Glue Crawler cannot determine data structure from XML file

Hi,

I've created Glue Crawler to determine data sctructure from XML file uploaded to s3, and write Table into Data Catalog. I tried 2 approaches:

1. Use Glue default classifier - this is preferred option as I might get different xml files with different structures - don't want to create custom classifiers for each of them.
2. Use Glue Custom classifier (with Feed set as a row tag).

For above two approaches, the table is created with below schema:

None of the approaches seems to reflect the real structure of the XML file (I would expect to see something similar to schema extracted by the crawler in https://aws.amazon.com/blogs/big-data/process-and-analyze-highly-nested-and-large-xml-files-using-aws-glue-and-amazon-athena/ - each tag from my file should be represented in the table)

Not sure if I misconfigured something or Glue has some limitations on discovering schema from XML files?

The xml file with data has below structure (file size is below 1 KB):

<Feed>
  <Document Type="RESULTS Latest">
    <Data id="t123">
      <Info MatchDay="1">
        <Date>2023-08-05 15:00:00</Date>
      </Info>
     </Data>
    <Data id="t456">
     <Info MatchDay="1">
       <Date>2023-08-05 15:00:00</Date>
     </Info>
    </Data>
    <Team id="44">
      <Name>Bradford City</Name>
      <OfficialName>Bradford City</OfficialName>
    </Team>
    <Team id="12">
      <Name>Mansfield Town</Name>
      <OfficialName>Mansfield Town</OfficialName>
    </Team>
  </Document>
</Feed>