Examples with WordExtractor org.apache.poi.hwpf.extractor.WordExtractor used on opensource projects

Example 1 with WordExtractor

use of org.apache.poi.hwpf.extractor.WordExtractor in project Xponents by OpenSextant.

the class MSDocConverter method conversionImplementation.

/** 
     */
@Override
protected ConvertedDocument conversionImplementation(InputStream input, java.io.File doc) throws IOException {
    org.apache.poi.hwpf.extractor.WordExtractor ex = new WordExtractor(input);
    String[] ps = ex.getParagraphText();
    input.close();
    StringBuilder sb = new StringBuilder();
    for (int i = 0; i < ps.length; i++) {
        sb.append(WordExtractor.stripFields(ps[i]).trim());
        sb.append('\n');
    }
    ConvertedDocument textdoc = new ConvertedDocument(doc);
    textdoc.setText(sb.toString());
    ex.close();
    return textdoc;
}

Example 2 with WordExtractor

use of org.apache.poi.hwpf.extractor.WordExtractor in project poi by apache.

the class ExtractorFactory method getEmbededDocsTextExtractors.

/**
     * Returns an array of text extractors, one for each of
     *  the embedded documents in the file (if there are any).
     * If there are no embedded documents, you'll get back an
     *  empty array. Otherwise, you'll get one open
     *  {@link POITextExtractor} for each embedded file.
     */
public static POITextExtractor[] getEmbededDocsTextExtractors(POIOLE2TextExtractor ext) throws IOException, OpenXML4JException, XmlException {
    // All the embedded directories we spotted
    ArrayList<Entry> dirs = new ArrayList<Entry>();
    // For anything else not directly held in as a POIFS directory
    ArrayList<InputStream> nonPOIFS = new ArrayList<InputStream>();
    // Find all the embedded directories
    DirectoryEntry root = ext.getRoot();
    if (root == null) {
        throw new IllegalStateException("The extractor didn't know which POIFS it came from!");
    }
    if (ext instanceof ExcelExtractor) {
        // These are in MBD... under the root
        Iterator<Entry> it = root.getEntries();
        while (it.hasNext()) {
            Entry entry = it.next();
            if (entry.getName().startsWith("MBD")) {
                dirs.add(entry);
            }
        }
    } else if (ext instanceof WordExtractor) {
        // These are in ObjectPool -> _... under the root
        try {
            DirectoryEntry op = (DirectoryEntry) root.getEntry("ObjectPool");
            Iterator<Entry> it = op.getEntries();
            while (it.hasNext()) {
                Entry entry = it.next();
                if (entry.getName().startsWith("_")) {
                    dirs.add(entry);
                }
            }
        } catch (FileNotFoundException e) {
            logger.log(POILogger.INFO, "Ignoring FileNotFoundException while extracting Word document", e.getLocalizedMessage());
        // ignored here
        }
    //} else if(ext instanceof PowerPointExtractor) {
    // Tricky, not stored directly in poifs
    // TODO
    } else if (ext instanceof OutlookTextExtactor) {
        // Stored in the Attachment blocks
        MAPIMessage msg = ((OutlookTextExtactor) ext).getMAPIMessage();
        for (AttachmentChunks attachment : msg.getAttachmentFiles()) {
            if (attachment.getAttachData() != null) {
                byte[] data = attachment.getAttachData().getValue();
                nonPOIFS.add(new ByteArrayInputStream(data));
            } else if (attachment.getAttachmentDirectory() != null) {
                dirs.add(attachment.getAttachmentDirectory().getDirectory());
            }
        }
    }
    // Create the extractors
    if (dirs.size() == 0 && nonPOIFS.size() == 0) {
        return new POITextExtractor[0];
    }
    ArrayList<POITextExtractor> textExtractors = new ArrayList<POITextExtractor>();
    for (Entry dir : dirs) {
        textExtractors.add(createExtractor((DirectoryNode) dir));
    }
    for (InputStream nonPOIF : nonPOIFS) {
        try {
            textExtractors.add(createExtractor(nonPOIF));
        } catch (IllegalArgumentException e) {
            // Ignore, just means it didn't contain
            //  a format we support as yet
            logger.log(POILogger.INFO, "Format not supported yet", e.getLocalizedMessage());
        } catch (XmlException e) {
            throw new IOException(e.getMessage(), e);
        } catch (OpenXML4JException e) {
            throw new IOException(e.getMessage(), e);
        }
    }
    return textExtractors.toArray(new POITextExtractor[textExtractors.size()]);
}

Example 3 with WordExtractor

use of org.apache.poi.hwpf.extractor.WordExtractor in project poi by apache.

the class TestExtractorFactory method testOPOIFS.

@Test
public void testOPOIFS() throws Exception {
    // Excel
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(xls))) instanceof ExcelExtractor);
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(xls))).getText().length() > 200);
    // Word
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(doc))) instanceof WordExtractor);
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(doc))).getText().length() > 120);
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(doc6))) instanceof Word6Extractor);
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(doc6))).getText().length() > 20);
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(doc95))) instanceof Word6Extractor);
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(doc95))).getText().length() > 120);
    // PowerPoint
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(ppt))) instanceof PowerPointExtractor);
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(ppt))).getText().length() > 120);
    // Visio
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(vsd))) instanceof VisioTextExtractor);
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(vsd))).getText().length() > 50);
    // Publisher
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(pub))) instanceof PublisherTextExtractor);
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(pub))).getText().length() > 50);
    // Outlook msg
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(msg))) instanceof OutlookTextExtactor);
    assertTrue(ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(msg))).getText().length() > 50);
    // Text
    try {
        ExtractorFactory.createExtractor(new OPOIFSFileSystem(new FileInputStream(txt)));
        fail();
    } catch (IOException e) {
    // Good
    }
}

Example 4 with WordExtractor

use of org.apache.poi.hwpf.extractor.WordExtractor in project poi by apache.

the class TestExtractorFactory method testInputStream.

@Test
public void testInputStream() throws Exception {
    // Excel
    POITextExtractor extractor = ExtractorFactory.createExtractor(new FileInputStream(xls));
    assertTrue(extractor instanceof ExcelExtractor);
    assertTrue(extractor.getText().length() > 200);
    extractor.close();
    extractor = ExtractorFactory.createExtractor(new FileInputStream(xlsx));
    assertTrue(extractor.getClass().getName(), extractor instanceof XSSFExcelExtractor);
    assertTrue(extractor.getText().length() > 200);
    // TODO Support OOXML-Strict, see bug #57699
    //        assertTrue(
    //                ExtractorFactory.createExtractor(new FileInputStream(xlsxStrict))
    //                instanceof XSSFExcelExtractor
    //        );
    //        assertTrue(
    //                ExtractorFactory.createExtractor(new FileInputStream(xlsxStrict)).getText().length() > 200
    //        );
    extractor.close();
    // Word
    extractor = ExtractorFactory.createExtractor(new FileInputStream(doc));
    assertTrue(extractor.getClass().getName(), extractor instanceof WordExtractor);
    assertTrue(extractor.getText().length() > 120);
    extractor.close();
    extractor = ExtractorFactory.createExtractor(new FileInputStream(doc6));
    assertTrue(extractor.getClass().getName(), extractor instanceof Word6Extractor);
    assertTrue(extractor.getText().length() > 20);
    extractor.close();
    extractor = ExtractorFactory.createExtractor(new FileInputStream(doc95));
    assertTrue(extractor.getClass().getName(), extractor instanceof Word6Extractor);
    assertTrue(extractor.getText().length() > 120);
    extractor.close();
    extractor = ExtractorFactory.createExtractor(new FileInputStream(docx));
    assertTrue(extractor instanceof XWPFWordExtractor);
    assertTrue(extractor.getText().length() > 120);
    extractor.close();
    // PowerPoint
    extractor = ExtractorFactory.createExtractor(new FileInputStream(ppt));
    assertTrue(extractor instanceof PowerPointExtractor);
    assertTrue(extractor.getText().length() > 120);
    extractor.close();
    extractor = ExtractorFactory.createExtractor(new FileInputStream(pptx));
    assertTrue(extractor instanceof XSLFPowerPointExtractor);
    assertTrue(extractor.getText().length() > 120);
    extractor.close();
    // Visio
    extractor = ExtractorFactory.createExtractor(new FileInputStream(vsd));
    assertTrue(extractor instanceof VisioTextExtractor);
    assertTrue(extractor.getText().length() > 50);
    extractor.close();
    // Visio - vsdx
    extractor = ExtractorFactory.createExtractor(new FileInputStream(vsdx));
    assertTrue(extractor instanceof XDGFVisioExtractor);
    assertTrue(extractor.getText().length() > 20);
    extractor.close();
    // Publisher
    extractor = ExtractorFactory.createExtractor(new FileInputStream(pub));
    assertTrue(extractor instanceof PublisherTextExtractor);
    assertTrue(extractor.getText().length() > 50);
    extractor.close();
    // Outlook msg
    extractor = ExtractorFactory.createExtractor(new FileInputStream(msg));
    assertTrue(extractor instanceof OutlookTextExtactor);
    assertTrue(extractor.getText().length() > 50);
    extractor.close();
    // Text
    try {
        FileInputStream stream = new FileInputStream(txt);
        try {
            ExtractorFactory.createExtractor(stream);
            fail();
        } finally {
            IOUtils.closeQuietly(stream);
        }
    } catch (IllegalArgumentException e) {
    // Good
    }
}

Example 5 with WordExtractor

use of org.apache.poi.hwpf.extractor.WordExtractor in project poi by apache.

the class TestExtractorFactory method testEmbeded.

/**
     * Test embeded docs text extraction. For now, only
     *  does poifs embeded, but will do ooxml ones 
     *  at some point.
     */
@Test
public void testEmbeded() throws Exception {
    POIOLE2TextExtractor ext;
    POITextExtractor[] embeds;
    // No embedings
    ext = (POIOLE2TextExtractor) ExtractorFactory.createExtractor(xls);
    embeds = ExtractorFactory.getEmbededDocsTextExtractors(ext);
    assertEquals(0, embeds.length);
    ext.close();
    // Excel
    ext = (POIOLE2TextExtractor) ExtractorFactory.createExtractor(xlsEmb);
    embeds = ExtractorFactory.getEmbededDocsTextExtractors(ext);
    assertEquals(6, embeds.length);
    int numWord = 0, numXls = 0, numPpt = 0, numMsg = 0, numWordX;
    for (POITextExtractor embed : embeds) {
        assertTrue(embed.getText().length() > 20);
        if (embed instanceof PowerPointExtractor)
            numPpt++;
        else if (embed instanceof ExcelExtractor)
            numXls++;
        else if (embed instanceof WordExtractor)
            numWord++;
        else if (embed instanceof OutlookTextExtactor)
            numMsg++;
    }
    assertEquals(2, numPpt);
    assertEquals(2, numXls);
    assertEquals(2, numWord);
    assertEquals(0, numMsg);
    ext.close();
    // Word
    ext = (POIOLE2TextExtractor) ExtractorFactory.createExtractor(docEmb);
    embeds = ExtractorFactory.getEmbededDocsTextExtractors(ext);
    numWord = 0;
    numXls = 0;
    numPpt = 0;
    numMsg = 0;
    assertEquals(4, embeds.length);
    for (POITextExtractor embed : embeds) {
        assertTrue(embed.getText().length() > 20);
        if (embed instanceof PowerPointExtractor)
            numPpt++;
        else if (embed instanceof ExcelExtractor)
            numXls++;
        else if (embed instanceof WordExtractor)
            numWord++;
        else if (embed instanceof OutlookTextExtactor)
            numMsg++;
    }
    assertEquals(1, numPpt);
    assertEquals(2, numXls);
    assertEquals(1, numWord);
    assertEquals(0, numMsg);
    ext.close();
    // Word which contains an OOXML file
    ext = (POIOLE2TextExtractor) ExtractorFactory.createExtractor(docEmbOOXML);
    embeds = ExtractorFactory.getEmbededDocsTextExtractors(ext);
    numWord = 0;
    numXls = 0;
    numPpt = 0;
    numMsg = 0;
    numWordX = 0;
    assertEquals(3, embeds.length);
    for (POITextExtractor embed : embeds) {
        assertTrue(embed.getText().length() > 20);
        if (embed instanceof PowerPointExtractor)
            numPpt++;
        else if (embed instanceof ExcelExtractor)
            numXls++;
        else if (embed instanceof WordExtractor)
            numWord++;
        else if (embed instanceof OutlookTextExtactor)
            numMsg++;
        else if (embed instanceof XWPFWordExtractor)
            numWordX++;
    }
    assertEquals(1, numPpt);
    assertEquals(1, numXls);
    assertEquals(0, numWord);
    assertEquals(1, numWordX);
    assertEquals(0, numMsg);
    ext.close();
    // Outlook
    ext = (OutlookTextExtactor) ExtractorFactory.createExtractor(msgEmb);
    embeds = ExtractorFactory.getEmbededDocsTextExtractors(ext);
    numWord = 0;
    numXls = 0;
    numPpt = 0;
    numMsg = 0;
    assertEquals(1, embeds.length);
    for (POITextExtractor embed : embeds) {
        assertTrue(embed.getText().length() > 20);
        if (embed instanceof PowerPointExtractor)
            numPpt++;
        else if (embed instanceof ExcelExtractor)
            numXls++;
        else if (embed instanceof WordExtractor)
            numWord++;
        else if (embed instanceof OutlookTextExtactor)
            numMsg++;
    }
    assertEquals(0, numPpt);
    assertEquals(0, numXls);
    assertEquals(1, numWord);
    assertEquals(0, numMsg);
    ext.close();
    // Outlook with another outlook file in it
    ext = (OutlookTextExtactor) ExtractorFactory.createExtractor(msgEmbMsg);
    embeds = ExtractorFactory.getEmbededDocsTextExtractors(ext);
    numWord = 0;
    numXls = 0;
    numPpt = 0;
    numMsg = 0;
    assertEquals(1, embeds.length);
    for (POITextExtractor embed : embeds) {
        assertTrue(embed.getText().length() > 20);
        if (embed instanceof PowerPointExtractor)
            numPpt++;
        else if (embed instanceof ExcelExtractor)
            numXls++;
        else if (embed instanceof WordExtractor)
            numWord++;
        else if (embed instanceof OutlookTextExtactor)
            numMsg++;
    }
    assertEquals(0, numPpt);
    assertEquals(0, numXls);
    assertEquals(0, numWord);
    assertEquals(1, numMsg);
    ext.close();
// TODO - PowerPoint
// TODO - Publisher
// TODO - Visio
}