*/
package org.codelibs.fess.crawler.transformer.impl;

import org.codelibs.fess.crawler.Constants;
import org.codelibs.fess.crawler.container.StandardCrawlerContainer;
import org.codelibs.fess.crawler.entity.AccessResultDataImpl;
import org.codelibs.fess.crawler.entity.ResponseData;
import org.codelibs.fess.crawler.entity.ResultData;
import org.codelibs.fess.crawler.exception.CrawlerSystemException;

import org.codelibs.fess.crawler.entity.Sitemap;
import org.codelibs.fess.crawler.entity.SitemapSet;
import org.codelibs.fess.crawler.exception.ChildUrlsException;
import org.codelibs.fess.crawler.helper.SitemapsHelper;
import org.codelibs.fess.crawler.processor.ResponseProcessor;

import jakarta.annotation.Resource;

/**
 * @author shinsuke
 *
 */
public class SitemapsResponseProcessor implements ResponseProcessor {

<!DOCTYPE components PUBLIC "-//DBFLUTE//DTD LastaDi 1.0//EN"
	"http://dbflute.org/meta/lastadi10.dtd">
<components namespace="fessCrawler">
	<include path="crawler/transformer_basic.xml"/>

	<component name="fessXpathTransformer" class="org.codelibs.fess.crawler.transformer.FessXpathTransformer" instance="singleton">
		<property name="name">"fessXpathTransformer"</property>
		<property name="featureMap">defaultFeatureMap</property>

import org.codelibs.core.lang.StringUtil;
import org.codelibs.fess.Constants;
import org.codelibs.fess.app.service.FileAuthenticationService;
import org.codelibs.fess.crawler.client.CrawlerClientFactory;
import org.codelibs.fess.crawler.client.ftp.FtpAuthentication;
import org.codelibs.fess.crawler.client.smb.SmbAuthentication;
import org.codelibs.fess.es.config.bsentity.BsFileConfig;
import org.codelibs.fess.helper.SystemHelper;

    /** The key of the configuration. e.g. false */
    String CRAWLER_DOCUMENT_APPEND_FILENAME = "crawler.document.append.filename";

    /** The key of the configuration. e.g. 20 */
    String CRAWLER_DOCUMENT_MAX_ALPHANUM_TERM_SIZE = "crawler.document.max.alphanum.term.size";

    /** The key of the configuration. e.g. 10 */

 * governing permissions and limitations under the License.
 */
package org.codelibs.fess.crawler.service.impl;

import java.util.List;
import java.util.regex.Pattern;

import org.codelibs.fess.crawler.helper.MemoryDataHelper;
import org.codelibs.fess.crawler.service.UrlFilterService;

import jakarta.annotation.Resource;

/**
 * @author shinsuke
 *
 */

crawler.ignore.robots.tags=false
crawler.ignore.content.exception=true
crawler.failure.url.status.codes=404
crawler.system.monitor.interval=60
crawler.hotthread.ignore_idle_threads=true
crawler.hotthread.interval=500ms
crawler.hotthread.snapshots=10
crawler.hotthread.threads=3
crawler.hotthread.timeout=30s
crawler.hotthread.type=cpu
crawler.metadata.content.excludes=resourceName,X-Parsed-By,Content-Encoding.*,Content-Type.*,X-TIKA.*,X-FESS.*

 * either express or implied. See the License for the specific language
 * governing permissions and limitations under the License.
 */
package org.codelibs.fess.crawler.client;

import java.util.Map;

import org.codelibs.fess.crawler.entity.RequestData;
import org.codelibs.fess.crawler.entity.ResponseData;

/**
 * @author shinsuke
 *
 */
public interface CrawlerClient extends AutoCloseable {

 */
package org.codelibs.fess.crawler.extractor.impl;

import java.io.IOException;
import java.io.InputStream;

import org.codelibs.core.io.ResourceUtil;
import org.codelibs.fess.crawler.container.StandardCrawlerContainer;
import org.codelibs.fess.crawler.entity.ExtractData;
import org.codelibs.fess.crawler.exception.CrawlerSystemException;
import org.codelibs.fess.crawler.extractor.ExtractorFactory;

 */
package org.codelibs.fess.crawler.helper.impl;

import java.util.Set;

import org.codelibs.fess.crawler.entity.RequestData;
import org.codelibs.fess.crawler.entity.ResponseData;
import org.codelibs.fess.crawler.entity.UrlQueue;
import org.codelibs.fess.crawler.exception.CrawlingAccessException;
import org.codelibs.fess.crawler.helper.LogHelper;
import org.codelibs.fess.crawler.log.LogType;
import org.slf4j.Logger;