1 use crate::common::Common; 2 use crate::common_directory::Directories; 3 use crate::common_extensions::Extensions; 4 use crate::common_items::ExcludedItems; 5 use crate::common_messages::Messages; 6 use crate::common_traits::{DebugPrint, PrintResults, SaveResults}; 7 use crossbeam_channel::Receiver; 8 use humansize::{file_size_opts as options, FileSize}; 9 use std::collections::BTreeMap; 10 use std::ffi::OsStr; 11 use std::fs::{File, Metadata}; 12 use std::io::{BufWriter, Write}; 13 use std::path::PathBuf; 14 use std::sync::atomic::Ordering; 15 use std::sync::atomic::{AtomicBool, AtomicU64}; 16 use std::sync::Arc; 17 use std::thread::sleep; 18 use std::time::Duration; 19 use std::time::{SystemTime, UNIX_EPOCH}; 20 use std::{fs, thread}; 21 22 #[derive(Debug)] 23 pub struct ProgressData { 24 pub files_checked: usize, 25 } 26 27 #[derive(Clone)] 28 pub struct FileEntry { 29 pub path: PathBuf, 30 pub size: u64, 31 pub modified_date: u64, 32 } 33 34 #[derive(Eq, PartialEq, Clone, Debug)] 35 pub enum DeleteMethod { 36 None, 37 Delete, 38 } 39 /// Info struck with helpful information's about results 40 #[derive(Default)] 41 pub struct Info { 42 pub taken_space: u64, 43 pub number_of_real_files: usize, 44 } 45 46 impl Info { new() -> Self47 pub fn new() -> Self { 48 Default::default() 49 } 50 } 51 52 /// Struct with required information's to work 53 pub struct BigFile { 54 text_messages: Messages, 55 information: Info, 56 big_files: BTreeMap<u64, Vec<FileEntry>>, 57 excluded_items: ExcludedItems, 58 directories: Directories, 59 allowed_extensions: Extensions, 60 recursive_search: bool, 61 number_of_files_to_check: usize, 62 delete_method: DeleteMethod, 63 stopped_search: bool, 64 } 65 66 impl BigFile { new() -> Self67 pub fn new() -> Self { 68 Self { 69 text_messages: Default::default(), 70 information: Info::new(), 71 big_files: Default::default(), 72 excluded_items: ExcludedItems::new(), 73 directories: Directories::new(), 74 allowed_extensions: Extensions::new(), 75 recursive_search: true, 76 number_of_files_to_check: 50, 77 delete_method: DeleteMethod::None, 78 stopped_search: false, 79 } 80 } 81 find_big_files(&mut self, stop_receiver: Option<&Receiver<()>>, progress_sender: Option<&futures::channel::mpsc::UnboundedSender<ProgressData>>)82 pub fn find_big_files(&mut self, stop_receiver: Option<&Receiver<()>>, progress_sender: Option<&futures::channel::mpsc::UnboundedSender<ProgressData>>) { 83 self.optimize_directories(); 84 if !self.look_for_big_files(stop_receiver, progress_sender) { 85 self.stopped_search = true; 86 return; 87 } 88 self.delete_files(); 89 self.debug_print(); 90 } get_stopped_search(&self) -> bool91 pub fn get_stopped_search(&self) -> bool { 92 self.stopped_search 93 } 94 get_big_files(&self) -> &BTreeMap<u64, Vec<FileEntry>>95 pub const fn get_big_files(&self) -> &BTreeMap<u64, Vec<FileEntry>> { 96 &self.big_files 97 } 98 get_text_messages(&self) -> &Messages99 pub const fn get_text_messages(&self) -> &Messages { 100 &self.text_messages 101 } 102 get_information(&self) -> &Info103 pub const fn get_information(&self) -> &Info { 104 &self.information 105 } 106 set_delete_method(&mut self, delete_method: DeleteMethod)107 pub fn set_delete_method(&mut self, delete_method: DeleteMethod) { 108 self.delete_method = delete_method; 109 } 110 set_recursive_search(&mut self, recursive_search: bool)111 pub fn set_recursive_search(&mut self, recursive_search: bool) { 112 self.recursive_search = recursive_search; 113 } 114 115 /// List of allowed extensions, only files with this extensions will be checking if are duplicates set_allowed_extensions(&mut self, allowed_extensions: String)116 pub fn set_allowed_extensions(&mut self, allowed_extensions: String) { 117 self.allowed_extensions.set_allowed_extensions(allowed_extensions, &mut self.text_messages); 118 } 119 look_for_big_files(&mut self, stop_receiver: Option<&Receiver<()>>, progress_sender: Option<&futures::channel::mpsc::UnboundedSender<ProgressData>>) -> bool120 fn look_for_big_files(&mut self, stop_receiver: Option<&Receiver<()>>, progress_sender: Option<&futures::channel::mpsc::UnboundedSender<ProgressData>>) -> bool { 121 let start_time: SystemTime = SystemTime::now(); 122 let mut folders_to_check: Vec<PathBuf> = Vec::with_capacity(1024 * 2); // This should be small enough too not see to big difference and big enough to store most of paths without needing to resize vector 123 124 // Add root folders for finding 125 for id in &self.directories.included_directories { 126 folders_to_check.push(id.clone()); 127 } 128 129 //// PROGRESS THREAD START 130 const LOOP_DURATION: u32 = 200; //in ms 131 let progress_thread_run = Arc::new(AtomicBool::new(true)); 132 133 let atomic_file_counter = Arc::new(AtomicU64::new(0)); 134 135 let progress_thread_handle; 136 if let Some(progress_sender) = progress_sender { 137 let progress_send = progress_sender.clone(); 138 let progress_thread_run = progress_thread_run.clone(); 139 let atomic_file_counter = atomic_file_counter.clone(); 140 progress_thread_handle = thread::spawn(move || loop { 141 progress_send 142 .unbounded_send(ProgressData { 143 files_checked: atomic_file_counter.load(Ordering::Relaxed) as usize, 144 }) 145 .unwrap(); 146 if !progress_thread_run.load(Ordering::Relaxed) { 147 break; 148 } 149 sleep(Duration::from_millis(LOOP_DURATION as u64)); 150 }); 151 } else { 152 progress_thread_handle = thread::spawn(|| {}); 153 } 154 155 //// PROGRESS THREAD END 156 157 while !folders_to_check.is_empty() { 158 if stop_receiver.is_some() && stop_receiver.unwrap().try_recv().is_ok() { 159 // Be sure that every thread is closed 160 progress_thread_run.store(false, Ordering::Relaxed); 161 progress_thread_handle.join().unwrap(); 162 return false; 163 } 164 165 let current_folder = folders_to_check.pop().unwrap(); 166 let read_dir = match fs::read_dir(¤t_folder) { 167 Ok(t) => t, 168 Err(e) => { 169 self.text_messages.warnings.push(format!("Cannot open dir {}, reason {}", current_folder.display(), e)); 170 continue; 171 } // Permissions denied 172 }; 173 'dir: for entry in read_dir { 174 let entry_data = match entry { 175 Ok(t) => t, 176 Err(e) => { 177 self.text_messages.warnings.push(format!("Cannot read entry in dir {}, reason {}", current_folder.display(), e)); 178 continue; 179 } //Permissions denied 180 }; 181 let metadata: Metadata = match entry_data.metadata() { 182 Ok(t) => t, 183 Err(e) => { 184 self.text_messages.warnings.push(format!("Cannot read metadata in dir {}, reason {}", current_folder.display(), e)); 185 continue; 186 } //Permissions denied 187 }; 188 if metadata.is_dir() { 189 if !self.recursive_search { 190 continue; 191 } 192 193 let next_folder = current_folder.join(entry_data.file_name()); 194 if self.directories.is_excluded(&next_folder) || self.excluded_items.is_excluded(&next_folder) { 195 continue 'dir; 196 } 197 198 folders_to_check.push(next_folder); 199 } else if metadata.is_file() { 200 atomic_file_counter.fetch_add(1, Ordering::Relaxed); 201 // Extracting file extension 202 let file_extension = entry_data.path().extension().and_then(OsStr::to_str).map(str::to_lowercase); 203 204 // Checking allowed extensions 205 if !self.allowed_extensions.file_extensions.is_empty() { 206 let allowed = self.allowed_extensions.file_extensions.iter().map(|e| e.to_lowercase()).any(|e| file_extension == Some(e)); 207 if !allowed { 208 // Not an allowed extension, ignore it. 209 continue 'dir; 210 } 211 } 212 213 // Checking expressions 214 let current_file_name = current_folder.join(entry_data.file_name()); 215 if self.excluded_items.is_excluded(¤t_file_name) { 216 continue 'dir; 217 } 218 219 // Creating new file entry 220 let fe: FileEntry = FileEntry { 221 path: current_file_name.clone(), 222 size: metadata.len(), 223 modified_date: match metadata.modified() { 224 Ok(t) => match t.duration_since(UNIX_EPOCH) { 225 Ok(d) => d.as_secs(), 226 Err(_inspected) => { 227 self.text_messages.warnings.push(format!("File {} seems to be modified before Unix Epoch.", current_file_name.display())); 228 0 229 } 230 }, 231 Err(e) => { 232 self.text_messages.warnings.push(format!("Unable to get modification date from file {}, reason {}", current_file_name.display(), e)); 233 0 234 } 235 }, 236 }; 237 238 self.big_files.entry(metadata.len()).or_insert_with(Vec::new); 239 self.big_files.get_mut(&metadata.len()).unwrap().push(fe); 240 } 241 } 242 } 243 244 // End thread which send info to gui 245 progress_thread_run.store(false, Ordering::Relaxed); 246 progress_thread_handle.join().unwrap(); 247 248 // Extract n biggest files to new TreeMap 249 let mut new_map: BTreeMap<u64, Vec<FileEntry>> = Default::default(); 250 251 for (size, vector) in self.big_files.iter().rev() { 252 if self.information.number_of_real_files < self.number_of_files_to_check { 253 for file in vector { 254 if self.information.number_of_real_files < self.number_of_files_to_check { 255 new_map.entry(*size).or_insert_with(Vec::new); 256 new_map.get_mut(size).unwrap().push(file.clone()); 257 self.information.taken_space += size; 258 self.information.number_of_real_files += 1; 259 } else { 260 break; 261 } 262 } 263 } else { 264 break; 265 } 266 } 267 self.big_files = new_map; 268 269 Common::print_time(start_time, SystemTime::now(), "look_for_big_files".to_string()); 270 true 271 } 272 set_number_of_files_to_check(&mut self, number_of_files_to_check: usize)273 pub fn set_number_of_files_to_check(&mut self, number_of_files_to_check: usize) { 274 self.number_of_files_to_check = number_of_files_to_check; 275 } 276 277 /// Setting excluded items which needs to contains * wildcard 278 /// Are a lot of slower than absolute path, so it should be used to heavy set_excluded_items(&mut self, excluded_items: Vec<String>)279 pub fn set_excluded_items(&mut self, excluded_items: Vec<String>) { 280 self.excluded_items.set_excluded_items(excluded_items, &mut self.text_messages); 281 } 282 283 /// Remove unused entries when included or excluded overlaps with each other or are duplicated etc. optimize_directories(&mut self)284 fn optimize_directories(&mut self) { 285 self.directories.optimize_directories(self.recursive_search, &mut self.text_messages); 286 } 287 288 /// Setting included directories, at least one must be provided set_included_directory(&mut self, included_directory: Vec<PathBuf>)289 pub fn set_included_directory(&mut self, included_directory: Vec<PathBuf>) { 290 self.directories.set_included_directory(included_directory, &mut self.text_messages); 291 } 292 293 /// Setting absolute path to exclude set_excluded_directory(&mut self, excluded_directory: Vec<PathBuf>)294 pub fn set_excluded_directory(&mut self, excluded_directory: Vec<PathBuf>) { 295 self.directories.set_excluded_directory(excluded_directory, &mut self.text_messages); 296 } 297 298 /// Function to delete files, from filed Vector delete_files(&mut self)299 fn delete_files(&mut self) { 300 let start_time: SystemTime = SystemTime::now(); 301 302 match self.delete_method { 303 DeleteMethod::Delete => { 304 for vec_file_entry in self.big_files.values() { 305 for file_entry in vec_file_entry { 306 if fs::remove_file(file_entry.path.clone()).is_err() { 307 self.text_messages.warnings.push(file_entry.path.display().to_string()); 308 } 309 } 310 } 311 } 312 DeleteMethod::None => { 313 //Just do nothing 314 } 315 } 316 317 Common::print_time(start_time, SystemTime::now(), "delete_files".to_string()); 318 } 319 } 320 321 impl Default for BigFile { default() -> Self322 fn default() -> Self { 323 Self::new() 324 } 325 } 326 327 impl DebugPrint for BigFile { 328 #[allow(dead_code)] 329 #[allow(unreachable_code)] 330 /// Debugging printing - only available on debug build debug_print(&self)331 fn debug_print(&self) { 332 #[cfg(not(debug_assertions))] 333 { 334 return; 335 } 336 println!("---------------DEBUG PRINT---------------"); 337 println!("### Information's"); 338 339 println!("Errors size - {}", self.text_messages.errors.len()); 340 println!("Warnings size - {}", self.text_messages.warnings.len()); 341 println!("Messages size - {}", self.text_messages.messages.len()); 342 343 println!("### Other"); 344 println!("Big files size {} in {} groups", self.information.number_of_real_files, self.big_files.len()); 345 println!("Allowed extensions - {:?}", self.allowed_extensions.file_extensions); 346 println!("Excluded items - {:?}", self.excluded_items.items); 347 println!("Included directories - {:?}", self.directories.included_directories); 348 println!("Excluded directories - {:?}", self.directories.excluded_directories); 349 println!("Recursive search - {}", self.recursive_search.to_string()); 350 println!("Number of files to check - {:?}", self.number_of_files_to_check); 351 println!("-----------------------------------------"); 352 } 353 } 354 355 impl SaveResults for BigFile { 356 /// Saving results to provided file save_results_to_file(&mut self, file_name: &str) -> bool357 fn save_results_to_file(&mut self, file_name: &str) -> bool { 358 let start_time: SystemTime = SystemTime::now(); 359 let file_name: String = match file_name { 360 "" => "results.txt".to_string(), 361 k => k.to_string(), 362 }; 363 364 let file_handler = match File::create(&file_name) { 365 Ok(t) => t, 366 Err(e) => { 367 self.text_messages.errors.push(format!("Failed to create file {}, reason {}", file_name, e)); 368 return false; 369 } 370 }; 371 let mut writer = BufWriter::new(file_handler); 372 373 if let Err(e) = writeln!( 374 writer, 375 "Results of searching {:?} with excluded directories {:?} and excluded items {:?}", 376 self.directories.included_directories, self.directories.excluded_directories, self.excluded_items.items 377 ) { 378 self.text_messages.errors.push(format!("Failed to save results to file {}, reason {}", file_name, e)); 379 return false; 380 } 381 382 if self.information.number_of_real_files != 0 { 383 write!(writer, "{} the biggest files.\n\n", self.information.number_of_real_files).unwrap(); 384 385 for (size, files) in self.big_files.iter().rev() { 386 for file_entry in files { 387 writeln!(writer, "{} ({}) - {}", size.file_size(options::BINARY).unwrap(), size, file_entry.path.display()).unwrap(); 388 } 389 } 390 } else { 391 write!(writer, "Not found any files.").unwrap(); 392 } 393 Common::print_time(start_time, SystemTime::now(), "save_results_to_file".to_string()); 394 true 395 } 396 } 397 398 impl PrintResults for BigFile { print_results(&self)399 fn print_results(&self) { 400 let start_time: SystemTime = SystemTime::now(); 401 println!("Found {} files which take {}:", self.information.number_of_real_files, self.information.taken_space.file_size(options::BINARY).unwrap()); 402 for (size, vector) in self.big_files.iter().rev() { 403 // TODO Align all to same width 404 for entry in vector { 405 println!("{} ({} bytes) - {}", size.file_size(options::BINARY).unwrap(), size, entry.path.display()); 406 } 407 } 408 Common::print_time(start_time, SystemTime::now(), "print_entries".to_string()); 409 } 410 } 411