CODEWITHSUNDEEP
pythonxmlelementtree
Zbigniew
Zbigniew

Reputation: 428

ElementTree XPath strange behaviour

Hi
I'm using ElementTree (1.3) with Python 2.7 and enjoy XPath functionality,
however one of search results surprised me.

My XML example:

<OTF>
  <FECS state="disabled" version="2.2.0.0">
    <BackEndCompatibility major="2.2" state="disabled">
        <BackEnd state="disabled" version="2.2.0.0"/>
    </BackEndCompatibility>
  </FECS>
</OTF>

Question 1:
When I use findall to get first found element

version = "2.2.0.0"
found = list(txml.findall(".//BackEnd[@version='%s']" % version))
return found and found[0] or None

it finds nothing.

However when I change XML file, so that BackEnd element contains subelements, 

        <BackEnd state="disabled" version="2.2.0.0">
           <any_dummy_element/> 
        </BackEnd>

then searched element is found properly.

Did you face such a behaviour?
Am I doing sth wrong or this is a bug in ElementTree implementation?

Question 2:
Another issue I have is xmlns.
Let's assume I change XML first line to contain xmlns:

<OTF xmlns="http://si-wiki/OTFCompatibility">
</OTF>

In such a case I have to change find string to: 

".//{http://si-wiki/OTFCompatibility}BackEnd[@version='%s']"

Is there any way to tell ElementTree to ignore xmlns during parsing and treat all elements' names (including root) like they had no prefix?

Regards,
Zbigniew

Upvotes: 3

Views: 554

Answers (1)

Zbigniew
Zbigniew

Reputation: 428

For question No1:
When I replaced lines 

    found = list(txml.findall(".//BackEnd[@version='%s']" % version))
    return found and found[0] or None

with

    found = txml.findall(".//BackEnd[@version='%s']" % version)
    if found:
        return found[0]
    return None

then correct result is returned without dummy children hack.

Upvotes: 1

Related Questions