[ENH] Add orient argument and split option to to_dict. (GH7840) #8486
Conversation
jreback
changed the title
orient argument and split option to to_dict. (GH7840)
|
@rpetchler ok! pls add a release note in v0.15.0 (in the deprection section) and we can get this in. |
jreback
added
API Design
IO JSON
| orient : str {'dict', 'list', 'series', 'split', 'records'} | ||
| Determines the type of the values of the dictionary. | ||
|
|
||
| - dict (default) : dict like {column -> {index -> value}} |
There was a problem hiding this comment.
@jorisvandenbossche how is this formatting?
There was a problem hiding this comment.
yep, that is good
the blank line is indeed needed to have the list rendered as a list
There was a problem hiding this comment.
only, there is no indentation needed for the list relative from the Determines ....
DOC: warning about copying for df.append/concat. Fixes #7967
|
pls squash: https://github.com/pydata/pandas/wiki/Using-Git @jorisvandenbossche any other comments? |
Update documentation with deprecation and enhancement notices. Remove indentation from list in docstring.
…into to_dict_gh7840
|
Hmm... squashed the commits but accidentally merged as well. Advice on how to undo all this? Thanks for your patience-- |
|
|
|
For the rest looking good! |
|
merged via 5995dc3 @rpetchler I squashed and merged it. thanks! |
Follows up from #8486 in 0.15.0 by removing outtype in DataFrame.to_dict()
Follows up from pandas-dev#8486 in 0.15.0 by removing outtype in DataFrame.to_dict()
closes #7840
Unlike the implementation offered in #7840, this version preserves the option to pass prefixes to the
orientparameter.I also rewrote the docstring to use the more readable list format from
pd.DataFrame.to_json.