Examples with TarArchiveInputStream com.zimbra.cs.service.formatter.TarArchiveInputStream used on opensource projects

Example 1 with TarArchiveInputStream

use of com.zimbra.cs.service.formatter.TarArchiveInputStream in project zm-mailbox by Zimbra.

the class SpamExtract method extractMessages.

private static List<String> extractMessages(HttpClient hc, GetMethod gm, String path, File outdir, boolean raw) throws HttpException, IOException {
    List<String> extractedIds = new ArrayList<String>();
    gm.setPath(path);
    if (LOG.isDebugEnabled()) {
        LOG.debug("Fetching " + path);
    }
    HttpClientUtil.executeMethod(hc, gm);
    if (gm.getStatusCode() != HttpStatus.SC_OK) {
        throw new IOException("HTTP GET failed: " + gm.getPath() + ": " + gm.getStatusCode() + ": " + gm.getStatusText());
    }
    ArchiveInputStream tgzStream = null;
    try {
        tgzStream = new TarArchiveInputStream(new GZIPInputStream(gm.getResponseBodyAsStream()), Charsets.UTF_8.name());
        ArchiveInputEntry entry = null;
        while ((entry = tgzStream.getNextEntry()) != null) {
            LOG.debug("got entry name %s", entry.getName());
            if (entry.getName().endsWith(".meta")) {
                ItemData itemData = new ItemData(readArchiveEntry(tgzStream, entry));
                UnderlyingData ud = itemData.ud;
                //.meta always followed by .eml
                entry = tgzStream.getNextEntry();
                if (raw) {
                    // Write the message as-is.
                    File file = new File(outdir, mOutputPrefix + "-" + mExtractIndex++);
                    OutputStream os = null;
                    try {
                        os = new BufferedOutputStream(new FileOutputStream(file));
                        byte[] data = readArchiveEntry(tgzStream, entry);
                        ByteUtil.copy(new ByteArrayInputStream(data), true, os, false);
                        if (verbose) {
                            LOG.info("Wrote: " + file);
                        }
                        extractedIds.add(ud.id + "");
                    } catch (java.io.IOException e) {
                        String fileName = outdir + "/" + mOutputPrefix + "-" + mExtractIndex;
                        LOG.error("Cannot write to " + fileName, e);
                    } finally {
                        if (os != null) {
                            os.close();
                        }
                    }
                } else {
                    // Write the attached message to the output directory.
                    BufferStream buffer = new BufferStream(entry.getSize(), MAX_BUFFER_SIZE);
                    buffer.setSequenced(false);
                    MimeMessage mm = null;
                    InputStream fis = null;
                    try {
                        byte[] data = readArchiveEntry(tgzStream, entry);
                        ByteUtil.copy(new ByteArrayInputStream(data), true, buffer, false);
                        if (buffer.isSpooled()) {
                            fis = new ZSharedFileInputStream(buffer.getFile());
                            mm = new ZMimeMessage(mJMSession, fis);
                        } else {
                            mm = new ZMimeMessage(mJMSession, buffer.getInputStream());
                        }
                        writeAttachedMessages(mm, outdir, entry.getName());
                        extractedIds.add(ud.id + "");
                    } catch (MessagingException me) {
                        LOG.warn("exception occurred fetching message", me);
                    } finally {
                        ByteUtil.closeStream(fis);
                    }
                }
            }
        }
    } finally {
        Closeables.closeQuietly(tgzStream);
    }
    return extractedIds;
}

Example 2 with TarArchiveInputStream

use of com.zimbra.cs.service.formatter.TarArchiveInputStream in project zm-mailbox by Zimbra.

the class SpamExtract method extractMessages.

private static List<String> extractMessages(HttpClientBuilder hc, HttpGet gm, String path, File outdir, boolean raw) throws HttpException, IOException {
    List<String> extractedIds = new ArrayList<String>();
    HttpClient client = hc.build();
    if (LOG.isDebugEnabled()) {
        LOG.debug("Fetching " + path);
    }
    try {
        URI uri = new URI(path);
        gm.setURI(uri);
    } catch (URISyntaxException e) {
        LOG.warn("exception occurred in URI path", e);
    }
    HttpResponse httpResp = HttpClientUtil.executeMethod(client, gm);
    if (httpResp.getStatusLine().getStatusCode() != HttpStatus.SC_OK) {
        throw new IOException("HTTP GET failed: " + gm.getRequestLine().getUri() + ": " + httpResp.getStatusLine().getStatusCode() + ": " + httpResp.getStatusLine().getReasonPhrase());
    }
    try (ArchiveInputStream tgzStream = new TarArchiveInputStream(new GZIPInputStream(httpResp.getEntity().getContent()), Charsets.UTF_8.name())) {
        ArchiveInputEntry entry = null;
        while ((entry = tgzStream.getNextEntry()) != null) {
            LOG.debug("got entry name %s", entry.getName());
            if (entry.getName().endsWith(".meta")) {
                ItemData itemData = new ItemData(readArchiveEntry(tgzStream, entry));
                UnderlyingData ud = itemData.ud;
                // .meta always followed
                entry = tgzStream.getNextEntry();
                // by .eml
                if (raw) {
                    // Write the message as-is.
                    File file = new File(outdir, mOutputPrefix + "-" + mExtractIndex++);
                    OutputStream os = null;
                    try {
                        os = new BufferedOutputStream(new FileOutputStream(file));
                        byte[] data = readArchiveEntry(tgzStream, entry);
                        ByteUtil.copy(new ByteArrayInputStream(data), true, os, false);
                        if (verbose) {
                            LOG.info("Wrote: " + file);
                        }
                        extractedIds.add(ud.id + "");
                    } catch (java.io.IOException e) {
                        String fileName = outdir + "/" + mOutputPrefix + "-" + mExtractIndex;
                        LOG.error("Cannot write to " + fileName, e);
                    } finally {
                        if (os != null) {
                            os.close();
                        }
                    }
                } else {
                    // Write the attached message to the output directory.
                    BufferStream buffer = new BufferStream(entry.getSize(), MAX_BUFFER_SIZE);
                    buffer.setSequenced(false);
                    MimeMessage mm = null;
                    InputStream fis = null;
                    try {
                        byte[] data = readArchiveEntry(tgzStream, entry);
                        ByteUtil.copy(new ByteArrayInputStream(data), true, buffer, false);
                        if (buffer.isSpooled()) {
                            fis = new ZSharedFileInputStream(buffer.getFile());
                            mm = new ZMimeMessage(mJMSession, fis);
                        } else {
                            mm = new ZMimeMessage(mJMSession, buffer.getInputStream());
                        }
                        writeAttachedMessages(mm, outdir, entry.getName());
                        extractedIds.add(ud.id + "");
                    } catch (MessagingException me) {
                        LOG.warn("exception occurred fetching message", me);
                    } finally {
                        ByteUtil.closeStream(fis);
                    }
                }
            }
        }
    }
    return extractedIds;
}

Example 3 with TarArchiveInputStream

use of com.zimbra.cs.service.formatter.TarArchiveInputStream in project zm-mailbox by Zimbra.

the class ContactTest method testTruncatedContactsTgzImport.

@Test
public void testTruncatedContactsTgzImport() throws IOException {
    File file = new File(MailboxTestUtil.getZimbraServerDir("") + "src/java-test/Truncated.tgz");
    System.out.println(file.getAbsolutePath());
    InputStream is = new FileInputStream(file);
    ArchiveInputStream ais = new TarArchiveInputStream(new GZIPInputStream(is), "UTF-8");
    ArchiveInputEntry aie;
    boolean errorCaught = false;
    while ((aie = ais.getNextEntry()) != null) {
        try {
            ArchiveFormatter.readArchiveEntry(ais, aie);
        } catch (IOException e) {
            e.printStackTrace();
            errorCaught = true;
            break;
        }
    }
    Assert.assertTrue(errorCaught);
}