[ https://issues.apache.org/jira/browse/CASSANDRA-19528?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Francisco Guerrero updated CASSANDRA-19528: ------------------------------------------- Source Control Link: https://github.com/apache/cassandra-analytics/commit/47fdb6448b6956249790d5dc7bb76b699d35c079 Resolution: Fixed Status: Resolved (was: Ready to Commit) > [Analytics] Use a classloader to isolate in-jvm dtest classes in integration > test > --------------------------------------------------------------------------------- > > Key: CASSANDRA-19528 > URL: https://issues.apache.org/jira/browse/CASSANDRA-19528 > Project: Cassandra > Issue Type: Improvement > Components: Analytics Library > Reporter: Francisco Guerrero > Assignee: Francisco Guerrero > Priority: Normal > Labels: analytics, test-framework > Fix For: NA > > Time Spent: 50m > Remaining Estimate: 0h > > We need to isolate the in-jvm dtest classes (coming from the shaded dtest > jar) from the classes in Cassandra Analytics when running integration tests. > The isolation will allow us to test the Analytics code in the > {{AppClassloader}} without any class pollution from classes coming from the > shaded dtest jar. In some cases, the pollution can spill to classes that are > shaded in the analytics project (metrics, json serialization, guava, etc). > With the isolation in place we can be more confident that the code we are > testing is the code that will be running in a Spark environment. -- This message was sent by Atlassian Jira (v8.20.10#820010) --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org For additional commands, e-mail: commits-h...@cassandra.apache.org