Python XML Parsing the child tags in order

Question

My XML file is similar to the one as below:

<suite name="regression_1">
    <test name="Login check" id="s1-t1">
        <keyword name="Valid Username and Password">
            <keyword name="Invalid Username or Password">
                <keyword name="Invalid password">
                    <message level="TRACE" >Return error</message>
                    <status status="PASS"/>
                </keyword>
                <message level="INFO">Return error</message>
                <status status="FAIL"/>
            </keyword>
            <message level="INFO">Return: None</message>
            <status status="PASS"/>
        </keyword>
        <status status="FAIL"/>
    </test>
    <test name="test-2" id="s1-t1">
        <keyword name="abc">
            <keyword name="def">
                <message level="INFO">Return error</message>
                <status status="FAIL"/>
            </keyword>
            <message level="INFO">Return: None</message>
            <status status="PASS"/>
        </keyword>
        <status status="FAIL"/>
    </test>
</suite>

My output should check for the keywords and give the keyword structure for those whose status is "FAIL". There will be many keywords in a test and there may or may not be child keywords.

**** Sample Output *******

Suite: regression_1

Test Name: Login check

Keyword failed: ["Valid Username & Password", "Invalid Username or Password"]

Failure test case message : Return error

Suite: regression_1

Test Name: test-2

Keyword failed: ["abc","def"]

Failure test case message : Return error

---

My code is able to dig till the last child to collect the fail status. But not able to parse the proper path which is required for analysis. Also I think the complete loop is not getting executed. ie if 3rd child is "PASS", its not coming back to the 2nd child to check its status.

def getStatusForNode(tc):
    status_to_be_returned = []
    is_just_father = False


    for child in tc.childNodes:
        if child.nodeName == "keyword":
            is_just_father = True
            status_to_be_returned.append(getStatusForNode(child)[0])
            keyword_track.append(child.getAttribute("name"))

    if not is_just_father:
        status = tc.getElementsByTagName('status')
        return [(tc, status)]


    return  status_to_be_returned


DOMTree = xml.dom.minidom.parse("output.xml")
collection = DOMTree.documentElement
tc_entry = collection.getElementsByTagName("suite")

top = Element('tests')
comment = Comment("This xml is generated only for failing tests")
top.append(comment)


for tc in tc_entry:
    if tc.hasAttribute("name"):
       print("Suite name: {}".format(tc.getAttribute("name")))

    tests = tc.getElementsByTagName('test')
    for test in tests:
        keyword_track = []
        for child in test.childNodes:
            if child.nodeName == "keyword":
                children_status = getStatusForNode(child)
                for (tc_name, status) in children_status:
                    for state in status:
                        if state.getAttribute("status") != "PASS":
                            print("---")
                            print("Test name: {}".format(test.getAttribute("name")))
                            print("Keyword failed: {}".format(tc_name.getAttribute("name")))
                            print("Status: {}".format(state.getAttribute("status")))
                            messages = tc_name.getElementsByTagName('msg')
                            print("Failure test case messages:")
                            for message in messages:
                                print(message.childNodes[0].data)
                            print ("")

Output received from this code:

Test name: ABC

Keyword name: keyword_1-2-3

Status: FAIL

Failure test case messages: Failed in level 3

Any suggested optimisations for the code?

Answer 1

Question : XML Parsing the child tags in order

Solution with xml.etree.ElementTree , for instance:

Note : Still makes no sense to have the First <keyword> in Keyword faild: , both have PASS . If you want to have the First <keyword> in Output, remove # .

from xml.etree import ElementTree as ET

with open('output.xml') as fh:
    suite = ET.fromstring(fh.read())

# Find all <test>"
for test in suite.findall('./test'):
    keyword_failed = []
    # first_keyword = test.find('./keyword')
    # keyword_failed = [first_keyword.attrib['name']]
    message = None

    # Find all <test><keyword> <status status="FAIL">
    for keyword in test.findall('.//keyword/status[@status="FAIL"]/..'):
        keyword_failed.append(keyword.attrib['name'])
        message = keyword.find('./message')

        print('Suite: {}'.format(suite.attrib['name']))
        print('\tTest Name: {}'.format(test.attrib['name']))
        print('\tKeyword failed: {}'.format(keyword_failed))
        print('\tFailure test case message : level={} {}'.format(message.attrib['level'], message.text))

Output :
Suite: regression_1
Test Name: Login check
Keyword failed: ['Invalid Username or Password']
Failure test case message : level=INFO Return error
Suite: regression_1
Test Name: test-2
Keyword failed: ['def']
Failure test case message : level=INFO Return error

Tested with Python: 3.4.2