Files
Oxicloud/src/infrastructure/services/zip_service.rs
T
Claude 944c833787 Sweep aborted-upload orphans from the periodic trash job; pipeline ZIP reads
Two follow-ups to the streaming-upload work:

The dedup garbage_collect() pass only ran when a user manually emptied
their trash. Zero-reference rows — chunks orphaned by aborted streaming
uploads (registered at ref_count 0 by the ingest rollback) and blobs
dereferenced by trash expiry itself — could linger indefinitely on
instances where nobody empties trash. The periodic TrashCleanupService
sweep now ends every run with garbage_collect() (maintenance pool,
batched), bounding orphan lifetime to the cleanup interval.

Folder-ZIP creation was strictly sequential: open blob stream, deflate,
close, repeat — every per-file blob-store round-trip (PG lookup + backend
open; a full HTTP round-trip on S3/Azure) added to the wall clock. It now
runs as a 2-stage pipeline: a prefetch task streams the planned files'
content ahead of the writer through a bounded channel (~4 MiB), so the
next file's read latency overlaps the current file's compression. ZIP
entries are still written strictly in order, peak RAM stays flat, and a
writer error hangs up the channel so the prefetcher stops on its own.

Verified end-to-end against PostgreSQL 16: an upload aborted at ~14 MB
left exactly 30 ref_count=0 chunk rows which the GC then reclaimed
(8.5 MB, rows + physical files); a chunked upload completed with a wrong
MD5 returned 400 with the tee-computed digest and the SAME session then
completed successfully with the right checksum (parts persist — the old
assembly deleted them, so the documented retry never actually worked);
a 4-file folder ZIP downloaded and extracted byte-identical.

https://claude.ai/code/session_01WdNenpnujNR2sc32XVvwfS
2026-06-11 13:19:29 +00:00

348 lines
13 KiB
Rust

use crate::application::services::file_retrieval_service::FileRetrievalService;
use crate::application::services::folder_service::FolderService;
use crate::{
application::dtos::file_dto::FileDto,
application::ports::file_ports::FileRetrievalUseCase,
application::ports::folder_ports::FolderUseCase,
application::ports::zip_ports::ZipPort,
common::errors::{DomainError, ErrorKind, Result},
};
use async_zip::base::write::ZipFileWriter;
use async_zip::{Compression, ZipEntryBuilder};
use futures::StreamExt;
use futures::io::AsyncWriteExt as FuturesWriteExt;
use std::collections::HashMap;
use std::sync::Arc;
use tempfile::NamedTempFile;
use thiserror::Error;
use tokio::io::BufWriter;
use tokio_util::compat::Compat;
use tracing::*;
/// Error related to ZIP file creation
#[derive(Debug, Error)]
pub enum ZipError {
#[error("IO error: {0}")]
IoError(#[from] std::io::Error),
#[error("ZIP error: {0}")]
AsyncZipError(#[from] async_zip::error::ZipError),
#[error("Error reading file: {0}")]
FileReadError(String),
#[error("Error getting folder contents: {0}")]
FolderContentsError(String),
#[error("Folder not found: {0}")]
FolderNotFound(String),
}
impl From<ZipError> for DomainError {
fn from(err: ZipError) -> Self {
DomainError::new(ErrorKind::InternalError, "zip_service", err.to_string())
}
}
/// Type alias for the fully-async ZIP writer backed by a buffered tokio file.
type AsyncZipWriter = ZipFileWriter<Compat<BufWriter<tokio::fs::File>>>;
/// One planned archive entry, in final ZIP order.
enum ZipPlanEntry {
/// Directory entry (Stored, zero-length body).
Dir(String),
/// File entry: ZIP-relative path + file id to stream from the blob store.
File { zip_path: String, file_id: String },
}
/// Message protocol from the prefetch task to the ZIP writer. For each
/// planned file, in order: zero or more `Chunk`s, then exactly one `End`;
/// `Err` aborts the whole archive.
enum Prefetched {
Chunk(bytes::Bytes),
End,
Err(String),
}
/// Bound on the prefetch channel (messages of ≤ ~64 KB blob-stream chunks):
/// ~4 MiB of read-ahead. Enough to hide the per-file open latency of the
/// blob store (PG lookup + backend round-trip — significant on S3/Azure)
/// behind the deflate of the previous entry, while keeping RAM flat.
const PREFETCH_BUFFER_CHUNKS: usize = 64;
/// Service for creating ZIP files.
///
/// Uses `async_zip` for fully-async archive creation. Every write (headers,
/// compressed chunk data, central directory) goes through
/// `tokio::io::BufWriter` → `tokio::fs::File`, so **no Tokio worker is ever
/// blocked** by disk I/O or compression.
///
/// Archive creation is a 2-stage pipeline: a prefetch task reads file
/// content from the blob store ahead of the writer, so the next file's
/// read latency overlaps the current file's compression instead of adding
/// to it. The ZIP entries themselves are still written strictly in order
/// (the format requires it).
pub struct ZipService {
file_service: Arc<FileRetrievalService>,
folder_service: Arc<FolderService>,
}
impl ZipService {
/// Creates a new instance of the ZIP service
pub fn new(
file_service: Arc<FileRetrievalService>,
folder_service: Arc<FolderService>,
) -> Self {
Self {
file_service,
folder_service,
}
}
/// Creates a ZIP file backed by a temporary file, containing the contents
/// of a folder and all its subfolders. Returns the `NamedTempFile` so the
/// caller can stream it and let the OS clean up on drop.
///
/// Uses **2 SQL queries** (ltree `<@`) to fetch the entire subtree instead
/// of the previous N+1 BFS traversal.
pub async fn create_folder_zip(
&self,
folder_id: &str,
folder_name: &str,
) -> Result<NamedTempFile> {
info!(
"Creating ZIP for folder: {} (ID: {})",
folder_name, folder_id
);
// Verify the folder exists and get its path for prefix stripping
let root_folder = match self.folder_service.get_folder(folder_id).await {
Ok(f) => f,
Err(e) => {
error!("Error getting folder {}: {}", folder_id, e);
return Err(ZipError::FolderNotFound(folder_id.to_string()).into());
}
};
// ── 1. Bulk-fetch folder tree (small — one entry per folder) ────
let all_folders = self
.folder_service
.list_subtree_folders(folder_id)
.await
.map_err(|e| ZipError::FolderContentsError(format!("subtree folders: {}", e)))?;
// ── 2. Stream files from DB cursor — O(1) per row ───────────────
let mut file_stream = self
.file_service
.stream_files_in_subtree(folder_id)
.await
.map_err(|e| ZipError::FolderContentsError(format!("subtree files: {}", e)))?;
// Group files by folder_id incrementally from the stream
let mut files_by_folder: HashMap<String, Vec<FileDto>> =
HashMap::with_capacity(all_folders.len());
while let Some(file) = file_stream.next().await {
let file =
file.map_err(|e| ZipError::FolderContentsError(format!("subtree file: {}", e)))?;
let fid = file.folder_id.clone().unwrap_or_default();
files_by_folder.entry(fid).or_default().push(file);
}
info!(
"ZIP subtree: {} folders, {} files",
all_folders.len(),
files_by_folder.values().map(|v| v.len()).sum::<usize>()
);
// ── 3. Build a mapping: folder_id → ZIP-relative path ────────────
//
// The root folder's DB path is e.g. "/users/alice/Documents".
// We want ZIP entries relative to `folder_name`, so we strip the
// root prefix and prepend `folder_name`.
let root_path = root_folder.path.trim_end_matches('/');
let folder_zip_path = |db_path: &str| -> String {
let db_path = db_path.trim_end_matches('/');
if db_path == root_path {
folder_name.to_string()
} else {
let suffix = db_path
.strip_prefix(root_path)
.unwrap_or(db_path)
.trim_start_matches('/');
format!("{}/{}", folder_name, suffix)
}
};
// ── 4. Plan the archive (folders are already sorted by path) ─────
let mut plan: Vec<ZipPlanEntry> = Vec::new();
for folder in &all_folders {
let zip_dir = format!("{}/", folder_zip_path(&folder.path));
plan.push(ZipPlanEntry::Dir(zip_dir.clone()));
if let Some(files) = files_by_folder.get(&folder.id) {
for file in files {
plan.push(ZipPlanEntry::File {
zip_path: format!("{}{}", zip_dir, file.name),
file_id: file.id.to_string(),
});
}
}
}
// ── 5. Open the temp file + ZIP writer ───────────────────────────
let temp = NamedTempFile::new().map_err(ZipError::IoError)?;
let tokio_file = tokio::fs::File::create(temp.path())
.await
.map_err(ZipError::IoError)?;
let buf_writer = BufWriter::with_capacity(256 * 1024, tokio_file);
let mut zip = ZipFileWriter::with_tokio(buf_writer);
// ── 6. Write entries: 2-stage pipeline ───────────────────────────
// The prefetch task reads blob streams for the planned files, in
// order, ahead of the writer — the next file's blob-store latency
// overlaps the current file's deflate. If the writer bails out,
// dropping the receiver makes the prefetcher's next send fail and
// it stops on its own.
let file_ids: Vec<String> = plan
.iter()
.filter_map(|entry| match entry {
ZipPlanEntry::File { file_id, .. } => Some(file_id.clone()),
ZipPlanEntry::Dir(_) => None,
})
.collect();
let (tx, mut rx) = tokio::sync::mpsc::channel::<Prefetched>(PREFETCH_BUFFER_CHUNKS);
let _prefetcher = tokio::spawn(Self::prefetch_files(
self.file_service.clone(),
file_ids,
tx,
));
for entry in &plan {
match entry {
ZipPlanEntry::Dir(zip_dir) => {
let dir_entry =
ZipEntryBuilder::new(zip_dir.clone().into(), Compression::Stored);
match zip.write_entry_whole(dir_entry, &[]).await {
Ok(()) => debug!("Folder added to ZIP: {}", zip_dir),
Err(e) => {
warn!("Could not add folder entry (may already exist): {}", e);
}
}
}
ZipPlanEntry::File { zip_path, .. } => {
Self::write_prefetched_file(&mut zip, zip_path, &mut rx).await?;
}
}
}
// ── 7. Finalize ──────────────────────────────────────────────────
let mut compat_writer = zip.close().await.map_err(ZipError::AsyncZipError)?;
compat_writer.close().await.map_err(ZipError::IoError)?;
Ok(temp)
}
/// Prefetch stage: streams each planned file's content from the blob
/// store, in plan order, into the bounded channel. Stops on the first
/// read error (after forwarding it) or when the writer hangs up.
async fn prefetch_files(
file_service: Arc<FileRetrievalService>,
file_ids: Vec<String>,
tx: tokio::sync::mpsc::Sender<Prefetched>,
) {
for file_id in file_ids {
let stream = match file_service.get_file_stream(&file_id).await {
Ok(s) => s,
Err(e) => {
error!("Error opening file stream {}: {}", file_id, e);
let _ = tx
.send(Prefetched::Err(format!(
"Error streaming file {}: {}",
file_id, e
)))
.await;
return;
}
};
let mut stream = std::pin::Pin::from(stream);
while let Some(chunk_result) = stream.next().await {
let message = match chunk_result {
Ok(bytes) => Prefetched::Chunk(bytes),
Err(e) => Prefetched::Err(format!("Error streaming file {}: {}", file_id, e)),
};
let abort = matches!(message, Prefetched::Err(_));
if tx.send(message).await.is_err() || abort {
return; // writer gone, or fatal read error forwarded
}
}
if tx.send(Prefetched::End).await.is_err() {
return; // writer gone
}
}
}
/// Writer stage: drains one file's prefetched chunks into a Deflate
/// ZIP entry. Peak memory stays bounded by the channel, independent
/// of individual file sizes.
async fn write_prefetched_file(
zip: &mut AsyncZipWriter,
zip_path: &str,
rx: &mut tokio::sync::mpsc::Receiver<Prefetched>,
) -> Result<()> {
info!("Adding file to ZIP: {}", zip_path);
let entry = ZipEntryBuilder::new(zip_path.to_string().into(), Compression::Deflate);
let mut entry_writer = zip
.write_entry_stream(entry)
.await
.map_err(ZipError::AsyncZipError)?;
loop {
match rx.recv().await {
Some(Prefetched::Chunk(bytes)) => {
entry_writer
.write_all(&bytes)
.await
.map_err(ZipError::IoError)?;
}
Some(Prefetched::End) => break,
Some(Prefetched::Err(message)) => {
// Close the partially-written entry before bailing out.
let _ = entry_writer.close().await;
return Err(ZipError::FileReadError(message).into());
}
None => {
let _ = entry_writer.close().await;
return Err(ZipError::FileReadError(format!(
"Prefetch stage ended unexpectedly while writing {}",
zip_path
))
.into());
}
}
}
// Finalize the entry (writes data descriptor with CRC + sizes)
entry_writer
.close()
.await
.map_err(ZipError::AsyncZipError)?;
debug!("File added to ZIP: {}", zip_path);
Ok(())
}
}
// ─── Port implementation ─────────────────────────────────────────────────────
impl ZipPort for ZipService {
async fn create_folder_zip(
&self,
folder_id: &str,
folder_name: &str,
) -> std::result::Result<NamedTempFile, DomainError> {
self.create_folder_zip(folder_id, folder_name).await
}
}