Selecting from decimal (18,6) return NaN in pandas #109

Activity

1. Diyago commented on Sep 3, 2018

   

   Diyago

   on Sep 3, 2018

   Author

   More actions

   My problem easily fixed my using another library:

   from turbodbc import connect
   driver="Teradata Database ODBC Driver 16.10"
   connection = connect(driver=driver,
                         server=ip, 
                         DBCName='.com', 
                         uid=username,
                         pwd=password)
   
   cursor = connection.cursor()
   query = 'sel * from tb'
   cursor.execute(query)
   data = pd.DataFrame(cursor.fetchallnumpy()) #to get pandas dataframe
   data.to_csv('tb.csv')
   cursor.close()
   

   Is this library actually supported?
2. tetraptych commented on Sep 3, 2018

   

   tetraptych

   on Sep 3, 2018

   Contributor

   More actions

   Note that in the second example you are avoiding pd.read_sql entirely, suggesting this method might be causing the problem.

   I suggest fetching the rows from the database using the Teradata library (via cursor.execute(query).fetchall()), then creating a dataframe using the list of returned Row objects. Depending on what the pandas constructor expects, you may need to access the .columns and .values of individual rows.