Example usage for org.apache.poi.util LittleEndian getInt

List of usage examples for org.apache.poi.util LittleEndian getInt

  1. HOME
  2. Java
  3. org.apache.poi
  4. org.apache.poi.util.*
  5. LittleEndian
  6. getInt

Introduction

In this page you can find the example usage for org.apache.poi.util LittleEndian getInt.

Prototype

public static int getInt(byte[] data, int offset)

Source Link

Document

get an int value from a byte array

Usage

From source file:com.krawler.esp.fileparser.word.ExtractWordFile.java

License:Open Source License

public String extractText(String filepath) throws FastSavedException, IOException {
    InputStream iStream = new BufferedInputStream(new FileInputStream(filepath));
    POIFSFileSystem fsys = new POIFSFileSystem(iStream);

    // load our POIFS document streams.
    DocumentEntry headerProps = (DocumentEntry) fsys.getRoot().getEntry("WordDocument");
    DocumentInputStream din = fsys.createDocumentInputStream("WordDocument");
    byte[] header = new byte[headerProps.getSize()];

    din.read(header);/* w w w.  j  a  v a  2s . c  om*/
    din.close();

    int info = LittleEndian.getShort(header, 0xa);
    if ((info & 0x4) != 0) {
        throw new FastSavedException("Fast-saved files are unsupported at this time");
    }
    if ((info & 0x100) != 0) {
        System.out.println("This document is password protected");
    }

    // determine the version of Word this document came from.
    int nFib = LittleEndian.getShort(header, 0x2);
    switch (nFib) {
    case 101:
    case 102:
    case 103:
    case 104:
        // this is a Word 6.0 doc send it to the extractor for that version.
        Word6Extractor oldExtractor = new Word6Extractor();
        return oldExtractor.extractText(header);
    }

    // Get the information we need from the header
    boolean useTable1 = (info & 0x200) != 0;

    // get the location of the piece table
    int complexOffset = LittleEndian.getInt(header, 0x1a2);

    // determine which table stream we must use.
    String tableName = null;
    if (useTable1) {
        tableName = "1Table";
    } else {
        tableName = "0Table";
    }

    DocumentEntry table = (DocumentEntry) fsys.getRoot().getEntry(tableName);
    byte[] tableStream = new byte[table.getSize()];

    din = fsys.createDocumentInputStream(tableName);

    din.read(tableStream);
    din.close();

    int chpOffset = LittleEndian.getInt(header, 0xfa);
    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();
    List textPieces = tpt.getTextPieces();

    CHPBinTable cbt = new CHPBinTable(header, tableStream, chpOffset, chpSize, fcMin, tpt);

    // make the POIFS objects available for garbage collection
    din = null;
    fsys = null;
    table = null;
    headerProps = null;

    List textRuns = cbt.getTextRuns();
    Iterator runIt = textRuns.iterator();
    Iterator textIt = textPieces.iterator();

    TextPiece currentPiece = (TextPiece) textIt.next();
    int currentTextStart = currentPiece.getStart();
    int currentTextEnd = currentPiece.getEnd();

    WordTextBuffer finalTextBuf = new WordTextBuffer();

    // iterate through all text runs extract the text only if they haven't
    // been
    // deleted
    while (runIt.hasNext()) {
        CHPX chpx = (CHPX) runIt.next();
        boolean deleted = isDeleted(chpx.getGrpprl());
        if (deleted) {
            continue;
        }

        int runStart = chpx.getStart();
        int runEnd = chpx.getEnd();

        while (runStart >= currentTextEnd) {
            currentPiece = (TextPiece) textIt.next();
            currentTextStart = currentPiece.getStart();
            currentTextEnd = currentPiece.getEnd();
        }

        if (runEnd < currentTextEnd) {
            String str = currentPiece.substring(runStart - currentTextStart, runEnd - currentTextStart);
            finalTextBuf.append(str);
        } else if (runEnd > currentTextEnd) {
            while (runEnd > currentTextEnd) {
                String str = currentPiece.substring(runStart - currentTextStart,
                        currentTextEnd - currentTextStart);
                finalTextBuf.append(str);
                if (textIt.hasNext()) {
                    currentPiece = (TextPiece) textIt.next();
                    currentTextStart = currentPiece.getStart();
                    runStart = currentTextStart;
                    currentTextEnd = currentPiece.getEnd();
                } else {
                    return finalTextBuf.toString();
                }
            }
            String str = currentPiece.substring(0, runEnd - currentTextStart);
            finalTextBuf.append(str);
        } else {
            String str = currentPiece.substring(runStart - currentTextStart, runEnd - currentTextStart);
            if (textIt.hasNext()) {
                currentPiece = (TextPiece) textIt.next();
                currentTextStart = currentPiece.getStart();
                currentTextEnd = currentPiece.getEnd();
            }
            finalTextBuf.append(str);
        }
    }
    return finalTextBuf.toString();
}

From source file:com.krawler.esp.fileparser.word.Word6Extractor.java

License:Open Source License

public String extractText(byte[] mainStream) throws IOException {
    int fcMin = LittleEndian.getInt(mainStream, 0x18);
    int fcMax = LittleEndian.getInt(mainStream, 0x1C);

    int chpTableOffset = LittleEndian.getInt(mainStream, 0xb8);
    int chpTableSize = LittleEndian.getInt(mainStream, 0xbc);

    // get a list of character properties
    Word6CHPBinTable chpTable = new Word6CHPBinTable(mainStream, chpTableOffset, chpTableSize, fcMin);
    List textRuns = chpTable.getTextRuns();

    // iterate through the
    WordTextBuffer finalTextBuf = new WordTextBuffer();
    Iterator runsIt = textRuns.iterator();
    while (runsIt.hasNext()) {
        CHPX chpx = (CHPX) runsIt.next();
        int runStart = chpx.getStart() + fcMin;
        int runEnd = chpx.getEnd() + fcMin;

        if (!isDeleted(chpx.getGrpprl())) {
            String s = new String(mainStream, runStart, Math.min(runEnd, fcMax) - runStart, "Cp1252");
            finalTextBuf.append(s);/*www .  j  a  v  a 2 s.c om*/
            if (runEnd >= fcMax) {
                break;
            }
        }
    }

    return finalTextBuf.toString();
}

From source file:com.krawler.esp.fileparser.word.Word6Extractor.java

License:Open Source License

public String extractText(byte[] mainStream, TextPieceTable tpt) throws IOException {
    int fcMin = LittleEndian.getInt(mainStream, 0x18);
    int fcMax = LittleEndian.getInt(mainStream, 0x1C);

    int chpTableOffset = LittleEndian.getInt(mainStream, 0xb8);
    int chpTableSize = LittleEndian.getInt(mainStream, 0xbc);

    // get a list of character properties
    Word6CHPBinTable chpTable = new Word6CHPBinTable(mainStream, chpTableOffset, chpTableSize, fcMin, tpt);
    List textRuns = chpTable.getTextRuns();

    // iterate through the
    WordTextBuffer finalTextBuf = new WordTextBuffer();
    Iterator runsIt = textRuns.iterator();
    while (runsIt.hasNext()) {
        CHPX chpx = (CHPX) runsIt.next();
        int runStart = chpx.getStart() + fcMin;
        int runEnd = chpx.getEnd() + fcMin;

        if (!isDeleted(chpx.getGrpprl())) {
            String s = new String(mainStream, runStart, Math.min(runEnd, fcMax) - runStart, "Cp1252");
            finalTextBuf.append(s);/*from  w ww .  j  av  a2s .  c om*/
            if (runEnd >= fcMax) {
                break;
            }
        }
    }

    return finalTextBuf.toString();
}

From source file:com.krawler.esp.fileparser.wordparser.ExtractWordFile.java

License:Open Source License

public String extractText(String filepath) throws FastSavedException, IOException {
    InputStream iStream = new BufferedInputStream(new FileInputStream(filepath));

    ArrayList text = new ArrayList();
    POIFSFileSystem fsys = new POIFSFileSystem(iStream);

    // load our POIFS document streams.
    DocumentEntry headerProps = (DocumentEntry) fsys.getRoot().getEntry("WordDocument");
    DocumentInputStream din = fsys.createDocumentInputStream("WordDocument");
    byte[] header = new byte[headerProps.getSize()];

    din.read(header);//from  ww w . j  av a  2s.c  o m
    din.close();

    int info = LittleEndian.getShort(header, 0xa);
    if ((info & 0x4) != 0) {
        throw new FastSavedException("Fast-saved files are unsupported at this time");
    }
    if ((info & 0x100) != 0) {
        System.out.println("This document is password protected");
    }

    // determine the version of Word this document came from.
    int nFib = LittleEndian.getShort(header, 0x2);
    // Get the information we need from the header
    boolean useTable1 = (info & 0x200) != 0;

    // get the location of the piece table
    int complexOffset = LittleEndian.getInt(header, 0x1a2);

    // determine which table stream we must use.
    String tableName = null;
    if (useTable1) {
        tableName = "1Table";
    } else {
        tableName = "0Table";
    }

    DocumentEntry table = (DocumentEntry) fsys.getRoot().getEntry(tableName);
    byte[] tableStream = new byte[table.getSize()];

    din = fsys.createDocumentInputStream(tableName);

    din.read(tableStream);
    din.close();

    int chpOffset = LittleEndian.getInt(header, 0xfa);
    int chpSize = LittleEndian.getInt(header, 0xfe);
    int fcMin = LittleEndian.getInt(header, 0x18);

    ComplexFileTable cft = new ComplexFileTable(header, tableStream, complexOffset, fcMin);
    TextPieceTable tpt = cft.getTextPieceTable();
    switch (nFib) {
    case 101:
    case 102:
    case 103:
    case 104:
        // this is a Word 6.0 doc send it to the extractor for that version.
        Word6Extractor oldExtractor = new Word6Extractor();
        return oldExtractor.extractText(header, tpt);
    }
    CHPBinTable cbt = new CHPBinTable(header, tableStream, chpOffset, chpSize, fcMin, tpt);
    // load our text pieces and our character runs

    List textPieces = tpt.getTextPieces();

    // make the POIFS objects available for garbage collection
    din = null;
    fsys = null;
    table = null;
    headerProps = null;

    List textRuns = cbt.getTextRuns();
    Iterator runIt = textRuns.iterator();
    Iterator textIt = textPieces.iterator();

    TextPiece currentPiece = (TextPiece) textIt.next();
    int currentTextStart = currentPiece.getStart();
    int currentTextEnd = currentPiece.getEnd();

    WordTextBuffer finalTextBuf = new WordTextBuffer();

    // iterate through all text runs extract the text only if they haven't
    // been
    // deleted
    while (runIt.hasNext()) {
        CHPX chpx = (CHPX) runIt.next();
        boolean deleted = isDeleted(chpx.getGrpprl());
        if (deleted) {
            continue;
        }

        int runStart = chpx.getStart();
        int runEnd = chpx.getEnd();

        while (runStart >= currentTextEnd) {
            currentPiece = (TextPiece) textIt.next();
            currentTextStart = currentPiece.getStart();
            currentTextEnd = currentPiece.getEnd();
        }

        if (runEnd < currentTextEnd) {
            String str = currentPiece.substring(runStart - currentTextStart, runEnd - currentTextStart);
            finalTextBuf.append(str);
        } else if (runEnd > currentTextEnd) {
            while (runEnd > currentTextEnd) {
                String str = currentPiece.substring(runStart - currentTextStart,
                        currentTextEnd - currentTextStart);
                finalTextBuf.append(str);
                if (textIt.hasNext()) {
                    currentPiece = (TextPiece) textIt.next();
                    currentTextStart = currentPiece.getStart();
                    runStart = currentTextStart;
                    currentTextEnd = currentPiece.getEnd();
                } else {
                    return finalTextBuf.toString();
                }
            }
            String str = currentPiece.substring(0, runEnd - currentTextStart);
            finalTextBuf.append(str);
        } else {
            String str = currentPiece.substring(runStart - currentTextStart, runEnd - currentTextStart);
            if (textIt.hasNext()) {
                currentPiece = (TextPiece) textIt.next();
                currentTextStart = currentPiece.getStart();
                currentTextEnd = currentPiece.getEnd();
            }
            finalTextBuf.append(str);
        }
    }
    return finalTextBuf.toString();
}

From source file:com.progdan.doc2txt.sprm.SprmOperation.java

License:Apache License

public SprmOperation(byte[] grpprl, int offset) {
    short sprmStart = LittleEndian.getShort(grpprl, offset);
    offset += 2;//w  w w .j a  va  2 s.c  om

    _operation = OP_BITFIELD.getValue(sprmStart);
    _type = TYPE_BITFIELD.getValue(sprmStart);
    int sizeCode = SIZECODE_BITFIELD.getValue(sprmStart);

    switch (sizeCode) {
    case 0:
    case 1:
        _operand = LittleEndian.getUnsignedByte(grpprl, offset);
        _sizeNeeded = 3;
        break;
    case 2:
    case 4:
    case 5:
        _operand = LittleEndian.getShort(grpprl, offset);
        _sizeNeeded = 4;
        break;
    case 3:
        _operand = LittleEndian.getInt(grpprl, offset);
        _sizeNeeded = 6;
        break;
    case 6:
        _varOperand = new byte[grpprl[offset++]];
        System.arraycopy(grpprl, offset, _varOperand, 0, _varOperand.length);
        _sizeNeeded = _varOperand.length + 3;
        break;
    case 7:
        byte threeByteInt[] = new byte[4];
        threeByteInt[0] = grpprl[offset];
        threeByteInt[1] = grpprl[offset + 1];
        threeByteInt[2] = grpprl[offset + 2];
        threeByteInt[3] = (byte) 0;
        _operand = LittleEndian.getInt(threeByteInt, 0);
        _sizeNeeded = 5;
        break;

    }
}

From source file:com.progdan.doc2txt.Word6Extractor.java

License:Apache License

/**
 * Extracts the text/*from  www .  j a  va  2s  .com*/
 *
 * @param mainStream The POIFS document stream entitled "WordDocument".
 *
 * @return The text from the document
 * @throws Exception If there are any unexpected exceptions.
 */
public String extractText(byte[] mainStream) throws Exception {
    int fcMin = LittleEndian.getInt(mainStream, 0x18);
    int fcMax = LittleEndian.getInt(mainStream, 0x1C);

    int chpTableOffset = LittleEndian.getInt(mainStream, 0xb8);
    int chpTableSize = LittleEndian.getInt(mainStream, 0xbc);

    // get a list of character properties
    Word6CHPBinTable chpTable = new Word6CHPBinTable(mainStream, chpTableOffset, chpTableSize, fcMin);
    List textRuns = chpTable.getTextRuns();

    // iterate through the
    WordTextBuffer finalTextBuf = new WordTextBuffer();
    Iterator runsIt = textRuns.iterator();
    while (runsIt.hasNext()) {
        CHPX chpx = (CHPX) runsIt.next();
        int runStart = chpx.getStart() + fcMin;
        int runEnd = chpx.getEnd() + fcMin;

        if (!isDeleted(chpx.getGrpprl())) {
            String s = new String(mainStream, runStart, Math.min(runEnd, fcMax) - runStart, "Cp1252");
            finalTextBuf.append(s);
            if (runEnd >= fcMax) {
                break;
            }
        }
    }

    return finalTextBuf.toString();
}

From source file:com.progdan.doc2txt.WordExtractor.java

License:Apache License

/**
 * Gets the text from a Word document./*www  .j  a v a2s.c  o m*/
 *
 * @param in The InputStream representing the Word file.
 */
public String extractText(InputStream in) throws Exception {
    ArrayList text = new ArrayList();
    POIFSFileSystem fsys = new POIFSFileSystem(in);

    // load our POIFS document streams.
    DocumentEntry headerProps = (DocumentEntry) fsys.getRoot().getEntry("WordDocument");
    DocumentInputStream din = fsys.createDocumentInputStream("WordDocument");
    byte[] header = new byte[headerProps.getSize()];

    din.read(header);
    din.close();

    int info = LittleEndian.getShort(header, 0xa);
    if ((info & 0x4) != 0) {
        throw new FastSavedException("Fast-saved files are unsupported at this time");
    }
    if ((info & 0x100) != 0) {
        throw new PasswordProtectedException("This document is password protected");
    }

    // determine the version of Word this document came from.
    int nFib = LittleEndian.getShort(header, 0x2);
    switch (nFib) {
    case 101:
    case 102:
    case 103:
    case 104:
        // this is a Word 6.0 doc send it to the extractor for that version.
        Word6Extractor oldExtractor = new Word6Extractor();
        return oldExtractor.extractText(header);
    }

    //Get the information we need from the header
    boolean useTable1 = (info & 0x200) != 0;

    //get the location of the piece table
    int complexOffset = LittleEndian.getInt(header, 0x1a2);

    // determine which table stream we must use.
    String tableName = null;
    if (useTable1) {
        tableName = "1Table";
    } else {
        tableName = "0Table";
    }

    DocumentEntry table = (DocumentEntry) fsys.getRoot().getEntry(tableName);
    byte[] tableStream = new byte[table.getSize()];

    din = fsys.createDocumentInputStream(tableName);

    din.read(tableStream);
    din.close();

    int chpOffset = LittleEndian.getInt(header, 0xfa);
    int chpSize = LittleEndian.getInt(header, 0xfe);
    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;
    headerProps = null;

    List textRuns = cbt.getTextRuns();
    Iterator runIt = textRuns.iterator();
    Iterator textIt = textPieces.iterator();

    TextPiece currentPiece = (TextPiece) textIt.next();
    int currentTextStart = currentPiece.getStart();
    int currentTextEnd = currentPiece.getEnd();

    WordTextBuffer finalTextBuf = new WordTextBuffer();

    // iterate through all text runs extract the text only if they haven't been
    // deleted
    while (runIt.hasNext()) {
        CHPX chpx = (CHPX) runIt.next();
        boolean deleted = isDeleted(chpx.getGrpprl());
        if (deleted) {
            continue;
        }

        int runStart = chpx.getStart();
        int runEnd = chpx.getEnd();

        while (runStart >= currentTextEnd) {
            currentPiece = (TextPiece) textIt.next();
            currentTextStart = currentPiece.getStart();
            currentTextEnd = currentPiece.getEnd();
        }

        if (runEnd < currentTextEnd) {
            String str = currentPiece.substring(runStart - currentTextStart, runEnd - currentTextStart);
            finalTextBuf.append(str);
        } else if (runEnd > currentTextEnd) {
            while (runEnd > currentTextEnd) {
                String str = currentPiece.substring(runStart - currentTextStart,
                        currentTextEnd - currentTextStart);
                finalTextBuf.append(str);
                if (textIt.hasNext()) {
                    currentPiece = (TextPiece) textIt.next();
                    currentTextStart = currentPiece.getStart();
                    runStart = currentTextStart;
                    currentTextEnd = currentPiece.getEnd();
                } else {
                    return finalTextBuf.toString();
                }
            }
            String str = currentPiece.substring(0, runEnd - currentTextStart);
            finalTextBuf.append(str);
        } else {
            String str = currentPiece.substring(runStart - currentTextStart, runEnd - currentTextStart);
            if (textIt.hasNext()) {
                currentPiece = (TextPiece) textIt.next();
                currentTextStart = currentPiece.getStart();
                currentTextEnd = currentPiece.getEnd();
            }
            finalTextBuf.append(str);
        }
    }
    return finalTextBuf.toString();
}

From source file:org.apache.nutch.parse.mspowerpoint.ContentReaderListener.java

License:Apache License

/**
 * Extracts the client text boxes of a slide.
 * /*from  www  .  j av a 2  s .  co  m*/
 * @param containerTextBox
 * @param offset
 * @param pptdata
 * @param offsetPD
 * @return Hashtable
 * @see TextBox
 */
protected Hashtable/* <Long, TextBox> */ extractTextBoxes(final Hashtable/* <Long, TextBox> */ containerTextBox,
        final int offset, final byte[] pptdata, final long offsetPD) {

    // To hold temporary data
    FilteredStringWriter outStream = new FilteredStringWriter();

    TextBox textBox;

    // Traversing the bytearray up to Presist directory position
    for (int i = offset; i < offsetPD - 20; i++) {
        try {
            // Record info
            // final long rinfo = LittleEndian.getUShort(pptdata, (int) i);
            // Record Type
            final long recordType = LittleEndian.getUShort(pptdata, i + 2);
            // Record Size
            final long recordSize = LittleEndian.getUInt(pptdata, i + 4);

            if (recordType == PPTConstants.PPT_ATOM_DRAWINGGROUP) {
                /*
                 * Record type is of Drawing Group
                 */

                // Total number of objects
                // final long objectCount = LittleEndian.getUInt(pptdata, (int) i +
                // 8);
                // currentID = Group ID+number of objects
                long currentID = LittleEndian.getInt(pptdata, i + 12);
                currentID = ((int) (currentID / 1024)) * 1024;

                if (currentID == PPTConstants.PPT_MASTERSLIDE) {
                    // Ignore Master Slide objects
                    if (LOG.isTraceEnabled()) {
                        LOG.trace("Ignore master slide.");
                    }
                    i++;
                    continue;
                }

                // Check for the ClientTextBox GroupID existence
                if (containerTextBox.containsKey(new Long(currentID))) {
                    // If exists get Client Textbox Group
                    textBox = (TextBox) containerTextBox.get(new Long(currentID));
                    textBox.setContent("");

                } else {
                    textBox = new TextBox(currentID);
                    containerTextBox.put(new Long(currentID), textBox);
                }

                /*
                 * Iterating the bytearray for TextCharAtoms and TextBytesAtom
                 */
                if ((offsetPD - 20) != recordSize) {
                    // TODO something wrong? Probably an OLE-Object, which we ignore.
                    if (LOG.isDebugEnabled()) {
                        LOG.debug("offsetPD - 20=" + (offsetPD - 20) + " recordsize=" + recordSize);
                    }
                } else {
                    for (int startPos = i + 8; startPos < offsetPD - 20 && startPos < recordSize; startPos++) { // && startPos <
                        // recordSize??
                        try {

                            // Record info
                            // final long nrinfo = LittleEndian.getUShort(pptdata, (int) j);

                            // Record Type
                            final long ntype = LittleEndian.getUShort(pptdata, startPos + 2);

                            // Record size
                            // Note that the size doesn't include the 8 byte atom header
                            final long nsize = LittleEndian.getUInt(pptdata, startPos + 4);

                            if (ntype == PPTConstants.PPT_ATOM_DRAWINGGROUP) {
                                /*
                                 * Break the loop if next GroupID found
                                 */
                                i = startPos - 1;
                                break;
                            } else if (ntype == PPTConstants.PPT_ATOM_TEXTBYTE) {
                                // TextByteAtom record
                                outStream = new FilteredStringWriter();
                                long ii = 0;
                                for (ii = startPos + 6; ii <= startPos + 6 + nsize; ii++) {
                                    // For loop to changed to a function
                                    // if ((ii + 2) >= pptdata.length)
                                    // break; // FIXME
                                    outStream.write((char) (pptdata[(int) ii + 2]));
                                }

                                // Setting the identified text for Current
                                // groupID
                                textBox.setContent(textBox.getContent() + outStream.toString());

                            } else if (ntype == PPTConstants.PPT_ATOM_TEXTCHAR) {
                                // TextCharAtom record

                                final String strTempContent = new String(pptdata, startPos + 6,
                                        (int) (nsize) + 2);
                                final byte bytes[] = strTempContent.getBytes();
                                if (true) {
                                    outStream = new FilteredStringWriter();
                                    for (int ii = 0; ii < bytes.length - 1; ii += 2) {
                                        // For loop to changed to a function
                                        outStream.write((char) (pptdata[ii + 2]));
                                    }
                                    textBox.setContent(textBox.getContent() + outStream.toString());
                                } else {
                                    // this version is used within POI
                                    String text = StringUtil.getFromCompressedUnicode(bytes, 0, bytes.length);
                                    textBox.setContent(textBox.getContent() + text);
                                }

                            } else {
                                // ignored
                                // if (LOG.isTraceEnabled()) {
                                //   LOG.trace("Ignored atom type: " + type);
                                // }
                            }
                        } catch (Throwable e) {
                            if (LOG.isErrorEnabled()) {
                                LOG.error("extractTextBoxes", e);
                            }
                            break;
                        }
                    }
                }
            } else {
                // Record type is ignored
                // if (LOG.isTraceEnabled()) {
                //   LOG.trace("Ignored record type: " + type);
                // }
            }
        } catch (Throwable ee) {
            if (LOG.isErrorEnabled()) {
                LOG.error("extractClientTextBoxes", ee);
            }
            break;
        }
    }
    return containerTextBox;
}

From source file:org.apache.nutch.parse.msword.WordExtractor.java

License:Apache License

/**
 * Gets the text from a Word document.//from   ww  w .j  av a 2  s.c  om
 *
 * @param in The InputStream representing the Word file.
 */
protected String extractText(InputStream in) throws Exception {

    ArrayList text = new ArrayList();
    POIFSFileSystem fsys = new POIFSFileSystem(in);

    // load our POIFS document streams.
    DocumentEntry headerProps = (DocumentEntry) fsys.getRoot().getEntry("WordDocument");
    DocumentInputStream din = fsys.createDocumentInputStream("WordDocument");
    byte[] header = new byte[headerProps.getSize()];

    din.read(header);
    din.close();

    int info = LittleEndian.getShort(header, 0xa);
    if ((info & 0x4) != 0) {
        throw new FastSavedException("Fast-saved files are unsupported at this time");
    }
    if ((info & 0x100) != 0) {
        throw new PasswordProtectedException("This document is password protected");
    }

    // determine the version of Word this document came from.
    int nFib = LittleEndian.getShort(header, 0x2);
    switch (nFib) {
    case 101:
    case 102:
    case 103:
    case 104:
        // this is a Word 6.0 doc send it to the extractor for that version.
        Word6Extractor oldExtractor = new Word6Extractor();
        return oldExtractor.extractText(header);
    }

    //Get the information we need from the header
    boolean useTable1 = (info & 0x200) != 0;

    //get the location of the piece table
    int complexOffset = LittleEndian.getInt(header, 0x1a2);

    // determine which table stream we must use.
    String tableName = null;
    if (useTable1) {
        tableName = "1Table";
    } else {
        tableName = "0Table";
    }

    DocumentEntry table = (DocumentEntry) fsys.getRoot().getEntry(tableName);
    byte[] tableStream = new byte[table.getSize()];

    din = fsys.createDocumentInputStream(tableName);

    din.read(tableStream);
    din.close();

    int chpOffset = LittleEndian.getInt(header, 0xfa);
    int chpSize = LittleEndian.getInt(header, 0xfe);
    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;
    headerProps = null;

    List textRuns = cbt.getTextRuns();
    Iterator runIt = textRuns.iterator();
    Iterator textIt = textPieces.iterator();

    TextPiece currentPiece = (TextPiece) textIt.next();
    int currentTextStart = currentPiece.getStart();
    int currentTextEnd = currentPiece.getEnd();

    WordTextBuffer finalTextBuf = new WordTextBuffer();

    // iterate through all text runs extract the text only if they haven't been
    // deleted
    while (runIt.hasNext()) {
        CHPX chpx = (CHPX) runIt.next();
        boolean deleted = isDeleted(chpx.getGrpprl());
        if (deleted) {
            continue;
        }

        int runStart = chpx.getStart();
        int runEnd = chpx.getEnd();

        while (runStart >= currentTextEnd) {
            currentPiece = (TextPiece) textIt.next();
            currentTextStart = currentPiece.getStart();
            currentTextEnd = currentPiece.getEnd();
        }

        if (runEnd < currentTextEnd) {
            String str = currentPiece.substring(runStart - currentTextStart, runEnd - currentTextStart);
            finalTextBuf.append(str);
        } else if (runEnd > currentTextEnd) {
            while (runEnd > currentTextEnd) {
                String str = currentPiece.substring(runStart - currentTextStart,
                        currentTextEnd - currentTextStart);
                finalTextBuf.append(str);
                if (textIt.hasNext()) {
                    currentPiece = (TextPiece) textIt.next();
                    currentTextStart = currentPiece.getStart();
                    runStart = currentTextStart;
                    currentTextEnd = currentPiece.getEnd();
                } else {
                    return finalTextBuf.toString();
                }
            }
            String str = currentPiece.substring(0, runEnd - currentTextStart);
            finalTextBuf.append(str);
        } else {
            String str = currentPiece.substring(runStart - currentTextStart, runEnd - currentTextStart);
            if (textIt.hasNext()) {
                currentPiece = (TextPiece) textIt.next();
                currentTextStart = currentPiece.getStart();
                currentTextEnd = currentPiece.getEnd();
            }
            finalTextBuf.append(str);
        }
    }
    return finalTextBuf.toString();
}

From source file:org.apache.tika.parser.executable.ExecutableParser.java

License:Apache License

/**
 * Parses a DOS or Windows PE file//  w  w w .  j  a  va2 s .  com
 */
public void parsePE(XHTMLContentHandler xhtml, Metadata metadata, InputStream stream, byte[] first4)
        throws TikaException, IOException {
    metadata.add(Metadata.CONTENT_TYPE, PE_EXE.toString());
    metadata.set(PLATFORM, PLATFORM_WINDOWS);

    // Skip over the MS-DOS bit
    byte[] msdosSection = new byte[0x3c - 4];
    IOUtils.readFully(stream, msdosSection);

    // Grab the PE header offset
    int peOffset = LittleEndian.readInt(stream);

    // Sanity check - while it may go anywhere, it's normally in the first few kb
    if (peOffset > 4096 || peOffset < 0x3f)
        return;

    // Skip the rest of the MS-DOS stub (if PE), until we reach what should
    //  be the PE header (if this is a PE executable)
    stream.skip(peOffset - 0x40);

    // Read the PE header
    byte[] pe = new byte[24];
    IOUtils.readFully(stream, pe);

    // Check it really is a PE header
    if (pe[0] == (byte) 'P' && pe[1] == (byte) 'E' && pe[2] == 0 && pe[3] == 0) {
        // Good, has a valid PE signature
    } else {
        // Old style MS-DOS
        return;
    }

    // Read the header values
    int machine = LittleEndian.getUShort(pe, 4);
    int numSectors = LittleEndian.getUShort(pe, 6);
    long createdAt = LittleEndian.getInt(pe, 8);
    long symbolTableOffset = LittleEndian.getInt(pe, 12);
    long numSymbols = LittleEndian.getInt(pe, 16);
    int sizeOptHdrs = LittleEndian.getUShort(pe, 20);
    int characteristcs = LittleEndian.getUShort(pe, 22);

    // Turn this into helpful metadata
    Date createdAtD = new Date(createdAt * 1000l);
    metadata.set(Metadata.CREATION_DATE, createdAtD);

    switch (machine) {
    case 0x14c:
        metadata.set(MACHINE_TYPE, MACHINE_x86_32);
        metadata.set(ENDIAN, Endian.LITTLE.getName());
        metadata.set(ARCHITECTURE_BITS, "32");
        break;
    case 0x8664:
        metadata.set(MACHINE_TYPE, MACHINE_x86_32);
        metadata.set(ENDIAN, Endian.LITTLE.getName());
        metadata.set(ARCHITECTURE_BITS, "64");
        break;
    case 0x200:
        metadata.set(MACHINE_TYPE, MACHINE_IA_64);
        metadata.set(ENDIAN, Endian.LITTLE.getName());
        metadata.set(ARCHITECTURE_BITS, "64");
        break;

    case 0x184:
        metadata.set(MACHINE_TYPE, MACHINE_ALPHA);
        metadata.set(ENDIAN, Endian.LITTLE.getName());
        metadata.set(ARCHITECTURE_BITS, "32");
        break;
    case 0x284:
        metadata.set(MACHINE_TYPE, MACHINE_ALPHA);
        metadata.set(ENDIAN, Endian.LITTLE.getName());
        metadata.set(ARCHITECTURE_BITS, "64");
        break;

    case 0x1c0:
    case 0x1c4:
        metadata.set(MACHINE_TYPE, MACHINE_ARM);
        metadata.set(ENDIAN, Endian.LITTLE.getName());
        metadata.set(ARCHITECTURE_BITS, "32");
        break;

    case 0x268:
        metadata.set(MACHINE_TYPE, MACHINE_M68K);
        metadata.set(ENDIAN, Endian.BIG.getName());
        metadata.set(ARCHITECTURE_BITS, "32");
        break;

    case 0x266:
    case 0x366:
    case 0x466:
        metadata.set(MACHINE_TYPE, MACHINE_MIPS);
        metadata.set(ENDIAN, Endian.BIG.getName());
        metadata.set(ARCHITECTURE_BITS, "16");
        break;
    case 0x162:
    case 0x166:
    case 0x168:
    case 0x169:
        metadata.set(MACHINE_TYPE, MACHINE_MIPS);
        metadata.set(ENDIAN, Endian.LITTLE.getName());
        metadata.set(ARCHITECTURE_BITS, "16");
        break;

    case 0x1f0:
    case 0x1f1:
        metadata.set(MACHINE_TYPE, MACHINE_PPC);
        metadata.set(ENDIAN, Endian.LITTLE.getName());
        metadata.set(ARCHITECTURE_BITS, "32");
        break;

    case 0x1a2:
    case 0x1a3:
        metadata.set(MACHINE_TYPE, MACHINE_SH3);
        metadata.set(ENDIAN, Endian.BIG.getName());
        metadata.set(ARCHITECTURE_BITS, "32");
        break;
    case 0x1a6:
        metadata.set(MACHINE_TYPE, MACHINE_SH4);
        metadata.set(ENDIAN, Endian.BIG.getName());
        metadata.set(ARCHITECTURE_BITS, "32");
        break;
    case 0x1a8:
        metadata.set(MACHINE_TYPE, MACHINE_SH3);
        metadata.set(ENDIAN, Endian.BIG.getName());
        metadata.set(ARCHITECTURE_BITS, "32");
        break;

    case 0x9041:
        metadata.set(MACHINE_TYPE, MACHINE_M32R);
        metadata.set(ENDIAN, Endian.BIG.getName());
        metadata.set(ARCHITECTURE_BITS, "32");
        break;

    case 0xebc:
        metadata.set(MACHINE_TYPE, MACHINE_EFI);
        break;

    default:
        metadata.set(MACHINE_TYPE, MACHINE_UNKNOWN);
        break;
    }
}