org.jpedal.fonts.tt.CMAP Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of OpenViewerFX Show documentation
Show all versions of OpenViewerFX Show documentation
Open Source (LGPL) JavaFX PDF Viewer for NetBeans plugin
The newest version!
/*
* ===========================================
* Java Pdf Extraction Decoding Access Library
* ===========================================
*
* Project Info: http://www.idrsolutions.com
* Help section for developers at http://www.idrsolutions.com/support/
*
* (C) Copyright 1997-2017 IDRsolutions and Contributors.
*
* This file is part of JPedal/JPDF2HTML5
*
@LICENSE@
*
* ---------------
* CMAP.java
* ---------------
*/
package org.jpedal.fonts.tt;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.Map;
import org.jpedal.fonts.StandardFonts;
import org.jpedal.utils.LogWriter;
public class CMAP extends Table {
protected int[][] glyphIndexToChar;
private boolean remapType4;
private boolean hasFormatZero;
private int[] glyphToIndex;
//flag 6 and use if not able to map elsewhere
private boolean hasSix;
//flag 4
private boolean hasFormat4;
private boolean hasFormat6;
private int lastFormat4Found = -1;
//used by format 6
private int firstCode = -1;
private int entryCount = -1;
//used by format 4
private int segCount;
/**
* which type of mapping to use
*/
private int fontMapping;
//used by format 4
protected int[] endCode;
protected int[] startCode;
protected int[] idDelta;
protected int[] idRangeOffset;
protected int[] glyphIdArray;
private int[] f6glyphIdArray;
private int[] offset;
//used by Format 12
int nGroups;
private int[] startCharCode;
private int[] endCharCode;
private int[] startGlyphCode;
/**
* CMap format used -1 shows not set
*/
protected int[] CMAPformats, CMAPlength, CMAPlang, CMAPsegCount, CMAPsearchRange, CMAPentrySelector, CMAPrangeShift, CMAPreserved;
/*Platform-specific ID list*/
// private static String[] PlatformSpecificID={"Roman","Japanese","Traditional Chinese","Korean",
// "Arabic","Hebrew","Greek","Russian",
// "RSymbol","Devanagari","Gurmukhi","Gujarati",
// "Oriya","Bengali","Tamil","Telugu",
// "Kannada","Malayalam","Sinhalese","Burmese",
// "Khmer","Thai","Laotian","Georgian",
// "Armenian","Simplified Chinese","Tibetan","Mongolian",
// "Geez","Slavic","Vietnamese","Sindhi","(Uninterpreted)"};
//
/*Platform-specific ID list*/
//private static String[] PlatformIDName={"Unicode","Macintosh","Reserved","Microsoft"};
/**
* shows which encoding used
*/
protected int[] platformID;
private static final Map exceptions;
/*set up differences from Mac Roman*/
static {
exceptions = new HashMap();
final String[] keys = {"notequal", "infinity", "lessequal", "greaterequal",
"partialdiff", "summation", "product", "pi",
"integral", "Omega", "radical", "approxequal",
"Delta", "lozenge", "Euro", "apple"};
final int[] values = {173, 176, 178, 179,
182, 183, 184, 185,
186, 189, 195, 197,
198, 215, 219, 240};
for (int i = 0; i < values.length; i++) {
exceptions.put(keys[i], values[i]);
}
StandardFonts.checkLoaded(StandardFonts.WIN);
}
/**
* which CMAP to use to decode the font
*/
private int formatToUse;
protected int id, numberSubtables;
protected int[] CMAPsubtables, platformSpecificID;
private int fontEncoding;
public CMAP(final FontFile2 currentFontFile, final int startPointer) {
final boolean debug = false;
if (debug) {
System.out.println("CMAP " + this);
}
//LogWriter.writeMethod("{readCMAPTable}", 0);
//read 'cmap' table
if (startPointer == 0) {
LogWriter.writeLog("No CMAP table found");
} else {
id = currentFontFile.getNextUint16(); //id
numberSubtables = currentFontFile.getNextUint16();
//read the subtables
CMAPsubtables = new int[numberSubtables];
platformID = new int[numberSubtables];
platformSpecificID = new int[numberSubtables];
CMAPformats = new int[numberSubtables];
CMAPsearchRange = new int[numberSubtables];
CMAPentrySelector = new int[numberSubtables];
CMAPrangeShift = new int[numberSubtables];
CMAPreserved = new int[numberSubtables];
CMAPsegCount = new int[numberSubtables];
CMAPlength = new int[numberSubtables];
CMAPlang = new int[numberSubtables];
glyphIndexToChar = new int[numberSubtables][256];
glyphToIndex = new int[256];
for (int i = 0; i < numberSubtables; i++) {
platformID[i] = currentFontFile.getNextUint16();
platformSpecificID[i] = currentFontFile.getNextUint16();
CMAPsubtables[i] = currentFontFile.getNextUint32();
if (debug) {
System.out.println("IDs platformID=" + platformID[i] + " platformSpecificID=" + platformSpecificID[i] + " CMAPsubtables=" + CMAPsubtables[i]);
}
//System.out.println(PlatformID[platformID[i]]+" "+PlatformSpecificID[platformSpecificID[i]]+CMAPsubtables[i]);
}
//now read each subtable
for (int j = 0; j < numberSubtables; j++) {
currentFontFile.selectTable(FontFile2.CMAP);
currentFontFile.skip(CMAPsubtables[j]);
//assume 16 bit format to start
CMAPformats[j] = currentFontFile.getNextUint16();
CMAPlength[j] = currentFontFile.getNextUint16();
CMAPlang[j] = currentFontFile.getNextUint16(); //lang
if (debug) {
System.out.println(j + " type=" + CMAPformats[j] + " length=" + CMAPlength[j] + " lang=" + CMAPlang[j]);
}
//flag if present
if (CMAPformats[j] == 6) {
hasSix = true;
}
if (CMAPformats[j] == 0 && CMAPlength[j] == 262) {
readFormatZeroTable(currentFontFile, j);
} else if (CMAPformats[j] == 4) {
readFormat4Table(j, currentFontFile);
} else if (CMAPformats[j] == 6) {
readFormat6Table(currentFontFile);
} else if (CMAPformats[j] == 12) {
readFormat12Table(currentFontFile);
} else {
//System.out.println("Unsupported Format "+CMAPformats[j]);
//reset to avoid setting
CMAPformats[j] = -1;
}
}
}
}
private void readFormat4Table(final int j, final FontFile2 currentFontFile) {
//read values
CMAPsegCount[j] = currentFontFile.getNextUint16();
segCount = CMAPsegCount[j] / 2;
CMAPsearchRange[j] = currentFontFile.getNextUint16(); //searchrange
CMAPentrySelector[j] = currentFontFile.getNextUint16(); //entrySelector
CMAPrangeShift[j] = currentFontFile.getNextUint16(); //rangeShift
//check current format 4 is greater than previous or vice versa and act accordingly
//because some font files have more than one format 4 subtables with different length
if (hasFormat4) {
if (CMAPlength[lastFormat4Found] > CMAPlength[j]) {
CMAPlength[j] = CMAPlength[lastFormat4Found];
CMAPsegCount[j] = CMAPsegCount[lastFormat4Found];
CMAPsearchRange[j] = CMAPsearchRange[lastFormat4Found]; //searchrange
CMAPentrySelector[j] = CMAPentrySelector[lastFormat4Found]; //entrySelector
CMAPrangeShift[j] = CMAPrangeShift[lastFormat4Found]; //rangeShift
return;
} else if (CMAPlength[lastFormat4Found] < CMAPlength[j]) {
CMAPlength[lastFormat4Found] = CMAPlength[j];
CMAPsegCount[lastFormat4Found] = CMAPsegCount[j];
CMAPsearchRange[lastFormat4Found] = CMAPsearchRange[j]; //searchrange
CMAPentrySelector[lastFormat4Found] = CMAPentrySelector[j]; //entrySelector
CMAPrangeShift[lastFormat4Found] = CMAPrangeShift[j]; //rangeShift
}
}
lastFormat4Found = j;
hasFormat4 = true;
//read tables and initialise size of arrays
endCode = new int[segCount];
for (int i = 0; i < segCount; i++) {
endCode[i] = currentFontFile.getNextUint16();
}
CMAPreserved[j] = currentFontFile.getNextUint16(); //reserved (should be zero)
startCode = new int[segCount];
for (int i = 0; i < segCount; i++) {
startCode[i] = currentFontFile.getNextUint16();
}
idDelta = new int[segCount];
for (int i = 0; i < segCount; i++) {
idDelta[i] = currentFontFile.getNextUint16();
}
idRangeOffset = new int[segCount];
for (int i = 0; i < segCount; i++) {
idRangeOffset[i] = currentFontFile.getNextUint16();
}
/*create offsets*/
offset = new int[segCount];
int diff, cumulative = 0;
for (int i = 0; i < segCount; i++) {
if (idDelta[i] == 0) { // && startCode[i]!=endCode[i]){
offset[i] = cumulative;
diff = 1 + endCode[i] - startCode[i];
//fixes bug in mapping theSansOffice tff font
if (startCode[i] == endCode[i] && idRangeOffset[i] == 0) {
diff = 0;
}
cumulative += diff;
}
}
// glyphIdArray at end
final int count = (CMAPlength[j] - 16 - (segCount * 8)) / 2;
glyphIdArray = new int[count];
for (int i = 0; i < count; i++) {
glyphIdArray[i] = currentFontFile.getNextUint16();
}
}
private void readFormat6Table(final FontFile2 currentFontFile) {
hasFormat6 = true;
firstCode = currentFontFile.getNextUint16();
entryCount = currentFontFile.getNextUint16();
f6glyphIdArray = new int[firstCode + entryCount];
for (int jj = 0; jj < entryCount; jj++) {
f6glyphIdArray[jj + firstCode] = currentFontFile.getNextUint16();
}
}
private void readFormat12Table(final FontFile2 currentFontFile) {
currentFontFile.getNextUint16(); //length //not what it says in spec but what I found in file
currentFontFile.getNextUint32(); //lang
nGroups = currentFontFile.getNextUint32();
startCharCode = new int[nGroups];
endCharCode = new int[nGroups];
startGlyphCode = new int[nGroups];
for (int ii = 0; ii < nGroups; ii++) {
startCharCode[ii] = currentFontFile.getNextUint32();
endCharCode[ii] = currentFontFile.getNextUint32();
startGlyphCode[ii] = currentFontFile.getNextUint32();
}
}
private void readFormatZeroTable(final FontFile2 currentFontFile, final int j) {
hasFormatZero = true;
for (int glyphNum = 0; glyphNum < 256; glyphNum++) {
final int index = currentFontFile.getNextUint8();
glyphIndexToChar[j][glyphNum] = index;
glyphToIndex[index] = glyphNum;
}
}
public CMAP() {
}
/**
* convert raw glyph number to Character code
*/
public int convertIndexToCharacterCode(final String glyph, int index) {
int index2 = -1;
final int rawIndex = index;
int format = CMAPformats[formatToUse];
final boolean debugMapping = false; //(index==223);
if (debugMapping) {
System.out.println(glyph + " fontMapping=" + fontMapping + " index=" + index + ' ' + remapType4);
}
/*
* convert index if needed
*/
if ((fontMapping == 1 || fontMapping == 2 || fontMapping == 3 || (fontMapping == 4 && remapType4))) { //) && (!"notdef".equals(glyph))) {
if (glyph != null && !"notdef".equals(glyph)) {
index2 = index; //StandardFonts.lookupCharacterIndex(glyph,StandardFonts.WIN);
index = StandardFonts.getAdobeMap(glyph);
} else if (exceptions.containsKey(glyph)) {
index = exceptions.get(glyph);
}
}
int value = -1;
//exception found in Itext
if (rawIndex == 128 && endCode != null && "Euro".equals(glyph)) {
value = getFormat4Value(8364, value);
} else if (format == 0) { //if no cmap use also identity
//hack
if (index > 255) {
index = 0;
}
value = glyphIndexToChar[formatToUse][index];
if (value == 0 && index2 != -1) {
value = glyphIndexToChar[formatToUse][index2];
}
//if no Format 0 check format 4 (see case 26831)
if (value == 0) {
final int format4Value = getFormat4Value(index, 0);
if (format4Value > 0) {
value = format4Value;
}
}
} else if (format == 4) {
value = getFormat4Value(index, value);
//hack for odd value in customer file
if (value == -1) {
if (index > 0xf000) {
value = getFormat4Value(index - 0xf000, value);
} else {
value = getFormat4Value(index + 0xf000, value);
}
}
//see 18113 fixes ligatures on page
if (value == -1) {
value = getFormat4Value(rawIndex + 0xf000, value);
}
} else if (format == 12) {
value = getFormat12Value(index, debugMapping, value);
}
//second attempt if no value found
if (value == -1 && hasSix) {
index = rawIndex;
format = 6;
}
if (format == 6) {
if (fontEncoding != 1) {
index = StandardFonts.lookupCharacterIndex(glyph, StandardFonts.MAC);
}
if (index >= f6glyphIdArray.length) {
value = 0;
} else {
value = f6glyphIdArray[index];
}
}
if (debugMapping) {
System.out.println("returns " + value + ' ' + this);
}
return value;
}
/**
* lookup tables similar to format 4
* see https://developer.apple.com/fonts/TTRefMan/RM06/Chap6cmap.html
*/
private int getFormat12Value(final int index, final boolean debugMapping, int value) {
/*
* cycle through tables and then add offset to Glyph start
*/
for (int i = 0; i < nGroups; i++) {
if (debugMapping) {
System.out.println("table=" + i + " start=" + startCharCode[i] + ' ' + index +
" end=" + endCharCode[i] + " glypgStartCode[i]=" + startGlyphCode[i]);
}
if (endCharCode[i] >= index && startCharCode[i] <= index) {
value = startGlyphCode[i] + index - startCharCode[i];
i = nGroups; //exit loop
}
}
return value;
}
private int getFormat4Value(final int index, int value) {
final boolean debugMapping = false;
for (int i = 0; i < segCount; i++) {
if (debugMapping) {
System.out.println("Segtable=" + i + " start=" + startCode[i] + ' ' + index +
" end=" + endCode[i] + " idRangeOffset[i]=" + idRangeOffset[i] +
" offset[i]=" + offset[i] + " idRangeOffset[i]=" + idRangeOffset[i] + " idDelta[i]=" + idDelta[i]);
}
if (endCode[i] >= index && startCode[i] <= index) {
final int idx;
if (idRangeOffset[i] == 0) {
if (debugMapping) {
System.out.println("xxx=" + (idDelta[i] + index));
}
value = (idDelta[i] + index) % 65536;
i = segCount;
} else {
idx = offset[i] + (index - startCode[i]);
if (idx < glyphIdArray.length) {
value = glyphIdArray[idx];
}
if (debugMapping) {
System.out.println("value=" + value + " idx=" +
idx + " glyphIdArrays=" + glyphIdArray[0] + ' ' +
glyphIdArray[1] + ' ' + glyphIdArray[2] + " offset[i]=" + offset[i] +
" index=" + index + " startCode[" + i + "]=" + startCode[i] + " i=" + i);
}
i = segCount;
}
}
}
return value;
}
/**
* work out correct CMAP table to use.
*/
public void setEncodingToUse(final boolean hasEncoding, final int fontEncoding, final boolean isCID) {
final boolean encodingDebug = false;
this.fontEncoding = fontEncoding;
if (encodingDebug) {
System.out.println(this + "hasEncoding=" + hasEncoding + " fontEncoding=" + fontEncoding + " isCID=" + isCID);
}
formatToUse = -1;
final int count = platformID.length;
/*case 1 */
for (int i = 0; i < count; i++) {
if ((platformID[i] == 3) && (CMAPformats[i] == 1 || CMAPformats[i] == 0)) {
formatToUse = i;
this.fontMapping = 1;
i = count;
if (encodingDebug) {
System.out.println("case1");
}
}
}
/*case 2*/
boolean wasCase2 = false;
if (formatToUse == -1 && hasFormatZero && !isCID) {
for (int i = 0; i < count; i++) {
if (platformID[i] == 1 && CMAPformats[i] == 0) {
formatToUse = i;
wasCase2 = !(glyphIndexToChar[formatToUse][223] != 0 && getFormat4Value(223, 0) == 0);
if (hasEncoding) {
fontMapping = 2;
StandardFonts.checkLoaded(StandardFonts.MAC);
} else if (!wasCase2 && platformSpecificID[formatToUse] == 0 && platformID[formatToUse] == 1) {
fontMapping = 3;
StandardFonts.checkLoaded(StandardFonts.WIN);
} else {
fontMapping = -1;
}
i = count;
if (encodingDebug) {
System.out.println("case2 fontMapping=" + fontMapping + " formatToUse=" + formatToUse + ' ' + platformSpecificID[formatToUse] + ' ' + platformID[formatToUse] + ' ' + hasEncoding + ' ' + fontEncoding + ' ' + wasCase2);
}
}
}
}
/*case 4 - no simple maps or prefer to last 1*/
/*last check uses fl glyph and sticks to case 1 if found*/
if (formatToUse == -1 || fontMapping == 3 || wasCase2) {
//if((formatToUse==-1)){
for (int i = 0; i < count; i++) {
if ((CMAPformats[i] == 4)) {
formatToUse = i;
fontMapping = 4;
i = count;
if (encodingDebug) {
System.out.println("case4 fontMapping=" + fontMapping + " formatToUse=" + formatToUse + ' ' + platformSpecificID[formatToUse] + ' ' + platformID[formatToUse] + ' ' + hasEncoding + ' ' + fontEncoding);
}
if (platformSpecificID[formatToUse] == 3 && platformID[formatToUse] == 0 && count > 1 && (hasEncoding || fontEncoding == StandardFonts.STD)) {
remapType4 = true;
// System.out.println("a "+wasCase2);
} else if ((platformSpecificID[formatToUse] == 1 || platformSpecificID[formatToUse] == 3) &&
platformID[formatToUse] == 0 && hasEncoding && fontEncoding == StandardFonts.WIN) {
remapType4 = true;
// System.out.println("b");
} else if (platformSpecificID[formatToUse] == 0 && platformID[formatToUse] == 0 && hasEncoding && fontEncoding == StandardFonts.WIN) {
remapType4 = true;
// System.out.println("c");
} else if (platformSpecificID[formatToUse] == 1 && platformID[formatToUse] == 3 && hasEncoding && fontEncoding == StandardFonts.WIN &&
(wasCase2 || getFormat4Value(223, 0) == 0)) {
remapType4 = true;
// System.out.println("d "+wasCase2+ " "+isCase2+" "+(glyphIndexToChar[formatToUse][223]+" "+getFormat4Value(223, 0)));
} else if (platformSpecificID[formatToUse] == 1 && platformID[formatToUse] == 3 && hasEncoding && (fontEncoding == StandardFonts.MAC || (count == 1 && fontEncoding == StandardFonts.WIN))) {
remapType4 = true;
// System.out.println("e");
} else if (!hasEncoding && fontEncoding != 1) {
remapType4 = true;
// System.out.println("g");
}
}
}
}
/*case 3 - no MAC cmap in other ranges and substituting font */
if (formatToUse == -1) {
for (int i = 0; i < count; i++) {
if ((CMAPformats[i] == 6)) {
formatToUse = i;
if (!hasEncoding) {
fontMapping = 2;
StandardFonts.checkLoaded(StandardFonts.MAC);
} else {
fontMapping = 6;
}
i = count;
if (encodingDebug) {
System.out.println("case3 fontMapping=" + fontMapping + " formatToUse=" + formatToUse + ' ' + platformID[formatToUse]);
}
}
}
}
/*case 5 - type12*/
if (formatToUse == -1) {
for (int i = 0; i < count; i++) {
if ((CMAPformats[i] == 12)) {
formatToUse = i;
if (!hasEncoding) {
fontMapping = 2;
StandardFonts.checkLoaded(StandardFonts.MAC);
} else {
fontMapping = 12;
}
i = count;
if (encodingDebug) {
System.out.println("case5");
}
}
}
}
if (fontEncoding == StandardFonts.ZAPF) {
fontMapping = 2;
StandardFonts.checkLoaded(StandardFonts.MAC);
if (encodingDebug) {
System.out.println("Zapf");
}
}
}
/**
* turn type 0 table into a list of glyph
*/
public Map buildCharStringTable() {
final Map glyfValues = new HashMap();
// for(int i : glyphToIndex){
// if(i>0){
// glyfValues.put(glyphToIndex[i],i);
// //System.out.println("i=" + i + " " + StandardFonts.getUnicodeChar(encodingToUse, i));
// }
// }
if (hasFormat4) {
final ArrayList list4 = new ArrayList();
for (int z = 0; z < segCount; z++) {
final int total = endCode[z] - startCode[z] + 1;
for (int q = 0; q < total; q++) {
list4.add(startCode[z] + q);
}
}
for (final Integer i : list4) {
glyfValues.put(i, getFormat4Value(i, 0));
}
} else if (hasFormat6) {
for (int z = 0; z < entryCount; z++) {
//System.out.println(firstCode+z+" ==> "+f6glyphIdArray[firstCode+z]);
glyfValues.put(firstCode + z, f6glyphIdArray[firstCode + z]);
}
} else {
for (int z = 0; z < glyphToIndex.length; z++) {
if (glyphToIndex[z] > 0) {
glyfValues.put(glyphToIndex[z], z);
}
}
}
return glyfValues;
}
public int getGlyphToIndex(final int rawInt) {
return glyphToIndex[rawInt];
}
}