X-Git-Url: https://git.saurik.com/apple/icu.git/blobdiff_plain/374ca955a76ecab1204ca8bfa63ff9238d998416..0f5d89e82340278ed3d7d50029f37cab2c41a57e:/icuSources/common/rbbidata.cpp?ds=sidebyside diff --git a/icuSources/common/rbbidata.cpp b/icuSources/common/rbbidata.cpp index aa33c610..1d4c9e58 100644 --- a/icuSources/common/rbbidata.cpp +++ b/icuSources/common/rbbidata.cpp @@ -1,6 +1,8 @@ +// © 2016 and later: Unicode, Inc. and others. +// License & terms of use: http://www.unicode.org/copyright.html /* *************************************************************************** -* Copyright (C) 1999-2004 International Business Machines Corporation * +* Copyright (C) 1999-2014 International Business Machines Corporation * * and others. All rights reserved. * *************************************************************************** */ @@ -12,7 +14,7 @@ #include "unicode/utypes.h" #include "rbbidata.h" #include "rbbirb.h" -#include "utrie.h" +#include "utrie2.h" #include "udatamem.h" #include "cmemory.h" #include "cstring.h" @@ -21,23 +23,6 @@ #include "uassert.h" -//----------------------------------------------------------------------------------- -// -// Trie access folding function. Copied as-is from properties code in uchar.c -// -//----------------------------------------------------------------------------------- -U_CDECL_BEGIN -static int32_t U_CALLCONV -getFoldingOffset(uint32_t data) { - /* if bit 15 is set, then the folding offset is in bits 14..0 of the 16-bit trie result */ - if(data&0x8000) { - return (int32_t)(data&0x7fff); - } else { - return 0; - } -} -U_CDECL_END - U_NAMESPACE_BEGIN //----------------------------------------------------------------------------- @@ -46,61 +31,93 @@ U_NAMESPACE_BEGIN // //----------------------------------------------------------------------------- RBBIDataWrapper::RBBIDataWrapper(const RBBIDataHeader *data, UErrorCode &status) { + init0(); + init(data, status); +} + +RBBIDataWrapper::RBBIDataWrapper(const RBBIDataHeader *data, enum EDontAdopt, UErrorCode &status) { + init0(); init(data, status); + fDontFreeData = TRUE; } RBBIDataWrapper::RBBIDataWrapper(UDataMemory* udm, UErrorCode &status) { - const RBBIDataHeader *d = (const RBBIDataHeader *) - // ((char *)&(udm->pHeader->info) + udm->pHeader->info.size); - // taking into consideration the padding added in by udata_write - ((char *)(udm->pHeader) + udm->pHeader->dataHeader.headerSize); - init(d, status); + init0(); + if (U_FAILURE(status)) { + return; + } + const DataHeader *dh = udm->pHeader; + int32_t headerSize = dh->dataHeader.headerSize; + if ( !(headerSize >= 20 && + dh->info.isBigEndian == U_IS_BIG_ENDIAN && + dh->info.charsetFamily == U_CHARSET_FAMILY && + dh->info.dataFormat[0] == 0x42 && // dataFormat="Brk " + dh->info.dataFormat[1] == 0x72 && + dh->info.dataFormat[2] == 0x6b && + dh->info.dataFormat[3] == 0x20 && + isDataVersionAcceptable(dh->info.formatVersion)) + ) { + status = U_INVALID_FORMAT_ERROR; + return; + } + const char *dataAsBytes = reinterpret_cast(dh); + const RBBIDataHeader *rbbidh = reinterpret_cast(dataAsBytes + headerSize); + init(rbbidh, status); fUDataMem = udm; } +UBool RBBIDataWrapper::isDataVersionAcceptable(const UVersionInfo version) { + return RBBI_DATA_FORMAT_VERSION[0] == version[0]; +} + + //----------------------------------------------------------------------------- // // init(). Does most of the work of construction, shared between the // constructors. // //----------------------------------------------------------------------------- +void RBBIDataWrapper::init0() { + fHeader = NULL; + fForwardTable = NULL; + fReverseTable = NULL; + fRuleSource = NULL; + fRuleStatusTable = NULL; + fTrie = NULL; + fUDataMem = NULL; + fRefCount = 0; + fDontFreeData = TRUE; +} + void RBBIDataWrapper::init(const RBBIDataHeader *data, UErrorCode &status) { if (U_FAILURE(status)) { return; } fHeader = data; - if (fHeader->fMagic != 0xb1a0) { - status = U_BRK_INTERNAL_ERROR; + if (fHeader->fMagic != 0xb1a0 || !isDataVersionAcceptable(fHeader->fFormatVersion)) { + status = U_INVALID_FORMAT_ERROR; return; } + // Note: in ICU version 3.2 and earlier, there was a formatVersion 1 + // that is no longer supported. At that time fFormatVersion was + // an int32_t field, rather than an array of 4 bytes. - fUDataMem = NULL; - fReverseTable = NULL; - fSafeFwdTable = NULL; - fSafeRevTable = NULL; + fDontFreeData = FALSE; if (data->fFTableLen != 0) { fForwardTable = (RBBIStateTable *)((char *)data + fHeader->fFTable); } if (data->fRTableLen != 0) { fReverseTable = (RBBIStateTable *)((char *)data + fHeader->fRTable); } - if (data->fSFTableLen != 0) { - fSafeFwdTable = (RBBIStateTable *)((char *)data + fHeader->fSFTable); - } - if (data->fSRTableLen != 0) { - fSafeRevTable = (RBBIStateTable *)((char *)data + fHeader->fSRTable); - } - - utrie_unserialize(&fTrie, - (uint8_t *)data + fHeader->fTrie, - fHeader->fTrieLen, - &status); + fTrie = utrie2_openFromSerialized(UTRIE2_16_VALUE_BITS, + (uint8_t *)data + fHeader->fTrie, + fHeader->fTrieLen, + NULL, // *actual length + &status); if (U_FAILURE(status)) { return; } - fTrie.getFoldingOffset=getFoldingOffset; - fRuleSource = (UChar *)((char *)data + fHeader->fRuleSource); fRuleString.setTo(TRUE, fRuleSource, -1); @@ -125,9 +142,11 @@ void RBBIDataWrapper::init(const RBBIDataHeader *data, UErrorCode &status) { //----------------------------------------------------------------------------- RBBIDataWrapper::~RBBIDataWrapper() { U_ASSERT(fRefCount == 0); + utrie2_close(fTrie); + fTrie = NULL; if (fUDataMem) { udata_close(fUDataMem); - } else { + } else if (!fDontFreeData) { uprv_free((void *)fHeader); } } @@ -231,28 +250,28 @@ void RBBIDataWrapper::printTable(const char *heading, const RBBIStateTable *tab #endif -#ifdef RBBI_DEBUG void RBBIDataWrapper::printData() { +#ifdef RBBI_DEBUG RBBIDebugPrintf("RBBI Data at %p\n", (void *)fHeader); - RBBIDebugPrintf(" Version = %d\n", fHeader->fVersion); + RBBIDebugPrintf(" Version = {%d %d %d %d}\n", fHeader->fFormatVersion[0], fHeader->fFormatVersion[1], + fHeader->fFormatVersion[2], fHeader->fFormatVersion[3]); RBBIDebugPrintf(" total length of data = %d\n", fHeader->fLength); RBBIDebugPrintf(" number of character categories = %d\n\n", fHeader->fCatCount); printTable("Forward State Transition Table", fForwardTable); printTable("Reverse State Transition Table", fReverseTable); - printTable("Safe Forward State Transition Table", fSafeFwdTable); - printTable("Safe Reverse State Transition Table", fSafeRevTable); RBBIDebugPrintf("\nOrignal Rules source:\n"); for (int32_t c=0; fRuleSource[c] != 0; c++) { RBBIDebugPrintf("%c", fRuleSource[c]); } RBBIDebugPrintf("\n\n"); -} #endif +} U_NAMESPACE_END +U_NAMESPACE_USE //----------------------------------------------------------------------------- // @@ -267,6 +286,10 @@ ubrk_swap(const UDataSwapper *ds, const void *inData, int32_t length, void *outD if (status == NULL || U_FAILURE(*status)) { return 0; } + if(ds==NULL || inData==NULL || length<-1 || (length>0 && outData==NULL)) { + *status=U_ILLEGAL_ARGUMENT_ERROR; + return 0; + } // // Check that the data header is for for break data. @@ -277,7 +300,7 @@ ubrk_swap(const UDataSwapper *ds, const void *inData, int32_t length, void *outD pInfo->dataFormat[1]==0x72 && pInfo->dataFormat[2]==0x6b && pInfo->dataFormat[3]==0x20 && - pInfo->formatVersion[0]==3 )) { + RBBIDataWrapper::isDataVersionAcceptable(pInfo->formatVersion) )) { udata_printError(ds, "ubrk_swap(): data format %02x.%02x.%02x.%02x (format version %02x) is not recognized\n", pInfo->dataFormat[0], pInfo->dataFormat[1], pInfo->dataFormat[2], pInfo->dataFormat[3], @@ -300,10 +323,9 @@ ubrk_swap(const UDataSwapper *ds, const void *inData, int32_t length, void *outD // const uint8_t *inBytes =(const uint8_t *)inData+headerSize; RBBIDataHeader *rbbiDH = (RBBIDataHeader *)inBytes; - if (ds->readUInt32(rbbiDH->fMagic) != 0xb1a0 || - ds->readUInt32(rbbiDH->fVersion) != 1 || - ds->readUInt32(rbbiDH->fLength) < sizeof(RBBIDataHeader)) - { + if (ds->readUInt32(rbbiDH->fMagic) != 0xb1a0 || + !RBBIDataWrapper::isDataVersionAcceptable(rbbiDH->fFormatVersion) || + ds->readUInt32(rbbiDH->fLength) < sizeof(RBBIDataHeader)) { udata_printError(ds, "ubrk_swap(): RBBI Data header is invalid.\n"); *status=U_UNSUPPORTED_ERROR; return 0; @@ -312,7 +334,8 @@ ubrk_swap(const UDataSwapper *ds, const void *inData, int32_t length, void *outD // // Prefight operation? Just return the size // - int32_t totalSize = headerSize + ds->readUInt32(rbbiDH->fLength); + int32_t breakDataLength = ds->readUInt32(rbbiDH->fLength); + int32_t totalSize = headerSize + breakDataLength; if (length < 0) { return totalSize; } @@ -320,15 +343,12 @@ ubrk_swap(const UDataSwapper *ds, const void *inData, int32_t length, void *outD // // Check that length passed in is consistent with length from RBBI data header. // - if (length > 0) { - length -= headerSize; - if ((uint32_t)length < ds->readUInt32(rbbiDH->fLength)) { - udata_printError(ds, "ubrk_swap(): too few bytes (%d after ICU Data header) for break data.\n", - length); - *status=U_INDEX_OUTOFBOUNDS_ERROR; - return 0; + if (length < totalSize) { + udata_printError(ds, "ubrk_swap(): too few bytes (%d after ICU Data header) for break data.\n", + breakDataLength); + *status=U_INDEX_OUTOFBOUNDS_ERROR; + return 0; } - } // @@ -336,7 +356,9 @@ ubrk_swap(const UDataSwapper *ds, const void *inData, int32_t length, void *outD // we need to reference the header to locate the data, and an // inplace swap of the header leaves it unusable. // - uint8_t *outBytes = (uint8_t *)outData + headerSize; + uint8_t *outBytes = (uint8_t *)outData + headerSize; + RBBIDataHeader *outputDH = (RBBIDataHeader *)outBytes; + int32_t tableStartOffset; int32_t tableLength; @@ -346,15 +368,14 @@ ubrk_swap(const UDataSwapper *ds, const void *inData, int32_t length, void *outD // when originally created. Any unused space between items needs to be zero. // if (inBytes != outBytes) { - uprv_memset(outBytes, 0, length); + uprv_memset(outBytes, 0, breakDataLength); } // // Each state table begins with several 32 bit fields. Calculate the size // in bytes of these. // - RBBIStateTable *stp = NULL; - int32_t topSize = (char *)stp->fTableData - (char *)stp; + int32_t topSize = offsetof(RBBIStateTable, fTableData); // Forward state table. tableStartOffset = ds->readUInt32(rbbiDH->fFTable); @@ -378,31 +399,9 @@ ubrk_swap(const UDataSwapper *ds, const void *inData, int32_t length, void *outD outBytes+tableStartOffset+topSize, status); } - // Safe Forward state table. Same layout as forward table, above. - tableStartOffset = ds->readUInt32(rbbiDH->fSFTable); - tableLength = ds->readUInt32(rbbiDH->fSFTableLen); - - if (tableLength > 0) { - ds->swapArray32(ds, inBytes+tableStartOffset, topSize, - outBytes+tableStartOffset, status); - ds->swapArray16(ds, inBytes+tableStartOffset+topSize, tableLength-topSize, - outBytes+tableStartOffset+topSize, status); - } - - // Safe Reverse state table. Same layout as forward table, above. - tableStartOffset = ds->readUInt32(rbbiDH->fSRTable); - tableLength = ds->readUInt32(rbbiDH->fSRTableLen); - - if (tableLength > 0) { - ds->swapArray32(ds, inBytes+tableStartOffset, topSize, - outBytes+tableStartOffset, status); - ds->swapArray16(ds, inBytes+tableStartOffset+topSize, tableLength-topSize, - outBytes+tableStartOffset+topSize, status); - } - // Trie table for character categories - utrie_swap(ds, inBytes+ds->readUInt32(rbbiDH->fTrie), ds->readUInt32(rbbiDH->fTrieLen), - outBytes+ds->readUInt32(rbbiDH->fTrie), status); + utrie2_swap(ds, inBytes+ds->readUInt32(rbbiDH->fTrie), ds->readUInt32(rbbiDH->fTrieLen), + outBytes+ds->readUInt32(rbbiDH->fTrie), status); // Source Rules Text. It's UChar data ds->swapArray16(ds, inBytes+ds->readUInt32(rbbiDH->fRuleSource), ds->readUInt32(rbbiDH->fRuleSourceLen), @@ -412,8 +411,12 @@ ubrk_swap(const UDataSwapper *ds, const void *inData, int32_t length, void *outD ds->swapArray32(ds, inBytes+ds->readUInt32(rbbiDH->fStatusTable), ds->readUInt32(rbbiDH->fStatusTableLen), outBytes+ds->readUInt32(rbbiDH->fStatusTable), status); - // And, last, the header. All 32 bit values. - ds->swapArray32(ds, inBytes, sizeof(RBBIDataHeader), outBytes, status); + // And, last, the header. + // It is all int32_t values except for fFormataVersion, which is an array of four bytes. + // Swap the whole thing as int32_t, then re-swap the one field. + // + ds->swapArray32(ds, inBytes, sizeof(RBBIDataHeader), outBytes, status); + ds->swapArray32(ds, outputDH->fFormatVersion, 4, outputDH->fFormatVersion, status); return totalSize; }