Removed Ix indexer and updated Index/MultiIndex path

[tk3369]

Fixes #64

[sethaxen]

Thanks for this fix! Is this still backwards-compatible with pre-1.0.0 pandas?

[sethaxen]

I've resolved the Travis errors on #63.

[tk3369]

Thanks for this fix! Is this still backwards-compatible with pre-1.0.0 pandas?

Technically, there are two changes in this PR:

1. Remove _IXIndexer
2. Update the python paths for Index and MultiIndex

The Ix indexer has been deprecated by Pandas for several years and finally they removed it in 1.0.0. Since we're removing functionality, I would say it's backward compatible. The second change is unlikely going to be compatible since the paths to find the Python classes are different.

Unless we want to support both pre- and post-1.0 configurations with the same code, I would suggest that we put something in the README and ask people to pin the release if they need to use an older version of Pandas.

I've resolved the Travis errors on #63.

I suppose that you mean the Travis error here :-) I think the dependent Python package is pytables rather than tables. I don't know much about the history but it seems that it's been renamed. I can fix that with the same PR.

[tk3369]

Oh, I see what you mean about the Travis fix in #63.

It seems odd that it used to work before.
https://travis-ci.org/JuliaPy/Pandas.jl/jobs/556473084?utm_medium=notification&utm_source=github_status

P.S. I was a bit confused earlier. It looks like the package is named tables for pip but pytables for conda. Another reason why we like Julia 😄

[sethaxen]

It seems odd that it used to work before.
https://travis-ci.org/JuliaPy/Pandas.jl/jobs/556473084?utm_medium=notification&utm_source=github_status

Yes, although that was 7 months ago. One of the dependencies could have changed, or as has happened to me before, Travis could have changed something with their setup.

[sethaxen]

@malmaud it would be great to get this and #63 merged soon with a new release. I have a package that depends on Pandas.jl that currently can't be used with pandas 1.0.0.

[malmaud]

Ok, I'll check it out today.

[sethaxen]

Thanks!

[malmaud]

Thanks for this PR! If it's not too hard, I think I'd like to see if we can get this backwards-compatible. I'll try adding a commit to this since it doesn't look like it will be too hard.

[tk3369]

That's great! It's the best of both worlds! Thanks @malmaud