0

无论如何使用Pentaho从 html 页面解析表 td 吗？假设我有这个 html 内容

<html>
  <body>
    <table>
      <tr>
        <td>info1</td>
        <td>info2</td>
      </tr>
      <tr>
        <td>info3</td>
        <td>info4</td>
      </tr>
    </table>
  </body>
</html>
我在 Pentaho 中使用带有以下数据的“从 XML 获取数据”：
Content::
Loop XPath: /html/body/table/tr
Fields::
Name: tableData
XPath: td
 我想要的数据信息是 
info1 info2 info3 info4
以任何方式。
任何帮助将不胜感激！

4

2 回答 2

0

我通过将文件的每一行读取为行来解决它。然后我添加了一个 Pentaho 步骤“用户定义的 Java 类”，并使用 XSLT 将我的表格内容解析为一个新的 XML 文件。使用该 XML，我能够获得完成任务所需的数据。
这是我在“用户定义的 Java 类”中写的：


import java.util.*;
import java.io.FileOutputStream;

import javax.xml.transform.Transformer;
import javax.xml.transform.TransformerFactory;

private int infilenameIndex;
private int xsltfilenameIndex;
private int outfilenameIndex;


public boolean processRow(StepMetaInterface smi, StepDataInterface sdi) throws KettleException {
  Object[] r=getRow();
  if (r==null) {
    setOutputDone();
    return false;
  }


  if (first == false) {
     infilenameIndex = getInputRowMeta().indexOfValue(getParameter("infilename"));
     if (infilenameIndex < 0) {
         throw new KettleException("Field not found in the input row, check parameter 'infilename'!");
     }
     xsltfilenameIndex = getInputRowMeta().indexOfValue(getParameter("xsltfilename"));
     if (xsltfilenameIndex < 0) {
         throw new KettleException("Field not found in the input row, check parameter 'xsltfilename'!");
     }
     outfilenameIndex = getInputRowMeta().indexOfValue(getParameter("outfilename"));
     if (outfilenameIndex < 0) {
         throw new KettleException("Field not found in the input row, check parameter 'outfilename'!");
     }

     first=false;
  }

  String infilename = get(Fields.In, "infilename").getString(r);
  String xsltfilename = get(Fields.In, "xsltfilename").getString(r);
  String outfilename = get(Fields.In, "outfilename").getString(r);

  Object[] outputRowData = RowDataUtil.resizeArray(r, data.outputRowMeta.size());
  int outputIndex = getInputRowMeta().size();

  transform(infilename, xsltfilename, outfilename);


  putRow(data.outputRowMeta, outputRowData);

  return true;
}
public void transform(String infilename, String xsltfilename, String outfilename) throws KettleException {

    javax.xml.transform.stream.StreamSource inss = null;
    javax.xml.transform.stream.StreamSource xsltss = null;
    javax.xml.transform.stream.StreamResult outss = null;

    logBasic("");
    logBasic("Transformerar " +  infilename + " med " + xsltfilename + " till " + outfilename );
    logBasic("");

    try {
       inss = new javax.xml.transform.stream.StreamSource(infilename);
    }     
    catch (Exception e) {
       logError("Infil saknas " +  infilename);
       throw new KettleException(e);
    }

    try {
       xsltss = new javax.xml.transform.stream.StreamSource(xsltfilename);
    }     
    catch (Exception e) {
       logError("Xsltfil saknas " +  xsltfilename);
       throw new KettleException(e);
    }

    try {
       outss = new javax.xml.transform.stream.StreamResult(outfilename);
    }     
    catch (Exception e) {
       logError("Outfil saknas " +  outfilename);
       throw new KettleException(e);
    }

    try {       
        TransformerFactory tFactory = TransformerFactory.newInstance();

        // Set the TransformerFactory to the SAXON implementation.
        //tFactory = new net.sf.saxon.TransformerFactoryImpl();

        Transformer transformer = tFactory.newTransformer(xsltss);

        // Do the transfromtation
        transformer.transform(inss, outss);
    }
    catch (Exception e) {
       throw new KettleException(e);
    }
    return;
}
于 2011-04-18T13:06:49.807 回答
-1

看到了这个。对于现在来这里的任何人。可以使用 jsoup 使用适当的路径解析为 xml。它是一个简单的插件，可以在用户定义的类中使用您拥有的任何其他方法。它是一个 CSS 选择器。

于 2013-05-23T23:32:56.463 回答