From ce52a1176c580a619969805eba63cb9417e3b4b9 Mon Sep 17 00:00:00 2001 From: mutouyun Date: Sun, 7 Aug 2022 22:29:11 +0800 Subject: [PATCH] add: [codecvt] character set encoding conversion interface --- include/libimp/codecvt.h | 26 +++ src/libimp/codecvt.cpp | 6 + src/libimp/platform/posix/codecvt.h | 319 ++++++++++++++++++++++++++++ src/libimp/platform/win/codecvt.h | 51 +++++ test/test_imp_codecvt.cpp | 29 +++ 5 files changed, 431 insertions(+) create mode 100644 include/libimp/codecvt.h create mode 100644 src/libimp/codecvt.cpp create mode 100644 src/libimp/platform/posix/codecvt.h create mode 100644 src/libimp/platform/win/codecvt.h create mode 100644 test/test_imp_codecvt.cpp diff --git a/include/libimp/codecvt.h b/include/libimp/codecvt.h new file mode 100644 index 0000000..e867529 --- /dev/null +++ b/include/libimp/codecvt.h @@ -0,0 +1,26 @@ +/** + * @file libimp/codecvt.h + * @author mutouyun (orz@orzz.org) + * @brief Character set conversion interface + * @date 2022-08-07 + */ +#pragma once + +#include +#include + +#include "libimp/def.h" +#include "libimp/export.h" + +LIBIMP_NAMESPACE_BEG_ + +/** + * @brief The transform between local-character-set(UTF-8/GBK/...) and UTF-16/32 + * + * @param des The target string pointer can be nullptr + * @param dlen The target string length can be 0 + */ +template +LIBIMP_EXPORT std::size_t cvt_cstr(CharT const *src, std::size_t slen, CharU *des, std::size_t dlen) noexcept; + +LIBIMP_NAMESPACE_END_ diff --git a/src/libimp/codecvt.cpp b/src/libimp/codecvt.cpp new file mode 100644 index 0000000..b10c5b5 --- /dev/null +++ b/src/libimp/codecvt.cpp @@ -0,0 +1,6 @@ +#include "libimp/detect_plat.h" +#if defined(LIBIMP_OS_WIN) +#include "libimp/platform/win/codecvt.h" +#else +#include "libimp/platform/posix/codecvt.h" +#endif \ No newline at end of file diff --git a/src/libimp/platform/posix/codecvt.h b/src/libimp/platform/posix/codecvt.h new file mode 100644 index 0000000..0ba52c0 --- /dev/null +++ b/src/libimp/platform/posix/codecvt.h @@ -0,0 +1,319 @@ +/** + * @file libimp/platform/win/codecvt.h + * @author mutouyun (orz@orzz.org) + */ +#pragma once + +#include +#include + +#include "libimp/codecvt.h" +#include "libimp/log.h" + +LIBIMP_NAMESPACE_BEG_ + +/** + * @brief The transform between local-character-set(UTF-8/GBK/...) and UTF-16/32 + * + * Modified from UnicodeConverter + * Copyright (c) 2010. Jianhui Qin (http://blog.csdn.net/jhqin) + */ +namespace { + +/// @brief X-bit unicode transformation format +enum class ufmt { + utf8, + utf16, + utf32, +}; + +template +struct utf_compatible : std::false_type {}; + +template +struct utf_compatible::value && (sizeof(T) == 1)>> : std::true_type {}; + +template +struct utf_compatible::value && (sizeof(T) == 2)>> : std::true_type {}; + +template +struct utf_compatible::value && (sizeof(T) == 4)>> : std::true_type {}; + +template +constexpr bool utf_compatible_v = utf_compatible::value; + +/** + * @brief UTF-32 --> UTF-8 + */ +template +auto cvt_char(T src, U* des, std::size_t dlen) noexcept + -> std::enable_if_t && + utf_compatible_v, std::size_t> { + if (src == 0) return 0; + + constexpr std::uint8_t prefix[] = { + 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC + }; + constexpr std::uint32_t codeup[] = { + 0x80, // U+00000000 - U+0000007F + 0x800, // U+00000080 - U+000007FF + 0x10000, // U+00000800 - U+0000FFFF + 0x200000, // U+00010000 - U+001FFFFF + 0x4000000, // U+00200000 - U+03FFFFFF + 0x80000000 // U+04000000 - U+7FFFFFFF + }; + + std::size_t i, len = sizeof(codeup) / sizeof(std::uint32_t); + for(i = 0; i < len; ++i) { + if (static_cast(src) < codeup[i]) break; + } + if (i == len) return 0; // the src is invalid + + len = i + 1; + if (des != nullptr) { + if (dlen > i) for (; i > 0; --i) { + des[i] = static_cast((src & 0x3F) | 0x80); + src >>= 6; + } + des[0] = static_cast(src | prefix[len - 1]); + } + return len; +} + +/** + * @brief UTF-8 --> UTF-32 + */ +template +auto cvt_char(T const *src, std::size_t slen, U &des) noexcept + -> std::enable_if_t && + utf_compatible_v, std::size_t> { + if ((src == nullptr) || (*src) == 0) return 0; + if (slen == 0) return 0; + + std::uint8_t b = (std::uint8_t)*(src++); + + if (b < 0x80) { + des = b; + return 1; + } + + if (b < 0xC0 || b > 0xFD) return 0; // the src is invalid + + std::size_t len; + if (b < 0xE0) { + des = b & 0x1F; + len = 2; + } else if (b < 0xF0) { + des = b & 0x0F; + len = 3; + } else if (b < 0xF8) { + des = b & 0x07; + len = 4; + } else if (b < 0xFC) { + des = b & 0x03; + len = 5; + } else { + des = b & 0x01; + len = 6; + } + + if (slen < len) return 0; + std::size_t i = 1; + for(; i < len; ++i) { + b = *(src++); + if ((b < 0x80) || (b > 0xBF)) return 0; // the src is invalid + des = (des << 6) + (b & 0x3F); + } + return len; +} + +/** + * @brief UTF-32 --> UTF-16 + */ +template +auto cvt_char(T src, U *des, std::size_t dlen) noexcept + -> std::enable_if_t && + utf_compatible_v, std::size_t> { + if (src == 0) return 0; + + if (src <= 0xFFFF) { + if ((des != nullptr) && (dlen != 0)) { + (*des) = static_cast(src); + } + return 1; + } else if (src <= 0xEFFFF) { + if ((des != nullptr) && (dlen > 1)) { + des[0] = static_cast(0xD800 + (src >> 10) - 0x40); // high + des[1] = static_cast(0xDC00 + (src & 0x03FF)); // low + } + return 2; + } + return 0; +} + +/** + * @brief UTF-16 --> UTF-32 +*/ +template +auto cvt_char(T const *src, std::size_t slen, U &des) + -> std::enable_if_t && + utf_compatible_v, std::size_t> { + if ((src == nullptr) || (*src) == 0) return 0; + if (slen == 0) return 0; + + std::uint16_t w1 = src[0]; + if ((w1 >= 0xD800) && (w1 <= 0xDFFF)) { + if (w1 < 0xDC00) { + if (slen < 2) return 0; + std::uint16_t w2 = src[1]; + if ((w2 >= 0xDC00) && (w2 <= 0xDFFF)) { + des = (w2 & 0x03FF) + (((w1 & 0x03FF) + 0x40) << 10); + return 2; + } + } + return 0; // the src is invalid + } + des = w1; + return 1; +} + +/** + * @brief UTF-16 --> UTF-8 +*/ +template +auto cvt_char(T src, U *des, std::size_t dlen) noexcept + -> std::enable_if_t && + utf_compatible_v, std::size_t> { + // make utf-16 to utf-32 + std::uint32_t tmp; + if (cvt_char(&src, 1, tmp) != 1) return 0; + // make utf-32 to utf-8 + return cvt_char(tmp, des, dlen); +} + +/** + * @brief UTF-8 --> UTF-16 +*/ +template +auto cvt_char(T const *src, std::size_t slen, U &des) + -> std::enable_if_t && + utf_compatible_v, std::size_t> { + // make utf-8 to utf-32 + std::uint32_t tmp; + std::size_t len = cvt_char(src, slen, tmp); + if (len == 0) return 0; + // make utf-32 to utf-16 + if (cvt_char(tmp, &des, 1) != 1) return 0; + return len; +} + +/** + * @brief UTF-32 string --> UTF-8/16 string +*/ +template +auto cvt_cstr_utf(T const *src, std::size_t slen, U *des, std::size_t dlen) noexcept + -> std::enable_if_t && + (utf_compatible_v || utf_compatible_v), std::size_t> { + LIBIMP_LOG_(); + if ((src == nullptr) || ((*src) == 0) || (slen == 0)) { + log.error("source string is empty."); + return 0; + } + std::size_t num = 0, len = 0; + for (std::size_t i = 0; (i < slen) && ((*src) != 0); ++src, ++i) { + len = cvt_char(*src, des, dlen); + if (len == 0) return 0; + if (des != nullptr) { + des += len; + if (dlen < len) { + dlen = 0; + } else { + dlen -= len; + } + } + num += len; + } + return num; +} + +/** + * @brief UTF-8/16 string --> UTF-32 string +*/ +template +auto cvt_cstr_utf(T const *src, std::size_t slen, U *des, std::size_t dlen) noexcept + -> std::enable_if_t && + (utf_compatible_v || utf_compatible_v), std::size_t> { + LIBIMP_LOG_(); + if ((src == nullptr) || ((*src) == 0) || (slen == 0)) { + log.error("source string is empty."); + return 0; + } + std::size_t num = 0; + for (std::size_t i = 0; (i < slen) && ((*src) != 0);) { + std::uint32_t tmp; + std::size_t len = cvt_char(src, slen - i, tmp); + if (len == 0) return 0; + if ((des != nullptr) && (dlen > 0)) { + (*des) = tmp; + ++des; + dlen -= 1; + } + src += len; + i += len; + num += 1; + } + return num; +} + +/** + * @brief UTF-8/16 string --> UTF-16/8 string +*/ +template +auto cvt_cstr_utf(T const *src, std::size_t slen, U *des, std::size_t dlen) noexcept + -> std::enable_if_t<(utf_compatible_v && utf_compatible_v) || + (utf_compatible_v && utf_compatible_v), std::size_t> { + LIBIMP_LOG_(); + if ((src == nullptr) || ((*src) == 0) || (slen == 0)) { + log.error("source string is empty."); + return 0; + } + std::size_t num = 0; + for (std::size_t i = 0; (i < slen) && ((*src) != 0);) { + // make utf-x to utf-32 + std::uint32_t tmp; + std::size_t len = cvt_char(src, slen - i, tmp); + if (len == 0) return 0; + src += len; + i += len; + // make utf-32 to utf-y + len = cvt_char(tmp, des, dlen); + if (len == 0) return 0; + if (des != nullptr) { + des += len; + if (dlen < len) { + dlen = 0; + } else { + dlen -= len; + } + } + num += len; + } + return num; +} + +} // namespace + +template <> +std::size_t cvt_cstr(char const *src, std::size_t slen, wchar_t *des, std::size_t dlen) noexcept { + return cvt_cstr_utf(src, slen, des, dlen); +} + +template <> +std::size_t cvt_cstr(wchar_t const *src, std::size_t slen, char *des, std::size_t dlen) noexcept { + return cvt_cstr_utf(src, slen, des, dlen); +} + +LIBIMP_NAMESPACE_END_ diff --git a/src/libimp/platform/win/codecvt.h b/src/libimp/platform/win/codecvt.h new file mode 100644 index 0000000..86a545c --- /dev/null +++ b/src/libimp/platform/win/codecvt.h @@ -0,0 +1,51 @@ +/** + * @file libimp/platform/win/codecvt.h + * @author mutouyun (orz@orzz.org) + */ +#pragma once + +#include + +#include "libimp/codecvt.h" +#include "libimp/log.h" + +LIBIMP_NAMESPACE_BEG_ + +/** + * @brief The transform between local-character-set(UTF-8/GBK/...) and UTF-16/32 + * + * https://docs.microsoft.com/en-us/windows/win32/api/stringapiset/nf-stringapiset-multibytetowidechar + * https://docs.microsoft.com/en-us/windows/win32/api/stringapiset/nf-stringapiset-widechartomultibyte + * + * CP_ACP : The system default Windows ANSI code page. + * CP_MACCP : The current system Macintosh code page. + * CP_OEMCP : The current system OEM code page. + * CP_SYMBOL : Symbol code page (42). + * CP_THREAD_ACP: The Windows ANSI code page for the current thread. + * CP_UTF7 : UTF-7. Use this value only when forced by a 7-bit transport mechanism. Use of UTF-8 is preferred. + * CP_UTF8 : UTF-8. + */ + +template <> +std::size_t cvt_cstr(char const *src, std::size_t slen, wchar_t *des, std::size_t dlen) noexcept { + LIBIMP_LOG_(); + int r = ::MultiByteToWideChar(CP_UTF8, 0, src, (int)slen, des, (int)dlen); + if (r <= 0) { + log.error("MultiByteToWideChar fails. return = {}", ::GetLastError()); + return 0; + } + return (std::size_t)r; +} + +template <> +std::size_t cvt_cstr(wchar_t const *src, std::size_t slen, char *des, std::size_t dlen) noexcept { + LIBIMP_LOG_(); + int r = ::WideCharToMultiByte(CP_UTF8, 0, src, (int)slen, des, (int)dlen, nullptr, nullptr); + if (r <= 0) { + log.error("WideCharToMultiByte fails. return = {}", ::GetLastError()); + return 0; + } + return (std::size_t)r; +} + +LIBIMP_NAMESPACE_END_ diff --git a/test/test_imp_codecvt.cpp b/test/test_imp_codecvt.cpp new file mode 100644 index 0000000..722d58c --- /dev/null +++ b/test/test_imp_codecvt.cpp @@ -0,0 +1,29 @@ + +#include +#include + +#include "gtest/gtest.h" + +#include "libimp/codecvt.h" + +TEST(codecvt, cvt_cstr) { + char const *utf8 = "hello world, " + "\xe4\xbd\xa0\xe5\xa5\xbd\xef\xbc" + "\x8c\xe3\x81\x93\xe3\x82\x93\xe3" + "\x81\xab\xe3\x81\xa1\xe3\x81\xaf"; + wchar_t const *utf16 = L"hello world, \u4f60\u597d\uff0c\u3053\u3093\u306b\u3061\u306f"; + { + auto cvt_len = imp::cvt_cstr(utf8, std::strlen(utf8), (wchar_t *)nullptr, 0); + EXPECT_NE(cvt_len, 0); + std::wstring wstr(cvt_len, L'\0'); + EXPECT_EQ(imp::cvt_cstr(utf8, std::strlen(utf8), &wstr[0], wstr.size()), cvt_len); + EXPECT_STREQ(wstr.c_str(), utf16); + } + { + auto cvt_len = imp::cvt_cstr(utf16, std::wcslen(utf16), (char *)nullptr, 0); + EXPECT_NE(cvt_len, 0); + std::string str(cvt_len, L'\0'); + EXPECT_EQ(imp::cvt_cstr(utf16, std::wcslen(utf16), &str[0], str.size()), cvt_len); + EXPECT_STREQ(str.c_str(), utf8); + } +} \ No newline at end of file