使用 STAX API 进行 XML 解析

Question

我正在尝试使用 Stax 解析开放街道地图的 XML 结构。在我的实现中，我使用 XMLStreamConstants.START_ELEMENT 和 XMLStreamConstants.END_ELEMENT 识别元素。

OpenStreetMaps 结构具有 tag 等元素，这些元素描述了 node 以及 way。下面是一个结构示例：

      <node id="2311741639" ... lat="50.7756648" lon="6.0844948">
       <tag k="entrance" v="yes"/>
      </node>
      <way id="4964449" visible="true" ... uid="67862">
       <nd ref="27290865"/>
        ...
       <tag k="highway" v="residential"/>
        ...
      </way>

如果解析器读取Tag-Element，如何区分节点和方式？

Answer 1

您可以使用节点表示的ArrayDeque，如果层次结构的深度很小，甚至可以构建一个临时的类似 DOM 的结构。

这是ArrayDeque的示例...

假设这个 XML 文件名为 stuff.xml：

<?xml version="1.0" encoding="UTF-8"?>

<stuff>

<node id="2311741639" lat="50.7756648" lon="6.0844948">
    <tag k="entrance" v="yes"/>
</node>

<way id="4964449" visible="true" uid="67862">
    <nd ref="27290865"/>
    <tag k="highway" v="residential"/>
</way>

</stuff>

假设文件在路径上：/my/path/

这里是代码（try/catch Java 6 风格）：

InputStream is = null;
XMLStreamReader reader = null;
try {
    is = new FileInputStream(new File("/my/path/stuff.xml"));
    XMLInputFactory xif = XMLInputFactory.newInstance();
    reader = xif.createXMLStreamReader(is);
    ArrayDeque<String> nodes = new ArrayDeque<String>();
    while (reader.hasNext()) {
        int current = reader.next();
        switch (current) {
            case XMLStreamConstants.START_ELEMENT: {
                nodes.add(reader.getLocalName());
                System.out.println("START: " + nodes.getLast());
                if (nodes.size() > 1) {
                    Iterator<String> iterator = nodes.descendingIterator();
                    // skipping first one as it's already represented
                    iterator.next();
                    while (iterator.hasNext()) {
                        System.out.println("\t in " + iterator.next());
                    }
                }
                break;
            }
            case XMLStreamConstants.END_ELEMENT: {
                System.out.println("END: " + nodes.removeLast());
                Iterator<String> iterator = nodes.descendingIterator();
                while (iterator.hasNext()) {
                    System.out.println("\t in " + iterator.next());
                }
                break;
            }
        }
    }

}
        catch (FileNotFoundException fnfe) {
            fnfe.printStackTrace();
        }
        catch (XMLStreamException xse) {
            xse.printStackTrace();
        }
        finally {
            if (reader != null) {
                try {
                    reader.close();
                    is.close();
                }
                catch (XMLStreamException xse) {
                    xse.printStackTrace();
                }
                catch (IOException ioe) {
                    ioe.printStackTrace();
                }
            }
        }

输出：

START: stuff
START: node
     in stuff
START: tag
     in node
     in stuff
END: tag
     in node
     in stuff
END: node
     in stuff
START: way
     in stuff
START: nd
     in way
     in stuff
END: nd
     in way
     in stuff
START: tag
     in way
     in stuff
END: tag
     in way
     in stuff
END: way
     in stuff
END: stuff