SAS 将数据集设置为 Python

Question

I have 2 sas datasets and I'm using SET statement to combine both of them into a new one by a key column. Here is how the data looks (note: '.' is a missing or null value)

data1       data2  
ID X  Y     ID  X  Z
01 12 11    01  .  4
02 15 .     03 17  6
            03 18  .

data combine;
set data1 data2;
by id;
run;

Result of the output dataset 'combine' is as below:

combine
ID  X  Y  Z
01 12  11  .
01 .   .   4
02 15  .   .
03 17  .   6
03 18  .   .

Can anyone let me know how do I do this in Pandas/Python. I tried using pd.concat() but that's not giving the above desired output. Appreciate if anybody can help.

Answer 1

concat

pd.concat([data1, data2], ignore_index=True).sort_values('ID')

   ID     X     Y    Z
0  01  12.0  11.0  NaN
2  01   NaN   NaN  4.0
1  02  15.0   NaN  NaN
3  03  17.0   NaN  6.0
4  03  18.0   NaN  NaN

---

append

data1.append(data2, ignore_index=True).sort_values('ID')

   ID     X     Y    Z
0  01  12.0  11.0  NaN
2  01   NaN   NaN  4.0
1  02  15.0   NaN  NaN
3  03  17.0   NaN  6.0
4  03  18.0   NaN  NaN

---

Response to Comment

Sorry...basically, all I need is just a 'flag' to indicate the data coming from 'data1' and data2. in simple words, creating a new variable called 'flag' and assign it to 1 if the data is coming from data1 and assign value 2 for the data coming from data2. hope it's clear...Thanks again! – user11580242

You can use concat

pd.concat({1: data1, 2: data2}, names=['flag']).reset_index('flag').sort_values('ID')

   flag  ID     X     Y    Z
0     1  01  12.0  11.0  NaN
0     2  01   NaN   NaN  4.0
1     1  02  15.0   NaN  NaN
1     2  03  17.0   NaN  6.0
2     2  03  18.0   NaN  NaN