tustvold commented on code in PR #5281:
URL: https://github.com/apache/arrow-rs/pull/5281#discussion_r1442762521
##########
object_store/src/local.rs:
##########
@@ -1082,6 +1097,58 @@ fn convert_walkdir_result(
}
}
+
+/// Download a remote object to a local [`File`]
+pub async fn upload(store: &dyn ObjectStore, location: &Path, opts:
PutOptions, file: &mut std::fs::File) -> Result<()> {
+ // Determine file size
+ let metadata = file.metadata().map_err(|e| Error::FileMetadata {
+ source: e.into(),
+ })?;
+ let file_size = metadata.len();
+
+ // Set a threshold for when to switch to multipart_put
+ let multipart_threshold: u64 = 50 * 1024 * 1024;
+
+ if file_size <= multipart_threshold {
+ let mut buffer = Vec::with_capacity(file_size as usize);
+ file.read_to_end(&mut buffer).map_err(|e|
Error::UnableToReadBytesFromFile{
+ source: e
+ })?;
+ let bytes = Bytes::from(buffer);
+ store.put_opts(&location, bytes, opts).await?;
+ Ok(())
+ } else {
+ let (_id, mut writer) = store.put_multipart(&location).await?;
+ let mut buffer = vec![0u8; 5 * 1024 * 1024];
+ while let Ok(size) = file.read(&mut buffer) {
+ if size == 0 {
+ break;
+ }
+ writer.write_all(&buffer[..size]).await.unwrap();
+ }
+
+ writer.flush().await.unwrap();
+ writer.shutdown().await.unwrap();
+ Ok(())
+ }
+}
+
+
+/// Upload a local [`File`] to a remote object store
+pub async fn download(store: &dyn ObjectStore, location: &Path, opts:
GetOptions, file: &mut File) -> Result<()> {
Review Comment:
I think the intention of using `File` directly is to handle the specifics
that arise from filesystems - e.g. the need for buffering and the lack of async
APIs (unless you count platform specific APIs like io_uring)
--
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]
