CODEWITHSUNDEEP
xmlxpathjmeter
user2917629
user2917629

Reputation: 942

Parsing a JMeter response (an XML-formatted .srx file)

I'm using JMeter to run a test where the response is returned as an XML-formatted .srx SPARQL file. The error I'm getting is:

2015/04/28 1:30:39 ERROR - jmeter.util.XPathUtil: Type=Val=false
    Tol=false org.xml.sax.SAXParseException; lineNumber: 1; columnNumber: 1;
    Content is not allowed in prolog.

2015/04/28 11:30:39 WARN  - jmeter.extractor.XPathExtractor: SAXException
    while processing (sparql/results/result/binding/literal) Content is not
    allowed in prolog.

I'm pretty new to JMeter. My guess is that my XPath Extractor somehow doing its magic in a place I'm not expecting. The XML file response looks like this (a snipplet from a file), without any space before or any extra characters:

<?xml version='1.0' encoding='UTF-8'?>
<sparql xmlns='http://www.w3.org/2005/sparql-results#'>
  <head>
    <variable name='p'/>
  </head>
  <results>
    <result>
      <binding name='p'>
        <literal datatype='http://www.w3.org/2001/XMLSchema#date'>1999-08-16</literal>
      </binding>

I set up my XPathExtractor this way:

Reference Name:Results
XPath query:sparql/results/result/binding/literal
default Value:Not Found

Can anyone help?

Upvotes: 0

Views: 3552

Answers (3)

user2917629
user2917629

Reputation: 942

I had to replace the XPath Extractor with the Regular Expression Extractor. Then, to find the string 1999-08-16 date, I used the regex (\d{4}-\d{2}-\d{2}). Somehow the date data comes on a separate line.

Upvotes: 0

Dmitri T
Dmitri T

Reputation: 168247

I would go for the following approach:

  1. In user.properties file (located in /bin folder of your JMeter installation) add the next line:

    xpath.namespace.config=my.properties

  2. Create my.properties file in /bin folder of your JMeter installation and add the following line there:

    s=http://www.w3.org/2005/sparql-results#

  3. Configure your XPath Extractor the following way:

    • Check Use Namespaces box
    • Reference name: any variable name of your choice, i.e. literal
    • XPath Query: //s:literal/text()

In regards to your question: Content is not allowed in prolog message usually stands for not well-formed XML, my suggestion will be applicable to valid XML

References:

Upvotes: 1

Muhammad Hamed
Muhammad Hamed

Reputation: 1239

Your xml had a namespace without a prefix

<sparql xmlns='http://www.w3.org/2005/sparql-results#'>

while your XPATH is not considering the namespace .. you can choose one of two solutions

  1. Either change namespace to be with a prefix and adjust your XPATH.

  2. Or use this XPATH which considering the namespace

    /*[local-name()='sparql']/*[local-name()='results']/*[local-name()='result']/*[local-name()='binding']/*[local-name()='literal']

I hope this could help.

Upvotes: 0

Related Questions