Python extract XML data with inconsistant child tags

Question

I have an XML file that I need to extract data from and insert into a database table. My struggle is that the XML data structure could contain inconsistant child tags. Meaning that (in the example below) one parent <Field> tag, may or may not contain a <ListValue> tag.

This is a short example and I will be adding additional <Field> tags potentially containing another <ListValue> tag. Note: All <Field> tags are expected to remain at the same level below the <Record> tag.

I want to see if anyone has a more "pythonic" way of converting this data than my example below. Maybe with list comprehension?

I will need to insert up to 4,000,000 <Record> level rows of data into a database, so I don't want to waste more time looping through the XML than is necessary. Speed will be essential.

Any assistance will be appreciated.

<?xml version="1.0" encoding="utf-16"?>
<Records count="10">
    <Metadata>
        <FieldDefinitions>
            <FieldDefinition id="15084" guid="f3426157-cbcb-4293-94e5-9f1c993db4b5" name="CCR_ID" alias="CCR_ID" />
            <FieldDefinition id="16335" guid="5dfddb49-9a7a-46ee-9bd2-d5bbed97a48d" name="Coming Due" alias="Coming_Due" />
        </FieldDefinitions>
    </Metadata>
    <LevelCounts>
        <LevelCount id="35" guid="661c747f-7ce5-474a-b320-044aaec7a5b1" count="10" />
    </LevelCounts>
    <Record contentId="20196771" levelId="35" levelGuid="661c747f-7ce5-474a-b320-044aaec7a5b1" moduleId="265" parentId="0">
        <Field id="15084" guid="f3426157-cbcb-4293-94e5-9f1c993db4b5" type="1">100383-320-V0217111</Field>
        <Field id="16335" guid="5dfddb49-9a7a-46ee-9bd2-d5bbed97a48d" type="4">
            <ListValues>
                <ListValue id="136572" displayName="121 - 180 days out">121 - 180 days out</ListValue>
            </ListValues>
        </Field>
    </Record>
    <Record contentId="20205193" levelId="35" levelGuid="661c747f-7ce5-474a-b320-044aaec7a5b1" moduleId="265" parentId="0">
        <Field id="15084" guid="f3426157-cbcb-4293-94e5-9f1c993db4b5" type="1">100383-320-V0217267</Field>
        <Field id="16335" guid="5dfddb49-9a7a-46ee-9bd2-d5bbed97a48d" type="4">
            <ListValues>
                <ListValue id="136572" displayName="121 - 180 days out">121 - 180 days out</ListValue>
            </ListValues>
        </Field>
    </Record>
    <Record contentId="20196779" levelId="35" levelGuid="661c747f-7ce5-474a-b320-044aaec7a5b1" moduleId="265" parentId="0">
        <Field id="15084" guid="f3426157-cbcb-4293-94e5-9f1c993db4b5" type="1">100384-320-V0217111</Field>
        <Field id="16335" guid="5dfddb49-9a7a-46ee-9bd2-d5bbed97a48d" type="4">
            <ListValues>
                <ListValue id="136572" displayName="121 - 180 days out">121 - 180 days out</ListValue>
            </ListValues>
        </Field>
    </Record>
</Records>

Here is my code for parsing the data:

from xml.etree import ElementTree
import pandas as pd

xml_string = '''SEE STRING ABOVE'''

auth_token = ElementTree.fromstring(xml_string.text)

dct = []
cols = ['CCR_ID', 'Coming_Due']

for r in auth_token.findall("Record"):
    for f in r.findall("Field"):

        if f.attrib['id'] == '15084':
            ccr_id = f.text

        for l in f.findall(".//ListValue"):
            coming_due = l.text

    dct.append((ccr_id, coming_due))


df = pd.DataFrame(dct)
df.columns = cols

print(df)

Here are my results:

                CCR_ID Coming_Due
0  100383-320-V0217111    121 - 180 days out
1  100383-320-V0217267    121 - 180 days out
2  100384-320-V0217111    121 - 180 days out
3  100384-320-V0217267    121 - 180 days out
4  100681-320-V0217111    121 - 180 days out
5  100681-320-V0217267      11 - 30 days out
6  100684-320-V0217111    121 - 180 days out
7  100684-320-V0217267      11 - 30 days out
8  100685-320-V0217111    121 - 180 days out
9  100685-320-V0217267      11 - 30 days out

Answer 1

If I understand you correctly, using pandas read_xml() may help:

df = pd.read_xml(string,"//Record//*")
df2= df[['Field','displayName']].copy()
df2['displayName'] = df2['displayName'].shift(-3)
df2.set_axis(['CCR_ID', 'Coming_Due'], axis=1,inplace=True)
df2.dropna()

Output based on your sample xml:

    Field   displayName
0   100383-320-V0217111     121 - 180 days out
4   100383-320-V0217267     121 - 180 days out
8   100384-320-V0217111     121 - 180 days out