3) I could eliminate the libxml2 dependency by just grepping the xml
for "<address .*ip=". Is reduced correctness (which is probably 0%
different in practice) worth the reduced package dependencies?
Or by using the ElementTree module that is part of the standard library:
https://docs.python.org/3.7/library/xml.etree.elementtree.html
I just quickly skimmed through the script to check what kind of functionality
you actually need when parsing the network XMLs, because ElementTree is simple
and kinda convenient, but limited...however, from what I've seen in this
patch, ElementTree should be sufficient here. Just my 2c that we can eliminate
the libxml2 dependency in another way as well :).
Regards,
--
Erik Skultety