将日期插入大 dataframe

Question

I have df with 3 columns - a , b , dt . I want to insert rows in this df such that where there are two similar values of a and b all the dates in between are inserted with values of a and b repeated.

>>> import pandas as pd
>>> from datetime import datetime as dt
>>> df = pd.DataFrame({'a':['abd', 'abd', 'rds', 'rds', 'rsd', 'rsd', 'tsb'], 'b':['ar','ar','pr','pr','sg','sg','sg'], 'dt':[dt(2013,1,1), dt(2013,1,4), dt(2014,7,3), dt(2014,7,14), dt(2016,4,8), dt(2016,4,9), dt(2016,4,9)]})
>>> df
     a   b         dt
0  abd  ar 2013-01-01
1  abd  ar 2013-01-04
2  rds  pr 2014-07-03
3  rds  pr 2014-07-14
4  rsd  sg 2016-04-08
5  rsd  sg 2016-04-09
6  tsb  sg 2016-04-09
>>> 

The desired output df is as follows:

>>> df
      a   b         dt
0   abd  ar 2013-01-01
1   abd  ar 2013-01-02
2   abd  ar 2013-01-03
3   abd  ar 2013-01-04
4   rds  pr 2014-07-03
5   rds  pr 2014-07-04
6   rds  pr 2014-07-05
7   rds  pr 2014-07-06
8   rds  pr 2014-07-07
9   rds  pr 2014-07-08
10  rds  pr 2014-07-09
11  rds  pr 2014-07-10
12  rds  pr 2014-07-11
13  rds  pr 2014-07-12
14  rds  pr 2014-07-13
15  rds  pr 2014-07-14
16  rsd  sg 2016-04-08
17  rsd  sg 2016-04-09
18  tsb  sg 2016-04-09
>>> 

Answer 1

This is a groupBy and resample operation. Try:

(df.set_index('dt')
   .groupby(['a', 'b'], group_keys=False, as_index=False)
   .resample('D')
   .ffill()
   .reset_index())

           dt    a   b
0  2013-01-01  abd  ar
1  2013-01-02  abd  ar
2  2013-01-03  abd  ar
3  2013-01-04  abd  ar
4  2014-07-03  rds  pr
5  2014-07-04  rds  pr
6  2014-07-05  rds  pr
7  2014-07-06  rds  pr
8  2014-07-07  rds  pr
9  2014-07-08  rds  pr
10 2014-07-09  rds  pr
11 2014-07-10  rds  pr
12 2014-07-11  rds  pr
13 2014-07-12  rds  pr
14 2014-07-13  rds  pr
15 2014-07-14  rds  pr
16 2016-04-08  rsd  sg
17 2016-04-09  rsd  sg
18 2016-04-09  tsb  sg