How to add cassandra table column dynamically?

Question

I\'m trying to add a new columns to cassandra table dynamically. I\'m using below version -

cqlsh 5.0.1

I\'m using python to interact with Cass

Answer 1

Cassandra internally stores data as rows, each row has a key (Partition key) and dynamic number of columns (clustering key). So, you can use Clustering Key value for your column names, e.g

CREATE TABLE my_data (
     pk text,
     column text,
     value text,
     PRIMARY KEY (pk, column)
); 

Insert new columns and values by a regular INSERT query:

INSERT INTO my_data (pk, column, value) VALUES ('pk1', 'A', 'value A'); 
INSERT INTO my_data (pk, column, value) VALUES ('pk1', 'B', 'value B');
INSERT INTO my_data (pk, column, value) VALUES ('pk1', 'C', 'value C');
...  

Get all columns for pk1:

SELECT * FROM my_data WHERE pk='pk1';

Updated

Assume, you have table my_data as described above and you want to add some columns and data for a specific pk value. In python code perform insert query:

pk = 'pk'
columns_data = {'A':'value for A','B':'value  for B','C': 'value for C'} #dynamic column data
for col_name, col_value in columns_data.iteritems():
   try:
      session.execute("INSERT INTO my_data (pk, column, value) VALUES (%s, %s, %s)", (pk, col_name, col_value))
   except:
      pass

Moreover, you can use asynchronous driver's methods, to achieve more performance of inserting.