From 03398e94b013b78781ac77be3755374369cbec4b Mon Sep 17 00:00:00 2001 From: Arvid Norberg Date: Thu, 13 Nov 2008 06:39:08 +0000 Subject: [PATCH] replaced utf8 conversion with code from unicode.org --- Jamfile | 1 + include/libtorrent/ConvertUTF.h | 142 +++++++++ include/libtorrent/config.hpp | 2 +- include/libtorrent/utf8.hpp | 227 +++++--------- src/ConvertUTF.cpp | 539 ++++++++++++++++++++++++++++++++ src/Makefile.am | 5 +- src/storage.cpp | 33 +- 7 files changed, 788 insertions(+), 161 deletions(-) create mode 100644 include/libtorrent/ConvertUTF.h create mode 100644 src/ConvertUTF.cpp diff --git a/Jamfile b/Jamfile index 200366ec7..a8e9cbc4a 100755 --- a/Jamfile +++ b/Jamfile @@ -327,6 +327,7 @@ SOURCES = broadcast_socket magnet_uri parse_url + ConvertUTF # -- extensions -- metadata_transfer diff --git a/include/libtorrent/ConvertUTF.h b/include/libtorrent/ConvertUTF.h new file mode 100644 index 000000000..9a2954882 --- /dev/null +++ b/include/libtorrent/ConvertUTF.h @@ -0,0 +1,142 @@ +/* + * Copyright 2001-2004 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. + */ + +/* --------------------------------------------------------------------- + + Conversions between UTF32, UTF-16, and UTF-8. Header file. + + Several funtions 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. +------------------------------------------------------------------------ */ + +#include + +typedef boost::uint32_t UTF32; /* at least 32 bits */ +typedef boost::uint16_t UTF16; /* at least 16 bits */ +typedef boost::uint8_t UTF8; /* typically 8 bits */ +typedef unsigned char Boolean; /* 0 or 1 */ + +/* 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 +#define UNI_MAX_LEGAL_UTF32 (UTF32)0x0010FFFF + +typedef enum { + conversionOK, /* conversion successful */ + sourceExhausted, /* partial character in source, but hit end */ + targetExhausted, /* insuff. room in target for conversion */ + sourceIllegal /* 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); + +ConversionResult ConvertUTF8toUTF32 ( + const UTF8** sourceStart, const UTF8* sourceEnd, + UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags); + +ConversionResult ConvertUTF32toUTF8 ( + const UTF32** sourceStart, const UTF32* sourceEnd, + UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags); + +ConversionResult ConvertUTF16toUTF32 ( + const UTF16** sourceStart, const UTF16* sourceEnd, + UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags); + +ConversionResult ConvertUTF32toUTF16 ( + const UTF32** sourceStart, const UTF32* sourceEnd, + UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags); + +Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd); + +/* --------------------------------------------------------------------- */ diff --git a/include/libtorrent/config.hpp b/include/libtorrent/config.hpp index a1db4b412..d70e1b6bd 100644 --- a/include/libtorrent/config.hpp +++ b/include/libtorrent/config.hpp @@ -84,7 +84,7 @@ POSSIBILITY OF SUCH DAMAGE. // should wpath or path be used? #if defined UNICODE && !defined BOOST_FILESYSTEM_NARROW_ONLY \ - && BOOST_VERSION >= 103400 && defined WIN32 + && BOOST_VERSION >= 103400 #define TORRENT_USE_WPATH 1 #else #define TORRENT_USE_WPATH 0 diff --git a/include/libtorrent/utf8.hpp b/include/libtorrent/utf8.hpp index 157a7fdba..6c63c2417 100644 --- a/include/libtorrent/utf8.hpp +++ b/include/libtorrent/utf8.hpp @@ -1,161 +1,104 @@ /* - Copyright (C) 2004-2005 Cory Nelson - This software is provided 'as-is', without any express or implied - warranty. In no event will the authors be held liable for any damages - arising from the use of this software. +Copyright (c) 2006, Arvid Norberg +All rights reserved. - Permission is granted to anyone to use this software for any purpose, - including commercial applications, and to alter it and redistribute it - freely, subject to the following restrictions: +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions +are met: + + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in + the documentation and/or other materials provided with the distribution. + * Neither the name of the author nor the names of its + contributors may be used to endorse or promote products derived + from this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE. - 1. The origin of this software must not be misrepresented; you must not - claim that you wrote the original software. If you use this software - in a product, an acknowledgment in the product documentation would be - appreciated but is not required. - 2. Altered source versions must be plainly marked as such, and must not be - misrepresented as being the original software. - 3. This notice may not be removed or altered from any source distribution. */ -// namespaces added by Arvid Norberg - -#ifndef __UTF8_H__ -#define __UTF8_H__ +#ifndef TORRENT_UTF8_HPP_INCLUDED +#define TORRENT_UTF8_HPP_INCLUDED #include -#include -#include #include +#include "libtorrent/ConvertUTF.h" -namespace libtorrent { -namespace detail { - -template -wchar_t decode_utf8_mb(InputIterator &iter, InputIterator last) +namespace libtorrent { - if (iter == last) throw std::runtime_error("incomplete UTF-8 sequence"); - if (((*iter) & 0xc0) != 0x80) throw std::runtime_error("invalid UTF-8 sequence"); - return (wchar_t)((*iter++) & 0x3f); -} - -template -wchar_t decode_utf8(InputIterator &iter, InputIterator last) -{ - wchar_t ret; - - if (((*iter) & 0x80) == 0) // one byte + inline int utf8_wchar(const std::string &utf8, std::wstring &wide) { - ret = *iter++; + // allocate space for worst-case + wide.resize(utf8.size()); + wchar_t const* dst_start = &wide[0]; + char const* src_start = &utf8[0]; + ConversionResult ret; + if (sizeof(wchar_t) == sizeof(UTF32)) + { + ret = ConvertUTF8toUTF32((const UTF8**)&src_start, (const UTF8*)src_start + + utf8.size(), (UTF32**)&dst_start, (UTF32*)dst_start + wide.size() + , lenientConversion); + wide.resize(dst_start - &wide[0]); + return ret; + } + else if (sizeof(wchar_t) == sizeof(UTF16)) + { + ret = ConvertUTF8toUTF16((const UTF8**)&src_start, (const UTF8*)src_start + + utf8.size(), (UTF16**)&dst_start, (UTF16*)dst_start + wide.size() + , lenientConversion); + wide.resize(dst_start - &wide[0]); + return ret; + } + else + { + return sourceIllegal; + } } - else if (((*iter) & 0xe0) == 0xc0) // two bytes +/* + inline int wchar_utf8(const std::wstring &wide, std::string &utf8) { - wchar_t byte1 = (*iter++) & 0x1f; - wchar_t byte2 = decode_utf8_mb(iter, last); - ret = (byte1 << 6) | byte2; + // allocate space for worst-case + utf8.resize(wide.size() * 6); + char* dst_start = &utf8[0]; + wchar_t const* src_start = &wide[0]; + ConversionResult ret; + if (sizeof(wchar_t) == sizeof(UTF32)) + { + ret = ConvertUTF32toUTF8((const UTF32**)&src_start, (const UTF32*)src_start + + wide.size(), (UTF8**)&dst_start, (UTF8*)dst_start + utf8.size() + , lenientConversion); + utf8.resize(dst_start - &utf8[0]); + return ret; + } + else if (sizeof(wchar_t) == sizeof(UTF16)) + { + ret = ConvertUTF16toUTF8((const UTF16**)&src_start, (const UTF16*)src_start + + wide.size(), (UTF8**)&dst_start, (UTF8*)dst_start + utf8.size() + , lenientConversion); + utf8.resize(dst_start - &utf8[0]); + return ret; + } + else + { + return sourceIllegal; + } } - else if (((*iter) & 0xf0) == 0xe0) // three bytes - { - wchar_t byte1 = (*iter++) & 0x0f; - wchar_t byte2 = decode_utf8_mb(iter, last); - wchar_t byte3 = decode_utf8_mb(iter, last); - ret = (byte1 << 12) | (byte2 << 6) | byte3; - } - // TODO: support surrogate pairs - else throw std::runtime_error("UTF-8 not convertable to UTF-16"); - - return ret; -} - -template -OutputIterator utf8_wchar(InputIterator first, InputIterator last, OutputIterator dest) -{ - for(; first!=last; ++dest) - *dest = decode_utf8(first, last); - return dest; -} - -template -void encode_wchar(InputIterator iter, OutputIterator &dest) -{ - if(*iter <= 0x007F) - { - *dest=(char)*iter; - ++dest; - } - else if(*iter <= 0x07FF) - { - *dest = (char)( - 0xC0 | - ((*iter & 0x07C0) >> 6) - ); - ++dest; - - *dest = (char)( - 0x80 | - (*iter & 0x003F) - ); - ++dest; - } - else if(*iter <= 0xFFFF) - { - *dest = (char)( - 0xE0 | - ((*iter & 0xF000) >> 12) - ); - ++dest; - - *dest = (char)( - 0x80 | - ((*iter & 0x0FC0) >> 6) - ); - ++dest; - - *dest = (char)( - 0x80 | - (*iter & 0x003F) - ); - ++dest; - } -} - -template -OutputIterator wchar_utf8(InputIterator first, InputIterator last, OutputIterator dest) -{ - for(; first!=last; ++first) - encode_wchar(first, dest); - return dest; -} - -} - -inline void utf8_wchar(const std::string &utf8, std::wstring &wide) -{ - wide.clear(); - detail::utf8_wchar(utf8.begin(), utf8.end(), std::back_inserter(wide)); -} - -inline std::wstring utf8_wchar(const std::string &str) -{ - std::wstring ret; - utf8_wchar(str, ret); - return ret; -} - -inline void wchar_utf8(const std::wstring &wide, std::string &utf8) -{ - utf8.clear(); - detail::wchar_utf8(wide.begin(), wide.end(), std::back_inserter(utf8)); -} - -inline std::string wchar_utf8(const std::wstring &str) -{ - std::string ret; - wchar_utf8(str, ret); - return ret; -} - +*/ } #endif + diff --git a/src/ConvertUTF.cpp b/src/ConvertUTF.cpp new file mode 100644 index 000000000..9b3deebd6 --- /dev/null +++ b/src/ConvertUTF.cpp @@ -0,0 +1,539 @@ +/* + * Copyright 2001-2004 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. + */ + +/* --------------------------------------------------------------------- + + 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. + June 2002: Tim Dodd added detection and handling of incomplete + source sequences, enhanced error detection, added casts + to eliminate compiler warnings. + July 2003: slight mods to back out aggressive FFFE detection. + Jan 2004: updated switches in from-UTF8 conversions. + Oct 2004: updated to use UNI_MAX_LEGAL_UTF32 in UTF-32 conversions. + + See the header file "ConvertUTF.h" for complete documentation. + +------------------------------------------------------------------------ */ + + +#include "ConvertUTF.h" +#ifdef CVTUTF_DEBUG +#include +#endif + +static const int halfShift = 10; /* used for shifting by 10 bits */ + +static const UTF32 halfBase = 0x0010000UL; +static const UTF32 halfMask = 0x3FFUL; + +#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 + +/* --------------------------------------------------------------------- */ + +ConversionResult ConvertUTF32toUTF16 ( + const UTF32** sourceStart, const UTF32* sourceEnd, + UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags) { + ConversionResult result = conversionOK; + const UTF32* source = *sourceStart; + UTF16* target = *targetStart; + while (source < sourceEnd) { + UTF32 ch; + if (target >= targetEnd) { + result = targetExhausted; break; + } + ch = *source++; + if (ch <= UNI_MAX_BMP) { /* Target is a character <= 0xFFFF */ + /* UTF-16 surrogate values are illegal in UTF-32; 0xffff or 0xfffe are both reserved values */ + if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) { + if (flags == strictConversion) { + --source; /* return to the illegal value itself */ + result = sourceIllegal; + break; + } else { + *target++ = UNI_REPLACEMENT_CHAR; + } + } else { + *target++ = (UTF16)ch; /* normal case */ + } + } else if (ch > UNI_MAX_LEGAL_UTF32) { + if (flags == strictConversion) { + result = sourceIllegal; + } else { + *target++ = UNI_REPLACEMENT_CHAR; + } + } else { + /* target is a character in range 0xFFFF - 0x10FFFF. */ + if (target + 1 >= targetEnd) { + --source; /* Back up source pointer! */ + result = targetExhausted; break; + } + ch -= halfBase; + *target++ = (UTF16)((ch >> halfShift) + UNI_SUR_HIGH_START); + *target++ = (UTF16)((ch & halfMask) + UNI_SUR_LOW_START); + } + } + *sourceStart = source; + *targetStart = target; + return result; +} + +/* --------------------------------------------------------------------- */ + +ConversionResult ConvertUTF16toUTF32 ( + const UTF16** sourceStart, const UTF16* sourceEnd, + UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags) { + ConversionResult result = conversionOK; + const UTF16* source = *sourceStart; + UTF32* target = *targetStart; + UTF32 ch, ch2; + while (source < sourceEnd) { + 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) { + /* If the 16 bits following the high surrogate are in the source buffer... */ + if (source < sourceEnd) { + ch2 = *source; + /* If it's a low surrogate, convert to UTF32. */ + 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 { /* We don't have the 16 bits following the high surrogate. */ + --source; /* return to the high surrogate */ + result = sourceExhausted; + break; + } + } else if (flags == strictConversion) { + /* UTF-16 surrogate values are illegal in UTF-32 */ + if (ch >= UNI_SUR_LOW_START && ch <= UNI_SUR_LOW_END) { + --source; /* return to the illegal value itself */ + result = sourceIllegal; + break; + } + } + if (target >= targetEnd) { + source = oldSource; /* Back up source pointer! */ + result = targetExhausted; break; + } + *target++ = ch; + } + *sourceStart = source; + *targetStart = target; +#ifdef CVTUTF_DEBUG +if (result == sourceIllegal) { + fprintf(stderr, "ConvertUTF16toUTF32 illegal seq 0x%04x,%04x\n", ch, ch2); + fflush(stderr); +} +#endif + return result; +} + +/* --------------------------------------------------------------------- */ + +/* + * 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. + * Note that *legal* UTF-8 values can't have 4 or 5-bytes. The table is + * left as-is for anyone who may want to do such conversion, which was + * allowed in earlier algorithms. + */ +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... etc.). Remember that sequencs + * for *legal* UTF-8 will be 4 or fewer bytes total. + */ +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) { + /* If the 16 bits following the high surrogate are in the source buffer... */ + if (source < sourceEnd) { + UTF32 ch2 = *source; + /* If it's a low surrogate, convert to UTF32. */ + 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 { /* We don't have the 16 bits following the high surrogate. */ + --source; /* return to the high surrogate */ + result = sourceExhausted; + break; + } + } else if (flags == strictConversion) { + /* UTF-16 surrogate values are illegal in UTF-32 */ + if (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)0x110000) { bytesToWrite = 4; + } else { bytesToWrite = 3; + ch = UNI_REPLACEMENT_CHAR; + } + + 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 = (UTF8)((ch | byteMark) & byteMask); ch >>= 6; + case 3: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6; + case 2: *--target = (UTF8)((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 0xED: if (a > 0x9F) 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 5: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */ + case 4: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */ + 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 */ + /* UTF-16 surrogate values are illegal in UTF-32 */ + if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) { + if (flags == strictConversion) { + source -= (extraBytesToRead+1); /* return to the illegal value itself */ + result = sourceIllegal; + break; + } else { + *target++ = UNI_REPLACEMENT_CHAR; + } + } 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++ = (UTF16)((ch & halfMask) + UNI_SUR_LOW_START); + } + } + *sourceStart = source; + *targetStart = target; + return result; +} + +/* --------------------------------------------------------------------- */ + +ConversionResult ConvertUTF32toUTF8 ( + const UTF32** sourceStart, const UTF32* sourceEnd, + UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags) { + ConversionResult result = conversionOK; + const UTF32* source = *sourceStart; + UTF8* target = *targetStart; + while (source < sourceEnd) { + UTF32 ch; + unsigned short bytesToWrite = 0; + const UTF32 byteMask = 0xBF; + const UTF32 byteMark = 0x80; + ch = *source++; + if (flags == strictConversion ) { + /* UTF-16 surrogate values are illegal in UTF-32 */ + if (ch >= UNI_SUR_HIGH_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. Turn any + * illegally large UTF32 things (> Plane 17) into replacement chars. + */ + if (ch < (UTF32)0x80) { bytesToWrite = 1; + } else if (ch < (UTF32)0x800) { bytesToWrite = 2; + } else if (ch < (UTF32)0x10000) { bytesToWrite = 3; + } else if (ch <= UNI_MAX_LEGAL_UTF32) { bytesToWrite = 4; + } else { bytesToWrite = 3; + ch = UNI_REPLACEMENT_CHAR; + result = sourceIllegal; + } + + target += bytesToWrite; + if (target > targetEnd) { + --source; /* Back up source pointer! */ + target -= bytesToWrite; result = targetExhausted; break; + } + switch (bytesToWrite) { /* note: everything falls through. */ + case 4: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6; + case 3: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6; + case 2: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6; + case 1: *--target = (UTF8) (ch | firstByteMark[bytesToWrite]); + } + target += bytesToWrite; + } + *sourceStart = source; + *targetStart = target; + return result; +} + +/* --------------------------------------------------------------------- */ + +ConversionResult ConvertUTF8toUTF32 ( + const UTF8** sourceStart, const UTF8* sourceEnd, + UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags) { + ConversionResult result = conversionOK; + const UTF8* source = *sourceStart; + UTF32* 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 5: ch += *source++; ch <<= 6; + case 4: ch += *source++; ch <<= 6; + 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 the source pointer! */ + result = targetExhausted; break; + } + if (ch <= UNI_MAX_LEGAL_UTF32) { + /* + * UTF-16 surrogate values are illegal in UTF-32, and anything + * over Plane 17 (> 0x10FFFF) is illegal. + */ + if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) { + if (flags == strictConversion) { + source -= (extraBytesToRead+1); /* return to the illegal value itself */ + result = sourceIllegal; + break; + } else { + *target++ = UNI_REPLACEMENT_CHAR; + } + } else { + *target++ = ch; + } + } else { /* i.e., ch > UNI_MAX_LEGAL_UTF32 */ + result = sourceIllegal; + *target++ = UNI_REPLACEMENT_CHAR; + } + } + *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/src/Makefile.am b/src/Makefile.am index 42a9bac70..f48883b02 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -24,7 +24,7 @@ logger.cpp file_pool.cpp ut_pex.cpp lsd.cpp upnp.cpp instantiate_connection.cpp socks5_stream.cpp socks4_stream.cpp http_stream.cpp connection_queue.cpp \ disk_io_thread.cpp ut_metadata.cpp magnet_uri.cpp udp_socket.cpp smart_ban.cpp \ http_parser.cpp gzip.cpp disk_buffer_holder.cpp create_torrent.cpp GeoIP.c \ -parse_url.cpp file_storage.cpp error_code.cpp $(kademlia_sources) +parse_url.cpp file_storage.cpp error_code.cpp ConvertUTF.cpp $(kademlia_sources) # mapped_storage.cpp noinst_HEADERS = \ @@ -101,7 +101,8 @@ $(top_srcdir)/include/libtorrent/udp_tracker_connection.hpp \ $(top_srcdir)/include/libtorrent/utf8.hpp \ $(top_srcdir)/include/libtorrent/xml_parse.hpp \ $(top_srcdir)/include/libtorrent/variant_stream.hpp \ -$(top_srcdir)/include/libtorrent/version.hpp +$(top_srcdir)/include/libtorrent/version.hpp \ +$(top_srcdir)/include/libtorrent/ConvertUTF.h libtorrent_rasterbar_la_LDFLAGS = $(LDFLAGS) -version-info $(INTERFACE_VERSION_INFO) diff --git a/src/storage.cpp b/src/storage.cpp index 5e689cf63..ae334a389 100644 --- a/src/storage.cpp +++ b/src/storage.cpp @@ -115,24 +115,25 @@ namespace libtorrent { std::wstring safe_convert(std::string const& s) { - try + std::wstring ret; + int result = libtorrent::utf8_wchar(s, ret); +#ifndef BOOST_WINDOWS + return ret; +#else + if (result == 0) return ret; + + ret.clear(); + const char* end = &s[0] + s.size(); + for (const char* i = &s[0]; i < end;) { - return libtorrent::utf8_wchar(s); - } - catch (std::exception) - { - std::wstring ret; - const char* end = &s[0] + s.size(); - for (const char* i = &s[0]; i < end;) - { - wchar_t c = '.'; - int result = std::mbtowc(&c, i, end - i); - if (result > 0) i += result; - else ++i; - ret += c; - } - return ret; + wchar_t c = '.'; + int result = std::mbtowc(&c, i, end - i); + if (result > 0) i += result; + else ++i; + ret += c; } + return ret; +#endif } } #endif