https://pypi.python.org/pypi/xlwt-future/0.8.0 is py3 compatible fork of dev-python/xlwt. How about bumping to the fork to get py3 compatibility.
Would you want to provide it under dev-python/xlwt, or do some virtual?
Also, see https://github.com/python-excel/xlwt/pull/32.
(In reply to Dirkjan Ochtman from comment #2) > Also, see https://github.com/python-excel/xlwt/pull/32. So no response from original upstream.
(In reply to Justin Lecher from comment #3) > (In reply to Dirkjan Ochtman from comment #2) > > Also, see https://github.com/python-excel/xlwt/pull/32. > > So no response from original upstream. To be honest, if it were for me, I wouldn't be really happy with someone telling me to use his fancy compat library that no other package uses. Especially that there're 2to3 and 3to2 already.
(In reply to Michał Górny from comment #4) > To be honest, if it were for me, I wouldn't be really happy with someone > telling me to use his fancy compat library that no other package uses. > Especially that there're 2to3 and 3to2 already. Does this mean a 2to3 run will fix it?
I have no idea. Just saying that I doubt that upstream will really like this. And I'm not really convinced that the person who did the fork will actually maintain it. Not that I say that xlwt itself will be more maintained...
I agree with Michał that just adding this is probably not the best approach. The future package is definitely non-standard. Let's bet on the xlwt maintainer actually stepping up to the plate at some point.
dev-python/xlwt has py3 support now