| 1 |
|
| 2 |
|
| 3 |
|
| 4 |
|
| 5 |
|
| 6 |
use std::path::{Path, PathBuf}; |
| 7 |
|
| 8 |
use tokio::fs; |
| 9 |
|
| 10 |
|
| 11 |
#[derive(Debug, Clone)] |
| 12 |
pub struct StagedFile { |
| 13 |
pub filename: String, |
| 14 |
pub size: u64, |
| 15 |
pub modified: std::time::SystemTime, |
| 16 |
|
| 17 |
pub classification: Option<FileClassification>, |
| 18 |
} |
| 19 |
|
| 20 |
|
| 21 |
#[derive(Debug, Clone, Copy, PartialEq, Eq)] |
| 22 |
pub struct FileClassification { |
| 23 |
pub item_type: &'static str, |
| 24 |
pub file_type: &'static str, |
| 25 |
pub content_type: &'static str, |
| 26 |
} |
| 27 |
|
| 28 |
|
| 29 |
pub const STAGING_QUOTA_BYTES: u64 = 1024 * 1024 * 1024; |
| 30 |
|
| 31 |
|
| 32 |
pub fn user_staging_dir(base: &Path, user_id: &str) -> PathBuf { |
| 33 |
base.join(user_id) |
| 34 |
} |
| 35 |
|
| 36 |
|
| 37 |
pub async fn list_staged_files(dir: &Path) -> Vec<StagedFile> { |
| 38 |
let mut files = Vec::new(); |
| 39 |
let Ok(mut entries) = fs::read_dir(dir).await else { |
| 40 |
return files; |
| 41 |
}; |
| 42 |
|
| 43 |
while let Ok(Some(entry)) = entries.next_entry().await { |
| 44 |
let Ok(metadata) = entry.metadata().await else { |
| 45 |
continue; |
| 46 |
}; |
| 47 |
if !metadata.is_file() { |
| 48 |
continue; |
| 49 |
} |
| 50 |
|
| 51 |
let filename = entry.file_name().to_string_lossy().to_string(); |
| 52 |
let ext = filename.rsplit('.').next().unwrap_or("").to_lowercase(); |
| 53 |
let classification = classify_extension(&ext); |
| 54 |
|
| 55 |
files.push(StagedFile { |
| 56 |
filename, |
| 57 |
size: metadata.len(), |
| 58 |
modified: metadata.modified().unwrap_or(std::time::SystemTime::UNIX_EPOCH), |
| 59 |
classification, |
| 60 |
}); |
| 61 |
} |
| 62 |
|
| 63 |
files.sort_by(|a, b| b.modified.cmp(&a.modified)); |
| 64 |
files |
| 65 |
} |
| 66 |
|
| 67 |
|
| 68 |
pub async fn staging_usage(dir: &Path) -> u64 { |
| 69 |
let mut total = 0u64; |
| 70 |
let Ok(mut entries) = fs::read_dir(dir).await else { |
| 71 |
return 0; |
| 72 |
}; |
| 73 |
|
| 74 |
while let Ok(Some(entry)) = entries.next_entry().await { |
| 75 |
if let Ok(metadata) = entry.metadata().await |
| 76 |
&& metadata.is_file() |
| 77 |
{ |
| 78 |
total += metadata.len(); |
| 79 |
} |
| 80 |
} |
| 81 |
|
| 82 |
total |
| 83 |
} |
| 84 |
|
| 85 |
|
| 86 |
pub async fn cleanup_stale(base: &Path, ttl: std::time::Duration) { |
| 87 |
let Ok(mut users) = fs::read_dir(base).await else { |
| 88 |
return; |
| 89 |
}; |
| 90 |
|
| 91 |
let now = std::time::SystemTime::now(); |
| 92 |
let mut removed = 0u64; |
| 93 |
|
| 94 |
while let Ok(Some(user_dir)) = users.next_entry().await { |
| 95 |
let Ok(metadata) = user_dir.metadata().await else { |
| 96 |
continue; |
| 97 |
}; |
| 98 |
if !metadata.is_dir() { |
| 99 |
continue; |
| 100 |
} |
| 101 |
|
| 102 |
let Ok(mut files) = fs::read_dir(user_dir.path()).await else { |
| 103 |
continue; |
| 104 |
}; |
| 105 |
|
| 106 |
let mut dir_empty = true; |
| 107 |
while let Ok(Some(file)) = files.next_entry().await { |
| 108 |
let Ok(fmeta) = file.metadata().await else { |
| 109 |
dir_empty = false; |
| 110 |
continue; |
| 111 |
}; |
| 112 |
|
| 113 |
let age = fmeta |
| 114 |
.modified() |
| 115 |
.ok() |
| 116 |
.and_then(|m| now.duration_since(m).ok()) |
| 117 |
.unwrap_or_default(); |
| 118 |
|
| 119 |
if age > ttl { |
| 120 |
if fs::remove_file(file.path()).await.is_ok() { |
| 121 |
removed += 1; |
| 122 |
} |
| 123 |
} else { |
| 124 |
dir_empty = false; |
| 125 |
} |
| 126 |
} |
| 127 |
|
| 128 |
|
| 129 |
if dir_empty { |
| 130 |
let _ = fs::remove_dir(user_dir.path()).await; |
| 131 |
} |
| 132 |
} |
| 133 |
|
| 134 |
if removed > 0 { |
| 135 |
tracing::info!(removed, "cleaned up stale staging files"); |
| 136 |
} |
| 137 |
} |
| 138 |
|
| 139 |
|
| 140 |
pub fn sanitize_filename(name: &str) -> String { |
| 141 |
|
| 142 |
let base = name.rsplit('/').next().unwrap_or(name); |
| 143 |
let base = base.rsplit('\\').next().unwrap_or(base); |
| 144 |
|
| 145 |
let sanitized: String = base |
| 146 |
.chars() |
| 147 |
.filter(|c| c.is_alphanumeric() || *c == '.' || *c == '-' || *c == '_' || *c == ' ') |
| 148 |
.collect(); |
| 149 |
|
| 150 |
let sanitized = sanitized.trim().to_string(); |
| 151 |
|
| 152 |
if sanitized.is_empty() || sanitized == "." || sanitized == ".." { |
| 153 |
return "upload".to_string(); |
| 154 |
} |
| 155 |
|
| 156 |
|
| 157 |
if sanitized.len() > 200 { |
| 158 |
sanitized[..200].to_string() |
| 159 |
} else { |
| 160 |
sanitized |
| 161 |
} |
| 162 |
} |
| 163 |
|
| 164 |
|
| 165 |
pub fn classify_extension(ext: &str) -> Option<FileClassification> { |
| 166 |
match ext { |
| 167 |
|
| 168 |
"mp3" => Some(FileClassification { |
| 169 |
item_type: "audio", |
| 170 |
file_type: "audio", |
| 171 |
content_type: "audio/mpeg", |
| 172 |
}), |
| 173 |
"wav" => Some(FileClassification { |
| 174 |
item_type: "audio", |
| 175 |
file_type: "audio", |
| 176 |
content_type: "audio/wav", |
| 177 |
}), |
| 178 |
"flac" => Some(FileClassification { |
| 179 |
item_type: "audio", |
| 180 |
file_type: "audio", |
| 181 |
content_type: "audio/flac", |
| 182 |
}), |
| 183 |
"ogg" => Some(FileClassification { |
| 184 |
item_type: "audio", |
| 185 |
file_type: "audio", |
| 186 |
content_type: "audio/ogg", |
| 187 |
}), |
| 188 |
"m4a" => Some(FileClassification { |
| 189 |
item_type: "audio", |
| 190 |
file_type: "audio", |
| 191 |
content_type: "audio/mp4", |
| 192 |
}), |
| 193 |
"aac" => Some(FileClassification { |
| 194 |
item_type: "audio", |
| 195 |
file_type: "audio", |
| 196 |
content_type: "audio/aac", |
| 197 |
}), |
| 198 |
|
| 199 |
"zip" => Some(FileClassification { |
| 200 |
item_type: "digital", |
| 201 |
file_type: "download", |
| 202 |
content_type: "application/zip", |
| 203 |
}), |
| 204 |
"dmg" => Some(FileClassification { |
| 205 |
item_type: "digital", |
| 206 |
file_type: "download", |
| 207 |
content_type: "application/x-apple-diskimage", |
| 208 |
}), |
| 209 |
"exe" => Some(FileClassification { |
| 210 |
item_type: "digital", |
| 211 |
file_type: "download", |
| 212 |
content_type: "application/vnd.microsoft.portable-executable", |
| 213 |
}), |
| 214 |
"appimage" => Some(FileClassification { |
| 215 |
item_type: "digital", |
| 216 |
file_type: "download", |
| 217 |
content_type: "application/octet-stream", |
| 218 |
}), |
| 219 |
"deb" => Some(FileClassification { |
| 220 |
item_type: "digital", |
| 221 |
file_type: "download", |
| 222 |
content_type: "application/vnd.debian.binary-package", |
| 223 |
}), |
| 224 |
|
| 225 |
"clap" => Some(FileClassification { |
| 226 |
item_type: "plugin", |
| 227 |
file_type: "download", |
| 228 |
content_type: "application/octet-stream", |
| 229 |
}), |
| 230 |
"vst3" => Some(FileClassification { |
| 231 |
item_type: "plugin", |
| 232 |
file_type: "download", |
| 233 |
content_type: "application/octet-stream", |
| 234 |
}), |
| 235 |
_ => None, |
| 236 |
} |
| 237 |
} |
| 238 |
|
| 239 |
|
| 240 |
|
| 241 |
pub fn derive_title(filename: &str) -> String { |
| 242 |
let without_ext = match filename.rfind('.') { |
| 243 |
Some(pos) if pos > 0 => &filename[..pos], |
| 244 |
_ => filename, |
| 245 |
}; |
| 246 |
|
| 247 |
without_ext |
| 248 |
.replace(['-', '_'], " ") |
| 249 |
.split_whitespace() |
| 250 |
.map(title_case_word) |
| 251 |
.collect::<Vec<_>>() |
| 252 |
.join(" ") |
| 253 |
} |
| 254 |
|
| 255 |
fn title_case_word(word: &str) -> String { |
| 256 |
let mut chars = word.chars(); |
| 257 |
match chars.next() { |
| 258 |
None => String::new(), |
| 259 |
Some(first) => { |
| 260 |
let mut s = first.to_uppercase().to_string(); |
| 261 |
s.extend(chars.map(|c| c.to_ascii_lowercase())); |
| 262 |
s |
| 263 |
} |
| 264 |
} |
| 265 |
} |
| 266 |
|
| 267 |
|
| 268 |
pub fn is_allowed_extension(ext: &str) -> bool { |
| 269 |
classify_extension(&ext.to_lowercase()).is_some() |
| 270 |
} |
| 271 |
|
| 272 |
|
| 273 |
pub fn format_bytes(bytes: u64) -> String { |
| 274 |
if bytes < 1024 { |
| 275 |
format!("{} B", bytes) |
| 276 |
} else if bytes < 1024 * 1024 { |
| 277 |
format!("{:.1} KB", bytes as f64 / 1024.0) |
| 278 |
} else if bytes < 1024 * 1024 * 1024 { |
| 279 |
format!("{:.1} MB", bytes as f64 / (1024.0 * 1024.0)) |
| 280 |
} else { |
| 281 |
format!("{:.2} GB", bytes as f64 / (1024.0 * 1024.0 * 1024.0)) |
| 282 |
} |
| 283 |
} |
| 284 |
|
| 285 |
#[cfg(test)] |
| 286 |
mod tests { |
| 287 |
use super::*; |
| 288 |
|
| 289 |
#[test] |
| 290 |
fn sanitize_filename_basic() { |
| 291 |
assert_eq!(sanitize_filename("song.mp3"), "song.mp3"); |
| 292 |
} |
| 293 |
|
| 294 |
#[test] |
| 295 |
fn sanitize_filename_path_traversal() { |
| 296 |
assert_eq!(sanitize_filename("../../etc/passwd"), "passwd"); |
| 297 |
assert_eq!(sanitize_filename("..\\..\\secret.txt"), "secret.txt"); |
| 298 |
} |
| 299 |
|
| 300 |
#[test] |
| 301 |
fn sanitize_filename_strips_special_chars() { |
| 302 |
assert_eq!(sanitize_filename("my<>file|name.zip"), "myfilename.zip"); |
| 303 |
} |
| 304 |
|
| 305 |
#[test] |
| 306 |
fn sanitize_filename_empty_and_dots() { |
| 307 |
assert_eq!(sanitize_filename(""), "upload"); |
| 308 |
assert_eq!(sanitize_filename("."), "upload"); |
| 309 |
assert_eq!(sanitize_filename(".."), "upload"); |
| 310 |
} |
| 311 |
|
| 312 |
#[test] |
| 313 |
fn sanitize_filename_length_limit() { |
| 314 |
let long_name = "a".repeat(300); |
| 315 |
assert_eq!(sanitize_filename(&long_name).len(), 200); |
| 316 |
} |
| 317 |
|
| 318 |
#[test] |
| 319 |
fn classify_extension_audio() { |
| 320 |
let c = classify_extension("mp3").unwrap(); |
| 321 |
assert_eq!(c.item_type, "audio"); |
| 322 |
assert_eq!(c.content_type, "audio/mpeg"); |
| 323 |
} |
| 324 |
|
| 325 |
#[test] |
| 326 |
fn classify_extension_digital() { |
| 327 |
let c = classify_extension("zip").unwrap(); |
| 328 |
assert_eq!(c.item_type, "digital"); |
| 329 |
assert_eq!(c.file_type, "download"); |
| 330 |
} |
| 331 |
|
| 332 |
#[test] |
| 333 |
fn classify_extension_unknown() { |
| 334 |
assert!(classify_extension("txt").is_none()); |
| 335 |
assert!(classify_extension("").is_none()); |
| 336 |
} |
| 337 |
|
| 338 |
#[test] |
| 339 |
fn derive_title_basic() { |
| 340 |
assert_eq!(derive_title("my-cool-song.mp3"), "My Cool Song"); |
| 341 |
assert_eq!(derive_title("hello_world.zip"), "Hello World"); |
| 342 |
} |
| 343 |
|
| 344 |
#[test] |
| 345 |
fn derive_title_no_extension() { |
| 346 |
assert_eq!(derive_title("readme"), "Readme"); |
| 347 |
} |
| 348 |
|
| 349 |
#[test] |
| 350 |
fn is_allowed_extension_cases() { |
| 351 |
assert!(is_allowed_extension("MP3")); |
| 352 |
assert!(is_allowed_extension("wav")); |
| 353 |
assert!(!is_allowed_extension("txt")); |
| 354 |
assert!(!is_allowed_extension("")); |
| 355 |
} |
| 356 |
|
| 357 |
#[test] |
| 358 |
fn format_bytes_ranges() { |
| 359 |
assert_eq!(format_bytes(0), "0 B"); |
| 360 |
assert_eq!(format_bytes(512), "512 B"); |
| 361 |
assert_eq!(format_bytes(1024), "1.0 KB"); |
| 362 |
assert_eq!(format_bytes(1_048_576), "1.0 MB"); |
| 363 |
assert_eq!(format_bytes(1_073_741_824), "1.00 GB"); |
| 364 |
} |
| 365 |
} |
| 366 |
|