x80) - return codecvt_base::error; - uint16_t t = static_cast(((c1 & 0x0F) << 12) - | ((c2 & 0x3F) << 6) - | (c3 & 0x3F)); - if (t > Maxcode) - return codecvt_base::error; - *to_nxt = t; - frm_nxt += 3; - } - else - { - return codecvt_base::error; - } - } - return frm_nxt < frm_end ? codecvt_base::partial : codecvt_base::ok; -} - -static -int -utf8_to_ucs2_length(const uint8_t* frm, const uint8_t* frm_end, - size_t mx, unsigned long Maxcode = 0x10FFFF, - codecvt_mode mode = codecvt_mode(0)) -{ - const uint8_t* frm_nxt = frm; - if (mode & consume_header) - { - if (frm_end-frm_nxt >= 3 && frm_nxt[0] == 0xEF && frm_nxt[1] == 0xBB && - frm_nxt[2] == 0xBF) - frm_nxt += 3; - } - for (size_t nchar32_t = 0; frm_nxt < frm_end && nchar32_t < mx; ++nchar32_t) - { - uint8_t c1 = static_cast(*frm_nxt); - if (c1 < 0x80) - { - if (c1 > Maxcode) - break; - ++frm_nxt; - } - else if (c1 < 0xC2) - { - break; - } - else if (c1 < 0xE0) - { - if ((frm_end-frm_nxt < 2) || ((frm_nxt[1] & 0xC0) != 0x80)) - break; - if ((((c1 & 0x1Fu) << 6) | (frm_nxt[1] & 0x3Fu)) > Maxcode) - break; - frm_nxt += 2; - } - else if (c1 < 0xF0) - { - if (frm_end-frm_nxt < 3) - break; - uint8_t c2 = frm_nxt[1]; - uint8_t c3 = frm_nxt[2]; - switch (c1) - { - case 0xE0: - if ((c2 & 0xE0) != 0xA0) - return static_cast(frm_nxt - frm); - break; - case 0xED: - if ((c2 & 0xE0) != 0x80) - return static_cast(frm_nxt - frm); - break; - default: - if ((c2 & 0xC0) != 0x80) - return static_cast(frm_nxt - frm); - break; - } - if ((c3 & 0xC0) != 0x80) - break; - if ((((c1 & 0x0Fu) << 12) | ((c2 & 0x3Fu) << 6) | (c3 & 0x3Fu)) > Maxcode) - break; - frm_nxt += 3; - } - else - { - break; - } - } - return static_cast(frm_nxt - frm); -} - -static -codecvt_base::result -ucs4_to_utf16be(const uint32_t* frm, const uint32_t* frm_end, const uint32_t*& frm_nxt, - uint8_t* to, uint8_t* to_end, uint8_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & generate_header) - { - if (to_end-to_nxt < 2) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xFE); - *to_nxt++ = static_cast(0xFF); - } - for (; frm_nxt < frm_end; ++frm_nxt) - { - uint32_t wc = *frm_nxt; - if ((wc & 0xFFFFF800) == 0x00D800 || wc > Maxcode) - return codecvt_base::error; - if (wc < 0x010000) - { - if (to_end-to_nxt < 2) - return codecvt_base::partial; - *to_nxt++ = static_cast(wc >> 8); - *to_nxt++ = static_cast(wc); - } - else - { - if (to_end-to_nxt < 4) - return codecvt_base::partial; - uint16_t t = static_cast( - 0xD800 - | ((((wc & 0x1F0000) >> 16) - 1) << 6) - | ((wc & 0x00FC00) >> 10)); - *to_nxt++ = static_cast(t >> 8); - *to_nxt++ = static_cast(t); - t = static_cast(0xDC00 | (wc & 0x03FF)); - *to_nxt++ = static_cast(t >> 8); - *to_nxt++ = static_cast(t); - } - } - return codecvt_base::ok; -} - -static -codecvt_base::result -utf16be_to_ucs4(const uint8_t* frm, const uint8_t* frm_end, const uint8_t*& frm_nxt, - uint32_t* to, uint32_t* to_end, uint32_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & consume_header) - { - if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFE && frm_nxt[1] == 0xFF) - frm_nxt += 2; - } - for (; frm_nxt < frm_end - 1 && to_nxt < to_end; ++to_nxt) - { - uint16_t c1 = static_cast(frm_nxt[0] << 8 | frm_nxt[1]); - if ((c1 & 0xFC00) == 0xDC00) - return codecvt_base::error; - if ((c1 & 0xFC00) != 0xD800) - { - if (c1 > Maxcode) - return codecvt_base::error; - *to_nxt = static_cast(c1); - frm_nxt += 2; - } - else - { - if (frm_end-frm_nxt < 4) - return codecvt_base::partial; - uint16_t c2 = static_cast(frm_nxt[2] << 8 | frm_nxt[3]); - if ((c2 & 0xFC00) != 0xDC00) - return codecvt_base::error; - uint32_t t = static_cast( - ((((c1 & 0x03C0) >> 6) + 1) << 16) - | ((c1 & 0x003F) << 10) - | (c2 & 0x03FF)); - if (t > Maxcode) - return codecvt_base::error; - *to_nxt = t; - frm_nxt += 4; - } - } - return frm_nxt < frm_end ? codecvt_base::partial : codecvt_base::ok; -} - -static -int -utf16be_to_ucs4_length(const uint8_t* frm, const uint8_t* frm_end, - size_t mx, unsigned long Maxcode = 0x10FFFF, - codecvt_mode mode = codecvt_mode(0)) -{ - const uint8_t* frm_nxt = frm; - if (mode & consume_header) - { - if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFE && frm_nxt[1] == 0xFF) - frm_nxt += 2; - } - for (size_t nchar32_t = 0; frm_nxt < frm_end - 1 && nchar32_t < mx; ++nchar32_t) - { - uint16_t c1 = static_cast(frm_nxt[0] << 8 | frm_nxt[1]); - if ((c1 & 0xFC00) == 0xDC00) - break; - if ((c1 & 0xFC00) != 0xD800) - { - if (c1 > Maxcode) - break; - frm_nxt += 2; - } - else - { - if (frm_end-frm_nxt < 4) - break; - uint16_t c2 = static_cast(frm_nxt[2] << 8 | frm_nxt[3]); - if ((c2 & 0xFC00) != 0xDC00) - break; - uint32_t t = static_cast( - ((((c1 & 0x03C0) >> 6) + 1) << 16) - | ((c1 & 0x003F) << 10) - | (c2 & 0x03FF)); - if (t > Maxcode) - break; - frm_nxt += 4; - } - } - return static_cast(frm_nxt - frm); -} - -static -codecvt_base::result -ucs4_to_utf16le(const uint32_t* frm, const uint32_t* frm_end, const uint32_t*& frm_nxt, - uint8_t* to, uint8_t* to_end, uint8_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & generate_header) - { - if (to_end-to_nxt < 2) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xFF); - *to_nxt++ = static_cast(0xFE); - } - for (; frm_nxt < frm_end; ++frm_nxt) - { - uint32_t wc = *frm_nxt; - if ((wc & 0xFFFFF800) == 0x00D800 || wc > Maxcode) - return codecvt_base::error; - if (wc < 0x010000) - { - if (to_end-to_nxt < 2) - return codecvt_base::partial; - *to_nxt++ = static_cast(wc); - *to_nxt++ = static_cast(wc >> 8); - } - else - { - if (to_end-to_nxt < 4) - return codecvt_base::partial; - uint16_t t = static_cast( - 0xD800 - | ((((wc & 0x1F0000) >> 16) - 1) << 6) - | ((wc & 0x00FC00) >> 10)); - *to_nxt++ = static_cast(t); - *to_nxt++ = static_cast(t >> 8); - t = static_cast(0xDC00 | (wc & 0x03FF)); - *to_nxt++ = static_cast(t); - *to_nxt++ = static_cast(t >> 8); - } - } - return codecvt_base::ok; -} - -static -codecvt_base::result -utf16le_to_ucs4(const uint8_t* frm, const uint8_t* frm_end, const uint8_t*& frm_nxt, - uint32_t* to, uint32_t* to_end, uint32_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & consume_header) - { - if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFF && frm_nxt[1] == 0xFE) - frm_nxt += 2; - } - for (; frm_nxt < frm_end - 1 && to_nxt < to_end; ++to_nxt) - { - uint16_t c1 = static_cast(frm_nxt[1] << 8 | frm_nxt[0]); - if ((c1 & 0xFC00) == 0xDC00) - return codecvt_base::error; - if ((c1 & 0xFC00) != 0xD800) - { - if (c1 > Maxcode) - return codecvt_base::error; - *to_nxt = static_cast(c1); - frm_nxt += 2; - } - else - { - if (frm_end-frm_nxt < 4) - return codecvt_base::partial; - uint16_t c2 = static_cast(frm_nxt[3] << 8 | frm_nxt[2]); - if ((c2 & 0xFC00) != 0xDC00) - return codecvt_base::error; - uint32_t t = static_cast( - ((((c1 & 0x03C0) >> 6) + 1) << 16) - | ((c1 & 0x003F) << 10) - | (c2 & 0x03FF)); - if (t > Maxcode) - return codecvt_base::error; - *to_nxt = t; - frm_nxt += 4; - } - } - return frm_nxt < frm_end ? codecvt_base::partial : codecvt_base::ok; -} - -static -int -utf16le_to_ucs4_length(const uint8_t* frm, const uint8_t* frm_end, - size_t mx, unsigned long Maxcode = 0x10FFFF, - codecvt_mode mode = codecvt_mode(0)) -{ - const uint8_t* frm_nxt = frm; - if (mode & consume_header) - { - if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFF && frm_nxt[1] == 0xFE) - frm_nxt += 2; - } - for (size_t nchar32_t = 0; frm_nxt < frm_end - 1 && nchar32_t < mx; ++nchar32_t) - { - uint16_t c1 = static_cast(frm_nxt[1] << 8 | frm_nxt[0]); - if ((c1 & 0xFC00) == 0xDC00) - break; - if ((c1 & 0xFC00) != 0xD800) - { - if (c1 > Maxcode) - break; - frm_nxt += 2; - } - else - { - if (frm_end-frm_nxt < 4) - break; - uint16_t c2 = static_cast(frm_nxt[3] << 8 | frm_nxt[2]); - if ((c2 & 0xFC00) != 0xDC00) - break; - uint32_t t = static_cast( - ((((c1 & 0x03C0) >> 6) + 1) << 16) - | ((c1 & 0x003F) << 10) - | (c2 & 0x03FF)); - if (t > Maxcode) - break; - frm_nxt += 4; - } - } - return static_cast(frm_nxt - frm); -} - -static -codecvt_base::result -ucs2_to_utf16be(const uint16_t* frm, const uint16_t* frm_end, const uint16_t*& frm_nxt, - uint8_t* to, uint8_t* to_end, uint8_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & generate_header) - { - if (to_end-to_nxt < 2) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xFE); - *to_nxt++ = static_cast(0xFF); - } - for (; frm_nxt < frm_end; ++frm_nxt) - { - uint16_t wc = *frm_nxt; - if ((wc & 0xF800) == 0xD800 || wc > Maxcode) - return codecvt_base::error; - if (to_end-to_nxt < 2) - return codecvt_base::partial; - *to_nxt++ = static_cast(wc >> 8); - *to_nxt++ = static_cast(wc); - } - return codecvt_base::ok; -} - -static -codecvt_base::result -utf16be_to_ucs2(const uint8_t* frm, const uint8_t* frm_end, const uint8_t*& frm_nxt, - uint16_t* to, uint16_t* to_end, uint16_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & consume_header) - { - if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFE && frm_nxt[1] == 0xFF) - frm_nxt += 2; - } - for (; frm_nxt < frm_end - 1 && to_nxt < to_end; ++to_nxt) - { - uint16_t c1 = static_cast(frm_nxt[0] << 8 | frm_nxt[1]); - if ((c1 & 0xF800) == 0xD800 || c1 > Maxcode) - return codecvt_base::error; - *to_nxt = c1; - frm_nxt += 2; - } - return frm_nxt < frm_end ? codecvt_base::partial : codecvt_base::ok; -} - -static -int -utf16be_to_ucs2_length(const uint8_t* frm, const uint8_t* frm_end, - size_t mx, unsigned long Maxcode = 0x10FFFF, - codecvt_mode mode = codecvt_mode(0)) -{ - const uint8_t* frm_nxt = frm; - if (mode & consume_header) - { - if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFE && frm_nxt[1] == 0xFF) - frm_nxt += 2; - } - for (size_t nchar16_t = 0; frm_nxt < frm_end - 1 && nchar16_t < mx; ++nchar16_t) - { - uint16_t c1 = static_cast(frm_nxt[0] << 8 | frm_nxt[1]); - if ((c1 & 0xF800) == 0xD800 || c1 > Maxcode) - break; - frm_nxt += 2; - } - return static_cast(frm_nxt - frm); -} - -static -codecvt_base::result -ucs2_to_utf16le(const uint16_t* frm, const uint16_t* frm_end, const uint16_t*& frm_nxt, - uint8_t* to, uint8_t* to_end, uint8_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & generate_header) - { - if (to_end-to_nxt < 2) - return codecvt_base::partial; - *to_nxt++ = static_cast(0xFF); - *to_nxt++ = static_cast(0xFE); - } - for (; frm_nxt < frm_end; ++frm_nxt) - { - uint16_t wc = *frm_nxt; - if ((wc & 0xF800) == 0xD800 || wc > Maxcode) - return codecvt_base::error; - if (to_end-to_nxt < 2) - return codecvt_base::partial; - *to_nxt++ = static_cast(wc); - *to_nxt++ = static_cast(wc >> 8); - } - return codecvt_base::ok; -} - -static -codecvt_base::result -utf16le_to_ucs2(const uint8_t* frm, const uint8_t* frm_end, const uint8_t*& frm_nxt, - uint16_t* to, uint16_t* to_end, uint16_t*& to_nxt, - unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) -{ - frm_nxt = frm; - to_nxt = to; - if (mode & consume_header) - { - if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFF && frm_nxt[1] == 0xFE) - frm_nxt += 2; - } - for (; frm_nxt < frm_end - 1 && to_nxt < to_end; ++to_nxt) - { - uint16_t c1 = static_cast(frm_nxt[1] << 8 | frm_nxt[0]); - if ((c1 & 0xF800) == 0xD800 || c1 > Maxcode) - return codecvt_base::error; - *to_nxt = c1; - frm_nxt += 2; - } - return frm_nxt < frm_end ? codecvt_base::partial : codecvt_base::ok; -} - -static -int -utf16le_to_ucs2_length(const uint8_t* frm, const uint8_t* frm_end, - size_t mx, unsigned long Maxcode = 0x10FFFF, - codecvt_mode mode = codecvt_mode(0)) -{ - const uint8_t* frm_nxt = frm; - frm_nxt = frm; - if (mode & consume_header) - { - if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFF && frm_nxt[1] == 0xFE) - frm_nxt += 2; - } - for (size_t nchar16_t = 0; frm_nxt < frm_end - 1 && nchar16_t < mx; ++nchar16_t) - { - uint16_t c1 = static_cast(frm_nxt[1] << 8 | frm_nxt[0]); - if ((c1 & 0xF800) == 0xD800 || c1 > Maxcode) - break; - frm_nxt += 2; - } - return static_cast(frm_nxt - frm); -} - -// template <> class codecvt - -locale::id codecvt::id; - -codecvt::~codecvt() -{ -} - -codecvt::result -codecvt::do_out(state_type&, - const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ - const uint16_t* _frm = reinterpret_cast(frm); - const uint16_t* _frm_end = reinterpret_cast(frm_end); - const uint16_t* _frm_nxt = _frm; - uint8_t* _to = reinterpret_cast(to); - uint8_t* _to_end = reinterpret_cast(to_end); - uint8_t* _to_nxt = _to; - result r = utf16_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -codecvt::result -codecvt::do_in(state_type&, - const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, - intern_type* to, intern_type* to_end, intern_type*& to_nxt) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - const uint8_t* _frm_nxt = _frm; - uint16_t* _to = reinterpret_cast(to); - uint16_t* _to_end = reinterpret_cast(to_end); - uint16_t* _to_nxt = _to; - result r = utf8_to_utf16(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -codecvt::result -codecvt::do_unshift(state_type&, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - to_nxt = to; - return noconv; -} - -int -codecvt::do_encoding() const _NOEXCEPT -{ - return 0; -} - -bool -codecvt::do_always_noconv() const _NOEXCEPT -{ - return false; -} - -int -codecvt::do_length(state_type&, - const extern_type* frm, const extern_type* frm_end, size_t mx) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - return utf8_to_utf16_length(_frm, _frm_end, mx); -} - -int -codecvt::do_max_length() const _NOEXCEPT -{ - return 4; -} - -// template <> class codecvt - -locale::id codecvt::id; - -codecvt::~codecvt() -{ -} - -codecvt::result -codecvt::do_out(state_type&, - const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ - const uint32_t* _frm = reinterpret_cast(frm); - const uint32_t* _frm_end = reinterpret_cast(frm_end); - const uint32_t* _frm_nxt = _frm; - uint8_t* _to = reinterpret_cast(to); - uint8_t* _to_end = reinterpret_cast(to_end); - uint8_t* _to_nxt = _to; - result r = ucs4_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -codecvt::result -codecvt::do_in(state_type&, - const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, - intern_type* to, intern_type* to_end, intern_type*& to_nxt) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - const uint8_t* _frm_nxt = _frm; - uint32_t* _to = reinterpret_cast(to); - uint32_t* _to_end = reinterpret_cast(to_end); - uint32_t* _to_nxt = _to; - result r = utf8_to_ucs4(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -codecvt::result -codecvt::do_unshift(state_type&, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - to_nxt = to; - return noconv; -} - -int -codecvt::do_encoding() const _NOEXCEPT -{ - return 0; -} - -bool -codecvt::do_always_noconv() const _NOEXCEPT -{ - return false; -} - -int -codecvt::do_length(state_type&, - const extern_type* frm, const extern_type* frm_end, size_t mx) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - return utf8_to_ucs4_length(_frm, _frm_end, mx); -} - -int -codecvt::do_max_length() const _NOEXCEPT -{ - return 4; -} - -// __codecvt_utf8 - -__codecvt_utf8::result -__codecvt_utf8::do_out(state_type&, - const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ -#if _WIN32 - const uint16_t* _frm = reinterpret_cast(frm); - const uint16_t* _frm_end = reinterpret_cast(frm_end); - const uint16_t* _frm_nxt = _frm; -#else - const uint32_t* _frm = reinterpret_cast(frm); - const uint32_t* _frm_end = reinterpret_cast(frm_end); - const uint32_t* _frm_nxt = _frm; -#endif - uint8_t* _to = reinterpret_cast(to); - uint8_t* _to_end = reinterpret_cast(to_end); - uint8_t* _to_nxt = _to; -#if _WIN32 - result r = ucs2_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); -#else - result r = ucs4_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); -#endif - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf8::result -__codecvt_utf8::do_in(state_type&, - const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, - intern_type* to, intern_type* to_end, intern_type*& to_nxt) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - const uint8_t* _frm_nxt = _frm; -#if _WIN32 - uint16_t* _to = reinterpret_cast(to); - uint16_t* _to_end = reinterpret_cast(to_end); - uint16_t* _to_nxt = _to; - result r = utf8_to_ucs2(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); -#else - uint32_t* _to = reinterpret_cast(to); - uint32_t* _to_end = reinterpret_cast(to_end); - uint32_t* _to_nxt = _to; - result r = utf8_to_ucs4(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); -#endif - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf8::result -__codecvt_utf8::do_unshift(state_type&, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - to_nxt = to; - return noconv; -} - -int -__codecvt_utf8::do_encoding() const _NOEXCEPT -{ - return 0; -} - -bool -__codecvt_utf8::do_always_noconv() const _NOEXCEPT -{ - return false; -} - -int -__codecvt_utf8::do_length(state_type&, - const extern_type* frm, const extern_type* frm_end, size_t mx) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - return utf8_to_ucs4_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); -} - -int -__codecvt_utf8::do_max_length() const _NOEXCEPT -{ - if (_Mode_ & consume_header) - return 7; - return 4; -} - -// __codecvt_utf8 - -__codecvt_utf8::result -__codecvt_utf8::do_out(state_type&, - const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ - const uint16_t* _frm = reinterpret_cast(frm); - const uint16_t* _frm_end = reinterpret_cast(frm_end); - const uint16_t* _frm_nxt = _frm; - uint8_t* _to = reinterpret_cast(to); - uint8_t* _to_end = reinterpret_cast(to_end); - uint8_t* _to_nxt = _to; - result r = ucs2_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf8::result -__codecvt_utf8::do_in(state_type&, - const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, - intern_type* to, intern_type* to_end, intern_type*& to_nxt) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - const uint8_t* _frm_nxt = _frm; - uint16_t* _to = reinterpret_cast(to); - uint16_t* _to_end = reinterpret_cast(to_end); - uint16_t* _to_nxt = _to; - result r = utf8_to_ucs2(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf8::result -__codecvt_utf8::do_unshift(state_type&, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - to_nxt = to; - return noconv; -} - -int -__codecvt_utf8::do_encoding() const _NOEXCEPT -{ - return 0; -} - -bool -__codecvt_utf8::do_always_noconv() const _NOEXCEPT -{ - return false; -} - -int -__codecvt_utf8::do_length(state_type&, - const extern_type* frm, const extern_type* frm_end, size_t mx) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - return utf8_to_ucs2_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); -} - -int -__codecvt_utf8::do_max_length() const _NOEXCEPT -{ - if (_Mode_ & consume_header) - return 6; - return 3; -} - -// __codecvt_utf8 - -__codecvt_utf8::result -__codecvt_utf8::do_out(state_type&, - const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ - const uint32_t* _frm = reinterpret_cast(frm); - const uint32_t* _frm_end = reinterpret_cast(frm_end); - const uint32_t* _frm_nxt = _frm; - uint8_t* _to = reinterpret_cast(to); - uint8_t* _to_end = reinterpret_cast(to_end); - uint8_t* _to_nxt = _to; - result r = ucs4_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf8::result -__codecvt_utf8::do_in(state_type&, - const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, - intern_type* to, intern_type* to_end, intern_type*& to_nxt) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - const uint8_t* _frm_nxt = _frm; - uint32_t* _to = reinterpret_cast(to); - uint32_t* _to_end = reinterpret_cast(to_end); - uint32_t* _to_nxt = _to; - result r = utf8_to_ucs4(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf8::result -__codecvt_utf8::do_unshift(state_type&, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - to_nxt = to; - return noconv; -} - -int -__codecvt_utf8::do_encoding() const _NOEXCEPT -{ - return 0; -} - -bool -__codecvt_utf8::do_always_noconv() const _NOEXCEPT -{ - return false; -} - -int -__codecvt_utf8::do_length(state_type&, - const extern_type* frm, const extern_type* frm_end, size_t mx) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - return utf8_to_ucs4_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); -} - -int -__codecvt_utf8::do_max_length() const _NOEXCEPT -{ - if (_Mode_ & consume_header) - return 7; - return 4; -} - -// __codecvt_utf16 - -__codecvt_utf16::result -__codecvt_utf16::do_out(state_type&, - const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ - const uint32_t* _frm = reinterpret_cast(frm); - const uint32_t* _frm_end = reinterpret_cast(frm_end); - const uint32_t* _frm_nxt = _frm; - uint8_t* _to = reinterpret_cast(to); - uint8_t* _to_end = reinterpret_cast(to_end); - uint8_t* _to_nxt = _to; - result r = ucs4_to_utf16be(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf16::result -__codecvt_utf16::do_in(state_type&, - const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, - intern_type* to, intern_type* to_end, intern_type*& to_nxt) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - const uint8_t* _frm_nxt = _frm; - uint32_t* _to = reinterpret_cast(to); - uint32_t* _to_end = reinterpret_cast(to_end); - uint32_t* _to_nxt = _to; - result r = utf16be_to_ucs4(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf16::result -__codecvt_utf16::do_unshift(state_type&, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - to_nxt = to; - return noconv; -} - -int -__codecvt_utf16::do_encoding() const _NOEXCEPT -{ - return 0; -} - -bool -__codecvt_utf16::do_always_noconv() const _NOEXCEPT -{ - return false; -} - -int -__codecvt_utf16::do_length(state_type&, - const extern_type* frm, const extern_type* frm_end, size_t mx) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - return utf16be_to_ucs4_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); -} - -int -__codecvt_utf16::do_max_length() const _NOEXCEPT -{ - if (_Mode_ & consume_header) - return 6; - return 4; -} - -// __codecvt_utf16 - -__codecvt_utf16::result -__codecvt_utf16::do_out(state_type&, - const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ - const uint32_t* _frm = reinterpret_cast(frm); - const uint32_t* _frm_end = reinterpret_cast(frm_end); - const uint32_t* _frm_nxt = _frm; - uint8_t* _to = reinterpret_cast(to); - uint8_t* _to_end = reinterpret_cast(to_end); - uint8_t* _to_nxt = _to; - result r = ucs4_to_utf16le(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf16::result -__codecvt_utf16::do_in(state_type&, - const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, - intern_type* to, intern_type* to_end, intern_type*& to_nxt) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - const uint8_t* _frm_nxt = _frm; - uint32_t* _to = reinterpret_cast(to); - uint32_t* _to_end = reinterpret_cast(to_end); - uint32_t* _to_nxt = _to; - result r = utf16le_to_ucs4(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf16::result -__codecvt_utf16::do_unshift(state_type&, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - to_nxt = to; - return noconv; -} - -int -__codecvt_utf16::do_encoding() const _NOEXCEPT -{ - return 0; -} - -bool -__codecvt_utf16::do_always_noconv() const _NOEXCEPT -{ - return false; -} - -int -__codecvt_utf16::do_length(state_type&, - const extern_type* frm, const extern_type* frm_end, size_t mx) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - return utf16le_to_ucs4_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); -} - -int -__codecvt_utf16::do_max_length() const _NOEXCEPT -{ - if (_Mode_ & consume_header) - return 6; - return 4; -} - -// __codecvt_utf16 - -__codecvt_utf16::result -__codecvt_utf16::do_out(state_type&, - const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ - const uint16_t* _frm = reinterpret_cast(frm); - const uint16_t* _frm_end = reinterpret_cast(frm_end); - const uint16_t* _frm_nxt = _frm; - uint8_t* _to = reinterpret_cast(to); - uint8_t* _to_end = reinterpret_cast(to_end); - uint8_t* _to_nxt = _to; - result r = ucs2_to_utf16be(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf16::result -__codecvt_utf16::do_in(state_type&, - const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, - intern_type* to, intern_type* to_end, intern_type*& to_nxt) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - const uint8_t* _frm_nxt = _frm; - uint16_t* _to = reinterpret_cast(to); - uint16_t* _to_end = reinterpret_cast(to_end); - uint16_t* _to_nxt = _to; - result r = utf16be_to_ucs2(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf16::result -__codecvt_utf16::do_unshift(state_type&, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - to_nxt = to; - return noconv; -} - -int -__codecvt_utf16::do_encoding() const _NOEXCEPT -{ - return 0; -} - -bool -__codecvt_utf16::do_always_noconv() const _NOEXCEPT -{ - return false; -} - -int -__codecvt_utf16::do_length(state_type&, - const extern_type* frm, const extern_type* frm_end, size_t mx) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - return utf16be_to_ucs2_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); -} - -int -__codecvt_utf16::do_max_length() const _NOEXCEPT -{ - if (_Mode_ & consume_header) - return 4; - return 2; -} - -// __codecvt_utf16 - -__codecvt_utf16::result -__codecvt_utf16::do_out(state_type&, - const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ - const uint16_t* _frm = reinterpret_cast(frm); - const uint16_t* _frm_end = reinterpret_cast(frm_end); - const uint16_t* _frm_nxt = _frm; - uint8_t* _to = reinterpret_cast(to); - uint8_t* _to_end = reinterpret_cast(to_end); - uint8_t* _to_nxt = _to; - result r = ucs2_to_utf16le(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf16::result -__codecvt_utf16::do_in(state_type&, - const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, - intern_type* to, intern_type* to_end, intern_type*& to_nxt) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - const uint8_t* _frm_nxt = _frm; - uint16_t* _to = reinterpret_cast(to); - uint16_t* _to_end = reinterpret_cast(to_end); - uint16_t* _to_nxt = _to; - result r = utf16le_to_ucs2(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf16::result -__codecvt_utf16::do_unshift(state_type&, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - to_nxt = to; - return noconv; -} - -int -__codecvt_utf16::do_encoding() const _NOEXCEPT -{ - return 0; -} - -bool -__codecvt_utf16::do_always_noconv() const _NOEXCEPT -{ - return false; -} - -int -__codecvt_utf16::do_length(state_type&, - const extern_type* frm, const extern_type* frm_end, size_t mx) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - return utf16le_to_ucs2_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); -} - -int -__codecvt_utf16::do_max_length() const _NOEXCEPT -{ - if (_Mode_ & consume_header) - return 4; - return 2; -} - -// __codecvt_utf16 - -__codecvt_utf16::result -__codecvt_utf16::do_out(state_type&, - const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ - const uint32_t* _frm = reinterpret_cast(frm); - const uint32_t* _frm_end = reinterpret_cast(frm_end); - const uint32_t* _frm_nxt = _frm; - uint8_t* _to = reinterpret_cast(to); - uint8_t* _to_end = reinterpret_cast(to_end); - uint8_t* _to_nxt = _to; - result r = ucs4_to_utf16be(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf16::result -__codecvt_utf16::do_in(state_type&, - const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, - intern_type* to, intern_type* to_end, intern_type*& to_nxt) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - const uint8_t* _frm_nxt = _frm; - uint32_t* _to = reinterpret_cast(to); - uint32_t* _to_end = reinterpret_cast(to_end); - uint32_t* _to_nxt = _to; - result r = utf16be_to_ucs4(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf16::result -__codecvt_utf16::do_unshift(state_type&, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - to_nxt = to; - return noconv; -} - -int -__codecvt_utf16::do_encoding() const _NOEXCEPT -{ - return 0; -} - -bool -__codecvt_utf16::do_always_noconv() const _NOEXCEPT -{ - return false; -} - -int -__codecvt_utf16::do_length(state_type&, - const extern_type* frm, const extern_type* frm_end, size_t mx) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - return utf16be_to_ucs4_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); -} - -int -__codecvt_utf16::do_max_length() const _NOEXCEPT -{ - if (_Mode_ & consume_header) - return 6; - return 4; -} - -// __codecvt_utf16 - -__codecvt_utf16::result -__codecvt_utf16::do_out(state_type&, - const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ - const uint32_t* _frm = reinterpret_cast(frm); - const uint32_t* _frm_end = reinterpret_cast(frm_end); - const uint32_t* _frm_nxt = _frm; - uint8_t* _to = reinterpret_cast(to); - uint8_t* _to_end = reinterpret_cast(to_end); - uint8_t* _to_nxt = _to; - result r = ucs4_to_utf16le(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf16::result -__codecvt_utf16::do_in(state_type&, - const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, - intern_type* to, intern_type* to_end, intern_type*& to_nxt) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - const uint8_t* _frm_nxt = _frm; - uint32_t* _to = reinterpret_cast(to); - uint32_t* _to_end = reinterpret_cast(to_end); - uint32_t* _to_nxt = _to; - result r = utf16le_to_ucs4(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf16::result -__codecvt_utf16::do_unshift(state_type&, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - to_nxt = to; - return noconv; -} - -int -__codecvt_utf16::do_encoding() const _NOEXCEPT -{ - return 0; -} - -bool -__codecvt_utf16::do_always_noconv() const _NOEXCEPT -{ - return false; -} - -int -__codecvt_utf16::do_length(state_type&, - const extern_type* frm, const extern_type* frm_end, size_t mx) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - return utf16le_to_ucs4_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); -} - -int -__codecvt_utf16::do_max_length() const _NOEXCEPT -{ - if (_Mode_ & consume_header) - return 6; - return 4; -} - -// __codecvt_utf8_utf16 - -__codecvt_utf8_utf16::result -__codecvt_utf8_utf16::do_out(state_type&, - const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ - const uint32_t* _frm = reinterpret_cast(frm); - const uint32_t* _frm_end = reinterpret_cast(frm_end); - const uint32_t* _frm_nxt = _frm; - uint8_t* _to = reinterpret_cast(to); - uint8_t* _to_end = reinterpret_cast(to_end); - uint8_t* _to_nxt = _to; - result r = utf16_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf8_utf16::result -__codecvt_utf8_utf16::do_in(state_type&, - const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, - intern_type* to, intern_type* to_end, intern_type*& to_nxt) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - const uint8_t* _frm_nxt = _frm; - uint32_t* _to = reinterpret_cast(to); - uint32_t* _to_end = reinterpret_cast(to_end); - uint32_t* _to_nxt = _to; - result r = utf8_to_utf16(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf8_utf16::result -__codecvt_utf8_utf16::do_unshift(state_type&, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - to_nxt = to; - return noconv; -} - -int -__codecvt_utf8_utf16::do_encoding() const _NOEXCEPT -{ - return 0; -} - -bool -__codecvt_utf8_utf16::do_always_noconv() const _NOEXCEPT -{ - return false; -} - -int -__codecvt_utf8_utf16::do_length(state_type&, - const extern_type* frm, const extern_type* frm_end, size_t mx) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - return utf8_to_utf16_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); -} - -int -__codecvt_utf8_utf16::do_max_length() const _NOEXCEPT -{ - if (_Mode_ & consume_header) - return 7; - return 4; -} - -// __codecvt_utf8_utf16 - -__codecvt_utf8_utf16::result -__codecvt_utf8_utf16::do_out(state_type&, - const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ - const uint16_t* _frm = reinterpret_cast(frm); - const uint16_t* _frm_end = reinterpret_cast(frm_end); - const uint16_t* _frm_nxt = _frm; - uint8_t* _to = reinterpret_cast(to); - uint8_t* _to_end = reinterpret_cast(to_end); - uint8_t* _to_nxt = _to; - result r = utf16_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf8_utf16::result -__codecvt_utf8_utf16::do_in(state_type&, - const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, - intern_type* to, intern_type* to_end, intern_type*& to_nxt) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - const uint8_t* _frm_nxt = _frm; - uint16_t* _to = reinterpret_cast(to); - uint16_t* _to_end = reinterpret_cast(to_end); - uint16_t* _to_nxt = _to; - result r = utf8_to_utf16(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf8_utf16::result -__codecvt_utf8_utf16::do_unshift(state_type&, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - to_nxt = to; - return noconv; -} - -int -__codecvt_utf8_utf16::do_encoding() const _NOEXCEPT -{ - return 0; -} - -bool -__codecvt_utf8_utf16::do_always_noconv() const _NOEXCEPT -{ - return false; -} - -int -__codecvt_utf8_utf16::do_length(state_type&, - const extern_type* frm, const extern_type* frm_end, size_t mx) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - return utf8_to_utf16_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); -} - -int -__codecvt_utf8_utf16::do_max_length() const _NOEXCEPT -{ - if (_Mode_ & consume_header) - return 7; - return 4; -} - -// __codecvt_utf8_utf16 - -__codecvt_utf8_utf16::result -__codecvt_utf8_utf16::do_out(state_type&, - const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, - extern_type* to, extern_type* to_end, extern_type*& to_nxt) const -{ - const uint32_t* _frm = reinterpret_cast(frm); - const uint32_t* _frm_end = reinterpret_cast(frm_end); - const uint32_t* _frm_nxt = _frm; - uint8_t* _to = reinterpret_cast(to); - uint8_t* _to_end = reinterpret_cast(to_end); - uint8_t* _to_nxt = _to; - result r = utf16_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf8_utf16::result -__codecvt_utf8_utf16::do_in(state_type&, - const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, - intern_type* to, intern_type* to_end, intern_type*& to_nxt) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - const uint8_t* _frm_nxt = _frm; - uint32_t* _to = reinterpret_cast(to); - uint32_t* _to_end = reinterpret_cast(to_end); - uint32_t* _to_nxt = _to; - result r = utf8_to_utf16(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, - _Maxcode_, _Mode_); - frm_nxt = frm + (_frm_nxt - _frm); - to_nxt = to + (_to_nxt - _to); - return r; -} - -__codecvt_utf8_utf16::result -__codecvt_utf8_utf16::do_unshift(state_type&, - extern_type* to, extern_type*, extern_type*& to_nxt) const -{ - to_nxt = to; - return noconv; -} - -int -__codecvt_utf8_utf16::do_encoding() const _NOEXCEPT -{ - return 0; -} - -bool -__codecvt_utf8_utf16::do_always_noconv() const _NOEXCEPT -{ - return false; -} - -int -__codecvt_utf8_utf16::do_length(state_type&, - const extern_type* frm, const extern_type* frm_end, size_t mx) const -{ - const uint8_t* _frm = reinterpret_cast(frm); - const uint8_t* _frm_end = reinterpret_cast(frm_end); - return utf8_to_utf16_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); -} - -int -__codecvt_utf8_utf16::do_max_length() const _NOEXCEPT -{ - if (_Mode_ & consume_header) - return 7; - return 4; -} - -// __narrow_to_utf8<16> - -__narrow_to_utf8<16>::~__narrow_to_utf8() -{ -} - -// __narrow_to_utf8<32> - -__narrow_to_utf8<32>::~__narrow_to_utf8() -{ -} - -// __widen_from_utf8<16> - -__widen_from_utf8<16>::~__widen_from_utf8() -{ -} - -// __widen_from_utf8<32> - -__widen_from_utf8<32>::~__widen_from_utf8() -{ -} - -// numpunct && numpunct - -locale::id numpunct< char >::id; -locale::id numpunct::id; - -numpunct::numpunct(size_t refs) - : locale::facet(refs), - __decimal_point_('.'), - __thousands_sep_(',') -{ -} - -numpunct::numpunct(size_t refs) - : locale::facet(refs), - __decimal_point_(L'.'), - __thousands_sep_(L',') -{ -} - -numpunct::~numpunct() -{ -} - -numpunct::~numpunct() -{ -} - - char numpunct< char >::do_decimal_point() const {return __decimal_point_;} -wchar_t numpunct::do_decimal_point() const {return __decimal_point_;} - - char numpunct< char >::do_thousands_sep() const {return __thousands_sep_;} -wchar_t numpunct::do_thousands_sep() const {return __thousands_sep_;} - -string numpunct< char >::do_grouping() const {return __grouping_;} -string numpunct::do_grouping() const {return __grouping_;} - - string numpunct< char >::do_truename() const {return "true";} -wstring numpunct::do_truename() const {return L"true";} - - string numpunct< char >::do_falsename() const {return "false";} -wstring numpunct::do_falsename() const {return L"false";} - -// numpunct_byname - -numpunct_byname::numpunct_byname(const char* nm, size_t refs) - : numpunct(refs) -{ - __init(nm); -} - -numpunct_byname::numpunct_byname(const string& nm, size_t refs) - : numpunct(refs) -{ - __init(nm.c_str()); -} - -numpunct_byname::~numpunct_byname() -{ -} - -void -numpunct_byname::__init(const char* nm) -{ - if (strcmp(nm, "C") != 0) - { - __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); -#ifndef _LIBCPP_NO_EXCEPTIONS - if (loc == nullptr) - throw runtime_error("numpunct_byname::numpunct_byname" - " failed to construct for " + string(nm)); -#endif // _LIBCPP_NO_EXCEPTIONS -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - lconv* lc = localeconv_l(loc.get()); -#else - lconv* lc = __localeconv_l(loc.get()); -#endif - if (*lc->decimal_point) - __decimal_point_ = *lc->decimal_point; - if (*lc->thousands_sep) - __thousands_sep_ = *lc->thousands_sep; - __grouping_ = lc->grouping; - // localization for truename and falsename is not available - } -} - -// numpunct_byname - -numpunct_byname::numpunct_byname(const char* nm, size_t refs) - : numpunct(refs) -{ - __init(nm); -} - -numpunct_byname::numpunct_byname(const string& nm, size_t refs) - : numpunct(refs) -{ - __init(nm.c_str()); -} - -numpunct_byname::~numpunct_byname() -{ -} - -void -numpunct_byname::__init(const char* nm) -{ - if (strcmp(nm, "C") != 0) - { - __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); -#ifndef _LIBCPP_NO_EXCEPTIONS - if (loc == nullptr) - throw runtime_error("numpunct_byname::numpunct_byname" - " failed to construct for " + string(nm)); -#endif // _LIBCPP_NO_EXCEPTIONS -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - lconv* lc = localeconv_l(loc.get()); -#else - lconv* lc = __localeconv_l(loc.get()); -#endif - if (*lc->decimal_point) - __decimal_point_ = *lc->decimal_point; - if (*lc->thousands_sep) - __thousands_sep_ = *lc->thousands_sep; - __grouping_ = lc->grouping; - // locallization for truename and falsename is not available - } -} - -// num_get helpers - -int -__num_get_base::__get_base(ios_base& iob) -{ - ios_base::fmtflags __basefield = iob.flags() & ios_base::basefield; - if (__basefield == ios_base::oct) - return 8; - else if (__basefield == ios_base::hex) - return 16; - else if (__basefield == 0) - return 0; - return 10; -} - -const char __num_get_base::__src[33] = "0123456789abcdefABCDEFxX+-pPiInN"; - -void -__check_grouping(const string& __grouping, unsigned* __g, unsigned* __g_end, - ios_base::iostate& __err) -{ - if (__grouping.size() != 0) - { - reverse(__g, __g_end); - const char* __ig = __grouping.data(); - const char* __eg = __ig + __grouping.size(); - for (unsigned* __r = __g; __r < __g_end-1; ++__r) - { - if (0 < *__ig && *__ig < numeric_limits::max()) - { - if (static_cast(*__ig) != *__r) - { - __err = ios_base::failbit; - return; - } - } - if (__eg - __ig > 1) - ++__ig; - } - if (0 < *__ig && *__ig < numeric_limits::max()) - { - if (static_cast(*__ig) < __g_end[-1] || __g_end[-1] == 0) - __err = ios_base::failbit; - } - } -} - -void -__num_put_base::__format_int(char* __fmtp, const char* __len, bool __signd, - ios_base::fmtflags __flags) -{ - if (__flags & ios_base::showpos) - *__fmtp++ = '+'; - if (__flags & ios_base::showbase) - *__fmtp++ = '#'; - while(*__len) - *__fmtp++ = *__len++; - if ((__flags & ios_base::basefield) == ios_base::oct) - *__fmtp = 'o'; - else if ((__flags & ios_base::basefield) == ios_base::hex) - { - if (__flags & ios_base::uppercase) - *__fmtp = 'X'; - else - *__fmtp = 'x'; - } - else if (__signd) - *__fmtp = 'd'; - else - *__fmtp = 'u'; -} - -bool -__num_put_base::__format_float(char* __fmtp, const char* __len, - ios_base::fmtflags __flags) -{ - bool specify_precision = true; - if (__flags & ios_base::showpos) - *__fmtp++ = '+'; - if (__flags & ios_base::showpoint) - *__fmtp++ = '#'; - ios_base::fmtflags floatfield = __flags & ios_base::floatfield; - bool uppercase = (__flags & ios_base::uppercase) != 0; - if (floatfield == (ios_base::fixed | ios_base::scientific)) - specify_precision = false; - else - { - *__fmtp++ = '.'; - *__fmtp++ = '*'; - } - while(*__len) - *__fmtp++ = *__len++; - if (floatfield == ios_base::fixed) - { - if (uppercase) - *__fmtp = 'F'; - else - *__fmtp = 'f'; - } - else if (floatfield == ios_base::scientific) - { - if (uppercase) - *__fmtp = 'E'; - else - *__fmtp = 'e'; - } - else if (floatfield == (ios_base::fixed | ios_base::scientific)) - { - if (uppercase) - *__fmtp = 'A'; - else - *__fmtp = 'a'; - } - else - { - if (uppercase) - *__fmtp = 'G'; - else - *__fmtp = 'g'; - } - return specify_precision; -} - -char* -__num_put_base::__identify_padding(char* __nb, char* __ne, - const ios_base& __iob) -{ - switch (__iob.flags() & ios_base::adjustfield) - { - case ios_base::internal: - if (__nb[0] == '-' || __nb[0] == '+') - return __nb+1; - if (__ne - __nb >= 2 && __nb[0] == '0' - && (__nb[1] == 'x' || __nb[1] == 'X')) - return __nb+2; - break; - case ios_base::left: - return __ne; - case ios_base::right: - default: - break; - } - return __nb; -} - -// time_get - -static -string* -init_weeks() -{ - static string weeks[14]; - weeks[0] = "Sunday"; - weeks[1] = "Monday"; - weeks[2] = "Tuesday"; - weeks[3] = "Wednesday"; - weeks[4] = "Thursday"; - weeks[5] = "Friday"; - weeks[6] = "Saturday"; - weeks[7] = "Sun"; - weeks[8] = "Mon"; - weeks[9] = "Tue"; - weeks[10] = "Wed"; - weeks[11] = "Thu"; - weeks[12] = "Fri"; - weeks[13] = "Sat"; - return weeks; -} - -static -wstring* -init_wweeks() -{ - static wstring weeks[14]; - weeks[0] = L"Sunday"; - weeks[1] = L"Monday"; - weeks[2] = L"Tuesday"; - weeks[3] = L"Wednesday"; - weeks[4] = L"Thursday"; - weeks[5] = L"Friday"; - weeks[6] = L"Saturday"; - weeks[7] = L"Sun"; - weeks[8] = L"Mon"; - weeks[9] = L"Tue"; - weeks[10] = L"Wed"; - weeks[11] = L"Thu"; - weeks[12] = L"Fri"; - weeks[13] = L"Sat"; - return weeks; -} - -template <> -const string* -__time_get_c_storage::__weeks() const -{ - static const string* weeks = init_weeks(); - return weeks; -} - -template <> -const wstring* -__time_get_c_storage::__weeks() const -{ - static const wstring* weeks = init_wweeks(); - return weeks; -} - -static -string* -init_months() -{ - static string months[24]; - months[0] = "January"; - months[1] = "February"; - months[2] = "March"; - months[3] = "April"; - months[4] = "May"; - months[5] = "June"; - months[6] = "July"; - months[7] = "August"; - months[8] = "September"; - months[9] = "October"; - months[10] = "November"; - months[11] = "December"; - months[12] = "Jan"; - months[13] = "Feb"; - months[14] = "Mar"; - months[15] = "Apr"; - months[16] = "May"; - months[17] = "Jun"; - months[18] = "Jul"; - months[19] = "Aug"; - months[20] = "Sep"; - months[21] = "Oct"; - months[22] = "Nov"; - months[23] = "Dec"; - return months; -} - -static -wstring* -init_wmonths() -{ - static wstring months[24]; - months[0] = L"January"; - months[1] = L"February"; - months[2] = L"March"; - months[3] = L"April"; - months[4] = L"May"; - months[5] = L"June"; - months[6] = L"July"; - months[7] = L"August"; - months[8] = L"September"; - months[9] = L"October"; - months[10] = L"November"; - months[11] = L"December"; - months[12] = L"Jan"; - months[13] = L"Feb"; - months[14] = L"Mar"; - months[15] = L"Apr"; - months[16] = L"May"; - months[17] = L"Jun"; - months[18] = L"Jul"; - months[19] = L"Aug"; - months[20] = L"Sep"; - months[21] = L"Oct"; - months[22] = L"Nov"; - months[23] = L"Dec"; - return months; -} - -template <> -const string* -__time_get_c_storage::__months() const -{ - static const string* months = init_months(); - return months; -} - -template <> -const wstring* -__time_get_c_storage::__months() const -{ - static const wstring* months = init_wmonths(); - return months; -} - -static -string* -init_am_pm() -{ - static string am_pm[24]; - am_pm[0] = "AM"; - am_pm[1] = "PM"; - return am_pm; -} - -static -wstring* -init_wam_pm() -{ - static wstring am_pm[24]; - am_pm[0] = L"AM"; - am_pm[1] = L"PM"; - return am_pm; -} - -template <> -const string* -__time_get_c_storage::__am_pm() const -{ - static const string* am_pm = init_am_pm(); - return am_pm; -} - -template <> -const wstring* -__time_get_c_storage::__am_pm() const -{ - static const wstring* am_pm = init_wam_pm(); - return am_pm; -} - -template <> -const string& -__time_get_c_storage::__x() const -{ - static string s("%m/%d/%y"); - return s; -} - -template <> -const wstring& -__time_get_c_storage::__x() const -{ - static wstring s(L"%m/%d/%y"); - return s; -} - -template <> -const string& -__time_get_c_storage::__X() const -{ - static string s("%H:%M:%S"); - return s; -} - -template <> -const wstring& -__time_get_c_storage::__X() const -{ - static wstring s(L"%H:%M:%S"); - return s; -} - -template <> -const string& -__time_get_c_storage::__c() const -{ - static string s("%a %b %d %H:%M:%S %Y"); - return s; -} - -template <> -const wstring& -__time_get_c_storage::__c() const -{ - static wstring s(L"%a %b %d %H:%M:%S %Y"); - return s; -} - -template <> -const string& -__time_get_c_storage::__r() const -{ - static string s("%I:%M:%S %p"); - return s; -} - -template <> -const wstring& -__time_get_c_storage::__r() const -{ - static wstring s(L"%I:%M:%S %p"); - return s; -} - -// time_get_byname - -__time_get::__time_get(const char* nm) - : __loc_(newlocale(LC_ALL_MASK, nm, 0)) -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - if (__loc_ == 0) - throw runtime_error("time_get_byname" - " failed to construct for " + string(nm)); -#endif // _LIBCPP_NO_EXCEPTIONS -} - -__time_get::__time_get(const string& nm) - : __loc_(newlocale(LC_ALL_MASK, nm.c_str(), 0)) -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - if (__loc_ == 0) - throw runtime_error("time_get_byname" - " failed to construct for " + nm); -#endif // _LIBCPP_NO_EXCEPTIONS -} - -__time_get::~__time_get() -{ - freelocale(__loc_); -} -#if defined(__clang__) -#pragma clang diagnostic ignored "-Wmissing-field-initializers" -#endif -#if defined(__GNUG__) -#pragma GCC diagnostic ignored "-Wmissing-field-initializers" -#endif - -template <> -string -__time_get_storage::__analyze(char fmt, const ctype& ct) -{ - tm t = {0}; - t.tm_sec = 59; - t.tm_min = 55; - t.tm_hour = 23; - t.tm_mday = 31; - t.tm_mon = 11; - t.tm_year = 161; - t.tm_wday = 6; - t.tm_yday = 364; - t.tm_isdst = -1; - char buf[100]; - char f[3] = {0}; - f[0] = '%'; - f[1] = fmt; - size_t n = strftime_l(buf, countof(buf), f, &t, __loc_); - char* bb = buf; - char* be = buf + n; - string result; - while (bb != be) - { - if (ct.is(ctype_base::space, *bb)) - { - result.push_back(' '); - for (++bb; bb != be && ct.is(ctype_base::space, *bb); ++bb) - ; - continue; - } - char* w = bb; - ios_base::iostate err = ios_base::goodbit; - ptrdiff_t i = __scan_keyword(w, be, this->__weeks_, this->__weeks_+14, - ct, err, false) - - this->__weeks_; - if (i < 14) - { - result.push_back('%'); - if (i < 7) - result.push_back('A'); - else - result.push_back('a'); - bb = w; - continue; - } - w = bb; - i = __scan_keyword(w, be, this->__months_, this->__months_+24, - ct, err, false) - - this->__months_; - if (i < 24) - { - result.push_back('%'); - if (i < 12) - result.push_back('B'); - else - result.push_back('b'); - if (fmt == 'x' && ct.is(ctype_base::digit, this->__months_[i][0])) - result.back() = 'm'; - bb = w; - continue; - } - if (this->__am_pm_[0].size() + this->__am_pm_[1].size() > 0) - { - w = bb; - i = __scan_keyword(w, be, this->__am_pm_, this->__am_pm_+2, - ct, err, false) - this->__am_pm_; - if (i < 2) - { - result.push_back('%'); - result.push_back('p'); - bb = w; - continue; - } - } - w = bb; - if (ct.is(ctype_base::digit, *bb)) - { - switch(__get_up_to_n_digits(bb, be, err, ct, 4)) - { - case 6: - result.push_back('%'); - result.push_back('w'); - break; - case 7: - result.push_back('%'); - result.push_back('u'); - break; - case 11: - result.push_back('%'); - result.push_back('I'); - break; - case 12: - result.push_back('%'); - result.push_back('m'); - break; - case 23: - result.push_back('%'); - result.push_back('H'); - break; - case 31: - result.push_back('%'); - result.push_back('d'); - break; - case 55: - result.push_back('%'); - result.push_back('M'); - break; - case 59: - result.push_back('%'); - result.push_back('S'); - break; - case 61: - result.push_back('%'); - result.push_back('y'); - break; - case 364: - result.push_back('%'); - result.push_back('j'); - break; - case 2061: - result.push_back('%'); - result.push_back('Y'); - break; - default: - for (; w != bb; ++w) - result.push_back(*w); - break; - } - continue; - } - if (*bb == '%') - { - result.push_back('%'); - result.push_back('%'); - ++bb; - continue; - } - result.push_back(*bb); - ++bb; - } - return result; -} - -#if defined(__clang__) -#pragma clang diagnostic ignored "-Wmissing-braces" -#endif - -template <> -wstring -__time_get_storage::__analyze(char fmt, const ctype& ct) -{ - tm t = {0}; - t.tm_sec = 59; - t.tm_min = 55; - t.tm_hour = 23; - t.tm_mday = 31; - t.tm_mon = 11; - t.tm_year = 161; - t.tm_wday = 6; - t.tm_yday = 364; - t.tm_isdst = -1; - char buf[100]; - char f[3] = {0}; - f[0] = '%'; - f[1] = fmt; - strftime_l(buf, countof(buf), f, &t, __loc_); - wchar_t wbuf[100]; - wchar_t* wbb = wbuf; - mbstate_t mb = {0}; - const char* bb = buf; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - size_t j = mbsrtowcs_l( wbb, &bb, countof(wbuf), &mb, __loc_); -#else - size_t j = __mbsrtowcs_l( wbb, &bb, countof(wbuf), &mb, __loc_); -#endif - if (j == size_t(-1)) - __throw_runtime_error("locale not supported"); - wchar_t* wbe = wbb + j; - wstring result; - while (wbb != wbe) - { - if (ct.is(ctype_base::space, *wbb)) - { - result.push_back(L' '); - for (++wbb; wbb != wbe && ct.is(ctype_base::space, *wbb); ++wbb) - ; - continue; - } - wchar_t* w = wbb; - ios_base::iostate err = ios_base::goodbit; - ptrdiff_t i = __scan_keyword(w, wbe, this->__weeks_, this->__weeks_+14, - ct, err, false) - - this->__weeks_; - if (i < 14) - { - result.push_back(L'%'); - if (i < 7) - result.push_back(L'A'); - else - result.push_back(L'a'); - wbb = w; - continue; - } - w = wbb; - i = __scan_keyword(w, wbe, this->__months_, this->__months_+24, - ct, err, false) - - this->__months_; - if (i < 24) - { - result.push_back(L'%'); - if (i < 12) - result.push_back(L'B'); - else - result.push_back(L'b'); - if (fmt == 'x' && ct.is(ctype_base::digit, this->__months_[i][0])) - result.back() = L'm'; - wbb = w; - continue; - } - if (this->__am_pm_[0].size() + this->__am_pm_[1].size() > 0) - { - w = wbb; - i = __scan_keyword(w, wbe, this->__am_pm_, this->__am_pm_+2, - ct, err, false) - this->__am_pm_; - if (i < 2) - { - result.push_back(L'%'); - result.push_back(L'p'); - wbb = w; - continue; - } - } - w = wbb; - if (ct.is(ctype_base::digit, *wbb)) - { - switch(__get_up_to_n_digits(wbb, wbe, err, ct, 4)) - { - case 6: - result.push_back(L'%'); - result.push_back(L'w'); - break; - case 7: - result.push_back(L'%'); - result.push_back(L'u'); - break; - case 11: - result.push_back(L'%'); - result.push_back(L'I'); - break; - case 12: - result.push_back(L'%'); - result.push_back(L'm'); - break; - case 23: - result.push_back(L'%'); - result.push_back(L'H'); - break; - case 31: - result.push_back(L'%'); - result.push_back(L'd'); - break; - case 55: - result.push_back(L'%'); - result.push_back(L'M'); - break; - case 59: - result.push_back(L'%'); - result.push_back(L'S'); - break; - case 61: - result.push_back(L'%'); - result.push_back(L'y'); - break; - case 364: - result.push_back(L'%'); - result.push_back(L'j'); - break; - case 2061: - result.push_back(L'%'); - result.push_back(L'Y'); - break; - default: - for (; w != wbb; ++w) - result.push_back(*w); - break; - } - continue; - } - if (ct.narrow(*wbb, 0) == '%') - { - result.push_back(L'%'); - result.push_back(L'%'); - ++wbb; - continue; - } - result.push_back(*wbb); - ++wbb; - } - return result; -} - -template <> -void -__time_get_storage::init(const ctype& ct) -{ - tm t = {0}; - char buf[100]; - // __weeks_ - for (int i = 0; i < 7; ++i) - { - t.tm_wday = i; - strftime_l(buf, countof(buf), "%A", &t, __loc_); - __weeks_[i] = buf; - strftime_l(buf, countof(buf), "%a", &t, __loc_); - __weeks_[i+7] = buf; - } - // __months_ - for (int i = 0; i < 12; ++i) - { - t.tm_mon = i; - strftime_l(buf, countof(buf), "%B", &t, __loc_); - __months_[i] = buf; - strftime_l(buf, countof(buf), "%b", &t, __loc_); - __months_[i+12] = buf; - } - // __am_pm_ - t.tm_hour = 1; - strftime_l(buf, countof(buf), "%p", &t, __loc_); - __am_pm_[0] = buf; - t.tm_hour = 13; - strftime_l(buf, countof(buf), "%p", &t, __loc_); - __am_pm_[1] = buf; - __c_ = __analyze('c', ct); - __r_ = __analyze('r', ct); - __x_ = __analyze('x', ct); - __X_ = __analyze('X', ct); -} - -template <> -void -__time_get_storage::init(const ctype& ct) -{ - tm t = {0}; - char buf[100]; - wchar_t wbuf[100]; - wchar_t* wbe; - mbstate_t mb = {0}; - // __weeks_ - for (int i = 0; i < 7; ++i) - { - t.tm_wday = i; - strftime_l(buf, countof(buf), "%A", &t, __loc_); - mb = mbstate_t(); - const char* bb = buf; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - size_t j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); -#else - size_t j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); -#endif - if (j == size_t(-1)) - __throw_runtime_error("locale not supported"); - wbe = wbuf + j; - __weeks_[i].assign(wbuf, wbe); - strftime_l(buf, countof(buf), "%a", &t, __loc_); - mb = mbstate_t(); - bb = buf; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); -#else - j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); -#endif - if (j == size_t(-1)) - __throw_runtime_error("locale not supported"); - wbe = wbuf + j; - __weeks_[i+7].assign(wbuf, wbe); - } - // __months_ - for (int i = 0; i < 12; ++i) - { - t.tm_mon = i; - strftime_l(buf, countof(buf), "%B", &t, __loc_); - mb = mbstate_t(); - const char* bb = buf; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - size_t j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); -#else - size_t j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); -#endif - if (j == size_t(-1)) - __throw_runtime_error("locale not supported"); - wbe = wbuf + j; - __months_[i].assign(wbuf, wbe); - strftime_l(buf, countof(buf), "%b", &t, __loc_); - mb = mbstate_t(); - bb = buf; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); -#else - j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); -#endif - if (j == size_t(-1)) - __throw_runtime_error("locale not supported"); - wbe = wbuf + j; - __months_[i+12].assign(wbuf, wbe); - } - // __am_pm_ - t.tm_hour = 1; - strftime_l(buf, countof(buf), "%p", &t, __loc_); - mb = mbstate_t(); - const char* bb = buf; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - size_t j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); -#else - size_t j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); -#endif - if (j == size_t(-1)) - __throw_runtime_error("locale not supported"); - wbe = wbuf + j; - __am_pm_[0].assign(wbuf, wbe); - t.tm_hour = 13; - strftime_l(buf, countof(buf), "%p", &t, __loc_); - mb = mbstate_t(); - bb = buf; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); -#else - j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); -#endif - if (j == size_t(-1)) - __throw_runtime_error("locale not supported"); - wbe = wbuf + j; - __am_pm_[1].assign(wbuf, wbe); - __c_ = __analyze('c', ct); - __r_ = __analyze('r', ct); - __x_ = __analyze('x', ct); - __X_ = __analyze('X', ct); -} - -template -struct _LIBCPP_HIDDEN __time_get_temp - : public ctype_byname -{ - explicit __time_get_temp(const char* nm) - : ctype_byname(nm, 1) {} - explicit __time_get_temp(const string& nm) - : ctype_byname(nm, 1) {} -}; - -template <> -__time_get_storage::__time_get_storage(const char* __nm) - : __time_get(__nm) -{ - const __time_get_temp ct(__nm); - init(ct); -} - -template <> -__time_get_storage::__time_get_storage(const string& __nm) - : __time_get(__nm) -{ - const __time_get_temp ct(__nm); - init(ct); -} - -template <> -__time_get_storage::__time_get_storage(const char* __nm) - : __time_get(__nm) -{ - const __time_get_temp ct(__nm); - init(ct); -} - -template <> -__time_get_storage::__time_get_storage(const string& __nm) - : __time_get(__nm) -{ - const __time_get_temp ct(__nm); - init(ct); -} - -template <> -time_base::dateorder -__time_get_storage::__do_date_order() const -{ - unsigned i; - for (i = 0; i < __x_.size(); ++i) - if (__x_[i] == '%') - break; - ++i; - switch (__x_[i]) - { - case 'y': - case 'Y': - for (++i; i < __x_.size(); ++i) - if (__x_[i] == '%') - break; - if (i == __x_.size()) - break; - ++i; - switch (__x_[i]) - { - case 'm': - for (++i; i < __x_.size(); ++i) - if (__x_[i] == '%') - break; - if (i == __x_.size()) - break; - ++i; - if (__x_[i] == 'd') - return time_base::ymd; - break; - case 'd': - for (++i; i < __x_.size(); ++i) - if (__x_[i] == '%') - break; - if (i == __x_.size()) - break; - ++i; - if (__x_[i] == 'm') - return time_base::ydm; - break; - } - break; - case 'm': - for (++i; i < __x_.size(); ++i) - if (__x_[i] == '%') - break; - if (i == __x_.size()) - break; - ++i; - if (__x_[i] == 'd') - { - for (++i; i < __x_.size(); ++i) - if (__x_[i] == '%') - break; - if (i == __x_.size()) - break; - ++i; - if (__x_[i] == 'y' || __x_[i] == 'Y') - return time_base::mdy; - break; - } - break; - case 'd': - for (++i; i < __x_.size(); ++i) - if (__x_[i] == '%') - break; - if (i == __x_.size()) - break; - ++i; - if (__x_[i] == 'm') - { - for (++i; i < __x_.size(); ++i) - if (__x_[i] == '%') - break; - if (i == __x_.size()) - break; - ++i; - if (__x_[i] == 'y' || __x_[i] == 'Y') - return time_base::dmy; - break; - } - break; - } - return time_base::no_order; -} - -template <> -time_base::dateorder -__time_get_storage::__do_date_order() const -{ - unsigned i; - for (i = 0; i < __x_.size(); ++i) - if (__x_[i] == L'%') - break; - ++i; - switch (__x_[i]) - { - case L'y': - case L'Y': - for (++i; i < __x_.size(); ++i) - if (__x_[i] == L'%') - break; - if (i == __x_.size()) - break; - ++i; - switch (__x_[i]) - { - case L'm': - for (++i; i < __x_.size(); ++i) - if (__x_[i] == L'%') - break; - if (i == __x_.size()) - break; - ++i; - if (__x_[i] == L'd') - return time_base::ymd; - break; - case L'd': - for (++i; i < __x_.size(); ++i) - if (__x_[i] == L'%') - break; - if (i == __x_.size()) - break; - ++i; - if (__x_[i] == L'm') - return time_base::ydm; - break; - } - break; - case L'm': - for (++i; i < __x_.size(); ++i) - if (__x_[i] == L'%') - break; - if (i == __x_.size()) - break; - ++i; - if (__x_[i] == L'd') - { - for (++i; i < __x_.size(); ++i) - if (__x_[i] == L'%') - break; - if (i == __x_.size()) - break; - ++i; - if (__x_[i] == L'y' || __x_[i] == L'Y') - return time_base::mdy; - break; - } - break; - case L'd': - for (++i; i < __x_.size(); ++i) - if (__x_[i] == L'%') - break; - if (i == __x_.size()) - break; - ++i; - if (__x_[i] == L'm') - { - for (++i; i < __x_.size(); ++i) - if (__x_[i] == L'%') - break; - if (i == __x_.size()) - break; - ++i; - if (__x_[i] == L'y' || __x_[i] == L'Y') - return time_base::dmy; - break; - } - break; - } - return time_base::no_order; -} - -// time_put - -__time_put::__time_put(const char* nm) - : __loc_(newlocale(LC_ALL_MASK, nm, 0)) -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - if (__loc_ == 0) - throw runtime_error("time_put_byname" - " failed to construct for " + string(nm)); -#endif // _LIBCPP_NO_EXCEPTIONS -} - -__time_put::__time_put(const string& nm) - : __loc_(newlocale(LC_ALL_MASK, nm.c_str(), 0)) -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - if (__loc_ == 0) - throw runtime_error("time_put_byname" - " failed to construct for " + nm); -#endif // _LIBCPP_NO_EXCEPTIONS -} - -__time_put::~__time_put() -{ - if (__loc_ != _LIBCPP_GET_C_LOCALE) - freelocale(__loc_); -} - -void -__time_put::__do_put(char* __nb, char*& __ne, const tm* __tm, - char __fmt, char __mod) const -{ - char fmt[] = {'%', __fmt, __mod, 0}; - if (__mod != 0) - swap(fmt[1], fmt[2]); - size_t n = strftime_l(__nb, countof(__nb, __ne), fmt, __tm, __loc_); - __ne = __nb + n; -} - -void -__time_put::__do_put(wchar_t* __wb, wchar_t*& __we, const tm* __tm, - char __fmt, char __mod) const -{ - char __nar[100]; - char* __ne = __nar + 100; - __do_put(__nar, __ne, __tm, __fmt, __mod); - mbstate_t mb = {0}; - const char* __nb = __nar; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - size_t j = mbsrtowcs_l(__wb, &__nb, countof(__wb, __we), &mb, __loc_); -#else - size_t j = __mbsrtowcs_l(__wb, &__nb, countof(__wb, __we), &mb, __loc_); -#endif - if (j == size_t(-1)) - __throw_runtime_error("locale not supported"); - __we = __wb + j; -} - -// moneypunct_byname - -template -static -void -__init_pat(money_base::pattern& pat, basic_string& __curr_symbol_, - bool intl, char cs_precedes, char sep_by_space, char sign_posn, - charT space_char) -{ - const char sign = static_cast(money_base::sign); - const char space = static_cast(money_base::space); - const char none = static_cast(money_base::none); - const char symbol = static_cast(money_base::symbol); - const char value = static_cast(money_base::value); - const bool symbol_contains_sep = intl && __curr_symbol_.size() == 4; - - // Comments on case branches reflect 'C11 7.11.2.1 The localeconv - // function'. "Space between sign and symbol or value" means that - // if the sign is adjacent to the symbol, there's a space between - // them, and otherwise there's a space between the sign and value. - // - // C11's localeconv specifies that the fourth character of an - // international curr_symbol is used to separate the sign and - // value when sep_by_space says to do so. C++ can't represent - // that, so we just use a space. When sep_by_space says to - // separate the symbol and value-or-sign with a space, we rearrange the - // curr_symbol to put its spacing character on the correct side of - // the symbol. - // - // We also need to avoid adding an extra space between the sign - // and value when the currency symbol is suppressed (by not - // setting showbase). We match glibc's strfmon by interpreting - // sep_by_space==1 as "omit the space when the currency symbol is - // absent". - // - // Users who want to get this right should use ICU instead. - - switch (cs_precedes) - { - case 0: // value before curr_symbol - if (symbol_contains_sep) { - // Move the separator to before the symbol, to place it - // between the value and symbol. - rotate(__curr_symbol_.begin(), __curr_symbol_.begin() + 3, - __curr_symbol_.end()); - } - switch (sign_posn) - { - case 0: // Parentheses surround the quantity and currency symbol. - pat.field[0] = sign; - pat.field[1] = value; - pat.field[2] = none; // Any space appears in the symbol. - pat.field[3] = symbol; - switch (sep_by_space) - { - case 0: // No space separates the currency symbol and value. - // This case may have changed between C99 and C11; - // assume the currency symbol matches the intention. - case 2: // Space between sign and currency or value. - // The "sign" is two parentheses, so no space here either. - return; - case 1: // Space between currency-and-sign or currency and value. - if (!symbol_contains_sep) { - // We insert the space into the symbol instead of - // setting pat.field[2]=space so that when - // showbase is not set, the space goes away too. - __curr_symbol_.insert(0, 1, space_char); - } - return; - default: - break; - } - break; - case 1: // The sign string precedes the quantity and currency symbol. - pat.field[0] = sign; - pat.field[3] = symbol; - switch (sep_by_space) - { - case 0: // No space separates the currency symbol and value. - pat.field[1] = value; - pat.field[2] = none; - return; - case 1: // Space between currency-and-sign or currency and value. - pat.field[1] = value; - pat.field[2] = none; - if (!symbol_contains_sep) { - // We insert the space into the symbol instead of - // setting pat.field[2]=space so that when - // showbase is not set, the space goes away too. - __curr_symbol_.insert(0, 1, space_char); - } - return; - case 2: // Space between sign and currency or value. - pat.field[1] = space; - pat.field[2] = value; - if (symbol_contains_sep) { - // Remove the separator from the symbol, since it - // has already appeared after the sign. - __curr_symbol_.erase(__curr_symbol_.begin()); - } - return; - default: - break; - } - break; - case 2: // The sign string succeeds the quantity and currency symbol. - pat.field[0] = value; - pat.field[3] = sign; - switch (sep_by_space) - { - case 0: // No space separates the currency symbol and value. - pat.field[1] = none; - pat.field[2] = symbol; - return; - case 1: // Space between currency-and-sign or currency and value. - if (!symbol_contains_sep) { - // We insert the space into the symbol instead of - // setting pat.field[1]=space so that when - // showbase is not set, the space goes away too. - __curr_symbol_.insert(0, 1, space_char); - } - pat.field[1] = none; - pat.field[2] = symbol; - return; - case 2: // Space between sign and currency or value. - pat.field[1] = symbol; - pat.field[2] = space; - if (symbol_contains_sep) { - // Remove the separator from the symbol, since it - // should not be removed if showbase is absent. - __curr_symbol_.erase(__curr_symbol_.begin()); - } - return; - default: - break; - } - break; - case 3: // The sign string immediately precedes the currency symbol. - pat.field[0] = value; - pat.field[3] = symbol; - switch (sep_by_space) - { - case 0: // No space separates the currency symbol and value. - pat.field[1] = none; - pat.field[2] = sign; - return; - case 1: // Space between currency-and-sign or currency and value. - pat.field[1] = space; - pat.field[2] = sign; - if (symbol_contains_sep) { - // Remove the separator from the symbol, since it - // has already appeared before the sign. - __curr_symbol_.erase(__curr_symbol_.begin()); - } - return; - case 2: // Space between sign and currency or value. - pat.field[1] = sign; - pat.field[2] = none; - if (!symbol_contains_sep) { - // We insert the space into the symbol instead of - // setting pat.field[2]=space so that when - // showbase is not set, the space goes away too. - __curr_symbol_.insert(0, 1, space_char); - } - return; - default: - break; - } - break; - case 4: // The sign string immediately succeeds the currency symbol. - pat.field[0] = value; - pat.field[3] = sign; - switch (sep_by_space) - { - case 0: // No space separates the currency symbol and value. - pat.field[1] = none; - pat.field[2] = symbol; - return; - case 1: // Space between currency-and-sign or currency and value. - pat.field[1] = none; - pat.field[2] = symbol; - if (!symbol_contains_sep) { - // We insert the space into the symbol instead of - // setting pat.field[1]=space so that when - // showbase is not set, the space goes away too. - __curr_symbol_.insert(0, 1, space_char); - } - return; - case 2: // Space between sign and currency or value. - pat.field[1] = symbol; - pat.field[2] = space; - if (symbol_contains_sep) { - // Remove the separator from the symbol, since it - // should not disappear when showbase is absent. - __curr_symbol_.erase(__curr_symbol_.begin()); - } - return; - default: - break; - } - break; - default: - break; - } - break; - case 1: // curr_symbol before value - switch (sign_posn) - { - case 0: // Parentheses surround the quantity and currency symbol. - pat.field[0] = sign; - pat.field[1] = symbol; - pat.field[2] = none; // Any space appears in the symbol. - pat.field[3] = value; - switch (sep_by_space) - { - case 0: // No space separates the currency symbol and value. - // This case may have changed between C99 and C11; - // assume the currency symbol matches the intention. - case 2: // Space between sign and currency or value. - // The "sign" is two parentheses, so no space here either. - return; - case 1: // Space between currency-and-sign or currency and value. - if (!symbol_contains_sep) { - // We insert the space into the symbol instead of - // setting pat.field[2]=space so that when - // showbase is not set, the space goes away too. - __curr_symbol_.insert(0, 1, space_char); - } - return; - default: - break; - } - break; - case 1: // The sign string precedes the quantity and currency symbol. - pat.field[0] = sign; - pat.field[3] = value; - switch (sep_by_space) - { - case 0: // No space separates the currency symbol and value. - pat.field[1] = symbol; - pat.field[2] = none; - return; - case 1: // Space between currency-and-sign or currency and value. - pat.field[1] = symbol; - pat.field[2] = none; - if (!symbol_contains_sep) { - // We insert the space into the symbol instead of - // setting pat.field[2]=space so that when - // showbase is not set, the space goes away too. - __curr_symbol_.push_back(space_char); - } - return; - case 2: // Space between sign and currency or value. - pat.field[1] = space; - pat.field[2] = symbol; - if (symbol_contains_sep) { - // Remove the separator from the symbol, since it - // has already appeared after the sign. - __curr_symbol_.pop_back(); - } - return; - default: - break; - } - break; - case 2: // The sign string succeeds the quantity and currency symbol. - pat.field[0] = symbol; - pat.field[3] = sign; - switch (sep_by_space) - { - case 0: // No space separates the currency symbol and value. - pat.field[1] = none; - pat.field[2] = value; - return; - case 1: // Space between currency-and-sign or currency and value. - pat.field[1] = none; - pat.field[2] = value; - if (!symbol_contains_sep) { - // We insert the space into the symbol instead of - // setting pat.field[1]=space so that when - // showbase is not set, the space goes away too. - __curr_symbol_.push_back(space_char); - } - return; - case 2: // Space between sign and currency or value. - pat.field[1] = value; - pat.field[2] = space; - if (symbol_contains_sep) { - // Remove the separator from the symbol, since it - // will appear before the sign. - __curr_symbol_.pop_back(); - } - return; - default: - break; - } - break; - case 3: // The sign string immediately precedes the currency symbol. - pat.field[0] = sign; - pat.field[3] = value; - switch (sep_by_space) - { - case 0: // No space separates the currency symbol and value. - pat.field[1] = symbol; - pat.field[2] = none; - return; - case 1: // Space between currency-and-sign or currency and value. - pat.field[1] = symbol; - pat.field[2] = none; - if (!symbol_contains_sep) { - // We insert the space into the symbol instead of - // setting pat.field[2]=space so that when - // showbase is not set, the space goes away too. - __curr_symbol_.push_back(space_char); - } - return; - case 2: // Space between sign and currency or value. - pat.field[1] = space; - pat.field[2] = symbol; - if (symbol_contains_sep) { - // Remove the separator from the symbol, since it - // has already appeared after the sign. - __curr_symbol_.pop_back(); - } - return; - default: - break; - } - break; - case 4: // The sign string immediately succeeds the currency symbol. - pat.field[0] = symbol; - pat.field[3] = value; - switch (sep_by_space) - { - case 0: // No space separates the currency symbol and value. - pat.field[1] = sign; - pat.field[2] = none; - return; - case 1: // Space between currency-and-sign or currency and value. - pat.field[1] = sign; - pat.field[2] = space; - if (symbol_contains_sep) { - // Remove the separator from the symbol, since it - // should not disappear when showbase is absent. - __curr_symbol_.pop_back(); - } - return; - case 2: // Space between sign and currency or value. - pat.field[1] = none; - pat.field[2] = sign; - if (!symbol_contains_sep) { - // We insert the space into the symbol instead of - // setting pat.field[1]=space so that when - // showbase is not set, the space goes away too. - __curr_symbol_.push_back(space_char); - } - return; - default: - break; - } - break; - default: - break; - } - break; - default: - break; - } - pat.field[0] = symbol; - pat.field[1] = sign; - pat.field[2] = none; - pat.field[3] = value; -} - -template<> -void -moneypunct_byname::init(const char* nm) -{ - typedef moneypunct base; - __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); -#ifndef _LIBCPP_NO_EXCEPTIONS - if (loc == nullptr) - throw runtime_error("moneypunct_byname" - " failed to construct for " + string(nm)); -#endif // _LIBCPP_NO_EXCEPTIONS -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - lconv* lc = localeconv_l(loc.get()); -#else - lconv* lc = __localeconv_l(loc.get()); -#endif - if (*lc->mon_decimal_point) - __decimal_point_ = *lc->mon_decimal_point; - else - __decimal_point_ = base::do_decimal_point(); - if (*lc->mon_thousands_sep) - __thousands_sep_ = *lc->mon_thousands_sep; - else - __thousands_sep_ = base::do_thousands_sep(); - __grouping_ = lc->mon_grouping; - __curr_symbol_ = lc->currency_symbol; - if (lc->frac_digits != CHAR_MAX) - __frac_digits_ = lc->frac_digits; - else - __frac_digits_ = base::do_frac_digits(); - if (lc->p_sign_posn == 0) - __positive_sign_ = "()"; - else - __positive_sign_ = lc->positive_sign; - if (lc->n_sign_posn == 0) - __negative_sign_ = "()"; - else - __negative_sign_ = lc->negative_sign; - // Assume the positive and negative formats will want spaces in - // the same places in curr_symbol since there's no way to - // represent anything else. - string_type __dummy_curr_symbol = __curr_symbol_; - __init_pat(__pos_format_, __dummy_curr_symbol, false, - lc->p_cs_precedes, lc->p_sep_by_space, lc->p_sign_posn, ' '); - __init_pat(__neg_format_, __curr_symbol_, false, - lc->n_cs_precedes, lc->n_sep_by_space, lc->n_sign_posn, ' '); -} - -template<> -void -moneypunct_byname::init(const char* nm) -{ - typedef moneypunct base; - __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); -#ifndef _LIBCPP_NO_EXCEPTIONS - if (loc == nullptr) - throw runtime_error("moneypunct_byname" - " failed to construct for " + string(nm)); -#endif // _LIBCPP_NO_EXCEPTIONS -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - lconv* lc = localeconv_l(loc.get()); -#else - lconv* lc = __localeconv_l(loc.get()); -#endif - if (*lc->mon_decimal_point) - __decimal_point_ = *lc->mon_decimal_point; - else - __decimal_point_ = base::do_decimal_point(); - if (*lc->mon_thousands_sep) - __thousands_sep_ = *lc->mon_thousands_sep; - else - __thousands_sep_ = base::do_thousands_sep(); - __grouping_ = lc->mon_grouping; - __curr_symbol_ = lc->int_curr_symbol; - if (lc->int_frac_digits != CHAR_MAX) - __frac_digits_ = lc->int_frac_digits; - else - __frac_digits_ = base::do_frac_digits(); -#if defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) - if (lc->p_sign_posn == 0) -#else // _LIBCPP_MSVCRT - if (lc->int_p_sign_posn == 0) -#endif // !_LIBCPP_MSVCRT - __positive_sign_ = "()"; - else - __positive_sign_ = lc->positive_sign; -#if defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) - if(lc->n_sign_posn == 0) -#else // _LIBCPP_MSVCRT - if (lc->int_n_sign_posn == 0) -#endif // !_LIBCPP_MSVCRT - __negative_sign_ = "()"; - else - __negative_sign_ = lc->negative_sign; - // Assume the positive and negative formats will want spaces in - // the same places in curr_symbol since there's no way to - // represent anything else. - string_type __dummy_curr_symbol = __curr_symbol_; -#if defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) - __init_pat(__pos_format_, __dummy_curr_symbol, true, - lc->p_cs_precedes, lc->p_sep_by_space, lc->p_sign_posn, ' '); - __init_pat(__neg_format_, __curr_symbol_, true, - lc->n_cs_precedes, lc->n_sep_by_space, lc->n_sign_posn, ' '); -#else // _LIBCPP_MSVCRT - __init_pat(__pos_format_, __dummy_curr_symbol, true, - lc->int_p_cs_precedes, lc->int_p_sep_by_space, - lc->int_p_sign_posn, ' '); - __init_pat(__neg_format_, __curr_symbol_, true, - lc->int_n_cs_precedes, lc->int_n_sep_by_space, - lc->int_n_sign_posn, ' '); -#endif // !_LIBCPP_MSVCRT -} - -template<> -void -moneypunct_byname::init(const char* nm) -{ - typedef moneypunct base; - __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); -#ifndef _LIBCPP_NO_EXCEPTIONS - if (loc == nullptr) - throw runtime_error("moneypunct_byname" - " failed to construct for " + string(nm)); -#endif // _LIBCPP_NO_EXCEPTIONS -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - lconv* lc = localeconv_l(loc.get()); -#else - lconv* lc = __localeconv_l(loc.get()); -#endif - if (*lc->mon_decimal_point) - __decimal_point_ = static_cast(*lc->mon_decimal_point); - else - __decimal_point_ = base::do_decimal_point(); - if (*lc->mon_thousands_sep) - __thousands_sep_ = static_cast(*lc->mon_thousands_sep); - else - __thousands_sep_ = base::do_thousands_sep(); - __grouping_ = lc->mon_grouping; - wchar_t wbuf[100]; - mbstate_t mb = {0}; - const char* bb = lc->currency_symbol; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - size_t j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); -#else - size_t j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); -#endif - if (j == size_t(-1)) - __throw_runtime_error("locale not supported"); - wchar_t* wbe = wbuf + j; - __curr_symbol_.assign(wbuf, wbe); - if (lc->frac_digits != CHAR_MAX) - __frac_digits_ = lc->frac_digits; - else - __frac_digits_ = base::do_frac_digits(); - if (lc->p_sign_posn == 0) - __positive_sign_ = L"()"; - else - { - mb = mbstate_t(); - bb = lc->positive_sign; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); -#else - j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); -#endif - if (j == size_t(-1)) - __throw_runtime_error("locale not supported"); - wbe = wbuf + j; - __positive_sign_.assign(wbuf, wbe); - } - if (lc->n_sign_posn == 0) - __negative_sign_ = L"()"; - else - { - mb = mbstate_t(); - bb = lc->negative_sign; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); -#else - j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); -#endif - if (j == size_t(-1)) - __throw_runtime_error("locale not supported"); - wbe = wbuf + j; - __negative_sign_.assign(wbuf, wbe); - } - // Assume the positive and negative formats will want spaces in - // the same places in curr_symbol since there's no way to - // represent anything else. - string_type __dummy_curr_symbol = __curr_symbol_; - __init_pat(__pos_format_, __dummy_curr_symbol, false, - lc->p_cs_precedes, lc->p_sep_by_space, lc->p_sign_posn, L' '); - __init_pat(__neg_format_, __curr_symbol_, false, - lc->n_cs_precedes, lc->n_sep_by_space, lc->n_sign_posn, L' '); -} - -template<> -void -moneypunct_byname::init(const char* nm) -{ - typedef moneypunct base; - __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); -#ifndef _LIBCPP_NO_EXCEPTIONS - if (loc == nullptr) - throw runtime_error("moneypunct_byname" - " failed to construct for " + string(nm)); -#endif // _LIBCPP_NO_EXCEPTIONS -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - lconv* lc = localeconv_l(loc.get()); -#else - lconv* lc = __localeconv_l(loc.get()); -#endif - if (*lc->mon_decimal_point) - __decimal_point_ = static_cast(*lc->mon_decimal_point); - else - __decimal_point_ = base::do_decimal_point(); - if (*lc->mon_thousands_sep) - __thousands_sep_ = static_cast(*lc->mon_thousands_sep); - else - __thousands_sep_ = base::do_thousands_sep(); - __grouping_ = lc->mon_grouping; - wchar_t wbuf[100]; - mbstate_t mb = {0}; - const char* bb = lc->int_curr_symbol; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - size_t j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); -#else - size_t j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); -#endif - if (j == size_t(-1)) - __throw_runtime_error("locale not supported"); - wchar_t* wbe = wbuf + j; - __curr_symbol_.assign(wbuf, wbe); - if (lc->int_frac_digits != CHAR_MAX) - __frac_digits_ = lc->int_frac_digits; - else - __frac_digits_ = base::do_frac_digits(); -#if defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) - if (lc->p_sign_posn == 0) -#else // _LIBCPP_MSVCRT - if (lc->int_p_sign_posn == 0) -#endif // !_LIBCPP_MSVCRT - __positive_sign_ = L"()"; - else - { - mb = mbstate_t(); - bb = lc->positive_sign; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); -#else - j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); -#endif - if (j == size_t(-1)) - __throw_runtime_error("locale not supported"); - wbe = wbuf + j; - __positive_sign_.assign(wbuf, wbe); - } -#if defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) - if (lc->n_sign_posn == 0) -#else // _LIBCPP_MSVCRT - if (lc->int_n_sign_posn == 0) -#endif // !_LIBCPP_MSVCRT - __negative_sign_ = L"()"; - else - { - mb = mbstate_t(); - bb = lc->negative_sign; -#ifdef _LIBCPP_LOCALE__L_EXTENSIONS - j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); -#else - j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); -#endif - if (j == size_t(-1)) - __throw_runtime_error("locale not supported"); - wbe = wbuf + j; - __negative_sign_.assign(wbuf, wbe); - } - // Assume the positive and negative formats will want spaces in - // the same places in curr_symbol since there's no way to - // represent anything else. - string_type __dummy_curr_symbol = __curr_symbol_; -#if defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) - __init_pat(__pos_format_, __dummy_curr_symbol, true, - lc->p_cs_precedes, lc->p_sep_by_space, lc->p_sign_posn, L' '); - __init_pat(__neg_format_, __curr_symbol_, true, - lc->n_cs_precedes, lc->n_sep_by_space, lc->n_sign_posn, L' '); -#else // _LIBCPP_MSVCRT - __init_pat(__pos_format_, __dummy_curr_symbol, true, - lc->int_p_cs_precedes, lc->int_p_sep_by_space, - lc->int_p_sign_posn, L' '); - __init_pat(__neg_format_, __curr_symbol_, true, - lc->int_n_cs_precedes, lc->int_n_sep_by_space, - lc->int_n_sign_posn, L' '); -#endif // !_LIBCPP_MSVCRT -} - -void __do_nothing(void*) {} - -void __throw_runtime_error(const char* msg) -{ -#ifndef _LIBCPP_NO_EXCEPTIONS - throw runtime_error(msg); -#else - (void)msg; -#endif -} - -template class collate; -template class collate; - -template class num_get; -template class num_get; - -template struct __num_get; -template struct __num_get; - -template class num_put; -template class num_put; - -template struct __num_put; -template struct __num_put; - -template class time_get; -template class time_get; - -template class time_get_byname; -template class time_get_byname; - -template class time_put; -template class time_put; - -template class time_put_byname; -template class time_put_byname; - -template class moneypunct; -template class moneypunct; -template class moneypunct; -template class moneypunct; - -template class moneypunct_byname; -template class moneypunct_byname; -template class moneypunct_byname; -template class moneypunct_byname; - -template class money_get; -template class money_get; - -template class __money_get; -template class __money_get; - -template class money_put; -template class money_put; - -template class __money_put; -template class __money_put; - -template class messages; -template class messages; - -template class messages_byname; -template class messages_byname; - -template class codecvt_byname; -template class codecvt_byname; -template class codecvt_byname; -template class codecvt_byname; - -template class __vector_base_common; - -_LIBCPP_END_NAMESPACE_STD +//===------------------------- locale.cpp ---------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// On Solaris, we need to define something to make the C99 parts of localeconv +// visible. +#ifdef __sun__ +#define _LCONV_C99 +#endif + +#include "string" +#include "locale" +#include "codecvt" +#include "vector" +#include "algorithm" +#include "typeinfo" +#ifndef _LIBCPP_NO_EXCEPTIONS +# include "type_traits" +#endif +#include "clocale" +#include "cstring" +#include "cwctype" +#include "__sso_allocator" +#if defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) +#include +#elif !defined(__ANDROID__) +#include +#endif +#include +#include + +// On Linux, wint_t and wchar_t have different signed-ness, and this causes +// lots of noise in the build log, but no bugs that I know of. +#if defined(__clang__) +#pragma clang diagnostic ignored "-Wsign-conversion" +#endif + +_LIBCPP_BEGIN_NAMESPACE_STD + +#ifdef __cloc_defined +locale_t __cloc() { + // In theory this could create a race condition. In practice + // the race condition is non-fatal since it will just create + // a little resource leak. Better approach would be appreciated. + static locale_t result = newlocale(LC_ALL_MASK, "C", 0); + return result; +} +#endif // __cloc_defined + +namespace { + +struct release +{ + void operator()(locale::facet* p) {p->__release_shared();} +}; + +template +inline +T& +make(A0 a0) +{ + static typename aligned_storage::type buf; + ::new (&buf) T(a0); + return *reinterpret_cast(&buf); +} + +template +inline +T& +make(A0 a0, A1 a1) +{ + static typename aligned_storage::type buf; + ::new (&buf) T(a0, a1); + return *reinterpret_cast(&buf); +} + +template +inline +T& +make(A0 a0, A1 a1, A2 a2) +{ + static typename aligned_storage::type buf; + ::new (&buf) T(a0, a1, a2); + return *reinterpret_cast(&buf); +} + +template +inline +_LIBCPP_CONSTEXPR +size_t +countof(const T (&)[N]) +{ + return N; +} + +template +inline +_LIBCPP_CONSTEXPR +size_t +countof(const T * const begin, const T * const end) +{ + return static_cast(end - begin); +} + +} + +#if defined(_AIX) +// Set priority to INT_MIN + 256 + 150 +# pragma priority ( -2147483242 ) +#endif + +const locale::category locale::none; +const locale::category locale::collate; +const locale::category locale::ctype; +const locale::category locale::monetary; +const locale::category locale::numeric; +const locale::category locale::time; +const locale::category locale::messages; +const locale::category locale::all; + +#if defined(__clang__) +#pragma clang diagnostic push +#pragma clang diagnostic ignored "-Wpadded" +#endif + +#if _WIN32 || __SIZEOF_WCHAR_T__ == 2 +# define SHORT_WCHAR 1 +#elif __SIZEOF_WCHAR_T__ == 4 +# define SHORT_WCHAR 0 +#else +# error Can't tell how big wchar_t is on this platform +#endif + +#if SHORT_WCHAR + typedef uint16_t ucs_char_t; +#else + typedef uint32_t ucs_char_t; +#endif + +class _LIBCPP_HIDDEN locale::__imp + : public facet +{ + enum {N = 28}; +#if defined(_LIBCPP_MSVC) +// FIXME: MSVC doesn't support aligned parameters by value. +// I can't get the __sso_allocator to work here +// for MSVC I think for this reason. + vector facets_; +#else + vector > facets_; +#endif + string name_; +public: + explicit __imp(size_t refs = 0); + explicit __imp(const string& name, size_t refs = 0); + __imp(const __imp&); + __imp(const __imp&, const string&, locale::category c); + __imp(const __imp& other, const __imp& one, locale::category c); + __imp(const __imp&, facet* f, long id); + ~__imp(); + + const string& name() const {return name_;} + bool has_facet(long id) const + {return static_cast(id) < facets_.size() && facets_[static_cast(id)];} + const locale::facet* use_facet(long id) const; + + static const locale& make_classic(); + static locale& make_global(); +private: + void install(facet* f, long id); + template void install(F* f) {install(f, f->id.__get());} + template void install_from(const __imp& other); +}; + +#if defined(__clang__) +#pragma clang diagnostic pop +#endif + +locale::__imp::__imp(size_t refs) + : facet(refs), + facets_(N), + name_("C") +{ + facets_.clear(); + install(&make<_VSTD::collate >(1u)); + install(&make<_VSTD::collate >(1u)); + install(&make<_VSTD::ctype >(nullptr, false, 1u)); + install(&make<_VSTD::ctype >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make >(1u)); + install(&make<_VSTD::messages >(1u)); + install(&make<_VSTD::messages >(1u)); +} + +locale::__imp::__imp(const string& name, size_t refs) + : facet(refs), + facets_(N), + name_(name) +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + try + { +#endif // _LIBCPP_NO_EXCEPTIONS + facets_ = locale::classic().__locale_->facets_; + for (unsigned i = 0; i < facets_.size(); ++i) + if (facets_[i]) + facets_[i]->__add_shared(); + install(new collate_byname(name_)); + install(new collate_byname(name_)); + install(new ctype_byname(name_)); + install(new ctype_byname(name_)); + install(new codecvt_byname(name_)); + install(new codecvt_byname(name_)); + install(new codecvt_byname(name_)); + install(new codecvt_byname(name_)); + install(new numpunct_byname(name_)); + install(new numpunct_byname(name_)); + install(new moneypunct_byname(name_)); + install(new moneypunct_byname(name_)); + install(new moneypunct_byname(name_)); + install(new moneypunct_byname(name_)); + install(new time_get_byname(name_)); + install(new time_get_byname(name_)); + install(new time_put_byname(name_)); + install(new time_put_byname(name_)); + install(new messages_byname(name_)); + install(new messages_byname(name_)); +#ifndef _LIBCPP_NO_EXCEPTIONS + } + catch (...) + { + for (unsigned i = 0; i < facets_.size(); ++i) + if (facets_[i]) + facets_[i]->__release_shared(); + throw; + } +#endif // _LIBCPP_NO_EXCEPTIONS +} + +// NOTE avoid the `base class should be explicitly initialized in the +// copy constructor` warning emitted by GCC +#if defined(__clang__) || _GNUC_VER >= 406 +#pragma GCC diagnostic push +#pragma GCC diagnostic ignored "-Wextra" +#endif + +locale::__imp::__imp(const __imp& other) + : facets_(max(N, other.facets_.size())), + name_(other.name_) +{ + facets_ = other.facets_; + for (unsigned i = 0; i < facets_.size(); ++i) + if (facets_[i]) + facets_[i]->__add_shared(); +} + +#if defined(__clang__) || _GNUC_VER >= 406 +#pragma GCC diagnostic pop +#endif + +locale::__imp::__imp(const __imp& other, const string& name, locale::category c) + : facets_(N), + name_("*") +{ + facets_ = other.facets_; + for (unsigned i = 0; i < facets_.size(); ++i) + if (facets_[i]) + facets_[i]->__add_shared(); +#ifndef _LIBCPP_NO_EXCEPTIONS + try + { +#endif // _LIBCPP_NO_EXCEPTIONS + if (c & locale::collate) + { + install(new collate_byname(name)); + install(new collate_byname(name)); + } + if (c & locale::ctype) + { + install(new ctype_byname(name)); + install(new ctype_byname(name)); + install(new codecvt_byname(name)); + install(new codecvt_byname(name)); + install(new codecvt_byname(name)); + install(new codecvt_byname(name)); + } + if (c & locale::monetary) + { + install(new moneypunct_byname(name)); + install(new moneypunct_byname(name)); + install(new moneypunct_byname(name)); + install(new moneypunct_byname(name)); + } + if (c & locale::numeric) + { + install(new numpunct_byname(name)); + install(new numpunct_byname(name)); + } + if (c & locale::time) + { + install(new time_get_byname(name)); + install(new time_get_byname(name)); + install(new time_put_byname(name)); + install(new time_put_byname(name)); + } + if (c & locale::messages) + { + install(new messages_byname(name)); + install(new messages_byname(name)); + } +#ifndef _LIBCPP_NO_EXCEPTIONS + } + catch (...) + { + for (unsigned i = 0; i < facets_.size(); ++i) + if (facets_[i]) + facets_[i]->__release_shared(); + throw; + } +#endif // _LIBCPP_NO_EXCEPTIONS +} + +template +inline +void +locale::__imp::install_from(const locale::__imp& one) +{ + long id = F::id.__get(); + install(const_cast(static_cast(one.use_facet(id))), id); +} + +locale::__imp::__imp(const __imp& other, const __imp& one, locale::category c) + : facets_(N), + name_("*") +{ + facets_ = other.facets_; + for (unsigned i = 0; i < facets_.size(); ++i) + if (facets_[i]) + facets_[i]->__add_shared(); +#ifndef _LIBCPP_NO_EXCEPTIONS + try + { +#endif // _LIBCPP_NO_EXCEPTIONS + if (c & locale::collate) + { + install_from<_VSTD::collate >(one); + install_from<_VSTD::collate >(one); + } + if (c & locale::ctype) + { + install_from<_VSTD::ctype >(one); + install_from<_VSTD::ctype >(one); + install_from<_VSTD::codecvt >(one); + install_from<_VSTD::codecvt >(one); + install_from<_VSTD::codecvt >(one); + install_from<_VSTD::codecvt >(one); + } + if (c & locale::monetary) + { + install_from >(one); + install_from >(one); + install_from >(one); + install_from >(one); + install_from >(one); + install_from >(one); + install_from >(one); + install_from >(one); + } + if (c & locale::numeric) + { + install_from >(one); + install_from >(one); + install_from >(one); + install_from >(one); + install_from >(one); + install_from >(one); + } + if (c & locale::time) + { + install_from >(one); + install_from >(one); + install_from >(one); + install_from >(one); + } + if (c & locale::messages) + { + install_from<_VSTD::messages >(one); + install_from<_VSTD::messages >(one); + } +#ifndef _LIBCPP_NO_EXCEPTIONS + } + catch (...) + { + for (unsigned i = 0; i < facets_.size(); ++i) + if (facets_[i]) + facets_[i]->__release_shared(); + throw; + } +#endif // _LIBCPP_NO_EXCEPTIONS +} + +locale::__imp::__imp(const __imp& other, facet* f, long id) + : facets_(max(N, other.facets_.size()+1)), + name_("*") +{ + f->__add_shared(); + unique_ptr hold(f); + facets_ = other.facets_; + for (unsigned i = 0; i < other.facets_.size(); ++i) + if (facets_[i]) + facets_[i]->__add_shared(); + install(hold.get(), id); +} + +locale::__imp::~__imp() +{ + for (unsigned i = 0; i < facets_.size(); ++i) + if (facets_[i]) + facets_[i]->__release_shared(); +} + +void +locale::__imp::install(facet* f, long id) +{ + f->__add_shared(); + unique_ptr hold(f); + if (static_cast(id) >= facets_.size()) + facets_.resize(static_cast(id+1)); + if (facets_[static_cast(id)]) + facets_[static_cast(id)]->__release_shared(); + facets_[static_cast(id)] = hold.release(); +} + +const locale::facet* +locale::__imp::use_facet(long id) const +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + if (!has_facet(id)) + throw bad_cast(); +#endif // _LIBCPP_NO_EXCEPTIONS + return facets_[static_cast(id)]; +} + +// locale + +const locale& +locale::__imp::make_classic() +{ + // only one thread can get in here and it only gets in once + static aligned_storage::type buf; + locale* c = reinterpret_cast(&buf); + c->__locale_ = &make<__imp>(1u); + return *c; +} + +const locale& +locale::classic() +{ + static const locale& c = __imp::make_classic(); + return c; +} + +locale& +locale::__imp::make_global() +{ + // only one thread can get in here and it only gets in once + static aligned_storage::type buf; + ::new (&buf) locale(locale::classic()); + return *reinterpret_cast(&buf); +} + +locale& +locale::__global() +{ + static locale& g = __imp::make_global(); + return g; +} + +locale::locale() _NOEXCEPT + : __locale_(__global().__locale_) +{ + __locale_->__add_shared(); +} + +locale::locale(const locale& l) _NOEXCEPT + : __locale_(l.__locale_) +{ + __locale_->__add_shared(); +} + +locale::~locale() +{ + __locale_->__release_shared(); +} + +const locale& +locale::operator=(const locale& other) _NOEXCEPT +{ + other.__locale_->__add_shared(); + __locale_->__release_shared(); + __locale_ = other.__locale_; + return *this; +} + +locale::locale(const char* name) +#ifndef _LIBCPP_NO_EXCEPTIONS + : __locale_(name ? new __imp(name) + : throw runtime_error("locale constructed with null")) +#else // _LIBCPP_NO_EXCEPTIONS + : __locale_(new __imp(name)) +#endif +{ + __locale_->__add_shared(); +} + +locale::locale(const string& name) + : __locale_(new __imp(name)) +{ + __locale_->__add_shared(); +} + +locale::locale(const locale& other, const char* name, category c) +#ifndef _LIBCPP_NO_EXCEPTIONS + : __locale_(name ? new __imp(*other.__locale_, name, c) + : throw runtime_error("locale constructed with null")) +#else // _LIBCPP_NO_EXCEPTIONS + : __locale_(new __imp(*other.__locale_, name, c)) +#endif +{ + __locale_->__add_shared(); +} + +locale::locale(const locale& other, const string& name, category c) + : __locale_(new __imp(*other.__locale_, name, c)) +{ + __locale_->__add_shared(); +} + +locale::locale(const locale& other, const locale& one, category c) + : __locale_(new __imp(*other.__locale_, *one.__locale_, c)) +{ + __locale_->__add_shared(); +} + +string +locale::name() const +{ + return __locale_->name(); +} + +void +locale::__install_ctor(const locale& other, facet* f, long id) +{ + if (f) + __locale_ = new __imp(*other.__locale_, f, id); + else + __locale_ = other.__locale_; + __locale_->__add_shared(); +} + +locale +locale::global(const locale& loc) +{ + locale& g = __global(); + locale r = g; + g = loc; + if (g.name() != "*") + setlocale(LC_ALL, g.name().c_str()); + return r; +} + +bool +locale::has_facet(id& x) const +{ + return __locale_->has_facet(x.__get()); +} + +const locale::facet* +locale::use_facet(id& x) const +{ + return __locale_->use_facet(x.__get()); +} + +bool +locale::operator==(const locale& y) const +{ + return (__locale_ == y.__locale_) + || (__locale_->name() != "*" && __locale_->name() == y.__locale_->name()); +} + +// locale::facet + +locale::facet::~facet() +{ +} + +void +locale::facet::__on_zero_shared() _NOEXCEPT +{ + delete this; +} + +// locale::id + +int32_t locale::id::__next_id = 0; + +namespace +{ + +class __fake_bind +{ + locale::id* id_; + void (locale::id::* pmf_)(); +public: + __fake_bind(void (locale::id::* pmf)(), locale::id* id) + : id_(id), pmf_(pmf) {} + + void operator()() const + { + (id_->*pmf_)(); + } +}; + +} + +long +locale::id::__get() +{ + call_once(__flag_, __fake_bind(&locale::id::__init, this)); + return __id_ - 1; +} + +void +locale::id::__init() +{ + __id_ = __sync_add_and_fetch(&__next_id, 1); +} + +// template <> class collate_byname + +collate_byname::collate_byname(const char* n, size_t refs) + : collate(refs), + __l(newlocale(LC_ALL_MASK, n, 0)) +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + if (__l == 0) + throw runtime_error("collate_byname::collate_byname" + " failed to construct for " + string(n)); +#endif // _LIBCPP_NO_EXCEPTIONS +} + +collate_byname::collate_byname(const string& name, size_t refs) + : collate(refs), + __l(newlocale(LC_ALL_MASK, name.c_str(), 0)) +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + if (__l == 0) + throw runtime_error("collate_byname::collate_byname" + " failed to construct for " + name); +#endif // _LIBCPP_NO_EXCEPTIONS +} + +collate_byname::~collate_byname() +{ + freelocale(__l); +} + +int +collate_byname::do_compare(const char_type* __lo1, const char_type* __hi1, + const char_type* __lo2, const char_type* __hi2) const +{ + string_type lhs(__lo1, __hi1); + string_type rhs(__lo2, __hi2); + int r = strcoll_l(lhs.c_str(), rhs.c_str(), __l); + if (r < 0) + return -1; + if (r > 0) + return 1; + return r; +} + +collate_byname::string_type +collate_byname::do_transform(const char_type* lo, const char_type* hi) const +{ + const string_type in(lo, hi); + string_type out(strxfrm_l(0, in.c_str(), 0, __l), char()); + strxfrm_l(const_cast(out.c_str()), in.c_str(), out.size()+1, __l); + return out; +} + +// template <> class collate_byname + +collate_byname::collate_byname(const char* n, size_t refs) + : collate(refs), + __l(newlocale(LC_ALL_MASK, n, 0)) +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + if (__l == 0) + throw runtime_error("collate_byname::collate_byname(size_t refs)" + " failed to construct for " + string(n)); +#endif // _LIBCPP_NO_EXCEPTIONS +} + +collate_byname::collate_byname(const string& name, size_t refs) + : collate(refs), + __l(newlocale(LC_ALL_MASK, name.c_str(), 0)) +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + if (__l == 0) + throw runtime_error("collate_byname::collate_byname(size_t refs)" + " failed to construct for " + name); +#endif // _LIBCPP_NO_EXCEPTIONS +} + +collate_byname::~collate_byname() +{ + freelocale(__l); +} + +int +collate_byname::do_compare(const char_type* __lo1, const char_type* __hi1, + const char_type* __lo2, const char_type* __hi2) const +{ + string_type lhs(__lo1, __hi1); + string_type rhs(__lo2, __hi2); + int r = wcscoll_l(lhs.c_str(), rhs.c_str(), __l); + if (r < 0) + return -1; + if (r > 0) + return 1; + return r; +} + +collate_byname::string_type +collate_byname::do_transform(const char_type* lo, const char_type* hi) const +{ + const string_type in(lo, hi); + string_type out(wcsxfrm_l(0, in.c_str(), 0, __l), wchar_t()); + wcsxfrm_l(const_cast(out.c_str()), in.c_str(), out.size()+1, __l); + return out; +} + +// template <> class ctype; + +const ctype_base::mask ctype_base::space; +const ctype_base::mask ctype_base::print; +const ctype_base::mask ctype_base::cntrl; +const ctype_base::mask ctype_base::upper; +const ctype_base::mask ctype_base::lower; +const ctype_base::mask ctype_base::alpha; +const ctype_base::mask ctype_base::digit; +const ctype_base::mask ctype_base::punct; +const ctype_base::mask ctype_base::xdigit; +const ctype_base::mask ctype_base::blank; +const ctype_base::mask ctype_base::alnum; +const ctype_base::mask ctype_base::graph; + +locale::id ctype::id; + +ctype::~ctype() +{ +} + +bool +ctype::do_is(mask m, char_type c) const +{ + return isascii(c) ? (ctype::classic_table()[c] & m) != 0 : false; +} + +const wchar_t* +ctype::do_is(const char_type* low, const char_type* high, mask* vec) const +{ + for (; low != high; ++low, ++vec) + *vec = static_cast(isascii(*low) ? + ctype::classic_table()[*low] : 0); + return low; +} + +const wchar_t* +ctype::do_scan_is(mask m, const char_type* low, const char_type* high) const +{ + for (; low != high; ++low) + if (isascii(*low) && (ctype::classic_table()[*low] & m)) + break; + return low; +} + +const wchar_t* +ctype::do_scan_not(mask m, const char_type* low, const char_type* high) const +{ + for (; low != high; ++low) + if (!(isascii(*low) && (ctype::classic_table()[*low] & m))) + break; + return low; +} + +wchar_t +ctype::do_toupper(char_type c) const +{ +#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE + return isascii(c) ? _DefaultRuneLocale.__mapupper[c] : c; +#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__NetBSD__) + return isascii(c) ? ctype::__classic_upper_table()[c] : c; +#else + return (isascii(c) && iswlower_l(c, __cloc())) ? c-L'a'+L'A' : c; +#endif +} + +const wchar_t* +ctype::do_toupper(char_type* low, const char_type* high) const +{ + for (; low != high; ++low) +#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE + *low = isascii(*low) ? _DefaultRuneLocale.__mapupper[*low] : *low; +#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__NetBSD__) + *low = isascii(*low) ? ctype::__classic_upper_table()[*low] + : *low; +#else + *low = (isascii(*low) && islower_l(*low, __cloc())) ? (*low-L'a'+L'A') : *low; +#endif + return low; +} + +wchar_t +ctype::do_tolower(char_type c) const +{ +#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE + return isascii(c) ? _DefaultRuneLocale.__maplower[c] : c; +#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__NetBSD__) + return isascii(c) ? ctype::__classic_lower_table()[c] : c; +#else + return (isascii(c) && isupper_l(c, __cloc())) ? c-L'A'+'a' : c; +#endif +} + +const wchar_t* +ctype::do_tolower(char_type* low, const char_type* high) const +{ + for (; low != high; ++low) +#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE + *low = isascii(*low) ? _DefaultRuneLocale.__maplower[*low] : *low; +#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__NetBSD__) + *low = isascii(*low) ? ctype::__classic_lower_table()[*low] + : *low; +#else + *low = (isascii(*low) && isupper_l(*low, __cloc())) ? *low-L'A'+L'a' : *low; +#endif + return low; +} + +wchar_t +ctype::do_widen(char c) const +{ + return c; +} + +const char* +ctype::do_widen(const char* low, const char* high, char_type* dest) const +{ + for (; low != high; ++low, ++dest) + *dest = *low; + return low; +} + +char +ctype::do_narrow(char_type c, char dfault) const +{ + if (isascii(c)) + return static_cast(c); + return dfault; +} + +const wchar_t* +ctype::do_narrow(const char_type* low, const char_type* high, char dfault, char* dest) const +{ + for (; low != high; ++low, ++dest) + if (isascii(*low)) + *dest = static_cast(*low); + else + *dest = dfault; + return low; +} + +// template <> class ctype; + +locale::id ctype::id; + +ctype::ctype(const mask* tab, bool del, size_t refs) + : locale::facet(refs), + __tab_(tab), + __del_(del) +{ + if (__tab_ == 0) + __tab_ = classic_table(); +} + +ctype::~ctype() +{ + if (__tab_ && __del_) + delete [] __tab_; +} + +char +ctype::do_toupper(char_type c) const +{ +#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE + return isascii(c) ? + static_cast(_DefaultRuneLocale.__mapupper[static_cast(c)]) : c; +#elif defined(__NetBSD__) + return static_cast(__classic_upper_table()[static_cast(c)]); +#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) + return isascii(c) ? + static_cast(__classic_upper_table()[static_cast(c)]) : c; +#else + return (isascii(c) && islower_l(c, __cloc())) ? c-'a'+'A' : c; +#endif +} + +const char* +ctype::do_toupper(char_type* low, const char_type* high) const +{ + for (; low != high; ++low) +#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE + *low = isascii(*low) ? + static_cast(_DefaultRuneLocale.__mapupper[static_cast(*low)]) : *low; +#elif defined(__NetBSD__) + *low = static_cast(__classic_upper_table()[static_cast(*low)]); +#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) + *low = isascii(*low) ? + static_cast(__classic_upper_table()[static_cast(*low)]) : *low; +#else + *low = (isascii(*low) && islower_l(*low, __cloc())) ? *low-'a'+'A' : *low; +#endif + return low; +} + +char +ctype::do_tolower(char_type c) const +{ +#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE + return isascii(c) ? + static_cast(_DefaultRuneLocale.__maplower[static_cast(c)]) : c; +#elif defined(__NetBSD__) + return static_cast(__classic_lower_table()[static_cast(c)]); +#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) || defined(__NetBSD__) + return isascii(c) ? + static_cast(__classic_lower_table()[static_cast(c)]) : c; +#else + return (isascii(c) && isupper_l(c, __cloc())) ? c-'A'+'a' : c; +#endif +} + +const char* +ctype::do_tolower(char_type* low, const char_type* high) const +{ + for (; low != high; ++low) +#ifdef _LIBCPP_HAS_DEFAULTRUNELOCALE + *low = isascii(*low) ? static_cast(_DefaultRuneLocale.__maplower[static_cast(*low)]) : *low; +#elif defined(__NetBSD__) + *low = static_cast(__classic_lower_table()[static_cast(*low)]); +#elif defined(__GLIBC__) || defined(__EMSCRIPTEN__) + *low = isascii(*low) ? static_cast(__classic_lower_table()[static_cast(*low)]) : *low; +#else + *low = (isascii(*low) && isupper_l(*low, __cloc())) ? *low-'A'+'a' : *low; +#endif + return low; +} + +char +ctype::do_widen(char c) const +{ + return c; +} + +const char* +ctype::do_widen(const char* low, const char* high, char_type* dest) const +{ + for (; low != high; ++low, ++dest) + *dest = *low; + return low; +} + +char +ctype::do_narrow(char_type c, char dfault) const +{ + if (isascii(c)) + return static_cast(c); + return dfault; +} + +const char* +ctype::do_narrow(const char_type* low, const char_type* high, char dfault, char* dest) const +{ + for (; low != high; ++low, ++dest) + if (isascii(*low)) + *dest = *low; + else + *dest = dfault; + return low; +} + +#ifdef __EMSCRIPTEN__ +extern "C" const unsigned short ** __ctype_b_loc(); +extern "C" const int ** __ctype_tolower_loc(); +extern "C" const int ** __ctype_toupper_loc(); +#endif + +#ifdef _LIBCPP_PROVIDES_DEFAULT_RUNE_TABLE +const ctype::mask* +ctype::classic_table() _NOEXCEPT +{ + static _LIBCPP_CONSTEXPR const ctype::mask builtin_table[table_size] = { + cntrl, cntrl, + cntrl, cntrl, + cntrl, cntrl, + cntrl, cntrl, + cntrl, cntrl | space | blank, + cntrl | space, cntrl | space, + cntrl | space, cntrl | space, + cntrl, cntrl, + cntrl, cntrl, + cntrl, cntrl, + cntrl, cntrl, + cntrl, cntrl, + cntrl, cntrl, + cntrl, cntrl, + cntrl, cntrl, + cntrl, cntrl, + space | blank | print, punct | print, + punct | print, punct | print, + punct | print, punct | print, + punct | print, punct | print, + punct | print, punct | print, + punct | print, punct | print, + punct | print, punct | print, + punct | print, punct | print, + digit | print | xdigit, digit | print | xdigit, + digit | print | xdigit, digit | print | xdigit, + digit | print | xdigit, digit | print | xdigit, + digit | print | xdigit, digit | print | xdigit, + digit | print | xdigit, digit | print | xdigit, + punct | print, punct | print, + punct | print, punct | print, + punct | print, punct | print, + punct | print, upper | xdigit | print | alpha, + upper | xdigit | print | alpha, upper | xdigit | print | alpha, + upper | xdigit | print | alpha, upper | xdigit | print | alpha, + upper | xdigit | print | alpha, upper | print | alpha, + upper | print | alpha, upper | print | alpha, + upper | print | alpha, upper | print | alpha, + upper | print | alpha, upper | print | alpha, + upper | print | alpha, upper | print | alpha, + upper | print | alpha, upper | print | alpha, + upper | print | alpha, upper | print | alpha, + upper | print | alpha, upper | print | alpha, + upper | print | alpha, upper | print | alpha, + upper | print | alpha, upper | print | alpha, + upper | print | alpha, punct | print, + punct | print, punct | print, + punct | print, punct | print, + punct | print, lower | xdigit | print | alpha, + lower | xdigit | print | alpha, lower | xdigit | print | alpha, + lower | xdigit | print | alpha, lower | xdigit | print | alpha, + lower | xdigit | print | alpha, lower | print | alpha, + lower | print | alpha, lower | print | alpha, + lower | print | alpha, lower | print | alpha, + lower | print | alpha, lower | print | alpha, + lower | print | alpha, lower | print | alpha, + lower | print | alpha, lower | print | alpha, + lower | print | alpha, lower | print | alpha, + lower | print | alpha, lower | print | alpha, + lower | print | alpha, lower | print | alpha, + lower | print | alpha, lower | print | alpha, + lower | print | alpha, punct | print, + punct | print, punct | print, + punct | print, cntrl, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + }; + return builtin_table; +} +#else +const ctype::mask* +ctype::classic_table() _NOEXCEPT +{ +#if defined(__APPLE__) || defined(__FreeBSD__) + return _DefaultRuneLocale.__runetype; +#elif defined(__NetBSD__) + return _C_ctype_tab_ + 1; +#elif defined(__GLIBC__) + return __cloc()->__ctype_b; +#elif __sun__ + return __ctype_mask; +#elif defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) + return _ctype+1; // internal ctype mask table defined in msvcrt.dll +// This is assumed to be safe, which is a nonsense assumption because we're +// going to end up dereferencing it later... +#elif defined(__EMSCRIPTEN__) + return *__ctype_b_loc(); +#elif defined(_NEWLIB_VERSION) + // Newlib has a 257-entry table in ctype_.c, where (char)0 starts at [1]. + return _ctype_ + 1; +#elif defined(_AIX) + return (const unsigned int *)__lc_ctype_ptr->obj->mask; +#else + // Platform not supported: abort so the person doing the port knows what to + // fix +# warning ctype::classic_table() is not implemented + printf("ctype::classic_table() is not implemented\n"); + abort(); + return NULL; +#endif +} +#endif + +#if defined(__GLIBC__) +const int* +ctype::__classic_lower_table() _NOEXCEPT +{ + return __cloc()->__ctype_tolower; +} + +const int* +ctype::__classic_upper_table() _NOEXCEPT +{ + return __cloc()->__ctype_toupper; +} +#elif __NetBSD__ +const short* +ctype::__classic_lower_table() _NOEXCEPT +{ + return _C_tolower_tab_ + 1; +} + +const short* +ctype::__classic_upper_table() _NOEXCEPT +{ + return _C_toupper_tab_ + 1; +} + +#elif defined(__EMSCRIPTEN__) +const int* +ctype::__classic_lower_table() _NOEXCEPT +{ + return *__ctype_tolower_loc(); +} + +const int* +ctype::__classic_upper_table() _NOEXCEPT +{ + return *__ctype_toupper_loc(); +} +#endif // __GLIBC__ || __EMSCRIPTEN__ || __NETBSD__ + +// template <> class ctype_byname + +ctype_byname::ctype_byname(const char* name, size_t refs) + : ctype(0, false, refs), + __l(newlocale(LC_ALL_MASK, name, 0)) +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + if (__l == 0) + throw runtime_error("ctype_byname::ctype_byname" + " failed to construct for " + string(name)); +#endif // _LIBCPP_NO_EXCEPTIONS +} + +ctype_byname::ctype_byname(const string& name, size_t refs) + : ctype(0, false, refs), + __l(newlocale(LC_ALL_MASK, name.c_str(), 0)) +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + if (__l == 0) + throw runtime_error("ctype_byname::ctype_byname" + " failed to construct for " + name); +#endif // _LIBCPP_NO_EXCEPTIONS +} + +ctype_byname::~ctype_byname() +{ + freelocale(__l); +} + +char +ctype_byname::do_toupper(char_type c) const +{ + return static_cast(toupper_l(static_cast(c), __l)); +} + +const char* +ctype_byname::do_toupper(char_type* low, const char_type* high) const +{ + for (; low != high; ++low) + *low = static_cast(toupper_l(static_cast(*low), __l)); + return low; +} + +char +ctype_byname::do_tolower(char_type c) const +{ + return static_cast(tolower_l(static_cast(c), __l)); +} + +const char* +ctype_byname::do_tolower(char_type* low, const char_type* high) const +{ + for (; low != high; ++low) + *low = static_cast(tolower_l(static_cast(*low), __l)); + return low; +} + +// template <> class ctype_byname + +ctype_byname::ctype_byname(const char* name, size_t refs) + : ctype(refs), + __l(newlocale(LC_ALL_MASK, name, 0)) +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + if (__l == 0) + throw runtime_error("ctype_byname::ctype_byname" + " failed to construct for " + string(name)); +#endif // _LIBCPP_NO_EXCEPTIONS +} + +ctype_byname::ctype_byname(const string& name, size_t refs) + : ctype(refs), + __l(newlocale(LC_ALL_MASK, name.c_str(), 0)) +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + if (__l == 0) + throw runtime_error("ctype_byname::ctype_byname" + " failed to construct for " + name); +#endif // _LIBCPP_NO_EXCEPTIONS +} + +ctype_byname::~ctype_byname() +{ + freelocale(__l); +} + +bool +ctype_byname::do_is(mask m, char_type c) const +{ +#ifdef _LIBCPP_WCTYPE_IS_MASK + return static_cast(iswctype_l(c, m, __l)); +#else + bool result = false; + wint_t ch = static_cast(c); + if ((m & space) == space) result |= (iswspace_l(ch, __l) != 0); + if ((m & print) == print) result |= (iswprint_l(ch, __l) != 0); + if ((m & cntrl) == cntrl) result |= (iswcntrl_l(ch, __l) != 0); + if ((m & upper) == upper) result |= (iswupper_l(ch, __l) != 0); + if ((m & lower) == lower) result |= (iswlower_l(ch, __l) != 0); + if ((m & alpha) == alpha) result |= (iswalpha_l(ch, __l) != 0); + if ((m & digit) == digit) result |= (iswdigit_l(ch, __l) != 0); + if ((m & punct) == punct) result |= (iswpunct_l(ch, __l) != 0); + if ((m & xdigit) == xdigit) result |= (iswxdigit_l(ch, __l) != 0); + if ((m & blank) == blank) result |= (iswblank_l(ch, __l) != 0); + return result; +#endif +} + +const wchar_t* +ctype_byname::do_is(const char_type* low, const char_type* high, mask* vec) const +{ + for (; low != high; ++low, ++vec) + { + if (isascii(*low)) + *vec = static_cast(ctype::classic_table()[*low]); + else + { + *vec = 0; + wint_t ch = static_cast(*low); + if (iswspace_l(ch, __l)) + *vec |= space; +#ifndef _LIBCPP_CTYPE_MASK_IS_COMPOSITE_PRINT + if (iswprint_l(ch, __l)) + *vec |= print; +#endif + if (iswcntrl_l(ch, __l)) + *vec |= cntrl; + if (iswupper_l(ch, __l)) + *vec |= upper; + if (iswlower_l(ch, __l)) + *vec |= lower; +#ifndef _LIBCPP_CTYPE_MASK_IS_COMPOSITE_ALPHA + if (iswalpha_l(ch, __l)) + *vec |= alpha; +#endif + if (iswdigit_l(ch, __l)) + *vec |= digit; + if (iswpunct_l(ch, __l)) + *vec |= punct; +#ifndef _LIBCPP_CTYPE_MASK_IS_COMPOSITE_XDIGIT + if (iswxdigit_l(ch, __l)) + *vec |= xdigit; +#endif +#if !defined(__sun__) + if (iswblank_l(ch, __l)) + *vec |= blank; +#endif + } + } + return low; +} + +const wchar_t* +ctype_byname::do_scan_is(mask m, const char_type* low, const char_type* high) const +{ + for (; low != high; ++low) + { +#ifdef _LIBCPP_WCTYPE_IS_MASK + if (iswctype_l(*low, m, __l)) + break; +#else + wint_t ch = static_cast(*low); + if ((m & space) == space && iswspace_l(ch, __l)) break; + if ((m & print) == print && iswprint_l(ch, __l)) break; + if ((m & cntrl) == cntrl && iswcntrl_l(ch, __l)) break; + if ((m & upper) == upper && iswupper_l(ch, __l)) break; + if ((m & lower) == lower && iswlower_l(ch, __l)) break; + if ((m & alpha) == alpha && iswalpha_l(ch, __l)) break; + if ((m & digit) == digit && iswdigit_l(ch, __l)) break; + if ((m & punct) == punct && iswpunct_l(ch, __l)) break; + if ((m & xdigit) == xdigit && iswxdigit_l(ch, __l)) break; + if ((m & blank) == blank && iswblank_l(ch, __l)) break; +#endif + } + return low; +} + +const wchar_t* +ctype_byname::do_scan_not(mask m, const char_type* low, const char_type* high) const +{ + for (; low != high; ++low) + { +#ifdef _LIBCPP_WCTYPE_IS_MASK + if (!iswctype_l(*low, m, __l)) + break; +#else + wint_t ch = static_cast(*low); + if ((m & space) == space && iswspace_l(ch, __l)) continue; + if ((m & print) == print && iswprint_l(ch, __l)) continue; + if ((m & cntrl) == cntrl && iswcntrl_l(ch, __l)) continue; + if ((m & upper) == upper && iswupper_l(ch, __l)) continue; + if ((m & lower) == lower && iswlower_l(ch, __l)) continue; + if ((m & alpha) == alpha && iswalpha_l(ch, __l)) continue; + if ((m & digit) == digit && iswdigit_l(ch, __l)) continue; + if ((m & punct) == punct && iswpunct_l(ch, __l)) continue; + if ((m & xdigit) == xdigit && iswxdigit_l(ch, __l)) continue; + if ((m & blank) == blank && iswblank_l(ch, __l)) continue; + break; +#endif + } + return low; +} + +wchar_t +ctype_byname::do_toupper(char_type c) const +{ + return towupper_l(c, __l); +} + +const wchar_t* +ctype_byname::do_toupper(char_type* low, const char_type* high) const +{ + for (; low != high; ++low) + *low = towupper_l(*low, __l); + return low; +} + +wchar_t +ctype_byname::do_tolower(char_type c) const +{ + return towlower_l(c, __l); +} + +const wchar_t* +ctype_byname::do_tolower(char_type* low, const char_type* high) const +{ + for (; low != high; ++low) + *low = towlower_l(*low, __l); + return low; +} + +wchar_t +ctype_byname::do_widen(char c) const +{ +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + return btowc_l(c, __l); +#else + return __btowc_l(c, __l); +#endif +} + +const char* +ctype_byname::do_widen(const char* low, const char* high, char_type* dest) const +{ + for (; low != high; ++low, ++dest) +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + *dest = btowc_l(*low, __l); +#else + *dest = __btowc_l(*low, __l); +#endif + return low; +} + +char +ctype_byname::do_narrow(char_type c, char dfault) const +{ +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + int r = wctob_l(c, __l); +#else + int r = __wctob_l(c, __l); +#endif + return r != static_cast(WEOF) ? static_cast(r) : dfault; +} + +const wchar_t* +ctype_byname::do_narrow(const char_type* low, const char_type* high, char dfault, char* dest) const +{ + for (; low != high; ++low, ++dest) + { +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + int r = wctob_l(*low, __l); +#else + int r = __wctob_l(*low, __l); +#endif + *dest = r != static_cast(WEOF) ? static_cast(r) : dfault; + } + return low; +} + +// template <> class codecvt + +locale::id codecvt::id; + +codecvt::~codecvt() +{ +} + +codecvt::result +codecvt::do_out(state_type&, + const intern_type* frm, const intern_type*, const intern_type*& frm_nxt, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + frm_nxt = frm; + to_nxt = to; + return noconv; +} + +codecvt::result +codecvt::do_in(state_type&, + const extern_type* frm, const extern_type*, const extern_type*& frm_nxt, + intern_type* to, intern_type*, intern_type*& to_nxt) const +{ + frm_nxt = frm; + to_nxt = to; + return noconv; +} + +codecvt::result +codecvt::do_unshift(state_type&, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + to_nxt = to; + return noconv; +} + +int +codecvt::do_encoding() const _NOEXCEPT +{ + return 1; +} + +bool +codecvt::do_always_noconv() const _NOEXCEPT +{ + return true; +} + +int +codecvt::do_length(state_type&, + const extern_type* frm, const extern_type* end, size_t mx) const +{ + return static_cast(min(mx, static_cast(end-frm))); +} + +int +codecvt::do_max_length() const _NOEXCEPT +{ + return 1; +} + +// template <> class codecvt + +locale::id codecvt::id; + +codecvt::codecvt(size_t refs) + : locale::facet(refs), + __l(_LIBCPP_GET_C_LOCALE) +{ +} + +codecvt::codecvt(const char* nm, size_t refs) + : locale::facet(refs), + __l(newlocale(LC_ALL_MASK, nm, 0)) +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + if (__l == 0) + throw runtime_error("codecvt_byname::codecvt_byname" + " failed to construct for " + string(nm)); +#endif // _LIBCPP_NO_EXCEPTIONS +} + +codecvt::~codecvt() +{ + if (__l != _LIBCPP_GET_C_LOCALE) + freelocale(__l); +} + +codecvt::result +codecvt::do_out(state_type& st, + const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + // look for first internal null in frm + const intern_type* fend = frm; + for (; fend != frm_end; ++fend) + if (*fend == 0) + break; + // loop over all null-terminated sequences in frm + to_nxt = to; + for (frm_nxt = frm; frm != frm_end && to != to_end; frm = frm_nxt, to = to_nxt) + { + // save state in case it is needed to recover to_nxt on error + mbstate_t save_state = st; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + size_t n = wcsnrtombs_l(to, &frm_nxt, static_cast(fend-frm), + static_cast(to_end-to), &st, __l); +#else + size_t n = __wcsnrtombs_l(to, &frm_nxt, fend-frm, to_end-to, &st, __l); +#endif + if (n == size_t(-1)) + { + // need to recover to_nxt + for (to_nxt = to; frm != frm_nxt; ++frm) + { +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + n = wcrtomb_l(to_nxt, *frm, &save_state, __l); +#else + n = __wcrtomb_l(to_nxt, *frm, &save_state, __l); +#endif + if (n == size_t(-1)) + break; + to_nxt += n; + } + frm_nxt = frm; + return error; + } + if (n == 0) + return partial; + to_nxt += n; + if (to_nxt == to_end) + break; + if (fend != frm_end) // set up next null terminated sequence + { + // Try to write the terminating null + extern_type tmp[MB_LEN_MAX]; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + n = wcrtomb_l(tmp, intern_type(), &st, __l); +#else + n = __wcrtomb_l(tmp, intern_type(), &st, __l); +#endif + if (n == size_t(-1)) // on error + return error; + if (n > static_cast(to_end-to_nxt)) // is there room? + return partial; + for (extern_type* p = tmp; n; --n) // write it + *to_nxt++ = *p++; + ++frm_nxt; + // look for next null in frm + for (fend = frm_nxt; fend != frm_end; ++fend) + if (*fend == 0) + break; + } + } + return frm_nxt == frm_end ? ok : partial; +} + +codecvt::result +codecvt::do_in(state_type& st, + const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, + intern_type* to, intern_type* to_end, intern_type*& to_nxt) const +{ + // look for first internal null in frm + const extern_type* fend = frm; + for (; fend != frm_end; ++fend) + if (*fend == 0) + break; + // loop over all null-terminated sequences in frm + to_nxt = to; + for (frm_nxt = frm; frm != frm_end && to != to_end; frm = frm_nxt, to = to_nxt) + { + // save state in case it is needed to recover to_nxt on error + mbstate_t save_state = st; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + size_t n = mbsnrtowcs_l(to, &frm_nxt, static_cast(fend-frm), + static_cast(to_end-to), &st, __l); +#else + size_t n = __mbsnrtowcs_l(to, &frm_nxt, fend-frm, to_end-to, &st, __l); +#endif + if (n == size_t(-1)) + { + // need to recover to_nxt + for (to_nxt = to; frm != frm_nxt; ++to_nxt) + { +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + n = mbrtowc_l(to_nxt, frm, static_cast(fend-frm), + &save_state, __l); +#else + n = __mbrtowc_l(to_nxt, frm, fend-frm, &save_state, __l); +#endif + switch (n) + { + case 0: + ++frm; + break; + case size_t(-1): + frm_nxt = frm; + return error; + case size_t(-2): + frm_nxt = frm; + return partial; + default: + frm += n; + break; + } + } + frm_nxt = frm; + return frm_nxt == frm_end ? ok : partial; + } + if (n == (size_t)-1) + return error; + to_nxt += n; + if (to_nxt == to_end) + break; + if (fend != frm_end) // set up next null terminated sequence + { + // Try to write the terminating null +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + n = mbrtowc_l(to_nxt, frm_nxt, 1, &st, __l); +#else + n = __mbrtowc_l(to_nxt, frm_nxt, 1, &st, __l); +#endif + if (n != 0) // on error + return error; + ++to_nxt; + ++frm_nxt; + // look for next null in frm + for (fend = frm_nxt; fend != frm_end; ++fend) + if (*fend == 0) + break; + } + } + return frm_nxt == frm_end ? ok : partial; +} + +codecvt::result +codecvt::do_unshift(state_type& st, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + to_nxt = to; + extern_type tmp[MB_LEN_MAX]; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + size_t n = wcrtomb_l(tmp, intern_type(), &st, __l); +#else + size_t n = __wcrtomb_l(tmp, intern_type(), &st, __l); +#endif + if (n == size_t(-1) || n == 0) // on error + return error; + --n; + if (n > static_cast(to_end-to_nxt)) // is there room? + return partial; + for (extern_type* p = tmp; n; --n) // write it + *to_nxt++ = *p++; + return ok; +} + +int +codecvt::do_encoding() const _NOEXCEPT +{ +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + if (mbtowc_l(nullptr, nullptr, MB_LEN_MAX, __l) == 0) +#else + if (__mbtowc_l(nullptr, nullptr, MB_LEN_MAX, __l) == 0) +#endif + { + // stateless encoding +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + if (__l == 0 || MB_CUR_MAX_L(__l) == 1) // there are no known constant length encodings +#else + if (__l == 0 || __mb_cur_max_l(__l) == 1) // there are no known constant length encodings +#endif + return 1; // which take more than 1 char to form a wchar_t + return 0; + } + return -1; +} + +bool +codecvt::do_always_noconv() const _NOEXCEPT +{ + return false; +} + +int +codecvt::do_length(state_type& st, + const extern_type* frm, const extern_type* frm_end, size_t mx) const +{ + int nbytes = 0; + for (size_t nwchar_t = 0; nwchar_t < mx && frm != frm_end; ++nwchar_t) + { +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + size_t n = mbrlen_l(frm, static_cast(frm_end-frm), &st, __l); +#else + size_t n = __mbrlen_l(frm, frm_end-frm, &st, __l); +#endif + switch (n) + { + case 0: + ++nbytes; + ++frm; + break; + case size_t(-1): + case size_t(-2): + return nbytes; + default: + nbytes += n; + frm += n; + break; + } + } + return nbytes; +} + +int +codecvt::do_max_length() const _NOEXCEPT +{ +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + return __l == 0 ? 1 : static_cast( MB_CUR_MAX_L(__l)); +#else + return __l == 0 ? 1 : static_cast(__mb_cur_max_l(__l)); +#endif +} + +// Valid UTF ranges +// UTF-32 UTF-16 UTF-8 # of code points +// first second first second third fourth +// 000000 - 00007F 0000 - 007F 00 - 7F 127 +// 000080 - 0007FF 0080 - 07FF C2 - DF, 80 - BF 1920 +// 000800 - 000FFF 0800 - 0FFF E0 - E0, A0 - BF, 80 - BF 2048 +// 001000 - 00CFFF 1000 - CFFF E1 - EC, 80 - BF, 80 - BF 49152 +// 00D000 - 00D7FF D000 - D7FF ED - ED, 80 - 9F, 80 - BF 2048 +// 00D800 - 00DFFF invalid +// 00E000 - 00FFFF E000 - FFFF EE - EF, 80 - BF, 80 - BF 8192 +// 010000 - 03FFFF D800 - D8BF, DC00 - DFFF F0 - F0, 90 - BF, 80 - BF, 80 - BF 196608 +// 040000 - 0FFFFF D8C0 - DBBF, DC00 - DFFF F1 - F3, 80 - BF, 80 - BF, 80 - BF 786432 +// 100000 - 10FFFF DBC0 - DBFF, DC00 - DFFF F4 - F4, 80 - 8F, 80 - BF, 80 - BF 65536 + +template +typename enable_if +< + is_same<_FrmType, uint16_t>::value || is_same<_FrmType, uint32_t>::value, + codecvt_base::result +>::type +utf16_to_utf8(const _FrmType* frm, const _FrmType* frm_end, const _FrmType*& frm_nxt, + uint8_t* to, uint8_t* to_end, uint8_t*& to_nxt, + unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) +{ + frm_nxt = frm; + to_nxt = to; + if (mode & generate_header) + { + if (to_end-to_nxt < 3) + return codecvt_base::partial; + *to_nxt++ = static_cast(0xEF); + *to_nxt++ = static_cast(0xBB); + *to_nxt++ = static_cast(0xBF); + } + for (; frm_nxt < frm_end; ++frm_nxt) + { + uint16_t wc1 = static_cast(*frm_nxt); + if (wc1 > Maxcode) + return codecvt_base::error; + if (wc1 < 0x0080) + { + if (to_end-to_nxt < 1) + return codecvt_base::partial; + *to_nxt++ = static_cast(wc1); + } + else if (wc1 < 0x0800) + { + if (to_end-to_nxt < 2) + return codecvt_base::partial; + *to_nxt++ = static_cast(0xC0 | (wc1 >> 6)); + *to_nxt++ = static_cast(0x80 | (wc1 & 0x03F)); + } + else if (wc1 < 0xD800) + { + if (to_end-to_nxt < 3) + return codecvt_base::partial; + *to_nxt++ = static_cast(0xE0 | (wc1 >> 12)); + *to_nxt++ = static_cast(0x80 | ((wc1 & 0x0FC0) >> 6)); + *to_nxt++ = static_cast(0x80 | (wc1 & 0x003F)); + } + else if (wc1 < 0xDC00) + { + if (frm_end-frm_nxt < 2) + return codecvt_base::partial; + uint16_t wc2 = static_cast(frm_nxt[1]); + if ((wc2 & 0xFC00) != 0xDC00) + return codecvt_base::error; + if (to_end-to_nxt < 4) + return codecvt_base::partial; + if (((((wc1 & 0x03C0UL) >> 6) + 1) << 16) + + ((wc1 & 0x003FUL) << 10) + (wc2 & 0x03FF) > Maxcode) + return codecvt_base::error; + ++frm_nxt; + uint8_t z = ((wc1 & 0x03C0) >> 6) + 1; + *to_nxt++ = static_cast(0xF0 | (z >> 2)); + *to_nxt++ = static_cast(0x80 | ((z & 0x03) << 4) | ((wc1 & 0x003C) >> 2)); + *to_nxt++ = static_cast(0x80 | ((wc1 & 0x0003) << 4) | ((wc2 & 0x03C0) >> 6)); + *to_nxt++ = static_cast(0x80 | (wc2 & 0x003F)); + } + else if (wc1 < 0xE000) + { + return codecvt_base::error; + } + else + { + if (to_end-to_nxt < 3) + return codecvt_base::partial; + *to_nxt++ = static_cast(0xE0 | (wc1 >> 12)); + *to_nxt++ = static_cast(0x80 | ((wc1 & 0x0FC0) >> 6)); + *to_nxt++ = static_cast(0x80 | (wc1 & 0x003F)); + } + } + return codecvt_base::ok; +} + +template +typename enable_if +< + is_same<_ToType, uint16_t>::value || is_same<_ToType, uint32_t>::value, + codecvt_base::result +>::type +utf8_to_utf16(const uint8_t* frm, const uint8_t* frm_end, const uint8_t*& frm_nxt, + _ToType* to, _ToType* to_end, _ToType*& to_nxt, + unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) +{ + frm_nxt = frm; + to_nxt = to; + if (mode & consume_header) + { + if (frm_end-frm_nxt >= 3 && frm_nxt[0] == 0xEF && frm_nxt[1] == 0xBB && + frm_nxt[2] == 0xBF) + frm_nxt += 3; + } + for (; frm_nxt < frm_end && to_nxt < to_end; ++to_nxt) + { + uint8_t c1 = *frm_nxt; + if (c1 > Maxcode) + return codecvt_base::error; + if (c1 < 0x80) + { + *to_nxt = static_cast<_ToType>(c1); + ++frm_nxt; + } + else if (c1 < 0xC2) + { + return codecvt_base::error; + } + else if (c1 < 0xE0) + { + if (frm_end-frm_nxt < 2) + return codecvt_base::partial; + uint8_t c2 = frm_nxt[1]; + if ((c2 & 0xC0) != 0x80) + return codecvt_base::error; + uint16_t t = static_cast(((c1 & 0x1F) << 6) | (c2 & 0x3F)); + if (t > Maxcode) + return codecvt_base::error; + *to_nxt = static_cast<_ToType>(t); + frm_nxt += 2; + } + else if (c1 < 0xF0) + { + if (frm_end-frm_nxt < 3) + return codecvt_base::partial; + uint8_t c2 = frm_nxt[1]; + uint8_t c3 = frm_nxt[2]; + switch (c1) + { + case 0xE0: + if ((c2 & 0xE0) != 0xA0) + return codecvt_base::error; + break; + case 0xED: + if ((c2 & 0xE0) != 0x80) + return codecvt_base::error; + break; + default: + if ((c2 & 0xC0) != 0x80) + return codecvt_base::error; + break; + } + if ((c3 & 0xC0) != 0x80) + return codecvt_base::error; + uint16_t t = static_cast(((c1 & 0x0F) << 12) + | ((c2 & 0x3F) << 6) + | (c3 & 0x3F)); + if (t > Maxcode) + return codecvt_base::error; + *to_nxt = static_cast<_ToType>(t); + frm_nxt += 3; + } + else if (c1 < 0xF5) + { + if (frm_end-frm_nxt < 4) + return codecvt_base::partial; + uint8_t c2 = frm_nxt[1]; + uint8_t c3 = frm_nxt[2]; + uint8_t c4 = frm_nxt[3]; + switch (c1) + { + case 0xF0: + if (!(0x90 <= c2 && c2 <= 0xBF)) + return codecvt_base::error; + break; + case 0xF4: + if ((c2 & 0xF0) != 0x80) + return codecvt_base::error; + break; + default: + if ((c2 & 0xC0) != 0x80) + return codecvt_base::error; + break; + } + if ((c3 & 0xC0) != 0x80 || (c4 & 0xC0) != 0x80) + return codecvt_base::error; + if (to_end-to_nxt < 2) + return codecvt_base::partial; + if ((((c1 & 7UL) << 18) + + ((c2 & 0x3FUL) << 12) + + ((c3 & 0x3FUL) << 6) + (c4 & 0x3F)) > Maxcode) + return codecvt_base::error; + *to_nxt = static_cast<_ToType>( + 0xD800 + | (((((c1 & 0x07) << 2) | ((c2 & 0x30) >> 4)) - 1) << 6) + | ((c2 & 0x0F) << 2) + | ((c3 & 0x30) >> 4)); + *++to_nxt = static_cast<_ToType>( + 0xDC00 + | ((c3 & 0x0F) << 6) + | (c4 & 0x3F)); + frm_nxt += 4; + } + else + { + return codecvt_base::error; + } + } + return frm_nxt < frm_end ? codecvt_base::partial : codecvt_base::ok; +} + +static +int +utf8_to_utf16_length(const uint8_t* frm, const uint8_t* frm_end, + size_t mx, unsigned long Maxcode = 0x10FFFF, + codecvt_mode mode = codecvt_mode(0)) +{ + const uint8_t* frm_nxt = frm; + if (mode & consume_header) + { + if (frm_end-frm_nxt >= 3 && frm_nxt[0] == 0xEF && frm_nxt[1] == 0xBB && + frm_nxt[2] == 0xBF) + frm_nxt += 3; + } + for (size_t nchar16_t = 0; frm_nxt < frm_end && nchar16_t < mx; ++nchar16_t) + { + uint8_t c1 = *frm_nxt; + if (c1 > Maxcode) + break; + if (c1 < 0x80) + { + ++frm_nxt; + } + else if (c1 < 0xC2) + { + break; + } + else if (c1 < 0xE0) + { + if ((frm_end-frm_nxt < 2) || (frm_nxt[1] & 0xC0) != 0x80) + break; + uint16_t t = static_cast(((c1 & 0x1F) << 6) | (frm_nxt[1] & 0x3F)); + if (t > Maxcode) + break; + frm_nxt += 2; + } + else if (c1 < 0xF0) + { + if (frm_end-frm_nxt < 3) + break; + uint8_t c2 = frm_nxt[1]; + uint8_t c3 = frm_nxt[2]; + switch (c1) + { + case 0xE0: + if ((c2 & 0xE0) != 0xA0) + return static_cast(frm_nxt - frm); + break; + case 0xED: + if ((c2 & 0xE0) != 0x80) + return static_cast(frm_nxt - frm); + break; + default: + if ((c2 & 0xC0) != 0x80) + return static_cast(frm_nxt - frm); + break; + } + if ((c3 & 0xC0) != 0x80) + break; + if ((((c1 & 0x0Fu) << 12) | ((c2 & 0x3Fu) << 6) | (c3 & 0x3Fu)) > Maxcode) + break; + frm_nxt += 3; + } + else if (c1 < 0xF5) + { + if (frm_end-frm_nxt < 4 || mx-nchar16_t < 2) + break; + uint8_t c2 = frm_nxt[1]; + uint8_t c3 = frm_nxt[2]; + uint8_t c4 = frm_nxt[3]; + switch (c1) + { + case 0xF0: + if (!(0x90 <= c2 && c2 <= 0xBF)) + return static_cast(frm_nxt - frm); + break; + case 0xF4: + if ((c2 & 0xF0) != 0x80) + return static_cast(frm_nxt - frm); + break; + default: + if ((c2 & 0xC0) != 0x80) + return static_cast(frm_nxt - frm); + break; + } + if ((c3 & 0xC0) != 0x80 || (c4 & 0xC0) != 0x80) + break; + if ((((c1 & 7UL) << 18) + + ((c2 & 0x3FUL) << 12) + + ((c3 & 0x3FUL) << 6) + (c4 & 0x3F)) > Maxcode) + break; + ++nchar16_t; + frm_nxt += 4; + } + else + { + break; + } + } + return static_cast(frm_nxt - frm); +} + +static +codecvt_base::result +ucs4_to_utf8(const uint32_t* frm, const uint32_t* frm_end, const uint32_t*& frm_nxt, + uint8_t* to, uint8_t* to_end, uint8_t*& to_nxt, + unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) +{ + frm_nxt = frm; + to_nxt = to; + if (mode & generate_header) + { + if (to_end-to_nxt < 3) + return codecvt_base::partial; + *to_nxt++ = static_cast(0xEF); + *to_nxt++ = static_cast(0xBB); + *to_nxt++ = static_cast(0xBF); + } + for (; frm_nxt < frm_end; ++frm_nxt) + { + uint32_t wc = *frm_nxt; + if ((wc & 0xFFFFF800) == 0x00D800 || wc > Maxcode) + return codecvt_base::error; + if (wc < 0x000080) + { + if (to_end-to_nxt < 1) + return codecvt_base::partial; + *to_nxt++ = static_cast(wc); + } + else if (wc < 0x000800) + { + if (to_end-to_nxt < 2) + return codecvt_base::partial; + *to_nxt++ = static_cast(0xC0 | (wc >> 6)); + *to_nxt++ = static_cast(0x80 | (wc & 0x03F)); + } + else if (wc < 0x010000) + { + if (to_end-to_nxt < 3) + return codecvt_base::partial; + *to_nxt++ = static_cast(0xE0 | (wc >> 12)); + *to_nxt++ = static_cast(0x80 | ((wc & 0x0FC0) >> 6)); + *to_nxt++ = static_cast(0x80 | (wc & 0x003F)); + } + else // if (wc < 0x110000) + { + if (to_end-to_nxt < 4) + return codecvt_base::partial; + *to_nxt++ = static_cast(0xF0 | (wc >> 18)); + *to_nxt++ = static_cast(0x80 | ((wc & 0x03F000) >> 12)); + *to_nxt++ = static_cast(0x80 | ((wc & 0x000FC0) >> 6)); + *to_nxt++ = static_cast(0x80 | (wc & 0x00003F)); + } + } + return codecvt_base::ok; +} + +static +codecvt_base::result +utf8_to_ucs4(const uint8_t* frm, const uint8_t* frm_end, const uint8_t*& frm_nxt, + uint32_t* to, uint32_t* to_end, uint32_t*& to_nxt, + unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) +{ + frm_nxt = frm; + to_nxt = to; + if (mode & consume_header) + { + if (frm_end-frm_nxt >= 3 && frm_nxt[0] == 0xEF && frm_nxt[1] == 0xBB && + frm_nxt[2] == 0xBF) + frm_nxt += 3; + } + for (; frm_nxt < frm_end && to_nxt < to_end; ++to_nxt) + { + uint8_t c1 = static_cast(*frm_nxt); + if (c1 < 0x80) + { + if (c1 > Maxcode) + return codecvt_base::error; + *to_nxt = static_cast(c1); + ++frm_nxt; + } + else if (c1 < 0xC2) + { + return codecvt_base::error; + } + else if (c1 < 0xE0) + { + if (frm_end-frm_nxt < 2) + return codecvt_base::partial; + uint8_t c2 = frm_nxt[1]; + if ((c2 & 0xC0) != 0x80) + return codecvt_base::error; + uint32_t t = static_cast(((c1 & 0x1F) << 6) + | (c2 & 0x3F)); + if (t > Maxcode) + return codecvt_base::error; + *to_nxt = t; + frm_nxt += 2; + } + else if (c1 < 0xF0) + { + if (frm_end-frm_nxt < 3) + return codecvt_base::partial; + uint8_t c2 = frm_nxt[1]; + uint8_t c3 = frm_nxt[2]; + switch (c1) + { + case 0xE0: + if ((c2 & 0xE0) != 0xA0) + return codecvt_base::error; + break; + case 0xED: + if ((c2 & 0xE0) != 0x80) + return codecvt_base::error; + break; + default: + if ((c2 & 0xC0) != 0x80) + return codecvt_base::error; + break; + } + if ((c3 & 0xC0) != 0x80) + return codecvt_base::error; + uint32_t t = static_cast(((c1 & 0x0F) << 12) + | ((c2 & 0x3F) << 6) + | (c3 & 0x3F)); + if (t > Maxcode) + return codecvt_base::error; + *to_nxt = t; + frm_nxt += 3; + } + else if (c1 < 0xF5) + { + if (frm_end-frm_nxt < 4) + return codecvt_base::partial; + uint8_t c2 = frm_nxt[1]; + uint8_t c3 = frm_nxt[2]; + uint8_t c4 = frm_nxt[3]; + switch (c1) + { + case 0xF0: + if (!(0x90 <= c2 && c2 <= 0xBF)) + return codecvt_base::error; + break; + case 0xF4: + if ((c2 & 0xF0) != 0x80) + return codecvt_base::error; + break; + default: + if ((c2 & 0xC0) != 0x80) + return codecvt_base::error; + break; + } + if ((c3 & 0xC0) != 0x80 || (c4 & 0xC0) != 0x80) + return codecvt_base::error; + uint32_t t = static_cast(((c1 & 0x07) << 18) + | ((c2 & 0x3F) << 12) + | ((c3 & 0x3F) << 6) + | (c4 & 0x3F)); + if (t > Maxcode) + return codecvt_base::error; + *to_nxt = t; + frm_nxt += 4; + } + else + { + return codecvt_base::error; + } + } + return frm_nxt < frm_end ? codecvt_base::partial : codecvt_base::ok; +} + +static +int +utf8_to_ucs4_length(const uint8_t* frm, const uint8_t* frm_end, + size_t mx, unsigned long Maxcode = 0x10FFFF, + codecvt_mode mode = codecvt_mode(0)) +{ + const uint8_t* frm_nxt = frm; + if (mode & consume_header) + { + if (frm_end-frm_nxt >= 3 && frm_nxt[0] == 0xEF && frm_nxt[1] == 0xBB && + frm_nxt[2] == 0xBF) + frm_nxt += 3; + } + for (size_t nchar32_t = 0; frm_nxt < frm_end && nchar32_t < mx; ++nchar32_t) + { + uint8_t c1 = static_cast(*frm_nxt); + if (c1 < 0x80) + { + if (c1 > Maxcode) + break; + ++frm_nxt; + } + else if (c1 < 0xC2) + { + break; + } + else if (c1 < 0xE0) + { + if ((frm_end-frm_nxt < 2) || ((frm_nxt[1] & 0xC0) != 0x80)) + break; + if ((((c1 & 0x1Fu) << 6) | (frm_nxt[1] & 0x3Fu)) > Maxcode) + break; + frm_nxt += 2; + } + else if (c1 < 0xF0) + { + if (frm_end-frm_nxt < 3) + break; + uint8_t c2 = frm_nxt[1]; + uint8_t c3 = frm_nxt[2]; + switch (c1) + { + case 0xE0: + if ((c2 & 0xE0) != 0xA0) + return static_cast(frm_nxt - frm); + break; + case 0xED: + if ((c2 & 0xE0) != 0x80) + return static_cast(frm_nxt - frm); + break; + default: + if ((c2 & 0xC0) != 0x80) + return static_cast(frm_nxt - frm); + break; + } + if ((c3 & 0xC0) != 0x80) + break; + if ((((c1 & 0x0Fu) << 12) | ((c2 & 0x3Fu) << 6) | (c3 & 0x3Fu)) > Maxcode) + break; + frm_nxt += 3; + } + else if (c1 < 0xF5) + { + if (frm_end-frm_nxt < 4) + break; + uint8_t c2 = frm_nxt[1]; + uint8_t c3 = frm_nxt[2]; + uint8_t c4 = frm_nxt[3]; + switch (c1) + { + case 0xF0: + if (!(0x90 <= c2 && c2 <= 0xBF)) + return static_cast(frm_nxt - frm); + break; + case 0xF4: + if ((c2 & 0xF0) != 0x80) + return static_cast(frm_nxt - frm); + break; + default: + if ((c2 & 0xC0) != 0x80) + return static_cast(frm_nxt - frm); + break; + } + if ((c3 & 0xC0) != 0x80 || (c4 & 0xC0) != 0x80) + break; + if ((((c1 & 0x07u) << 18) | ((c2 & 0x3Fu) << 12) | + ((c3 & 0x3Fu) << 6) | (c4 & 0x3Fu)) > Maxcode) + break; + frm_nxt += 4; + } + else + { + break; + } + } + return static_cast(frm_nxt - frm); +} + +static +codecvt_base::result +ucs2_to_utf8(const uint16_t* frm, const uint16_t* frm_end, const uint16_t*& frm_nxt, + uint8_t* to, uint8_t* to_end, uint8_t*& to_nxt, + unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) +{ + frm_nxt = frm; + to_nxt = to; + if (mode & generate_header) + { + if (to_end-to_nxt < 3) + return codecvt_base::partial; + *to_nxt++ = static_cast(0xEF); + *to_nxt++ = static_cast(0xBB); + *to_nxt++ = static_cast(0xBF); + } + for (; frm_nxt < frm_end; ++frm_nxt) + { + uint16_t wc = *frm_nxt; + if ((wc & 0xF800) == 0xD800 || wc > Maxcode) + return codecvt_base::error; + if (wc < 0x0080) + { + if (to_end-to_nxt < 1) + return codecvt_base::partial; + *to_nxt++ = static_cast(wc); + } + else if (wc < 0x0800) + { + if (to_end-to_nxt < 2) + return codecvt_base::partial; + *to_nxt++ = static_cast(0xC0 | (wc >> 6)); + *to_nxt++ = static_cast(0x80 | (wc & 0x03F)); + } + else // if (wc <= 0xFFFF) + { + if (to_end-to_nxt < 3) + return codecvt_base::partial; + *to_nxt++ = static_cast(0xE0 | (wc >> 12)); + *to_nxt++ = static_cast(0x80 | ((wc & 0x0FC0) >> 6)); + *to_nxt++ = static_cast(0x80 | (wc & 0x003F)); + } + } + return codecvt_base::ok; +} + +static +codecvt_base::result +utf8_to_ucs2(const uint8_t* frm, const uint8_t* frm_end, const uint8_t*& frm_nxt, + uint16_t* to, uint16_t* to_end, uint16_t*& to_nxt, + unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) +{ + frm_nxt = frm; + to_nxt = to; + if (mode & consume_header) + { + if (frm_end-frm_nxt >= 3 && frm_nxt[0] == 0xEF && frm_nxt[1] == 0xBB && + frm_nxt[2] == 0xBF) + frm_nxt += 3; + } + for (; frm_nxt < frm_end && to_nxt < to_end; ++to_nxt) + { + uint8_t c1 = static_cast(*frm_nxt); + if (c1 < 0x80) + { + if (c1 > Maxcode) + return codecvt_base::error; + *to_nxt = static_cast(c1); + ++frm_nxt; + } + else if (c1 < 0xC2) + { + return codecvt_base::error; + } + else if (c1 < 0xE0) + { + if (frm_end-frm_nxt < 2) + return codecvt_base::partial; + uint8_t c2 = frm_nxt[1]; + if ((c2 & 0xC0) != 0x80) + return codecvt_base::error; + uint16_t t = static_cast(((c1 & 0x1F) << 6) + | (c2 & 0x3F)); + if (t > Maxcode) + return codecvt_base::error; + *to_nxt = t; + frm_nxt += 2; + } + else if (c1 < 0xF0) + { + if (frm_end-frm_nxt < 3) + return codecvt_base::partial; + uint8_t c2 = frm_nxt[1]; + uint8_t c3 = frm_nxt[2]; + switch (c1) + { + case 0xE0: + if ((c2 & 0xE0) != 0xA0) + return codecvt_base::error; + break; + case 0xED: + if ((c2 & 0xE0) != 0x80) + return codecvt_base::error; + break; + default: + if ((c2 & 0xC0) != 0x80) + return codecvt_base::error; + break; + } + if ((c3 & 0xC0) != 0x80) + return codecvt_base::error; + uint16_t t = static_cast(((c1 & 0x0F) << 12) + | ((c2 & 0x3F) << 6) + | (c3 & 0x3F)); + if (t > Maxcode) + return codecvt_base::error; + *to_nxt = t; + frm_nxt += 3; + } + else + { + return codecvt_base::error; + } + } + return frm_nxt < frm_end ? codecvt_base::partial : codecvt_base::ok; +} + +static +int +utf8_to_ucs2_length(const uint8_t* frm, const uint8_t* frm_end, + size_t mx, unsigned long Maxcode = 0x10FFFF, + codecvt_mode mode = codecvt_mode(0)) +{ + const uint8_t* frm_nxt = frm; + if (mode & consume_header) + { + if (frm_end-frm_nxt >= 3 && frm_nxt[0] == 0xEF && frm_nxt[1] == 0xBB && + frm_nxt[2] == 0xBF) + frm_nxt += 3; + } + for (size_t nchar32_t = 0; frm_nxt < frm_end && nchar32_t < mx; ++nchar32_t) + { + uint8_t c1 = static_cast(*frm_nxt); + if (c1 < 0x80) + { + if (c1 > Maxcode) + break; + ++frm_nxt; + } + else if (c1 < 0xC2) + { + break; + } + else if (c1 < 0xE0) + { + if ((frm_end-frm_nxt < 2) || ((frm_nxt[1] & 0xC0) != 0x80)) + break; + if ((((c1 & 0x1Fu) << 6) | (frm_nxt[1] & 0x3Fu)) > Maxcode) + break; + frm_nxt += 2; + } + else if (c1 < 0xF0) + { + if (frm_end-frm_nxt < 3) + break; + uint8_t c2 = frm_nxt[1]; + uint8_t c3 = frm_nxt[2]; + switch (c1) + { + case 0xE0: + if ((c2 & 0xE0) != 0xA0) + return static_cast(frm_nxt - frm); + break; + case 0xED: + if ((c2 & 0xE0) != 0x80) + return static_cast(frm_nxt - frm); + break; + default: + if ((c2 & 0xC0) != 0x80) + return static_cast(frm_nxt - frm); + break; + } + if ((c3 & 0xC0) != 0x80) + break; + if ((((c1 & 0x0Fu) << 12) | ((c2 & 0x3Fu) << 6) | (c3 & 0x3Fu)) > Maxcode) + break; + frm_nxt += 3; + } + else + { + break; + } + } + return static_cast(frm_nxt - frm); +} + +static +codecvt_base::result +ucs4_to_utf16be(const uint32_t* frm, const uint32_t* frm_end, const uint32_t*& frm_nxt, + uint8_t* to, uint8_t* to_end, uint8_t*& to_nxt, + unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) +{ + frm_nxt = frm; + to_nxt = to; + if (mode & generate_header) + { + if (to_end-to_nxt < 2) + return codecvt_base::partial; + *to_nxt++ = static_cast(0xFE); + *to_nxt++ = static_cast(0xFF); + } + for (; frm_nxt < frm_end; ++frm_nxt) + { + uint32_t wc = *frm_nxt; + if ((wc & 0xFFFFF800) == 0x00D800 || wc > Maxcode) + return codecvt_base::error; + if (wc < 0x010000) + { + if (to_end-to_nxt < 2) + return codecvt_base::partial; + *to_nxt++ = static_cast(wc >> 8); + *to_nxt++ = static_cast(wc); + } + else + { + if (to_end-to_nxt < 4) + return codecvt_base::partial; + uint16_t t = static_cast( + 0xD800 + | ((((wc & 0x1F0000) >> 16) - 1) << 6) + | ((wc & 0x00FC00) >> 10)); + *to_nxt++ = static_cast(t >> 8); + *to_nxt++ = static_cast(t); + t = static_cast(0xDC00 | (wc & 0x03FF)); + *to_nxt++ = static_cast(t >> 8); + *to_nxt++ = static_cast(t); + } + } + return codecvt_base::ok; +} + +static +codecvt_base::result +utf16be_to_ucs4(const uint8_t* frm, const uint8_t* frm_end, const uint8_t*& frm_nxt, + uint32_t* to, uint32_t* to_end, uint32_t*& to_nxt, + unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) +{ + frm_nxt = frm; + to_nxt = to; + if (mode & consume_header) + { + if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFE && frm_nxt[1] == 0xFF) + frm_nxt += 2; + } + for (; frm_nxt < frm_end - 1 && to_nxt < to_end; ++to_nxt) + { + uint16_t c1 = static_cast(frm_nxt[0] << 8 | frm_nxt[1]); + if ((c1 & 0xFC00) == 0xDC00) + return codecvt_base::error; + if ((c1 & 0xFC00) != 0xD800) + { + if (c1 > Maxcode) + return codecvt_base::error; + *to_nxt = static_cast(c1); + frm_nxt += 2; + } + else + { + if (frm_end-frm_nxt < 4) + return codecvt_base::partial; + uint16_t c2 = static_cast(frm_nxt[2] << 8 | frm_nxt[3]); + if ((c2 & 0xFC00) != 0xDC00) + return codecvt_base::error; + uint32_t t = static_cast( + ((((c1 & 0x03C0) >> 6) + 1) << 16) + | ((c1 & 0x003F) << 10) + | (c2 & 0x03FF)); + if (t > Maxcode) + return codecvt_base::error; + *to_nxt = t; + frm_nxt += 4; + } + } + return frm_nxt < frm_end ? codecvt_base::partial : codecvt_base::ok; +} + +static +int +utf16be_to_ucs4_length(const uint8_t* frm, const uint8_t* frm_end, + size_t mx, unsigned long Maxcode = 0x10FFFF, + codecvt_mode mode = codecvt_mode(0)) +{ + const uint8_t* frm_nxt = frm; + if (mode & consume_header) + { + if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFE && frm_nxt[1] == 0xFF) + frm_nxt += 2; + } + for (size_t nchar32_t = 0; frm_nxt < frm_end - 1 && nchar32_t < mx; ++nchar32_t) + { + uint16_t c1 = static_cast(frm_nxt[0] << 8 | frm_nxt[1]); + if ((c1 & 0xFC00) == 0xDC00) + break; + if ((c1 & 0xFC00) != 0xD800) + { + if (c1 > Maxcode) + break; + frm_nxt += 2; + } + else + { + if (frm_end-frm_nxt < 4) + break; + uint16_t c2 = static_cast(frm_nxt[2] << 8 | frm_nxt[3]); + if ((c2 & 0xFC00) != 0xDC00) + break; + uint32_t t = static_cast( + ((((c1 & 0x03C0) >> 6) + 1) << 16) + | ((c1 & 0x003F) << 10) + | (c2 & 0x03FF)); + if (t > Maxcode) + break; + frm_nxt += 4; + } + } + return static_cast(frm_nxt - frm); +} + +static +codecvt_base::result +ucs4_to_utf16le(const uint32_t* frm, const uint32_t* frm_end, const uint32_t*& frm_nxt, + uint8_t* to, uint8_t* to_end, uint8_t*& to_nxt, + unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) +{ + frm_nxt = frm; + to_nxt = to; + if (mode & generate_header) + { + if (to_end-to_nxt < 2) + return codecvt_base::partial; + *to_nxt++ = static_cast(0xFF); + *to_nxt++ = static_cast(0xFE); + } + for (; frm_nxt < frm_end; ++frm_nxt) + { + uint32_t wc = *frm_nxt; + if ((wc & 0xFFFFF800) == 0x00D800 || wc > Maxcode) + return codecvt_base::error; + if (wc < 0x010000) + { + if (to_end-to_nxt < 2) + return codecvt_base::partial; + *to_nxt++ = static_cast(wc); + *to_nxt++ = static_cast(wc >> 8); + } + else + { + if (to_end-to_nxt < 4) + return codecvt_base::partial; + uint16_t t = static_cast( + 0xD800 + | ((((wc & 0x1F0000) >> 16) - 1) << 6) + | ((wc & 0x00FC00) >> 10)); + *to_nxt++ = static_cast(t); + *to_nxt++ = static_cast(t >> 8); + t = static_cast(0xDC00 | (wc & 0x03FF)); + *to_nxt++ = static_cast(t); + *to_nxt++ = static_cast(t >> 8); + } + } + return codecvt_base::ok; +} + +static +codecvt_base::result +utf16le_to_ucs4(const uint8_t* frm, const uint8_t* frm_end, const uint8_t*& frm_nxt, + uint32_t* to, uint32_t* to_end, uint32_t*& to_nxt, + unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) +{ + frm_nxt = frm; + to_nxt = to; + if (mode & consume_header) + { + if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFF && frm_nxt[1] == 0xFE) + frm_nxt += 2; + } + for (; frm_nxt < frm_end - 1 && to_nxt < to_end; ++to_nxt) + { + uint16_t c1 = static_cast(frm_nxt[1] << 8 | frm_nxt[0]); + if ((c1 & 0xFC00) == 0xDC00) + return codecvt_base::error; + if ((c1 & 0xFC00) != 0xD800) + { + if (c1 > Maxcode) + return codecvt_base::error; + *to_nxt = static_cast(c1); + frm_nxt += 2; + } + else + { + if (frm_end-frm_nxt < 4) + return codecvt_base::partial; + uint16_t c2 = static_cast(frm_nxt[3] << 8 | frm_nxt[2]); + if ((c2 & 0xFC00) != 0xDC00) + return codecvt_base::error; + uint32_t t = static_cast( + ((((c1 & 0x03C0) >> 6) + 1) << 16) + | ((c1 & 0x003F) << 10) + | (c2 & 0x03FF)); + if (t > Maxcode) + return codecvt_base::error; + *to_nxt = t; + frm_nxt += 4; + } + } + return frm_nxt < frm_end ? codecvt_base::partial : codecvt_base::ok; +} + +static +int +utf16le_to_ucs4_length(const uint8_t* frm, const uint8_t* frm_end, + size_t mx, unsigned long Maxcode = 0x10FFFF, + codecvt_mode mode = codecvt_mode(0)) +{ + const uint8_t* frm_nxt = frm; + if (mode & consume_header) + { + if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFF && frm_nxt[1] == 0xFE) + frm_nxt += 2; + } + for (size_t nchar32_t = 0; frm_nxt < frm_end - 1 && nchar32_t < mx; ++nchar32_t) + { + uint16_t c1 = static_cast(frm_nxt[1] << 8 | frm_nxt[0]); + if ((c1 & 0xFC00) == 0xDC00) + break; + if ((c1 & 0xFC00) != 0xD800) + { + if (c1 > Maxcode) + break; + frm_nxt += 2; + } + else + { + if (frm_end-frm_nxt < 4) + break; + uint16_t c2 = static_cast(frm_nxt[3] << 8 | frm_nxt[2]); + if ((c2 & 0xFC00) != 0xDC00) + break; + uint32_t t = static_cast( + ((((c1 & 0x03C0) >> 6) + 1) << 16) + | ((c1 & 0x003F) << 10) + | (c2 & 0x03FF)); + if (t > Maxcode) + break; + frm_nxt += 4; + } + } + return static_cast(frm_nxt - frm); +} + +static +codecvt_base::result +ucs2_to_utf16be(const uint16_t* frm, const uint16_t* frm_end, const uint16_t*& frm_nxt, + uint8_t* to, uint8_t* to_end, uint8_t*& to_nxt, + unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) +{ + frm_nxt = frm; + to_nxt = to; + if (mode & generate_header) + { + if (to_end-to_nxt < 2) + return codecvt_base::partial; + *to_nxt++ = static_cast(0xFE); + *to_nxt++ = static_cast(0xFF); + } + for (; frm_nxt < frm_end; ++frm_nxt) + { + uint16_t wc = *frm_nxt; + if ((wc & 0xF800) == 0xD800 || wc > Maxcode) + return codecvt_base::error; + if (to_end-to_nxt < 2) + return codecvt_base::partial; + *to_nxt++ = static_cast(wc >> 8); + *to_nxt++ = static_cast(wc); + } + return codecvt_base::ok; +} + +static +codecvt_base::result +utf16be_to_ucs2(const uint8_t* frm, const uint8_t* frm_end, const uint8_t*& frm_nxt, + uint16_t* to, uint16_t* to_end, uint16_t*& to_nxt, + unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) +{ + frm_nxt = frm; + to_nxt = to; + if (mode & consume_header) + { + if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFE && frm_nxt[1] == 0xFF) + frm_nxt += 2; + } + for (; frm_nxt < frm_end - 1 && to_nxt < to_end; ++to_nxt) + { + uint16_t c1 = static_cast(frm_nxt[0] << 8 | frm_nxt[1]); + if ((c1 & 0xF800) == 0xD800 || c1 > Maxcode) + return codecvt_base::error; + *to_nxt = c1; + frm_nxt += 2; + } + return frm_nxt < frm_end ? codecvt_base::partial : codecvt_base::ok; +} + +static +int +utf16be_to_ucs2_length(const uint8_t* frm, const uint8_t* frm_end, + size_t mx, unsigned long Maxcode = 0x10FFFF, + codecvt_mode mode = codecvt_mode(0)) +{ + const uint8_t* frm_nxt = frm; + if (mode & consume_header) + { + if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFE && frm_nxt[1] == 0xFF) + frm_nxt += 2; + } + for (size_t nchar16_t = 0; frm_nxt < frm_end - 1 && nchar16_t < mx; ++nchar16_t) + { + uint16_t c1 = static_cast(frm_nxt[0] << 8 | frm_nxt[1]); + if ((c1 & 0xF800) == 0xD800 || c1 > Maxcode) + break; + frm_nxt += 2; + } + return static_cast(frm_nxt - frm); +} + +static +codecvt_base::result +ucs2_to_utf16le(const uint16_t* frm, const uint16_t* frm_end, const uint16_t*& frm_nxt, + uint8_t* to, uint8_t* to_end, uint8_t*& to_nxt, + unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) +{ + frm_nxt = frm; + to_nxt = to; + if (mode & generate_header) + { + if (to_end-to_nxt < 2) + return codecvt_base::partial; + *to_nxt++ = static_cast(0xFF); + *to_nxt++ = static_cast(0xFE); + } + for (; frm_nxt < frm_end; ++frm_nxt) + { + uint16_t wc = *frm_nxt; + if ((wc & 0xF800) == 0xD800 || wc > Maxcode) + return codecvt_base::error; + if (to_end-to_nxt < 2) + return codecvt_base::partial; + *to_nxt++ = static_cast(wc); + *to_nxt++ = static_cast(wc >> 8); + } + return codecvt_base::ok; +} + +static +codecvt_base::result +utf16le_to_ucs2(const uint8_t* frm, const uint8_t* frm_end, const uint8_t*& frm_nxt, + uint16_t* to, uint16_t* to_end, uint16_t*& to_nxt, + unsigned long Maxcode = 0x10FFFF, codecvt_mode mode = codecvt_mode(0)) +{ + frm_nxt = frm; + to_nxt = to; + if (mode & consume_header) + { + if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFF && frm_nxt[1] == 0xFE) + frm_nxt += 2; + } + for (; frm_nxt < frm_end - 1 && to_nxt < to_end; ++to_nxt) + { + uint16_t c1 = static_cast(frm_nxt[1] << 8 | frm_nxt[0]); + if ((c1 & 0xF800) == 0xD800 || c1 > Maxcode) + return codecvt_base::error; + *to_nxt = c1; + frm_nxt += 2; + } + return frm_nxt < frm_end ? codecvt_base::partial : codecvt_base::ok; +} + +static +int +utf16le_to_ucs2_length(const uint8_t* frm, const uint8_t* frm_end, + size_t mx, unsigned long Maxcode = 0x10FFFF, + codecvt_mode mode = codecvt_mode(0)) +{ + const uint8_t* frm_nxt = frm; + frm_nxt = frm; + if (mode & consume_header) + { + if (frm_end-frm_nxt >= 2 && frm_nxt[0] == 0xFF && frm_nxt[1] == 0xFE) + frm_nxt += 2; + } + for (size_t nchar16_t = 0; frm_nxt < frm_end - 1 && nchar16_t < mx; ++nchar16_t) + { + uint16_t c1 = static_cast(frm_nxt[1] << 8 | frm_nxt[0]); + if ((c1 & 0xF800) == 0xD800 || c1 > Maxcode) + break; + frm_nxt += 2; + } + return static_cast(frm_nxt - frm); +} + +// template <> class codecvt + +locale::id codecvt::id; + +codecvt::~codecvt() +{ +} + +codecvt::result +codecvt::do_out(state_type&, + const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + const uint16_t* _frm = reinterpret_cast(frm); + const uint16_t* _frm_end = reinterpret_cast(frm_end); + const uint16_t* _frm_nxt = _frm; + uint8_t* _to = reinterpret_cast(to); + uint8_t* _to_end = reinterpret_cast(to_end); + uint8_t* _to_nxt = _to; + result r = utf16_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +codecvt::result +codecvt::do_in(state_type&, + const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, + intern_type* to, intern_type* to_end, intern_type*& to_nxt) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + const uint8_t* _frm_nxt = _frm; + uint16_t* _to = reinterpret_cast(to); + uint16_t* _to_end = reinterpret_cast(to_end); + uint16_t* _to_nxt = _to; + result r = utf8_to_utf16(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +codecvt::result +codecvt::do_unshift(state_type&, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + to_nxt = to; + return noconv; +} + +int +codecvt::do_encoding() const _NOEXCEPT +{ + return 0; +} + +bool +codecvt::do_always_noconv() const _NOEXCEPT +{ + return false; +} + +int +codecvt::do_length(state_type&, + const extern_type* frm, const extern_type* frm_end, size_t mx) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + return utf8_to_utf16_length(_frm, _frm_end, mx); +} + +int +codecvt::do_max_length() const _NOEXCEPT +{ + return 4; +} + +// template <> class codecvt + +locale::id codecvt::id; + +codecvt::~codecvt() +{ +} + +codecvt::result +codecvt::do_out(state_type&, + const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + const uint32_t* _frm = reinterpret_cast(frm); + const uint32_t* _frm_end = reinterpret_cast(frm_end); + const uint32_t* _frm_nxt = _frm; + uint8_t* _to = reinterpret_cast(to); + uint8_t* _to_end = reinterpret_cast(to_end); + uint8_t* _to_nxt = _to; + result r = ucs4_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +codecvt::result +codecvt::do_in(state_type&, + const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, + intern_type* to, intern_type* to_end, intern_type*& to_nxt) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + const uint8_t* _frm_nxt = _frm; + uint32_t* _to = reinterpret_cast(to); + uint32_t* _to_end = reinterpret_cast(to_end); + uint32_t* _to_nxt = _to; + result r = utf8_to_ucs4(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +codecvt::result +codecvt::do_unshift(state_type&, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + to_nxt = to; + return noconv; +} + +int +codecvt::do_encoding() const _NOEXCEPT +{ + return 0; +} + +bool +codecvt::do_always_noconv() const _NOEXCEPT +{ + return false; +} + +int +codecvt::do_length(state_type&, + const extern_type* frm, const extern_type* frm_end, size_t mx) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + return utf8_to_ucs4_length(_frm, _frm_end, mx); +} + +int +codecvt::do_max_length() const _NOEXCEPT +{ + return 4; +} + +// __codecvt_utf8 + +__codecvt_utf8::result +__codecvt_utf8::do_out(state_type&, + const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + const ucs_char_t* _frm = reinterpret_cast(frm); + const ucs_char_t* _frm_end = reinterpret_cast(frm_end); + const ucs_char_t* _frm_nxt = _frm; + uint8_t* _to = reinterpret_cast(to); + uint8_t* _to_end = reinterpret_cast(to_end); + uint8_t* _to_nxt = _to; +#if SHORT_WCHAR + result r = ucs2_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); +#else + result r = ucs4_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); +#endif + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf8::result +__codecvt_utf8::do_in(state_type&, + const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, + intern_type* to, intern_type* to_end, intern_type*& to_nxt) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + const uint8_t* _frm_nxt = _frm; + ucs_char_t* _to = reinterpret_cast(to); + ucs_char_t* _to_end = reinterpret_cast(to_end); + ucs_char_t* _to_nxt = _to; +#if SHORT_WCHAR + result r = utf8_to_ucs2(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); +#else + result r = utf8_to_ucs4(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); +#endif + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf8::result +__codecvt_utf8::do_unshift(state_type&, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + to_nxt = to; + return noconv; +} + +int +__codecvt_utf8::do_encoding() const _NOEXCEPT +{ + return 0; +} + +bool +__codecvt_utf8::do_always_noconv() const _NOEXCEPT +{ + return false; +} + +int +__codecvt_utf8::do_length(state_type&, + const extern_type* frm, const extern_type* frm_end, size_t mx) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); +#if SHORT_WCHAR + return utf8_to_ucs2_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); +#else + return utf8_to_ucs4_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); +#endif +} + +int +__codecvt_utf8::do_max_length() const _NOEXCEPT +{ + if (_Mode_ & consume_header) + return 7; + return 4; +} + +// __codecvt_utf8 + +__codecvt_utf8::result +__codecvt_utf8::do_out(state_type&, + const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + const uint16_t* _frm = reinterpret_cast(frm); + const uint16_t* _frm_end = reinterpret_cast(frm_end); + const uint16_t* _frm_nxt = _frm; + uint8_t* _to = reinterpret_cast(to); + uint8_t* _to_end = reinterpret_cast(to_end); + uint8_t* _to_nxt = _to; + result r = ucs2_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf8::result +__codecvt_utf8::do_in(state_type&, + const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, + intern_type* to, intern_type* to_end, intern_type*& to_nxt) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + const uint8_t* _frm_nxt = _frm; + uint16_t* _to = reinterpret_cast(to); + uint16_t* _to_end = reinterpret_cast(to_end); + uint16_t* _to_nxt = _to; + result r = utf8_to_ucs2(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf8::result +__codecvt_utf8::do_unshift(state_type&, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + to_nxt = to; + return noconv; +} + +int +__codecvt_utf8::do_encoding() const _NOEXCEPT +{ + return 0; +} + +bool +__codecvt_utf8::do_always_noconv() const _NOEXCEPT +{ + return false; +} + +int +__codecvt_utf8::do_length(state_type&, + const extern_type* frm, const extern_type* frm_end, size_t mx) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + return utf8_to_ucs2_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); +} + +int +__codecvt_utf8::do_max_length() const _NOEXCEPT +{ + if (_Mode_ & consume_header) + return 6; + return 3; +} + +// __codecvt_utf8 + +__codecvt_utf8::result +__codecvt_utf8::do_out(state_type&, + const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + const uint32_t* _frm = reinterpret_cast(frm); + const uint32_t* _frm_end = reinterpret_cast(frm_end); + const uint32_t* _frm_nxt = _frm; + uint8_t* _to = reinterpret_cast(to); + uint8_t* _to_end = reinterpret_cast(to_end); + uint8_t* _to_nxt = _to; + result r = ucs4_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf8::result +__codecvt_utf8::do_in(state_type&, + const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, + intern_type* to, intern_type* to_end, intern_type*& to_nxt) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + const uint8_t* _frm_nxt = _frm; + uint32_t* _to = reinterpret_cast(to); + uint32_t* _to_end = reinterpret_cast(to_end); + uint32_t* _to_nxt = _to; + result r = utf8_to_ucs4(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf8::result +__codecvt_utf8::do_unshift(state_type&, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + to_nxt = to; + return noconv; +} + +int +__codecvt_utf8::do_encoding() const _NOEXCEPT +{ + return 0; +} + +bool +__codecvt_utf8::do_always_noconv() const _NOEXCEPT +{ + return false; +} + +int +__codecvt_utf8::do_length(state_type&, + const extern_type* frm, const extern_type* frm_end, size_t mx) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + return utf8_to_ucs4_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); +} + +int +__codecvt_utf8::do_max_length() const _NOEXCEPT +{ + if (_Mode_ & consume_header) + return 7; + return 4; +} + +// __codecvt_utf16 + +__codecvt_utf16::result +__codecvt_utf16::do_out(state_type&, + const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + const ucs_char_t* _frm = reinterpret_cast(frm); + const ucs_char_t* _frm_end = reinterpret_cast(frm_end); + const ucs_char_t* _frm_nxt = _frm; + uint8_t* _to = reinterpret_cast(to); + uint8_t* _to_end = reinterpret_cast(to_end); + uint8_t* _to_nxt = _to; +#if SHORT_WCHAR + result r = ucs2_to_utf16be(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); +#else + result r = ucs4_to_utf16be(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); +#endif + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf16::result +__codecvt_utf16::do_in(state_type&, + const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, + intern_type* to, intern_type* to_end, intern_type*& to_nxt) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + const uint8_t* _frm_nxt = _frm; + ucs_char_t* _to = reinterpret_cast(to); + ucs_char_t* _to_end = reinterpret_cast(to_end); + ucs_char_t* _to_nxt = _to; +#if SHORT_WCHAR + result r = utf16be_to_ucs2(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); +#else + result r = utf16be_to_ucs4(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); +#endif + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf16::result +__codecvt_utf16::do_unshift(state_type&, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + to_nxt = to; + return noconv; +} + +int +__codecvt_utf16::do_encoding() const _NOEXCEPT +{ + return 0; +} + +bool +__codecvt_utf16::do_always_noconv() const _NOEXCEPT +{ + return false; +} + +int +__codecvt_utf16::do_length(state_type&, + const extern_type* frm, const extern_type* frm_end, size_t mx) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); +#if SHORT_WCHAR + return utf16be_to_ucs2_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); +#else + return utf16be_to_ucs4_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); +#endif +} + +int +__codecvt_utf16::do_max_length() const _NOEXCEPT +{ + if (_Mode_ & consume_header) + return 6; + return 4; +} + +// __codecvt_utf16 + +__codecvt_utf16::result +__codecvt_utf16::do_out(state_type&, + const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + const ucs_char_t* _frm = reinterpret_cast(frm); + const ucs_char_t* _frm_end = reinterpret_cast(frm_end); + const ucs_char_t* _frm_nxt = _frm; + uint8_t* _to = reinterpret_cast(to); + uint8_t* _to_end = reinterpret_cast(to_end); + uint8_t* _to_nxt = _to; +#if SHORT_WCHAR + result r = ucs2_to_utf16le(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); +#else + result r = ucs4_to_utf16le(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); +#endif + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf16::result +__codecvt_utf16::do_in(state_type&, + const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, + intern_type* to, intern_type* to_end, intern_type*& to_nxt) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + const uint8_t* _frm_nxt = _frm; + ucs_char_t* _to = reinterpret_cast(to); + ucs_char_t* _to_end = reinterpret_cast(to_end); + ucs_char_t* _to_nxt = _to; +#if SHORT_WCHAR + result r = utf16le_to_ucs2(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); +#else + result r = utf16le_to_ucs4(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); +#endif + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf16::result +__codecvt_utf16::do_unshift(state_type&, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + to_nxt = to; + return noconv; +} + +int +__codecvt_utf16::do_encoding() const _NOEXCEPT +{ + return 0; +} + +bool +__codecvt_utf16::do_always_noconv() const _NOEXCEPT +{ + return false; +} + +int +__codecvt_utf16::do_length(state_type&, + const extern_type* frm, const extern_type* frm_end, size_t mx) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); +#if SHORT_WCHAR + return utf16le_to_ucs2_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); +#else + return utf16le_to_ucs4_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); +#endif +} + +int +__codecvt_utf16::do_max_length() const _NOEXCEPT +{ + if (_Mode_ & consume_header) + return 6; + return 4; +} + +// __codecvt_utf16 + +__codecvt_utf16::result +__codecvt_utf16::do_out(state_type&, + const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + const uint16_t* _frm = reinterpret_cast(frm); + const uint16_t* _frm_end = reinterpret_cast(frm_end); + const uint16_t* _frm_nxt = _frm; + uint8_t* _to = reinterpret_cast(to); + uint8_t* _to_end = reinterpret_cast(to_end); + uint8_t* _to_nxt = _to; + result r = ucs2_to_utf16be(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf16::result +__codecvt_utf16::do_in(state_type&, + const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, + intern_type* to, intern_type* to_end, intern_type*& to_nxt) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + const uint8_t* _frm_nxt = _frm; + uint16_t* _to = reinterpret_cast(to); + uint16_t* _to_end = reinterpret_cast(to_end); + uint16_t* _to_nxt = _to; + result r = utf16be_to_ucs2(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf16::result +__codecvt_utf16::do_unshift(state_type&, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + to_nxt = to; + return noconv; +} + +int +__codecvt_utf16::do_encoding() const _NOEXCEPT +{ + return 0; +} + +bool +__codecvt_utf16::do_always_noconv() const _NOEXCEPT +{ + return false; +} + +int +__codecvt_utf16::do_length(state_type&, + const extern_type* frm, const extern_type* frm_end, size_t mx) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + return utf16be_to_ucs2_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); +} + +int +__codecvt_utf16::do_max_length() const _NOEXCEPT +{ + if (_Mode_ & consume_header) + return 4; + return 2; +} + +// __codecvt_utf16 + +__codecvt_utf16::result +__codecvt_utf16::do_out(state_type&, + const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + const uint16_t* _frm = reinterpret_cast(frm); + const uint16_t* _frm_end = reinterpret_cast(frm_end); + const uint16_t* _frm_nxt = _frm; + uint8_t* _to = reinterpret_cast(to); + uint8_t* _to_end = reinterpret_cast(to_end); + uint8_t* _to_nxt = _to; + result r = ucs2_to_utf16le(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf16::result +__codecvt_utf16::do_in(state_type&, + const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, + intern_type* to, intern_type* to_end, intern_type*& to_nxt) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + const uint8_t* _frm_nxt = _frm; + uint16_t* _to = reinterpret_cast(to); + uint16_t* _to_end = reinterpret_cast(to_end); + uint16_t* _to_nxt = _to; + result r = utf16le_to_ucs2(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf16::result +__codecvt_utf16::do_unshift(state_type&, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + to_nxt = to; + return noconv; +} + +int +__codecvt_utf16::do_encoding() const _NOEXCEPT +{ + return 0; +} + +bool +__codecvt_utf16::do_always_noconv() const _NOEXCEPT +{ + return false; +} + +int +__codecvt_utf16::do_length(state_type&, + const extern_type* frm, const extern_type* frm_end, size_t mx) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + return utf16le_to_ucs2_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); +} + +int +__codecvt_utf16::do_max_length() const _NOEXCEPT +{ + if (_Mode_ & consume_header) + return 4; + return 2; +} + +// __codecvt_utf16 + +__codecvt_utf16::result +__codecvt_utf16::do_out(state_type&, + const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + const uint32_t* _frm = reinterpret_cast(frm); + const uint32_t* _frm_end = reinterpret_cast(frm_end); + const uint32_t* _frm_nxt = _frm; + uint8_t* _to = reinterpret_cast(to); + uint8_t* _to_end = reinterpret_cast(to_end); + uint8_t* _to_nxt = _to; + result r = ucs4_to_utf16be(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf16::result +__codecvt_utf16::do_in(state_type&, + const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, + intern_type* to, intern_type* to_end, intern_type*& to_nxt) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + const uint8_t* _frm_nxt = _frm; + uint32_t* _to = reinterpret_cast(to); + uint32_t* _to_end = reinterpret_cast(to_end); + uint32_t* _to_nxt = _to; + result r = utf16be_to_ucs4(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf16::result +__codecvt_utf16::do_unshift(state_type&, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + to_nxt = to; + return noconv; +} + +int +__codecvt_utf16::do_encoding() const _NOEXCEPT +{ + return 0; +} + +bool +__codecvt_utf16::do_always_noconv() const _NOEXCEPT +{ + return false; +} + +int +__codecvt_utf16::do_length(state_type&, + const extern_type* frm, const extern_type* frm_end, size_t mx) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + return utf16be_to_ucs4_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); +} + +int +__codecvt_utf16::do_max_length() const _NOEXCEPT +{ + if (_Mode_ & consume_header) + return 6; + return 4; +} + +// __codecvt_utf16 + +__codecvt_utf16::result +__codecvt_utf16::do_out(state_type&, + const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + const uint32_t* _frm = reinterpret_cast(frm); + const uint32_t* _frm_end = reinterpret_cast(frm_end); + const uint32_t* _frm_nxt = _frm; + uint8_t* _to = reinterpret_cast(to); + uint8_t* _to_end = reinterpret_cast(to_end); + uint8_t* _to_nxt = _to; + result r = ucs4_to_utf16le(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf16::result +__codecvt_utf16::do_in(state_type&, + const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, + intern_type* to, intern_type* to_end, intern_type*& to_nxt) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + const uint8_t* _frm_nxt = _frm; + uint32_t* _to = reinterpret_cast(to); + uint32_t* _to_end = reinterpret_cast(to_end); + uint32_t* _to_nxt = _to; + result r = utf16le_to_ucs4(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf16::result +__codecvt_utf16::do_unshift(state_type&, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + to_nxt = to; + return noconv; +} + +int +__codecvt_utf16::do_encoding() const _NOEXCEPT +{ + return 0; +} + +bool +__codecvt_utf16::do_always_noconv() const _NOEXCEPT +{ + return false; +} + +int +__codecvt_utf16::do_length(state_type&, + const extern_type* frm, const extern_type* frm_end, size_t mx) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + return utf16le_to_ucs4_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); +} + +int +__codecvt_utf16::do_max_length() const _NOEXCEPT +{ + if (_Mode_ & consume_header) + return 6; + return 4; +} + +// __codecvt_utf8_utf16 + +__codecvt_utf8_utf16::result +__codecvt_utf8_utf16::do_out(state_type&, + const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + const ucs_char_t* _frm = reinterpret_cast(frm); + const ucs_char_t* _frm_end = reinterpret_cast(frm_end); + const ucs_char_t* _frm_nxt = _frm; + uint8_t* _to = reinterpret_cast(to); + uint8_t* _to_end = reinterpret_cast(to_end); + uint8_t* _to_nxt = _to; + result r = utf16_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf8_utf16::result +__codecvt_utf8_utf16::do_in(state_type&, + const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, + intern_type* to, intern_type* to_end, intern_type*& to_nxt) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + const uint8_t* _frm_nxt = _frm; + ucs_char_t* _to = reinterpret_cast(to); + ucs_char_t* _to_end = reinterpret_cast(to_end); + ucs_char_t* _to_nxt = _to; + result r = utf8_to_utf16(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf8_utf16::result +__codecvt_utf8_utf16::do_unshift(state_type&, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + to_nxt = to; + return noconv; +} + +int +__codecvt_utf8_utf16::do_encoding() const _NOEXCEPT +{ + return 0; +} + +bool +__codecvt_utf8_utf16::do_always_noconv() const _NOEXCEPT +{ + return false; +} + +int +__codecvt_utf8_utf16::do_length(state_type&, + const extern_type* frm, const extern_type* frm_end, size_t mx) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + return utf8_to_utf16_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); +} + +int +__codecvt_utf8_utf16::do_max_length() const _NOEXCEPT +{ + if (_Mode_ & consume_header) + return 7; + return 4; +} + +// __codecvt_utf8_utf16 + +__codecvt_utf8_utf16::result +__codecvt_utf8_utf16::do_out(state_type&, + const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + const uint16_t* _frm = reinterpret_cast(frm); + const uint16_t* _frm_end = reinterpret_cast(frm_end); + const uint16_t* _frm_nxt = _frm; + uint8_t* _to = reinterpret_cast(to); + uint8_t* _to_end = reinterpret_cast(to_end); + uint8_t* _to_nxt = _to; + result r = utf16_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf8_utf16::result +__codecvt_utf8_utf16::do_in(state_type&, + const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, + intern_type* to, intern_type* to_end, intern_type*& to_nxt) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + const uint8_t* _frm_nxt = _frm; + uint16_t* _to = reinterpret_cast(to); + uint16_t* _to_end = reinterpret_cast(to_end); + uint16_t* _to_nxt = _to; + result r = utf8_to_utf16(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf8_utf16::result +__codecvt_utf8_utf16::do_unshift(state_type&, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + to_nxt = to; + return noconv; +} + +int +__codecvt_utf8_utf16::do_encoding() const _NOEXCEPT +{ + return 0; +} + +bool +__codecvt_utf8_utf16::do_always_noconv() const _NOEXCEPT +{ + return false; +} + +int +__codecvt_utf8_utf16::do_length(state_type&, + const extern_type* frm, const extern_type* frm_end, size_t mx) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + return utf8_to_utf16_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); +} + +int +__codecvt_utf8_utf16::do_max_length() const _NOEXCEPT +{ + if (_Mode_ & consume_header) + return 7; + return 4; +} + +// __codecvt_utf8_utf16 + +__codecvt_utf8_utf16::result +__codecvt_utf8_utf16::do_out(state_type&, + const intern_type* frm, const intern_type* frm_end, const intern_type*& frm_nxt, + extern_type* to, extern_type* to_end, extern_type*& to_nxt) const +{ + const uint32_t* _frm = reinterpret_cast(frm); + const uint32_t* _frm_end = reinterpret_cast(frm_end); + const uint32_t* _frm_nxt = _frm; + uint8_t* _to = reinterpret_cast(to); + uint8_t* _to_end = reinterpret_cast(to_end); + uint8_t* _to_nxt = _to; + result r = utf16_to_utf8(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf8_utf16::result +__codecvt_utf8_utf16::do_in(state_type&, + const extern_type* frm, const extern_type* frm_end, const extern_type*& frm_nxt, + intern_type* to, intern_type* to_end, intern_type*& to_nxt) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + const uint8_t* _frm_nxt = _frm; + uint32_t* _to = reinterpret_cast(to); + uint32_t* _to_end = reinterpret_cast(to_end); + uint32_t* _to_nxt = _to; + result r = utf8_to_utf16(_frm, _frm_end, _frm_nxt, _to, _to_end, _to_nxt, + _Maxcode_, _Mode_); + frm_nxt = frm + (_frm_nxt - _frm); + to_nxt = to + (_to_nxt - _to); + return r; +} + +__codecvt_utf8_utf16::result +__codecvt_utf8_utf16::do_unshift(state_type&, + extern_type* to, extern_type*, extern_type*& to_nxt) const +{ + to_nxt = to; + return noconv; +} + +int +__codecvt_utf8_utf16::do_encoding() const _NOEXCEPT +{ + return 0; +} + +bool +__codecvt_utf8_utf16::do_always_noconv() const _NOEXCEPT +{ + return false; +} + +int +__codecvt_utf8_utf16::do_length(state_type&, + const extern_type* frm, const extern_type* frm_end, size_t mx) const +{ + const uint8_t* _frm = reinterpret_cast(frm); + const uint8_t* _frm_end = reinterpret_cast(frm_end); + return utf8_to_utf16_length(_frm, _frm_end, mx, _Maxcode_, _Mode_); +} + +int +__codecvt_utf8_utf16::do_max_length() const _NOEXCEPT +{ + if (_Mode_ & consume_header) + return 7; + return 4; +} + +// __narrow_to_utf8<16> + +__narrow_to_utf8<16>::~__narrow_to_utf8() +{ +} + +// __narrow_to_utf8<32> + +__narrow_to_utf8<32>::~__narrow_to_utf8() +{ +} + +// __widen_from_utf8<16> + +__widen_from_utf8<16>::~__widen_from_utf8() +{ +} + +// __widen_from_utf8<32> + +__widen_from_utf8<32>::~__widen_from_utf8() +{ +} + +// numpunct && numpunct + +locale::id numpunct< char >::id; +locale::id numpunct::id; + +numpunct::numpunct(size_t refs) + : locale::facet(refs), + __decimal_point_('.'), + __thousands_sep_(',') +{ +} + +numpunct::numpunct(size_t refs) + : locale::facet(refs), + __decimal_point_(L'.'), + __thousands_sep_(L',') +{ +} + +numpunct::~numpunct() +{ +} + +numpunct::~numpunct() +{ +} + + char numpunct< char >::do_decimal_point() const {return __decimal_point_;} +wchar_t numpunct::do_decimal_point() const {return __decimal_point_;} + + char numpunct< char >::do_thousands_sep() const {return __thousands_sep_;} +wchar_t numpunct::do_thousands_sep() const {return __thousands_sep_;} + +string numpunct< char >::do_grouping() const {return __grouping_;} +string numpunct::do_grouping() const {return __grouping_;} + + string numpunct< char >::do_truename() const {return "true";} +wstring numpunct::do_truename() const {return L"true";} + + string numpunct< char >::do_falsename() const {return "false";} +wstring numpunct::do_falsename() const {return L"false";} + +// numpunct_byname + +numpunct_byname::numpunct_byname(const char* nm, size_t refs) + : numpunct(refs) +{ + __init(nm); +} + +numpunct_byname::numpunct_byname(const string& nm, size_t refs) + : numpunct(refs) +{ + __init(nm.c_str()); +} + +numpunct_byname::~numpunct_byname() +{ +} + +void +numpunct_byname::__init(const char* nm) +{ + if (strcmp(nm, "C") != 0) + { + __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); +#ifndef _LIBCPP_NO_EXCEPTIONS + if (loc == nullptr) + throw runtime_error("numpunct_byname::numpunct_byname" + " failed to construct for " + string(nm)); +#endif // _LIBCPP_NO_EXCEPTIONS +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + lconv* lc = localeconv_l(loc.get()); +#else + lconv* lc = __localeconv_l(loc.get()); +#endif + if (*lc->decimal_point) + __decimal_point_ = *lc->decimal_point; + if (*lc->thousands_sep) + __thousands_sep_ = *lc->thousands_sep; + __grouping_ = lc->grouping; + // localization for truename and falsename is not available + } +} + +// numpunct_byname + +numpunct_byname::numpunct_byname(const char* nm, size_t refs) + : numpunct(refs) +{ + __init(nm); +} + +numpunct_byname::numpunct_byname(const string& nm, size_t refs) + : numpunct(refs) +{ + __init(nm.c_str()); +} + +numpunct_byname::~numpunct_byname() +{ +} + +void +numpunct_byname::__init(const char* nm) +{ + if (strcmp(nm, "C") != 0) + { + __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); +#ifndef _LIBCPP_NO_EXCEPTIONS + if (loc == nullptr) + throw runtime_error("numpunct_byname::numpunct_byname" + " failed to construct for " + string(nm)); +#endif // _LIBCPP_NO_EXCEPTIONS +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + lconv* lc = localeconv_l(loc.get()); +#else + lconv* lc = __localeconv_l(loc.get()); +#endif + if (*lc->decimal_point) + __decimal_point_ = *lc->decimal_point; + if (*lc->thousands_sep) + __thousands_sep_ = *lc->thousands_sep; + __grouping_ = lc->grouping; + // locallization for truename and falsename is not available + } +} + +// num_get helpers + +int +__num_get_base::__get_base(ios_base& iob) +{ + ios_base::fmtflags __basefield = iob.flags() & ios_base::basefield; + if (__basefield == ios_base::oct) + return 8; + else if (__basefield == ios_base::hex) + return 16; + else if (__basefield == 0) + return 0; + return 10; +} + +const char __num_get_base::__src[33] = "0123456789abcdefABCDEFxX+-pPiInN"; + +void +__check_grouping(const string& __grouping, unsigned* __g, unsigned* __g_end, + ios_base::iostate& __err) +{ + if (__grouping.size() != 0) + { + reverse(__g, __g_end); + const char* __ig = __grouping.data(); + const char* __eg = __ig + __grouping.size(); + for (unsigned* __r = __g; __r < __g_end-1; ++__r) + { + if (0 < *__ig && *__ig < numeric_limits::max()) + { + if (static_cast(*__ig) != *__r) + { + __err = ios_base::failbit; + return; + } + } + if (__eg - __ig > 1) + ++__ig; + } + if (0 < *__ig && *__ig < numeric_limits::max()) + { + if (static_cast(*__ig) < __g_end[-1] || __g_end[-1] == 0) + __err = ios_base::failbit; + } + } +} + +void +__num_put_base::__format_int(char* __fmtp, const char* __len, bool __signd, + ios_base::fmtflags __flags) +{ + if (__flags & ios_base::showpos) + *__fmtp++ = '+'; + if (__flags & ios_base::showbase) + *__fmtp++ = '#'; + while(*__len) + *__fmtp++ = *__len++; + if ((__flags & ios_base::basefield) == ios_base::oct) + *__fmtp = 'o'; + else if ((__flags & ios_base::basefield) == ios_base::hex) + { + if (__flags & ios_base::uppercase) + *__fmtp = 'X'; + else + *__fmtp = 'x'; + } + else if (__signd) + *__fmtp = 'd'; + else + *__fmtp = 'u'; +} + +bool +__num_put_base::__format_float(char* __fmtp, const char* __len, + ios_base::fmtflags __flags) +{ + bool specify_precision = true; + if (__flags & ios_base::showpos) + *__fmtp++ = '+'; + if (__flags & ios_base::showpoint) + *__fmtp++ = '#'; + ios_base::fmtflags floatfield = __flags & ios_base::floatfield; + bool uppercase = (__flags & ios_base::uppercase) != 0; + if (floatfield == (ios_base::fixed | ios_base::scientific)) + specify_precision = false; + else + { + *__fmtp++ = '.'; + *__fmtp++ = '*'; + } + while(*__len) + *__fmtp++ = *__len++; + if (floatfield == ios_base::fixed) + { + if (uppercase) + *__fmtp = 'F'; + else + *__fmtp = 'f'; + } + else if (floatfield == ios_base::scientific) + { + if (uppercase) + *__fmtp = 'E'; + else + *__fmtp = 'e'; + } + else if (floatfield == (ios_base::fixed | ios_base::scientific)) + { + if (uppercase) + *__fmtp = 'A'; + else + *__fmtp = 'a'; + } + else + { + if (uppercase) + *__fmtp = 'G'; + else + *__fmtp = 'g'; + } + return specify_precision; +} + +char* +__num_put_base::__identify_padding(char* __nb, char* __ne, + const ios_base& __iob) +{ + switch (__iob.flags() & ios_base::adjustfield) + { + case ios_base::internal: + if (__nb[0] == '-' || __nb[0] == '+') + return __nb+1; + if (__ne - __nb >= 2 && __nb[0] == '0' + && (__nb[1] == 'x' || __nb[1] == 'X')) + return __nb+2; + break; + case ios_base::left: + return __ne; + case ios_base::right: + default: + break; + } + return __nb; +} + +// time_get + +static +string* +init_weeks() +{ + static string weeks[14]; + weeks[0] = "Sunday"; + weeks[1] = "Monday"; + weeks[2] = "Tuesday"; + weeks[3] = "Wednesday"; + weeks[4] = "Thursday"; + weeks[5] = "Friday"; + weeks[6] = "Saturday"; + weeks[7] = "Sun"; + weeks[8] = "Mon"; + weeks[9] = "Tue"; + weeks[10] = "Wed"; + weeks[11] = "Thu"; + weeks[12] = "Fri"; + weeks[13] = "Sat"; + return weeks; +} + +static +wstring* +init_wweeks() +{ + static wstring weeks[14]; + weeks[0] = L"Sunday"; + weeks[1] = L"Monday"; + weeks[2] = L"Tuesday"; + weeks[3] = L"Wednesday"; + weeks[4] = L"Thursday"; + weeks[5] = L"Friday"; + weeks[6] = L"Saturday"; + weeks[7] = L"Sun"; + weeks[8] = L"Mon"; + weeks[9] = L"Tue"; + weeks[10] = L"Wed"; + weeks[11] = L"Thu"; + weeks[12] = L"Fri"; + weeks[13] = L"Sat"; + return weeks; +} + +template <> +const string* +__time_get_c_storage::__weeks() const +{ + static const string* weeks = init_weeks(); + return weeks; +} + +template <> +const wstring* +__time_get_c_storage::__weeks() const +{ + static const wstring* weeks = init_wweeks(); + return weeks; +} + +static +string* +init_months() +{ + static string months[24]; + months[0] = "January"; + months[1] = "February"; + months[2] = "March"; + months[3] = "April"; + months[4] = "May"; + months[5] = "June"; + months[6] = "July"; + months[7] = "August"; + months[8] = "September"; + months[9] = "October"; + months[10] = "November"; + months[11] = "December"; + months[12] = "Jan"; + months[13] = "Feb"; + months[14] = "Mar"; + months[15] = "Apr"; + months[16] = "May"; + months[17] = "Jun"; + months[18] = "Jul"; + months[19] = "Aug"; + months[20] = "Sep"; + months[21] = "Oct"; + months[22] = "Nov"; + months[23] = "Dec"; + return months; +} + +static +wstring* +init_wmonths() +{ + static wstring months[24]; + months[0] = L"January"; + months[1] = L"February"; + months[2] = L"March"; + months[3] = L"April"; + months[4] = L"May"; + months[5] = L"June"; + months[6] = L"July"; + months[7] = L"August"; + months[8] = L"September"; + months[9] = L"October"; + months[10] = L"November"; + months[11] = L"December"; + months[12] = L"Jan"; + months[13] = L"Feb"; + months[14] = L"Mar"; + months[15] = L"Apr"; + months[16] = L"May"; + months[17] = L"Jun"; + months[18] = L"Jul"; + months[19] = L"Aug"; + months[20] = L"Sep"; + months[21] = L"Oct"; + months[22] = L"Nov"; + months[23] = L"Dec"; + return months; +} + +template <> +const string* +__time_get_c_storage::__months() const +{ + static const string* months = init_months(); + return months; +} + +template <> +const wstring* +__time_get_c_storage::__months() const +{ + static const wstring* months = init_wmonths(); + return months; +} + +static +string* +init_am_pm() +{ + static string am_pm[24]; + am_pm[0] = "AM"; + am_pm[1] = "PM"; + return am_pm; +} + +static +wstring* +init_wam_pm() +{ + static wstring am_pm[24]; + am_pm[0] = L"AM"; + am_pm[1] = L"PM"; + return am_pm; +} + +template <> +const string* +__time_get_c_storage::__am_pm() const +{ + static const string* am_pm = init_am_pm(); + return am_pm; +} + +template <> +const wstring* +__time_get_c_storage::__am_pm() const +{ + static const wstring* am_pm = init_wam_pm(); + return am_pm; +} + +template <> +const string& +__time_get_c_storage::__x() const +{ + static string s("%m/%d/%y"); + return s; +} + +template <> +const wstring& +__time_get_c_storage::__x() const +{ + static wstring s(L"%m/%d/%y"); + return s; +} + +template <> +const string& +__time_get_c_storage::__X() const +{ + static string s("%H:%M:%S"); + return s; +} + +template <> +const wstring& +__time_get_c_storage::__X() const +{ + static wstring s(L"%H:%M:%S"); + return s; +} + +template <> +const string& +__time_get_c_storage::__c() const +{ + static string s("%a %b %d %H:%M:%S %Y"); + return s; +} + +template <> +const wstring& +__time_get_c_storage::__c() const +{ + static wstring s(L"%a %b %d %H:%M:%S %Y"); + return s; +} + +template <> +const string& +__time_get_c_storage::__r() const +{ + static string s("%I:%M:%S %p"); + return s; +} + +template <> +const wstring& +__time_get_c_storage::__r() const +{ + static wstring s(L"%I:%M:%S %p"); + return s; +} + +// time_get_byname + +__time_get::__time_get(const char* nm) + : __loc_(newlocale(LC_ALL_MASK, nm, 0)) +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + if (__loc_ == 0) + throw runtime_error("time_get_byname" + " failed to construct for " + string(nm)); +#endif // _LIBCPP_NO_EXCEPTIONS +} + +__time_get::__time_get(const string& nm) + : __loc_(newlocale(LC_ALL_MASK, nm.c_str(), 0)) +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + if (__loc_ == 0) + throw runtime_error("time_get_byname" + " failed to construct for " + nm); +#endif // _LIBCPP_NO_EXCEPTIONS +} + +__time_get::~__time_get() +{ + freelocale(__loc_); +} +#if defined(__clang__) +#pragma clang diagnostic ignored "-Wmissing-field-initializers" +#endif +#if defined(__GNUG__) +#pragma GCC diagnostic ignored "-Wmissing-field-initializers" +#endif + +template <> +string +__time_get_storage::__analyze(char fmt, const ctype& ct) +{ + tm t = {0}; + t.tm_sec = 59; + t.tm_min = 55; + t.tm_hour = 23; + t.tm_mday = 31; + t.tm_mon = 11; + t.tm_year = 161; + t.tm_wday = 6; + t.tm_yday = 364; + t.tm_isdst = -1; + char buf[100]; + char f[3] = {0}; + f[0] = '%'; + f[1] = fmt; + size_t n = strftime_l(buf, countof(buf), f, &t, __loc_); + char* bb = buf; + char* be = buf + n; + string result; + while (bb != be) + { + if (ct.is(ctype_base::space, *bb)) + { + result.push_back(' '); + for (++bb; bb != be && ct.is(ctype_base::space, *bb); ++bb) + ; + continue; + } + char* w = bb; + ios_base::iostate err = ios_base::goodbit; + ptrdiff_t i = __scan_keyword(w, be, this->__weeks_, this->__weeks_+14, + ct, err, false) + - this->__weeks_; + if (i < 14) + { + result.push_back('%'); + if (i < 7) + result.push_back('A'); + else + result.push_back('a'); + bb = w; + continue; + } + w = bb; + i = __scan_keyword(w, be, this->__months_, this->__months_+24, + ct, err, false) + - this->__months_; + if (i < 24) + { + result.push_back('%'); + if (i < 12) + result.push_back('B'); + else + result.push_back('b'); + if (fmt == 'x' && ct.is(ctype_base::digit, this->__months_[i][0])) + result.back() = 'm'; + bb = w; + continue; + } + if (this->__am_pm_[0].size() + this->__am_pm_[1].size() > 0) + { + w = bb; + i = __scan_keyword(w, be, this->__am_pm_, this->__am_pm_+2, + ct, err, false) - this->__am_pm_; + if (i < 2) + { + result.push_back('%'); + result.push_back('p'); + bb = w; + continue; + } + } + w = bb; + if (ct.is(ctype_base::digit, *bb)) + { + switch(__get_up_to_n_digits(bb, be, err, ct, 4)) + { + case 6: + result.push_back('%'); + result.push_back('w'); + break; + case 7: + result.push_back('%'); + result.push_back('u'); + break; + case 11: + result.push_back('%'); + result.push_back('I'); + break; + case 12: + result.push_back('%'); + result.push_back('m'); + break; + case 23: + result.push_back('%'); + result.push_back('H'); + break; + case 31: + result.push_back('%'); + result.push_back('d'); + break; + case 55: + result.push_back('%'); + result.push_back('M'); + break; + case 59: + result.push_back('%'); + result.push_back('S'); + break; + case 61: + result.push_back('%'); + result.push_back('y'); + break; + case 364: + result.push_back('%'); + result.push_back('j'); + break; + case 2061: + result.push_back('%'); + result.push_back('Y'); + break; + default: + for (; w != bb; ++w) + result.push_back(*w); + break; + } + continue; + } + if (*bb == '%') + { + result.push_back('%'); + result.push_back('%'); + ++bb; + continue; + } + result.push_back(*bb); + ++bb; + } + return result; +} + +#if defined(__clang__) +#pragma clang diagnostic ignored "-Wmissing-braces" +#endif + +template <> +wstring +__time_get_storage::__analyze(char fmt, const ctype& ct) +{ + tm t = {0}; + t.tm_sec = 59; + t.tm_min = 55; + t.tm_hour = 23; + t.tm_mday = 31; + t.tm_mon = 11; + t.tm_year = 161; + t.tm_wday = 6; + t.tm_yday = 364; + t.tm_isdst = -1; + char buf[100]; + char f[3] = {0}; + f[0] = '%'; + f[1] = fmt; + strftime_l(buf, countof(buf), f, &t, __loc_); + wchar_t wbuf[100]; + wchar_t* wbb = wbuf; + mbstate_t mb = {0}; + const char* bb = buf; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + size_t j = mbsrtowcs_l( wbb, &bb, countof(wbuf), &mb, __loc_); +#else + size_t j = __mbsrtowcs_l( wbb, &bb, countof(wbuf), &mb, __loc_); +#endif + if (j == size_t(-1)) + __throw_runtime_error("locale not supported"); + wchar_t* wbe = wbb + j; + wstring result; + while (wbb != wbe) + { + if (ct.is(ctype_base::space, *wbb)) + { + result.push_back(L' '); + for (++wbb; wbb != wbe && ct.is(ctype_base::space, *wbb); ++wbb) + ; + continue; + } + wchar_t* w = wbb; + ios_base::iostate err = ios_base::goodbit; + ptrdiff_t i = __scan_keyword(w, wbe, this->__weeks_, this->__weeks_+14, + ct, err, false) + - this->__weeks_; + if (i < 14) + { + result.push_back(L'%'); + if (i < 7) + result.push_back(L'A'); + else + result.push_back(L'a'); + wbb = w; + continue; + } + w = wbb; + i = __scan_keyword(w, wbe, this->__months_, this->__months_+24, + ct, err, false) + - this->__months_; + if (i < 24) + { + result.push_back(L'%'); + if (i < 12) + result.push_back(L'B'); + else + result.push_back(L'b'); + if (fmt == 'x' && ct.is(ctype_base::digit, this->__months_[i][0])) + result.back() = L'm'; + wbb = w; + continue; + } + if (this->__am_pm_[0].size() + this->__am_pm_[1].size() > 0) + { + w = wbb; + i = __scan_keyword(w, wbe, this->__am_pm_, this->__am_pm_+2, + ct, err, false) - this->__am_pm_; + if (i < 2) + { + result.push_back(L'%'); + result.push_back(L'p'); + wbb = w; + continue; + } + } + w = wbb; + if (ct.is(ctype_base::digit, *wbb)) + { + switch(__get_up_to_n_digits(wbb, wbe, err, ct, 4)) + { + case 6: + result.push_back(L'%'); + result.push_back(L'w'); + break; + case 7: + result.push_back(L'%'); + result.push_back(L'u'); + break; + case 11: + result.push_back(L'%'); + result.push_back(L'I'); + break; + case 12: + result.push_back(L'%'); + result.push_back(L'm'); + break; + case 23: + result.push_back(L'%'); + result.push_back(L'H'); + break; + case 31: + result.push_back(L'%'); + result.push_back(L'd'); + break; + case 55: + result.push_back(L'%'); + result.push_back(L'M'); + break; + case 59: + result.push_back(L'%'); + result.push_back(L'S'); + break; + case 61: + result.push_back(L'%'); + result.push_back(L'y'); + break; + case 364: + result.push_back(L'%'); + result.push_back(L'j'); + break; + case 2061: + result.push_back(L'%'); + result.push_back(L'Y'); + break; + default: + for (; w != wbb; ++w) + result.push_back(*w); + break; + } + continue; + } + if (ct.narrow(*wbb, 0) == '%') + { + result.push_back(L'%'); + result.push_back(L'%'); + ++wbb; + continue; + } + result.push_back(*wbb); + ++wbb; + } + return result; +} + +template <> +void +__time_get_storage::init(const ctype& ct) +{ + tm t = {0}; + char buf[100]; + // __weeks_ + for (int i = 0; i < 7; ++i) + { + t.tm_wday = i; + strftime_l(buf, countof(buf), "%A", &t, __loc_); + __weeks_[i] = buf; + strftime_l(buf, countof(buf), "%a", &t, __loc_); + __weeks_[i+7] = buf; + } + // __months_ + for (int i = 0; i < 12; ++i) + { + t.tm_mon = i; + strftime_l(buf, countof(buf), "%B", &t, __loc_); + __months_[i] = buf; + strftime_l(buf, countof(buf), "%b", &t, __loc_); + __months_[i+12] = buf; + } + // __am_pm_ + t.tm_hour = 1; + strftime_l(buf, countof(buf), "%p", &t, __loc_); + __am_pm_[0] = buf; + t.tm_hour = 13; + strftime_l(buf, countof(buf), "%p", &t, __loc_); + __am_pm_[1] = buf; + __c_ = __analyze('c', ct); + __r_ = __analyze('r', ct); + __x_ = __analyze('x', ct); + __X_ = __analyze('X', ct); +} + +template <> +void +__time_get_storage::init(const ctype& ct) +{ + tm t = {0}; + char buf[100]; + wchar_t wbuf[100]; + wchar_t* wbe; + mbstate_t mb = {0}; + // __weeks_ + for (int i = 0; i < 7; ++i) + { + t.tm_wday = i; + strftime_l(buf, countof(buf), "%A", &t, __loc_); + mb = mbstate_t(); + const char* bb = buf; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + size_t j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); +#else + size_t j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); +#endif + if (j == size_t(-1)) + __throw_runtime_error("locale not supported"); + wbe = wbuf + j; + __weeks_[i].assign(wbuf, wbe); + strftime_l(buf, countof(buf), "%a", &t, __loc_); + mb = mbstate_t(); + bb = buf; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); +#else + j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); +#endif + if (j == size_t(-1)) + __throw_runtime_error("locale not supported"); + wbe = wbuf + j; + __weeks_[i+7].assign(wbuf, wbe); + } + // __months_ + for (int i = 0; i < 12; ++i) + { + t.tm_mon = i; + strftime_l(buf, countof(buf), "%B", &t, __loc_); + mb = mbstate_t(); + const char* bb = buf; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + size_t j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); +#else + size_t j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); +#endif + if (j == size_t(-1)) + __throw_runtime_error("locale not supported"); + wbe = wbuf + j; + __months_[i].assign(wbuf, wbe); + strftime_l(buf, countof(buf), "%b", &t, __loc_); + mb = mbstate_t(); + bb = buf; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); +#else + j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); +#endif + if (j == size_t(-1)) + __throw_runtime_error("locale not supported"); + wbe = wbuf + j; + __months_[i+12].assign(wbuf, wbe); + } + // __am_pm_ + t.tm_hour = 1; + strftime_l(buf, countof(buf), "%p", &t, __loc_); + mb = mbstate_t(); + const char* bb = buf; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + size_t j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); +#else + size_t j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); +#endif + if (j == size_t(-1)) + __throw_runtime_error("locale not supported"); + wbe = wbuf + j; + __am_pm_[0].assign(wbuf, wbe); + t.tm_hour = 13; + strftime_l(buf, countof(buf), "%p", &t, __loc_); + mb = mbstate_t(); + bb = buf; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); +#else + j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, __loc_); +#endif + if (j == size_t(-1)) + __throw_runtime_error("locale not supported"); + wbe = wbuf + j; + __am_pm_[1].assign(wbuf, wbe); + __c_ = __analyze('c', ct); + __r_ = __analyze('r', ct); + __x_ = __analyze('x', ct); + __X_ = __analyze('X', ct); +} + +template +struct _LIBCPP_HIDDEN __time_get_temp + : public ctype_byname +{ + explicit __time_get_temp(const char* nm) + : ctype_byname(nm, 1) {} + explicit __time_get_temp(const string& nm) + : ctype_byname(nm, 1) {} +}; + +template <> +__time_get_storage::__time_get_storage(const char* __nm) + : __time_get(__nm) +{ + const __time_get_temp ct(__nm); + init(ct); +} + +template <> +__time_get_storage::__time_get_storage(const string& __nm) + : __time_get(__nm) +{ + const __time_get_temp ct(__nm); + init(ct); +} + +template <> +__time_get_storage::__time_get_storage(const char* __nm) + : __time_get(__nm) +{ + const __time_get_temp ct(__nm); + init(ct); +} + +template <> +__time_get_storage::__time_get_storage(const string& __nm) + : __time_get(__nm) +{ + const __time_get_temp ct(__nm); + init(ct); +} + +template <> +time_base::dateorder +__time_get_storage::__do_date_order() const +{ + unsigned i; + for (i = 0; i < __x_.size(); ++i) + if (__x_[i] == '%') + break; + ++i; + switch (__x_[i]) + { + case 'y': + case 'Y': + for (++i; i < __x_.size(); ++i) + if (__x_[i] == '%') + break; + if (i == __x_.size()) + break; + ++i; + switch (__x_[i]) + { + case 'm': + for (++i; i < __x_.size(); ++i) + if (__x_[i] == '%') + break; + if (i == __x_.size()) + break; + ++i; + if (__x_[i] == 'd') + return time_base::ymd; + break; + case 'd': + for (++i; i < __x_.size(); ++i) + if (__x_[i] == '%') + break; + if (i == __x_.size()) + break; + ++i; + if (__x_[i] == 'm') + return time_base::ydm; + break; + } + break; + case 'm': + for (++i; i < __x_.size(); ++i) + if (__x_[i] == '%') + break; + if (i == __x_.size()) + break; + ++i; + if (__x_[i] == 'd') + { + for (++i; i < __x_.size(); ++i) + if (__x_[i] == '%') + break; + if (i == __x_.size()) + break; + ++i; + if (__x_[i] == 'y' || __x_[i] == 'Y') + return time_base::mdy; + break; + } + break; + case 'd': + for (++i; i < __x_.size(); ++i) + if (__x_[i] == '%') + break; + if (i == __x_.size()) + break; + ++i; + if (__x_[i] == 'm') + { + for (++i; i < __x_.size(); ++i) + if (__x_[i] == '%') + break; + if (i == __x_.size()) + break; + ++i; + if (__x_[i] == 'y' || __x_[i] == 'Y') + return time_base::dmy; + break; + } + break; + } + return time_base::no_order; +} + +template <> +time_base::dateorder +__time_get_storage::__do_date_order() const +{ + unsigned i; + for (i = 0; i < __x_.size(); ++i) + if (__x_[i] == L'%') + break; + ++i; + switch (__x_[i]) + { + case L'y': + case L'Y': + for (++i; i < __x_.size(); ++i) + if (__x_[i] == L'%') + break; + if (i == __x_.size()) + break; + ++i; + switch (__x_[i]) + { + case L'm': + for (++i; i < __x_.size(); ++i) + if (__x_[i] == L'%') + break; + if (i == __x_.size()) + break; + ++i; + if (__x_[i] == L'd') + return time_base::ymd; + break; + case L'd': + for (++i; i < __x_.size(); ++i) + if (__x_[i] == L'%') + break; + if (i == __x_.size()) + break; + ++i; + if (__x_[i] == L'm') + return time_base::ydm; + break; + } + break; + case L'm': + for (++i; i < __x_.size(); ++i) + if (__x_[i] == L'%') + break; + if (i == __x_.size()) + break; + ++i; + if (__x_[i] == L'd') + { + for (++i; i < __x_.size(); ++i) + if (__x_[i] == L'%') + break; + if (i == __x_.size()) + break; + ++i; + if (__x_[i] == L'y' || __x_[i] == L'Y') + return time_base::mdy; + break; + } + break; + case L'd': + for (++i; i < __x_.size(); ++i) + if (__x_[i] == L'%') + break; + if (i == __x_.size()) + break; + ++i; + if (__x_[i] == L'm') + { + for (++i; i < __x_.size(); ++i) + if (__x_[i] == L'%') + break; + if (i == __x_.size()) + break; + ++i; + if (__x_[i] == L'y' || __x_[i] == L'Y') + return time_base::dmy; + break; + } + break; + } + return time_base::no_order; +} + +// time_put + +__time_put::__time_put(const char* nm) + : __loc_(newlocale(LC_ALL_MASK, nm, 0)) +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + if (__loc_ == 0) + throw runtime_error("time_put_byname" + " failed to construct for " + string(nm)); +#endif // _LIBCPP_NO_EXCEPTIONS +} + +__time_put::__time_put(const string& nm) + : __loc_(newlocale(LC_ALL_MASK, nm.c_str(), 0)) +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + if (__loc_ == 0) + throw runtime_error("time_put_byname" + " failed to construct for " + nm); +#endif // _LIBCPP_NO_EXCEPTIONS +} + +__time_put::~__time_put() +{ + if (__loc_ != _LIBCPP_GET_C_LOCALE) + freelocale(__loc_); +} + +void +__time_put::__do_put(char* __nb, char*& __ne, const tm* __tm, + char __fmt, char __mod) const +{ + char fmt[] = {'%', __fmt, __mod, 0}; + if (__mod != 0) + swap(fmt[1], fmt[2]); + size_t n = strftime_l(__nb, countof(__nb, __ne), fmt, __tm, __loc_); + __ne = __nb + n; +} + +void +__time_put::__do_put(wchar_t* __wb, wchar_t*& __we, const tm* __tm, + char __fmt, char __mod) const +{ + char __nar[100]; + char* __ne = __nar + 100; + __do_put(__nar, __ne, __tm, __fmt, __mod); + mbstate_t mb = {0}; + const char* __nb = __nar; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + size_t j = mbsrtowcs_l(__wb, &__nb, countof(__wb, __we), &mb, __loc_); +#else + size_t j = __mbsrtowcs_l(__wb, &__nb, countof(__wb, __we), &mb, __loc_); +#endif + if (j == size_t(-1)) + __throw_runtime_error("locale not supported"); + __we = __wb + j; +} + +// moneypunct_byname + +template +static +void +__init_pat(money_base::pattern& pat, basic_string& __curr_symbol_, + bool intl, char cs_precedes, char sep_by_space, char sign_posn, + charT space_char) +{ + const char sign = static_cast(money_base::sign); + const char space = static_cast(money_base::space); + const char none = static_cast(money_base::none); + const char symbol = static_cast(money_base::symbol); + const char value = static_cast(money_base::value); + const bool symbol_contains_sep = intl && __curr_symbol_.size() == 4; + + // Comments on case branches reflect 'C11 7.11.2.1 The localeconv + // function'. "Space between sign and symbol or value" means that + // if the sign is adjacent to the symbol, there's a space between + // them, and otherwise there's a space between the sign and value. + // + // C11's localeconv specifies that the fourth character of an + // international curr_symbol is used to separate the sign and + // value when sep_by_space says to do so. C++ can't represent + // that, so we just use a space. When sep_by_space says to + // separate the symbol and value-or-sign with a space, we rearrange the + // curr_symbol to put its spacing character on the correct side of + // the symbol. + // + // We also need to avoid adding an extra space between the sign + // and value when the currency symbol is suppressed (by not + // setting showbase). We match glibc's strfmon by interpreting + // sep_by_space==1 as "omit the space when the currency symbol is + // absent". + // + // Users who want to get this right should use ICU instead. + + switch (cs_precedes) + { + case 0: // value before curr_symbol + if (symbol_contains_sep) { + // Move the separator to before the symbol, to place it + // between the value and symbol. + rotate(__curr_symbol_.begin(), __curr_symbol_.begin() + 3, + __curr_symbol_.end()); + } + switch (sign_posn) + { + case 0: // Parentheses surround the quantity and currency symbol. + pat.field[0] = sign; + pat.field[1] = value; + pat.field[2] = none; // Any space appears in the symbol. + pat.field[3] = symbol; + switch (sep_by_space) + { + case 0: // No space separates the currency symbol and value. + // This case may have changed between C99 and C11; + // assume the currency symbol matches the intention. + case 2: // Space between sign and currency or value. + // The "sign" is two parentheses, so no space here either. + return; + case 1: // Space between currency-and-sign or currency and value. + if (!symbol_contains_sep) { + // We insert the space into the symbol instead of + // setting pat.field[2]=space so that when + // showbase is not set, the space goes away too. + __curr_symbol_.insert(0, 1, space_char); + } + return; + default: + break; + } + break; + case 1: // The sign string precedes the quantity and currency symbol. + pat.field[0] = sign; + pat.field[3] = symbol; + switch (sep_by_space) + { + case 0: // No space separates the currency symbol and value. + pat.field[1] = value; + pat.field[2] = none; + return; + case 1: // Space between currency-and-sign or currency and value. + pat.field[1] = value; + pat.field[2] = none; + if (!symbol_contains_sep) { + // We insert the space into the symbol instead of + // setting pat.field[2]=space so that when + // showbase is not set, the space goes away too. + __curr_symbol_.insert(0, 1, space_char); + } + return; + case 2: // Space between sign and currency or value. + pat.field[1] = space; + pat.field[2] = value; + if (symbol_contains_sep) { + // Remove the separator from the symbol, since it + // has already appeared after the sign. + __curr_symbol_.erase(__curr_symbol_.begin()); + } + return; + default: + break; + } + break; + case 2: // The sign string succeeds the quantity and currency symbol. + pat.field[0] = value; + pat.field[3] = sign; + switch (sep_by_space) + { + case 0: // No space separates the currency symbol and value. + pat.field[1] = none; + pat.field[2] = symbol; + return; + case 1: // Space between currency-and-sign or currency and value. + if (!symbol_contains_sep) { + // We insert the space into the symbol instead of + // setting pat.field[1]=space so that when + // showbase is not set, the space goes away too. + __curr_symbol_.insert(0, 1, space_char); + } + pat.field[1] = none; + pat.field[2] = symbol; + return; + case 2: // Space between sign and currency or value. + pat.field[1] = symbol; + pat.field[2] = space; + if (symbol_contains_sep) { + // Remove the separator from the symbol, since it + // should not be removed if showbase is absent. + __curr_symbol_.erase(__curr_symbol_.begin()); + } + return; + default: + break; + } + break; + case 3: // The sign string immediately precedes the currency symbol. + pat.field[0] = value; + pat.field[3] = symbol; + switch (sep_by_space) + { + case 0: // No space separates the currency symbol and value. + pat.field[1] = none; + pat.field[2] = sign; + return; + case 1: // Space between currency-and-sign or currency and value. + pat.field[1] = space; + pat.field[2] = sign; + if (symbol_contains_sep) { + // Remove the separator from the symbol, since it + // has already appeared before the sign. + __curr_symbol_.erase(__curr_symbol_.begin()); + } + return; + case 2: // Space between sign and currency or value. + pat.field[1] = sign; + pat.field[2] = none; + if (!symbol_contains_sep) { + // We insert the space into the symbol instead of + // setting pat.field[2]=space so that when + // showbase is not set, the space goes away too. + __curr_symbol_.insert(0, 1, space_char); + } + return; + default: + break; + } + break; + case 4: // The sign string immediately succeeds the currency symbol. + pat.field[0] = value; + pat.field[3] = sign; + switch (sep_by_space) + { + case 0: // No space separates the currency symbol and value. + pat.field[1] = none; + pat.field[2] = symbol; + return; + case 1: // Space between currency-and-sign or currency and value. + pat.field[1] = none; + pat.field[2] = symbol; + if (!symbol_contains_sep) { + // We insert the space into the symbol instead of + // setting pat.field[1]=space so that when + // showbase is not set, the space goes away too. + __curr_symbol_.insert(0, 1, space_char); + } + return; + case 2: // Space between sign and currency or value. + pat.field[1] = symbol; + pat.field[2] = space; + if (symbol_contains_sep) { + // Remove the separator from the symbol, since it + // should not disappear when showbase is absent. + __curr_symbol_.erase(__curr_symbol_.begin()); + } + return; + default: + break; + } + break; + default: + break; + } + break; + case 1: // curr_symbol before value + switch (sign_posn) + { + case 0: // Parentheses surround the quantity and currency symbol. + pat.field[0] = sign; + pat.field[1] = symbol; + pat.field[2] = none; // Any space appears in the symbol. + pat.field[3] = value; + switch (sep_by_space) + { + case 0: // No space separates the currency symbol and value. + // This case may have changed between C99 and C11; + // assume the currency symbol matches the intention. + case 2: // Space between sign and currency or value. + // The "sign" is two parentheses, so no space here either. + return; + case 1: // Space between currency-and-sign or currency and value. + if (!symbol_contains_sep) { + // We insert the space into the symbol instead of + // setting pat.field[2]=space so that when + // showbase is not set, the space goes away too. + __curr_symbol_.insert(0, 1, space_char); + } + return; + default: + break; + } + break; + case 1: // The sign string precedes the quantity and currency symbol. + pat.field[0] = sign; + pat.field[3] = value; + switch (sep_by_space) + { + case 0: // No space separates the currency symbol and value. + pat.field[1] = symbol; + pat.field[2] = none; + return; + case 1: // Space between currency-and-sign or currency and value. + pat.field[1] = symbol; + pat.field[2] = none; + if (!symbol_contains_sep) { + // We insert the space into the symbol instead of + // setting pat.field[2]=space so that when + // showbase is not set, the space goes away too. + __curr_symbol_.push_back(space_char); + } + return; + case 2: // Space between sign and currency or value. + pat.field[1] = space; + pat.field[2] = symbol; + if (symbol_contains_sep) { + // Remove the separator from the symbol, since it + // has already appeared after the sign. + __curr_symbol_.pop_back(); + } + return; + default: + break; + } + break; + case 2: // The sign string succeeds the quantity and currency symbol. + pat.field[0] = symbol; + pat.field[3] = sign; + switch (sep_by_space) + { + case 0: // No space separates the currency symbol and value. + pat.field[1] = none; + pat.field[2] = value; + return; + case 1: // Space between currency-and-sign or currency and value. + pat.field[1] = none; + pat.field[2] = value; + if (!symbol_contains_sep) { + // We insert the space into the symbol instead of + // setting pat.field[1]=space so that when + // showbase is not set, the space goes away too. + __curr_symbol_.push_back(space_char); + } + return; + case 2: // Space between sign and currency or value. + pat.field[1] = value; + pat.field[2] = space; + if (symbol_contains_sep) { + // Remove the separator from the symbol, since it + // will appear before the sign. + __curr_symbol_.pop_back(); + } + return; + default: + break; + } + break; + case 3: // The sign string immediately precedes the currency symbol. + pat.field[0] = sign; + pat.field[3] = value; + switch (sep_by_space) + { + case 0: // No space separates the currency symbol and value. + pat.field[1] = symbol; + pat.field[2] = none; + return; + case 1: // Space between currency-and-sign or currency and value. + pat.field[1] = symbol; + pat.field[2] = none; + if (!symbol_contains_sep) { + // We insert the space into the symbol instead of + // setting pat.field[2]=space so that when + // showbase is not set, the space goes away too. + __curr_symbol_.push_back(space_char); + } + return; + case 2: // Space between sign and currency or value. + pat.field[1] = space; + pat.field[2] = symbol; + if (symbol_contains_sep) { + // Remove the separator from the symbol, since it + // has already appeared after the sign. + __curr_symbol_.pop_back(); + } + return; + default: + break; + } + break; + case 4: // The sign string immediately succeeds the currency symbol. + pat.field[0] = symbol; + pat.field[3] = value; + switch (sep_by_space) + { + case 0: // No space separates the currency symbol and value. + pat.field[1] = sign; + pat.field[2] = none; + return; + case 1: // Space between currency-and-sign or currency and value. + pat.field[1] = sign; + pat.field[2] = space; + if (symbol_contains_sep) { + // Remove the separator from the symbol, since it + // should not disappear when showbase is absent. + __curr_symbol_.pop_back(); + } + return; + case 2: // Space between sign and currency or value. + pat.field[1] = none; + pat.field[2] = sign; + if (!symbol_contains_sep) { + // We insert the space into the symbol instead of + // setting pat.field[1]=space so that when + // showbase is not set, the space goes away too. + __curr_symbol_.push_back(space_char); + } + return; + default: + break; + } + break; + default: + break; + } + break; + default: + break; + } + pat.field[0] = symbol; + pat.field[1] = sign; + pat.field[2] = none; + pat.field[3] = value; +} + +template<> +void +moneypunct_byname::init(const char* nm) +{ + typedef moneypunct base; + __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); +#ifndef _LIBCPP_NO_EXCEPTIONS + if (loc == nullptr) + throw runtime_error("moneypunct_byname" + " failed to construct for " + string(nm)); +#endif // _LIBCPP_NO_EXCEPTIONS +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + lconv* lc = localeconv_l(loc.get()); +#else + lconv* lc = __localeconv_l(loc.get()); +#endif + if (*lc->mon_decimal_point) + __decimal_point_ = *lc->mon_decimal_point; + else + __decimal_point_ = base::do_decimal_point(); + if (*lc->mon_thousands_sep) + __thousands_sep_ = *lc->mon_thousands_sep; + else + __thousands_sep_ = base::do_thousands_sep(); + __grouping_ = lc->mon_grouping; + __curr_symbol_ = lc->currency_symbol; + if (lc->frac_digits != CHAR_MAX) + __frac_digits_ = lc->frac_digits; + else + __frac_digits_ = base::do_frac_digits(); + if (lc->p_sign_posn == 0) + __positive_sign_ = "()"; + else + __positive_sign_ = lc->positive_sign; + if (lc->n_sign_posn == 0) + __negative_sign_ = "()"; + else + __negative_sign_ = lc->negative_sign; + // Assume the positive and negative formats will want spaces in + // the same places in curr_symbol since there's no way to + // represent anything else. + string_type __dummy_curr_symbol = __curr_symbol_; + __init_pat(__pos_format_, __dummy_curr_symbol, false, + lc->p_cs_precedes, lc->p_sep_by_space, lc->p_sign_posn, ' '); + __init_pat(__neg_format_, __curr_symbol_, false, + lc->n_cs_precedes, lc->n_sep_by_space, lc->n_sign_posn, ' '); +} + +template<> +void +moneypunct_byname::init(const char* nm) +{ + typedef moneypunct base; + __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); +#ifndef _LIBCPP_NO_EXCEPTIONS + if (loc == nullptr) + throw runtime_error("moneypunct_byname" + " failed to construct for " + string(nm)); +#endif // _LIBCPP_NO_EXCEPTIONS +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + lconv* lc = localeconv_l(loc.get()); +#else + lconv* lc = __localeconv_l(loc.get()); +#endif + if (*lc->mon_decimal_point) + __decimal_point_ = *lc->mon_decimal_point; + else + __decimal_point_ = base::do_decimal_point(); + if (*lc->mon_thousands_sep) + __thousands_sep_ = *lc->mon_thousands_sep; + else + __thousands_sep_ = base::do_thousands_sep(); + __grouping_ = lc->mon_grouping; + __curr_symbol_ = lc->int_curr_symbol; + if (lc->int_frac_digits != CHAR_MAX) + __frac_digits_ = lc->int_frac_digits; + else + __frac_digits_ = base::do_frac_digits(); +#if defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) + if (lc->p_sign_posn == 0) +#else // _LIBCPP_MSVCRT + if (lc->int_p_sign_posn == 0) +#endif // !_LIBCPP_MSVCRT + __positive_sign_ = "()"; + else + __positive_sign_ = lc->positive_sign; +#if defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) + if(lc->n_sign_posn == 0) +#else // _LIBCPP_MSVCRT + if (lc->int_n_sign_posn == 0) +#endif // !_LIBCPP_MSVCRT + __negative_sign_ = "()"; + else + __negative_sign_ = lc->negative_sign; + // Assume the positive and negative formats will want spaces in + // the same places in curr_symbol since there's no way to + // represent anything else. + string_type __dummy_curr_symbol = __curr_symbol_; +#if defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) + __init_pat(__pos_format_, __dummy_curr_symbol, true, + lc->p_cs_precedes, lc->p_sep_by_space, lc->p_sign_posn, ' '); + __init_pat(__neg_format_, __curr_symbol_, true, + lc->n_cs_precedes, lc->n_sep_by_space, lc->n_sign_posn, ' '); +#else // _LIBCPP_MSVCRT + __init_pat(__pos_format_, __dummy_curr_symbol, true, + lc->int_p_cs_precedes, lc->int_p_sep_by_space, + lc->int_p_sign_posn, ' '); + __init_pat(__neg_format_, __curr_symbol_, true, + lc->int_n_cs_precedes, lc->int_n_sep_by_space, + lc->int_n_sign_posn, ' '); +#endif // !_LIBCPP_MSVCRT +} + +template<> +void +moneypunct_byname::init(const char* nm) +{ + typedef moneypunct base; + __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); +#ifndef _LIBCPP_NO_EXCEPTIONS + if (loc == nullptr) + throw runtime_error("moneypunct_byname" + " failed to construct for " + string(nm)); +#endif // _LIBCPP_NO_EXCEPTIONS +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + lconv* lc = localeconv_l(loc.get()); +#else + lconv* lc = __localeconv_l(loc.get()); +#endif + if (*lc->mon_decimal_point) + __decimal_point_ = static_cast(*lc->mon_decimal_point); + else + __decimal_point_ = base::do_decimal_point(); + if (*lc->mon_thousands_sep) + __thousands_sep_ = static_cast(*lc->mon_thousands_sep); + else + __thousands_sep_ = base::do_thousands_sep(); + __grouping_ = lc->mon_grouping; + wchar_t wbuf[100]; + mbstate_t mb = {0}; + const char* bb = lc->currency_symbol; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + size_t j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); +#else + size_t j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); +#endif + if (j == size_t(-1)) + __throw_runtime_error("locale not supported"); + wchar_t* wbe = wbuf + j; + __curr_symbol_.assign(wbuf, wbe); + if (lc->frac_digits != CHAR_MAX) + __frac_digits_ = lc->frac_digits; + else + __frac_digits_ = base::do_frac_digits(); + if (lc->p_sign_posn == 0) + __positive_sign_ = L"()"; + else + { + mb = mbstate_t(); + bb = lc->positive_sign; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); +#else + j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); +#endif + if (j == size_t(-1)) + __throw_runtime_error("locale not supported"); + wbe = wbuf + j; + __positive_sign_.assign(wbuf, wbe); + } + if (lc->n_sign_posn == 0) + __negative_sign_ = L"()"; + else + { + mb = mbstate_t(); + bb = lc->negative_sign; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); +#else + j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); +#endif + if (j == size_t(-1)) + __throw_runtime_error("locale not supported"); + wbe = wbuf + j; + __negative_sign_.assign(wbuf, wbe); + } + // Assume the positive and negative formats will want spaces in + // the same places in curr_symbol since there's no way to + // represent anything else. + string_type __dummy_curr_symbol = __curr_symbol_; + __init_pat(__pos_format_, __dummy_curr_symbol, false, + lc->p_cs_precedes, lc->p_sep_by_space, lc->p_sign_posn, L' '); + __init_pat(__neg_format_, __curr_symbol_, false, + lc->n_cs_precedes, lc->n_sep_by_space, lc->n_sign_posn, L' '); +} + +template<> +void +moneypunct_byname::init(const char* nm) +{ + typedef moneypunct base; + __locale_unique_ptr loc(newlocale(LC_ALL_MASK, nm, 0), freelocale); +#ifndef _LIBCPP_NO_EXCEPTIONS + if (loc == nullptr) + throw runtime_error("moneypunct_byname" + " failed to construct for " + string(nm)); +#endif // _LIBCPP_NO_EXCEPTIONS +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + lconv* lc = localeconv_l(loc.get()); +#else + lconv* lc = __localeconv_l(loc.get()); +#endif + if (*lc->mon_decimal_point) + __decimal_point_ = static_cast(*lc->mon_decimal_point); + else + __decimal_point_ = base::do_decimal_point(); + if (*lc->mon_thousands_sep) + __thousands_sep_ = static_cast(*lc->mon_thousands_sep); + else + __thousands_sep_ = base::do_thousands_sep(); + __grouping_ = lc->mon_grouping; + wchar_t wbuf[100]; + mbstate_t mb = {0}; + const char* bb = lc->int_curr_symbol; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + size_t j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); +#else + size_t j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); +#endif + if (j == size_t(-1)) + __throw_runtime_error("locale not supported"); + wchar_t* wbe = wbuf + j; + __curr_symbol_.assign(wbuf, wbe); + if (lc->int_frac_digits != CHAR_MAX) + __frac_digits_ = lc->int_frac_digits; + else + __frac_digits_ = base::do_frac_digits(); +#if defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) + if (lc->p_sign_posn == 0) +#else // _LIBCPP_MSVCRT + if (lc->int_p_sign_posn == 0) +#endif // !_LIBCPP_MSVCRT + __positive_sign_ = L"()"; + else + { + mb = mbstate_t(); + bb = lc->positive_sign; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); +#else + j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); +#endif + if (j == size_t(-1)) + __throw_runtime_error("locale not supported"); + wbe = wbuf + j; + __positive_sign_.assign(wbuf, wbe); + } +#if defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) + if (lc->n_sign_posn == 0) +#else // _LIBCPP_MSVCRT + if (lc->int_n_sign_posn == 0) +#endif // !_LIBCPP_MSVCRT + __negative_sign_ = L"()"; + else + { + mb = mbstate_t(); + bb = lc->negative_sign; +#ifdef _LIBCPP_LOCALE__L_EXTENSIONS + j = mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); +#else + j = __mbsrtowcs_l(wbuf, &bb, countof(wbuf), &mb, loc.get()); +#endif + if (j == size_t(-1)) + __throw_runtime_error("locale not supported"); + wbe = wbuf + j; + __negative_sign_.assign(wbuf, wbe); + } + // Assume the positive and negative formats will want spaces in + // the same places in curr_symbol since there's no way to + // represent anything else. + string_type __dummy_curr_symbol = __curr_symbol_; +#if defined(_LIBCPP_MSVCRT) || defined(__MINGW32__) + __init_pat(__pos_format_, __dummy_curr_symbol, true, + lc->p_cs_precedes, lc->p_sep_by_space, lc->p_sign_posn, L' '); + __init_pat(__neg_format_, __curr_symbol_, true, + lc->n_cs_precedes, lc->n_sep_by_space, lc->n_sign_posn, L' '); +#else // _LIBCPP_MSVCRT + __init_pat(__pos_format_, __dummy_curr_symbol, true, + lc->int_p_cs_precedes, lc->int_p_sep_by_space, + lc->int_p_sign_posn, L' '); + __init_pat(__neg_format_, __curr_symbol_, true, + lc->int_n_cs_precedes, lc->int_n_sep_by_space, + lc->int_n_sign_posn, L' '); +#endif // !_LIBCPP_MSVCRT +} + +void __do_nothing(void*) {} + +void __throw_runtime_error(const char* msg) +{ +#ifndef _LIBCPP_NO_EXCEPTIONS + throw runtime_error(msg); +#else + (void)msg; +#endif +} + +template class collate; +template class collate; + +template class num_get; +template class num_get; + +template struct __num_get; +template struct __num_get; + +template class num_put; +template class num_put; + +template struct __num_put; +template struct __num_put; + +template class time_get; +template class time_get; + +template class time_get_byname; +template class time_get_byname; + +template class time_put; +template class time_put; + +template class time_put_byname; +template class time_put_byname; + +template class moneypunct; +template class moneypunct; +template class moneypunct; +template class moneypunct; + +template class moneypunct_byname; +template class moneypunct_byname; +template class moneypunct_byname; +template class moneypunct_byname; + +template class money_get; +template class money_get; + +template class __money_get; +template class __money_get; + +template class money_put; +template class money_put; + +template class __money_put; +template class __money_put; + +template class messages; +template class messages; + +template class messages_byname; +template class messages_byname; + +template class codecvt_byname; +template class codecvt_byname; +template class codecvt_byname; +template class codecvt_byname; + +template class __vector_base_common; + +_LIBCPP_END_NAMESPACE_STD Index: test/CMakeLists.txt =================================================================== --- test/CMakeLists.txt +++ test/CMakeLists.txt @@ -60,6 +60,10 @@ set(LIBCXX_EXECUTOR "None" CACHE STRING "Executor to use when running tests.") + # Detect size of wchar_t + include(CheckTypeSize) + check_type_size("wchar_t" WCHAR_SIZE) + set(AUTO_GEN_COMMENT "## Autogenerated by libcxx configuration.\n# Do not edit!") configure_file( Index: test/libcxx/test/config.py =================================================================== --- test/libcxx/test/config.py +++ test/libcxx/test/config.py @@ -330,6 +330,9 @@ if self.cxx.hasCompileFlag('-fsized-deallocation'): self.config.available_features.add('fsized-deallocation') + if self.config.wchar_size == '2': + self.config.available_features.add('short_wchar') + def configure_compile_flags(self): no_default_flags = self.get_lit_bool('no_default_flags', False) if not no_default_flags: Index: test/lit.site.cfg.in =================================================================== --- test/lit.site.cfg.in +++ test/lit.site.cfg.in @@ -1,29 +1,30 @@ -@AUTO_GEN_COMMENT@ -config.cxx_under_test = "@LIBCXX_COMPILER@" -config.std = "@LIBCXX_STD_VERSION@" -config.libcxx_src_root = "@LIBCXX_SOURCE_DIR@" -config.libcxx_obj_root = "@LIBCXX_BINARY_DIR@" -config.cxx_library_root = "@LIBCXX_LIBRARY_DIR@" -config.enable_exceptions = "@LIBCXX_ENABLE_EXCEPTIONS@" -config.enable_rtti = "@LIBCXX_ENABLE_RTTI@" -config.enable_shared = "@LIBCXX_ENABLE_SHARED@" -config.enable_32bit = "@LIBCXX_BUILD_32_BITS@" -config.enable_global_filesystem_namespace = "@LIBCXX_ENABLE_GLOBAL_FILESYSTEM_NAMESPACE@" -config.enable_stdin = "@LIBCXX_ENABLE_STDIN@" -config.enable_stdout = "@LIBCXX_ENABLE_STDOUT@" -config.enable_threads = "@LIBCXX_ENABLE_THREADS@" -config.enable_monotonic_clock = "@LIBCXX_ENABLE_MONOTONIC_CLOCK@" -config.cxx_abi = "@LIBCXX_CXX_ABI_LIBNAME@" -config.use_sanitizer = "@LLVM_USE_SANITIZER@" -config.abi_library_path = "@LIBCXX_CXX_ABI_LIBRARY_PATH@" -config.configuration_variant = "@LIBCXX_LIT_VARIANT@" -config.target_triple = "@LIBCXX_TARGET_TRIPLE@" -config.sysroot = "@LIBCXX_SYSROOT@" -config.gcc_toolchain = "@LIBCXX_GCC_TOOLCHAIN@" -config.generate_coverage = "@LIBCXX_GENERATE_COVERAGE@" -config.target_info = "@LIBCXX_TARGET_INFO@" -config.executor = "@LIBCXX_EXECUTOR@" -config.llvm_unwinder = "@LIBCXXABI_USE_LLVM_UNWINDER@" - -# Let the main config do the real work. -lit_config.load_config(config, "@LIBCXX_SOURCE_DIR@/test/lit.cfg") +@AUTO_GEN_COMMENT@ +config.cxx_under_test = "@LIBCXX_COMPILER@" +config.std = "@LIBCXX_STD_VERSION@" +config.libcxx_src_root = "@LIBCXX_SOURCE_DIR@" +config.libcxx_obj_root = "@LIBCXX_BINARY_DIR@" +config.cxx_library_root = "@LIBCXX_LIBRARY_DIR@" +config.enable_exceptions = "@LIBCXX_ENABLE_EXCEPTIONS@" +config.enable_rtti = "@LIBCXX_ENABLE_RTTI@" +config.enable_shared = "@LIBCXX_ENABLE_SHARED@" +config.enable_32bit = "@LIBCXX_BUILD_32_BITS@" +config.enable_global_filesystem_namespace = "@LIBCXX_ENABLE_GLOBAL_FILESYSTEM_NAMESPACE@" +config.enable_stdin = "@LIBCXX_ENABLE_STDIN@" +config.enable_stdout = "@LIBCXX_ENABLE_STDOUT@" +config.enable_threads = "@LIBCXX_ENABLE_THREADS@" +config.enable_monotonic_clock = "@LIBCXX_ENABLE_MONOTONIC_CLOCK@" +config.cxx_abi = "@LIBCXX_CXX_ABI_LIBNAME@" +config.use_sanitizer = "@LLVM_USE_SANITIZER@" +config.abi_library_path = "@LIBCXX_CXX_ABI_LIBRARY_PATH@" +config.configuration_variant = "@LIBCXX_LIT_VARIANT@" +config.target_triple = "@LIBCXX_TARGET_TRIPLE@" +config.sysroot = "@LIBCXX_SYSROOT@" +config.gcc_toolchain = "@LIBCXX_GCC_TOOLCHAIN@" +config.wchar_size = "@WCHAR_SIZE@" +config.generate_coverage = "@LIBCXX_GENERATE_COVERAGE@" +config.target_info = "@LIBCXX_TARGET_INFO@" +config.executor = "@LIBCXX_EXECUTOR@" +config.llvm_unwinder = "@LIBCXXABI_USE_LLVM_UNWINDER@" + +# Let the main config do the real work. +lit_config.load_config(config, "@LIBCXX_SOURCE_DIR@/test/lit.cfg") Index: test/std/localization/locale.stdcvt/codecvt_utf16_in.pass.cpp =================================================================== --- test/std/localization/locale.stdcvt/codecvt_utf16_in.pass.cpp +++ test/std/localization/locale.stdcvt/codecvt_utf16_in.pass.cpp @@ -1,739 +1,741 @@ -//===----------------------------------------------------------------------===// -// -// The LLVM Compiler Infrastructure -// -// This file is dual licensed under the MIT and the University of Illinois Open -// Source Licenses. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - -// - -// template -// class codecvt_utf16 -// : public codecvt -// { -// // unspecified -// }; - -// result -// in(stateT& state, -// const externT* from, const externT* from_end, const externT*& from_next, -// internT* to, internT* to_end, internT*& to_next) const; - -#include -#include - -int main() -{ - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0; - char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - wchar_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(w == 0x40003); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x1005); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x00); - n[1] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0; - char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - wchar_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x00); - n[1] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0; - char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - wchar_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+6); - assert(w == 0x40003); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x1005); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x00); - n[1] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0; - char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - wchar_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(w == 0x40003); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x1005); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[1] = char(0x00); - n[0] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0; - char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - wchar_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[1] = char(0x00); - n[0] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0; - char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - wchar_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+6); - assert(w == 0x40003); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x1005); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[1] = char(0x00); - n[0] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - char32_t w = 0; - char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - char32_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(w == 0x40003); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x1005); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x00); - n[1] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - char32_t w = 0; - char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - char32_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x00); - n[1] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - char32_t w = 0; - char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - char32_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+6); - assert(w == 0x40003); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x1005); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x00); - n[1] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - char32_t w = 0; - char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - char32_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(w == 0x40003); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x1005); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[1] = char(0x00); - n[0] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - char32_t w = 0; - char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - char32_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[1] = char(0x00); - n[0] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - char32_t w = 0; - char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - char32_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+6); - assert(w == 0x40003); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x1005); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[1] = char(0x00); - n[0] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - - { - typedef std::codecvt_utf16 C; - C c; - char16_t w = 0; - char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - char16_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x1005); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x00); - n[1] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - char16_t w = 0; - char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - char16_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x00); - n[1] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - char16_t w = 0; - char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - char16_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n+2); - assert(w == 0); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x1005); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x00); - n[1] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - char16_t w = 0; - char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - char16_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x1005); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[1] = char(0x00); - n[0] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - char16_t w = 0; - char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - char16_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[1] = char(0x00); - n[0] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } - { - typedef std::codecvt_utf16 C; - C c; - char16_t w = 0; - char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - char16_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n+2); - assert(w == 0); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x1005); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[1] = char(0x00); - n[0] = char(0x56); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x56); - } -} +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// template +// class codecvt_utf16 +// : public codecvt +// { +// // unspecified +// }; + +// result +// in(stateT& state, +// const externT* from, const externT* from_end, const externT*& from_next, +// internT* to, internT* to_end, internT*& to_next) const; + +// UNSUPPORTED: short_wchar + +#include +#include + +int main() +{ + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(w == 0x40003); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0; + char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(w == 0x40003); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(w == 0x40003); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0; + char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(w == 0x40003); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char32_t w = 0; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(w == 0x40003); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char32_t w = 0; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char32_t w = 0; + char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(w == 0x40003); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char32_t w = 0; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(w == 0x40003); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char32_t w = 0; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char32_t w = 0; + char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(w == 0x40003); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + + { + typedef std::codecvt_utf16 C; + C c; + char16_t w = 0; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char16_t w = 0; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char16_t w = 0; + char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n+2); + assert(w == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char16_t w = 0; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char16_t w = 0; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char16_t w = 0; + char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n+2); + assert(w == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } +} Index: test/std/localization/locale.stdcvt/codecvt_utf16_in_short_wchar.pass.cpp =================================================================== --- test/std/localization/locale.stdcvt/codecvt_utf16_in_short_wchar.pass.cpp +++ test/std/localization/locale.stdcvt/codecvt_utf16_in_short_wchar.pass.cpp @@ -0,0 +1,693 @@ +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// template +// class codecvt_utf16 +// : public codecvt +// { +// // unspecified +// }; + +// result +// in(stateT& state, +// const externT* from, const externT* from_end, const externT*& from_next, +// internT* to, internT* to_end, internT*& to_next) const; + +// REQUIRES: short_wchar + +#include +#include + +int main() +{ + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0; + char n[2] = {char(0x10), char(0x05)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0; + char n[2] = {char(0x10), char(0x05)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0; + char n[4] = {char(0xFE), char(0xFF), char(0x10), char(0x05)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(w == 0x1005); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0; + char n[2] = {char(0x05), char(0x10)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0; + char n[2] = {char(0x05), char(0x10)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0; + char n[6] = {char(0xFF), char(0xFE), char(0x05), char(0x10)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(w == 0x1005); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char32_t w = 0; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(w == 0x40003); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char32_t w = 0; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char32_t w = 0; + char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(w == 0x40003); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char32_t w = 0; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(w == 0x40003); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char32_t w = 0; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char32_t w = 0; + char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(w == 0x40003); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + + { + typedef std::codecvt_utf16 C; + C c; + char16_t w = 0; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char16_t w = 0; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char16_t w = 0; + char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n+2); + assert(w == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x00); + n[1] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char16_t w = 0; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char16_t w = 0; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } + { + typedef std::codecvt_utf16 C; + C c; + char16_t w = 0; + char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n+2); + assert(w == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x1005); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[1] = char(0x00); + n[0] = char(0x56); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x56); + } +} Index: test/std/localization/locale.stdcvt/codecvt_utf16_length.pass.cpp =================================================================== --- test/std/localization/locale.stdcvt/codecvt_utf16_length.pass.cpp +++ test/std/localization/locale.stdcvt/codecvt_utf16_length.pass.cpp @@ -1,449 +1,451 @@ -//===----------------------------------------------------------------------===// -// -// The LLVM Compiler Infrastructure -// -// This file is dual licensed under the MIT and the University of Illinois Open -// Source Licenses. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - -// - -// template -// class codecvt_utf16 -// : public codecvt -// { -// // unspecified -// }; - -// int length(stateT& state, const externT* from, const externT* from_end, -// size_t max) const; - -#include -#include - -int main() -{ - { - typedef std::codecvt_utf16 C; - C c; - char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 2); - assert(r == 4); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[0] = char(0x00); - n[1] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 2); - assert(r == 0); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 0); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[0] = char(0x00); - n[1] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - std::mbstate_t m; - int r = c.length(m, n, n+6, 2); - assert(r == 6); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[0] = char(0x00); - n[1] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 2); - assert(r == 4); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[1] = char(0x00); - n[0] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 2); - assert(r == 0); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 0); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[1] = char(0x00); - n[0] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - std::mbstate_t m; - int r = c.length(m, n, n+6, 2); - assert(r == 6); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[1] = char(0x00); - n[0] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 2); - assert(r == 4); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[0] = char(0x00); - n[1] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 2); - assert(r == 0); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 0); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[0] = char(0x00); - n[1] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - std::mbstate_t m; - int r = c.length(m, n, n+6, 2); - assert(r == 6); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[0] = char(0x00); - n[1] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 2); - assert(r == 4); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[1] = char(0x00); - n[0] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 2); - assert(r == 0); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 0); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[1] = char(0x00); - n[0] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - std::mbstate_t m; - int r = c.length(m, n, n+6, 2); - assert(r == 6); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[1] = char(0x00); - n[0] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - - { - typedef std::codecvt_utf16 C; - C c; - char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 2); - assert(r == 0); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[0] = char(0x00); - n[1] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 2); - assert(r == 0); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 0); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[0] = char(0x00); - n[1] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; - std::mbstate_t m; - int r = c.length(m, n, n+6, 2); - assert(r == 2); - - n[0] = char(0x10); - n[1] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[0] = char(0x04); - n[1] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[0] = char(0x00); - n[1] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 2); - assert(r == 0); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[1] = char(0x00); - n[0] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 2); - assert(r == 0); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 0); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[1] = char(0x00); - n[0] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } - { - typedef std::codecvt_utf16 C; - C c; - char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; - std::mbstate_t m; - int r = c.length(m, n, n+6, 2); - assert(r == 2); - - n[1] = char(0x10); - n[0] = char(0x05); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[1] = char(0x04); - n[0] = char(0x53); - r = c.length(m, n, n+2, 2); - assert(r == 2); - - n[1] = char(0x00); - n[0] = char(0x56); - r = c.length(m, n, n+2, 2); - assert(r == 2); - } -} +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// template +// class codecvt_utf16 +// : public codecvt +// { +// // unspecified +// }; + +// int length(stateT& state, const externT* from, const externT* from_end, +// size_t max) const; + +// UNSUPPORTED: short_wchar + +#include +#include + +int main() +{ + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 4); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 0); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + std::mbstate_t m; + int r = c.length(m, n, n+6, 2); + assert(r == 6); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 4); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 0); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + std::mbstate_t m; + int r = c.length(m, n, n+6, 2); + assert(r == 6); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 4); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 0); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + std::mbstate_t m; + int r = c.length(m, n, n+6, 2); + assert(r == 6); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 4); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 0); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + std::mbstate_t m; + int r = c.length(m, n, n+6, 2); + assert(r == 6); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 0); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + std::mbstate_t m; + int r = c.length(m, n, n+6, 2); + assert(r == 2); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 0); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + std::mbstate_t m; + int r = c.length(m, n, n+6, 2); + assert(r == 2); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } +} Index: test/std/localization/locale.stdcvt/codecvt_utf16_length_short_wchar.pass.cpp =================================================================== --- test/std/localization/locale.stdcvt/codecvt_utf16_length_short_wchar.pass.cpp +++ test/std/localization/locale.stdcvt/codecvt_utf16_length_short_wchar.pass.cpp @@ -0,0 +1,426 @@ +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// template +// class codecvt_utf16 +// : public codecvt +// { +// // unspecified +// }; + +// int length(stateT& state, const externT* from, const externT* from_end, +// size_t max) const; + +// REQUIRES: short_wchar + +#include +#include + +int main() +{ + { + typedef std::codecvt_utf16 C; + C c; + char n[2] = {char(0x10), char(0x05)}; + std::mbstate_t m; + int r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[2] = {char(0x10), char(0x05)}; + std::mbstate_t m; + int r = c.length(m, n, n+2, 2); + assert(r == 0); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xFE), char(0xFF), char(0x10), char(0x05)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 4); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[2] = {char(0x05), char(0x10)}; + std::mbstate_t m; + int r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[2] = {char(0x05), char(0x10)}; + std::mbstate_t m; + int r = c.length(m, n, n+2, 2); + assert(r == 0); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xFF), char(0xFE), char(0x05), char(0x10)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 4); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 4); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 0); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + std::mbstate_t m; + int r = c.length(m, n, n+6, 2); + assert(r == 6); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 4); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 0); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + std::mbstate_t m; + int r = c.length(m, n, n+6, 2); + assert(r == 6); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 0); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 0); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[6] = {char(0xFE), char(0xFF), char(0xD8), char(0xC0), char(0xDC), char(0x03)}; + std::mbstate_t m; + int r = c.length(m, n, n+6, 2); + assert(r == 2); + + n[0] = char(0x10); + n[1] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x04); + n[1] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[0] = char(0x00); + n[1] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[4] = {char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 2); + assert(r == 0); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 0); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } + { + typedef std::codecvt_utf16 C; + C c; + char n[6] = {char(0xFF), char(0xFE), char(0xC0), char(0xD8), char(0x03), char(0xDC)}; + std::mbstate_t m; + int r = c.length(m, n, n+6, 2); + assert(r == 2); + + n[1] = char(0x10); + n[0] = char(0x05); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x04); + n[0] = char(0x53); + r = c.length(m, n, n+2, 2); + assert(r == 2); + + n[1] = char(0x00); + n[0] = char(0x56); + r = c.length(m, n, n+2, 2); + assert(r == 2); + } +} Index: test/std/localization/locale.stdcvt/codecvt_utf16_out.pass.cpp =================================================================== --- test/std/localization/locale.stdcvt/codecvt_utf16_out.pass.cpp +++ test/std/localization/locale.stdcvt/codecvt_utf16_out.pass.cpp @@ -1,331 +1,333 @@ -//===----------------------------------------------------------------------===// -// -// The LLVM Compiler Infrastructure -// -// This file is dual licensed under the MIT and the University of Illinois Open -// Source Licenses. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - -// - -// template -// class codecvt_utf16 -// : public codecvt -// { -// // unspecified -// }; - -// result -// out(stateT& state, -// const internT* from, const internT* from_end, const internT*& from_next, -// externT* to, externT* to_end, externT*& to_next) const; - -#include -#include - -int main() -{ - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0x40003; - char n[4] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xD8)); - assert(n[1] == char(0xC0)); - assert(n[2] == char(0xDC)); - assert(n[3] == char(0x03)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0x10)); - assert(n[1] == char(0x05)); - assert(n[2] == char(0xDC)); - assert(n[3] == char(0x03)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0x04)); - assert(n[1] == char(0x53)); - assert(n[2] == char(0xDC)); - assert(n[3] == char(0x03)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0x00)); - assert(n[1] == char(0x56)); - assert(n[2] == char(0xDC)); - assert(n[3] == char(0x03)); - } - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0x40003; - char n[4] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[0] == char(0)); - assert(n[1] == char(0)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[0] == char(0)); - assert(n[1] == char(0)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0x04)); - assert(n[1] == char(0x53)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0x00)); - assert(n[1] == char(0x56)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - } - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0x40003; - char n[6] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+6); - assert(n[0] == char(0xFE)); - assert(n[1] == char(0xFF)); - assert(n[2] == char(0xD8)); - assert(n[3] == char(0xC0)); - assert(n[4] == char(0xDC)); - assert(n[5] == char(0x03)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xFE)); - assert(n[1] == char(0xFF)); - assert(n[2] == char(0x10)); - assert(n[3] == char(0x05)); - assert(n[4] == char(0xDC)); - assert(n[5] == char(0x03)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xFE)); - assert(n[1] == char(0xFF)); - assert(n[2] == char(0x04)); - assert(n[3] == char(0x53)); - assert(n[4] == char(0xDC)); - assert(n[5] == char(0x03)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xFE)); - assert(n[1] == char(0xFF)); - assert(n[2] == char(0x00)); - assert(n[3] == char(0x56)); - assert(n[4] == char(0xDC)); - assert(n[5] == char(0x03)); - } - - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0x40003; - char n[4] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[1] == char(0xD8)); - assert(n[0] == char(0xC0)); - assert(n[3] == char(0xDC)); - assert(n[2] == char(0x03)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[1] == char(0x10)); - assert(n[0] == char(0x05)); - assert(n[3] == char(0xDC)); - assert(n[2] == char(0x03)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[1] == char(0x04)); - assert(n[0] == char(0x53)); - assert(n[3] == char(0xDC)); - assert(n[2] == char(0x03)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[1] == char(0x00)); - assert(n[0] == char(0x56)); - assert(n[3] == char(0xDC)); - assert(n[2] == char(0x03)); - } - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0x40003; - char n[4] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[1] == char(0)); - assert(n[0] == char(0)); - assert(n[3] == char(0)); - assert(n[2] == char(0)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[1] == char(0)); - assert(n[0] == char(0)); - assert(n[3] == char(0)); - assert(n[2] == char(0)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[1] == char(0x04)); - assert(n[0] == char(0x53)); - assert(n[3] == char(0)); - assert(n[2] == char(0)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[1] == char(0x00)); - assert(n[0] == char(0x56)); - assert(n[3] == char(0)); - assert(n[2] == char(0)); - } - { - typedef std::codecvt_utf16 C; - C c; - wchar_t w = 0x40003; - char n[6] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+6); - assert(n[1] == char(0xFE)); - assert(n[0] == char(0xFF)); - assert(n[3] == char(0xD8)); - assert(n[2] == char(0xC0)); - assert(n[5] == char(0xDC)); - assert(n[4] == char(0x03)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[1] == char(0xFE)); - assert(n[0] == char(0xFF)); - assert(n[3] == char(0x10)); - assert(n[2] == char(0x05)); - assert(n[5] == char(0xDC)); - assert(n[4] == char(0x03)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[1] == char(0xFE)); - assert(n[0] == char(0xFF)); - assert(n[3] == char(0x04)); - assert(n[2] == char(0x53)); - assert(n[5] == char(0xDC)); - assert(n[4] == char(0x03)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+6, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[1] == char(0xFE)); - assert(n[0] == char(0xFF)); - assert(n[3] == char(0x00)); - assert(n[2] == char(0x56)); - assert(n[5] == char(0xDC)); - assert(n[4] == char(0x03)); - } -} +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// template +// class codecvt_utf16 +// : public codecvt +// { +// // unspecified +// }; + +// result +// out(stateT& state, +// const internT* from, const internT* from_end, const internT*& from_next, +// externT* to, externT* to_end, externT*& to_next) const; + +// UNSUPPORTED: short_wchar + +#include +#include + +int main() +{ + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0x40003; + char n[4] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xD8)); + assert(n[1] == char(0xC0)); + assert(n[2] == char(0xDC)); + assert(n[3] == char(0x03)); + + w = 0x1005; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0x10)); + assert(n[1] == char(0x05)); + assert(n[2] == char(0xDC)); + assert(n[3] == char(0x03)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0x04)); + assert(n[1] == char(0x53)); + assert(n[2] == char(0xDC)); + assert(n[3] == char(0x03)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0x00)); + assert(n[1] == char(0x56)); + assert(n[2] == char(0xDC)); + assert(n[3] == char(0x03)); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0x40003; + char n[4] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x1005; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0x04)); + assert(n[1] == char(0x53)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0x00)); + assert(n[1] == char(0x56)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0x40003; + char n[6] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(n[0] == char(0xFE)); + assert(n[1] == char(0xFF)); + assert(n[2] == char(0xD8)); + assert(n[3] == char(0xC0)); + assert(n[4] == char(0xDC)); + assert(n[5] == char(0x03)); + + w = 0x1005; + r = c.out(m, &w, &w+1, wp, n, n+6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xFE)); + assert(n[1] == char(0xFF)); + assert(n[2] == char(0x10)); + assert(n[3] == char(0x05)); + assert(n[4] == char(0xDC)); + assert(n[5] == char(0x03)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xFE)); + assert(n[1] == char(0xFF)); + assert(n[2] == char(0x04)); + assert(n[3] == char(0x53)); + assert(n[4] == char(0xDC)); + assert(n[5] == char(0x03)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xFE)); + assert(n[1] == char(0xFF)); + assert(n[2] == char(0x00)); + assert(n[3] == char(0x56)); + assert(n[4] == char(0xDC)); + assert(n[5] == char(0x03)); + } + + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0x40003; + char n[4] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[1] == char(0xD8)); + assert(n[0] == char(0xC0)); + assert(n[3] == char(0xDC)); + assert(n[2] == char(0x03)); + + w = 0x1005; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[1] == char(0x10)); + assert(n[0] == char(0x05)); + assert(n[3] == char(0xDC)); + assert(n[2] == char(0x03)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[1] == char(0x04)); + assert(n[0] == char(0x53)); + assert(n[3] == char(0xDC)); + assert(n[2] == char(0x03)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[1] == char(0x00)); + assert(n[0] == char(0x56)); + assert(n[3] == char(0xDC)); + assert(n[2] == char(0x03)); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0x40003; + char n[4] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[1] == char(0)); + assert(n[0] == char(0)); + assert(n[3] == char(0)); + assert(n[2] == char(0)); + + w = 0x1005; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[1] == char(0)); + assert(n[0] == char(0)); + assert(n[3] == char(0)); + assert(n[2] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[1] == char(0x04)); + assert(n[0] == char(0x53)); + assert(n[3] == char(0)); + assert(n[2] == char(0)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[1] == char(0x00)); + assert(n[0] == char(0x56)); + assert(n[3] == char(0)); + assert(n[2] == char(0)); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0x40003; + char n[6] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(n[1] == char(0xFE)); + assert(n[0] == char(0xFF)); + assert(n[3] == char(0xD8)); + assert(n[2] == char(0xC0)); + assert(n[5] == char(0xDC)); + assert(n[4] == char(0x03)); + + w = 0x1005; + r = c.out(m, &w, &w+1, wp, n, n+6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[1] == char(0xFE)); + assert(n[0] == char(0xFF)); + assert(n[3] == char(0x10)); + assert(n[2] == char(0x05)); + assert(n[5] == char(0xDC)); + assert(n[4] == char(0x03)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[1] == char(0xFE)); + assert(n[0] == char(0xFF)); + assert(n[3] == char(0x04)); + assert(n[2] == char(0x53)); + assert(n[5] == char(0xDC)); + assert(n[4] == char(0x03)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[1] == char(0xFE)); + assert(n[0] == char(0xFF)); + assert(n[3] == char(0x00)); + assert(n[2] == char(0x56)); + assert(n[5] == char(0xDC)); + assert(n[4] == char(0x03)); + } +} Index: test/std/localization/locale.stdcvt/codecvt_utf16_out_short_wchar.pass.cpp =================================================================== --- test/std/localization/locale.stdcvt/codecvt_utf16_out_short_wchar.pass.cpp +++ test/std/localization/locale.stdcvt/codecvt_utf16_out_short_wchar.pass.cpp @@ -0,0 +1,233 @@ +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// template +// class codecvt_utf16 +// : public codecvt +// { +// // unspecified +// }; + +// result +// out(stateT& state, +// const internT* from, const internT* from_end, const internT*& from_next, +// externT* to, externT* to_end, externT*& to_next) const; + +// REQUIRES: short_wchar + +#include +#include + +int main() +{ + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0x1005; + char n[2] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+2, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0x10)); + assert(n[1] == char(0x05)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+2, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0x04)); + assert(n[1] == char(0x53)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+2, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0x00)); + assert(n[1] == char(0x56)); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0x1005; + char n[2] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+2, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+2, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0x04)); + assert(n[1] == char(0x53)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+2, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0x00)); + assert(n[1] == char(0x56)); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0x1005; + char n[4] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xFE)); + assert(n[1] == char(0xFF)); + assert(n[2] == char(0x10)); + assert(n[3] == char(0x05)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xFE)); + assert(n[1] == char(0xFF)); + assert(n[2] == char(0x04)); + assert(n[3] == char(0x53)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xFE)); + assert(n[1] == char(0xFF)); + assert(n[2] == char(0x00)); + assert(n[3] == char(0x56)); + } + + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0x1005; + char n[2] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+2, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[1] == char(0x10)); + assert(n[0] == char(0x05)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+2, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[1] == char(0x04)); + assert(n[0] == char(0x53)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+2, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[1] == char(0x00)); + assert(n[0] == char(0x56)); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0x1005; + char n[2] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+2, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[1] == char(0)); + assert(n[0] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+2, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[1] == char(0x04)); + assert(n[0] == char(0x53)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+2, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[1] == char(0x00)); + assert(n[0] == char(0x56)); + } + { + typedef std::codecvt_utf16 C; + C c; + wchar_t w = 0x1005; + char n[4] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[1] == char(0xFE)); + assert(n[0] == char(0xFF)); + assert(n[3] == char(0x10)); + assert(n[2] == char(0x05)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[1] == char(0xFE)); + assert(n[0] == char(0xFF)); + assert(n[3] == char(0x04)); + assert(n[2] == char(0x53)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[1] == char(0xFE)); + assert(n[0] == char(0xFF)); + assert(n[3] == char(0x00)); + assert(n[2] == char(0x56)); + } +} Index: test/std/localization/locale.stdcvt/codecvt_utf8_in.pass.cpp =================================================================== --- test/std/localization/locale.stdcvt/codecvt_utf8_in.pass.cpp +++ test/std/localization/locale.stdcvt/codecvt_utf8_in.pass.cpp @@ -1,360 +1,362 @@ -//===----------------------------------------------------------------------===// -// -// The LLVM Compiler Infrastructure -// -// This file is dual licensed under the MIT and the University of Illinois Open -// Source Licenses. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - -// - -// template -// class codecvt_utf8 -// : public codecvt -// { -// // unspecified -// }; - -// result -// in(stateT& state, -// const externT* from, const externT* from_end, const externT*& from_next, -// internT* to, internT* to_end, internT*& to_next) const; - -#include -#include - -int main() -{ - { - typedef std::codecvt_utf8 C; - C c; - wchar_t w = 0; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - wchar_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(w == 0x40003); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.in(m, n, n+3, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+3); - assert(w == 0x1005); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x56); - r = c.in(m, n, n+1, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(w == 0x56); - } - { - typedef std::codecvt_utf8 C; - C c; - wchar_t w = 0; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - wchar_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.in(m, n, n+3, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x56); - r = c.in(m, n, n+1, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(w == 0x56); - } - { - typedef std::codecvt_utf8 C; - C c; - wchar_t w = 0; - char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; - wchar_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+7, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+7); - assert(w == 0x40003); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.in(m, n, n+3, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+3); - assert(w == 0x1005); - - n[0] = char(0xEF); - n[1] = char(0xBB); - n[2] = char(0xBF); - n[3] = char(0xD1); - n[4] = char(0x93); - r = c.in(m, n, n+5, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+5); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x56); - r = c.in(m, n, n+1, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(w == 0x56); - } - { - typedef std::codecvt_utf8 C; - C c; - char32_t w = 0; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - char32_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(w == 0x40003); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.in(m, n, n+3, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+3); - assert(w == 0x1005); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x56); - r = c.in(m, n, n+1, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(w == 0x56); - } - { - typedef std::codecvt_utf8 C; - C c; - char32_t w = 0; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - char32_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.in(m, n, n+3, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x56); - r = c.in(m, n, n+1, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(w == 0x56); - } - { - typedef std::codecvt_utf8 C; - C c; - char32_t w = 0; - char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; - char32_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+7, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+7); - assert(w == 0x40003); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.in(m, n, n+3, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+3); - assert(w == 0x1005); - - n[0] = char(0xEF); - n[1] = char(0xBB); - n[2] = char(0xBF); - n[3] = char(0xD1); - n[4] = char(0x93); - r = c.in(m, n, n+5, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+5); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x56); - r = c.in(m, n, n+1, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(w == 0x56); - } - { - typedef std::codecvt_utf8 C; - C c; - char16_t w = 0; - char n[3] = {char(0xE1), char(0x80), char(0x85)}; - char16_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+3, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+3); - assert(w == 0x1005); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x56); - r = c.in(m, n, n+1, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(w == 0x56); - } - { - typedef std::codecvt_utf8 C; - C c; - char16_t w = 0; - char n[3] = {char(0xE1), char(0x80), char(0x85)}; - char16_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+3, np, &w, &w+1, wp); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(w == 0); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x56); - r = c.in(m, n, n+1, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(w == 0x56); - } - { - typedef std::codecvt_utf8 C; - C c; - char16_t w = 0; - char n[6] = {char(0xEF), char(0xBB), char(0xBF), char(0xE1), char(0x80), char(0x85)}; - char16_t* wp = nullptr; - std::mbstate_t m; - const char* np = nullptr; - std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+6); - assert(w == 0x1005); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.in(m, n, n+2, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(w == 0x453); - - w = 0x56; - n[0] = char(0x56); - r = c.in(m, n, n+1, np, &w, &w+1, wp); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(w == 0x56); - } -} +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// template +// class codecvt_utf8 +// : public codecvt +// { +// // unspecified +// }; + +// result +// in(stateT& state, +// const externT* from, const externT* from_end, const externT*& from_next, +// internT* to, internT* to_end, internT*& to_next) const; + +// UNSUPPORTED: short_wchar + +#include +#include + +int main() +{ + { + typedef std::codecvt_utf8 C; + C c; + wchar_t w = 0; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(w == 0x40003); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.in(m, n, n+3, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+3); + assert(w == 0x1005); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + wchar_t w = 0; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.in(m, n, n+3, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + wchar_t w = 0; + char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+7, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+7); + assert(w == 0x40003); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.in(m, n, n+3, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+3); + assert(w == 0x1005); + + n[0] = char(0xEF); + n[1] = char(0xBB); + n[2] = char(0xBF); + n[3] = char(0xD1); + n[4] = char(0x93); + r = c.in(m, n, n+5, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+5); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + char32_t w = 0; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(w == 0x40003); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.in(m, n, n+3, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+3); + assert(w == 0x1005); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + char32_t w = 0; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.in(m, n, n+3, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + char32_t w = 0; + char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+7, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+7); + assert(w == 0x40003); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.in(m, n, n+3, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+3); + assert(w == 0x1005); + + n[0] = char(0xEF); + n[1] = char(0xBB); + n[2] = char(0xBF); + n[3] = char(0xD1); + n[4] = char(0x93); + r = c.in(m, n, n+5, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+5); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + char16_t w = 0; + char n[3] = {char(0xE1), char(0x80), char(0x85)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+3, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+3); + assert(w == 0x1005); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + char16_t w = 0; + char n[3] = {char(0xE1), char(0x80), char(0x85)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+3, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + char16_t w = 0; + char n[6] = {char(0xEF), char(0xBB), char(0xBF), char(0xE1), char(0x80), char(0x85)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(w == 0x1005); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } +} Index: test/std/localization/locale.stdcvt/codecvt_utf8_in_short_wchar.pass.cpp =================================================================== --- test/std/localization/locale.stdcvt/codecvt_utf8_in_short_wchar.pass.cpp +++ test/std/localization/locale.stdcvt/codecvt_utf8_in_short_wchar.pass.cpp @@ -0,0 +1,338 @@ +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// template +// class codecvt_utf8 +// : public codecvt +// { +// // unspecified +// }; + +// result +// in(stateT& state, +// const externT* from, const externT* from_end, const externT*& from_next, +// internT* to, internT* to_end, internT*& to_next) const; + +// REQUIRES: short_wchar + +#include +#include + +int main() +{ + { + typedef std::codecvt_utf8 C; + C c; + wchar_t w = 0; + char n[3] = {char(0xE1), char(0x80), char(0x85)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + std::codecvt_base::result r = c.in(m, n, n+3, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+3); + assert(w == 0x1005); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + wchar_t w = 0; + char n[3] = {char(0xE1), char(0x80), char(0x85)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+3, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + wchar_t w = 0; + char n[6] = {char(0xEF), char(0xBB), char(0xBF), char(0xE1), char(0x80), char(0x85)}; + wchar_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(w == 0x1005); + + n[0] = char(0xEF); + n[1] = char(0xBB); + n[2] = char(0xBF); + n[3] = char(0xD1); + n[4] = char(0x93); + r = c.in(m, n, n+5, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+5); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + char32_t w = 0; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(w == 0x40003); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.in(m, n, n+3, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+3); + assert(w == 0x1005); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + char32_t w = 0; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+4, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.in(m, n, n+3, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + char32_t w = 0; + char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; + char32_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+7, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+7); + assert(w == 0x40003); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.in(m, n, n+3, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+3); + assert(w == 0x1005); + + n[0] = char(0xEF); + n[1] = char(0xBB); + n[2] = char(0xBF); + n[3] = char(0xD1); + n[4] = char(0x93); + r = c.in(m, n, n+5, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+5); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + char16_t w = 0; + char n[3] = {char(0xE1), char(0x80), char(0x85)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+3, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+3); + assert(w == 0x1005); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + char16_t w = 0; + char n[3] = {char(0xE1), char(0x80), char(0x85)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+3, np, &w, &w+1, wp); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(w == 0); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } + { + typedef std::codecvt_utf8 C; + C c; + char16_t w = 0; + char n[6] = {char(0xEF), char(0xBB), char(0xBF), char(0xE1), char(0x80), char(0x85)}; + char16_t* wp = nullptr; + std::mbstate_t m; + const char* np = nullptr; + std::codecvt_base::result r = c.in(m, n, n+6, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(w == 0x1005); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.in(m, n, n+2, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(w == 0x453); + + w = 0x56; + n[0] = char(0x56); + r = c.in(m, n, n+1, np, &w, &w+1, wp); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(w == 0x56); + } +} Index: test/std/localization/locale.stdcvt/codecvt_utf8_length.pass.cpp =================================================================== --- test/std/localization/locale.stdcvt/codecvt_utf8_length.pass.cpp +++ test/std/localization/locale.stdcvt/codecvt_utf8_length.pass.cpp @@ -1,244 +1,246 @@ -//===----------------------------------------------------------------------===// -// -// The LLVM Compiler Infrastructure -// -// This file is dual licensed under the MIT and the University of Illinois Open -// Source Licenses. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - -// - -// template -// class codecvt_utf8 -// : public codecvt -// { -// // unspecified -// }; - -// int length(stateT& state, const externT* from, const externT* from_end, -// size_t max) const; - -#include -#include - -int main() -{ - { - typedef std::codecvt_utf8 C; - C c; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 1); - assert(r == 4); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.length(m, n, n+3, 2); - assert(r == 3); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.length(m, n, n+2, 3); - assert(r == 2); - - n[0] = char(0x56); - r = c.length(m, n, n+1, 3); - assert(r == 1); - } - { - typedef std::codecvt_utf8 C; - C c; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 1); - assert(r == 0); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.length(m, n, n+3, 2); - assert(r == 0); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.length(m, n, n+2, 3); - assert(r == 2); - - n[0] = char(0x56); - r = c.length(m, n, n+1, 3); - assert(r == 1); - } - { - typedef std::codecvt_utf8 C; - C c; - char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; - std::mbstate_t m; - int r = c.length(m, n, n+7, 1); - assert(r == 7); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.length(m, n, n+3, 2); - assert(r == 3); - - n[0] = char(0xEF); - n[1] = char(0xBB); - n[2] = char(0xBF); - n[3] = char(0xD1); - n[4] = char(0x93); - r = c.length(m, n, n+5, 3); - assert(r == 5); - - n[0] = char(0x56); - r = c.length(m, n, n+1, 3); - assert(r == 1); - } - { - typedef std::codecvt_utf8 C; - C c; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 1); - assert(r == 4); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.length(m, n, n+3, 2); - assert(r == 3); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.length(m, n, n+2, 3); - assert(r == 2); - - n[0] = char(0x56); - r = c.length(m, n, n+1, 3); - assert(r == 1); - } - { - typedef std::codecvt_utf8 C; - C c; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 1); - assert(r == 0); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.length(m, n, n+3, 2); - assert(r == 0); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.length(m, n, n+2, 3); - assert(r == 2); - - n[0] = char(0x56); - r = c.length(m, n, n+1, 3); - assert(r == 1); - } - { - typedef std::codecvt_utf8 C; - C c; - char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; - std::mbstate_t m; - int r = c.length(m, n, n+7, 1); - assert(r == 7); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.length(m, n, n+3, 2); - assert(r == 3); - - n[0] = char(0xEF); - n[1] = char(0xBB); - n[2] = char(0xBF); - n[3] = char(0xD1); - n[4] = char(0x93); - r = c.length(m, n, n+5, 3); - assert(r == 5); - - n[0] = char(0x56); - r = c.length(m, n, n+1, 3); - assert(r == 1); - } - { - typedef std::codecvt_utf8 C; - C c; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 1); - assert(r == 0); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.length(m, n, n+3, 2); - assert(r == 3); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.length(m, n, n+2, 3); - assert(r == 2); - - n[0] = char(0x56); - r = c.length(m, n, n+1, 3); - assert(r == 1); - } - { - typedef std::codecvt_utf8 C; - C c; - char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; - std::mbstate_t m; - int r = c.length(m, n, n+4, 1); - assert(r == 0); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.length(m, n, n+3, 2); - assert(r == 0); - - n[0] = char(0xD1); - n[1] = char(0x93); - r = c.length(m, n, n+2, 3); - assert(r == 2); - - n[0] = char(0x56); - r = c.length(m, n, n+1, 3); - assert(r == 1); - } - { - typedef std::codecvt_utf8 C; - C c; - char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; - std::mbstate_t m; - int r = c.length(m, n, n+7, 1); - assert(r == 3); - - n[0] = char(0xE1); - n[1] = char(0x80); - n[2] = char(0x85); - r = c.length(m, n, n+3, 2); - assert(r == 3); - - n[0] = char(0xEF); - n[1] = char(0xBB); - n[2] = char(0xBF); - n[3] = char(0xD1); - n[4] = char(0x93); - r = c.length(m, n, n+5, 3); - assert(r == 5); - - n[0] = char(0x56); - r = c.length(m, n, n+1, 3); - assert(r == 1); - } -} +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// template +// class codecvt_utf8 +// : public codecvt +// { +// // unspecified +// }; + +// int length(stateT& state, const externT* from, const externT* from_end, +// size_t max) const; + +// UNSUPPORTED: short_wchar + +#include +#include + +int main() +{ + { + typedef std::codecvt_utf8 C; + C c; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 1); + assert(r == 4); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 3); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.length(m, n, n+2, 3); + assert(r == 2); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 1); + assert(r == 0); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 0); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.length(m, n, n+2, 3); + assert(r == 2); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; + std::mbstate_t m; + int r = c.length(m, n, n+7, 1); + assert(r == 7); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 3); + + n[0] = char(0xEF); + n[1] = char(0xBB); + n[2] = char(0xBF); + n[3] = char(0xD1); + n[4] = char(0x93); + r = c.length(m, n, n+5, 3); + assert(r == 5); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 1); + assert(r == 4); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 3); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.length(m, n, n+2, 3); + assert(r == 2); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 1); + assert(r == 0); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 0); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.length(m, n, n+2, 3); + assert(r == 2); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; + std::mbstate_t m; + int r = c.length(m, n, n+7, 1); + assert(r == 7); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 3); + + n[0] = char(0xEF); + n[1] = char(0xBB); + n[2] = char(0xBF); + n[3] = char(0xD1); + n[4] = char(0x93); + r = c.length(m, n, n+5, 3); + assert(r == 5); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 1); + assert(r == 0); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 3); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.length(m, n, n+2, 3); + assert(r == 2); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 1); + assert(r == 0); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 0); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.length(m, n, n+2, 3); + assert(r == 2); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; + std::mbstate_t m; + int r = c.length(m, n, n+7, 1); + assert(r == 3); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 3); + + n[0] = char(0xEF); + n[1] = char(0xBB); + n[2] = char(0xBF); + n[3] = char(0xD1); + n[4] = char(0x93); + r = c.length(m, n, n+5, 3); + assert(r == 5); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } +} Index: test/std/localization/locale.stdcvt/codecvt_utf8_length_short_wchar.pass.cpp =================================================================== --- test/std/localization/locale.stdcvt/codecvt_utf8_length_short_wchar.pass.cpp +++ test/std/localization/locale.stdcvt/codecvt_utf8_length_short_wchar.pass.cpp @@ -0,0 +1,234 @@ +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// template +// class codecvt_utf8 +// : public codecvt +// { +// // unspecified +// }; + +// int length(stateT& state, const externT* from, const externT* from_end, +// size_t max) const; + +// REQUIRES: short_wchar + +#include +#include + +int main() +{ + { + typedef std::codecvt_utf8 C; + C c; + char n[3] = {char(0xE1), char(0x80), char(0x85)}; + std::mbstate_t m; + int r = c.length(m, n, n+3, 2); + assert(r == 3); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.length(m, n, n+2, 3); + assert(r == 2); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[3] = {char(0xE1), char(0x80), char(0x85)}; + std::mbstate_t m; + int r = c.length(m, n, n+3, 2); + assert(r == 0); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.length(m, n, n+2, 3); + assert(r == 2); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[6] = {char(0xEF), char(0xBB), char(0xBF), char(0xE1), char(0x80), char(0x85)}; + std::mbstate_t m; + int r = c.length(m, n, n+6, 2); + assert(r == 6); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 3); + + n[0] = char(0xEF); + n[1] = char(0xBB); + n[2] = char(0xBF); + n[3] = char(0xD1); + n[4] = char(0x93); + r = c.length(m, n, n+5, 3); + assert(r == 5); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 1); + assert(r == 4); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 3); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.length(m, n, n+2, 3); + assert(r == 2); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 1); + assert(r == 0); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 0); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.length(m, n, n+2, 3); + assert(r == 2); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; + std::mbstate_t m; + int r = c.length(m, n, n+7, 1); + assert(r == 7); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 3); + + n[0] = char(0xEF); + n[1] = char(0xBB); + n[2] = char(0xBF); + n[3] = char(0xD1); + n[4] = char(0x93); + r = c.length(m, n, n+5, 3); + assert(r == 5); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 1); + assert(r == 0); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 3); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.length(m, n, n+2, 3); + assert(r == 2); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[4] = {char(0xF1), char(0x80), char(0x80), char(0x83)}; + std::mbstate_t m; + int r = c.length(m, n, n+4, 1); + assert(r == 0); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 0); + + n[0] = char(0xD1); + n[1] = char(0x93); + r = c.length(m, n, n+2, 3); + assert(r == 2); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } + { + typedef std::codecvt_utf8 C; + C c; + char n[7] = {char(0xEF), char(0xBB), char(0xBF), char(0xF1), char(0x80), char(0x80), char(0x83)}; + std::mbstate_t m; + int r = c.length(m, n, n+7, 1); + assert(r == 3); + + n[0] = char(0xE1); + n[1] = char(0x80); + n[2] = char(0x85); + r = c.length(m, n, n+3, 2); + assert(r == 3); + + n[0] = char(0xEF); + n[1] = char(0xBB); + n[2] = char(0xBF); + n[3] = char(0xD1); + n[4] = char(0x93); + r = c.length(m, n, n+5, 3); + assert(r == 5); + + n[0] = char(0x56); + r = c.length(m, n, n+1, 3); + assert(r == 1); + } +} Index: test/std/localization/locale.stdcvt/codecvt_utf8_out.pass.cpp =================================================================== --- test/std/localization/locale.stdcvt/codecvt_utf8_out.pass.cpp +++ test/std/localization/locale.stdcvt/codecvt_utf8_out.pass.cpp @@ -1,456 +1,458 @@ -//===----------------------------------------------------------------------===// -// -// The LLVM Compiler Infrastructure -// -// This file is dual licensed under the MIT and the University of Illinois Open -// Source Licenses. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - -// - -// template -// class codecvt_utf8 -// : public codecvt -// { -// // unspecified -// }; - -// result -// out(stateT& state, -// const internT* from, const internT* from_end, const internT*& from_next, -// externT* to, externT* to_end, externT*& to_next) const; - -#include -#include - -int main() -{ - { - typedef std::codecvt_utf8 C; - C c; - wchar_t w = 0x40003; - char n[4] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xF1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x80)); - assert(n[3] == char(0x83)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+3); - assert(n[0] == char(0xE1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0x83)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0x83)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0x83)); - } - { - typedef std::codecvt_utf8 C; - C c; - wchar_t w = 0x40003; - char n[4] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[0] == char(0)); - assert(n[1] == char(0)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[0] == char(0)); - assert(n[1] == char(0)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - } - { - typedef std::codecvt_utf8 C; - C c; - wchar_t w = 0x40003; - char n[7] = {0}; - const wchar_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+7); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xF1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x80)); - assert(n[6] == char(0x83)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+6); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xE1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0x83)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+5); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xD1)); - assert(n[4] == char(0x93)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0x83)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0x56)); - assert(n[4] == char(0x93)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0x83)); - } - { - typedef std::codecvt_utf8 C; - C c; - char32_t w = 0x40003; - char n[4] = {0}; - const char32_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xF1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x80)); - assert(n[3] == char(0x83)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+3); - assert(n[0] == char(0xE1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0x83)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0x83)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0x83)); - } - { - typedef std::codecvt_utf8 C; - C c; - char32_t w = 0x40003; - char n[4] = {0}; - const char32_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[0] == char(0)); - assert(n[1] == char(0)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[0] == char(0)); - assert(n[1] == char(0)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - } - { - typedef std::codecvt_utf8 C; - C c; - char32_t w = 0x40003; - char n[7] = {0}; - const char32_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+7); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xF1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x80)); - assert(n[6] == char(0x83)); - - w = 0x1005; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+6); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xE1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0x83)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+5); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xD1)); - assert(n[4] == char(0x93)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0x83)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0x56)); - assert(n[4] == char(0x93)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0x83)); - } - { - typedef std::codecvt_utf8 C; - C c; - char16_t w = 0x1005; - char n[4] = {0}; - const char16_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+3); - assert(n[0] == char(0xE1)); - assert(n[1] == char(0x80)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0x85)); - assert(n[3] == char(0)); - } - { - typedef std::codecvt_utf8 C; - C c; - char16_t w = 0x1005; - char n[4] = {0}; - const char16_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::error); - assert(wp == &w); - assert(np == n); - assert(n[0] == char(0)); - assert(n[1] == char(0)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+2); - assert(n[0] == char(0xD1)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+4, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+1); - assert(n[0] == char(0x56)); - assert(n[1] == char(0x93)); - assert(n[2] == char(0)); - assert(n[3] == char(0)); - } - { - typedef std::codecvt_utf8 C; - C c; - char16_t w = 0x1005; - char n[7] = {0}; - const char16_t* wp = nullptr; - std::mbstate_t m; - char* np = nullptr; - std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+6); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xE1)); - assert(n[4] == char(0x80)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0)); - - w = 0x453; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+5); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0xD1)); - assert(n[4] == char(0x93)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0)); - - w = 0x56; - r = c.out(m, &w, &w+1, wp, n, n+7, np); - assert(r == std::codecvt_base::ok); - assert(wp == &w+1); - assert(np == n+4); - assert(n[0] == char(0xEF)); - assert(n[1] == char(0xBB)); - assert(n[2] == char(0xBF)); - assert(n[3] == char(0x56)); - assert(n[4] == char(0x93)); - assert(n[5] == char(0x85)); - assert(n[6] == char(0)); - } -} +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// template +// class codecvt_utf8 +// : public codecvt +// { +// // unspecified +// }; + +// result +// out(stateT& state, +// const internT* from, const internT* from_end, const internT*& from_next, +// externT* to, externT* to_end, externT*& to_next) const; + +// UNSUPPORTED: short_wchar + +#include +#include + +int main() +{ + { + typedef std::codecvt_utf8 C; + C c; + wchar_t w = 0x40003; + char n[4] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xF1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x80)); + assert(n[3] == char(0x83)); + + w = 0x1005; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+3); + assert(n[0] == char(0xE1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0x83)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0x83)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0x83)); + } + { + typedef std::codecvt_utf8 C; + C c; + wchar_t w = 0x40003; + char n[4] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x1005; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + } + { + typedef std::codecvt_utf8 C; + C c; + wchar_t w = 0x40003; + char n[7] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+7); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xF1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x80)); + assert(n[6] == char(0x83)); + + w = 0x1005; + r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xE1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0x83)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+5); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xD1)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0x83)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0x56)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0x83)); + } + { + typedef std::codecvt_utf8 C; + C c; + char32_t w = 0x40003; + char n[4] = {0}; + const char32_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xF1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x80)); + assert(n[3] == char(0x83)); + + w = 0x1005; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+3); + assert(n[0] == char(0xE1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0x83)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0x83)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0x83)); + } + { + typedef std::codecvt_utf8 C; + C c; + char32_t w = 0x40003; + char n[4] = {0}; + const char32_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x1005; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + } + { + typedef std::codecvt_utf8 C; + C c; + char32_t w = 0x40003; + char n[7] = {0}; + const char32_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+7); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xF1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x80)); + assert(n[6] == char(0x83)); + + w = 0x1005; + r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xE1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0x83)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+5); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xD1)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0x83)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0x56)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0x83)); + } + { + typedef std::codecvt_utf8 C; + C c; + char16_t w = 0x1005; + char n[4] = {0}; + const char16_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+3); + assert(n[0] == char(0xE1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0)); + } + { + typedef std::codecvt_utf8 C; + C c; + char16_t w = 0x1005; + char n[4] = {0}; + const char16_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + } + { + typedef std::codecvt_utf8 C; + C c; + char16_t w = 0x1005; + char n[7] = {0}; + const char16_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xE1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+5); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xD1)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0x56)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0)); + } +} Index: test/std/localization/locale.stdcvt/codecvt_utf8_out_short_wchar.pass.cpp =================================================================== --- test/std/localization/locale.stdcvt/codecvt_utf8_out_short_wchar.pass.cpp +++ test/std/localization/locale.stdcvt/codecvt_utf8_out_short_wchar.pass.cpp @@ -0,0 +1,413 @@ +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// template +// class codecvt_utf8 +// : public codecvt +// { +// // unspecified +// }; + +// result +// out(stateT& state, +// const internT* from, const internT* from_end, const internT*& from_next, +// externT* to, externT* to_end, externT*& to_next) const; + +// REQUIRES: short_wchar + +#include +#include + +int main() +{ + { + typedef std::codecvt_utf8 C; + C c; + wchar_t w = 0x1005; + char n[3] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+3, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+3); + assert(n[0] == char(0xE1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x85)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+3, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+3, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + } + { + typedef std::codecvt_utf8 C; + C c; + wchar_t w = 0x1005; + char n[2] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+2, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+2, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+2, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + } + { + typedef std::codecvt_utf8 C; + C c; + wchar_t w = 0x1005; + char n[6] = {0}; + const wchar_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xE1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x85)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+5); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xD1)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+6, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0x56)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + } + { + typedef std::codecvt_utf8 C; + C c; + char32_t w = 0x40003; + char n[4] = {0}; + const char32_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xF1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x80)); + assert(n[3] == char(0x83)); + + w = 0x1005; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+3); + assert(n[0] == char(0xE1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0x83)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0x83)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0x83)); + } + { + typedef std::codecvt_utf8 C; + C c; + char32_t w = 0x40003; + char n[4] = {0}; + const char32_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x1005; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + } + { + typedef std::codecvt_utf8 C; + C c; + char32_t w = 0x40003; + char n[7] = {0}; + const char32_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+7); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xF1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x80)); + assert(n[6] == char(0x83)); + + w = 0x1005; + r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xE1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0x83)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+5); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xD1)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0x83)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0x56)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0x83)); + } + { + typedef std::codecvt_utf8 C; + C c; + char16_t w = 0x1005; + char n[4] = {0}; + const char16_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+3); + assert(n[0] == char(0xE1)); + assert(n[1] == char(0x80)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0x85)); + assert(n[3] == char(0)); + } + { + typedef std::codecvt_utf8 C; + C c; + char16_t w = 0x1005; + char n[4] = {0}; + const char16_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::error); + assert(wp == &w); + assert(np == n); + assert(n[0] == char(0)); + assert(n[1] == char(0)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+2); + assert(n[0] == char(0xD1)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+4, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+1); + assert(n[0] == char(0x56)); + assert(n[1] == char(0x93)); + assert(n[2] == char(0)); + assert(n[3] == char(0)); + } + { + typedef std::codecvt_utf8 C; + C c; + char16_t w = 0x1005; + char n[7] = {0}; + const char16_t* wp = nullptr; + std::mbstate_t m; + char* np = nullptr; + std::codecvt_base::result r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+6); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xE1)); + assert(n[4] == char(0x80)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0)); + + w = 0x453; + r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+5); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0xD1)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0)); + + w = 0x56; + r = c.out(m, &w, &w+1, wp, n, n+7, np); + assert(r == std::codecvt_base::ok); + assert(wp == &w+1); + assert(np == n+4); + assert(n[0] == char(0xEF)); + assert(n[1] == char(0xBB)); + assert(n[2] == char(0xBF)); + assert(n[3] == char(0x56)); + assert(n[4] == char(0x93)); + assert(n[5] == char(0x85)); + assert(n[6] == char(0)); + } +} Index: test/std/localization/locales/locale.convenience/conversions/conversions.string/converted.pass.cpp =================================================================== --- test/std/localization/locales/locale.convenience/conversions/conversions.string/converted.pass.cpp +++ test/std/localization/locales/locale.convenience/conversions/conversions.string/converted.pass.cpp @@ -1,32 +1,34 @@ -//===----------------------------------------------------------------------===// -// -// The LLVM Compiler Infrastructure -// -// This file is dual licensed under the MIT and the University of Illinois Open -// Source Licenses. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - -// - -// wstring_convert - -// size_t converted() const; - -#include -#include -#include - -int main() -{ - typedef std::codecvt_utf8 Codecvt; - typedef std::wstring_convert Myconv; - Myconv myconv; - assert(myconv.converted() == 0); - std::string bs = myconv.to_bytes(L"\x40003"); - assert(myconv.converted() == 1); - bs = myconv.to_bytes(L"\x40003\x65"); - assert(myconv.converted() == 2); - std::wstring ws = myconv.from_bytes("\xF1\x80\x80\x83"); - assert(myconv.converted() == 4); -} +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// wstring_convert + +// size_t converted() const; + +// UNSUPPORTED: short_wchar + +#include +#include +#include + +int main() +{ + typedef std::codecvt_utf8 Codecvt; + typedef std::wstring_convert Myconv; + Myconv myconv; + assert(myconv.converted() == 0); + std::string bs = myconv.to_bytes(L"\x40003"); + assert(myconv.converted() == 1); + bs = myconv.to_bytes(L"\x40003\x65"); + assert(myconv.converted() == 2); + std::wstring ws = myconv.from_bytes("\xF1\x80\x80\x83"); + assert(myconv.converted() == 4); +} Index: test/std/localization/locales/locale.convenience/conversions/conversions.string/converted_short_wchar.pass.cpp =================================================================== --- test/std/localization/locales/locale.convenience/conversions/conversions.string/converted_short_wchar.pass.cpp +++ test/std/localization/locales/locale.convenience/conversions/conversions.string/converted_short_wchar.pass.cpp @@ -0,0 +1,34 @@ +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// wstring_convert + +// size_t converted() const; + +// REQUIRES: short_wchar + +#include +#include +#include + +int main() +{ + typedef std::codecvt_utf8 Codecvt; + typedef std::wstring_convert Myconv; + Myconv myconv; + assert(myconv.converted() == 0); + std::string bs = myconv.to_bytes(L"\x4003"); + assert(myconv.converted() == 1); + bs = myconv.to_bytes(L"\x4003\x65"); + assert(myconv.converted() == 2); + std::wstring ws = myconv.from_bytes("\xE4\x80\x83"); + assert(myconv.converted() == 3); +} Index: test/std/localization/locales/locale.convenience/conversions/conversions.string/from_bytes.pass.cpp =================================================================== --- test/std/localization/locales/locale.convenience/conversions/conversions.string/from_bytes.pass.cpp +++ test/std/localization/locales/locale.convenience/conversions/conversions.string/from_bytes.pass.cpp @@ -1,39 +1,41 @@ -//===----------------------------------------------------------------------===// -// -// The LLVM Compiler Infrastructure -// -// This file is dual licensed under the MIT and the University of Illinois Open -// Source Licenses. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - -// - -// wstring_convert - -// wide_string from_bytes(char byte); -// wide_string from_bytes(const char* ptr); -// wide_string from_bytes(const byte_string& str); -// wide_string from_bytes(const char* first, const char* last); - -#include -#include -#include - -int main() -{ - { - std::wstring_convert > myconv; - std::string bs("\xF1\x80\x80\x83"); - std::wstring ws = myconv.from_bytes('a'); - assert(ws == L"a"); - ws = myconv.from_bytes(bs.c_str()); - assert(ws == L"\x40003"); - ws = myconv.from_bytes(bs); - assert(ws == L"\x40003"); - ws = myconv.from_bytes(bs.data(), bs.data() + bs.size()); - assert(ws == L"\x40003"); - ws = myconv.from_bytes(""); - assert(ws.size() == 0); - } -} +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// wstring_convert + +// wide_string from_bytes(char byte); +// wide_string from_bytes(const char* ptr); +// wide_string from_bytes(const byte_string& str); +// wide_string from_bytes(const char* first, const char* last); + +// UNSUPPORTED: short_wchar + +#include +#include +#include + +int main() +{ + { + std::wstring_convert > myconv; + std::string bs("\xF1\x80\x80\x83"); + std::wstring ws = myconv.from_bytes('a'); + assert(ws == L"a"); + ws = myconv.from_bytes(bs.c_str()); + assert(ws == L"\x40003"); + ws = myconv.from_bytes(bs); + assert(ws == L"\x40003"); + ws = myconv.from_bytes(bs.data(), bs.data() + bs.size()); + assert(ws == L"\x40003"); + ws = myconv.from_bytes(""); + assert(ws.size() == 0); + } +} Index: test/std/localization/locales/locale.convenience/conversions/conversions.string/from_bytes_short_wchar.pass.cpp =================================================================== --- test/std/localization/locales/locale.convenience/conversions/conversions.string/from_bytes_short_wchar.pass.cpp +++ test/std/localization/locales/locale.convenience/conversions/conversions.string/from_bytes_short_wchar.pass.cpp @@ -0,0 +1,41 @@ +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// wstring_convert + +// wide_string from_bytes(char byte); +// wide_string from_bytes(const char* ptr); +// wide_string from_bytes(const byte_string& str); +// wide_string from_bytes(const char* first, const char* last); + +// REQUIRES: short_wchar + +#include +#include +#include + +int main() +{ + { + std::wstring_convert > myconv; + std::string bs("\xE4\x80\x83"); + std::wstring ws = myconv.from_bytes('a'); + assert(ws == L"a"); + ws = myconv.from_bytes(bs.c_str()); + assert(ws == L"\x4003"); + ws = myconv.from_bytes(bs); + assert(ws == L"\x4003"); + ws = myconv.from_bytes(bs.data(), bs.data() + bs.size()); + assert(ws == L"\x4003"); + ws = myconv.from_bytes(""); + assert(ws.size() == 0); + } +} Index: test/std/localization/locales/locale.convenience/conversions/conversions.string/to_bytes.pass.cpp =================================================================== --- test/std/localization/locales/locale.convenience/conversions/conversions.string/to_bytes.pass.cpp +++ test/std/localization/locales/locale.convenience/conversions/conversions.string/to_bytes.pass.cpp @@ -1,39 +1,41 @@ -//===----------------------------------------------------------------------===// -// -// The LLVM Compiler Infrastructure -// -// This file is dual licensed under the MIT and the University of Illinois Open -// Source Licenses. See LICENSE.TXT for details. -// -//===----------------------------------------------------------------------===// - -// - -// wstring_convert - -// byte_string to_bytes(Elem wchar); -// byte_string to_bytes(const Elem* wptr); -// byte_string to_bytes(const wide_string& wstr); -// byte_string to_bytes(const Elem* first, const Elem* last); - -#include -#include -#include - -int main() -{ - { - std::wstring_convert > myconv; - std::wstring ws(1, L'\x40003'); - std::string bs = myconv.to_bytes(ws[0]); - assert(bs == "\xF1\x80\x80\x83"); - bs = myconv.to_bytes(ws.c_str()); - assert(bs == "\xF1\x80\x80\x83"); - bs = myconv.to_bytes(ws); - assert(bs == "\xF1\x80\x80\x83"); - bs = myconv.to_bytes(ws.data(), ws.data() + ws.size()); - assert(bs == "\xF1\x80\x80\x83"); - bs = myconv.to_bytes(L""); - assert(bs.size() == 0); - } -} +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// wstring_convert + +// byte_string to_bytes(Elem wchar); +// byte_string to_bytes(const Elem* wptr); +// byte_string to_bytes(const wide_string& wstr); +// byte_string to_bytes(const Elem* first, const Elem* last); + +// UNSUPPORTED: short_wchar + +#include +#include +#include + +int main() +{ + { + std::wstring_convert > myconv; + std::wstring ws(1, L'\x40003'); + std::string bs = myconv.to_bytes(ws[0]); + assert(bs == "\xF1\x80\x80\x83"); + bs = myconv.to_bytes(ws.c_str()); + assert(bs == "\xF1\x80\x80\x83"); + bs = myconv.to_bytes(ws); + assert(bs == "\xF1\x80\x80\x83"); + bs = myconv.to_bytes(ws.data(), ws.data() + ws.size()); + assert(bs == "\xF1\x80\x80\x83"); + bs = myconv.to_bytes(L""); + assert(bs.size() == 0); + } +} Index: test/std/localization/locales/locale.convenience/conversions/conversions.string/to_bytes_short_wchar.pass.cpp =================================================================== --- test/std/localization/locales/locale.convenience/conversions/conversions.string/to_bytes_short_wchar.pass.cpp +++ test/std/localization/locales/locale.convenience/conversions/conversions.string/to_bytes_short_wchar.pass.cpp @@ -0,0 +1,41 @@ +//===----------------------------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is dual licensed under the MIT and the University of Illinois Open +// Source Licenses. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +// + +// wstring_convert + +// byte_string to_bytes(Elem wchar); +// byte_string to_bytes(const Elem* wptr); +// byte_string to_bytes(const wide_string& wstr); +// byte_string to_bytes(const Elem* first, const Elem* last); + +// REQUIRES: short_wchar + +#include +#include +#include + +int main() +{ + { + std::wstring_convert > myconv; + std::wstring ws(1, L'\x4003'); + std::string bs = myconv.to_bytes(ws[0]); + assert(bs == "\xE4\x80\x83"); + bs = myconv.to_bytes(ws.c_str()); + assert(bs == "\xE4\x80\x83"); + bs = myconv.to_bytes(ws); + assert(bs == "\xE4\x80\x83"); + bs = myconv.to_bytes(ws.data(), ws.data() + ws.size()); + assert(bs == "\xE4\x80\x83"); + bs = myconv.to_bytes(L""); + assert(bs.size() == 0); + } +}