Joining Dataframe with multiple dates (Vlookup)

Question

I have the following in a file below and was wondering how to join them together, I know how to join if they were two separate dataframe but the values range from D1 up to D20+ and will not be practical to create so many multiple dataframes. My objective is to join them based on D1 date as per below.

    D1           D1value  D2           D2value  D3           D3value   
    1/2/2018     21.14    1/2/2018     11.14    1/6/2018     1.55
    1/3/2018     19.13    1/3/2018     51.14    1/13/2018    2.66
    1/6/2018     19.89    1/14/2018    31.14    1/14/2018    3.77
    1/13/2018    20.24   
    1/14/2018    20.91

After joining it should be

    D1           D1value  D2value  D3value   
    1/2/2018     21.14    11.14    NaN
    1/3/2018     19.13    51.14    NaN
    1/6/2018     19.89    NaN      1.55 
    1/13/2018    20.24    NaN      2.66
    1/14/2018    20.91    31.14    3.77

Any kind advise please?

For piRSquared answer add df:

df = pd.concat([proc(d) for k, d in df.groupby(g, 1)], axis=1)

Both piRSquared and jp_data_analysis answer works The issue is I want to select both as the right answer!

Answer 1

I'm assuming that columns come in adjacent pairs.

def proc(d):
    v = d.dropna().values
    return pd.Series(v[:, 1], pd.to_datetime(v[:, 0]), name=d.columns[1])

g = np.arange(len(df.columns)) // 2
pd.concat([proc(d) for k, d in df.groupby(g, 1)], axis=1)

           D1value D2value D3value
2018-01-02   21.14   11.14     NaN
2018-01-03   19.13   51.14     NaN
2018-01-06   19.89     NaN    1.55
2018-01-13   20.24     NaN    2.66
2018-01-14   20.91   31.14    3.77

---

Setup
I'm assuming your file looks exactly like this:

D1,D1value,D2,D2value,D3,D3value
1/2/2018,21.14,1/2/2018,11.14,1/6/2018,1.55
1/3/2018,19.13,1/3/2018,51.14,1/13/2018,2.66
1/6/2018,19.89,1/14/2018,31.14,1/14/2018,3.77
1/13/2018,20.24,,,,
1/14/2018,20.91,,,,

I read it in with

df = pd.read_csv('test.csv')