BUG: resample with TimedeltaIndex, fenceposts are off

[jreback]

brought up in #12925
some examples in #15167

Resampling fence posts are off when upsampling < 1 Day.

In [5]: df = DataFrame(data=[[1, 3], [-5, 10], [0, 0]],index=pd.timedelta_range('00:00:00','00:10:00',freq='5T'))

In [6]: df.resample('1T').asfreq()
Out[6]: 
            0     1
00:00:00  1.0   3.0
00:01:00  NaN   NaN
00:02:00  NaN   NaN
00:03:00  NaN   NaN
00:04:00  NaN   NaN
00:05:00 -5.0  10.0
00:06:00  NaN   NaN
00:07:00  NaN   NaN
00:08:00  NaN   NaN
00:09:00  NaN   NaN

In [8]: df.resample('1T').mean()
Out[8]: 
            0     1
00:00:00  1.0   3.0
00:01:00  NaN   NaN
00:02:00  NaN   NaN
00:03:00  NaN   NaN
00:04:00  NaN   NaN
00:05:00 -5.0  10.0
00:06:00  NaN   NaN
00:07:00  NaN   NaN
00:08:00  NaN   NaN
00:09:00  NaN   NaN
00:10:00  0.0   0.0

[jorisvandenbossche]

Isn't this the same as #12926, which you fixed already?

[jreback]

very similar, but this test case fails as my original soln didn't work.

[wcwagner]

Same issue appears when downsampling:

In [6]: from pandas import *
In [7]: rng = timedelta_range(start='1d', periods=10, freq='d')
In [8]: df = DataFrame({'value':range(10)}, index=rng)
In [9]: df.resample('2D').count()
Out[9]: 
         value
1 days       2
3 days       2
5 days       2
7 days       2
9 days       2
11 days      0

In [10]: df.resample('2D').asfreq()
Out[10]: 
        value
1 days      0
3 days      2
5 days      4
7 days      6
9 days      8