Append not working with DataFrames in for loop

Question

I have a time series with 3 columns. I need to copy column 2 after the last row and the same with the column 3.

I created a for loop and append the dataframes, but the append doesn't seem to work. There is no error or warning, just doesn't work.

Initial DataFrame dataImport_selVar100:

    val01_ambient_temperature   val01_ambient_winddir   val01_ambient_windspeed
measure_time            
2019-03-24 07:30:00 12.956060   108.200005  4.166667
2019-03-24 07:40:00 12.999207   103.000000  3.666667
2019-03-24 07:50:00 12.761206   106.500000  4.533333
2019-03-24 08:00:00 12.523205   98.413330   3.916667
2019-03-24 08:10:00 12.285204   97.853333   4.055000

Code:

counterTest=0
for column in dataImport_selVar100:
    if counterTest==0: #initialize
        result0=pd.DataFrame(dataImport_selVar100.iloc[:,counterTest])
    else:
        result1=pd.DataFrame(dataImport_selVar100.iloc[:,counterTest])
        result0.append(result1,ignore_index=True,sort=False)

    #print(result[column])
    counterTest +=1

The actual results are just the ones from result0 (100 rows)

<class 'pandas.core.frame.DataFrame'>
DatetimeIndex: 100 entries, 2019-03-24 07:30:00 to 2019-03-25 00:00:00
Data columns (total 1 columns):
val01_ambient_temperature    100 non-null float64
dtypes: float64(1)
memory usage: 6.6 KB

The expected results are the sum of all the rows

<class 'pandas.core.frame.DataFrame'>
DatetimeIndex: 300 entries, 2019-03-24 07:30:00 to 2019-03-25 00:00:00
Data columns (total 3 columns):
val01_ambient_temperature    100 non-null float64
val01_ambient_winddir        100 non-null float64
val01_ambient_windspeed      100 non-null float64
dtypes: float64(2)
memory usage: 7.0 KB

Answer 1

result0.append(result1,ignore_index=True,sort=False)

Append returns the new dataframe. It does not happen inplace. You'll need:

result0 = result0.append(result1,ignore_index=True,sort=False)

Also be aware that append is very costly. Might be worth it to look into pd.concat.