throw new FessSystemException("Could not find extractorFactory.");
        }
        final Extractor extractor = extractorFactory.getExtractor(responseData.getMimeType());
        if (logger.isDebugEnabled()) {
            logger.debug("url={}, extractor={}", responseData.getUrl(), extractor);
        }
        return extractor;
    }

    }

    /**
     * Gets the appropriate extractor for the given response data.
     * Selects an extractor based on the MIME type or falls back to the Tika extractor.
     *
     * @param responseData the response data containing the document to extract
     * @return the extractor instance for processing the document
     * @throws FessSystemException if no suitable extractor can be found
     */
    @Override

 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND,
 * either express or implied. See the License for the specific language
 * governing permissions and limitations under the License.
 */
package org.codelibs.fess.crawler.extractor.impl;

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

import org.codelibs.core.io.ResourceUtil;
import org.codelibs.fess.crawler.container.StandardCrawlerContainer;

        csvExtractor.setMaxRows(2);
        final InputStream in = ResourceUtil.getResourceAsStream("extractor/csv/test.csv");
        final ExtractData extractData = csvExtractor.getText(in, null);
        CloseableUtil.closeQuietly(in);

        final String rowCount = extractData.getValues("row_count")[0];
        // Should extract 2 data rows (header doesn't count toward maxRows)
        assertEquals("2", rowCount);
    }

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

import java.util.regex.Pattern;

/**
 * Extracts text content from XML documents.
 */
public class XmlExtractor extends AbstractXmlExtractor {

    /**
     * Creates a new XmlExtractor instance.
     */

```

4. **Add tests**: Unit + integration

### Adding a Content Extractor

1. **Implement `Extractor`**:
```java
public class MyExtractor extends AbstractExtractor {
    @Override
    public ExtractData getText(InputStream in, Map<String, String> params) {
        ExtractData data = new ExtractData();
        // Extract text
        data.setContent(extractedText);
        return data;
    }
}

    /**
     * Constructs a new CommandExtractor.
     */
    public CommandExtractor() {
        // NOP
    }

    /*
     * (non-Javadoc)
     *
     * @see org.codelibs.fess.crawler.extractor.Extractor#getText(java.io.InputStream,
     * java.util.Map)
     */
    @Override
    public ExtractData getText(final InputStream in, final Map<String, String> params) {

        this.relatedQueryMap = relatedQueryMap;
        return relatedQueryMap.size();
    }

    /**
     * Extracts the virtual host key from a RelatedQuery entity.
     * If the virtual host is blank or null, returns an empty string.
     *
     * @param entity the RelatedQuery entity to extract the host key from
     * @return the virtual host key, or empty string if blank or null
     */

    /** The key of the message: Crawling/Displaying */
    public static final String LABELS_pathmap_pt_both = "{labels.pathmap_pt_both}";

    /** The key of the message: Extracted URL Conversion */
    public static final String LABELS_pathmap_pt_stored = "{labels.pathmap_pt_stored}";

    /** The key of the message: Regular Name */

        assertContentType("application/pdf", "extractor/test.pdf", "hoge.pdf");

        assertContentType("application/gzip", "extractor/gz/test.tar.gz", "hoge.tar.gz");
        assertContentType("application/zip", "extractor/zip/test.zip", "hoge.zip");
        assertContentType("application/x-lharc", "extractor/lha/test.lzh", "hoge.lzh"); // TODO is it correct?

        assertContentType("application/xml", "extractor/test.mm", "hoge.mm");