wForget commented on code in PR #1359:
URL: https://github.com/apache/datafusion-comet/pull/1359#discussion_r1944107142
##########
native/core/src/parquet/parquet_support.rs:
##########
@@ -1861,6 +1863,42 @@ fn trim_end(s: &str) -> &str {
}
}
+// Default object store which is local filesystem
+#[cfg(not(feature = "hdfs"))]
+pub(crate) fn register_object_store(
+ session_context: Arc<SessionContext>,
+) -> Result<ObjectStoreUrl, ExecutionError> {
+ let object_store = object_store::local::LocalFileSystem::new();
+ let url = ObjectStoreUrl::parse("file://")?;
+ session_context
+ .runtime_env()
+ .register_object_store(url.as_ref(), Arc::new(object_store));
+ Ok(url)
+}
+
+// HDFS object store
+#[cfg(feature = "hdfs")]
+pub(crate) fn register_object_store(
+ session_context: Arc<SessionContext>,
+) -> Result<ObjectStoreUrl, ExecutionError> {
+ // TODO: read the namenode configuration from file schema or from
spark.defaultFS
Review Comment:
> do you mean the better place to register the object store will be inside
file_partitions iterator loop ?
Yes, is it possible that native scan paths correspond to multiple object
stores or are different from spark.defaultFs?
--
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]
