What are the exact differences between CLI and CQL , i.e.
Do the CLI and CQL support all featuresets offered by cassandra ?
Is there a difference between creating a keyspace in CLI v/s CQL ?
Similarly is there a difference between creating a columnfamily in CLI v/s CQL ?
Are objects created with one tool useable/operable with the another ?
Going forward what's the future of CLI and CQL ? Will they be both maintained and developed in parallel, or will CQL be the eventual winner ?
For a new development, what is the recommendation, use CLI or CQL ? and why ?
CLI v/s CQL
(2 posts) (2 voices)-
Posted 1 year ago #
-
By CLI, I think you mean cassandra-cli, and by CQL, I think you mean cqlsh. They have a lot of differences. Cqlsh is intended to be a command-line tool which natively speaks the CQL language. The CQL language is intended to be the main way to interface with Cassandra going forward, although the Thrift interface will be supported for some time still. The CQL language can be used in your application code using drivers (existing for Python, Java, PHP, node.js, Ruby, and more), so you can use the same syntax in your code and in the command-line inspection tool.
The cassandra-cli tool is lower-level than Cqlsh, and uses the Thrift interface instead of CQL. It has its own special syntax for creating columnfamilies and keyspaces, for inserting data, and for making queries. You'll find that while it's not as user-friendly, there are a few things it can still do that Cqlsh can't, like show column timestamps and show the storage engine rows underlying cql3 composite columnfamilies, as opposed to the transformed CQL rows. But for the most part, new users should probably be using the CQL interface and the Cqlsh tool.
Posted 1 year ago #
Reply
You must log in to post.
