diff --git a/README.md b/README.md index 076253e8..6c4c0275 100644 --- a/README.md +++ b/README.md @@ -17,7 +17,7 @@ Be sure to clone the repository properly:\ | llvm/clang (Linux) | UNKNOWN | 7.0.1 | 8.0.0 | | [MSVC](https://visualstudio.microsoft.com/downloads/) (Windows) | 2017 (15.9.30) | 2022 (17.11.5) | 2022 (17.12.3) | | [XCode](https://developer.apple.com/downloads/) (macOS) | 12.3 | 14.3 | 15.2 | -| [CMake](https://cmake.org/download/) | 3.15.5 | 3.26.3 | 3.29.0 | +| [CMake](https://cmake.org/download/) | 3.26.3 | 3.26.3 | 3.31.6 | | [Boost](https://www.boost.org/users/download/) | 1.75 | 1.84 | 1.84 | | [OpenSSL](https://www.openssl.org/source/) [(win)](https://slproweb.com/products/Win32OpenSSL.html) | 1.1.1n | 1.1.1w | 3.4 | | [Qt](https://download.qt.io/archive/qt/) (*only for GUI*) | 5.8.0 | 5.15.2 | 5.15.2 | @@ -42,14 +42,19 @@ Recommended OS versions: Ubuntu 20.04, 22.04 LTS. sudo apt-get install -y build-essential g++ python-dev autotools-dev libicu-dev libbz2-dev cmake git screen checkinstall zlib1g-dev libssl-dev bzip2 mesa-common-dev libglu1-mesa-dev -2. Clone Zano into a local folder\ + Make sure you have correct versions installed (see 'Dependencies' section above): + + cmake --version && gcc --version + + +3. Clone Zano into a local folder\ (If for some reason you need to use alternative Zano branch, change 'master' to the required branch name.) git clone --recursive https://github.com/hyle-team/zano.git -b master In the following steps we assume that you cloned Zano into '~/zano' folder in your home directory. -3. Download and build Boost\ +4. Download and build Boost\ (Assuming you have cloned Zano into the 'zano' folder. If you used a different location for Zano, **edit line 4** accordingly.) curl -OL https://archives.boost.io/release/1.84.0/source/boost_1_84_0.tar.bz2 @@ -59,7 +64,7 @@ Recommended OS versions: Ubuntu 20.04, 22.04 LTS. ./b2 && cd .. Make sure that you see "The Boost C++ Libraries were successfully built!" message at the end. -4. Install Qt\ +5. Install Qt\ (*GUI version only, skip this step if you're building server version*) [*GUI version*] @@ -70,7 +75,7 @@ Recommended OS versions: Ubuntu 20.04, 22.04 LTS. Then follow the instructions in Wizard. Don't forget to tick the WebEngine module checkbox! -5. Install OpenSSL +6. Install OpenSSL We recommend installing OpenSSL v1.1.1w locally unless you would like to use the same version system-wide.\ (Assuming that `$HOME` environment variable is set to your home directory. Otherwise, edit line 4 accordingly.) @@ -82,7 +87,7 @@ Recommended OS versions: Ubuntu 20.04, 22.04 LTS. make && make test && make install && cd .. -6. [*OPTIONAL*] Set global environment variables for convenient use\ +7. [*OPTIONAL*] Set global environment variables for convenient use\ For instance, by adding the following lines to `~/.bashrc` [*server version*] diff --git a/src/common/utf8.h b/src/common/utf8.h new file mode 100644 index 00000000..b5135309 --- /dev/null +++ b/src/common/utf8.h @@ -0,0 +1,46 @@ +// Copyright 2006 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_2675DCD0_9480_4c0c_B92A_CC14C027B731 +#define UTF8_FOR_CPP_2675DCD0_9480_4c0c_B92A_CC14C027B731 + +/* +To control the C++ language version used by the library, you can define UTF_CPP_CPLUSPLUS macro +and set it to one of the values used by the __cplusplus predefined macro. + +For instance, + #define UTF_CPP_CPLUSPLUS 199711L +will cause the UTF-8 CPP library to use only types and language features available in the C++ 98 standard. +Some library features will be disabled. + +If you leave UTF_CPP_CPLUSPLUS undefined, it will be internally assigned to __cplusplus. +*/ + +#include "utf8/checked.h" +#include "utf8/unchecked.h" + +#endif // header guard diff --git a/src/common/utf8/checked.h b/src/common/utf8/checked.h new file mode 100644 index 00000000..98949f8b --- /dev/null +++ b/src/common/utf8/checked.h @@ -0,0 +1,359 @@ +// Copyright 2006-2016 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_CHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 +#define UTF8_FOR_CPP_CHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 + +#include "core.h" +#include + +namespace utf8 +{ + // Base for the exceptions that may be thrown from the library + class exception : public ::std::exception { + }; + + // Exceptions that may be thrown from the library functions. + class invalid_code_point : public exception { + utfchar32_t cp; + public: + invalid_code_point(utfchar32_t codepoint) : cp(codepoint) {} + virtual const char* what() const UTF_CPP_NOEXCEPT UTF_CPP_OVERRIDE { return "Invalid code point"; } + utfchar32_t code_point() const {return cp;} + }; + + class invalid_utf8 : public exception { + utfchar8_t u8; + public: + invalid_utf8 (utfchar8_t u) : u8(u) {} + invalid_utf8 (char c) : u8(static_cast(c)) {} + virtual const char* what() const UTF_CPP_NOEXCEPT UTF_CPP_OVERRIDE { return "Invalid UTF-8"; } + utfchar8_t utf8_octet() const {return u8;} + }; + + class invalid_utf16 : public exception { + utfchar16_t u16; + public: + invalid_utf16 (utfchar16_t u) : u16(u) {} + virtual const char* what() const UTF_CPP_NOEXCEPT UTF_CPP_OVERRIDE { return "Invalid UTF-16"; } + utfchar16_t utf16_word() const {return u16;} + }; + + class not_enough_room : public exception { + public: + virtual const char* what() const UTF_CPP_NOEXCEPT UTF_CPP_OVERRIDE { return "Not enough space"; } + }; + + /// The library API - functions intended to be called by the users + + template + octet_iterator append(utfchar32_t cp, octet_iterator result) + { + if (!utf8::internal::is_code_point_valid(cp)) + throw invalid_code_point(cp); + + return internal::append(cp, result); + } + + inline void append(utfchar32_t cp, std::string& s) + { + append(cp, std::back_inserter(s)); + } + + template + word_iterator append16(utfchar32_t cp, word_iterator result) + { + if (!utf8::internal::is_code_point_valid(cp)) + throw invalid_code_point(cp); + + return internal::append16(cp, result); + } + + template + output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out, utfchar32_t replacement) + { + while (start != end) { + octet_iterator sequence_start = start; + internal::utf_error err_code = utf8::internal::validate_next(start, end); + switch (err_code) { + case internal::UTF8_OK : + for (octet_iterator it = sequence_start; it != start; ++it) + *out++ = *it; + break; + case internal::NOT_ENOUGH_ROOM: + out = utf8::append (replacement, out); + start = end; + break; + case internal::INVALID_LEAD: + out = utf8::append (replacement, out); + ++start; + break; + case internal::INCOMPLETE_SEQUENCE: + case internal::OVERLONG_SEQUENCE: + case internal::INVALID_CODE_POINT: + out = utf8::append (replacement, out); + ++start; + // just one replacement mark for the sequence + while (start != end && utf8::internal::is_trail(*start)) + ++start; + break; + } + } + return out; + } + + template + inline output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out) + { + static const utfchar32_t replacement_marker = utf8::internal::mask16(0xfffd); + return utf8::replace_invalid(start, end, out, replacement_marker); + } + + inline std::string replace_invalid(const std::string& s, utfchar32_t replacement) + { + std::string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result), replacement); + return result; + } + + inline std::string replace_invalid(const std::string& s) + { + std::string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + template + utfchar32_t next(octet_iterator& it, octet_iterator end) + { + utfchar32_t cp = 0; + internal::utf_error err_code = utf8::internal::validate_next(it, end, cp); + switch (err_code) { + case internal::UTF8_OK : + break; + case internal::NOT_ENOUGH_ROOM : + throw not_enough_room(); + case internal::INVALID_LEAD : + case internal::INCOMPLETE_SEQUENCE : + case internal::OVERLONG_SEQUENCE : + throw invalid_utf8(static_cast(*it)); + case internal::INVALID_CODE_POINT : + throw invalid_code_point(cp); + } + return cp; + } + + template + utfchar32_t next16(word_iterator& it, word_iterator end) + { + utfchar32_t cp = 0; + internal::utf_error err_code = utf8::internal::validate_next16(it, end, cp); + if (err_code == internal::NOT_ENOUGH_ROOM) + throw not_enough_room(); + return cp; + } + + template + utfchar32_t peek_next(octet_iterator it, octet_iterator end) + { + return utf8::next(it, end); + } + + template + utfchar32_t prior(octet_iterator& it, octet_iterator start) + { + // can't do much if it == start + if (it == start) + throw not_enough_room(); + + octet_iterator end = it; + // Go back until we hit either a lead octet or start + while (utf8::internal::is_trail(*(--it))) + if (it == start) + throw invalid_utf8(*it); // error - no lead byte in the sequence + return utf8::peek_next(it, end); + } + + template + void advance (octet_iterator& it, distance_type n, octet_iterator end) + { + const distance_type zero(0); + if (n < zero) { + // backward + for (distance_type i = n; i < zero; ++i) + utf8::prior(it, end); + } else { + // forward + for (distance_type i = zero; i < n; ++i) + utf8::next(it, end); + } + } + + template + typename std::iterator_traits::difference_type + distance (octet_iterator first, octet_iterator last) + { + typename std::iterator_traits::difference_type dist; + for (dist = 0; first < last; ++dist) + utf8::next(first, last); + return dist; + } + + template + octet_iterator utf16to8 (u16bit_iterator start, u16bit_iterator end, octet_iterator result) + { + while (start != end) { + utfchar32_t cp = utf8::internal::mask16(*start++); + // Take care of surrogate pairs first + if (utf8::internal::is_lead_surrogate(cp)) { + if (start != end) { + const utfchar32_t trail_surrogate = utf8::internal::mask16(*start++); + if (utf8::internal::is_trail_surrogate(trail_surrogate)) + cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET; + else + throw invalid_utf16(static_cast(trail_surrogate)); + } + else + throw invalid_utf16(static_cast(cp)); + + } + // Lone trail surrogate + else if (utf8::internal::is_trail_surrogate(cp)) + throw invalid_utf16(static_cast(cp)); + + result = utf8::append(cp, result); + } + return result; + } + + template + u16bit_iterator utf8to16 (octet_iterator start, octet_iterator end, u16bit_iterator result) + { + while (start < end) { + const utfchar32_t cp = utf8::next(start, end); + if (cp > 0xffff) { //make a surrogate pair + *result++ = static_cast((cp >> 10) + internal::LEAD_OFFSET); + *result++ = static_cast((cp & 0x3ff) + internal::TRAIL_SURROGATE_MIN); + } + else + *result++ = static_cast(cp); + } + return result; + } + + template + octet_iterator utf32to8 (u32bit_iterator start, u32bit_iterator end, octet_iterator result) + { + while (start != end) + result = utf8::append(*(start++), result); + + return result; + } + + template + u32bit_iterator utf8to32 (octet_iterator start, octet_iterator end, u32bit_iterator result) + { + while (start < end) + (*result++) = utf8::next(start, end); + + return result; + } + + // The iterator class + template + class iterator { + octet_iterator it; + octet_iterator range_start; + octet_iterator range_end; + public: + typedef utfchar32_t value_type; + typedef utfchar32_t* pointer; + typedef utfchar32_t& reference; + typedef std::ptrdiff_t difference_type; + typedef std::bidirectional_iterator_tag iterator_category; + iterator () {} + explicit iterator (const octet_iterator& octet_it, + const octet_iterator& rangestart, + const octet_iterator& rangeend) : + it(octet_it), range_start(rangestart), range_end(rangeend) + { + if (it < range_start || it > range_end) + throw std::out_of_range("Invalid utf-8 iterator position"); + } + // the default "big three" are OK + octet_iterator base () const { return it; } + utfchar32_t operator * () const + { + octet_iterator temp = it; + return utf8::next(temp, range_end); + } + bool operator == (const iterator& rhs) const + { + if (range_start != rhs.range_start || range_end != rhs.range_end) + throw std::logic_error("Comparing utf-8 iterators defined with different ranges"); + return (it == rhs.it); + } + bool operator != (const iterator& rhs) const + { + return !(operator == (rhs)); + } + iterator& operator ++ () + { + utf8::next(it, range_end); + return *this; + } + iterator operator ++ (int) + { + iterator temp = *this; + utf8::next(it, range_end); + return temp; + } + iterator& operator -- () + { + utf8::prior(it, range_start); + return *this; + } + iterator operator -- (int) + { + iterator temp = *this; + utf8::prior(it, range_start); + return temp; + } + }; // class iterator + +} // namespace utf8 + +#if UTF_CPP_CPLUSPLUS >= 202002L // C++ 20 or later +#include "cpp20.h" +#elif UTF_CPP_CPLUSPLUS >= 201703L // C++ 17 or later +#include "cpp17.h" +#elif UTF_CPP_CPLUSPLUS >= 201103L // C++ 11 or later +#include "cpp11.h" +#endif // C++ 11 or later + +#endif //header guard + diff --git a/src/common/utf8/core.h b/src/common/utf8/core.h new file mode 100644 index 00000000..af22969e --- /dev/null +++ b/src/common/utf8/core.h @@ -0,0 +1,502 @@ +// Copyright 2006 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_CORE_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 +#define UTF8_FOR_CPP_CORE_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 + +#include +#include +#include + +// Determine the C++ standard version. +// If the user defines UTF_CPP_CPLUSPLUS, use that. +// Otherwise, trust the unreliable predefined macro __cplusplus + +#if !defined UTF_CPP_CPLUSPLUS + #define UTF_CPP_CPLUSPLUS __cplusplus +#endif + +#if UTF_CPP_CPLUSPLUS >= 201103L // C++ 11 or later + #define UTF_CPP_OVERRIDE override + #define UTF_CPP_NOEXCEPT noexcept + #define UTF_CPP_STATIC_ASSERT(condition) static_assert(condition, "UTFCPP static assert"); +#else // C++ 98/03 + #define UTF_CPP_OVERRIDE + #define UTF_CPP_NOEXCEPT throw() + // Simulate static_assert: + template struct StaticAssert {static void assert() {int static_assert_impl[(Condition ? 1 : -1)];} }; + template <> struct StaticAssert {static void assert() {}}; + #define UTF_CPP_STATIC_ASSERT(condition) StaticAssert::assert(); +#endif // C++ 11 or later + + +namespace utf8 +{ +// The typedefs for 8-bit, 16-bit and 32-bit code units +#if UTF_CPP_CPLUSPLUS >= 201103L // C++ 11 or later + #if UTF_CPP_CPLUSPLUS >= 202002L // C++ 20 or later + typedef char8_t utfchar8_t; + #else // C++ 11/14/17 + typedef unsigned char utfchar8_t; + #endif + typedef char16_t utfchar16_t; + typedef char32_t utfchar32_t; +#else // C++ 98/03 + typedef unsigned char utfchar8_t; + typedef unsigned short utfchar16_t; + typedef unsigned int utfchar32_t; +#endif // C++ 11 or later + +// Helper code - not intended to be directly called by the library users. May be changed at any time +namespace internal +{ + // Unicode constants + // Leading (high) surrogates: 0xd800 - 0xdbff + // Trailing (low) surrogates: 0xdc00 - 0xdfff + const utfchar16_t LEAD_SURROGATE_MIN = 0xd800u; + const utfchar16_t LEAD_SURROGATE_MAX = 0xdbffu; + const utfchar16_t TRAIL_SURROGATE_MIN = 0xdc00u; + const utfchar16_t TRAIL_SURROGATE_MAX = 0xdfffu; + const utfchar16_t LEAD_OFFSET = 0xd7c0u; // LEAD_SURROGATE_MIN - (0x10000 >> 10) + const utfchar32_t SURROGATE_OFFSET = 0xfca02400u; // 0x10000u - (LEAD_SURROGATE_MIN << 10) - TRAIL_SURROGATE_MIN + + // Maximum valid value for a Unicode code point + const utfchar32_t CODE_POINT_MAX = 0x0010ffffu; + + template + inline utfchar8_t mask8(octet_type oc) + { + return static_cast(0xff & oc); + } + template + inline utfchar16_t mask16(u16_type oc) + { + return static_cast(0xffff & oc); + } + + template + inline bool is_trail(octet_type oc) + { + return ((utf8::internal::mask8(oc) >> 6) == 0x2); + } + + inline bool is_lead_surrogate(utfchar32_t cp) + { + return (cp >= LEAD_SURROGATE_MIN && cp <= LEAD_SURROGATE_MAX); + } + + inline bool is_trail_surrogate(utfchar32_t cp) + { + return (cp >= TRAIL_SURROGATE_MIN && cp <= TRAIL_SURROGATE_MAX); + } + + inline bool is_surrogate(utfchar32_t cp) + { + return (cp >= LEAD_SURROGATE_MIN && cp <= TRAIL_SURROGATE_MAX); + } + + inline bool is_code_point_valid(utfchar32_t cp) + { + return (cp <= CODE_POINT_MAX && !utf8::internal::is_surrogate(cp)); + } + + inline bool is_in_bmp(utfchar32_t cp) + { + return cp < utfchar32_t(0x10000); + } + + template + int sequence_length(octet_iterator lead_it) + { + const utfchar8_t lead = utf8::internal::mask8(*lead_it); + if (lead < 0x80) + return 1; + else if ((lead >> 5) == 0x6) + return 2; + else if ((lead >> 4) == 0xe) + return 3; + else if ((lead >> 3) == 0x1e) + return 4; + else + return 0; + } + + inline bool is_overlong_sequence(utfchar32_t cp, int length) + { + if (cp < 0x80) { + if (length != 1) + return true; + } + else if (cp < 0x800) { + if (length != 2) + return true; + } + else if (cp < 0x10000) { + if (length != 3) + return true; + } + return false; + } + + enum utf_error {UTF8_OK, NOT_ENOUGH_ROOM, INVALID_LEAD, INCOMPLETE_SEQUENCE, OVERLONG_SEQUENCE, INVALID_CODE_POINT}; + + /// Helper for get_sequence_x + template + utf_error increase_safely(octet_iterator& it, const octet_iterator end) + { + if (++it == end) + return NOT_ENOUGH_ROOM; + + if (!utf8::internal::is_trail(*it)) + return INCOMPLETE_SEQUENCE; + + return UTF8_OK; + } + + #define UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(IT, END) {utf_error ret = increase_safely(IT, END); if (ret != UTF8_OK) return ret;} + + /// get_sequence_x functions decode utf-8 sequences of the length x + template + utf_error get_sequence_1(octet_iterator& it, octet_iterator end, utfchar32_t& code_point) + { + if (it == end) + return NOT_ENOUGH_ROOM; + + code_point = utf8::internal::mask8(*it); + + return UTF8_OK; + } + + template + utf_error get_sequence_2(octet_iterator& it, octet_iterator end, utfchar32_t& code_point) + { + if (it == end) + return NOT_ENOUGH_ROOM; + + code_point = utf8::internal::mask8(*it); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point = ((code_point << 6) & 0x7ff) + ((*it) & 0x3f); + + return UTF8_OK; + } + + template + utf_error get_sequence_3(octet_iterator& it, octet_iterator end, utfchar32_t& code_point) + { + if (it == end) + return NOT_ENOUGH_ROOM; + + code_point = utf8::internal::mask8(*it); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point = ((code_point << 12) & 0xffff) + ((utf8::internal::mask8(*it) << 6) & 0xfff); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point = static_cast(code_point + ((*it) & 0x3f)); + + return UTF8_OK; + } + + template + utf_error get_sequence_4(octet_iterator& it, octet_iterator end, utfchar32_t& code_point) + { + if (it == end) + return NOT_ENOUGH_ROOM; + + code_point = utf8::internal::mask8(*it); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point = ((code_point << 18) & 0x1fffff) + ((utf8::internal::mask8(*it) << 12) & 0x3ffff); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point = static_cast(code_point + ((utf8::internal::mask8(*it) << 6) & 0xfff)); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point = static_cast(code_point + ((*it) & 0x3f)); + + return UTF8_OK; + } + + #undef UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR + + template + utf_error validate_next(octet_iterator& it, octet_iterator end, utfchar32_t& code_point) + { + if (it == end) + return NOT_ENOUGH_ROOM; + + // Save the original value of it so we can go back in case of failure + // Of course, it does not make much sense with i.e. stream iterators + octet_iterator original_it = it; + + utfchar32_t cp = 0; + // Determine the sequence length based on the lead octet + const int length = utf8::internal::sequence_length(it); + + // Get trail octets and calculate the code point + utf_error err = UTF8_OK; + switch (length) { + case 0: + return INVALID_LEAD; + case 1: + err = utf8::internal::get_sequence_1(it, end, cp); + break; + case 2: + err = utf8::internal::get_sequence_2(it, end, cp); + break; + case 3: + err = utf8::internal::get_sequence_3(it, end, cp); + break; + case 4: + err = utf8::internal::get_sequence_4(it, end, cp); + break; + } + + if (err == UTF8_OK) { + // Decoding succeeded. Now, security checks... + if (utf8::internal::is_code_point_valid(cp)) { + if (!utf8::internal::is_overlong_sequence(cp, length)){ + // Passed! Return here. + code_point = cp; + ++it; + return UTF8_OK; + } + else + err = OVERLONG_SEQUENCE; + } + else + err = INVALID_CODE_POINT; + } + + // Failure branch - restore the original value of the iterator + it = original_it; + return err; + } + + template + inline utf_error validate_next(octet_iterator& it, octet_iterator end) { + utfchar32_t ignored; + return utf8::internal::validate_next(it, end, ignored); + } + + template + utf_error validate_next16(word_iterator& it, word_iterator end, utfchar32_t& code_point) + { + // Make sure the iterator dereferences a large enough type + typedef typename std::iterator_traits::value_type word_type; + UTF_CPP_STATIC_ASSERT(sizeof(word_type) >= sizeof(utfchar16_t)); + // Check the edge case: + if (it == end) + return NOT_ENOUGH_ROOM; + // Save the original value of it so we can go back in case of failure + // Of course, it does not make much sense with i.e. stream iterators + word_iterator original_it = it; + + utf_error err = UTF8_OK; + + const utfchar16_t first_word = *it++; + if (!is_surrogate(first_word)) { + code_point = first_word; + return UTF8_OK; + } + else { + if (it == end) + err = NOT_ENOUGH_ROOM; + else if (is_lead_surrogate(first_word)) { + const utfchar16_t second_word = *it++; + if (is_trail_surrogate(second_word)) { + code_point = static_cast(first_word << 10) + second_word + SURROGATE_OFFSET; + return UTF8_OK; + } else + err = INCOMPLETE_SEQUENCE; + + } else { + err = INVALID_LEAD; + } + } + // error branch + it = original_it; + return err; + } + + // Internal implementation of both checked and unchecked append() function + // This function will be invoked by the overloads below, as they will know + // the octet_type. + template + octet_iterator append(utfchar32_t cp, octet_iterator result) { + if (cp < 0x80) // one octet + *(result++) = static_cast(cp); + else if (cp < 0x800) { // two octets + *(result++) = static_cast((cp >> 6) | 0xc0); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + else if (cp < 0x10000) { // three octets + *(result++) = static_cast((cp >> 12) | 0xe0); + *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + else { // four octets + *(result++) = static_cast((cp >> 18) | 0xf0); + *(result++) = static_cast(((cp >> 12) & 0x3f)| 0x80); + *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + return result; + } + + // One of the following overloads will be invoked from the API calls + + // A simple (but dangerous) case: the caller appends byte(s) to a char array + inline char* append(utfchar32_t cp, char* result) { + return append(cp, result); + } + + // Hopefully, most common case: the caller uses back_inserter + // i.e. append(cp, std::back_inserter(str)); + template + std::back_insert_iterator append + (utfchar32_t cp, std::back_insert_iterator result) { + return append, + typename container_type::value_type>(cp, result); + } + + // The caller uses some other kind of output operator - not covered above + // Note that in this case we are not able to determine octet_type + // so we assume it's utfchar8_t; that can cause a conversion warning if we are wrong. + template + octet_iterator append(utfchar32_t cp, octet_iterator result) { + return append(cp, result); + } + + // Internal implementation of both checked and unchecked append16() function + // This function will be invoked by the overloads below, as they will know + // the word_type. + template + word_iterator append16(utfchar32_t cp, word_iterator result) { + UTF_CPP_STATIC_ASSERT(sizeof(word_type) >= sizeof(utfchar16_t)); + if (is_in_bmp(cp)) + *(result++) = static_cast(cp); + else { + // Code points from the supplementary planes are encoded via surrogate pairs + *(result++) = static_cast(LEAD_OFFSET + (cp >> 10)); + *(result++) = static_cast(TRAIL_SURROGATE_MIN + (cp & 0x3FF)); + } + return result; + } + + // Hopefully, most common case: the caller uses back_inserter + // i.e. append16(cp, std::back_inserter(str)); + template + std::back_insert_iterator append16 + (utfchar32_t cp, std::back_insert_iterator result) { + return append16, + typename container_type::value_type>(cp, result); + } + + // The caller uses some other kind of output operator - not covered above + // Note that in this case we are not able to determine word_type + // so we assume it's utfchar16_t; that can cause a conversion warning if we are wrong. + template + word_iterator append16(utfchar32_t cp, word_iterator result) { + return append16(cp, result); + } + +} // namespace internal + + /// The library API - functions intended to be called by the users + + // Byte order mark + const utfchar8_t bom[] = {0xef, 0xbb, 0xbf}; + + template + octet_iterator find_invalid(octet_iterator start, octet_iterator end) + { + octet_iterator result = start; + while (result != end) { + utf8::internal::utf_error err_code = utf8::internal::validate_next(result, end); + if (err_code != internal::UTF8_OK) + return result; + } + return result; + } + + inline const char* find_invalid(const char* str) + { + const char* end = str + std::strlen(str); + return find_invalid(str, end); + } + + inline std::size_t find_invalid(const std::string& s) + { + std::string::const_iterator invalid = find_invalid(s.begin(), s.end()); + return (invalid == s.end()) ? std::string::npos : static_cast(invalid - s.begin()); + } + + template + inline bool is_valid(octet_iterator start, octet_iterator end) + { + return (utf8::find_invalid(start, end) == end); + } + + inline bool is_valid(const char* str) + { + return (*(utf8::find_invalid(str)) == '\0'); + } + + inline bool is_valid(const std::string& s) + { + return is_valid(s.begin(), s.end()); + } + + + + template + inline bool starts_with_bom (octet_iterator it, octet_iterator end) + { + return ( + ((it != end) && (utf8::internal::mask8(*it++)) == bom[0]) && + ((it != end) && (utf8::internal::mask8(*it++)) == bom[1]) && + ((it != end) && (utf8::internal::mask8(*it)) == bom[2]) + ); + } + + inline bool starts_with_bom(const std::string& s) + { + return starts_with_bom(s.begin(), s.end()); + } +} // namespace utf8 + +#endif // header guard + + diff --git a/src/common/utf8/cpp11.h b/src/common/utf8/cpp11.h new file mode 100644 index 00000000..691633c8 --- /dev/null +++ b/src/common/utf8/cpp11.h @@ -0,0 +1,70 @@ +// Copyright 2018 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_a184c22c_d012_11e8_a8d5_f2801f1b9fd1 +#define UTF8_FOR_CPP_a184c22c_d012_11e8_a8d5_f2801f1b9fd1 + +#include "checked.h" + +namespace utf8 +{ + inline void append16(utfchar32_t cp, std::u16string& s) + { + append16(cp, std::back_inserter(s)); + } + + inline std::string utf16to8(const std::u16string& s) + { + std::string result; + utf16to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u16string utf8to16(const std::string& s) + { + std::u16string result; + utf8to16(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::string utf32to8(const std::u32string& s) + { + std::string result; + utf32to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u32string utf8to32(const std::string& s) + { + std::u32string result; + utf8to32(s.begin(), s.end(), std::back_inserter(result)); + return result; + } +} // namespace utf8 + +#endif // header guard + diff --git a/src/common/utf8/cpp17.h b/src/common/utf8/cpp17.h new file mode 100644 index 00000000..6e2fcc23 --- /dev/null +++ b/src/common/utf8/cpp17.h @@ -0,0 +1,96 @@ +// Copyright 2018 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_7e906c01_03a3_4daf_b420_ea7ea952b3c9 +#define UTF8_FOR_CPP_7e906c01_03a3_4daf_b420_ea7ea952b3c9 + +#include "cpp11.h" + +namespace utf8 +{ + inline std::string utf16to8(std::u16string_view s) + { + std::string result; + utf16to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u16string utf8to16(std::string_view s) + { + std::u16string result; + utf8to16(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::string utf32to8(std::u32string_view s) + { + std::string result; + utf32to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u32string utf8to32(std::string_view s) + { + std::u32string result; + utf8to32(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::size_t find_invalid(std::string_view s) + { + std::string_view::const_iterator invalid = find_invalid(s.begin(), s.end()); + return (invalid == s.end()) ? std::string_view::npos : static_cast(invalid - s.begin()); + } + + inline bool is_valid(std::string_view s) + { + return is_valid(s.begin(), s.end()); + } + + inline std::string replace_invalid(std::string_view s, char32_t replacement) + { + std::string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result), replacement); + return result; + } + + inline std::string replace_invalid(std::string_view s) + { + std::string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline bool starts_with_bom(std::string_view s) + { + return starts_with_bom(s.begin(), s.end()); + } + +} // namespace utf8 + +#endif // header guard + diff --git a/src/common/utf8/cpp20.h b/src/common/utf8/cpp20.h new file mode 100644 index 00000000..07b61d0f --- /dev/null +++ b/src/common/utf8/cpp20.h @@ -0,0 +1,124 @@ +// Copyright 2022 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_207e906c01_03a3_4daf_b420_ea7ea952b3c9 +#define UTF8_FOR_CPP_207e906c01_03a3_4daf_b420_ea7ea952b3c9 + +#include "cpp17.h" + +namespace utf8 +{ + inline std::u8string utf16tou8(const std::u16string& s) + { + std::u8string result; + utf16to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u8string utf16tou8(std::u16string_view s) + { + std::u8string result; + utf16to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u16string utf8to16(const std::u8string& s) + { + std::u16string result; + utf8to16(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u16string utf8to16(const std::u8string_view& s) + { + std::u16string result; + utf8to16(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u8string utf32tou8(const std::u32string& s) + { + std::u8string result; + utf32to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u8string utf32tou8(const std::u32string_view& s) + { + std::u8string result; + utf32to8(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u32string utf8to32(const std::u8string& s) + { + std::u32string result; + utf8to32(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::u32string utf8to32(const std::u8string_view& s) + { + std::u32string result; + utf8to32(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline std::size_t find_invalid(const std::u8string& s) + { + std::u8string::const_iterator invalid = find_invalid(s.begin(), s.end()); + return (invalid == s.end()) ? std::string_view::npos : static_cast(invalid - s.begin()); + } + + inline bool is_valid(const std::u8string& s) + { + return is_valid(s.begin(), s.end()); + } + + inline std::u8string replace_invalid(const std::u8string& s, char32_t replacement) + { + std::u8string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result), replacement); + return result; + } + + inline std::u8string replace_invalid(const std::u8string& s) + { + std::u8string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + inline bool starts_with_bom(const std::u8string& s) + { + return starts_with_bom(s.begin(), s.end()); + } + +} // namespace utf8 + +#endif // header guard + diff --git a/src/common/utf8/unchecked.h b/src/common/utf8/unchecked.h new file mode 100644 index 00000000..bf291789 --- /dev/null +++ b/src/common/utf8/unchecked.h @@ -0,0 +1,287 @@ +// Copyright 2006 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_UNCHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 +#define UTF8_FOR_CPP_UNCHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 + +#include "core.h" + +namespace utf8 +{ + namespace unchecked + { + template + octet_iterator append(utfchar32_t cp, octet_iterator result) + { + return internal::append(cp, result); + } + + template + word_iterator append16(utfchar32_t cp, word_iterator result) + { + return internal::append16(cp, result); + } + + template + output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out, utfchar32_t replacement) + { + while (start != end) { + octet_iterator sequence_start = start; + internal::utf_error err_code = utf8::internal::validate_next(start, end); + switch (err_code) { + case internal::UTF8_OK : + for (octet_iterator it = sequence_start; it != start; ++it) + *out++ = *it; + break; + case internal::NOT_ENOUGH_ROOM: + out = utf8::unchecked::append(replacement, out); + start = end; + break; + case internal::INVALID_LEAD: + out = utf8::unchecked::append(replacement, out); + ++start; + break; + case internal::INCOMPLETE_SEQUENCE: + case internal::OVERLONG_SEQUENCE: + case internal::INVALID_CODE_POINT: + out = utf8::unchecked::append(replacement, out); + ++start; + // just one replacement mark for the sequence + while (start != end && utf8::internal::is_trail(*start)) + ++start; + break; + } + } + return out; + } + + template + inline output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out) + { + static const utfchar32_t replacement_marker = utf8::internal::mask16(0xfffd); + return utf8::unchecked::replace_invalid(start, end, out, replacement_marker); + } + + inline std::string replace_invalid(const std::string& s, utfchar32_t replacement) + { + std::string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result), replacement); + return result; + } + + inline std::string replace_invalid(const std::string& s) + { + std::string result; + replace_invalid(s.begin(), s.end(), std::back_inserter(result)); + return result; + } + + template + utfchar32_t next(octet_iterator& it) + { + utfchar32_t cp = utf8::internal::mask8(*it); + switch (utf8::internal::sequence_length(it)) { + case 1: + break; + case 2: + it++; + cp = ((cp << 6) & 0x7ff) + ((*it) & 0x3f); + break; + case 3: + ++it; + cp = ((cp << 12) & 0xffff) + ((utf8::internal::mask8(*it) << 6) & 0xfff); + ++it; + cp = static_cast(cp + ((*it) & 0x3f)); + break; + case 4: + ++it; + cp = ((cp << 18) & 0x1fffff) + ((utf8::internal::mask8(*it) << 12) & 0x3ffff); + ++it; + cp = static_cast(cp + ((utf8::internal::mask8(*it) << 6) & 0xfff)); + ++it; + cp = static_cast(cp + ((*it) & 0x3f)); + break; + } + ++it; + return cp; + } + + template + utfchar32_t peek_next(octet_iterator it) + { + return utf8::unchecked::next(it); + } + + template + utfchar32_t next16(word_iterator& it) + { + utfchar32_t cp = utf8::internal::mask16(*it++); + if (utf8::internal::is_lead_surrogate(cp)) + return (cp << 10) + *it++ + utf8::internal::SURROGATE_OFFSET; + return cp; + } + + template + utfchar32_t prior(octet_iterator& it) + { + while (utf8::internal::is_trail(*(--it))) ; + octet_iterator temp = it; + return utf8::unchecked::next(temp); + } + + template + void advance(octet_iterator& it, distance_type n) + { + const distance_type zero(0); + if (n < zero) { + // backward + for (distance_type i = n; i < zero; ++i) + utf8::unchecked::prior(it); + } else { + // forward + for (distance_type i = zero; i < n; ++i) + utf8::unchecked::next(it); + } + } + + template + typename std::iterator_traits::difference_type + distance(octet_iterator first, octet_iterator last) + { + typename std::iterator_traits::difference_type dist; + for (dist = 0; first < last; ++dist) + utf8::unchecked::next(first); + return dist; + } + + template + octet_iterator utf16to8(u16bit_iterator start, u16bit_iterator end, octet_iterator result) + { + while (start != end) { + utfchar32_t cp = utf8::internal::mask16(*start++); + // Take care of surrogate pairs first + if (utf8::internal::is_lead_surrogate(cp)) { + if (start == end) + return result; + utfchar32_t trail_surrogate = utf8::internal::mask16(*start++); + cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET; + } + result = utf8::unchecked::append(cp, result); + } + return result; + } + + template + u16bit_iterator utf8to16(octet_iterator start, octet_iterator end, u16bit_iterator result) + { + while (start < end) { + utfchar32_t cp = utf8::unchecked::next(start); + if (cp > 0xffff) { //make a surrogate pair + *result++ = static_cast((cp >> 10) + internal::LEAD_OFFSET); + *result++ = static_cast((cp & 0x3ff) + internal::TRAIL_SURROGATE_MIN); + } + else + *result++ = static_cast(cp); + } + return result; + } + + template + octet_iterator utf32to8(u32bit_iterator start, u32bit_iterator end, octet_iterator result) + { + while (start != end) + result = utf8::unchecked::append(*(start++), result); + + return result; + } + + template + u32bit_iterator utf8to32(octet_iterator start, octet_iterator end, u32bit_iterator result) + { + while (start < end) + (*result++) = utf8::unchecked::next(start); + + return result; + } + + // The iterator class + template + class iterator { + octet_iterator it; + public: + typedef utfchar32_t value_type; + typedef utfchar32_t* pointer; + typedef utfchar32_t& reference; + typedef std::ptrdiff_t difference_type; + typedef std::bidirectional_iterator_tag iterator_category; + iterator () {} + explicit iterator (const octet_iterator& octet_it): it(octet_it) {} + // the default "big three" are OK + octet_iterator base () const { return it; } + utfchar32_t operator * () const + { + octet_iterator temp = it; + return utf8::unchecked::next(temp); + } + bool operator == (const iterator& rhs) const + { + return (it == rhs.it); + } + bool operator != (const iterator& rhs) const + { + return !(operator == (rhs)); + } + iterator& operator ++ () + { + ::std::advance(it, utf8::internal::sequence_length(it)); + return *this; + } + iterator operator ++ (int) + { + iterator temp = *this; + ::std::advance(it, utf8::internal::sequence_length(it)); + return temp; + } + iterator& operator -- () + { + utf8::unchecked::prior(it); + return *this; + } + iterator operator -- (int) + { + iterator temp = *this; + utf8::unchecked::prior(it); + return temp; + } + }; // class iterator + + } // namespace utf8::unchecked +} // namespace utf8 + + +#endif // header guard + diff --git a/src/common/util.cpp b/src/common/util.cpp index 9d12b839..b2827d59 100644 --- a/src/common/util.cpp +++ b/src/common/util.cpp @@ -11,7 +11,8 @@ using namespace epee; #include "util.h" #include "currency_core/currency_config.h" #include "version.h" - +#define UTF_CPP_CPLUSPLUS 201703L +#include "utf8.h" #ifdef WIN32 #include #include @@ -840,113 +841,198 @@ std::string get_nix_version_display_string() return pr.first + " " + pr.second; } - // Replaces invalid UTF-8 sequences with the UTF-8 replacement character (U+FFFD). -// If the original string is already valid UTF-8, it remains untouched. -// Returns true if the function modified the string, false otherwise. - bool sanitize_utf8(std::string& str_to_sanitize) { - const unsigned char* data = reinterpret_cast(str_to_sanitize.data()); - const unsigned char* end = data + str_to_sanitize.size(); - std::string output; // remains empty unless invalid data is found - bool foundInvalid = false; - - // We'll track where the last "valid segment" started. - const unsigned char* segmentStart = data; - - // Helper lambda to append the replacement character U+FFFD (0xEF 0xBF 0xBD). - auto appendReplacementChar = [&](const unsigned char* pos) { - // If this is our first detected invalid sequence, start building 'output'. - // - Copy everything from segmentStart to 'pos' (not inclusive). - if (!foundInvalid) { - foundInvalid = true; - // Reserve approximate space to avoid repeated allocations - output.reserve(str_to_sanitize.size()); - output.append(reinterpret_cast(segmentStart), - reinterpret_cast(pos)); - } - // Append the UTF-8 replacement character - output.append("\xEF\xBF\xBD"); - }; - - while (data < end) { - unsigned char c = *data; - int extraBytes = 0; - - // 1) Determine how many bytes the current sequence should have, - // based on the leading byte. - if ((c & 0x80) == 0) { - // 1-byte sequence (ASCII) - extraBytes = 0; - } - else if ((c & 0xE0) == 0xC0) { - // 2-byte sequence - extraBytes = 1; - } - else if ((c & 0xF0) == 0xE0) { - // 3-byte sequence - extraBytes = 2; - } - else if ((c & 0xF8) == 0xF0) { - // 4-byte sequence - extraBytes = 3; - } - else { - // Invalid leading byte - appendReplacementChar(data); - data++; - // Next segment of "known good" data starts after this invalid byte - segmentStart = data; - continue; - } - - // 2) Check that we have enough bytes left in the string - if (data + extraBytes >= end) { - // We don't have enough continuation bytes for a complete sequence - appendReplacementChar(data); - // Nothing more we can parse, so we're done - data = end; - break; - } - - // 3) Validate the continuation bytes, which must match 10xxxxxx - bool invalidContinuation = false; - for (int i = 1; i <= extraBytes; ++i) { - if ((data[i] & 0xC0) != 0x80) { - invalidContinuation = true; - break; - } - } - - if (invalidContinuation) { - // One or more continuation bytes are invalid - appendReplacementChar(data); - data++; - segmentStart = data; - continue; - } - - // If we reach here, (c + continuation) forms a valid UTF-8 sequence - data += (extraBytes + 1); - } - - // If we never encountered invalid data, 'foundInvalid' is false. - // In that case, we do nothing and simply return false. - if (!foundInvalid) { - return false; // No change was made - } - - // Otherwise, we need to copy the last valid segment (if any) into 'output'. - // This covers the data from 'segmentStart' to the end. - if (segmentStart < end) { - output.append(reinterpret_cast(segmentStart), - reinterpret_cast(end)); - } - - // Now replace the original string with our sanitized version - str_to_sanitize.swap(output); - - // Indicate we performed a modification - return true; + bool isUtf8(std::string_view text, size_t* p_invalid_offset) + { + size_t local_invalid = 0; + size_t& invalid = p_invalid_offset ? *p_invalid_offset : local_invalid; + invalid = ::utf8::find_invalid(text); + return invalid == std::string_view::npos; } + bool sanitize_utf8(std::string& str_to_sanitize) + { + size_t invalid = 0; + if (!isUtf8(str_to_sanitize, &invalid)) + { + std::string temp; + utf8::replace_invalid(str_to_sanitize.begin(), str_to_sanitize.end(), back_inserter(temp)); + str_to_sanitize.swap(temp); + return true; + } + return false; + } + +// // UTF-8 encoding of the Unicode replacement character U+FFFD. +// static const char REPLACEMENT_CHAR_UTF8[] = "\xEF\xBF\xBD"; +// +// bool sanitize_utf8(std::string& str_to_sanitize) +// { +// const unsigned char* data = reinterpret_cast(str_to_sanitize.data()); +// std::size_t length = str_to_sanitize.size(); +// std::size_t i = 0; +// +// std::string sanitized; +// sanitized.reserve(length); // Reserve to avoid repeated allocations +// +// bool changed = false; +// +// while (i < length) +// { +// unsigned char c = data[i]; +// +// // 1) ASCII fast path: 0xxxxxxx +// if (c < 0x80) +// { +// sanitized.push_back(static_cast(c)); +// i++; +// } +// // 2) 2-byte sequence: 110xxxxx 10xxxxxx +// else if ((c & 0xE0) == 0xC0) +// { +// // Need at least 2 bytes total +// if (i + 1 < length) +// { +// unsigned char c2 = data[i + 1]; +// // Valid second byte: 10xxxxxx +// // Also ensure it’s not an overlong encoding (0xC0, 0xC1 are invalid starts) +// if ((c2 & 0xC0) == 0x80 && (c & 0xFE) != 0xC0) +// { +// sanitized.push_back(static_cast(c)); +// sanitized.push_back(static_cast(c2)); +// i += 2; +// } +// else +// { +// sanitized.append(REPLACEMENT_CHAR_UTF8); +// changed = true; +// // Advance by one or two bytes if second byte was somewhat valid +// i += 1 + ((c2 & 0xC0) == 0x80); +// } +// } +// else +// { +// // Truncated 2-byte sequence at the end +// sanitized.append(REPLACEMENT_CHAR_UTF8); +// changed = true; +// i++; +// } +// } +// // 3) 3-byte sequence: 1110xxxx 10xxxxxx 10xxxxxx +// else if ((c & 0xF0) == 0xE0) +// { +// // Need at least 3 bytes total +// if (i + 2 < length) +// { +// unsigned char c2 = data[i + 1]; +// unsigned char c3 = data[i + 2]; +// if (((c2 & 0xC0) == 0x80) && +// ((c3 & 0xC0) == 0x80)) +// { +// // Combine to form a code point: +// // 1110xxxx 10xxxxxx 10xxxxxx +// // Check for UTF-16 surrogates (invalid in UTF-8) +// unsigned int codepoint = ((c & 0x0F) << 12) +// | ((c2 & 0x3F) << 6) +// | (c3 & 0x3F); +// if (codepoint >= 0xD800 && codepoint <= 0xDFFF) +// { +// // Surrogate half, invalid in UTF-8 +// sanitized.append(REPLACEMENT_CHAR_UTF8); +// changed = true; +// } +// else +// { +// // Valid 3-byte sequence +// sanitized.push_back(static_cast(c)); +// sanitized.push_back(static_cast(c2)); +// sanitized.push_back(static_cast(c3)); +// } +// i += 3; +// } +// else +// { +// sanitized.append(REPLACEMENT_CHAR_UTF8); +// changed = true; +// // Advance by how many bytes look valid +// i += 1 + ((c2 & 0xC0) == 0x80) + ((c3 & 0xC0) == 0x80); +// } +// } +// else +// { +// // Truncated 3-byte sequence +// sanitized.append(REPLACEMENT_CHAR_UTF8); +// changed = true; +// i = length; // jump out +// } +// } +// // 4) 4-byte sequence: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx +// else if ((c & 0xF8) == 0xF0) +// { +// // Need at least 4 bytes total +// if (i + 3 < length) +// { +// unsigned char c2 = data[i + 1]; +// unsigned char c3 = data[i + 2]; +// unsigned char c4 = data[i + 3]; +// if (((c2 & 0xC0) == 0x80) && +// ((c3 & 0xC0) == 0x80) && +// ((c4 & 0xC0) == 0x80)) +// { +// // Combine to form a code point: +// // 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx +// unsigned int codepoint = ((c & 0x07) << 18) +// | ((c2 & 0x3F) << 12) +// | ((c3 & 0x3F) << 6) +// | (c4 & 0x3F); +// // Max valid code point is 0x10FFFF +// if (codepoint <= 0x10FFFF) +// { +// sanitized.push_back(static_cast(c)); +// sanitized.push_back(static_cast(c2)); +// sanitized.push_back(static_cast(c3)); +// sanitized.push_back(static_cast(c4)); +// } +// else +// { +// sanitized.append(REPLACEMENT_CHAR_UTF8); +// changed = true; +// } +// i += 4; +// } +// else +// { +// sanitized.append(REPLACEMENT_CHAR_UTF8); +// changed = true; +// i += 1 + ((c2 & 0xC0) == 0x80) + +// ((c3 & 0xC0) == 0x80) + +// ((c4 & 0xC0) == 0x80); +// } +// } +// else +// { +// // Truncated 4-byte sequence +// sanitized.append(REPLACEMENT_CHAR_UTF8); +// changed = true; +// i = length; // jump out +// } +// } +// else +// { +// // Invalid leading byte (> 0xF4 or some unexpected pattern) +// sanitized.append(REPLACEMENT_CHAR_UTF8); +// changed = true; +// i++; +// } +// } +// +// // Update original string only if changes were made +// if (changed) +// { +// str_to_sanitize = std::move(sanitized); +// return true; +// } +// return false; +// } + } // namespace tools diff --git a/src/version.h.in b/src/version.h.in index e4f5d197..a2c99556 100644 --- a/src/version.h.in +++ b/src/version.h.in @@ -8,6 +8,6 @@ #define PROJECT_REVISION "1" #define PROJECT_VERSION PROJECT_MAJOR_VERSION "." PROJECT_MINOR_VERSION "." PROJECT_REVISION -#define PROJECT_VERSION_BUILD_NO 391 +#define PROJECT_VERSION_BUILD_NO 392 #define PROJECT_VERSION_BUILD_NO_STR STRINGIFY_EXPAND(PROJECT_VERSION_BUILD_NO) #define PROJECT_VERSION_LONG PROJECT_VERSION "." PROJECT_VERSION_BUILD_NO_STR "[" BUILD_COMMIT_ID "]"