cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Aleksey Yeschenko (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-6833) Add json data type
Date Tue, 11 Mar 2014 13:23:44 GMT


Aleksey Yeschenko commented on CASSANDRA-6833:

I'm relatively strongly -1 on this. Even if it's just validation, it does send a wrong message
to users - especially the newcomers.

I'm afraid that people migrating from other databases, esp. with JSON-centric data models,
would just choose the 'easy' migration route and just continue sticking their stuff into C*
JSON columns (now using C* as a primitive key->JSON value store), instead of remodeling
it for wide C* partitions, collections, and user types.

We shouldn't be making the wrong way to do things easier (and it's already easy as it is -
you can stick all your JSON into a blob/text column). Adding an official JSON type on top
of that only legitimizes it and thus makes it worse.

I'm also strongly -1 on adding new CQL syntax for it, and even stronger -1 on making it cqlsh-only.
There is an expectation that CQL queries that work in cqlsh can be copied to the actual application
code and be used with the java/python-drivers, and this would violate that expectation.

> Add json data type
> ------------------
>                 Key: CASSANDRA-6833
>                 URL:
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>            Reporter: Jonathan Ellis
>            Priority: Minor
>             Fix For: 2.0.7
> While recognizing that UDT (CASSANDRA-5590) is the Right Way to store hierarchical data
in C*, it can still be useful to store json blobs as text.  Adding a json type would allow
validating that data.  (And adding formatting support in cqlsh?)

This message was sent by Atlassian JIRA

View raw message