Benedict created CASSANDRA-8840:
-----------------------------------
Summary: Classify our Assertions (like
com.google.base.Preconditions)
Key: CASSANDRA-8840
URL: https://issues.apache.org/jira/browse/CASSANDRA-8840
Project: Cassandra
Issue Type: Improvement
Components: Core
Reporter: Benedict
Priority: Minor
I raised this on IRC, then dropped it due to opposition, but it's possible the
opposition was due to my conflation of the act of classification with the
disabling of some of the assertions. The two aren't wed, and I think it _would_
improve readability significantly by itself, as Ariel reminded me with his use
of google's Preconditions class in CASSANDRA-8692.
I would prefer to use our own version of this class, that we can force the
\@Inline compiler hint onto, so that we have no negative performance
implications. Also, we can then introduce a class of data corruption checks,
etc. I think this would aid readability, and also permit easier analysis of the
codebase via IDE (right now it's very hard to say what data corruption checks
we actually perform, for instance).
Thoughts?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
