Added metadataframe.py which has MetaDataframe class. This class stores... #2695

Closed
wants to merge 16 commits into
from

Conversation

Projects
None yet
3 participants

... a dataframe and attempts to overload all operators and promote dataframe methods to object methods in hopes that users can subclass this effectively as if it were a Dataframe itself.

hugadams added some commits Jan 14, 2013

Added metadataframe.py which has MetaDataframe class. This class stor…
…es a dataframe and attempts to overload all operators and promote dataframe methods to object methods in hopes that users can subclass this effectively as if it were a Dataframe itself.

Hi guys. I'm not exactly sure what tests to run on this class. Any help would be appreciated.

One thing I'm worried about is that I didn't overload every attributes that DataFrame relies on. I did most of the basic ones, and all those of the public API, but there may be more that I missed because they were inherited from NDFrame or PandasObject.

Also, I decided to to call my deserialization functions mload and mloads to avoid name conflicts with the pandas API.

Adam and others added some commits Jan 21, 2013

Removed copy(deep=True) from _deepcopy, since returns that yielded a …
…series were unable in to implement this. Not sure of the ramifications this may have, but does not seem to affect my usecases.
Fixed indexing.. Solution is a bit of a pain, but seems to be the way…
… to go for ensuring that .ix always retains custom attributes.
Changed _deepcopy to _transfer to reflect that it is just a method to…
… transfer existing attributes between objects.
Contributor

stephenwlin commented Jan 28, 2013

Why not MetaDataFrame instead of MetaDataframe, to be consistent with DataFrame capitalization?

Not sure why I did this initially but will change it thanks for the heads
up.
On Jan 28, 2013 3:30 PM, "stephenwlin" notifications@github.com wrote:

Why not MetaDataFrame instead of MetaDataframe, to be consistent with
DataFrame capitalization?


Reply to this email directly or view it on GitHubhttps://github.com/pydata/pandas/pull/2695#issuecomment-12802717.

Contributor

y-p commented Mar 13, 2013

Lots of discussion (#2845) but no real concensus or champion. Users
are free to reuse this independently of it being merged into pandas.

If there are no objections, closing in 3..2..

I will keep the metadataframe module hosted separately in my PyUvVis (https://github.com/hugadams/pyuvvis) package and continue to use it and develop it with the intention of general use cases. I monitor the mailing list frequently so any requests for it can be directed to me:

hughesadam87@gmail.com

When I get around to writing the PyUvVis official documentation, I'll put a small writeup together on this class by itself and share. If there ever becomes a point that merging would be desired, that is fine.

Contributor

y-p commented Mar 13, 2013

Thank you. closing now.

@y-p y-p closed this Mar 14, 2013

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment