Getting text of parent, child and their child-CodePudding

<avis>
<numeroseao>1331795</numeroseao>
<numero>61628-3435560</numero>
<organisme>Ville de Québec</organisme>
<fournisseurs>
  <fournisseur>
    <nomorganisation>APEL ASSOCIATION POUT DU LA MARAISNORD</nomorganisation>
    <adjudicataire>1</adjudicataire>
    <montantsoumis>0.000000</montantsoumis>
    <montantssoumisunite>0</montantssoumisunite>
    <montantcontrat>89732.240000</montantcontrat>
    <montanttotalcontrat>0.000000</montanttotalcontrat>
  </fournisseur>
</fournisseurs>
</avis>

So there is avis, avis has fournisseurs, fournisseurs has further nodes. How do I get these values to a data frame?

I am using the below code

element_tree = ET.parse('D:\\python_script\\temp2\\AvisRevisions_20200201_20200229.xml')
root = element_tree.getroot()
for child in root.findall('.//avis/*/*/*'):

for child in root.findall('.//avis/*'):

but it only gets me either the parent nodes or child nodes, not all of them.

CodePudding user response：

Since your data is not flat it might cause issues when you import the xml directly into pandas. In this scenario a library like pandas_read_xml might be useful:

import pandas_read_xml as pdx

df = pdx.read_xml(xml)
df = pdx.fully_flatten(df)  # this should get you the structure you want

In the line above, the xml variable is your "AvisRevisions_20200201_20200229.xml" file.

For a flatter structure you could use Pandas as such:

import pandas as pd

df = pd.read_xml(xml, xpath="//fournisseurs")

If you are looking for the entire "avis" section, you can replace the xpath parameter by:

df = pd.read_xml(xml, xpath="//avis")

From this, pandas should create the dataframe with appropriate columns. Here is a link to the Pandas docs.

CodePudding user response：

Try the below

import xml.etree.ElementTree as ET
import pandas as pd

xml = '''<avis>
<numeroseao>1331795</numeroseao>
<numero>61628-3435560</numero>
<organisme>Ville de Québec</organisme>
<fournisseurs>
  <fournisseur>
    <nomorganisation>APEL ASSOCIATION POUT DU LA MARAISNORD</nomorganisation>
    <adjudicataire>1</adjudicataire>
    <montantsoumis>0.000000</montantsoumis>
    <montantssoumisunite>0</montantssoumisunite>
    <montantcontrat>89732.240000</montantcontrat>
    <montanttotalcontrat>0.000000</montanttotalcontrat>
  </fournisseur>
</fournisseurs>
</avis>'''
root = ET.fromstring(xml)

data = []
fournisseur = root.find('.//fournisseur')
data.append({e.tag:e.text for e in fournisseur})
df = pd.DataFrame(data)