if [[ -z "${im_cmd}" ]] ; then
    echo "ImageMagick (convert or magick) does not exist."
    exit 1
  fi
  check_command pdftoppm
  check_command unoconv
  tmp_pdf_file=/tmp/thumbnail.$$.pdf
  unoconv -e PageRange=1-1 -o ${tmp_pdf_file} -f pdf "${target_file}"
  if [[ ! -f ${tmp_pdf_file} ]] ; then
    echo "unoconv does not work."
    exit 1
  fi
  tmp_png_prefix=/tmp/thumbnail.png.$$

     * @throws CurlException if the content does not exist.
     */
    public InputStream getContentAsStream() throws IOException {
        if (contentCache == null) {
            if (contentException != null) {
                throw new CurlException("The content does not exist.", contentException);
            } else {
                throw new CurlException("The content does not exist.");
            }
        }

     * Stops the crawling process.
     * Sets the crawler status to DONE and interrupts all crawler threads.
     */
    public void stop() {
        if (logger.isInfoEnabled()) {
            logger.info("Stopping crawler: sessionId={}, status={}", crawlerContext.sessionId, crawlerContext.getStatus());
        }
        crawlerContext.setStatus(CrawlerStatus.DONE);
        try {
            if (crawlerThreadGroup != null) {

    /**
     * Recursively deletes a directory and all its contents.
     * Does not throw exceptions, only logs warnings if deletion fails.
     *
     * @param dir the directory to delete
     */
    protected void closeQuietly(final Path dir) {
        if (Files.notExists(dir)) {
            if (logger.isDebugEnabled()) {
                logger.debug("Path does not exist: path={}", dir);
            }
            return;
        }

                } finally {
                    done.set(true);
                }
            }, Thread.currentThread().getName() + "-pdf");
            task.setDaemon(isDaemonThread);
            task.start();
            task.join(timeout);
            if (!done.get()) {
                for (int i = 0; i < 100 && !done.get(); i++) {
                    task.interrupt();

     * The value is, e.g. false <br>
     * comment: Does it enable the Framework internal debug? (true only when emergency)
     * @return The value of found property. (NotNull: if not found, exception but basically no way)
     */
    String getFrameworkDebug();

    /**
     * Is the property for the key 'framework.debug' true? <br>
     * The value is, e.g. false <br>

    }

    public void test_putValue() {
        // Test putValue method
        ExtractData data = new ExtractData();

        data.putValue("title", "Test Title");
        data.putValue("author", "John Doe");

        String[] titleValues = data.getValues("title");
        assertNotNull(titleValues);
        assertEquals(1, titleValues.length);
        assertEquals("Test Title", titleValues[0]);

{
  "title": "Sample Document",
  "author": "John Doe",
  "version": "1.0",
  "published": "2025-01-15",
  "tags": ["crawler", "extractor", "json"],
  "content": {
    "summary": "This is a sample JSON document for testing",
    "body": "The extractor should handle nested objects and arrays properly"
  },
  "metadata": {
    "created_at": "2025-01-01T00:00:00Z",
    "updated_at": "2025-01-15T12:00:00Z"
  }

        assertTrue(content.contains("Email"));
        assertTrue(content.contains("Age"));
        assertTrue(content.contains("Department"));

        // Verify data extraction
        assertTrue(content.contains("John Doe"));
        assertTrue(content.contains("******@****.***"));
        assertTrue(content.contains("Engineering"));
    }

    public void test_columnMetadata() {

        assertEquals("mybucket", getField(conn, "bucketName"));
        assertEquals("", getField(conn, "objectName"));
    }

    /**
     * Test URL with multiple consecutive slashes.
     * Note: Java's URL class does NOT normalize multiple slashes in the path.
     */
    public void test_urlParsing_multipleSlashes() throws Exception {
        URL url = new URL("gcs://mybucket//path//to//object.txt");