DatetimeIndex, df.ix[date] and df.ix[[date]] failure

[lodagro]

from stackoverflow

The original frame can not be shared, but following code reproduces the issue.

import pandas
import numpy as np
import datetime

# create large list of non periodic datetime
dates = []
sec = datetime.timedelta(seconds=1)
half_sec = sec / 2
d = datetime.datetime(2011, 12, 5, 20, 30)
n = 100000
for i in range(n):
    dates.append(d)
    dates.append(d + sec)
    dates.append(d + sec + half_sec)
    dates.append(d + sec + sec + half_sec)
    d += 3 * sec

# duplicate some values in the list
duplicate_positions = np.random.randint(0, len(dates) - 1, 20)
for p in duplicate_positions:
    dates[p + 1] = dates[p]
print "duplicated values at %s" % str(duplicate_positions)

df = pandas.DataFrame(np.random.randn(len(dates), 4),
                      index=dates,
                      columns=list('ABCD'))

pos = n * 3
timestamp = df.index[pos]
print "timestamp = df.index[%d] = %s" % (pos, str(timestamp))
print "timestamp in index? %s" % str(timestamp in df.index)
print "df.ix[timestamp] = \n%s" % str(df.ix[timestamp])
print "df.ix[[timestamp]] = \n%s" % str(df.ix[[timestamp]])

running this script outputs:

duplicated values at [143779 365504 338898 192427  63827 333772 366805 378282 375136  77760
  59843 215934 173395 185449 310734 184004  48594 221298 348967  86615]
timestamp = df.index[300000] = 2011-12-08 11:00:00
timestamp in index? True
df.ix[timestamp] = 
A    0.959639
B    0.651874
C    1.059355
D   -1.296393
Name: 2011-12-08 11:00:00
df.ix[[timestamp]] = 
Empty DataFrame
Columns: array([A, B, C, D], dtype=object)
Index: <class 'pandas.tseries.index.DatetimeIndex'>
Length: 0, Freq: None, Timezone: None

for n = 1000000 it gets worse

duplicated values at [ 686452 3862593 3747433   63099 2422495 2191536  486238 1632442 2373460
 3266942 3127937 2538658 1405505  739509 1519644 2817907 1005119  755410
 1784244   86211]
timestamp = df.index[3000000] = 2011-12-31 21:30:00
timestamp in index? False
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
...
TypeError: Cannot compare Timestamp with 1326116999500000000

[wesm]

Fixed this one