* 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.InputStream;
import java.util.HashMap;
import java.util.Map;

import org.apache.logging.log4j.LogManager;
import org.apache.logging.log4j.Logger;
import org.codelibs.core.io.CloseableUtil;

        final InputStream in = ResourceUtil.getResourceAsStream("extractor/test_sjis.xml");
        final String content = xmlExtractor.getText(in, null).getContent();
        CloseableUtil.closeQuietly(in);
        logger.info(content);
        assertTrue(content.contains("テスト"));
    }

    public void test_getXml_entity() {
        final InputStream in = ResourceUtil.getResourceAsStream("extractor/test_entity.xml");

import org.codelibs.fess.crawler.extractor.Extractor;
import org.codelibs.fess.crawler.extractor.ExtractorFactory;

import jakarta.annotation.Resource;

/**
 * TextTransformer is a class that transforms a ResponseData object into a ResultData object containing the extracted text content.
 * It uses an Extractor to extract the text from the response body based on the MIME type.

* Fix broken link in docs about OAuth 2.0 with scopes. PR [#275](https://github.com/tiangolo/fastapi/pull/275) by [@dmontagu](https://github.com/dmontagu).

* Refactor param extraction using Pydantic `Field`:
    * Large refactor, improvement, and simplification of param extraction from *path operations*.

 * 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.IOException;
import java.io.InputStream;

import org.apache.logging.log4j.LogManager;
import org.apache.logging.log4j.Logger;
import org.codelibs.core.io.CloseableUtil;

import java.io.File;
import java.util.HashSet;
import java.util.Set;

class ClassMetaDataUtil {
    static void extractFromMetadata(File metaData, Set<String> excludedPackagePatterns, Action<ClassMetaData> extractor) {
        SimpleClassMetaDataRepository<ClassMetaData> repository = new SimpleClassMetaDataRepository<>();
        repository.load(metaData);

        final Set<String> excludedPrefixes = new HashSet<>();

        validateInputStream(in);
        try (final HSSFWorkbook workbook = new HSSFWorkbook(in);
                final org.apache.poi.hssf.extractor.ExcelExtractor excelExtractor =
                        new org.apache.poi.hssf.extractor.ExcelExtractor(workbook)) {
            return new ExtractData(excelExtractor.getText());
        } catch (final IOException e) {

     */
    @Override
    public ExtractData getText(final InputStream in, final Map<String, String> params) {
        validateInputStream(in);
        try (final org.apache.poi.hwpf.extractor.WordExtractor wordExtractor = new org.apache.poi.hwpf.extractor.WordExtractor(in)) {
            return new ExtractData(wordExtractor.getText());
        } catch (final IOException e) {

        final Extractor extractor = new Extractor() {
            public ExtractData getText(final InputStream in, final Map<String, String> params) {
                return null;
            }
        };

        assertNull(extractorFactory.getExtractor("test"));
        extractorFactory.addExtractor("test", extractor);
        assertEquals(extractor, extractorFactory.getExtractor("test"));
    }

 * 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.InputStream;

import org.apache.logging.log4j.LogManager;
import org.apache.logging.log4j.Logger;
import org.codelibs.core.io.CloseableUtil;
import org.codelibs.core.io.ResourceUtil;