BUG: Fix to_dict() problem when using datetime DataFrame #11247

[ethanluoyc]

closes #11247

This is my first time contributing through PR so forgive me if I make any silly mistake.
The issue can be referred to in GH11247
I followed @jreback 's suggestion to box things up in to_dict() but I am not sure if this is what he means because I do not know exactly when _maybe_datetime() should be used. Initially I tried to tweak internals.py but it fails.

Let me know how I can improve from here. Hope I can really learn from my first PR.

[jreback]

see guidelines here: http://pandas.pydata.org/pandas-docs/stable/contributing.html

dont' turn on PEP changes for your editor.

[ethanluoyc]

I think I forgot to fetch to update my local master before my first commit to this PR. Somehow I messed up is it true?

[jreback]

add tests! (ideally for each one of the orients if we dont' have already).

[ethanluoyc]

which test case shall I add in? I am new to this :)

On 15 Oct 2015, at 2:50 AM, Jeff Reback notifications@github.com wrote:

add tests! (ideally for each one of the orients if we dont' have already).

—
Reply to this email directly or view it on GitHub #11327 (comment).

[ethanluoyc]

@jreback I have added tests:)

[jreback]

can you make this a mixed frame ,e.g. add a float and a string say. obviously need to adjust the tests as well.

[ethanluoyc]

I adjusted the tests and now it tests both mixed and single dtypes

[jreback]

pls add a whatsnew note for 0.17.1 in bug fixes

[ethanluoyc]

Just asking what I should do if I merge updates locally with my branch and need to merge the 2 commits into one..

—
Sent from Mailbox

On Fri, Oct 16, 2015 at 6:25 AM, Jeff Reback notifications@github.com
wrote:

pls add a whatsnew note for 0.17.1 in bug fixes

Reply to this email directly or view it on GitHub:
#11327 (comment)

[jreback]

merged via 89b4e5b

thanks!