uint64 multicolumn sort

[scottupdike]

Sorting uint64 columns is incorrect when sorting on multiple columns:

>>> import pandas as pd
>>> import numpy as np
>>> df = pd.DataFrame({'a': pd.Series([18446637057563306014, 1162265347240853609]), 'b': pd.Series([1, 2])})
>>> df['a'] = df['a'].astype(np.uint64)
>>> df.sort(['a', 'b'])['a']
0    18446637057563306014
1     1162265347240853609
Name: a, dtype: uint64
>>> df.sort(['a'])['a']
1     1162265347240853609
0    18446637057563306014
Name: a, dtype: uint64
>>> pd.show_versions()

INSTALLED VERSIONS
------------------
commit: None
python: 2.7.9.final.0
python-bits: 64
OS: Linux
OS-release: 2.6.32-40.87-jump1.03
machine: x86_64
processor:
byteorder: little
LC_ALL: None
LANG: en_US.UTF-8

pandas: 0.15.2
nose: 1.3.4
Cython: None
numpy: 1.9.1
scipy: 0.15.1
statsmodels: None
IPython: 2.3.1
sphinx: None
patsy: None
dateutil: 2.4.0
pytz: 2014.10
bottleneck: None
tables: None
numexpr: 2.4
matplotlib: 1.4.2
openpyxl: None
xlrd: None
xlwt: None
xlsxwriter: None
lxml: None
bs4: 4.3.2
html5lib: None
httplib2: None
apiclient: None
rpy2: None
sqlalchemy: None
pymysql: None
psycopg2: None

[jreback]

uint64 not super-well supported; marked as a bug

[mroeschke]

This looks to work on master. Could use a test:

In [59]: >>> df = pd.DataFrame({'a': pd.Series([18446637057563306014, 1162265347240853609]), 'b': pd.Series([1, 2])})
    ...: >>> df['a'] = df['a'].astype(np.uint64)
    ...: >>> df.sort_values(['a', 'b'])['a']
Out[59]:
1     1162265347240853609
0    18446637057563306014
Name: a, dtype: uint64

In [60]: df.sort_values(['a'])['a']
Out[60]:
1     1162265347240853609
0    18446637057563306014
Name: a, dtype: uint64

In [61]: pd.__version__
Out[61]: '0.26.0.dev0+593.g9d45934af'