Extract data from xml to Excel (Python 2.7 )

Question

i'm attempting to extract some data from a XML file and create a Excel with the information.

XML File:

<UniversalTransaction>
    <TransactionInfo>
        <DataContext>
            <DataSourceCollection>
                <DataSource>
                    <Type>AccountingInvoice</Type>
                    <Key>AR INV 00001006</Key>
                </DataSource>
            </DataSourceCollection>

            <Company>
                <Code>DCL</Code>
                <Country>
                    <Code>CL</Code>
                    <Name>Chile</Name>
                </Country>
                <Name>Your Chile Corp</Name>
            </Company>
...etc

Then I made this Code in python 2.7

import xml.etree.ElementTree as ET
import xlwt
from datetime import datetime

tree = ET.parse('ar.xml')
root = tree.getroot()

#extract xml
invoice = root.findall('DataSource')
arinv = root.find('Key').text
country = root.findall('Company')
ctry = root.find('Name').text

wb = xlwt.Workbook()
ws = wb.add_sheet('A Test Sheet')

ws.write(0, 0, arinv)
ws.write(0, 1, ctry)

wb.save('example2.xls')

But I get this error:

arinv = root.find('Key').text
'NoneType' object has no attribute 'text' 

And i guess it will be the same with

ctry = root.find('Name').text

Also when I change the "extract xml" part of the code to this

for ar in root.findall('DataContext'):
    nro = []
    ctry = []
    inv = ar.find('Key').text
    nro.append(inv)
    country = ar.find('Name').text
    ctry.append(country)

i get the following error:

ws.write(0, 0, arinv)
name 'arinv' is not defined

then again, I guess its the same with "ctry"

Windows 10, python 2.7

I'll apreciate any help, thanks.

Answer 1

It is better to ask shortened questions - without yours bunch of context code. Probably you find a solution yourself when you carefully try to split out exact short question.

According to the docs, Element.find basically finds only in direct children. You need to use some XPath (look about XPath expressions in the docs) like

root.findall('.//Key')[0].text

(given with assumption the Key always exists, contains text and unique within a document; ie without validation)