BUG: output of df.to_string depends on whether columns is a CategoricalIndex or not

[topper-123]

The output of df.to_string is different if the columns is a CategoricalIndex rather than a normal Index:

>>> data = [[4, 2], [3, 2], [4, 3]]
>>> cols = ["aaaaaaaaa", "b"]
>>>  pd.DataFrame(data, columns=cols)
   aaaaaaaaa  b
0          4  2
1          3  2
2          4  3
>>> pd.DataFrame(data, columns=pd.CategoricalIndex(cols))
   aaaaaaaaa  b
0          4          2
1          3          2
2          4          3

We can see that the width of the "b" column depends on the width of the "a" column, if the columns are a CategoricalIndex.

These two examples should return the same repr.

[simonjayhawkins]

Thanks @topper-123 for the report. This was OK for earlier versions?

[topper-123]

Yes, this is a regression since v1.0.