Excel Copy Paste Way in Python

Question

I have a data frame as below

df = pd.DataFrame([[3,2,1],[4,5,6],[10,20,30]], columns = ['A','B','C'])

    A   B   C
0   3   2   1
1   4   5   6
2  10  20  30

Is there any way in python to mimic copy and paste function in excel? For example I want to copy paste row 0 column A and B and paste them into row 0 column B and C, such that it will become

    A   B   C
0   3   3   2
1   4   5   6
2  10  20  30

In a small data frame, I can use:

df.loc[0,'C'] = df.loc[0,'B']
df.loc[0,'B'] = df.loc[0,'A']

But my original data frame is sizable and I prefer not to do this one element by one element. I was also trying to do:

df.loc[0,['A','B']] = df.loc[0,['B','C']]

But my data in row 0 column A becomes NaN.

So is there a way of doing things similar like doing copy paste in excel in python (simply block a range of data, copy them and paste them on top of another existing data)? Thanks

The columns name do matter when you assign the value , so adding .values or to_numpy() to remove the impact of index — BENY
– BENY, Commented Nov 5, 2019 at 15:35
I did this, but like I mentioned above: I have a large data set, so I want the code to block certain range of row and paste it on top of another range of existing row. Thanks. — teteh May
– teteh May, Commented Nov 5, 2019 at 15:35
@anky_91 if you post an answer, let me know and I'll edit my post accordingly. — piRSquared
– piRSquared, Commented Nov 5, 2019 at 16:02

piRSquared · Accepted Answer · 2019-11-05 16:01:13Z

2

anky_91's answer

df.loc[0, ['B', 'C']] = df.loc[0, ['A', 'B']].to_numpy()

`shift`

There are many ways you can use shift. This is just one.

df.update(df.shift(axis=1)[['B', 'C']])

For reasons that I'm not happy about, you can provide a fill_value to shift to preserve the integer dtype

df.update(df.shift(axis=1, fill_value=0)[['B', 'C']])

answered Nov 5, 2019 at 16:01

piRSquared

296k68 gold badges509 silver badges654 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Randy · Accepted Answer · 2019-11-05 15:34:23Z

0

This mostly feels like a bad idea, but if it's what you really want to do, you can use .iloc to address columns by number and just shift them:

In [56]: df.iloc[0, 1:] = df.iloc[0, :-1].values

In [57]: df
Out[57]:
    A   B   C
0   3   3   2
1   4   5   6
2  10  20  30

answered Nov 5, 2019 at 15:34

Randy

14.9k3 gold badges40 silver badges44 bronze badges

2 Comments

teteh May Over a year ago

Hi Randy, is there any possibility to take a range of columns by their names instead of number? Thanks.

Randy Over a year ago

Yes, as noted in anky_91's comment, you can just flip your assignment around with .loc and use .values/.to_numpy() if you want to assign them by name.

Collectives™ on Stack Overflow

Excel Copy Paste Way in Python

2 Answers 2

anky_91's answer

`shift`

Comments

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

anky_91's answer

shift

Comments

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related

`shift`