BUG: DataFrame constructor with copy=False and missing columns creates columns that are views of each other

[jorisvandenbossche]

See #44889 (comment)

With copy=False and constructing a DataFrame with missing columns in the passed data, the columns get constructed as identical arrays, so mutating one also incorrectly mutates the other columns:

In [10]: df = pd.DataFrame(index=[1, 2, 3], columns=["a", "b", "c"], copy=False)

In [11]: df.iloc[0, 0] = 0

In [12]: df
Out[12]: 
     a    b    c
1    0    0    0
2  NaN  NaN  NaN
3  NaN  NaN  NaN

[simonjayhawkins]

With copy=False and constructing a DataFrame with missing columns in the passed data, the columns get constructed as identical arrays, so mutating one also incorrectly mutates the other columns:

This is a regression from 1.2.5

first bad commit: [66e6ba3] API: honor copy=True when passing dict to DataFrame (#38939)

cc @jbrockmendel