1.0 UTF-8 http://www.w3.org/1999/02/22-rdf-syntax-ns# http://www.recshop.fake/cd# http://www.recshop.fake/cd/Empire Burlesque Shinsuke Sugaya Japan テスト 100 2010 コメント...

CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. en http://usefulinc.com/ns/doap# http://www.w3.org/1999/02/22-rdf-syntax-ns# http://projects.apache.org/ns/asfext# http://xmlns.com/foaf/0.1/ http://Maven.rdf.apache.org/ 2001 http://www.apache.org/licenses/LICENSE-2.0.txt Apache Maven https://maven.apache.org/ https://maven.apache.org/ Maven is a project development management and comprehension tool. Maven is...

			<arg>"http[s]?:.*"</arg>
		</postConstruct>
		<postConstruct name="addRule">
			<arg>"mimeType"</arg>
			<!-- Supported MIME type -->
			<arg>
  "(application/xml"
+ "|application/xhtml+xml"
+ "|application/rdf+xml"
+ "|application/pdf"
+ "|application/x-freemind"
+ "|text/xml"
+ "|text/xml-external-parsed-entity)"
			</arg>
		</postConstruct>
	</component>

<!DOCTYPE [<!-- a comment --><!ENTITY rdf "http://www.w3.org/1999/02/22-rdf-syntax-ns#">]>
<!DOCTYPE [<!ENTITY go "Golang"><!-- a comment-->]>
<!DOCTYPE <!-> <!> <!----> <!-->--> <!--->--> [<!ENTITY go "Golang"><!-- a comment-->]>
`

var directivesWithCommentsTokens = []Token{
	CharData("\n"),
	Directive(`DOCTYPE [ <!ENTITY rdf "http://www.w3.org/1999/02/22-rdf-syntax-ns#">]`),
	CharData("\n"),

        CloseableUtil.closeQuietly(bis);
        assertEquals("UTF-8", encoding);
    }

    public void test_getRdf() {
        final InputStream in = ResourceUtil.getResourceAsStream("extractor/test.rdf");
        final String content = xmlExtractor.getText(in, null).getContent();
        CloseableUtil.closeQuietly(in);
        logger.info(content);
        assertTrue(content.contains("テスト"));

  /**
   * <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");

  /**
   * <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");

  </mime-type>

  <mime-type type="application/rdf+xml">
    <root-XML localName="RDF"/>
    <root-XML localName="RDF" namespaceURI="http://www.w3.org/1999/02/22-rdf-syntax-ns#"/>
    <sub-class-of type="application/xml"/>
    <acronym>RDF/XML</acronym>
    <_comment>XML syntax for RDF graphs</_comment>

    <glob pattern="*.rdf"/>
    <glob pattern="*.owl"/>
    <glob pattern="^rdf$" isregex="true"/>

	<component name="extractorFactory"
		class="org.codelibs.fess.crawler.extractor.ExtractorFactory">
		<postConstruct name="addExtractor">
			<arg>[
				"application/xml",
				"application/xhtml+xml",
				"application/rdf+xml",
				"application/x-freemind",
				"text/xml"
				]</arg>
			<arg>xmlExtractor</arg>
		</postConstruct>
		<postConstruct name="addExtractor">
			<arg>[
				"text/html"
				]</arg>