cross csp transfer support#29
Draft
lizdotsh wants to merge 5 commits into
Draft
Conversation
l1n
added a commit
that referenced
this pull request
Mar 2, 2026
CopyObject API has a 5 GiB limit. For objects exceeding this, Copy now falls back to multipart copy using CreateMultipartUpload + UploadPartCopy + CompleteMultipartUpload. The fallback is triggered when CopyObject returns EntityTooLarge or InvalidRequest with a "copy source is larger" message (the actual error S3 returns in practice). Uses 5 GiB part size to minimize API calls since the copy is entirely server-side. Closes #29
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
just uses tempfiles for now. I thought about using either:
but the WriteAt interface makes this quite annoying as I want to make sure i start uploading after a specific subset of the file has been written to. I could make another struct that larps as a tempfile instead but no mem benefits only fewer context switches
i think the best way to do this in parallel is to go directly to the s3api layer and directly ListParts and then stream each part concurrently. that can be followup, this might be good enough to be usable anyways