responseData.setCharSet(Constants.UTF_8);
        final ResultData resultData = xmlTransformer.transform(responseData);
        assertEquals(result, new String(resultData.getData(), Constants.UTF_8));
    }

    public void test_transformNs() throws Exception {
        final String result = "<?xml version=\"1.0\"?>\n"//
                + "<doc>\n"//

     * This method searches for 'fess_ds++.xml' configuration files within JAR files
     * in the data store plugin directory and extracts component class names.
     *
     * <p>The method uses secure XML parsing features to prevent XXE attacks and
     * other XML-based vulnerabilities. Component class names are extracted from
     * the 'class' attribute of 'component' elements in the XML files.</p>
     *

      override fun contentType(): MediaType = "application/xml".toMediaType()

      override fun writeTo(sink: BufferedSink) {
        sink.writeUtf8("<el")
        sink.flush()
        throw IOException("failed to stream the XML")
      }
    }

    class ValidRequestBody : RequestBody() {
      override fun contentType(): MediaType = "application/xml".toMediaType()

      override fun writeTo(sink: BufferedSink) {

crawler.addUrl("file:///path/to/directory");
crawler.urlFilter.addInclude("file:///path/to/directory/.*");
```

## Configuration

### XML Configuration

Fess Crawler uses XML-based configuration with LastaFlute DI. Place configuration files in your classpath:

```xml
<!-- crawler.xml -->
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE components PUBLIC "-//DBFLUTE//DTD LastaDi 1.0//EN"
    "http://dbflute.org/meta/lastadi10.dtd">

  /**
   * <a href="https://en.wikipedia.org/wiki/RDF/XML">RDF/XML</a> documents, which are XML
   * serializations of <a
   * href="https://en.wikipedia.org/wiki/Resource_Description_Framework">Resource Description
   * Framework</a> graphs.
   *
   * @since 14.0
   */
  public static final MediaType RDF_XML_UTF_8 = createConstantUtf8(APPLICATION_TYPE, "rdf+xml");

 * (such as an XML document). Typically (but not always), the inverse process of "unescaping" the
 * text is performed automatically by the relevant parser.
 *
 * <p>For example, an XML escaper would convert the literal string {@code "Foo<Bar>"} into {@code
 * "Foo&lt;Bar&gt;"} to prevent {@code "<Bar>"} from being confused with an XML tag. When the

/target/
/work/
/bin/
/.settings/
.project
.classpath
*.iml
.idea
.DS_Store
dependency-reduced-pom.xml
/.serena

    @Override
    public ResponseData execute(RequestData request) { /* ... */ }

    @Override
    public void close() { /* cleanup */ }
}
```

2. **Register in DI config** (`crawler.xml`):
```xml
<component name="myClient" class="...MyClient" instance="singleton"/>
```

3. **Add to factory**:
```java
clientFactory.addClient("^myprotocol://.*", myClient);
```

      with:
        java-version: '21'
        distribution: 'temurin'
    - uses: actions/cache@v4
      with:
        path: ~/.m2/repository
        key: ${{ runner.os }}-maven-${{ hashFiles('**/pom.xml') }}
        restore-keys: |
          ${{ runner.os }}-maven-
    - name: Checkout fess-parent
      uses: actions/checkout@v4
      with:
        repository: codelibs/fess-parent
        ref: ${{ env.PARENT_BRANCH }}

# Auto detect text files and perform LF normalization
* text=auto

# Known text files
*.java text
*.xml text
*.yml text
*.md text
*.sh text
*.css text
*.txt text

# Known binary files