CLN: Removed coerce param in pd.to_timedelta and pd.to_datetime #13602

gfyoung · 2016-07-09T21:51:53Z

Deprecated back in v0.17.0 here. Seems appropriate to carry though now.

jreback · 2016-07-09T21:55:28Z

doc/source/whatsnew/v0.19.0.txt

+Removal of prior version deprecations/changes
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+- ``pd.to_timedelta`` has dropped the ``coerce`` parameter (:issue:`13602`)


add what it's in favor of

Done. Old build to cancel here.

codecov-io · 2016-07-10T02:55:35Z

Current coverage is 85.30% (diff: 100%)

Merging #13602 into master will increase coverage by 0.04%

@@             master     #13602   diff @@
==========================================
  Files           140        139     -1   
  Lines         50634      50139   -495   
  Methods           0          0          
  Messages          0          0          
  Branches          0          0          
==========================================
- Hits          43173      42771   -402   
+ Misses         7461       7368    -93   
  Partials          0          0

Powered by Codecov. Last update d98e982...c8c5f49

gfyoung · 2016-07-10T03:00:23Z

@jreback : Travis is passing, so ready to merge if there are no other concerns.

jreback · 2016-07-10T20:51:53Z

actually, I don't think this can be done w/o doing for to_datetime as well. Either both or none.

jreback · 2016-07-10T20:54:51Z

and we generally are waiting much longer, see the list @jorisvandenbossche put up at the very bottom:
#6581

any of those are fair game. Yes we could/should do some of the 0.17.0 ones, but let's get some of the others first.

gfyoung · 2016-07-10T20:57:47Z

@jreback : so if I also remove the to_datetime param, will this be merged in for 0.19.0 or not? A duration of two major releases seems like a good amount of time for people to adjust their code I thought.

jreback · 2016-07-10T21:00:28Z

@gfyoung yes it is normally ok, but we SO many deprecations (esp from 0.17.0.), that I think let's clean everything up thats < 0.17.0 first. So I would say defer this to 0.20.0..

gfyoung · 2016-07-10T21:08:57Z

@jreback : Hmm...I'm not sure I totally follow that logic, especially since I do see a deprecation removal from 0.17.0 being listed in that issue that you linked to (the legacy offsets one).

jreback · 2016-07-10T21:15:30Z

yeah. I don't have a problem with anything from 0.17.0. BUT this is a 'biggie' (well to_datetime is really). So no objection if you include both (for consistency). But if you have a limited pool of PR energy, would prefer other ones to be addressed .....

gfyoung · 2016-07-10T21:18:09Z

@jreback : this one isn't that difficult to remove AFAICT - I'll also take a look at some of the other deprecation removals and put up PR's for those if I can.

jreback · 2016-07-10T21:20:06Z

right, add on to_datetime removal then would be ok, but @jorisvandenbossche might feel different.

gfyoung · 2016-07-10T21:21:55Z

@jreback : fair enough - in the meantime, I think we do have one deprecation removal that I think all of us can agree upon (#13419) 😄

gfyoung · 2016-07-12T16:24:56Z

@jreback : removed the coerce parameter in to_datetime, and Travis is still passing. @jorisvandenbossche thoughts?

jorisvandenbossche · 2016-07-12T17:40:52Z

Personally I would table this for 0.20 ..
The other deprecated things we are removing are all even from 0.15 and (the maybe's) from 0.16 (except for the offset/freq PR), and the 0.17 release is less than a year old.

jorisvandenbossche · 2016-07-12T17:42:07Z

Maybe we need some more 'formal' rules on how long deprecations are kept at the minimum? Eg minimal two feature releases and minimal a year.

gfyoung · 2016-07-13T02:07:06Z

@jorisvandenbossche: When has two feature releases ever taken more than a year? Two feature releases sure, but a year seems a little long too me, and not just because I would like to merge this ;)

jreback · 2016-07-13T02:13:08Z

I think let's defer this one to 0.20. I don't see any downside to delaying this.

jorisvandenbossche · 2016-07-13T09:34:25Z

but a year seems a little long too me, and not just because I would like to merge this ;)

We may not like it, but it is a fact that not all users upgrade each version but often skip a few at once ...
I understand you rather see it merged, but rebasing it in a few months will also not be that much work!

gfyoung · 2016-07-13T09:39:17Z

@jorisvandenbossche : this deprecation removal process is somewhat new to me, and I was not aware of user tendencies to skip version upgrades (I generally follow them quite closely for my projects). However, I perfectly understand! I was being somewhat facetious when I mentioned that I wanted to merge this now from the ;) I added at the end.

[ci skip]

jorisvandenbossche

This looks OK to me!

jorisvandenbossche · 2016-10-13T20:16:02Z

@gfyoung can you force push again? (possibly due to the repo transfer the PRs open before the transfer cannot be merged from the github interface)

jreback · 2016-10-13T20:54:49Z

@jorisvandenbossche if you merge this, we for sure need to branch 0.19.1 (which is ok)

jreback · 2016-10-13T20:55:10Z

also should remove coerce from pd.to_datetime for consistency.

gfyoung · 2016-10-13T21:26:42Z

@jreback : Ah, good point. I didn't realize it had been deprecated back in 0.17.0. Will do then.

gfyoung · 2016-10-14T03:46:22Z

@jreback : Actually, never mind! I did remove it in this PR! Need to change the title.

jreback · 2016-10-14T10:29:15Z

ok lgtm. our first non-0.19.1 PR!

jorisvandenbossche · 2016-10-15T19:58:52Z

Merged! Thanks @gfyoung

jreback · 2016-10-15T20:08:03Z

@jorisvandenbossche so need to branch 0.19.1 (as this should not be included)!

Deprecated back in `v0.17.0` <a href="https://github.com/pydata/pandas /commit/987b7e7e586b8df1d127406c69e0a9094a1a5322">here</a>. Seems appropriate to carry though now. Author: gfyoung <gfyoung17@gmail.com> Closes pandas-dev#13602 from gfyoung/remove-coerce-to-timedelta and squashes the following commits: 091af83 [gfyoung] CLN: Removed coerce param in pd.to_timedelta

gfyoung force-pushed the remove-coerce-to-timedelta branch from 9f0781d to 2d5eaeb Compare July 9, 2016 21:52

jreback reviewed Jul 9, 2016
View reviewed changes

gfyoung force-pushed the remove-coerce-to-timedelta branch from 2d5eaeb to 76e0bcd Compare July 9, 2016 21:59

jreback added Timedelta Timedelta data type Deprecate Functionality to remove in pandas labels Jul 10, 2016

gfyoung force-pushed the remove-coerce-to-timedelta branch 3 times, most recently from d0e4b05 to a0455c8 Compare July 12, 2016 02:26

gfyoung force-pushed the remove-coerce-to-timedelta branch from a0455c8 to 8f3e8a3 Compare July 13, 2016 02:07

gfyoung force-pushed the remove-coerce-to-timedelta branch from 8f3e8a3 to ed5f093 Compare July 13, 2016 06:03

jorisvandenbossche added this to the 0.20.0 milestone Jul 13, 2016

gfyoung force-pushed the remove-coerce-to-timedelta branch 4 times, most recently from a9c191f to a799398 Compare September 24, 2016 19:10

gfyoung force-pushed the remove-coerce-to-timedelta branch 6 times, most recently from 0d6aae4 to e3c54f5 Compare October 2, 2016 20:19

gfyoung force-pushed the remove-coerce-to-timedelta branch 3 times, most recently from 0a968ed to 18c4f9d Compare October 9, 2016 19:25

CLN: Removed coerce param in pd.to_timedelta

091af83

[ci skip]

gfyoung force-pushed the remove-coerce-to-timedelta branch from 18c4f9d to 091af83 Compare October 10, 2016 21:02

jorisvandenbossche approved these changes Oct 10, 2016

View reviewed changes

gfyoung force-pushed the remove-coerce-to-timedelta branch 2 times, most recently from 1363567 to eb62e50 Compare October 13, 2016 14:16

gfyoung changed the title ~~CLN: Removed coerce param in pd.to_timedelta~~ CLN: Removed coerce param in pd.to_timedelta and pd.to_datetime Oct 14, 2016

gfyoung force-pushed the remove-coerce-to-timedelta branch from eb62e50 to c43d28f Compare October 14, 2016 17:18

jorisvandenbossche closed this in 750b6ae Oct 15, 2016

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

CLN: Removed coerce param in pd.to_timedelta and pd.to_datetime #13602

CLN: Removed coerce param in pd.to_timedelta and pd.to_datetime #13602

gfyoung commented Jul 9, 2016

jreback Jul 9, 2016

gfyoung Jul 9, 2016

codecov-io commented Jul 10, 2016 •

edited

gfyoung commented Jul 10, 2016

jreback commented Jul 10, 2016

jreback commented Jul 10, 2016

gfyoung commented Jul 10, 2016 •

edited

jreback commented Jul 10, 2016

gfyoung commented Jul 10, 2016 •

edited

jreback commented Jul 10, 2016

gfyoung commented Jul 10, 2016

jreback commented Jul 10, 2016

gfyoung commented Jul 10, 2016

gfyoung commented Jul 12, 2016

jorisvandenbossche commented Jul 12, 2016

jorisvandenbossche commented Jul 12, 2016

gfyoung commented Jul 13, 2016

jreback commented Jul 13, 2016

jorisvandenbossche commented Jul 13, 2016

gfyoung commented Jul 13, 2016 •

edited

jorisvandenbossche left a comment

jorisvandenbossche commented Oct 13, 2016

jreback commented Oct 13, 2016

jreback commented Oct 13, 2016

gfyoung commented Oct 13, 2016

gfyoung commented Oct 14, 2016

jreback commented Oct 14, 2016

jorisvandenbossche commented Oct 15, 2016

jreback commented Oct 15, 2016

CLN: Removed coerce param in pd.to_timedelta and pd.to_datetime #13602

CLN: Removed coerce param in pd.to_timedelta and pd.to_datetime #13602

Conversation

gfyoung commented Jul 9, 2016

jreback Jul 9, 2016

Choose a reason for hiding this comment

gfyoung Jul 9, 2016

Choose a reason for hiding this comment

codecov-io commented Jul 10, 2016 • edited

Current coverage is 85.30% (diff: 100%)

gfyoung commented Jul 10, 2016

jreback commented Jul 10, 2016

jreback commented Jul 10, 2016

gfyoung commented Jul 10, 2016 • edited

jreback commented Jul 10, 2016

gfyoung commented Jul 10, 2016 • edited

jreback commented Jul 10, 2016

gfyoung commented Jul 10, 2016

jreback commented Jul 10, 2016

gfyoung commented Jul 10, 2016

gfyoung commented Jul 12, 2016

jorisvandenbossche commented Jul 12, 2016

jorisvandenbossche commented Jul 12, 2016

gfyoung commented Jul 13, 2016

jreback commented Jul 13, 2016

jorisvandenbossche commented Jul 13, 2016

gfyoung commented Jul 13, 2016 • edited

jorisvandenbossche left a comment

Choose a reason for hiding this comment

jorisvandenbossche commented Oct 13, 2016

jreback commented Oct 13, 2016

jreback commented Oct 13, 2016

gfyoung commented Oct 13, 2016

gfyoung commented Oct 14, 2016

jreback commented Oct 14, 2016

jorisvandenbossche commented Oct 15, 2016

jreback commented Oct 15, 2016

codecov-io commented Jul 10, 2016 •

edited

gfyoung commented Jul 10, 2016 •

edited

gfyoung commented Jul 10, 2016 •

edited

gfyoung commented Jul 13, 2016 •

edited