Attachment #23181 for bug #46610

View | Details | Raw Unified | Return to bug 46610
Collapse All | Expand All




	 * @param fcStart The start of the text for this property, in _bytes_
	 * @param fcEnd The end of the text for this property, in _bytes_
	 */
	public BytePropertyNode(int fcStart, int fcEnd, CharIndexTranslator translator, Object buf, boolean isUnicode) {
		super(
				translator.getCharIndex(fcStart),
				translator.getCharIndex(fcEnd),
				buf
		);
		this.isUnicode = isUnicode;

	}
	
	public boolean isUnicode() {
		//XXX Is this method useful?
		return isUnicode;
	}
	public int getStartBytes() {
		//XXX That is wrong in most cases!
		if(isUnicode)
			return getStart()*2;
		return getStart();
	}
	public int getEndBytes() {
		//XXX That is wrong in most cases!
		if(isUnicode)
			return getEnd()*2;
		return getEnd();




package org.apache.poi.hwpf.model;

public interface CharIndexTranslator {

	/**
	 * Calculates the char index of the given byte index.
	 * 
	 * @param byteStart
	 * @param bytePos
	 * @return
	 */
	int getCharIndex(int bytePos);

}

Lines 121-127 Link Here

(-)src/scratchpad/src/org/apache/poi/hwpf/model/CHPBinTable.java (-2 / +2 lines)
121	{	121	{
122	boolean needsToBeUnicode = tpt.isUnicodeAtCharOffset(cpStart);	122	boolean needsToBeUnicode = tpt.isUnicodeAtCharOffset(cpStart);
123		123
124	CHPX insertChpx = new CHPX(0, 0, buf, needsToBeUnicode);	124	CHPX insertChpx = new CHPX(0, 0, tpt,buf, needsToBeUnicode);
125		125
126	// Ensure character offsets are really characters	126	// Ensure character offsets are really characters
127	insertChpx.setStart(cpStart);	127	insertChpx.setStart(cpStart);
Lines 141-147 Link Here
141	// Original, until insert at point	141	// Original, until insert at point
142	// New one	142	// New one
143	// Clone of original, on to the old end	143	// Clone of original, on to the old end
144	CHPX clone = new CHPX(0, 0, chpx.getSprmBuf(), needsToBeUnicode);	144	CHPX clone = new CHPX(0, 0, tpt,chpx.getSprmBuf(), needsToBeUnicode);
145	// Again ensure contains character based offsets no matter what	145	// Again ensure contains character based offsets no matter what
146	clone.setStart(cpStart);	146	clone.setStart(cpStart);
147	clone.setEnd(chpx.getEnd());	147	clone.setEnd(chpx.getEnd());

Lines 61-68 Link Here

(-)src/scratchpad/src/org/apache/poi/hwpf/model/CHPFormattedDiskPage.java (-2 / +4 lines)
61		61
62	for (int x = 0; x < _crun; x++)	62	for (int x = 0; x < _crun; x++)
63	{	63	{
64	boolean isUnicode = tpt.isUnicodeAtByteOffset( getStart(x) );	64	int startAt = getStart(x) - fcMin;
65	_chpxList.add(new CHPX(getStart(x) - fcMin, getEnd(x) - fcMin, getGrpprl(x), isUnicode));	65	boolean isUnicode = tpt.isUnicodeAtByteOffset( startAt );
		66	int endAt = getEnd(x) - fcMin;
		67	_chpxList.add(new CHPX(startAt, endAt, tpt, getGrpprl(x), isUnicode));
66	}	68	}
67	}	69	}
68		70




public class CHPX extends BytePropertyNode
{

  public CHPX(int fcStart, int fcEnd, CharIndexTranslator translator, byte[] grpprl, boolean isUnicode)
  {
    super(fcStart, fcEnd, translator, new SprmBuffer(grpprl), isUnicode);
  }

  public CHPX(int fcStart, int fcEnd, CharIndexTranslator translator, SprmBuffer buf, boolean isUnicode)
  {
    super(fcStart, fcEnd, translator ,buf, isUnicode);
  }



Lines 78-84 Link Here

(-)src/scratchpad/src/org/apache/poi/hwpf/model/PAPBinTable.java (-2 / +2 lines)
78	{	78	{
79	boolean needsToBeUnicode = tpt.isUnicodeAtCharOffset(cpStart);	79	boolean needsToBeUnicode = tpt.isUnicodeAtCharOffset(cpStart);
80		80
81	PAPX forInsert = new PAPX(0, 0, buf, _dataStream, needsToBeUnicode);	81	PAPX forInsert = new PAPX(0, 0, tpt, buf, _dataStream, needsToBeUnicode);
82		82
83	// Ensure character offsets are really characters	83	// Ensure character offsets are really characters
84	forInsert.setStart(cpStart);	84	forInsert.setStart(cpStart);
Lines 108-114 Link Here
108	// Original, until insert at point	108	// Original, until insert at point
109	// New one	109	// New one
110	// Clone of original, on to the old end	110	// Clone of original, on to the old end
111	PAPX clone = new PAPX(0, 0, clonedBuf, _dataStream, needsToBeUnicode);	111	PAPX clone = new PAPX(0, 0, tpt, clonedBuf, _dataStream, needsToBeUnicode);
112	// Again ensure contains character based offsets no matter what	112	// Again ensure contains character based offsets no matter what
113	clone.setStart(cpStart);	113	clone.setStart(cpStart);
114	clone.setEnd(currentPap.getEnd());	114	clone.setEnd(currentPap.getEnd());

Lines 70-76 Link Here

(-)src/scratchpad/src/org/apache/poi/hwpf/model/PAPFormattedDiskPage.java (-1 / +1 lines)
70	boolean isUnicode = tpt.isUnicodeAtByteOffset(startAt);	70	boolean isUnicode = tpt.isUnicodeAtByteOffset(startAt);
71	//System.err.println(startAt + " -> " + endAt + " = " + isUnicode);	71	//System.err.println(startAt + " -> " + endAt + " = " + isUnicode);
72		72
73	_papxList.add(new PAPX(startAt, endAt, getGrpprl(x), getParagraphHeight(x), dataStream, isUnicode));	73	_papxList.add(new PAPX(startAt, endAt, tpt, getGrpprl(x), getParagraphHeight(x), dataStream, isUnicode));
74	}	74	}
75	_fkp = null;	75	_fkp = null;
76	_dataStream = dataStream;	76	_dataStream = dataStream;




  private ParagraphHeight _phe;
  private int _hugeGrpprlOffset = -1;

  public PAPX(int fcStart, int fcEnd, CharIndexTranslator translator, byte[] papx, ParagraphHeight phe, byte[] dataStream, boolean isUnicode)
  {
    super(fcStart, fcEnd, translator, new SprmBuffer(papx), isUnicode);
    _phe = phe;
    SprmBuffer buf = findHuge(new SprmBuffer(papx), dataStream);
    if(buf != null)
      _buf = buf;
  }

  public PAPX(int fcStart, int fcEnd, CharIndexTranslator translator, SprmBuffer buf, byte[] dataStream, boolean isUnicode)
  {
    super(fcStart, fcEnd, translator, buf, isUnicode);
    _phe = new ParagraphHeight();
    buf = findHuge(buf, dataStream);
    if(buf != null)

Lines 67-73 Link Here

(-)src/scratchpad/src/org/apache/poi/hwpf/model/SectionTable.java (-2 / +2 lines)
67	// check for the optimization	67	// check for the optimization
68	if (fileOffset == 0xffffffff)	68	if (fileOffset == 0xffffffff)
69	{	69	{
70	_sections.add(new SEPX(sed, startAt, endAt, new byte[0], isUnicodeAtStart));	70	_sections.add(new SEPX(sed, startAt, endAt, tpt, new byte[0], isUnicodeAtStart));
71	}	71	}
72	else	72	else
73	{	73	{
Lines 76-82 Link Here
76	byte[] buf = new byte[sepxSize];	76	byte[] buf = new byte[sepxSize];
77	fileOffset += LittleEndian.SHORT_SIZE;	77	fileOffset += LittleEndian.SHORT_SIZE;
78	System.arraycopy(documentStream, fileOffset, buf, 0, buf.length);	78	System.arraycopy(documentStream, fileOffset, buf, 0, buf.length);
79	_sections.add(new SEPX(sed, startAt, endAt, buf, isUnicodeAtStart));	79	_sections.add(new SEPX(sed, startAt, endAt, tpt, buf, isUnicodeAtStart));
80	}	80	}
81	}	81	}
82		82

Lines 31-39 Link Here

(-)src/scratchpad/src/org/apache/poi/hwpf/model/SEPX.java (-2 / +2 lines)
31		31
32	SectionDescriptor _sed;	32	SectionDescriptor _sed;
33		33
34	public SEPX(SectionDescriptor sed, int start, int end, byte[] grpprl, boolean isUnicode)	34	public SEPX(SectionDescriptor sed, int start, int end, CharIndexTranslator translator, byte[] grpprl, boolean isUnicode)
35	{	35	{
36	super(start, end, SectionSprmUncompressor.uncompressSEP(grpprl, 0), isUnicode);	36	super(start, end, translator, SectionSprmUncompressor.uncompressSEP(grpprl, 0), isUnicode);
37	_sed = sed;	37	_sed = sed;
38	}	38	}
39		39




 *  convertion.
 * @author Ryan Ackley
 */
public class TextPieceTable implements CharIndexTranslator 
{
  protected ArrayList _textPieces = new ArrayList();
  //int _multiple;

    }
    return false;
  }
  	/* (non-Javadoc)
	 * @see org.apache.poi.hwpf.model.CharIndexTranslator#getLengthInChars(int)
	 */
	public int getCharIndex(int bytePos) {
		int charCount = 0;
		int curByte = 0;

		Iterator it = _textPieces.iterator();
		while (it.hasNext() && curByte < bytePos) {
			TextPiece tp = (TextPiece) it.next();

			int bytesLength = tp.bytesLength();
			int nextByte = curByte + bytesLength;

			int toAdd = bytePos > nextByte ? bytesLength : bytesLength
					- (nextByte - bytePos);

			if (tp.isUnicode()) {
				charCount += toAdd / 2;
			} else {
				charCount += toAdd;
			}

			curByte = nextByte;
		}

		return charCount;
	}
}

Return to bug 46610

Lines 30-39 Link Here

(-)src/scratchpad/src/org/apache/poi/hwpf/model/BytePropertyNode.java (-3 / +6 lines)
30	* @param fcStart The start of the text for this property, in _bytes_	30	* @param fcStart The start of the text for this property, in _bytes_
31	* @param fcEnd The end of the text for this property, in _bytes_	31	* @param fcEnd The end of the text for this property, in _bytes_
32	*/	32	*/
33	public BytePropertyNode(int fcStart, int fcEnd, Object buf, boolean isUnicode) {	33	public BytePropertyNode(int fcStart, int fcEnd, CharIndexTranslator translator, Object buf, boolean isUnicode) {
34	super(	34	super(
35	generateCp(fcStart, isUnicode),	35	translator.getCharIndex(fcStart),
36	generateCp(fcEnd, isUnicode),	36	translator.getCharIndex(fcEnd),
37	buf	37	buf
38	);	38	);
39	this.isUnicode = isUnicode;	39	this.isUnicode = isUnicode;
Lines 45-58 Link Here
45	}	45	}
46		46
47	public boolean isUnicode() {	47	public boolean isUnicode() {
		48	//XXX Is this method useful?
48	return isUnicode;	49	return isUnicode;
49	}	50	}
50	public int getStartBytes() {	51	public int getStartBytes() {
		52	//XXX That is wrong in most cases!
51	if(isUnicode)	53	if(isUnicode)
52	return getStart()*2;	54	return getStart()*2;
53	return getStart();	55	return getStart();
54	}	56	}
55	public int getEndBytes() {	57	public int getEndBytes() {
		58	//XXX That is wrong in most cases!
56	if(isUnicode)	59	if(isUnicode)
57	return getEnd()*2;	60	return getEnd()*2;
58	return getEnd();	61	return getEnd();

Line 0 Link Here

(-)src/scratchpad/src/org/apache/poi/hwpf/model/CharIndexTranslator.java (+14 lines)
	1	package org.apache.poi.hwpf.model;
	2
	3	public interface CharIndexTranslator {
	4
	5	/**
	6	* Calculates the char index of the given byte index.
	7	*
	8	* @param byteStart
	9	* @param bytePos
	10	* @return
	11	*/
	12	int getCharIndex(int bytePos);
	13
	14	}

Lines 37-50 Link Here

(-)src/scratchpad/src/org/apache/poi/hwpf/model/CHPX.java (-4 / +4 lines)
37	public class CHPX extends BytePropertyNode	37	public class CHPX extends BytePropertyNode
38	{	38	{
39		39
40	public CHPX(int fcStart, int fcEnd, byte[] grpprl, boolean isUnicode)	40	public CHPX(int fcStart, int fcEnd, CharIndexTranslator translator, byte[] grpprl, boolean isUnicode)
41	{	41	{
42	super(fcStart, fcEnd, new SprmBuffer(grpprl), isUnicode);	42	super(fcStart, fcEnd, translator, new SprmBuffer(grpprl), isUnicode);
43	}	43	}
44		44
45	public CHPX(int fcStart, int fcEnd, SprmBuffer buf, boolean isUnicode)	45	public CHPX(int fcStart, int fcEnd, CharIndexTranslator translator, SprmBuffer buf, boolean isUnicode)
46	{	46	{
47	super(fcStart, fcEnd, buf, isUnicode);	47	super(fcStart, fcEnd, translator ,buf, isUnicode);
48	}	48	}
49		49
50		50

Lines 43-60 Link Here

(-)src/scratchpad/src/org/apache/poi/hwpf/model/PAPX.java (-4 / +4 lines)
43	private ParagraphHeight _phe;	43	private ParagraphHeight _phe;
44	private int _hugeGrpprlOffset = -1;	44	private int _hugeGrpprlOffset = -1;
45		45
46	public PAPX(int fcStart, int fcEnd, byte[] papx, ParagraphHeight phe, byte[] dataStream, boolean isUnicode)	46	public PAPX(int fcStart, int fcEnd, CharIndexTranslator translator, byte[] papx, ParagraphHeight phe, byte[] dataStream, boolean isUnicode)
47	{	47	{
48	super(fcStart, fcEnd, new SprmBuffer(papx), isUnicode);	48	super(fcStart, fcEnd, translator, new SprmBuffer(papx), isUnicode);
49	_phe = phe;	49	_phe = phe;
50	SprmBuffer buf = findHuge(new SprmBuffer(papx), dataStream);	50	SprmBuffer buf = findHuge(new SprmBuffer(papx), dataStream);
51	if(buf != null)	51	if(buf != null)
52	_buf = buf;	52	_buf = buf;
53	}	53	}
54		54
55	public PAPX(int fcStart, int fcEnd, SprmBuffer buf, byte[] dataStream, boolean isUnicode)	55	public PAPX(int fcStart, int fcEnd, CharIndexTranslator translator, SprmBuffer buf, byte[] dataStream, boolean isUnicode)
56	{	56	{
57	super(fcStart, fcEnd, buf, isUnicode);	57	super(fcStart, fcEnd, translator, buf, isUnicode);
58	_phe = new ParagraphHeight();	58	_phe = new ParagraphHeight();
59	buf = findHuge(buf, dataStream);	59	buf = findHuge(buf, dataStream);
60	if(buf != null)	60	if(buf != null)

Lines 38-44 Link Here

(-)src/scratchpad/src/org/apache/poi/hwpf/model/TextPieceTable.java (-1 / +29 lines)
38	* convertion.	38	* convertion.
39	* @author Ryan Ackley	39	* @author Ryan Ackley
40	*/	40	*/
41	public class TextPieceTable	41	public class TextPieceTable implements CharIndexTranslator
42	{	42	{
43	protected ArrayList _textPieces = new ArrayList();	43	protected ArrayList _textPieces = new ArrayList();
44	//int _multiple;	44	//int _multiple;
Lines 269-272 Link Here
269	}	269	}
270	return false;	270	return false;
271	}	271	}
		272	/* (non-Javadoc)
		273	* @see org.apache.poi.hwpf.model.CharIndexTranslator#getLengthInChars(int)
		274	*/
		275	public int getCharIndex(int bytePos) {
		276	int charCount = 0;
		277	int curByte = 0;
		278
		279	Iterator it = _textPieces.iterator();
		280	while (it.hasNext() && curByte < bytePos) {
		281	TextPiece tp = (TextPiece) it.next();
		282
		283	int bytesLength = tp.bytesLength();
		284	int nextByte = curByte + bytesLength;
		285
		286	int toAdd = bytePos > nextByte ? bytesLength : bytesLength
		287	- (nextByte - bytePos);
		288
		289	if (tp.isUnicode()) {
		290	charCount += toAdd / 2;
		291	} else {
		292	charCount += toAdd;
		293	}
		294
		295	curByte = nextByte;
		296	}
		297
		298	return charCount;
		299	}
272	}	300	}