From 0c45c63943ec70bcb8734299ca0b86f4f0131c6f Mon Sep 17 00:00:00 2001 From: Tsuda Kageyu Date: Fri, 27 Jan 2017 11:44:56 +0900 Subject: [PATCH] Replace unicode.h/unicode.cpp by the UTF8-CPP library. unicode.h/unicode.cpp are no longer maintained and incompatible with Debian's guideline. UTF8-CPP is maintained on GitHub and published under the Boost Software License which is compatible with either LGPL or MPL and should go along with Debian's guideline. --- taglib/CMakeLists.txt | 8 +- taglib/toolkit/tstring.cpp | 100 ++--------- taglib/toolkit/unicode.cpp | 303 ------------------------------- taglib/toolkit/unicode.h | 149 --------------- taglib/toolkit/utf8/checked.h | 327 +++++++++++++++++++++++++++++++++ taglib/toolkit/utf8/core.h | 329 ++++++++++++++++++++++++++++++++++ tests/test_string.cpp | 8 + 7 files changed, 682 insertions(+), 542 deletions(-) delete mode 100644 taglib/toolkit/unicode.cpp delete mode 100644 taglib/toolkit/unicode.h create mode 100644 taglib/toolkit/utf8/checked.h create mode 100644 taglib/toolkit/utf8/core.h diff --git a/taglib/CMakeLists.txt b/taglib/CMakeLists.txt index 4ea54304..13ba7892 100644 --- a/taglib/CMakeLists.txt +++ b/taglib/CMakeLists.txt @@ -312,12 +312,6 @@ set(toolkit_SRCS toolkit/tzlib.cpp ) -if(NOT WIN32) - set(unicode_SRCS - toolkit/unicode.cpp - ) -endif() - if(HAVE_ZLIB_SOURCE) set(zlib_SRCS ${ZLIB_SOURCE}/adler32.c @@ -334,7 +328,7 @@ set(tag_LIB_SRCS ${vorbis_SRCS} ${oggflacs_SRCS} ${mpc_SRCS} ${ape_SRCS} ${toolkit_SRCS} ${flacs_SRCS} ${wavpack_SRCS} ${speex_SRCS} ${trueaudio_SRCS} ${riff_SRCS} ${aiff_SRCS} ${wav_SRCS} ${asf_SRCS} ${mp4_SRCS} ${mod_SRCS} ${s3m_SRCS} ${it_SRCS} ${xm_SRCS} ${opus_SRCS} - ${unicode_SRCS} ${zlib_SRCS} + ${zlib_SRCS} tag.cpp tagunion.cpp fileref.cpp diff --git a/taglib/toolkit/tstring.cpp b/taglib/toolkit/tstring.cpp index 1639d7c5..b154cbae 100644 --- a/taglib/toolkit/tstring.cpp +++ b/taglib/toolkit/tstring.cpp @@ -31,16 +31,11 @@ #include #include -#ifdef _WIN32 -# include -#else -# include "unicode.h" -#endif - #include #include #include #include +#include #include "tstring.h" @@ -48,72 +43,6 @@ namespace { using namespace TagLib; - size_t UTF16toUTF8(const wchar_t *src, size_t srcLength, char *dst, size_t dstLength) - { - size_t len = 0; - -#ifdef _WIN32 - - len = ::WideCharToMultiByte( - CP_UTF8, 0, src, static_cast(srcLength), dst, static_cast(dstLength), NULL, NULL); - -#else - - using namespace Unicode; - - const UTF16 *srcBegin = src; - const UTF16 *srcEnd = srcBegin + srcLength; - - UTF8 *dstBegin = reinterpret_cast(dst); - UTF8 *dstEnd = dstBegin + dstLength; - - ConversionResult result = ConvertUTF16toUTF8( - &srcBegin, srcEnd, &dstBegin, dstEnd, lenientConversion); - - if(result == conversionOK) - len = dstBegin - reinterpret_cast(dst); - -#endif - - if(len == 0) - debug("String::UTF16toUTF8() - Unicode conversion error."); - - return len; - } - - size_t UTF8toUTF16(const char *src, size_t srcLength, wchar_t *dst, size_t dstLength) - { - size_t len = 0; - -#ifdef _WIN32 - - len = ::MultiByteToWideChar( - CP_UTF8, MB_ERR_INVALID_CHARS, src, static_cast(srcLength), dst, static_cast(dstLength)); - -#else - - using namespace Unicode; - - const UTF8 *srcBegin = reinterpret_cast(src); - const UTF8 *srcEnd = srcBegin + srcLength; - - UTF16 *dstBegin = dst; - UTF16 *dstEnd = dstBegin + dstLength; - - ConversionResult result = ConvertUTF8toUTF16( - &srcBegin, srcEnd, &dstBegin, dstEnd, lenientConversion); - - if(result == conversionOK) - len = dstBegin - dst; - -#endif - - if(len == 0) - debug("String::UTF8toUTF16() - Unicode conversion error."); - - return len; - } - // Returns the native format of std::wstring. String::Type wcharByteOrder() { @@ -139,9 +68,13 @@ namespace { data.resize(length); - if(length > 0) { - const size_t len = UTF8toUTF16(s, length, &data[0], data.size()); - data.resize(len); + try { + const std::wstring::iterator dstEnd = utf8::utf8to16(s, s + length, data.begin()); + data.resize(dstEnd - data.begin()); + } + catch(const utf8::exception &e) { + debug(String("String::copyFromUTF8() - UTF8-CPP error: ") + e.what()); + data.clear(); } } @@ -517,19 +450,20 @@ ByteVector String::data(Type t) const return v; } case UTF8: - if(!d->data.empty()) { - ByteVector v(size() * 4 + 1, 0); + ByteVector v(size() * 4, 0); - const size_t len = UTF16toUTF8( - d->data.c_str(), d->data.size(), v.data(), v.size()); - v.resize(static_cast(len)); + try { + const ByteVector::Iterator dstEnd = utf8::utf16to8(begin(), end(), v.begin()); + v.resize(static_cast(dstEnd - v.begin())); + } + catch(const utf8::exception &e) { + debug(String("String::data() - UTF8-CPP error: ") + e.what()); + v.clear(); + } return v; } - else { - return ByteVector(); - } case UTF16: { ByteVector v(2 + size() * 2, 0); diff --git a/taglib/toolkit/unicode.cpp b/taglib/toolkit/unicode.cpp deleted file mode 100644 index 1b26977e..00000000 --- a/taglib/toolkit/unicode.cpp +++ /dev/null @@ -1,303 +0,0 @@ -/******************************************************************************* - * * - * THIS FILE IS INCLUDED IN TAGLIB, BUT IS NOT COPYRIGHTED BY THE TAGLIB * - * AUTHORS, NOT PART OF THE TAGLIB API AND COULD GO AWAY AT ANY POINT IN TIME. * - * AS SUCH IT SHOULD BE CONSIERED FOR INTERNAL USE ONLY. * - * * - *******************************************************************************/ - -/* - * Copyright 2001 Unicode, Inc. - * - * Disclaimer - * - * This source code is provided as is by Unicode, Inc. No claims are - * made as to fitness for any particular purpose. No warranties of any - * kind are expressed or implied. The recipient agrees to determine - * applicability of information provided. If this file has been - * purchased on magnetic or optical media from Unicode, Inc., the - * sole remedy for any claim will be exchange of defective media - * within 90 days of receipt. - * - * Limitations on Rights to Redistribute This Code - * - * Unicode, Inc. hereby grants the right to freely use the information - * supplied in this file in the creation of products supporting the - * Unicode Standard, and to make copies of this file in any form - * for internal or external distribution as long as this notice - * remains attached. - */ - -/* - * This file has been modified by Scott Wheeler to remove - * the UTF32 conversion functions and to place the appropriate functions - * in their own C++ namespace. - */ - -/* --------------------------------------------------------------------- - - Conversions between UTF32, UTF-16, and UTF-8. Source code file. - Author: Mark E. Davis, 1994. - Rev History: Rick McGowan, fixes & updates May 2001. - Sept 2001: fixed const & error conditions per - mods suggested by S. Parent & A. Lillich. - - See the header file "ConvertUTF.h" for complete documentation. - ------------------------------------------------------------------------- */ - - -#include "unicode.h" -#include - -#define UNI_SUR_HIGH_START (UTF32)0xD800 -#define UNI_SUR_HIGH_END (UTF32)0xDBFF -#define UNI_SUR_LOW_START (UTF32)0xDC00 -#define UNI_SUR_LOW_END (UTF32)0xDFFF -#define false 0 -#define true 1 - -namespace Unicode { - -static const int halfShift = 10; /* used for shifting by 10 bits */ - -static const UTF32 halfBase = 0x0010000UL; -static const UTF32 halfMask = 0x3FFUL; - -/* - * Index into the table below with the first byte of a UTF-8 sequence to - * get the number of trailing bytes that are supposed to follow it. - */ -static const char trailingBytesForUTF8[256] = { - 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, - 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, - 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, - 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5 -}; - -/* - * Magic values subtracted from a buffer value during UTF8 conversion. - * This table contains as many values as there might be trailing bytes - * in a UTF-8 sequence. - */ -static const UTF32 offsetsFromUTF8[6] = { 0x00000000UL, 0x00003080UL, 0x000E2080UL, - 0x03C82080UL, 0xFA082080UL, 0x82082080UL }; - -/* - * Once the bits are split out into bytes of UTF-8, this is a mask OR-ed - * into the first byte, depending on how many bytes follow. There are - * as many entries in this table as there are UTF-8 sequence types. - * (I.e., one byte sequence, two byte... six byte sequence.) - */ -static const UTF8 firstByteMark[7] = { 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC }; - -/* --------------------------------------------------------------------- */ - -/* The interface converts a whole buffer to avoid function-call overhead. - * Constants have been gathered. Loops & conditionals have been removed as - * much as possible for efficiency, in favor of drop-through switches. - * (See "Note A" at the bottom of the file for equivalent code.) - * If your compiler supports it, the "isLegalUTF8" call can be turned - * into an inline function. - */ - -/* --------------------------------------------------------------------- */ - -ConversionResult ConvertUTF16toUTF8 ( - const UTF16** sourceStart, const UTF16* sourceEnd, - UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags) { - ConversionResult result = conversionOK; - const UTF16* source = *sourceStart; - UTF8* target = *targetStart; - while (source < sourceEnd) { - UTF32 ch; - unsigned short bytesToWrite = 0; - const UTF32 byteMask = 0xBF; - const UTF32 byteMark = 0x80; - const UTF16* oldSource = source; /* In case we have to back up because of target overflow. */ - ch = *source++; - /* If we have a surrogate pair, convert to UTF32 first. */ - if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_HIGH_END && source < sourceEnd) { - UTF32 ch2 = *source; - if (ch2 >= UNI_SUR_LOW_START && ch2 <= UNI_SUR_LOW_END) { - ch = ((ch - UNI_SUR_HIGH_START) << halfShift) - + (ch2 - UNI_SUR_LOW_START) + halfBase; - ++source; - } else if (flags == strictConversion) { /* it's an unpaired high surrogate */ - --source; /* return to the illegal value itself */ - result = sourceIllegal; - break; - } - } else if ((flags == strictConversion) && (ch >= UNI_SUR_LOW_START && ch <= UNI_SUR_LOW_END)) { - --source; /* return to the illegal value itself */ - result = sourceIllegal; - break; - } - /* Figure out how many bytes the result will require */ - if (ch < (UTF32)0x80) { bytesToWrite = 1; - } else if (ch < (UTF32)0x800) { bytesToWrite = 2; - } else if (ch < (UTF32)0x10000) { bytesToWrite = 3; - } else if (ch < (UTF32)0x200000) { bytesToWrite = 4; - } else { bytesToWrite = 2; - ch = UNI_REPLACEMENT_CHAR; - } - // printf("bytes to write = %i\n", bytesToWrite); - target += bytesToWrite; - if (target > targetEnd) { - source = oldSource; /* Back up source pointer! */ - target -= bytesToWrite; result = targetExhausted; break; - } - switch (bytesToWrite) { /* note: everything falls through. */ - case 4: *--target = (ch | byteMark) & byteMask; ch >>= 6; - case 3: *--target = (ch | byteMark) & byteMask; ch >>= 6; - case 2: *--target = (ch | byteMark) & byteMask; ch >>= 6; - case 1: *--target = (UTF8)(ch | firstByteMark[bytesToWrite]); - } - target += bytesToWrite; - } - *sourceStart = source; - *targetStart = target; - return result; -} - -/* --------------------------------------------------------------------- */ - -/* - * Utility routine to tell whether a sequence of bytes is legal UTF-8. - * This must be called with the length pre-determined by the first byte. - * If not calling this from ConvertUTF8to*, then the length can be set by: - * length = trailingBytesForUTF8[*source]+1; - * and the sequence is illegal right away if there aren't that many bytes - * available. - * If presented with a length > 4, this returns false. The Unicode - * definition of UTF-8 goes up to 4-byte sequences. - */ - -static Boolean isLegalUTF8(const UTF8 *source, int length) { - UTF8 a; - const UTF8 *srcptr = source+length; - switch (length) { - default: return false; - /* Everything else falls through when "true"... */ - case 4: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false; - case 3: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false; - case 2: if ((a = (*--srcptr)) > 0xBF) return false; - switch (*source) { - /* no fall-through in this inner switch */ - case 0xE0: if (a < 0xA0) return false; break; - case 0xF0: if (a < 0x90) return false; break; - case 0xF4: if (a > 0x8F) return false; break; - default: if (a < 0x80) return false; - } - case 1: if (*source >= 0x80 && *source < 0xC2) return false; - if (*source > 0xF4) return false; - } - return true; -} - -/* --------------------------------------------------------------------- */ - -/* - * Exported function to return whether a UTF-8 sequence is legal or not. - * This is not used here; it's just exported. - */ -Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd) { - int length = trailingBytesForUTF8[*source]+1; - if (source+length > sourceEnd) { - return false; - } - return isLegalUTF8(source, length); -} - -/* --------------------------------------------------------------------- */ - -ConversionResult ConvertUTF8toUTF16 ( - const UTF8** sourceStart, const UTF8* sourceEnd, - UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags) { - ConversionResult result = conversionOK; - const UTF8* source = *sourceStart; - UTF16* target = *targetStart; - while (source < sourceEnd) { - UTF32 ch = 0; - unsigned short extraBytesToRead = trailingBytesForUTF8[*source]; - if (source + extraBytesToRead >= sourceEnd) { - result = sourceExhausted; break; - } - /* Do this check whether lenient or strict */ - if (! isLegalUTF8(source, extraBytesToRead+1)) { - result = sourceIllegal; - break; - } - /* - * The cases all fall through. See "Note A" below. - */ - switch (extraBytesToRead) { - case 3: ch += *source++; ch <<= 6; - case 2: ch += *source++; ch <<= 6; - case 1: ch += *source++; ch <<= 6; - case 0: ch += *source++; - } - ch -= offsetsFromUTF8[extraBytesToRead]; - - if (target >= targetEnd) { - source -= (extraBytesToRead+1); /* Back up source pointer! */ - result = targetExhausted; break; - } - if (ch <= UNI_MAX_BMP) { /* Target is a character <= 0xFFFF */ - if ((flags == strictConversion) && (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END)) { - source -= (extraBytesToRead+1); /* return to the illegal value itself */ - result = sourceIllegal; - break; - } else { - *target++ = (UTF16)ch; /* normal case */ - } - } else if (ch > UNI_MAX_UTF16) { - if (flags == strictConversion) { - result = sourceIllegal; - source -= (extraBytesToRead+1); /* return to the start */ - break; /* Bail out; shouldn't continue */ - } else { - *target++ = UNI_REPLACEMENT_CHAR; - } - } else { - /* target is a character in range 0xFFFF - 0x10FFFF. */ - if (target + 1 >= targetEnd) { - source -= (extraBytesToRead+1); /* Back up source pointer! */ - result = targetExhausted; break; - } - ch -= halfBase; - *target++ = (UTF16)((ch >> halfShift) + UNI_SUR_HIGH_START); - *target++ = (ch & halfMask) + UNI_SUR_LOW_START; - } - } - *sourceStart = source; - *targetStart = target; - return result; -} - -} - -/* --------------------------------------------------------------------- - - Note A. - The fall-through switches in UTF-8 reading code save a - temp variable, some decrements & conditionals. The switches - are equivalent to the following loop: - { - int tmpBytesToRead = extraBytesToRead+1; - do { - ch += *source++; - --tmpBytesToRead; - if (tmpBytesToRead) ch <<= 6; - } while (tmpBytesToRead > 0); - } - In UTF-8 writing code, the switches on "bytesToWrite" are - similarly unrolled loops. - - --------------------------------------------------------------------- */ - - diff --git a/taglib/toolkit/unicode.h b/taglib/toolkit/unicode.h deleted file mode 100644 index d3a869f0..00000000 --- a/taglib/toolkit/unicode.h +++ /dev/null @@ -1,149 +0,0 @@ -#ifndef TAGLIB_UNICODE_H -#define TAGLIB_UNICODE_H - -/******************************************************************************* - * * - * THIS FILE IS INCLUDED IN TAGLIB, BUT IS NOT COPYRIGHTED BY THE TAGLIB * - * AUTHORS, NOT PART OF THE TAGLIB API AND COULD GO AWAY AT ANY POINT IN TIME. * - * AS SUCH IT SHOULD BE CONSIERED FOR INTERNAL USE ONLY. * - * * - *******************************************************************************/ - -#ifndef DO_NOT_DOCUMENT // tell Doxygen not to document this header - -/* - * Copyright 2001 Unicode, Inc. - * - * Disclaimer - * - * This source code is provided as is by Unicode, Inc. No claims are - * made as to fitness for any particular purpose. No warranties of any - * kind are expressed or implied. The recipient agrees to determine - * applicability of information provided. If this file has been - * purchased on magnetic or optical media from Unicode, Inc., the - * sole remedy for any claim will be exchange of defective media - * within 90 days of receipt. - * - * Limitations on Rights to Redistribute This Code - * - * Unicode, Inc. hereby grants the right to freely use the information - * supplied in this file in the creation of products supporting the - * Unicode Standard, and to make copies of this file in any form - * for internal or external distribution as long as this notice - * remains attached. - */ - -/* - * This file has been modified by Scott Wheeler to remove - * the UTF32 conversion functions and to place the appropriate functions - * in their own C++ namespace. - */ - -/* --------------------------------------------------------------------- - - Conversions between UTF32, UTF-16, and UTF-8. Header file. - - Several functions are included here, forming a complete set of - conversions between the three formats. UTF-7 is not included - here, but is handled in a separate source file. - - Each of these routines takes pointers to input buffers and output - buffers. The input buffers are const. - - Each routine converts the text between *sourceStart and sourceEnd, - putting the result into the buffer between *targetStart and - targetEnd. Note: the end pointers are *after* the last item: e.g. - *(sourceEnd - 1) is the last item. - - The return result indicates whether the conversion was successful, - and if not, whether the problem was in the source or target buffers. - (Only the first encountered problem is indicated.) - - After the conversion, *sourceStart and *targetStart are both - updated to point to the end of last text successfully converted in - the respective buffers. - - Input parameters: - sourceStart - pointer to a pointer to the source buffer. - The contents of this are modified on return so that - it points at the next thing to be converted. - targetStart - similarly, pointer to pointer to the target buffer. - sourceEnd, targetEnd - respectively pointers to the ends of the - two buffers, for overflow checking only. - - These conversion functions take a ConversionFlags argument. When this - flag is set to strict, both irregular sequences and isolated surrogates - will cause an error. When the flag is set to lenient, both irregular - sequences and isolated surrogates are converted. - - Whether the flag is strict or lenient, all illegal sequences will cause - an error return. This includes sequences such as: , , - or in UTF-8, and values above 0x10FFFF in UTF-32. Conformant code - must check for illegal sequences. - - When the flag is set to lenient, characters over 0x10FFFF are converted - to the replacement character; otherwise (when the flag is set to strict) - they constitute an error. - - Output parameters: - The value "sourceIllegal" is returned from some routines if the input - sequence is malformed. When "sourceIllegal" is returned, the source - value will point to the illegal value that caused the problem. E.g., - in UTF-8 when a sequence is malformed, it points to the start of the - malformed sequence. - - Author: Mark E. Davis, 1994. - Rev History: Rick McGowan, fixes & updates May 2001. - Fixes & updates, Sept 2001. - ------------------------------------------------------------------------- */ - -/* --------------------------------------------------------------------- - The following 4 definitions are compiler-specific. - The C standard does not guarantee that wchar_t has at least - 16 bits, so wchar_t is no less portable than unsigned short! - All should be unsigned values to avoid sign extension during - bit mask & shift operations. ------------------------------------------------------------------------- */ - -/* Some fundamental constants */ -#define UNI_REPLACEMENT_CHAR (UTF32)0x0000FFFD -#define UNI_MAX_BMP (UTF32)0x0000FFFF -#define UNI_MAX_UTF16 (UTF32)0x0010FFFF -#define UNI_MAX_UTF32 (UTF32)0x7FFFFFFF - -namespace Unicode { - -typedef unsigned long UTF32; /* at least 32 bits */ -typedef wchar_t UTF16; /* TagLib assumes that wchar_t is sufficient for UTF-16. */ -typedef unsigned char UTF8; /* typically 8 bits */ -typedef unsigned char Boolean; /* 0 or 1 */ - -typedef enum { - conversionOK = 0, /* conversion successful */ - sourceExhausted = 1, /* partial character in source, but hit end */ - targetExhausted = 2, /* insuff. room in target for conversion */ - sourceIllegal = 3 /* source sequence is illegal/malformed */ -} ConversionResult; - -typedef enum { - strictConversion = 0, - lenientConversion -} ConversionFlags; - -ConversionResult ConvertUTF8toUTF16 ( - const UTF8** sourceStart, const UTF8* sourceEnd, - UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags); - -ConversionResult ConvertUTF16toUTF8 ( - const UTF16** sourceStart, const UTF16* sourceEnd, - UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags); - -Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd); - -} // namespace Unicode - -/* --------------------------------------------------------------------- */ - -#endif -#endif diff --git a/taglib/toolkit/utf8/checked.h b/taglib/toolkit/utf8/checked.h new file mode 100644 index 00000000..13311551 --- /dev/null +++ b/taglib/toolkit/utf8/checked.h @@ -0,0 +1,327 @@ +// 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_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 { + uint32_t cp; + public: + invalid_code_point(uint32_t cp) : cp(cp) {} + virtual const char* what() const throw() { return "Invalid code point"; } + uint32_t code_point() const {return cp;} + }; + + class invalid_utf8 : public exception { + uint8_t u8; + public: + invalid_utf8 (uint8_t u) : u8(u) {} + virtual const char* what() const throw() { return "Invalid UTF-8"; } + uint8_t utf8_octet() const {return u8;} + }; + + class invalid_utf16 : public exception { + uint16_t u16; + public: + invalid_utf16 (uint16_t u) : u16(u) {} + virtual const char* what() const throw() { return "Invalid UTF-16"; } + uint16_t utf16_word() const {return u16;} + }; + + class not_enough_room : public exception { + public: + virtual const char* what() const throw() { return "Not enough space"; } + }; + + /// The library API - functions intended to be called by the users + + template + octet_iterator append(uint32_t cp, octet_iterator result) + { + if (!utf8::internal::is_code_point_valid(cp)) + throw invalid_code_point(cp); + + 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; + } + + template + output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out, uint32_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: + throw not_enough_room(); + 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 uint32_t replacement_marker = utf8::internal::mask16(0xfffd); + return utf8::replace_invalid(start, end, out, replacement_marker); + } + + template + uint32_t next(octet_iterator& it, octet_iterator end) + { + uint32_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(*it); + case internal::INVALID_CODE_POINT : + throw invalid_code_point(cp); + } + return cp; + } + + template + uint32_t peek_next(octet_iterator it, octet_iterator end) + { + return utf8::next(it, end); + } + + template + uint32_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); + } + + /// Deprecated in versions that include "prior" + template + uint32_t previous(octet_iterator& it, octet_iterator pass_start) + { + octet_iterator end = it; + while (utf8::internal::is_trail(*(--it))) + if (it == pass_start) + throw invalid_utf8(*it); // error - no lead byte in the sequence + octet_iterator temp = it; + return utf8::next(temp, end); + } + + template + void advance (octet_iterator& it, distance_type n, octet_iterator end) + { + for (distance_type i = 0; 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) { + uint32_t cp = utf8::internal::mask16(*start++); + // Take care of surrogate pairs first + if (utf8::internal::is_lead_surrogate(cp)) { + if (start != end) { + uint32_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) { + uint32_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 : public std::iterator { + octet_iterator it; + octet_iterator range_start; + octet_iterator range_end; + public: + iterator () {} + explicit iterator (const octet_iterator& octet_it, + const octet_iterator& range_start, + const octet_iterator& range_end) : + it(octet_it), range_start(range_start), range_end(range_end) + { + 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; } + uint32_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 + +#endif //header guard + + diff --git a/taglib/toolkit/utf8/core.h b/taglib/toolkit/utf8/core.h new file mode 100644 index 00000000..693d388c --- /dev/null +++ b/taglib/toolkit/utf8/core.h @@ -0,0 +1,329 @@ +// 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 + +namespace utf8 +{ + // The typedefs for 8-bit, 16-bit and 32-bit unsigned integers + // You may need to change them to match your system. + // These typedefs have the same names as ones from cstdint, or boost/cstdint + typedef unsigned char uint8_t; + typedef unsigned short uint16_t; + typedef unsigned int uint32_t; + +// 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 uint16_t LEAD_SURROGATE_MIN = 0xd800u; + const uint16_t LEAD_SURROGATE_MAX = 0xdbffu; + const uint16_t TRAIL_SURROGATE_MIN = 0xdc00u; + const uint16_t TRAIL_SURROGATE_MAX = 0xdfffu; + const uint16_t LEAD_OFFSET = LEAD_SURROGATE_MIN - (0x10000 >> 10); + const uint32_t SURROGATE_OFFSET = 0x10000u - (LEAD_SURROGATE_MIN << 10) - TRAIL_SURROGATE_MIN; + + // Maximum valid value for a Unicode code point + const uint32_t CODE_POINT_MAX = 0x0010ffffu; + + template + inline uint8_t mask8(octet_type oc) + { + return static_cast(0xff & oc); + } + template + inline uint16_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); + } + + template + inline bool is_lead_surrogate(u16 cp) + { + return (cp >= LEAD_SURROGATE_MIN && cp <= LEAD_SURROGATE_MAX); + } + + template + inline bool is_trail_surrogate(u16 cp) + { + return (cp >= TRAIL_SURROGATE_MIN && cp <= TRAIL_SURROGATE_MAX); + } + + template + inline bool is_surrogate(u16 cp) + { + return (cp >= LEAD_SURROGATE_MIN && cp <= TRAIL_SURROGATE_MAX); + } + + template + inline bool is_code_point_valid(u32 cp) + { + return (cp <= CODE_POINT_MAX && !utf8::internal::is_surrogate(cp)); + } + + template + inline typename std::iterator_traits::difference_type + sequence_length(octet_iterator lead_it) + { + uint8_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; + } + + template + inline bool is_overlong_sequence(uint32_t cp, octet_difference_type 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, 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, uint32_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, uint32_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, uint32_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 += (*it) & 0x3f; + + return UTF8_OK; + } + + template + utf_error get_sequence_4(octet_iterator& it, octet_iterator end, uint32_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 += (utf8::internal::mask8(*it) << 6) & 0xfff; + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + 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, uint32_t& code_point) + { + // 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; + + uint32_t cp = 0; + // Determine the sequence length based on the lead octet + typedef typename std::iterator_traits::difference_type octet_difference_type; + const octet_difference_type 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) { + uint32_t ignored; + return utf8::internal::validate_next(it, end, ignored); + } + +} // namespace internal + + /// The library API - functions intended to be called by the users + + // Byte order mark + const uint8_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; + } + + template + inline bool is_valid(octet_iterator start, octet_iterator end) + { + return (utf8::find_invalid(start, end) == 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]) + ); + } + + //Deprecated in release 2.3 + template + inline bool is_bom (octet_iterator it) + { + return ( + (utf8::internal::mask8(*it++)) == bom[0] && + (utf8::internal::mask8(*it++)) == bom[1] && + (utf8::internal::mask8(*it)) == bom[2] + ); + } +} // namespace utf8 + +#endif // header guard + + diff --git a/tests/test_string.cpp b/tests/test_string.cpp index d3661c24..20d96559 100644 --- a/tests/test_string.cpp +++ b/tests/test_string.cpp @@ -49,6 +49,7 @@ class TestString : public CppUnit::TestFixture CPPUNIT_TEST(testUpper); CPPUNIT_TEST(testEncodeNonLatin1); CPPUNIT_TEST(testEncodeEmpty); + CPPUNIT_TEST(testEncodeNonBMP); CPPUNIT_TEST(testIterator); CPPUNIT_TEST(testRedundantUTF8); CPPUNIT_TEST_SUITE_END(); @@ -313,6 +314,13 @@ public: CPPUNIT_ASSERT(empty.to8Bit(true).empty()); } + void testEncodeNonBMP() + { + const ByteVector a("\xFF\xFE\x3C\xD8\x50\xDD\x40\xD8\xF5\xDC\x3C\xD8\x00\xDE", 14); + const ByteVector b("\xF0\x9F\x85\x90\xF0\xA0\x83\xB5\xF0\x9F\x88\x80"); + CPPUNIT_ASSERT_EQUAL(b, String(a, String::UTF16).data(String::UTF8)); + } + void testIterator() { String s1 = "taglib string"; -- 2.40.0