python pandas: filter out rows with multiple conditions

Question

I need to fix a CSV file. When I read it via pandas it shows me just one column but it has multiple.

So I split the column:

df = df['test_column'].str.split(' ', expand = True)

and got 168 rows.

I changed the name of the columns: df.set_axis(list(range(1, 169)), axis = 1, inplace = True)

I am currently looking each column if it is completely empty or it has value in it with:

a = 122 #just a column name
df = df[df[a].notnull()]
print(df[a].to_string())

The Problem here is, even the specific row is empty it still shows me. I assume there is just spaces (" ").

So how can do multiple conditions?

Answer 1

IIUC replace all empty strings or spaces to missing values first:

#removed ' ', by default plitting by arbitrary space
df = df['test_column'].str.split(expand = True)
#starting columns by 1
df.columns += 1

df = df.replace(r'^\s*$', np.nan, regex=True)

a = 122 #just a column name
df = df[df[a].isna()]
print (df)