/************************************************************************** * * © 2016 and later: Unicode, Inc. and others. * License & terms of use: http://www.unicode.org/copyright.html#License * *************************************************************************** *************************************************************************** * * Copyright (C) 1998-2013, International Business Machines * Corporation and others. All Rights Reserved. * ************************************************************************/ #include "layout/LETypes.h" #include "layout/LESwaps.h" #include "sfnt.h" #include "cmaps.h" #include <stdio.h> #define SWAPU16(code) ((LEUnicode16) SWAPW(code)) #define SWAPU32(code) ((LEUnicode32) SWAPL(code)) // // Finds the high bit by binary searching // through the bits in value. // le_int8 highBit(le_uint32 value) { le_uint8 bit = 0; if (value >= 1 << 16) { value >>= 16; bit += 16; } if (value >= 1 << 8) { value >>= 8; bit += 8; } if (value >= 1 << 4) { value >>= 4; bit += 4; } if (value >= 1 << 2) { value >>= 2; bit += 2; } if (value >= 1 << 1) { value >>= 1; bit += 1; } return bit; } CMAPMapper *CMAPMapper::createUnicodeMapper(const CMAPTable *cmap) { le_uint16 i; le_uint16 nSubtables = SWAPW(cmap->numberSubtables); const CMAPEncodingSubtable *subtable = NULL; le_bool found = FALSE; le_uint16 foundPlatformID = 0xFFFF; le_uint16 foundPlatformSpecificID = 0xFFFF; le_uint32 foundOffset = 0; le_uint16 foundTable = 0xFFFF; // first pass, look for MS table. (preferred?) for (i = 0; i < nSubtables && !found; i += 1) { const CMAPEncodingSubtableHeader *esh = &cmap->encodingSubtableHeaders[i]; le_uint16 platformID = SWAPW(esh->platformID); le_uint16 platformSpecificID = SWAPW(esh->platformSpecificID); if (platformID == 3) { // microsoft switch (platformSpecificID) { case 1: // Unicode BMP (UCS-2) case 10: // Unicode UCS-4 foundOffset = SWAPL(esh->encodingOffset); foundPlatformID = platformID; foundPlatformSpecificID = platformSpecificID; found = TRUE; foundTable = i; break; //default: // printf("%s:%d: microsoft (3) platform specific ID %d (wanted 1 or 10) for subtable %d/%d\n", __FILE__, __LINE__, (SWAPW(esh->platformSpecificID)), i, nSubtables); } } else { //printf("%s:%d: platform ID %d (wanted 3, microsoft) for subtable %d/%d\n", __FILE__, __LINE__, (SWAPW(esh->platformID)), i, nSubtables); } } // second pass, allow non MS table // first pass, look for MS table. (preferred?) for (i = 0; i < nSubtables && !found; i += 1) { const CMAPEncodingSubtableHeader *esh = &cmap->encodingSubtableHeaders[i]; le_uint16 platformID = SWAPW(esh->platformID); le_uint16 platformSpecificID = SWAPW(esh->platformSpecificID); //printf("%s:%d: table %d/%d has platform:specific %d:%d\n", __FILE__, __LINE__, i, nSubtables, platformID, platformSpecificID); switch(platformID) { case 0: // Unicode platform switch(platformSpecificID) { case 0: case 1: case 2: case 3: foundOffset = SWAPL(esh->encodingOffset); foundPlatformID = platformID; foundPlatformSpecificID = platformSpecificID; foundTable = i; found = TRUE; break; default: printf("Error: table %d (psid %d) is unknown. Skipping.\n", i, platformSpecificID); break; } break; //default: //printf("Skipping platform id %d\n", platformID); } } if (found) { subtable = (const CMAPEncodingSubtable *) ((const char *) cmap + foundOffset); //printf("%s:%d: using subtable #%d/%d type %d:%d\n", __FILE__, __LINE__, foundTable, nSubtables, foundPlatformID, foundPlatformSpecificID); } else { printf("%s:%d: could not find subtable.\n", __FILE__, __LINE__); return NULL; } le_uint16 tableFormat = SWAPW(subtable->format); //printf("%s:%d: table format %d\n", __FILE__, __LINE__, tableFormat); switch (tableFormat) { case 4: return new CMAPFormat4Mapper(cmap, (const CMAPFormat4Encoding *) subtable); case 12: { const CMAPFormat12Encoding *encoding = (const CMAPFormat12Encoding *) subtable; return new CMAPGroupMapper(cmap, encoding->groups, SWAPL(encoding->nGroups)); } default: break; } printf("%s:%d: Unknown format %x.\n", __FILE__, __LINE__, (SWAPW(subtable->format))); return NULL; } CMAPFormat4Mapper::CMAPFormat4Mapper(const CMAPTable *cmap, const CMAPFormat4Encoding *header) : CMAPMapper(cmap) { le_uint16 segCount = SWAPW(header->segCountX2) / 2; fEntrySelector = SWAPW(header->entrySelector); fRangeShift = SWAPW(header->rangeShift) / 2; fEndCodes = &header->endCodes[0]; fStartCodes = &header->endCodes[segCount + 1]; // + 1 for reservedPad... fIdDelta = &fStartCodes[segCount]; fIdRangeOffset = &fIdDelta[segCount]; } LEGlyphID CMAPFormat4Mapper::unicodeToGlyph(LEUnicode32 unicode32) const { if (unicode32 >= 0x10000) { return 0; } LEUnicode16 unicode = (LEUnicode16) unicode32; le_uint16 index = 0; le_uint16 probe = 1 << fEntrySelector; TTGlyphID result = 0; if (SWAPU16(fStartCodes[fRangeShift]) <= unicode) { index = fRangeShift; } while (probe > (1 << 0)) { probe >>= 1; if (SWAPU16(fStartCodes[index + probe]) <= unicode) { index += probe; } } if (unicode >= SWAPU16(fStartCodes[index]) && unicode <= SWAPU16(fEndCodes[index])) { if (fIdRangeOffset[index] == 0) { result = (TTGlyphID) unicode; } else { le_uint16 offset = unicode - SWAPU16(fStartCodes[index]); le_uint16 rangeOffset = SWAPW(fIdRangeOffset[index]); le_uint16 *glyphIndexTable = (le_uint16 *) ((char *) &fIdRangeOffset[index] + rangeOffset); result = SWAPW(glyphIndexTable[offset]); } result += SWAPW(fIdDelta[index]); } else { result = 0; } return LE_SET_GLYPH(0, result); } CMAPFormat4Mapper::~CMAPFormat4Mapper() { // parent destructor does it all } CMAPGroupMapper::CMAPGroupMapper(const CMAPTable *cmap, const CMAPGroup *groups, le_uint32 nGroups) : CMAPMapper(cmap), fGroups(groups) { le_uint8 bit = highBit(nGroups); fPower = 1 << bit; fRangeOffset = nGroups - fPower; } LEGlyphID CMAPGroupMapper::unicodeToGlyph(LEUnicode32 unicode32) const { le_int32 probe = fPower; le_int32 range = 0; if (SWAPU32(fGroups[fRangeOffset].startCharCode) <= unicode32) { range = fRangeOffset; } while (probe > (1 << 0)) { probe >>= 1; if (SWAPU32(fGroups[range + probe].startCharCode) <= unicode32) { range += probe; } } if (SWAPU32(fGroups[range].startCharCode) <= unicode32 && SWAPU32(fGroups[range].endCharCode) >= unicode32) { return (LEGlyphID) (SWAPU32(fGroups[range].startGlyphCode) + unicode32 - SWAPU32(fGroups[range].startCharCode)); } return 0; } CMAPGroupMapper::~CMAPGroupMapper() { // parent destructor does it all }