Ivan Kanakarakis

@wahadatjan would you like to describe what you are trying to achieve?

Hello, this is known and will be fixed, but I don't have a specific timeframe. This time of the year is very busy with meetings but I hope I can...

Thank for filling in this issue. I agree that we should look into this.

Correct. This is due to the namespace prefix being embedded within the string value of the `type` XML attribute. We could setup a mechanism to process namespaces with specified prefix...

The XSD files are there and loaded; and they include the `PersonIdentifierType`. This is part of the eIDAS XSD files, here: https://github.com/IdentityPython/pysaml2/blob/14c649a/src/saml2/data/schemas/eidas-schema-attribute-naturalperson.xsd#L5 > The parser should be aware that xsi:type...

Doing the same with `lxml`, preserves the prefixes ```py import lxml.etree as let xmlstr = """CZ/CZ/f93fab3a-b132-4c21-ba05-f00a9988441e""" el = let.fromstring(xmlstr) let.tostring(el) ``` ```xml CZ/CZ/f93fab3a-b132-4c21-ba05-f00a9988441e ```

@melanger I do not see a way to configure the builtin XML parser - to stop removing the namespace-prefixes it does not recognize as referenced - or, to consider certain...

I did some work to hack the code and use lxml with pysaml2. Have a look at #940 This of course needs a lot more work; not all tests pass,...

Give it a try first, to ensure it can work for you. I had a minimal test case there just to get things started. Ideally this can become a configurable...

also found [this collection](http://www.roesler-ac.de/wolfram/hello.htm)