How to handle custom named index when copying a dataframe using pd.read_clipboard?

Question

Given this data frame from some other question:

         Constraint Name    TotalSP       Onpeak    Offpeak
Constraint_ID               
77127   aaaaaaaaaaaaaaaa         


        

Answer 1

read_clipboard by default uses whitespace to separate the columns. The problem you see is because of the whitespace in the first column. If you specify two or more spaces as the separator, based on the table format it will figure out the index column itself:

df = pd.read_clipboard(sep='\s{2,}')

df
Out: 
                  Constraint Name  TotalSP   Onpeak  Offpeak
Constraint_ID                                               
77127          aaaaaaaaaaaaaaaaaa -2174.50 -2027.21  -147.29
98333          bbbbbbbbbbbbbbbbbb -1180.62 -1180.62     0.00
1049           cccccccccccccccccc -1036.53  -886.77  -149.76

index_col argument can also be used to tell pandas the first column is the index, in case the structure cannot be inferred from the separator alone:

df = pd.read_clipboard(index_col=0, sep='\s{2,}')

Answer 2

This is not as cool as @ayhan's answer, but most of the time works pretty well. Assuming you are using ipython or jupyter, just copy and paste the data into %%file:

Then do some quick edits. With multi-indexes, just move the index up a line, something like this (also shortening "Constraint ID" to "ID" to save a little space in this case):

%%file foo.txt
ID       Constraint Name    TotalSP       Onpeak    Offpeak
77127   aaaaaaaaaaaaaaaaaa  -2174.5     -2027.21    -147.29
98333   bbbbbbbbbbbbbbbbbb  -1180.62    -1180.62     0
1049    cccccccccccccccccc  -1036.53    -886.77     -149.76

pd.read_fwf('foo.txt')
Out[338]: 
      ID     Constraint Name  TotalSP   Onpeak  Offpeak
0  77127  aaaaaaaaaaaaaaaaaa -2174.50 -2027.21  -147.29
1  98333  bbbbbbbbbbbbbbbbbb -1180.62 -1180.62     0.00
2   1049  cccccccccccccccccc -1036.53  -886.77  -149.76

read_fwf generally works pretty well on tabular stuff like this, correctly dealing with spaces in column names (usually). Of course, you can also use this basic method with read_csv.

The nice thing about this method is that for small sample data you can deal with just about any of the weird ways that users post data here. And there are a lot of weird ways. ;-)