#![crate_name = "uu_du"] /* * This file is part of the uutils coreutils package. * * (c) Derek Chiang * * For the full copyright and license information, please view the LICENSE * file that was distributed with this source code. */ extern crate time; #[macro_use] extern crate uucore; // XXX: remove when we no longer support 1.22.0 use std::ascii::AsciiExt; use std::collections::HashSet; use std::env; use std::fs; use std::io::{stderr, Result, Write}; use std::iter; use std::os::unix::fs::MetadataExt; use std::path::PathBuf; use time::Timespec; const NAME: &'static str = "du"; const SUMMARY: &'static str = "estimate file space usage"; const LONG_HELP: &'static str = " Display values are in units of the first available SIZE from --block-size, and the DU_BLOCK_SIZE, BLOCK_SIZE and BLOCKSIZE environ‐ ment variables. Otherwise, units default to 1024 bytes (or 512 if POSIXLY_CORRECT is set). SIZE is an integer and optional unit (example: 10M is 10*1024*1024). Units are K, M, G, T, P, E, Z, Y (powers of 1024) or KB, MB, ... (pow‐ ers of 1000). "; // TODO: Suport Z & Y (currently limited by size of u64) const UNITS: [(char, u32); 6] = [('E', 6), ('P', 5), ('T', 4), ('G', 3), ('M', 2), ('K', 1)]; struct Options { all: bool, program_name: String, max_depth: Option, total: bool, separate_dirs: bool, } struct Stat { path: PathBuf, is_dir: bool, size: u64, blocks: u64, nlink: u64, inode: u64, created: u64, accessed: u64, modified: u64, } impl Stat { fn new(path: PathBuf) -> Result { let metadata = fs::symlink_metadata(&path)?; Ok(Stat { path: path, is_dir: metadata.is_dir(), size: metadata.len(), blocks: metadata.blocks() as u64, nlink: metadata.nlink() as u64, inode: metadata.ino() as u64, created: metadata.mtime() as u64, accessed: metadata.atime() as u64, modified: metadata.mtime() as u64, }) } } fn unit_string_to_number(s: &str) -> Option { let mut offset = 0; let mut s_chars = s.chars().rev(); let (mut ch, multiple) = match s_chars.next() { Some('B') | Some('b') => ('B', 1000u64), Some(ch) => (ch, 1024u64), None => return None, }; if ch == 'B' { ch = s_chars.next()?; offset += 1; } ch = ch.to_ascii_uppercase(); let unit = UNITS .iter() .rev() .find(|&&(unit_ch, _)| unit_ch == ch) .map(|&(_, val)| { // we found a match, so increment offset offset += 1; val }) .or_else(|| if multiple == 1024 { Some(0) } else { None })?; let number = s[..s.len() - offset].parse::().ok()?; Some(number * multiple.pow(unit)) } fn translate_to_pure_number(s: &Option) -> Option { match s { &Some(ref s) => unit_string_to_number(s), &None => None, } } fn read_block_size(s: Option) -> u64 { match translate_to_pure_number(&s) { Some(v) => v, None => { if let Some(value) = s { show_error!("invalid --block-size argument '{}'", value); }; for env_var in &["DU_BLOCK_SIZE", "BLOCK_SIZE", "BLOCKSIZE"] { if let Some(quantity) = translate_to_pure_number(&env::var(env_var).ok()) { return quantity; } } if env::var("POSIXLY_CORRECT").is_ok() { 512 } else { 1024 } } } } // this takes `my_stat` to avoid having to stat files multiple times. // XXX: this should use the impl Trait return type when it is stabilized fn du( mut my_stat: Stat, options: &Options, depth: usize, inodes: &mut HashSet, ) -> Box> { let mut stats = vec![]; let mut futures = vec![]; if my_stat.is_dir { let read = match fs::read_dir(&my_stat.path) { Ok(read) => read, Err(e) => { safe_writeln!( stderr(), "{}: cannot read directory ‘{}‘: {}", options.program_name, my_stat.path.display(), e ); return Box::new(iter::once(my_stat)); } }; for f in read.into_iter() { match f { Ok(entry) => { match Stat::new(entry.path()) { Ok(this_stat) => { if this_stat.is_dir { futures.push(du(this_stat, options, depth + 1, inodes)); } else { if inodes.contains(&this_stat.inode) { continue; } inodes.insert(this_stat.inode); my_stat.size += this_stat.size; my_stat.blocks += this_stat.blocks; if options.all { stats.push(this_stat); } } } Err(error) => show_error!("{}", error), } } Err(error) => show_error!("{}", error), } } } stats.extend(futures.into_iter().flat_map(|val| val).rev().filter_map( |stat| { if !options.separate_dirs && stat.path.parent().unwrap() == my_stat.path { my_stat.size += stat.size; my_stat.blocks += stat.blocks; } if options.max_depth == None || depth < options.max_depth.unwrap() { Some(stat) } else { None } }, )); stats.push(my_stat); Box::new(stats.into_iter()) } fn convert_size_human(size: u64, multiplier: u64, _block_size: u64) -> String { for &(unit, power) in &UNITS { let limit = multiplier.pow(power); if size >= limit { return format!("{:.1}{}", (size as f64) / (limit as f64), unit); } } format!("{}B", size) } fn convert_size_k(size: u64, multiplier: u64, _block_size: u64) -> String { format!("{}", ((size as f64) / (multiplier as f64)).ceil()) } fn convert_size_m(size: u64, multiplier: u64, _block_size: u64) -> String { format!("{}", ((size as f64) / ((multiplier * multiplier) as f64)).ceil()) } fn convert_size_other(size: u64, _multiplier: u64, block_size: u64) -> String { format!("{}", ((size as f64) / (block_size as f64)).ceil()) } pub fn uumain(args: Vec) -> i32 { let syntax = format!( "[OPTION]... [FILE]... {0} [OPTION]... --files0-from=F", NAME ); let matches = new_coreopts!(&syntax, SUMMARY, LONG_HELP) // In task .optflag("a", "all", " write counts for all files, not just directories") // In main .optflag("", "apparent-size", "print apparent sizes, rather than disk usage although the apparent size is usually smaller, it may be larger due to holes in ('sparse') files, internal fragmentation, indirect blocks, and the like") // In main .optopt("B", "block-size", "scale sizes by SIZE before printing them. E.g., '-BM' prints sizes in units of 1,048,576 bytes. See SIZE format below.", "SIZE") // In main .optflag("b", "bytes", "equivalent to '--apparent-size --block-size=1'") // In main .optflag("c", "total", "produce a grand total") // In task // opts.optflag("D", "dereference-args", "dereference only symlinks that are listed // on the command line"), // In main // opts.optopt("", "files0-from", "summarize disk usage of the NUL-terminated file // names specified in file F; // If F is - then read names from standard input", "F"), // // In task // opts.optflag("H", "", "equivalent to --dereference-args (-D)"), // In main .optflag("h", "human-readable", "print sizes in human readable format (e.g., 1K 234M 2G)") // In main .optflag("", "si", "like -h, but use powers of 1000 not 1024") // In main .optflag("k", "", "like --block-size=1K") // In task .optflag("l", "count-links", "count sizes many times if hard linked") // // In main .optflag("m", "", "like --block-size=1M") // // In task // opts.optflag("L", "dereference", "dereference all symbolic links"), // // In task // opts.optflag("P", "no-dereference", "don't follow any symbolic links (this is the default)"), // // In main .optflag("0", "null", "end each output line with 0 byte rather than newline") // In main .optflag("S", "separate-dirs", "do not include size of subdirectories") // In main .optflag("s", "summarize", "display only a total for each argument") // // In task // opts.optflag("x", "one-file-system", "skip directories on different file systems"), // // In task // opts.optopt("X", "exclude-from", "exclude files that match any pattern in FILE", "FILE"), // // In task // opts.optopt("", "exclude", "exclude files that match PATTERN", "PATTERN"), // In main .optopt("d", "max-depth", "print the total for a directory (or file, with --all) only if it is N or fewer levels below the command line argument; --max-depth=0 is the same as --summarize", "N") // In main .optflagopt("", "time", "show time of the last modification of any file in the directory, or any of its subdirectories. If WORD is given, show time as WORD instead of modification time: atime, access, use, ctime or status", "WORD") // In main .optopt("", "time-style", "show times using style STYLE: full-iso, long-iso, iso, +FORMAT FORMAT is interpreted like 'date'", "STYLE") .parse(args); let summarize = matches.opt_present("summarize"); let max_depth_str = matches.opt_str("max-depth"); let max_depth = max_depth_str.as_ref().and_then(|s| s.parse::().ok()); match (max_depth_str, max_depth) { (Some(ref s), _) if summarize => { show_error!("summarizing conflicts with --max-depth={}", *s); return 1; } (Some(ref s), None) => { show_error!("invalid maximum depth '{}'", *s); return 1; } (Some(_), Some(_)) | (None, _) => { /* valid */ } } let options = Options { all: matches.opt_present("all"), program_name: NAME.to_owned(), max_depth: max_depth, total: matches.opt_present("total"), separate_dirs: matches.opt_present("S"), }; let strs = if matches.free.is_empty() { vec!["./".to_owned()] } else { matches.free.clone() }; let block_size = read_block_size(matches.opt_str("block-size")); let multiplier: u64 = if matches.opt_present("si") { 1000 } else { 1024 }; let convert_size_fn = { if matches.opt_present("human-readable") || matches.opt_present("si") { convert_size_human } else if matches.opt_present("k") { convert_size_k } else if matches.opt_present("m") { convert_size_m } else { convert_size_other } }; let convert_size = |size| convert_size_fn(size, multiplier, block_size); let time_format_str = match matches.opt_str("time-style") { Some(s) => { match &s[..] { "full-iso" => "%Y-%m-%d %H:%M:%S.%f %z", "long-iso" => "%Y-%m-%d %H:%M", "iso" => "%Y-%m-%d", _ => { show_error!( "invalid argument '{}' for 'time style' Valid arguments are: - 'full-iso' - 'long-iso' - 'iso' Try '{} --help' for more information.", s, NAME ); return 1; } } } None => "%Y-%m-%d %H:%M", }; let line_separator = if matches.opt_present("0") { "\0" } else { "\n" }; let mut grand_total = 0; for path_str in strs.into_iter() { let path = PathBuf::from(&path_str); match Stat::new(path) { Ok(stat) => { let mut inodes: HashSet = HashSet::new(); let iter = du(stat, &options, 0, &mut inodes).into_iter(); let (_, len) = iter.size_hint(); let len = len.unwrap(); for (index, stat) in iter.enumerate() { let size = if matches.opt_present("apparent-size") { stat.nlink * stat.size } else { // C's stat is such that each block is assume to be 512 bytes // See: http://linux.die.net/man/2/stat stat.blocks * 512 }; if matches.opt_present("time") { let tm = { let (secs, nsecs) = { let time = match matches.opt_str("time") { Some(s) => { match &s[..] { "accessed" => stat.accessed, "created" => stat.created, "modified" => stat.modified, _ => { show_error!( "invalid argument 'modified' for '--time' Valid arguments are: - 'accessed', 'created', 'modified' Try '{} --help' for more information.", NAME ); return 1; } } } None => stat.modified, }; ((time / 1000) as i64, (time % 1000 * 1000000) as i32) }; time::at(Timespec::new(secs, nsecs)) }; if !summarize || (summarize && index == len - 1) { let time_str = tm.strftime(time_format_str).unwrap(); print!( "{}\t{}\t{}{}", convert_size(size), time_str, stat.path.display(), line_separator ); } } else { if !summarize || (summarize && index == len - 1) { print!( "{}\t{}{}", convert_size(size), stat.path.display(), line_separator ); } } if options.total && index == (len - 1) { // The last element will be the total size of the the path under // path_str. We add it to the grand total. grand_total += size; } } } Err(_) => { show_error!("{}: {}", path_str, "No such file or directory"); } } } if options.total { print!("{}\ttotal", convert_size(grand_total)); print!("{}", line_separator); } 0 } #[cfg(test)] mod test_du { #[allow(unused_imports)] use super::*; #[test] fn test_translate_to_pure_number() { let test_data = [ (Some("10".to_string()), Some(10)), (Some("10K".to_string()), Some(10 * 1024)), (Some("5M".to_string()), Some(5 * 1024 * 1024)), (Some("900KB".to_string()), Some(900 * 1000)), (Some("BAD_STRING".to_string()), None), ]; for it in test_data.into_iter() { assert_eq!(translate_to_pure_number(&it.0), it.1); } } #[test] fn test_read_block_size() { let test_data = [ (Some("10".to_string()), 10), (None, 1024), (Some("BAD_STRING".to_string()), 1024), ]; for it in test_data.into_iter() { assert_eq!(read_block_size(it.0.clone()), it.1); } } }