I pull from a SQL database which sometimes looks like this.
A B C D _ E F _ G
So when I put it into a data frame and print it as a CSV I want it to look like
A, B, C, D, , E, F, , G
But instead it looks like
A, B, C, D, E, F, G
So when I have a full row and then a row with some data missing, if I put it into an excel sheet and deliminate by comma, they do not match up properly.
df5 = pd.DataFrame(SQLQuery) df5.to_csv(r'finalResults.csv', index = False)
Advertisement
Answer
You should use df5 = pd.read_sql_query(SQLQuery, connectionString)
Then if you do df5.to_csv or .to_excel it will properly format