Details
-
Sub-task
-
Status: Open
-
Major
-
Resolution: Unresolved
-
3.3.0
-
None
-
None
Description
As discussed in HADOOP-16090; having an S3AWriteOpContext to match the read context lets us do some interesting things
- if we record the state of the file (is this an overwrite)? we'll know not to bother with deleting any parent dir markers (Assumption: there are none(
- If a check for a parent dir was made (which we don't currently do, but...) then iff that was an empty dir marker, we'd know to only delete that marker and not go further up the tree.
Attachments
Issue Links
- is depended upon by
-
HADOOP-16456 Refactor the S3A codebase into a more maintainable and testable form
- Open
- relates to
-
HADOOP-16090 S3A Client to add explicit support for versioned stores
- Resolved
- links to