From 57cc94cd1456488fe42f23295c69fb1bee46337e Mon Sep 17 00:00:00 2001
From: Neur0toxine <pashok9825@gmail.com>
Date: Sun, 26 May 2024 14:28:42 +0300
Subject: [PATCH] cleanup for old files

---
 README.md     |  6 ++---
 src/main.rs   |  2 ++
 src/server.rs | 63 +++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 68 insertions(+), 3 deletions(-)

diff --git a/README.md b/README.md
index 12b22a7..2b18520 100644
--- a/README.md
+++ b/README.md
@@ -21,13 +21,13 @@ curl --location 'http://localhost:8090/enqueue' \
 3. You will receive JSON response with job ID. The transcoding result will be saved into `/tmp/{job_id}.out.atranscoder`
 
 # Roadmap
-- [ ] Implement acceptable error handling.
 - [ ] Restart threads in case of panic.
-- [ ] Remove old conversion results and input files every Nth hours.
+- [x] Implement somewhat acceptable error handling.
+- [x] Remove old conversion results and input files that are older than 1 hour.
 - [x] Remove input file after transcoding it.
 - [x] Implement file upload to `uploadUrl` (if `Content-Type: application/json` then conversion was not successful and body contains an error info).
 - [x] Remove transcoding result after uploading it to the `uploadUrl`.
 - [ ] (Optional) Make `uploadUrl` optional and allow the client to download the file on-demand.
 - [ ] Docker image for `amd64` and `aarch64`.
-- [ ] Statically linked binary for Docker image & result docker image based on `scratch` (reduce image size).
+- [ ] ~~Statically linked binary for Docker image & result docker image based on `scratch` (reduce image size).~~ Not yet, see [Dockerfile.scratch](Dockerfile.scratch).
 - [ ] Tests!
\ No newline at end of file
diff --git a/src/main.rs b/src/main.rs
index fcbe4cb..70f6c0c 100644
--- a/src/main.rs
+++ b/src/main.rs
@@ -1,4 +1,5 @@
 use std::env;
+use tracing::Instrument;
 
 use tracing_subscriber::EnvFilter;
 
@@ -32,6 +33,7 @@ async fn main() {
             .unwrap()
     });
     Server::new(pool, temp_dir)
+        .start_cleanup_task()
         .serve(&addr)
         .await
         .expect("Cannot bind the addr")
diff --git a/src/server.rs b/src/server.rs
index 730f80a..698666f 100644
--- a/src/server.rs
+++ b/src/server.rs
@@ -1,13 +1,18 @@
+use std::ffi::OsStr;
 use std::path::Path;
 use std::sync::Arc;
+use std::time::{Duration, SystemTime};
 
 use axum::extract::{DefaultBodyLimit, State};
 use axum::http::StatusCode;
 use axum::routing::post;
 use axum::{Json, Router};
 use axum_typed_multipart::TypedMultipart;
+use tokio::fs;
 use tokio::net::TcpListener;
+use tokio::time::interval;
 use tower_http::trace::TraceLayer;
+use tracing::{debug, error};
 use uuid::Uuid;
 
 use crate::dto::{ConvertRequest, ConvertResponse};
@@ -15,6 +20,7 @@ use crate::task::{Task, TaskParams};
 use crate::thread_pool::ThreadPool;
 
 const CONTENT_LENGTH_LIMIT: usize = 30 * 1024 * 1024;
+const WORK_DIR_IN_OUT_LIFETIME: u64 = 60 * 60;
 
 pub struct Server {
     thread_pool: Arc<ThreadPool>,
@@ -29,6 +35,21 @@ impl Server {
         }
     }
 
+    pub fn start_cleanup_task(self) -> Self {
+        let dir_path = self.work_dir.clone();
+        tokio::spawn(async move {
+            let mut interval = interval(Duration::from_secs(60));
+            loop {
+                interval.tick().await;
+
+                if let Err(err) = cleanup_directory(dir_path.as_str()).await {
+                    error!("could not perform working directory cleanup: {}", err);
+                }
+            }
+        });
+        self
+    }
+
     pub async fn serve(self, addr: &str) -> std::io::Result<()> {
         let this = Arc::new(self);
         let app = Router::new()
@@ -104,3 +125,45 @@ fn error_response(msg: &str) -> (StatusCode, Json<ConvertResponse>) {
         }),
     )
 }
+
+async fn cleanup_directory(dir_path: &str) -> Result<(), Box<dyn std::error::Error>> {
+    // Get the current time
+    let now = SystemTime::now();
+
+    // Read the directory
+    let mut entries = fs::read_dir(dir_path).await?;
+
+    // Iterate over directory entries
+    while let Some(entry) = entries.next_entry().await? {
+        let file_path = entry.path();
+
+        // Check if the entry is a file
+        if file_path.is_file() {
+            // Check if the file extension is ".atranscoder"
+            if let Some(extension) = file_path
+                .extension()
+                .and_then(OsStr::to_str)
+                .map(|ext| ext.to_lowercase())
+            {
+                if extension.eq("atranscoder") {
+                    // Get the metadata of the file
+                    let metadata = fs::metadata(&file_path).await?;
+
+                    // Get the last modified time of the file
+                    let modified_time = metadata.modified()?;
+
+                    // Calculate the duration since the last modification
+                    let duration_since_modified = now.duration_since(modified_time)?;
+
+                    // If the file is older than one hour, remove it
+                    if duration_since_modified > Duration::from_secs(WORK_DIR_IN_OUT_LIFETIME) {
+                        fs::remove_file(file_path.clone()).await?;
+                        debug!("removed file: {:?}", file_path);
+                    }
+                }
+            }
+        }
+    }
+
+    Ok(())
+}
\ No newline at end of file