TST/REF: collect tests by method

[jbrockmendel]

• closes #xxxx
• tests added / passed
• Ensure all linting tests pass, see here for how to run them
• whatsnew entry

[jbrockmendel]

@WillAyd do you have time to look at the npdev json failure (i think its affecting all PRs ATM)

[WillAyd]

Can’t dive too deep at the moment but could have some time tonight. If I had to guess the origin it may come from the PyArg_ParseTupleAndKeywords in objToJSON.c since the format string specifies int for the doublePrecision and indent arguments. Haven’t really debugged though

…

On Feb 16, 2021, at 11:39 AM, jbrockmendel ***@***.***> wrote: @WillAyd <https://github.com/WillAyd> do you have time to look at the npdev json failure (i think its affecting all PRs ATM) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#39832 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAEU4UJBNZHQ32377727YGDS7LCWBANCNFSM4XVXMQCQ>.

[jbrockmendel]

@seberg any ideas what in npdev could be causing a DeprecationWarning: an integer is required (got type float). Implicit conversion to integers using __int__ is deprecated, and may be removed in a future version of Python to be issued? I looked at recently-merged PRs and didn't see anything obvious.

Only occurs with datetime64tz columns

[seberg]

There is one stranger change in the recent dev version that might be related, would have to dig in

If you had a string + number or maybe also string + datetime64 promotion in an array creation, like np.array(["string", 1]), that might give you an object array, which in turn might run into this Deprecation warning? That is a bit around the corner though :). (numpy/numpy#18425 although that one focuses on masked arrays)

I also recently modified that float + timedelta64 always raises an error during a promotion. That could lead to a similar "to object" behaviour when using np.array([...]).

[WillAyd]

@jbrockmendel have you been able to reproduce locally? I tried with 3.8.5 and the DEV numpy but couldn't get the test to fail

[jbrockmendel]

have you been able to reproduce locally? I tried with 3.8.5 and the DEV numpy but couldn't get the test to fail

are you passing -W error? im getting the warning with np 1.19.5, which actually suggests its something on our end

[jreback]

FYI opened #39854 for the failure

[jreback]

@jbrockmendel same number tests before / after?

[jreback]

well your after will be higher (as you split some).

[jbrockmendel]

same number tests before / after?

nothing removed, just moved or split