datetime - Merge two time series with offset? -

i have 2 files time series data, this:

file_a.csv: t,x,y,z 00:00:00,1,1,1 00:00:01,2,2,2 00:00:02,3,3,3 00:00:03,4,4,4  file_b.csv: t,x,y,z 00:00:00,5,5,5 00:00:01,6,6,6 00:00:02,7,7,7 

and merge them in order get:

t,x,y,z 00:00:00,1,1,1 00:00:01,2,2,2 00:00:02,3,3,3 00:00:03,4,4,4 00:00:04,5,5,5 00:00:05,6,6,6 00:00:06,7,7,7 

basically, want offset "t" of data set n+1 last value of "t" of data set n.

how can that? combine_first not want: merges 2 columns.

needed guess stuff in question (exact format of time, times sorted, increments fixed), here general idea.

starting dataframes:

import pandas pd stringio import stringio import numpy np = pd.read_csv(stringio('t,x,y,z\n00:00:00,1,1,1\n00:00:01,2,2,2\n00:00:02,3,3,3\n00:00:03,4,4,4')) b = pd.read_csv(stringio('t,x,y,z\n00:00:00,5,5,5\n00:00:01,6,6,6\n00:00:02,7,7,7')) 

convert times , sort (possibly can skip last part):

a.t = pd.to_datetime(a.t) a.sort(columns=[a.t.name], inplace=true) b.t = pd.to_datetime(b.t) b.sort(columns=[b.t.name], inplace=true) 

increment each time in b difference between , first time in b + last time in a plus 1 sec.

b.t = b.t - b.t[0] + a.t.values[-1] + np.timedelta64(1, 's') 

finally, concat:

pd.concat([a, b]) 

(note times here full datetimes now, i.e., dates.)