Adding a new column in the first ordinal position in a pyspark dataframe

Question

I have a pyspark data frame like:

+--------+-------+-------+
| col1   | col2  | col3  |
+--------+-------+-------+
|  25    |  01   |     2 |
|  23    |  12   |     5 | 
|  11    |  22   |     8 |
+--------+-------+-------+

and I want to create new dataframe by adding a new column like this:

+--------------+-------+-------+-------+
| new_column   | col1  | col2  | col3  |
+--------------+-------+-------+-------+
|  0           |  01   |     2 |  0    |
|  0           |  12   |     5 |  0    |
|  0           |  22   |     8 |  0    |
+--------------+-------+-------+-------+

I know I can add column by:

df.withColumn("new_column", lit(0))

but it adds column at last like this:

+--------------+-------+-------+-------------+
| col1         | col1  | col2  | new_column  |
+--------------+-------+-------+-------------+
|  25          |  01   |     2 |  0          |
|  23          |  12   |     5 |  0          |
|  11          |  22   |     8 |  0          |
+--------------+-------+-------+-------------+

Answer 1

you can reorder columns using select.

df = df.select('new_column','col1','col2','col3')
df.show()

Answer 2

You can always reorder the columns in a spark DataFrame using select, as shown in this post.

In this case, you can also achieve the desired output in one step using select and alias as follows:

df = df.select(lit(0).alias("new_column"), "*")

Which is logically equivalent to the following SQL code:

SELECT 0 AS new_column, * FROM df