Skip to content

fix: perform streaming upload one chunk at a time #407

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 5 commits into from
Closed

fix: perform streaming upload one chunk at a time #407

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
8483f1d
fix: perform streaming upload one chunk at a time
kskalski Nov 6, 2024
a92d8b6
Response is already checked.
kskalski Nov 6, 2024
49a0bf7
Fix response_data var use.
kskalski Nov 6, 2024
8219fd8
Use to_string to get etag.
kskalski Nov 6, 2024
b3b5da1
Make part_number mut.
kskalski Nov 6, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
76 changes: 25 additions & 51 deletions s3/src/bucket.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1540,73 +1540,47 @@ impl Bucket {
let msg = self
.initiate_multipart_upload(s3_path, content_type)
.await?;
let path = msg.key;
let path = &msg.key;
let upload_id = &msg.upload_id;

let mut part_number: u32 = 0;
let mut etags = Vec::new();

// Collect request handles
let mut handles = vec![];
let mut total_size = 0;
loop {
let chunk = if part_number == 0 {
first_chunk.clone()
} else {
crate::utils::read_chunk_async(reader).await?
};
total_size += chunk.len();

let done = chunk.len() < CHUNK_SIZE;

// Start chunk upload
// Read chunks one by one making multi part request for each and gathering responses
let mut chunk = first_chunk;
let mut part_number = 0;
let mut parts = vec![];
while !chunk.is_empty() {
total_size += chunk.len();
part_number += 1;
handles.push(self.make_multipart_request(
&path,
chunk,

// Perform chunk upload and reading next chunk in parallel
let part_fut = self.make_multipart_request(
path,
std::mem::take(&mut chunk),
part_number,
upload_id,
content_type,
));

if done {
break;
}
}
);
let chunk_fut = crate::utils::read_chunk_async(reader);

// Wait for all chunks to finish (or fail)
let responses = futures::future::join_all(handles).await;
let (next_chunk, response) = futures::future::try_join(chunk_fut, part_fut).await?;
chunk = next_chunk;

for response in responses {
let response_data = response?;
if !(200..300).contains(&response_data.status_code()) {
// Analyze the response to fast fail bad upload
if !(200..300).contains(&response.status_code()) {
// if chunk upload failed - abort the upload
match self.abort_upload(&path, upload_id).await {
Ok(_) => {
return Err(error_from_response_data(response_data)?);
}
Err(error) => {
return Err(error);
}
}
self.abort_upload(&path, upload_id).await?;
return Err(error_from_response_data(response)?);
}

let etag = response_data.as_str()?;
etags.push(etag.to_string());
parts.push(Part {
etag: response.to_string()?,
Copy link

Copilot AI Sep 16, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The method call response.to_string()? appears incorrect. Based on the original code that used response_data.as_str()?, this should likely be response.as_str()?.to_string() to first convert to string slice then to owned String.

Suggested change
etag: response.to_string()?,
etag: response.as_str()?.to_string(),

Copilot uses AI. Check for mistakes.
part_number,
});
}

// Finish the upload
let inner_data = etags
.clone()
.into_iter()
.enumerate()
.map(|(i, x)| Part {
etag: x,
part_number: i as u32 + 1,
})
.collect::<Vec<Part>>();
let response_data = self
.complete_multipart_upload(&path, &msg.upload_id, inner_data)
.complete_multipart_upload(path, upload_id, parts)
.await?;

Ok(PutStreamResponse::new(
Expand Down