Data packaging and Directory Restructuring

[nickdelgrosso]

Cool package! However, when I "pip installed" dfply, the diamonds.csv flie didn't download as well, and I got import errors, since the file gets read in as part of the package load process. I took a look at the packaging, and tried to help out by making some changes that would make it more pip-compatible and would (I hope--I'm still a newbie at packaging, and this is actually my first pull request ever!) simplify the growth of the project by making "six" and "pandas_ply" explicit global requirements that could be downloaded during installation.

This pull request:

Moves data out of python package,
added setup.py and MANIFEST.in links to data
made data.py a module that loads the data, to keep code working without API change.
Added pandas_ply and six to requirements, to remove vendor code from python package as well.
I'm sorry that I made all these into a single commit (stupid, I know). If there are any fixes you'd like me to make, please let me know!

[kieferk]

This sounds like a good idea to me. Sorry about the data issue with pip, I was not aware of that.

I probably won't be able to merge this in until later on this evening or tomorrow morning, but will get to it as soon as I can. I am going to change the target branch to develop from master to make sure everything plays nice with the changes, then I'll merge it into the master branch as the next version.

[kieferk]

OK everything is checked out and merged into master branch as well as develop. Thanks for doing this!

[nickdelgrosso]

Hey, that's great! Glad I could contribute--this is a really cool
project!

Best wishes,

Nick

On 2016-10-11 05:17, Kiefer Katovich wrote:

OK everything is checked out and merged into master branch as well as
develop. Thanks for doing this!

You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub [1], or mute the
thread [2].

Links:

[1] #6 (comment)
[2]
https://github.com/notifications/unsubscribe-auth/ADoHd0mLIBM5ikJAxHmEmCp5Ti1G4vz0ks5qyv-0gaJpZM4KTD8x

[bmwilly]

Hi. Cool package indeed! I tried installing with

pip install git+https://github.com/kieferk/dfply.git

and still got the error you describe above when trying to import (diamonds.csv does not exist). Any ideas?

[nickdelgrosso]

Okay, good to know; it sounds like the data isn't being packaged correctly. Really, diamonds.csv shouldn't be read during the import in the first place, so maybe instead of improving the packaging, it would make more sense to change that aspect. Meanwhile, @bmwilly could you please try downloading the package using git and installing it via the setup.py package? I just tried it on my work computer and it worked out fine. If you're not familiar with the process, the commands would look something like this::

git clone https://github.com/kieferk/dfply.git
cd dfply
python setup.py install

[bmwilly]

@neuroneuro15 yes, that works!

[kieferk]

Hi all,

Sorry to hear this is not working via pip. I try and figure out how to fix
the csv packaging once github is back online.

On Fri, Oct 21, 2016 at 8:28 AM, Brandon Williams notifications@github.com
wrote:

@neuroneuro15 https://github.com/neuroneuro15 yes, that works!

—
You are receiving this because you modified the open/close state.
Reply to this email directly, view it on GitHub
#6 (comment), or mute
the thread
https://github.com/notifications/unsubscribe-auth/ARNAmFAR8yecyTBUTYh613kMRGB2xoXeks5q2NoKgaJpZM4KTD8x
.

[bmwilly]

@kieferk I think this was fixed already. I posted my comment before #8 was resolved (I'm not sure what commit actually fixed the problem).

[kieferk]

I think the github version was working but not the pip version. My re-org of the files was a fix for pip.