Pandas IO XML Issue Tracker

[ParfaitG]

Issue tracking for new pandas.io.xml module (after merge: #39516):

To-Do

• BUG: Fix declaration not showing for xml_declaration=True when pretty_print=False for etree parser.
• CLN: Centralize docstrings to avoid repetition in formats.py and frame.py.
• CLN: When etree supports pretty_print, remove xml.dom.minidom reliance.
• TYP: Refactor code for type hints on parse_doc methods for optional dependency, lxml.
• TST: Add tests for edge cases (ParserError, OSError, URLError, etc. ). See checklist in tests code.
• TST: Add more tests for storage_options (i.e., read/write to pandas-test S3 bucket).

Enhancements

• ENH: Add parse_dates and dtype converters similar to other IO methods.
• ENH: Add support for nullable dtyes in reading and exporting XML.
• ENH: Add iterparse for memory efficient parsing of large XML. See etree iterparse and lxml iterparse.
• ENH: Add xpath_vars to pass $ variables in xpath expression. See lxml xpath() method.
• ENH: Add xsl_params to pass values into XSLT script. See lxml stylesheet parameters.
• ENH: Add prefix_cols to specify which columns should have namespace prefixes.
• ENH: Add nested (bool) to write out nested node-sets for data frames with hierarchical columns or multindex.
• ENH: Add engine for external processors for XPath and XSLT 2.0 and 3.0, XQuery, streaming, and others.

[mroeschke]

It appears a lot of the issues have been address. Additionally IMO it's easier to use individual issues + labels to identify what needs working on so closing for now