pdftoxml in utils.py is not portable to Windows.

[StevenMaude]

1. The /dev/null needs to be NUL on Windows.
2. NamedTemporaryFile behaves differently in Windows to Unix.

[aparna06]

Is this still true?

I am trying to convert pdf to xml on a Windows machine , Python 3 and I am getting an error on the "return xmldata.decode('utf-8')"

Please let me know.

[StevenMaude]

This is still the case as no-one's changed the code there.

You could:

• just run pdftohtml.exe separately and dump the results to a file (either via a script, or via Python subprocess or however you like)
• or you can try using this as a starting point for replacing the code in this package.

[aparna06]

Thank you. I shall try both.