BF: define expected Series of correct for arch (eg i386) int #15044
Conversation
| expected = Series([0, 0]) | ||
| # Explicit dtype since Series produces int64 for ints, while cut | ||
| # (due to numpy.searchsorted) would use int32 on i386, so let's assure | ||
| # correct default to the architecture int |
There was a problem hiding this comment.
we don't test on 32-bit, but if this works that is ok
specify dtype='intp' here
There was a problem hiding this comment.
though this works fine on windows now (e.g. 64-bit points on 32-bit systems).
and are you on master? this looks like an older line number.
There was a problem hiding this comment.
There was a problem hiding this comment.
actually I think this is a bug, cut on ints should always be 64-bit regardless of arch.
There was a problem hiding this comment.
https://github.com/pandas-dev/pandas/blob/master/pandas/tools/tile.py#L218
needs an .astype('int64', copy=False) as these can get returned to the user in some paths; and they are 32-bit on 32-bit :< (while 64bit on others)
jreback
added
Compat
Current coverage is 84.75% (diff: 100%)@@ master #15044 diff @@
==========================================
Files 145 145
Lines 51146 51146
Methods 0 0
Messages 0 0
Branches 0 0
==========================================
- Hits 43351 43350 -1
- Misses 7795 7796 +1
Partials 0 0
|
… view of type object
| # date time is just accidental | ||
| if prop in ('itemsize', 'nbytes') \ | ||
| and values.dtype.name == 'object': | ||
| continue |
There was a problem hiding this comment.
@jreback - I have added one more "skip" which was failing on i386 -- described "incode". Does it make sense? ;)
|
can you rebase / update |
|
this was closed by / superseded #15766 |
git diff upstream/master | flake8 --diff