用于多个文件的Groovy脚本xml解析器

Question

Hi my groovy script strips out an xml tag from a file and writes to a file.

 import org.apache.commons.lang.RandomStringUtils
 import groovy.util.XmlSlurper



 inputFile = 'C:\\sample.xml'
 outputFile = 'C:\\ouput.txt'


 XMLTag='Details'

 fileContents = new File(inputFile).getText('UTF-8')

 def xmlFile=new XmlSlurper().parseText(fileContents)

 def myPayload= new String(xmlFile.'**'.find{node-> node.name() ==    XMLTag}   *.text().toString())


 file = new File(outputFile)
 w = file.newWriter() 
 w << myPayload.substring(1, myPayload.length()-1)
 w.close()

My question is how do I write it so the it goes through an entire directory and performs it on multiple xml files and creates multiple output as at the moment it is hard coded. ('C:\\sample.xml' and 'C:\\ouput.txt')

Thanks

Leon

Answer 1

First, I would recommend that you take what you have and put it into a single function; it's good coding practrice and improves readabililty.

Now to executing the function on each xml file in a directory, you can use groovy's File.eachFileMatch() . For example, if you want to run it on each xml file in the current directory, you could do:

import org.apache.commons.lang.RandomStringUtils
import groovy.util.XmlSlurper
import static groovy.io.FileType.*

void stripTag(File inputFile, String outputFile) {
    def XMLTag='Details'

    fileContents = inputFile.getText('UTF-8')

    def xmlFile=new XmlSlurper().parseText(fileContents)

    def myPayload= new String(xmlFile.'**'.find{node-> node.name() == XMLTag}*.text().toString())


    def file = new File(outputFile)
    w = file.newWriter() 
    w << myPayload.substring(1, myPayload.length()-1)
    w.close()
}

// This will match all files in the current directory with the file extension .xml
new File(".").eachFileMatch(FILES, ~/.*\.xml/) { File input ->
    // Set the output file name to be <file_name>_out.txt
    // Change this as you need
    stripTag(input, input.name + "_out.txt")
}

If you want to, you can add reading in the directory from the command line as well.