sunchao commented on code in PR #476:
URL: https://github.com/apache/datafusion-comet/pull/476#discussion_r1616194709
##########
common/src/main/scala/org/apache/spark/sql/comet/util/Utils.scala:
##########
@@ -264,4 +265,24 @@ object Utils {
throw new SparkException(s"Unsupported Arrow Vector for $reason:
${valueVector.getClass}")
}
}
+
+ /**
+ * Imports data from ArrowArray/ArrowSchema into a FieldVector. This is
basically the same as
+ * Java Arrow `Data.importVector`. `Data.importVector` initiates
`SchemaImporter` internally
+ * which is used to fill dictionary ids for dictionary encoded vectors.
Every call to
+ * `importVector` will begin with dictionary ids starting from 0. So,
separate calls to
+ * `importVector` will overwrite dictionary ids. To avoid this, we need to
use the same
+ * `SchemaImporter` instance for all calls to `importVector`.
+ */
+ def importVector(
Review Comment:
maybe just declare this as a method for `CometSchemaImporter`?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
