cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sylvain Lebresne (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-8261) Clean up schema metadata classes
Date Wed, 12 Nov 2014 10:58:34 GMT


Sylvain Lebresne commented on CASSANDRA-8261:

Looks good overall, though it moves the timestamp from milliseconds to microseconds for tracing,
which while a good thing per-se doesn't feel like something that should be sneaked in a big
refactoring patch. So I'd have a minor preference for pulling that part in a separate ticket,
but +1 otherwise.

> Clean up schema metadata classes
> --------------------------------
>                 Key: CASSANDRA-8261
>                 URL:
>             Project: Cassandra
>          Issue Type: Improvement
>            Reporter: Aleksey Yeschenko
>            Assignee: Aleksey Yeschenko
>            Priority: Minor
>             Fix For: 3.0
>         Attachments: 8261-isolate-hadcoded-system-tables.txt, 8261-isolate-thrift-code.txt
> While working on CASSANDRA-6717, I've made some general cleanup changes to schema metadata
classes - distracted from the core purpose. Also, being distracted from it by other things,
every time I come back to it gives me a bit of a rebase hell.
> Thus I'm isolating those changes into a separate issue here, hoping to commit them one
by one, before I go back and finalize CASSANDRA-6717.
> The changes include:
> - moving all the toThrift/fromThrift conversion code to ThriftConversion, where it belongs
> - moving the complied system CFMetaData objects away from CFMetaData (to SystemKeyspace
and TracesKeyspace)
> - isolating legacy toSchema/fromSchema code into a separate class (LegacySchemaTables
- former DefsTables)
> - refactoring CFMetaData/KSMetaData fields to match CQL CREATE TABLE syntax, and encapsulating
more things in CompactionOptions/CompressionOptions/ReplicationOptions classes
> - moving the definition classes to the new 'schema' package

This message was sent by Atlassian JIRA

View raw message