我在使用javax.swing.text.ElementIterator()时遇到了奇怪的行为。它永远不会显示所有元素,并且会显示不同数量的元素,具体取决于我使用哪种类型的ParserCallback。以下测试是使用我个人资料中的网站完成的,但可以使用任何其他较大的html文件完成。

// some imports shown in case its an import mixup
import javax.swing.text.AttributeSet;
import javax.swing.text.BadLocationException;
import javax.swing.text.ChangedCharSetException;
import javax.swing.text.Element;
import javax.swing.text.ElementIterator;
import javax.swing.text.MutableAttributeSet;
import javax.swing.text.StyleConstants;
import javax.swing.text.html.HTML;
import javax.swing.text.html.HTML.Tag;
import javax.swing.text.html.HTMLDocument;
import javax.swing.text.html.HTMLEditorKit;
import javax.swing.text.html.HTMLEditorKit.Parser;
import javax.swing.text.html.HTMLEditorKit.ParserCallback;
import javax.swing.text.html.parser.ParserDelegator;

// Shows whats in an element, recursively
public void printElement(HTMLDocument htmlDoc, Element element)
        throws BadLocationException
{
    AttributeSet attributes = element.getAttributes();
    System.out.println("element: '" + element.toString().trim() + "', name: '" + element.getName() + "', children: " + element.getElementCount() + ", attributes: " + attributes.getAttributeCount() + ", leaf: " + element.isLeaf());
    Enumeration attrEnum = attributes.getAttributeNames();
    while (attrEnum.hasMoreElements())
    {
        Object attr = attrEnum.nextElement();
        System.out.println("\tAttribute: '" + attr + "', Val: '" + attributes.getAttribute(attr) + "'");
        if (attr == StyleConstants.NameAttribute
                && attributes.getAttribute(StyleConstants.NameAttribute) == HTML.Tag.CONTENT)
        {
            int startOffset = element.getStartOffset();
            int endOffset = element.getEndOffset();
            int length = endOffset - startOffset;
            System.out.printf("\t\tContent (%d-%d): '%s'\n", startOffset, endOffset, htmlDoc.getText(startOffset, length).trim());
        }
    }
    for (int i = 0; i < element.getElementCount(); i++)
    {
        Element child = element.getElement(i);
        printElement(htmlDoc, child);
    }
}

public void tryParse(String filename)
        throws FileNotFoundException, IOException, BadLocationException
{
    BufferedReader in = new BufferedReader(new InputStreamReader(new FileInputStream(filename)));

    Parser parser = new ParserDelegator();
    HTMLEditorKit htmlKit = new HTMLEditorKit();
    HTMLDocument htmlDoc = (HTMLDocument) htmlKit.createDefaultDocument();
    ParserCallback callback2 = htmlDoc.getReader(0);
    ParserCallback callback1 =
            new HTMLEditorKit.ParserCallback()
            {
            };

    parser.parse(in, callback2, true);
    ElementIterator iterator = new ElementIterator(htmlDoc);
    Element element;
    while ((element = iterator.next()) != null)
        printElement(htmlDoc, element);
    in.close();
}


在上面的测试中,如果我使用callback1或callback2,结果会有所不同。即使很奇怪,如果我确实用适当的函数填充回调并让它们输出某些内容,它们也表明解析器确实可以处理整个网站,但是ElementIterator仍然不具备全部功能。

我也尝试使用htmlKit.read()代替parser.parse(),但是它仍然无法正常工作。

尽管我现在通过使用解析器回调函数(此处未显示)获得期望的结果,但我仍然想知道为什么ElementIterator不能按预期工作,以防我以后需要它,所以我想知道这里是否有人对那个ElementIterator有经验并可以回答。

更新:完整的Java源代码上传到这里:
http://home.snafu.de/tilman/tmp/Main.java

最佳答案

使用here中看到的方法,我没有注意到您描述的问题。我添加了println(),所有元素似乎都在那里。

附录:我不确定您的tryParse()如何失败,但是您的printElement()似乎可以在我的main()上正常工作:

import java.io.BufferedReader;
import java.io.FileReader;
import java.util.Enumeration;
import javax.swing.text.AttributeSet;
import javax.swing.text.BadLocationException;
import javax.swing.text.Element;
import javax.swing.text.ElementIterator;
import javax.swing.text.StyleConstants;
import javax.swing.text.html.HTML;
import javax.swing.text.html.HTMLDocument;
import javax.swing.text.html.HTMLEditorKit;

/** @see https://stackoverflow.com/questions/2882782 */
public class NewMain {

    public static void main(String args[]) throws Exception {
        HTMLEditorKit htmlKit = new HTMLEditorKit();
        HTMLDocument htmlDoc = (HTMLDocument) htmlKit.createDefaultDocument();
        htmlKit.read(new BufferedReader(new FileReader("test.html")), htmlDoc, 0);
        ElementIterator iterator = new ElementIterator(htmlDoc);
        Element element;
        while ((element = iterator.next()) != null) {
            printElement(htmlDoc, element);
        }
    }
    private static void printElement(HTMLDocument htmlDoc, Element element)
        throws BadLocationException {
        AttributeSet attrSet = element.getAttributes();
        System.out.println(""
            + "Element: '" + element.toString().trim()
            + "', name: '" + element.getName()
            + "', children: " + element.getElementCount()
            + ", attributes: " + attrSet.getAttributeCount()
            + ", leaf: " + element.isLeaf());
        Enumeration attrNames = attrSet.getAttributeNames();
        while (attrNames.hasMoreElements()) {
            Object attr = attrNames.nextElement();
            System.out.println("  Attribute: '" + attr + "', Value: '"
                + attrSet.getAttribute(attr) + "'");
            Object tag = attrSet.getAttribute(StyleConstants.NameAttribute);
            if (attr == StyleConstants.NameAttribute
                && tag == HTML.Tag.CONTENT) {
                int startOffset = element.getStartOffset();
                int endOffset = element.getEndOffset();
                int length = endOffset - startOffset;
                System.out.printf("    Content (%d-%d): '%s'\n", startOffset,
                    endOffset, htmlDoc.getText(startOffset, length).trim());
            }
        }
    }
}

07-26 03:08