New Version

Allow -n for number_of_lines to be used with -d 'max depth'

Remove depth specific functions, the job is now handled by the mainline

Add depth as a field onto the Node object.

.github/workflows/CICD.yml

@@ -81,6 +81,8 @@ jobs:
       matrix:
         job:
           # { os, target, cargo-options, features, use-cross, toolchain }
+          - { os: ubuntu-latest  , target: aarch64-unknown-linux-gnu   , use-cross: use-cross }
+          - { os: ubuntu-latest  , target: aarch64-unknown-linux-musl  , use-cross: use-cross }
           - { os: ubuntu-latest  , target: arm-unknown-linux-gnueabihf , use-cross: use-cross }
           - { os: ubuntu-20.04   , target: i686-unknown-linux-gnu      , use-cross: use-cross }
           - { os: ubuntu-20.04   , target: i686-unknown-linux-musl     , use-cross: use-cross }
@@ -99,6 +101,7 @@ jobs:
       run: |
         case ${{ matrix.job.target }} in
           arm-unknown-linux-gnueabihf) sudo apt-get -y update ; sudo apt-get -y install gcc-arm-linux-gnueabihf ;;
+          aarch64-unknown-linux-gnu) sudo apt-get -y update ; sudo apt-get -y install binutils-aarch64-linux-gnu ;;
         esac
     - name: Initialize workflow variables
       id: vars
@@ -134,7 +137,7 @@ jobs:
         echo ::set-output name=REF_TAG::${REF_TAG}
         echo ::set-output name=REF_SHAS::${REF_SHAS}
         # parse target
-        unset TARGET_ARCH ; case ${{ matrix.job.target }} in arm-unknown-linux-gnueabihf) TARGET_ARCH=arm ;; i686-*) TARGET_ARCH=i686 ;; x86_64-*) TARGET_ARCH=x86_64 ;; esac;
+        unset TARGET_ARCH ; case ${{ matrix.job.target }} in arm-unknown-linux-gnueabihf) TARGET_ARCH=arm ;; aarch-*) TARGET_ARCH=aarch64 ;; i686-*) TARGET_ARCH=i686 ;; x86_64-*) TARGET_ARCH=x86_64 ;; esac;
         echo set-output name=TARGET_ARCH::${TARGET_ARCH}
         echo ::set-output name=TARGET_ARCH::${TARGET_ARCH}
         unset TARGET_OS ; case ${{ matrix.job.target }} in *-linux-*) TARGET_OS=linux ;; *-apple-*) TARGET_OS=macos ;; *-windows-*) TARGET_OS=windows ;; esac;
@@ -166,16 +169,16 @@ jobs:
         echo ::set-output name=CARGO_USE_CROSS::${CARGO_USE_CROSS}
         # # * `arm` cannot be tested on ubuntu-* hosts (b/c testing is currently primarily done via comparison of target outputs with built-in outputs and the `arm` target is not executable on the host)
         JOB_DO_TESTING="true"
-        case ${{ matrix.job.target }} in arm-*) unset JOB_DO_TESTING ;; esac;
+        case ${{ matrix.job.target }} in arm-*|aarch64-*) unset JOB_DO_TESTING ;; esac;
         echo set-output name=JOB_DO_TESTING::${JOB_DO_TESTING:-<empty>/false}
         echo ::set-output name=JOB_DO_TESTING::${JOB_DO_TESTING}
         # # * test only binary for arm-type targets
         unset CARGO_TEST_OPTIONS
-        unset CARGO_TEST_OPTIONS ; case ${{ matrix.job.target }} in arm-*) CARGO_TEST_OPTIONS="--bin ${PROJECT_NAME}" ;; esac;
+        unset CARGO_TEST_OPTIONS ; case ${{ matrix.job.target }} in arm-*|aarch64-*) CARGO_TEST_OPTIONS="--bin ${PROJECT_NAME}" ;; esac;
         echo set-output name=CARGO_TEST_OPTIONS::${CARGO_TEST_OPTIONS}
         echo ::set-output name=CARGO_TEST_OPTIONS::${CARGO_TEST_OPTIONS}
         # * strip executable?
-        STRIP="strip" ; case ${{ matrix.job.target }} in arm-unknown-linux-gnueabihf) STRIP="arm-linux-gnueabihf-strip" ;; *-pc-windows-msvc) STRIP="" ;; esac;
+        STRIP="strip" ; case ${{ matrix.job.target }} in arm-unknown-linux-gnueabihf) STRIP="arm-linux-gnueabihf-strip" ;; *-pc-windows-msvc) STRIP="" ;; aarch64-unknown-linux-gnu) STRIP="aarch64-linux-gnu-strip" ;; aarch64-unknown-linux-musl) STRIP="" ;;esac;
         echo set-output name=STRIP::${STRIP}
         echo ::set-output name=STRIP::${STRIP}
     - name: Create all needed build/work directories
@@ -210,13 +213,13 @@ jobs:
       with:
           command: install
           args: cargo-deb
-      if: ${{ contains(matrix.job.target, 'musl') }}
+      if: matrix.job.target == 'i686-unknown-linux-musl' || matrix.job.target == 'x86_64-unknown-linux-musl'
     - name: Build deb
       uses: actions-rs/cargo@v1
       with:
           command: deb
           args: --no-build --target=${{ matrix.job.target }}
-      if: ${{ contains(matrix.job.target, 'musl') }}
+      if: matrix.job.target == 'i686-unknown-linux-musl' || matrix.job.target == 'x86_64-unknown-linux-musl'
     - name: Test
       uses: actions-rs/cargo@v1
       with:
@@ -233,7 +236,7 @@ jobs:
       with:
         name: ${{ env.PROJECT_NAME }}-${{ matrix.job.target }}.deb
         path: target/${{ matrix.job.target }}/debian
-      if: ${{ contains(matrix.job.target, 'musl') }}
+      if: matrix.job.target == 'i686-unknown-linux-musl' || matrix.job.target == 'x86_64-unknown-linux-musl'
     - name: Package
       shell: bash
       run: |

Cargo.lock

@@ -148,7 +148,7 @@ checksum = "fea41bba32d969b513997752735605054bc0dfa92b4c56bf1189f2e174be7a10"
 
 [[package]]
 name = "du-dust"
-version = "0.7.5"
+version = "0.8.1"
 dependencies = [
  "ansi_term",
  "assert_cmd",
@@ -340,9 +340,9 @@ dependencies = [
 
 [[package]]
 name = "regex"
-version = "1.5.4"
+version = "1.5.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d07a8629359eb56f1e2fb1652bb04212c072a87ba68546a04065d525673ac461"
+checksum = "1a11647b6b25ff05a515cb92c365cec08801e83423a235b51e231e1808747286"
 dependencies = [
  "aho-corasick",
  "memchr",

Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "du-dust"
 description = "A more intuitive version of du"
-version = "0.7.5"
+version = "0.8.1"
 authors = ["bootandy <bootandy@gmail.com>", "nebkor <code@ardent.nebcorp.com>"]
 edition = "2018"
 readme = "README.md"

README.md

@@ -32,7 +32,7 @@ Because I want an easy way to see where my disk is being used.
 
 #### Windows:
 * Windows GNU version - works
-* Windows MSVC - requires: VCRUNTIME140.dll
+* Windows MSVC - requires: [VCRUNTIME140.dll](https://docs.microsoft.com/en-gb/cpp/windows/latest-supported-vc-redist?view=msvc-170)
 
 #### Download
 

src/dir_walker.rs

@@ -34,7 +34,7 @@ pub fn walk_it(dirs: HashSet<PathBuf>, walk_data: WalkData) -> (Vec<Node>, bool)
     let top_level_nodes: Vec<_> = dirs
         .into_iter()
         .filter_map(|d| {
-            let n = walk(d, &permissions_flag, &walk_data);
+            let n = walk(d, &permissions_flag, &walk_data, 0);
             match n {
                 Some(n) => {
                     let mut inodes: HashSet<(u64, u64)> = HashSet::new();
@@ -73,6 +73,7 @@ fn clean_inodes(
         size: x.size + new_children.iter().map(|c| c.size).sum::<u64>(),
         children: new_children,
         inode_device: x.inode_device,
+        depth: x.depth,
     });
 }
 
@@ -108,7 +109,12 @@ fn ignore_file(entry: &DirEntry, walk_data: &WalkData) -> bool {
     (is_dot_file && walk_data.ignore_hidden) || is_ignored_path
 }
 
-fn walk(dir: PathBuf, permissions_flag: &AtomicBool, walk_data: &WalkData) -> Option<Node> {
+fn walk(
+    dir: PathBuf,
+    permissions_flag: &AtomicBool,
+    walk_data: &WalkData,
+    depth: usize,
+) -> Option<Node> {
     let mut children = vec![];
 
     if let Ok(entries) = fs::read_dir(dir.clone()) {
@@ -126,7 +132,7 @@ fn walk(dir: PathBuf, permissions_flag: &AtomicBool, walk_data: &WalkData) -> Op
                     if !ignore_file(entry, walk_data) {
                         if let Ok(data) = entry.file_type() {
                             if data.is_dir() && !data.is_symlink() {
-                                return walk(entry.path(), permissions_flag, walk_data);
+                                return walk(entry.path(), permissions_flag, walk_data, depth + 1);
                             }
                             return build_node(
                                 entry.path(),
@@ -137,6 +143,7 @@ fn walk(dir: PathBuf, permissions_flag: &AtomicBool, walk_data: &WalkData) -> Op
                                 data.is_symlink(),
                                 data.is_file(),
                                 walk_data.by_filecount,
+                                depth,
                             );
                         }
                     }
@@ -158,6 +165,7 @@ fn walk(dir: PathBuf, permissions_flag: &AtomicBool, walk_data: &WalkData) -> Op
         false,
         false,
         walk_data.by_filecount,
+        depth,
     )
 }
 
@@ -172,6 +180,7 @@ mod tests {
             size: 10,
             children: vec![],
             inode_device: Some((5, 6)),
+            depth: 0,
         }
     }
 

src/filter.rs

@@ -5,18 +5,10 @@ use std::collections::HashMap;
 use std::collections::HashSet;
 use std::path::PathBuf;
 
-pub fn get_by_depth(top_level_nodes: Vec<Node>, n: usize) -> Option<DisplayNode> {
-    if top_level_nodes.is_empty() {
-        // perhaps change this, bring back Error object?
-        return None;
-    }
-    let root = get_new_root(top_level_nodes);
-    Some(build_by_depth(&root, n - 1))
-}
-
 pub fn get_biggest(
     top_level_nodes: Vec<Node>,
     n: usize,
+    depth: usize,
     using_a_filter: bool,
 ) -> Option<DisplayNode> {
     if top_level_nodes.is_empty() {
@@ -30,14 +22,14 @@ pub fn get_biggest(
     let mut allowed_nodes = HashSet::new();
 
     allowed_nodes.insert(&root.name);
-    heap = add_children(using_a_filter, &root, heap);
+    heap = add_children(using_a_filter, &root, depth, heap);
 
     for _ in number_top_level_nodes..n {
         let line = heap.pop();
         match line {
             Some(line) => {
                 allowed_nodes.insert(&line.name);
-                heap = add_children(using_a_filter, line, heap);
+                heap = add_children(using_a_filter, line, depth, heap);
             }
             None => break,
         }
@@ -77,17 +69,20 @@ pub fn get_all_file_types(top_level_nodes: Vec<Node>, n: usize) -> Option<Displa
 
 fn add_children<'a>(
     using_a_filter: bool,
-    line: &'a Node,
+    file_or_folder: &'a Node,
+    depth: usize,
     mut heap: BinaryHeap<&'a Node>,
 ) -> BinaryHeap<&'a Node> {
-    if using_a_filter {
-        line.children.iter().for_each(|c| {
-            if c.name.is_file() || c.size > 0 {
-                heap.push(c)
-            }
-        });
-    } else {
-        line.children.iter().for_each(|c| heap.push(c));
+    if depth > file_or_folder.depth {
+        if using_a_filter {
+            file_or_folder.children.iter().for_each(|c| {
+                if c.name.is_file() || c.size > 0 {
+                    heap.push(c)
+                }
+            });
+        } else {
+            file_or_folder.children.iter().for_each(|c| heap.push(c));
+        }
     }
     heap
 }
@@ -111,29 +106,6 @@ fn build_by_all_file_types(top_level_nodes: Vec<Node>, counter: &mut HashMap<Str
     }
 }
 
-fn build_by_depth(node: &Node, depth: usize) -> DisplayNode {
-    let new_children = {
-        if depth == 0 {
-            vec![]
-        } else {
-            let mut new_children: Vec<_> = node
-                .children
-                .iter()
-                .map(|c| build_by_depth(c, depth - 1))
-                .collect();
-            new_children.sort();
-            new_children.reverse();
-            new_children
-        }
-    };
-
-    DisplayNode {
-        name: node.name.clone(),
-        size: node.size,
-        children: new_children,
-    }
-}
-
 fn get_new_root(top_level_nodes: Vec<Node>) -> Node {
     if top_level_nodes.len() > 1 {
         let total_size = top_level_nodes.iter().map(|node| node.size).sum();
@@ -142,6 +114,7 @@ fn get_new_root(top_level_nodes: Vec<Node>) -> Node {
             size: total_size,
             children: top_level_nodes,
             inode_device: None,
+            depth: 0,
         }
     } else {
         top_level_nodes.into_iter().next().unwrap()

src/main.rs

@@ -10,7 +10,7 @@ use self::display::draw_it;
 use clap::{crate_version, Arg};
 use clap::{Command, Values};
 use dir_walker::{walk_it, WalkData};
-use filter::{get_all_file_types, get_biggest, get_by_depth};
+use filter::{get_all_file_types, get_biggest};
 use regex::Regex;
 use std::cmp::max;
 use std::path::PathBuf;
@@ -99,9 +99,6 @@ fn get_regex_value(maybe_value: Option<Values>) -> Vec<Regex> {
 }
 
 fn main() {
-    let default_height = get_height_of_terminal();
-    let def_num_str = default_height.to_string();
-
     let options = Command::new("Dust")
         .about("Like du but more intuitive")
         .version(crate_version!())
@@ -112,7 +109,7 @@ fn main() {
                 .long("depth")
                 .help("Depth to show")
                 .takes_value(true)
-                .conflicts_with("number_of_lines"),
+                .default_value(usize::MAX.to_string().as_ref())
         )
         .arg(
             Arg::new("number_of_lines")
@@ -120,7 +117,6 @@ fn main() {
                 .long("number-of-lines")
                 .help("Number of lines of output to show. (Default is terminal_height - 10)")
                 .takes_value(true)
-                .default_value(def_num_str.as_ref()),
         )
         .arg(
             Arg::new("display_full_paths")
@@ -188,8 +184,7 @@ fn main() {
                 .multiple_occurrences(true)
                 .conflicts_with("filter")
                 .conflicts_with("types")
-                .conflicts_with("depth")
-                .help("Exclude files matching this regex. To ignore png files type: -v \"\\.png$\" "),
+                .help("Exclude filepaths matching this regex. To ignore png files type: -v \"\\.png$\" "),
         )
         .arg(
             Arg::new("filter")
@@ -199,8 +194,7 @@ fn main() {
                 .number_of_values(1)
                 .multiple_occurrences(true)
                 .conflicts_with("types")
-                .conflicts_with("depth")
-                .help("Only include files matching this regex. For png files type: -e \"\\.png$\" "),
+                .help("Only include filepaths matching this regex. For png files type: -e \"\\.png$\" "),
         )
         .arg(
             Arg::new("types")
@@ -236,26 +230,36 @@ fn main() {
     let filter_regexs = get_regex_value(options.values_of("filter"));
     let invert_filter_regexs = get_regex_value(options.values_of("invert_filter"));
 
-    let number_of_lines = match options.value_of_t("number_of_lines") {
-        Ok(v) => v,
-        Err(_) => {
-            eprintln!("Ignoring bad value for number_of_lines");
-            default_height
-        }
-    };
-
     let terminal_width = match options.value_of_t("width") {
         Ok(v) => v,
         Err(_) => get_width_of_terminal(),
     };
 
-    let depth = options.value_of("depth").and_then(|depth| {
-        depth
-            .parse::<usize>()
-            .map(|v| v + 1)
-            .map_err(|_| eprintln!("Ignoring bad value for depth"))
-            .ok()
-    });
+    let depth = match options.value_of_t("depth") {
+        Ok(v) => v,
+        Err(_) => {
+            eprintln!("Ignoring bad value for depth");
+            usize::MAX
+        }
+    };
+    // If depth is set we set the default number_of_lines to be max
+    // instead of screen height
+    let default_height = if depth != usize::MAX {
+        usize::MAX
+    } else {
+        get_height_of_terminal()
+    };
+
+    let number_of_lines = match options.value_of("number_of_lines") {
+        Some(v) => match v.parse::<usize>() {
+            Ok(num_lines) => num_lines,
+            Err(_) => {
+                eprintln!("Ignoring bad value for number_of_lines");
+                default_height
+            }
+        },
+        None => default_height,
+    };
 
     let no_colors = init_color(options.is_present("no_colors"));
     let use_apparent_size = options.is_present("display_apparent_size");
@@ -297,10 +301,10 @@ fn main() {
     let tree = {
         match (depth, summarize_file_types) {
             (_, true) => get_all_file_types(top_level_nodes, number_of_lines),
-            (Some(depth), _) => get_by_depth(top_level_nodes, depth),
-            (_, _) => get_biggest(
+            (depth, _) => get_biggest(
                 top_level_nodes,
                 number_of_lines,
+                depth,
                 options.values_of("filter").is_some()
                     || options.value_of("invert_filter").is_some(),
             ),

src/node.rs

@@ -12,6 +12,7 @@ pub struct Node {
     pub size: u64,
     pub children: Vec<Node>,
     pub inode_device: Option<(u64, u64)>,
+    pub depth: usize,
 }
 
 #[allow(clippy::too_many_arguments)]
@@ -24,6 +25,7 @@ pub fn build_node(
     is_symlink: bool,
     is_file: bool,
     by_filecount: bool,
+    depth: usize,
 ) -> Option<Node> {
     match get_metadata(&dir, use_apparent_size) {
         Some(data) => {
@@ -50,6 +52,7 @@ pub fn build_node(
                 size,
                 children,
                 inode_device,
+                depth,
             })
         }
         None => None,

src/platform.rs

@@ -114,8 +114,8 @@ pub fn get_metadata(d: &Path, _use_apparent_size: bool) -> Option<(u64, Option<(
 
             let attr_filtered = md.file_attributes()
                 & !(FILE_ATTRIBUTE_HIDDEN | FILE_ATTRIBUTE_READONLY | FILE_ATTRIBUTE_SYSTEM);
-            if attr_filtered == FILE_ATTRIBUTE_ARCHIVE
-                || attr_filtered == FILE_ATTRIBUTE_DIRECTORY
+            if (attr_filtered & FILE_ATTRIBUTE_ARCHIVE) != 0
+                || (attr_filtered & FILE_ATTRIBUTE_DIRECTORY) != 0
                 || md.file_attributes() == FILE_ATTRIBUTE_NORMAL
             {
                 Some((md.len(), None))

src/utils.rs

@@ -7,11 +7,10 @@ use regex::Regex;
 
 pub fn simplify_dir_names<P: AsRef<Path>>(filenames: Vec<P>) -> HashSet<PathBuf> {
     let mut top_level_names: HashSet<PathBuf> = HashSet::with_capacity(filenames.len());
-    let mut to_remove: Vec<PathBuf> = Vec::with_capacity(filenames.len());
-
     for t in filenames {
         let top_level_name = normalize_path(t);
         let mut can_add = true;
+        let mut to_remove: Vec<PathBuf> = Vec::new();
 
         for tt in top_level_names.iter() {
             if is_a_parent_of(&top_level_name, tt) {
@@ -20,14 +19,13 @@ pub fn simplify_dir_names<P: AsRef<Path>>(filenames: Vec<P>) -> HashSet<PathBuf>
                 can_add = false;
             }
         }
-        to_remove.sort_unstable();
-        top_level_names.retain(|tr| to_remove.binary_search(tr).is_err());
-        to_remove.clear();
+        for r in to_remove {
+            top_level_names.remove(&r);
+        }
         if can_add {
             top_level_names.insert(top_level_name);
         }
     }
-
     top_level_names
 }
 
@@ -94,6 +92,7 @@ mod tests {
     fn test_simplify_dir_rm_subdir() {
         let mut correct = HashSet::new();
         correct.insert(["a", "b"].iter().collect::<PathBuf>());
+        assert_eq!(simplify_dir_names(vec!["a/b/c", "a/b", "a/b/d/f"]), correct);
         assert_eq!(simplify_dir_names(vec!["a/b", "a/b/c", "a/b/d/f"]), correct);
     }