Java Code Examples for org.apache.tika.Tika#detect()

/**
     * 将图片文件转换成base64字符串，参数为该图片的路径
     *
     * @param file
     * @return java.lang.String
     */
    public static String fileBase64(File file) {
        try {
            // check content type of the file
            Tika tika = new Tika();
            String contentType =tika.detect(file);
            // read data as byte[]
            byte[] data = Files.readAllBytes(file.toPath());
            // convert byte[] to base64(java7)
            String base64str = DatatypeConverter.printBase64Binary(data);
            // convert byte[] to base64(java8)
//             String base64str = Base64.getEncoder().encodeToString(data);
            // cretate "data URI"
            StringBuilder sb = new StringBuilder();
            sb.append("data:");
            sb.append(contentType);
            sb.append(";base64,");
            sb.append(base64str);
            System.out.println(sb.toString());
            return sb.toString();

        } catch (IOException e) {
            e.printStackTrace();
        }
        return null;
    }
 

ActionResult<Wo> execute(EffectivePerson effectivePerson, byte[] bytes, FormDataContentDisposition disposition)
		throws Exception {
	ActionResult<Wo> result = new ActionResult<>();
	Tika tika = new Tika();
	String type = tika.detect(bytes);
	Wo wo = new Wo();
	switch (type) {
	case ("application/msword"):
		wo.setValue(this.doc(bytes));
		break;
	case ("application/vnd.openxmlformats-officedocument.wordprocessingml.document"):
		wo.setValue(this.docx(bytes));
		break;
	default:
		throw new ExceptionUnsupportType(type);
	}
	result.setData(wo);
	return result;
}
 

/**
 * 利用 Tika 分析 Mime Type
 * 因为 Tika 要解析 File 、 URL 数据流，所以解析需要一定时间。不要用解析扩展名的方法，无法动态判断，不准。
 * <p>
 * Parses the given file and returns the extracted text content.
 *
 * @param file
 * @return
 */
public static String detect(File file) throws Exception {

    //文件不存在
    if (!file.exists()) {
        throw new Exception("exception ! " + file.getAbsoluteFile() + " not existes.");
    }
    Tika t = new Tika();
    return t.detect(file);

}
 

/**
 * 利用 Tika 分析 Mime Type
 * 因为 Tika 要解析 File 、 URL 数据流，所以解析需要一定时间。不要用解析扩展名的方法，无法动态判断，不准。
 * Parses the resource at the given URL and returns the extracted text content.
 *
 * @param url
 * @return
 */
public static String detect(URL url, int timeout) throws Exception {

    //网址不存在
    if (!MyUrlUtils.isURLAvailable(url, timeout)) {
        throw new Exception("exception ! " + url.getAuthority() + " not available");
    }
    Tika t = new Tika();
    return t.detect(url);
}
 

/**
 * Test method demonstrating usage of Apache Tika.
 * 
 * @throws IOException
 */
@Test
public void whenUsingTika_thenSuccess() throws IOException {
    final File file = new File(FILE_LOC);
    final Tika tika = new Tika();
    final String mimeType = tika.detect(file);
    assertEquals(mimeType, PNG_EXT);
}
 

public static byte[] toPdf(String fileName, byte[] bytes, String stamp) throws Exception {

		Config.collect().validate();

		Tika tika = new Tika();
		String type = tika.detect(bytes, fileName);

		switch (Objects.toString(type, "")) {
		case MEDIATYPE_DOC:
			break;
		case MEDIATYPE_DOCX:
			break;
		default:
			throw new ExceptionUnsupportedMediaType(type);
		}

		URL serverUrl = new URL(Config.collect().url() + "/o2_collect_assemble/jaxrs/document/to/pdf");

		HttpURLConnection connection = (HttpURLConnection) serverUrl.openConnection();

		String boundary = "----" + StringTools.uniqueToken();

		connection.setRequestMethod("POST");
		connection.setDoOutput(true);
		connection.setUseCaches(false);
		connection.addRequestProperty("Content-Type", "multipart/form-data; boundary=" + boundary);

		try (OutputStream out = connection.getOutputStream();
				BufferedWriter writer = new BufferedWriter(new OutputStreamWriter(out))) {
			writer.write(twoHyphens + boundary);
			writer.write(CRLF);
			writer.write("Content-Disposition: form-data; name=\"file\"; filename=\""
					+ (StringUtils.isEmpty(fileName) ? StringTools.uniqueToken() : fileName) + "\"");
			writer.write(CRLF);
			writer.write("Content-Type: " + HttpMediaType.APPLICATION_OCTET_STREAM);
			writer.write(CRLF);
			writer.write(CRLF);
			writer.flush();
			out.write(bytes);
			out.flush();
			writer.write(CRLF);
			writer.write(twoHyphens + boundary);
			if (StringUtils.isNotEmpty(stamp)) {
				writer.write(CRLF);
				writer.write("Content-Disposition: form-data; name=\"stamp\"");
				writer.write(CRLF);
				writer.write("Content-Type: " + HttpMediaType.TEXT_PLAIN);
				writer.write(CRLF);
				writer.write(CRLF);
				writer.write(stamp);
				writer.write(CRLF);
				writer.write(twoHyphens + boundary);
			}
			writer.write(twoHyphens);
			writer.flush();
		}

		String respText = null;

		try (InputStream in = connection.getInputStream()) {
			respText = IOUtils.toString(in, DefaultCharset.charset_utf_8);
		}

		if (StringUtils.isNotEmpty(respText)) {
			ActionResponse response = XGsonBuilder.instance().fromJson(respText, ActionResponse.class);
			WrapString wrap = XGsonBuilder.instance().fromJson(response.getData(), WrapString.class);
			return Base64.decodeBase64(wrap.getValue());
		}
		return null;

	}
 

public static byte[] toImage(String fileName, byte[] bytes, String stamp, Integer page) throws Exception {

		Config.collect().validate();

		Tika tika = new Tika();
		String type = tika.detect(bytes, fileName);

		switch (Objects.toString(type, "")) {
		case MEDIATYPE_DOC:
			break;
		case MEDIATYPE_DOCX:
			break;
		default:
			throw new ExceptionUnsupportedMediaType(type);
		}

		URL serverUrl = new URL(Config.collect().url() + "/o2_collect_assemble/jaxrs/document/to/image");

		HttpURLConnection connection = (HttpURLConnection) serverUrl.openConnection();

		String boundary = "----" + StringTools.uniqueToken();

		connection.setRequestMethod("POST");
		connection.setDoOutput(true);
		connection.setUseCaches(false);
		connection.addRequestProperty("Content-Type", "multipart/form-data; boundary=" + boundary);

		try (OutputStream out = connection.getOutputStream();
				BufferedWriter writer = new BufferedWriter(new OutputStreamWriter(out))) {
			writer.write(twoHyphens + boundary);
			writer.write(CRLF);
			writer.write("Content-Disposition: form-data; name=\"file\"; filename=\""
					+ (StringUtils.isEmpty(fileName) ? StringTools.uniqueToken() : fileName) + "\"");
			writer.write(CRLF);
			writer.write("Content-Type: " + HttpMediaType.APPLICATION_OCTET_STREAM);
			writer.write(CRLF);
			writer.write(CRLF);
			writer.flush();
			out.write(bytes);
			out.flush();
			writer.write(CRLF);
			writer.write(twoHyphens + boundary);
			writer.write(CRLF);
			writer.write("Content-Disposition: form-data; name=\"page\"");
			writer.write(CRLF);
			writer.write("Content-Type: " + HttpMediaType.TEXT_PLAIN);
			writer.write(CRLF);
			writer.write(CRLF);
			writer.write("" + ((page == null || page < 0) ? 0 : page));
			writer.write(CRLF);
			writer.write(twoHyphens + boundary);
			if (StringUtils.isNotEmpty(stamp)) {
				writer.write(CRLF);
				writer.write("Content-Disposition: form-data; name=\"stamp\"");
				writer.write(CRLF);
				writer.write("Content-Type: " + HttpMediaType.TEXT_PLAIN);
				writer.write(CRLF);
				writer.write(CRLF);
				writer.write(stamp);
				writer.write(CRLF);
				writer.write(twoHyphens + boundary);
			}
			writer.write(twoHyphens);
			writer.flush();
		}

		String respText = null;

		try (InputStream in = connection.getInputStream()) {
			respText = IOUtils.toString(in, DefaultCharset.charset_utf_8);
		}

		if (StringUtils.isNotEmpty(respText)) {
			ActionResponse response = XGsonBuilder.instance().fromJson(respText, ActionResponse.class);
			WrapString wrap = XGsonBuilder.instance().fromJson(response.getData(), WrapString.class);
			return Base64.decodeBase64(wrap.getValue());
		}
		return null;

	}
 

public static String getMimeType(InputStream s, String fileName) throws IOException {
    final Tika t = new Tika();
    return t.detect(s, fileName);
}
 

public static String getMimeTypeWithByteBuffer(java.nio.ByteBuffer buffer) throws IOException {
    byte[] b = buffer.array();

    Tika tika = new Tika();
    return tika.detect(b);
}
 

public String extractText(InputStream input, String url, List<Outlink> outLinksList) throws IOException {
  String resultText = "";
  ZipInputStream zin = new ZipInputStream(input);
  ZipEntry entry;
  
  while ((entry = zin.getNextEntry()) != null) {
    
    if (!entry.isDirectory()) {
      int size = (int) entry.getSize();
      byte[] b = new byte[size];
      for(int x = 0; x < size; x++) {
        int err = zin.read();
        if(err != -1) {
          b[x] = (byte)err;
        }
      }
      String newurl = url + "/";
      String fname = entry.getName();
      newurl += fname;
      URL aURL = new URL(newurl);
      String base = aURL.toString();
      int i = fname.lastIndexOf('.');
      if (i != -1) {
        // Trying to resolve the Mime-Type
        Tika tika = new Tika();
        String contentType = tika.detect(fname);
        try {
          Metadata metadata = new Metadata();
          metadata.set(Response.CONTENT_LENGTH, Long.toString(entry.getSize()));
          metadata.set(Response.CONTENT_TYPE, contentType);
          Content content = new Content(newurl, base, b, contentType, metadata, this.conf);
          Parse parse = new ParseUtil(this.conf).parse(content).get(content.getUrl());
          ParseData theParseData = parse.getData();
          Outlink[] theOutlinks = theParseData.getOutlinks();
          
          for(int count = 0; count < theOutlinks.length; count++) {
            outLinksList.add(new Outlink(theOutlinks[count].getToUrl(), theOutlinks[count].getAnchor()));
          }
          
          resultText += entry.getName() + " " + parse.getText() + " ";
        } catch (ParseException e) {
          if (LOG.isInfoEnabled()) { 
            LOG.info("fetch okay, but can't parse " + fname + ", reason: " + e.getMessage());
          }
        }
      }
    }
  }
  
  return resultText;
}
 

public static String detectDocTypeUsingFacade(InputStream stream) throws IOException {
    Tika tika = new Tika();
    String mediaType = tika.detect(stream);
    return mediaType;
}