Github user pwendell commented on a diff in the pull request:
https://github.com/apache/spark/pull/4166#discussion_r24301030
--- Diff:
core/src/main/scala/org/apache/spark/executor/ExecutorURLClassLoader.scala ---
@@ -19,56 +19,51 @@ package org.apache.spark.executor
import java.net.{URLClassLoader, URL}
-import org.apache.spark.util.ParentClassLoader
+/**
+ * A ClassLoader trait that changes the normal delegation scheme.
+ * Normally, it's (cache, parent, self). GreedyClassLoader uses
+ * (cache, self, parent). This lets this CL "hide" or "override"
+ * class defs that also exist in the parent loader.
+ */
+private[spark] trait GreedyClassLoader extends ClassLoader {
--- End diff --
Rather than making this a trait, can you just modify
ChildExecutorURLClassLoader to have this logic? We don't need it anywhere else
at this point, and this is the main intended goal of
ChildExecutorURLClassLoader (to do greedy loading like this).
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
