使用 Xerces 的 getElementsByTagName 的 Java 代码不返回子节点

Question

我正在使用 Xerces 解析器来尝试解析以下 XML sn-p。当我使用 findElementsByTagName 方法查找 Circle 时，我可以找到“位置信息”元素，我得到一个空的 NodeList。有人可以抽查一下，看看我做错了什么吗？

<urn:locationResponse xmlns:urn="urn:ietf:params:xml:ns:geopriv:held">
<presence entity="pres:www.telecomsys.com" xmlns="urn:ietf:params:xml:ns:pidf"
          xmlns:gp="urn:ietf:params:xml:ns:pidf:geopriv10" xmlns:dm="urn:ietf:params:xml:ns:pidf:data-model"
          xmlns:gs="http://www.opengis.net/pidflo/1.0" xmlns:ca="urn:ietf:params:xml:ns:pidf:geopriv10:civicAddr"
          xmlns:gml="http://www.opengis.net/gml">
    <tuple id="FIRST_LOCATION">
        <status>
            <gp:geopriv>
                <location-info xmlns="urn:ietf:params:xml:ns:pidf:geopriv10">
                    <gs:Circle srsName="urn:ogc:def:crs:EPSG::4326">
                        <gml:pos>00.000000 -00.00000</gml:pos>
                        <gs:radius uom="urn:ogc:def:uom:EPSG::9001">200</gs:radius>
                    </gs:Circle>
                    <gp:confidence>95</gp:confidence>
                </location-info>
                <gp:usage-rules>
                    <gp:retransmission-allowed>yes</gp:retransmission-allowed>
                </gp:usage-rules>
                <gp:method>Derived</gp:method>
            </gp:geopriv>
        </status>
        <timestamp>2001-01-00T00:00Z</timestamp>
    </tuple>
</presence>

以下是我的代码，它试图从这个 XML 中获取“Circle”标签

    private static final Log logger = LogFactory.getLog(PIDFLOParser.class);
    private static final String LOCATION_INFO = "location-info";
    private static final String CIRCLE = "Circle";

    // Use of the Document BuilderFactory to create a DocumentBuilder class.
    DocumentBuilderFactory fact = DocumentBuilderFactory.newInstance();
    DocumentBuilder builder = fact.newDocumentBuilder();

    Document doc = builder.parse(new InputSource(new StringReader(xmlDoc)));
    doc.getDocumentElement().normalize();

    Node node = doc.getDocumentElement();

    String root = node.getNodeName();
    System.out.println("Root Node: " + root);

    NodeList listResponse = doc.getElementsByTagName(LOCATION_INFO);
    if (listResponse.getLength() == 0) {
        System.out.println(String.format("%s doesn't exist in the Document.", LOCATION_INFO));
    }

    Node firstNode = listResponse.item(0);
    listResponse = ((Element) firstNode).getElementsByTagName(CIRCLE);
    if (listResponse.getLength() == 0) {
        System.out.println(String.format("%s doesn't exist in the Document.", CIRCLE));
    }

    listResponse = ((Element) firstNode).getElementsByTagNameNS("gs", CIRCLE);
    if (listResponse.getLength() == 0) {
        System.out.println(String.format("%s doesn't exist in the Document when searching with namespace.", CIRCLE));
    }

这段代码的输出是：

    Root Node: urn:locationResponse
Circle doesn't exist in the Document.
Circle doesn't exist in the Document when searching with namespace.

我做错了什么？提前感谢您的帮助！

在 guido 关于命名空间的完整 URI 的评论后更新

...
    private static final String NS_GS = "http://www.opengis.net/pidflo/1.0";
...
    listResponse = ((Element) firstNode).getElementsByTagNameNS(NS_GS, CIRCLE);
    if (listResponse.getLength() == 0) {
        System.out.println(String.format("%s doesn't exist in the Document when searching with namespace.", CIRCLE));
    }

输出还是一样的：

Root Node: urn:locationResponse
Circle doesn't exist in the Document.
Circle doesn't exist in the Document when searching with namespace.

Answer 1

当您调用getElementsByTagNameNS 时，您应该指定命名空间的 URI，而不是 xml 中使用的前缀，因此：

getElementsByTagNameNS("gs", CIRCLE);

应该是：

getElementsByTagNameNS("http://www.opengis.net/pidflo/1.0", CIRCLE);

因为gs:Circle元素是在命名空间URI下定义的：

xmlns:gs="http://www.opengis.net/pidflo/1.0"

要使命名空间工作，您需要为其设置工厂：

 DocumentBuilderFactory fact = DocumentBuilderFactory.newInstance();
 fact.setNamespaceAware(true);

或者，您可以只使用（不带命名空间）完整的限定名称：

 getElementsByTagName("gs:Circle");

注意：另请注意，您的 xml 在您的问题中无效，因为它缺少结束根元素 </urn:locationResponse>