Examples with DocumentInputStream org.apache.poi.poifs.filesystem.DocumentInputStream used on opensource projects

Example 6 with DocumentInputStream

use of org.apache.poi.poifs.filesystem.DocumentInputStream in project poi by apache.

the class VBAMacroReader method readModule.

private static void readModule(DocumentInputStream dis, String name, ModuleMap modules) throws IOException {
    Module module = modules.get(name);
    // TODO Refactor this to fetch dir then do the rest
    if (module == null) {
        // no DIR stream with offsets yet, so store the compressed bytes for later
        module = new Module();
        modules.put(name, module);
        module.read(dis);
    } else if (module.buf == null) {
        //if we haven't already read the bytes for the module keyed off this name...
        if (module.offset == null) {
            //This should not happen. bug 59858
            throw new IOException("Module offset for '" + name + "' was never read.");
        }
        // we know the offset already, so decompress immediately on-the-fly
        long skippedBytes = dis.skip(module.offset);
        if (skippedBytes != module.offset) {
            throw new IOException("tried to skip " + module.offset + " bytes, but actually skipped " + skippedBytes + " bytes");
        }
        InputStream stream = new RLEDecompressingInputStream(dis);
        module.read(stream);
        stream.close();
    }
}

Example 7 with DocumentInputStream

use of org.apache.poi.poifs.filesystem.DocumentInputStream in project poi by apache.

the class CopyCompare method equal.

/**
     * <p>Compares two {@link DocumentEntry} instances of a POI file system.
     * Documents that are not property set streams must be bitwise identical.
     * Property set streams must be logically equal.</p>
     *
     * @param d1 The first document.
     * @param d2 The second document.
     * @param msg The method may append human-readable comparison messages to
     * this string buffer. 
     * @return <code>true</code> if the documents are equal, else
     * <code>false</code>.
     * @exception MarkUnsupportedException if a POI document stream does not
     * support the mark() operation.
     * @exception NoPropertySetStreamException if the application tries to
     * create a property set from a POI document stream that is not a property
     * set stream.
     * @throws UnsupportedEncodingException 
     * @exception IOException if any I/O exception occurs.
     */
private static boolean equal(final DocumentEntry d1, final DocumentEntry d2, final StringBuffer msg) throws NoPropertySetStreamException, MarkUnsupportedException, UnsupportedEncodingException, IOException {
    final DocumentInputStream dis1 = new DocumentInputStream(d1);
    final DocumentInputStream dis2 = new DocumentInputStream(d2);
    try {
        if (PropertySet.isPropertySetStream(dis1) && PropertySet.isPropertySetStream(dis2)) {
            final PropertySet ps1 = PropertySetFactory.create(dis1);
            final PropertySet ps2 = PropertySetFactory.create(dis2);
            if (!ps1.equals(ps2)) {
                msg.append("Property sets are not equal.\n");
                return false;
            }
        } else {
            int i1, i2;
            do {
                i1 = dis1.read();
                i2 = dis2.read();
                if (i1 != i2) {
                    msg.append("Documents are not equal.\n");
                    return false;
                }
            } while (i1 > -1);
        }
    } finally {
        dis2.close();
        dis1.close();
    }
    return true;
}

Example 8 with DocumentInputStream

use of org.apache.poi.poifs.filesystem.DocumentInputStream in project poi by apache.

the class AgileDecryptor method getDataStream.

@Override
public InputStream getDataStream(DirectoryNode dir) throws IOException, GeneralSecurityException {
    DocumentInputStream dis = dir.createDocumentInputStream(DEFAULT_POIFS_ENTRY);
    _length = dis.readLong();
    return new AgileCipherInputStream(dis, _length);
}

Example 9 with DocumentInputStream

use of org.apache.poi.poifs.filesystem.DocumentInputStream in project tika by apache.

the class RTFObjDataParser method handleEmbeddedPOIFS.

//will throw IOException if not actually POIFS
//can return null byte[]
private byte[] handleEmbeddedPOIFS(InputStream is, Metadata metadata, AtomicInteger unknownFilenameCount) throws IOException {
    byte[] ret = null;
    try (NPOIFSFileSystem fs = new NPOIFSFileSystem(is)) {
        DirectoryNode root = fs.getRoot();
        if (root == null) {
            return ret;
        }
        if (root.hasEntry("Package")) {
            Entry ooxml = root.getEntry("Package");
            TikaInputStream stream = TikaInputStream.get(new DocumentInputStream((DocumentEntry) ooxml));
            ByteArrayOutputStream out = new ByteArrayOutputStream();
            IOUtils.copy(stream, out);
            ret = out.toByteArray();
        } else {
            //try poifs
            POIFSDocumentType type = POIFSDocumentType.detectType(root);
            if (type == POIFSDocumentType.OLE10_NATIVE) {
                try {
                    // Try to un-wrap the OLE10Native record:
                    Ole10Native ole = Ole10Native.createFromEmbeddedOleObject(root);
                    ret = ole.getDataBuffer();
                } catch (Ole10NativeException ex) {
                // Not a valid OLE10Native record, skip it
                }
            } else if (type == POIFSDocumentType.COMP_OBJ) {
                DocumentEntry contentsEntry;
                try {
                    contentsEntry = (DocumentEntry) root.getEntry("CONTENTS");
                } catch (FileNotFoundException ioe) {
                    contentsEntry = (DocumentEntry) root.getEntry("Contents");
                }
                try (DocumentInputStream inp = new DocumentInputStream(contentsEntry)) {
                    ret = new byte[contentsEntry.getSize()];
                    inp.readFully(ret);
                }
            } else {
                ByteArrayOutputStream out = new ByteArrayOutputStream();
                is.reset();
                IOUtils.copy(is, out);
                ret = out.toByteArray();
                metadata.set(Metadata.RESOURCE_NAME_KEY, "file_" + unknownFilenameCount.getAndIncrement() + "." + type.getExtension());
                metadata.set(Metadata.CONTENT_TYPE, type.getType().toString());
            }
        }
    }
    return ret;
}

Example 10 with DocumentInputStream

use of org.apache.poi.poifs.filesystem.DocumentInputStream in project tika by apache.

the class POIFSContainerDetector method processCompObjFormatType.

/**
     * Is this one of the kinds of formats which uses CompObj to
     * store all of their data, eg Star Draw, Star Impress or
     * (older) Works?
     * If not, it's likely an embedded resource
     */
private static MediaType processCompObjFormatType(DirectoryEntry root) {
    try {
        Entry e = root.getEntry("
CompObj");
        if (e != null && e.isDocumentEntry()) {
            DocumentNode dn = (DocumentNode) e;
            DocumentInputStream stream = new DocumentInputStream(dn);
            byte[] bytes = IOUtils.toByteArray(stream);
            /*
                 * This array contains a string with a normal ASCII name of the
                 * application used to create this file. We want to search for that
                 * name.
                 */
            if (arrayContains(bytes, MS_GRAPH_CHART_BYTES)) {
                return MS_GRAPH_CHART;
            } else if (arrayContains(bytes, STAR_DRAW)) {
                return SDA;
            } else if (arrayContains(bytes, STAR_IMPRESS)) {
                return SDD;
            } else if (arrayContains(bytes, WORKS_QUILL96)) {
                return WPS;
            }
        }
    } catch (Exception e) {
    /*
             * "root.getEntry" can throw FileNotFoundException. The code inside
             * "if" can throw IOExceptions. Theoretically. Practically no
             * exceptions will likely ever appear.
             *
             * Swallow all of them. If any occur, we just assume that we can't
             * distinguish between Draw and Impress and return something safe:
             * x-tika-msoffice
             */
    }
    return OLE;
}