import org.codelibs.fess.crawler.client.http.RequestHeader;
import org.codelibs.fess.crawler.entity.ExtractData;
import org.codelibs.fess.crawler.exception.ExtractException;

import com.google.common.base.Charsets;

import jakarta.annotation.PostConstruct;
import jakarta.annotation.PreDestroy;

/**
 * Extract a text by using external http server.
 */
public class ApiExtractor extends AbstractExtractor {

     * The thread group for crawler threads.
     */
    protected ThreadGroup crawlerThreadGroup;

    /**
     * Constructs a new Crawler instance.
     * Initializes the crawler context with a new session ID based on the current timestamp.
     */
    public Crawler() {
        crawlerContext = new CrawlerContext();
        final SimpleDateFormat sdf = new SimpleDateFormat("yyyyMMddHHmmssSSS", Locale.ENGLISH);

 * <li>{@link #deleteElevateWord(String, boolean)} - Delete an elevate word and optionally apply it.
 * <li>{@link #restoreElevateWord()} - Restore elevate words.
 * <li>{@link #deleteOldWords(ZonedDateTime)} - Delete old words based on a threshold date.
 * <li>{@link #setIndexName(String)} - Set the index name.
 * <li>{@link #setSupportedFields(String[])} - Set the supported fields.
 * <li>{@link #setTagFieldNames(String[])} - Set the tag field names.

      copyright notice that is included in or attached to the work
      (an example is provided in the Appendix below).

      "Derivative Works" shall mean any work, whether in Source or Object
      form, that is based on (or derived from) the Work and for which the
      editorial revisions, annotations, elaborations, or other modifications
      represent, as a whole, an original work of authorship. For the purposes

        }

        @Override
        public ResponseProcessor getResponseProcessor() {
            return responseProcessor;
        }
    }

    /**
     * Conditional test rule that matches based on URL pattern
     */
    public static class UrlPatternRule implements Rule {
        private static final long serialVersionUID = 1L;
        private final String ruleId;
        private final String urlPattern;

				"application/vnd.dvb.esgcontainer",
				"application/vnd.dvb.ipdcdftnotifaccess",
				"application/vnd.dvb.ipdcesgaccess",
				"application/vnd.dvb.ipdcroaming",
				"application/vnd.dvb.iptv.alfec-base",
				"application/vnd.dvb.iptv.alfec-enhancement",
				"application/vnd.dvb.notif-aggregate-root+xml",
				"application/vnd.dvb.notif-container+xml",
				"application/vnd.dvb.notif-generic+xml",