Package org.apache.poi.hwpf.model

Examples of org.apache.poi.hwpf.model.TextPieceTable

104
105
106
107
108
109
110
111
112
113
114
115
        int fcMin = LittleEndian.getInt(header, 0x18);
        CHPBinTable cbt = new CHPBinTable(header, tableStream, chpOffset, chpSize, fcMin);

        // load our text pieces and our character runs
        ComplexFileTable cft = new ComplexFileTable(header, tableStream, complexOffset, fcMin);
        TextPieceTable tpt = cft.getTextPieceTable();
        List textPieces = tpt.getTextPieces();

        // make the POIFS objects available for garbage collection
        din = null;
        fsys = null;
        table = null;
View Full Code Here
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
    //fcMin = _fib.getFcMin()

    // Start to load up our standard structures.
    _dop = new DocumentProperties(_tableStream, _fib.getFcDop(), _fib.getLcbDop() );
    _cft = new ComplexFileTable(_mainStream, _tableStream, _fib.getFcClx(), fcMin);
    TextPieceTable _tpt = _cft.getTextPieceTable();

    // Now load the rest of the properties, which need to be adjusted
    //  for where text really begin
    _cbt = new CHPBinTable(_mainStream, _tableStream, _fib.getFcPlcfbteChpx(), _fib.getLcbPlcfbteChpx(), _tpt);
    _pbt = new PAPBinTable(_mainStream, _tableStream, _dataStream, _fib.getFcPlcfbtePapx(), _fib.getLcbPlcfbtePapx(), _tpt);

        _text = _tpt.getText();

        /*
         * in this mode we preserving PAPX/CHPX structure from file, so text may
         * miss from output, and text order may be corrupted
         */
        boolean preserveBinTables = false;
        try
        {
            preserveBinTables = Boolean.parseBoolean( System
                    .getProperty( PROPERTY_PRESERVE_BIN_TABLES ) );
        }
        catch ( Exception exc )
        {
            // ignore;
        }

        if ( !preserveBinTables )
        {
            _cbt.rebuild( _cft );
            _pbt.rebuild( _text, _cft );
        }

        /*
         * Property to disable text rebuilding. In this mode changing the text
         * will lead to unpredictable behavior
         */
        boolean preserveTextTable = false;
        try
        {
            preserveTextTable = Boolean.parseBoolean( System
                    .getProperty( PROPERTY_PRESERVE_TEXT_TABLE ) );
        }
        catch ( Exception exc )
        {
            // ignore;
        }
        if ( !preserveTextTable )
        {
            _cft = new ComplexFileTable();
            _tpt = _cft.getTextPieceTable();
            final TextPiece textPiece = new SinglentonTextPiece( _text );
            _tpt.add( textPiece );
            _text = textPiece.getStringBuilder();
        }

        // Read FSPA and Escher information
        // _fspa = new FSPATable(_tableStream, _fib.getFcPlcspaMom(),
View Full Code Here
121
122
123
124
125
126
127
128
129
130
131
132
    int fcMin = LittleEndian.getInt(header, 0x18);
    CHPBinTable cbt = new CHPBinTable(header, tableStream, chpOffset, chpSize, fcMin);

    // load our text pieces and our character runs
    ComplexFileTable cft = new ComplexFileTable(header, tableStream, complexOffset, fcMin);
    TextPieceTable tpt = cft.getTextPieceTable();
    List textPieces = tpt.getTextPieces();

    // make the POIFS objects available for garbage collection
    din = null;
    fsys = null;
    table = null;
View Full Code Here
120
121
122
123
124
125
126
127
128
129
130
131
132
133
    int chpSize = LittleEndian.getInt(header, 0xfe);
    int fcMin = LittleEndian.getInt(header, 0x18);

    // load our text pieces and our character runs
    ComplexFileTable cft = new ComplexFileTable(header, tableStream, complexOffset, fcMin);
    TextPieceTable tpt = cft.getTextPieceTable();
    CHPBinTable cbt = new CHPBinTable(header, tableStream, chpOffset, chpSize, fcMin, tpt);

    List textPieces = tpt.getTextPieces();

    // make the POIFS objects available for garbage collection
    din = null;
    fsys = null;
    table = null;
View Full Code Here
82
83
84
85
86
87
88
89
90
91
92
            PieceDescriptor pd = new PieceDescriptor(new byte[] {0,0, 0,0,0,127, 0,0}, 0);
            pd.setFilePosition(_fib.getFcMin());

            // Generate a single Text Piece Table, with a single Text Piece
            //  which covers all the (8 bit only) text in the file
            tpt = new TextPieceTable();
            byte[] textData = new byte[_fib.getFcMac()-_fib.getFcMin()];
            System.arraycopy(_mainStream, _fib.getFcMin(), textData, 0, textData.length);
            TextPiece tp = new TextPiece(
                    0, textData.length, textData, pd, 0
            );
View Full Code Here
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
    //fcMin = _fib.getFcMin()

    // Start to load up our standard structures.
    _dop = new DocumentProperties(_tableStream, _fib.getFcDop(), _fib.getLcbDop() );
    _cft = new ComplexFileTable(_mainStream, _tableStream, _fib.getFcClx(), fcMin);
    TextPieceTable _tpt = _cft.getTextPieceTable();

    // Now load the rest of the properties, which need to be adjusted
    //  for where text really begin
    _cbt = new CHPBinTable(_mainStream, _tableStream, _fib.getFcPlcfbteChpx(), _fib.getLcbPlcfbteChpx(), _tpt);
    _pbt = new PAPBinTable(_mainStream, _tableStream, _dataStream, _fib.getFcPlcfbtePapx(), _fib.getLcbPlcfbtePapx(), _tpt);

        _text = _tpt.getText();

        /*
         * in this mode we preserving PAPX/CHPX structure from file, so text may
         * miss from output, and text order may be corrupted
         */
        boolean preserveBinTables = false;
        try
        {
            preserveBinTables = Boolean.parseBoolean( System
                    .getProperty( PROPERTY_PRESERVE_BIN_TABLES ) );
        }
        catch ( Exception exc )
        {
            // ignore;
        }

        if ( !preserveBinTables )
        {
            _cbt.rebuild( _cft );
            _pbt.rebuild( _text, _cft );
        }

        /*
         * Property to disable text rebuilding. In this mode changing the text
         * will lead to unpredictable behavior
         */
        boolean preserveTextTable = false;
        try
        {
            preserveTextTable = Boolean.parseBoolean( System
                    .getProperty( PROPERTY_PRESERVE_TEXT_TABLE ) );
        }
        catch ( Exception exc )
        {
            // ignore;
        }
        if ( !preserveTextTable )
        {
            _cft = new ComplexFileTable();
            _tpt = _cft.getTextPieceTable();
            final TextPiece textPiece = new SinglentonTextPiece( _text );
            _tpt.add( textPiece );
            _text = textPiece.getStringBuilder();
        }

        // Read FSPA and Escher information
        // _fspa = new FSPATable(_tableStream, _fib.getFcPlcspaMom(),
View Full Code Here
86
87
88
89
90
91
92
93
94
95
96
            PieceDescriptor pd = new PieceDescriptor(new byte[] {0,0, 0,0,0,127, 0,0}, 0);
            pd.setFilePosition(_fib.getFibBase().getFcMin());

            // Generate a single Text Piece Table, with a single Text Piece
            //  which covers all the (8 bit only) text in the file
            tpt = new TextPieceTable();
            byte[] textData = new byte[_fib.getFibBase().getFcMac()-_fib.getFibBase().getFcMin()];
            System.arraycopy(_mainStream, _fib.getFibBase().getFcMin(), textData, 0, textData.length);
            TextPiece tp = new TextPiece(
                    0, textData.length, textData, pd
            );
View Full Code Here
78
79
80
81
82
83
84
85
86
87
88
            PieceDescriptor pd = new PieceDescriptor(new byte[] {0,0, 0,0,0,127, 0,0}, 0);
            pd.setFilePosition(_fib.getFcMin());

            // Generate a single Text Piece Table, with a single Text Piece
            //  which covers all the (8 bit only) text in the file
            tpt = new TextPieceTable();
            byte[] textData = new byte[_fib.getFcMac()-_fib.getFcMin()];
            System.arraycopy(_mainStream, _fib.getFcMin(), textData, 0, textData.length);
            TextPiece tp = new TextPiece(
                    0, textData.length, textData, pd, 0
            );
View Full Code Here
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
    //fcMin = _fib.getFcMin()

    // Start to load up our standard structures.
    _dop = new DocumentProperties(_tableStream, _fib.getFcDop(), _fib.getLcbDop() );
    _cft = new ComplexFileTable(_mainStream, _tableStream, _fib.getFcClx(), fcMin);
    TextPieceTable _tpt = _cft.getTextPieceTable();

    // Now load the rest of the properties, which need to be adjusted
    //  for where text really begin
    _cbt = new CHPBinTable(_mainStream, _tableStream, _fib.getFcPlcfbteChpx(), _fib.getLcbPlcfbteChpx(), _tpt);
    _pbt = new PAPBinTable(_mainStream, _tableStream, _dataStream, _fib.getFcPlcfbtePapx(), _fib.getLcbPlcfbtePapx(), _tpt);

        _text = _tpt.getText();

        /*
         * in this mode we preserving PAPX/CHPX structure from file, so text may
         * miss from output, and text order may be corrupted
         */
        boolean preserveBinTables = false;
        try
        {
            preserveBinTables = Boolean.parseBoolean( System
                    .getProperty( PROPERTY_PRESERVE_BIN_TABLES ) );
        }
        catch ( Exception exc )
        {
            // ignore;
        }

        if ( !preserveBinTables )
        {
            _cbt.rebuild( _cft );
            _pbt.rebuild( _text, _cft );
        }

        /*
         * Property to disable text rebuilding. In this mode changing the text
         * will lead to unpredictable behavior
         */
        boolean preserveTextTable = false;
        try
        {
            preserveTextTable = Boolean.parseBoolean( System
                    .getProperty( PROPERTY_PRESERVE_TEXT_TABLE ) );
        }
        catch ( Exception exc )
        {
            // ignore;
        }
        if ( !preserveTextTable )
        {
            _cft = new ComplexFileTable();
            _tpt = _cft.getTextPieceTable();
            final TextPiece textPiece = new SinglentonTextPiece( _text );
            _tpt.add( textPiece );
            _text = textPiece.getStringBuilder();
        }

        // Read FSPA and Escher information
        // _fspa = new FSPATable(_tableStream, _fib.getFcPlcspaMom(),
View Full Code Here
85
86
87
88
89
90
91
92
93
94
95
            PieceDescriptor pd = new PieceDescriptor(new byte[] {0,0, 0,0,0,127, 0,0}, 0);
            pd.setFilePosition(_fib.getFcMin());

            // Generate a single Text Piece Table, with a single Text Piece
            //  which covers all the (8 bit only) text in the file
            tpt = new TextPieceTable();
            byte[] textData = new byte[_fib.getFcMac()-_fib.getFcMin()];
            System.arraycopy(_mainStream, _fib.getFcMin(), textData, 0, textData.length);
            TextPiece tp = new TextPiece(
                    0, textData.length, textData, pd
            );
View Full Code Here
TOP

Related Classes of org.apache.poi.hwpf.model.TextPieceTable

  • org.apache.nutch.parse.msword.WordExtractor
  • org.apache.poi.hwpf.HWPFDocument
  • org.apache.poi.hwpf.HWPFOldDocument
  • org.apache.tika.parser.microsoft.WordParser
  • java.util.Iterator
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.