Fix: Try to stop panics on android.

Catch panics on thread initialization.

.github/workflows/CICD.yml

@@ -0,0 +1,347 @@
+name: CICD
+
+# spell-checker:ignore CICD CODECOV MSVC MacOS Peltoche SHAs buildable clippy esac fakeroot gnueabihf halium libssl mkdir musl popd printf pushd rustfmt softprops toolchain
+
+env:
+  PROJECT_NAME: dust
+  PROJECT_DESC: "du + rust = dust"
+  PROJECT_AUTH: "bootandy"
+  RUST_MIN_SRV: "1.31.0"
+
+on: [push, pull_request]
+
+jobs:
+  style:
+    name: Style
+    runs-on: ${{ matrix.job.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        job:
+          - { os: ubuntu-latest }
+          - { os: macos-latest }
+          - { os: windows-latest }
+    steps:
+      - uses: actions/checkout@v1
+      - name: Initialize workflow variables
+        id: vars
+        shell: bash
+        run: |
+          # 'windows-latest' `cargo fmt` is bugged for this project (see reasons @ GH:rust-lang/rustfmt #3324, #3590, #3688 ; waiting for repair)
+          JOB_DO_FORMAT_TESTING="true"
+          case ${{ matrix.job.os }} in windows-latest) unset JOB_DO_FORMAT_TESTING ;; esac;
+          echo set-output name=JOB_DO_FORMAT_TESTING::${JOB_DO_FORMAT_TESTING:-<empty>/false}
+          echo ::set-output name=JOB_DO_FORMAT_TESTING::${JOB_DO_FORMAT_TESTING}
+          # target-specific options
+          # * CARGO_FEATURES_OPTION
+          CARGO_FEATURES_OPTION='' ;
+          if [ -n "${{ matrix.job.features }}" ]; then CARGO_FEATURES_OPTION='--features "${{ matrix.job.features }}"' ; fi
+          echo set-output name=CARGO_FEATURES_OPTION::${CARGO_FEATURES_OPTION}
+          echo ::set-output name=CARGO_FEATURES_OPTION::${CARGO_FEATURES_OPTION}
+      - name: Install `rust` toolchain
+        uses: actions-rs/toolchain@v1
+        with:
+          toolchain: stable
+          override: true
+          profile: minimal # minimal component installation (ie, no documentation)
+          components: rustfmt, clippy
+      - name: "`fmt` testing"
+        if: steps.vars.outputs.JOB_DO_FORMAT_TESTING
+        uses: actions-rs/cargo@v1
+        with:
+          command: fmt
+          args: --all -- --check
+      - name: "`clippy` testing"
+        if: success() || failure() # run regardless of prior step ("`fmt` testing") success/failure
+        uses: actions-rs/cargo@v1
+        with:
+          command: clippy
+          args: ${{ matrix.job.cargo-options }} ${{ steps.vars.outputs.CARGO_FEATURES_OPTION }} -- -D warnings
+
+  min_version:
+    name: MinSRV # Minimum supported rust version
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v1
+      - name: Install `rust` toolchain (v${{ env.RUST_MIN_SRV }})
+        uses: actions-rs/toolchain@v1
+        with:
+          toolchain: ${{ env.RUST_MIN_SRV }}
+          profile: minimal # minimal component installation (ie, no documentation)
+      - name: Test
+        uses: actions-rs/cargo@v1
+        with:
+          command: test
+
+  build:
+    name: Build
+    runs-on: ${{ matrix.job.os }}
+    strategy:
+      fail-fast: false
+      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-latest,
+              target: i686-unknown-linux-gnu,
+              use-cross: use-cross,
+            }
+          - {
+              os: ubuntu-latest,
+              target: i686-unknown-linux-musl,
+              use-cross: use-cross,
+            }
+          - {
+              os: ubuntu-latest,
+              target: x86_64-unknown-linux-gnu,
+              use-cross: use-cross,
+            }
+          - {
+              os: ubuntu-latest,
+              target: x86_64-unknown-linux-musl,
+              use-cross: use-cross,
+            }
+          - { os: macos-latest, target: x86_64-apple-darwin }
+          - { os: windows-latest, target: i686-pc-windows-gnu }
+          - { os: windows-latest, target: i686-pc-windows-msvc }
+          - { os: windows-latest, target: x86_64-pc-windows-gnu } ## !maint: [rivy; 2020-01-21] may break due to rust bug; follow possible solution from GH:rust-lang/rust#47048 (refs: GH:rust-lang/rust#47048 , GH:rust-lang/rust#53454 , GH:bike-barn/hermit#172 )
+          - { os: windows-latest, target: x86_64-pc-windows-msvc }
+    steps:
+      - uses: actions/checkout@v1
+      - name: Install any prerequisites
+        shell: bash
+        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
+        shell: bash
+        run: |
+          # toolchain
+          TOOLCHAIN="stable" ## default to "stable" toolchain
+          # * specify alternate TOOLCHAIN for *-pc-windows-gnu targets; gnu targets on Windows are broken for the standard *-pc-windows-msvc toolchain (refs: <https://github.com/rust-lang/rust/issues/47048>, <https://github.com/rust-lang/rust/issues/53454>, <https://github.com/rust-lang/cargo/issues/6754>)
+          case ${{ matrix.job.target }} in *-pc-windows-gnu) TOOLCHAIN="stable-${{ matrix.job.target }}" ;; esac;
+          # * use requested TOOLCHAIN if specified
+          if [ -n "${{ matrix.job.toolchain }}" ]; then TOOLCHAIN="${{ matrix.job.toolchain }}" ; fi
+          echo set-output name=TOOLCHAIN::${TOOLCHAIN}
+          echo ::set-output name=TOOLCHAIN::${TOOLCHAIN}
+          # staging directory
+          STAGING='_staging'
+          echo set-output name=STAGING::${STAGING}
+          echo ::set-output name=STAGING::${STAGING}
+          # determine EXE suffix
+          EXE_suffix="" ; case ${{ matrix.job.target }} in *-pc-windows-*) EXE_suffix=".exe" ;; esac;
+          echo set-output name=EXE_suffix::${EXE_suffix}
+          echo ::set-output name=EXE_suffix::${EXE_suffix}
+          # parse commit reference info
+          REF_NAME=${GITHUB_REF#refs/*/}
+          unset REF_BRANCH ; case ${GITHUB_REF} in refs/heads/*) REF_BRANCH=${GITHUB_REF#refs/heads/} ;; esac;
+          unset REF_TAG ; case ${GITHUB_REF} in refs/tags/*) REF_TAG=${GITHUB_REF#refs/tags/} ;; esac;
+          REF_SHAS=${GITHUB_SHA:0:8}
+          echo set-output name=REF_NAME::${REF_NAME}
+          echo set-output name=REF_BRANCH::${REF_BRANCH}
+          echo set-output name=REF_TAG::${REF_TAG}
+          echo set-output name=REF_SHAS::${REF_SHAS}
+          echo ::set-output name=REF_NAME::${REF_NAME}
+          echo ::set-output name=REF_BRANCH::${REF_BRANCH}
+          echo ::set-output name=REF_TAG::${REF_TAG}
+          echo ::set-output name=REF_SHAS::${REF_SHAS}
+          # parse target
+          unset TARGET_ARCH ; case ${{ matrix.job.target }} in arm-unknown-linux-gnueabihf) TARGET_ARCH=arm ;; aarch-*) TARGET_ARCH=aarch64 ;; i686-*) TARGET_ARCH=i686 ;; x86_64-*) TARGET_ARCH=x86_64 ;; esac;
+          echo set-output name=TARGET_ARCH::${TARGET_ARCH}
+          echo ::set-output name=TARGET_ARCH::${TARGET_ARCH}
+          unset TARGET_OS ; case ${{ matrix.job.target }} in *-linux-*) TARGET_OS=linux ;; *-apple-*) TARGET_OS=macos ;; *-windows-*) TARGET_OS=windows ;; esac;
+          echo set-output name=TARGET_OS::${TARGET_OS}
+          echo ::set-output name=TARGET_OS::${TARGET_OS}
+          # package name
+          PKG_suffix=".tar.gz" ; case ${{ matrix.job.target }} in *-pc-windows-*) PKG_suffix=".zip" ;; esac;
+          PKG_BASENAME=${PROJECT_NAME}-${REF_TAG:-$REF_SHAS}-${{ matrix.job.target }}
+          PKG_NAME=${PKG_BASENAME}${PKG_suffix}
+          echo set-output name=PKG_suffix::${PKG_suffix}
+          echo set-output name=PKG_BASENAME::${PKG_BASENAME}
+          echo set-output name=PKG_NAME::${PKG_NAME}
+          echo ::set-output name=PKG_suffix::${PKG_suffix}
+          echo ::set-output name=PKG_BASENAME::${PKG_BASENAME}
+          echo ::set-output name=PKG_NAME::${PKG_NAME}
+          # deployable tag? (ie, leading "vM" or "M"; M == version number)
+          unset DEPLOY ; if [[ $REF_TAG =~ ^[vV]?[0-9].* ]]; then DEPLOY='true' ; fi
+          echo set-output name=DEPLOY::${DEPLOY:-<empty>/false}
+          echo ::set-output name=DEPLOY::${DEPLOY}
+          # target-specific options
+          # * CARGO_FEATURES_OPTION
+          CARGO_FEATURES_OPTION='' ;
+          if [ -n "${{ matrix.job.features }}" ]; then CARGO_FEATURES_OPTION='--features "${{ matrix.job.features }}"' ; fi
+          echo set-output name=CARGO_FEATURES_OPTION::${CARGO_FEATURES_OPTION}
+          echo ::set-output name=CARGO_FEATURES_OPTION::${CARGO_FEATURES_OPTION}
+          # * CARGO_USE_CROSS (truthy)
+          CARGO_USE_CROSS='true' ; case '${{ matrix.job.use-cross }}' in ''|0|f|false|n|no) unset CARGO_USE_CROSS ;; esac;
+          echo set-output name=CARGO_USE_CROSS::${CARGO_USE_CROSS:-<empty>/false}
+          echo ::set-output name=CARGO_USE_CROSS::${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 ${{ matrix.job.target }} in arm-*|aarch64-*) unset JOB_DO_TESTING ;; esac;
+          echo set-output name=JOB_DO_TESTING::${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-*|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 ${{ matrix.job.target }} in arm-unknown-linux-gnueabihf) STRIP="arm-linux-gnueabihf-strip" ;; *-pc-windows-msvc) STRIP="" ;; aarch64-unknown-linux-gnu) STRIP="aarch64-linux-gnu-strip" ;; aarch64-unknown-linux-musl) STRIP="" ;;esac;
+          echo set-output name=STRIP::${STRIP}
+          echo ::set-output name=STRIP::${STRIP}
+      - name: Create all needed build/work directories
+        shell: bash
+        run: |
+          mkdir -p '${{ steps.vars.outputs.STAGING }}'
+          mkdir -p '${{ steps.vars.outputs.STAGING }}/${{ steps.vars.outputs.PKG_BASENAME }}'
+      - name: rust toolchain ~ install
+        uses: actions-rs/toolchain@v1
+        with:
+          toolchain: ${{ steps.vars.outputs.TOOLCHAIN }}
+          target: ${{ matrix.job.target }}
+          override: true
+          profile: minimal # minimal component installation (ie, no documentation)
+      - name: Info
+        shell: bash
+        run: |
+          gcc --version || true
+          rustup -V
+          rustup toolchain list
+          rustup default
+          cargo -V
+          rustc -V
+      - name: Build
+        uses: actions-rs/cargo@v1
+        with:
+          use-cross: ${{ steps.vars.outputs.CARGO_USE_CROSS }}
+          command: build
+          args: --release --target=${{ matrix.job.target }} ${{ matrix.job.cargo-options }} ${{ steps.vars.outputs.CARGO_FEATURES_OPTION }}
+      - name: Install cargo-deb
+        uses: actions-rs/cargo@v1
+        with:
+          command: install
+          args: cargo-deb
+        if: matrix.job.target == 'i686-unknown-linux-musl' || matrix.job.target == 'x86_64-unknown-linux-musl'
+      - name: Build deb
+        uses: actions-rs/cargo@v1
+        with:
+          command: deb
+          args: --no-build --target=${{ matrix.job.target }}
+        if: matrix.job.target == 'i686-unknown-linux-musl' || matrix.job.target == 'x86_64-unknown-linux-musl'
+      - name: Test
+        uses: actions-rs/cargo@v1
+        with:
+          use-cross: ${{ steps.vars.outputs.CARGO_USE_CROSS }}
+          command: test
+          args: --target=${{ matrix.job.target }} ${{ steps.vars.outputs.CARGO_TEST_OPTIONS}} ${{ matrix.job.cargo-options }} ${{ steps.vars.outputs.CARGO_FEATURES_OPTION }}
+      - name: Archive executable artifacts
+        uses: actions/upload-artifact@master
+        with:
+          name: ${{ env.PROJECT_NAME }}-${{ matrix.job.target }}
+          path: target/${{ matrix.job.target }}/release/${{ env.PROJECT_NAME }}${{ steps.vars.outputs.EXE_suffix }}
+      - name: Archive deb artifacts
+        uses: actions/upload-artifact@master
+        with:
+          name: ${{ env.PROJECT_NAME }}-${{ matrix.job.target }}.deb
+          path: target/${{ matrix.job.target }}/debian
+        if: matrix.job.target == 'i686-unknown-linux-musl' || matrix.job.target == 'x86_64-unknown-linux-musl'
+      - name: Package
+        shell: bash
+        run: |
+          # binary
+          cp 'target/${{ matrix.job.target }}/release/${{ env.PROJECT_NAME }}${{ steps.vars.outputs.EXE_suffix }}' '${{ steps.vars.outputs.STAGING }}/${{ steps.vars.outputs.PKG_BASENAME }}/'
+          # `strip` binary (if needed)
+          if [ -n "${{ steps.vars.outputs.STRIP }}" ]; then "${{ steps.vars.outputs.STRIP }}" '${{ steps.vars.outputs.STAGING }}/${{ steps.vars.outputs.PKG_BASENAME }}/${{ env.PROJECT_NAME }}${{ steps.vars.outputs.EXE_suffix }}' ; fi
+          # README and LICENSE
+          cp README.md '${{ steps.vars.outputs.STAGING }}/${{ steps.vars.outputs.PKG_BASENAME }}/'
+          cp LICENSE '${{ steps.vars.outputs.STAGING }}/${{ steps.vars.outputs.PKG_BASENAME }}/'
+          # base compressed package
+          pushd '${{ steps.vars.outputs.STAGING }}/' >/dev/null
+          case ${{ matrix.job.target }} in
+            *-pc-windows-*) 7z -y a '${{ steps.vars.outputs.PKG_NAME }}' '${{ steps.vars.outputs.PKG_BASENAME }}'/* | tail -2 ;;
+            *) tar czf '${{ steps.vars.outputs.PKG_NAME }}' '${{ steps.vars.outputs.PKG_BASENAME }}'/* ;;
+          esac;
+          popd >/dev/null
+      - name: Publish
+        uses: softprops/action-gh-release@v1
+        if: steps.vars.outputs.DEPLOY
+        with:
+          files: |
+            ${{ steps.vars.outputs.STAGING }}/${{ steps.vars.outputs.PKG_NAME }}
+            target/${{ matrix.job.target }}/debian/*.deb
+
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+
+  ## fix! [rivy; 2020-22-01] `cargo tarpaulin` is unable to test this repo at the moment; alternate recipe or another testing framework?
+  # coverage:
+  #   name: Code Coverage
+  #   runs-on: ${{ matrix.job.os }}
+  #   strategy:
+  #     fail-fast: true
+  #     matrix:
+  #       # job: [ { os: ubuntu-latest }, { os: macos-latest }, { os: windows-latest } ]
+  #       job: [ { os: ubuntu-latest } ] ## cargo-tarpaulin is currently only available on linux
+  #   steps:
+  #   - uses: actions/checkout@v1
+  #   # - name: Reattach HEAD ## may be needed for accurate code coverage info
+  #   #   run: git checkout ${{ github.head_ref }}
+  #   - name: Initialize workflow variables
+  #     id: vars
+  #     shell: bash
+  #     run: |
+  #       # staging directory
+  #       STAGING='_staging'
+  #       echo set-output name=STAGING::${STAGING}
+  #       echo ::set-output name=STAGING::${STAGING}
+  #       # check for CODECOV_TOKEN availability (work-around for inaccessible 'secrets' object for 'if'; see <https://github.community/t5/GitHub-Actions/jobs-lt-job-id-gt-if-does-not-work-with-env-secrets/m-p/38549>)
+  #       unset HAS_CODECOV_TOKEN
+  #       if [ -n $CODECOV_TOKEN ]; then HAS_CODECOV_TOKEN='true' ; fi
+  #       echo set-output name=HAS_CODECOV_TOKEN::${HAS_CODECOV_TOKEN}
+  #       echo ::set-output name=HAS_CODECOV_TOKEN::${HAS_CODECOV_TOKEN}
+  #     env:
+  #       CODECOV_TOKEN: "${{ secrets.CODECOV_TOKEN }}"
+  #   - name: Create all needed build/work directories
+  #     shell: bash
+  #     run: |
+  #       mkdir -p '${{ steps.vars.outputs.STAGING }}/work'
+  #   - name: Install required packages
+  #     run: |
+  #       sudo apt-get -y install libssl-dev
+  #       pushd '${{ steps.vars.outputs.STAGING }}/work' >/dev/null
+  #       wget --no-verbose https://github.com/xd009642/tarpaulin/releases/download/0.9.3/cargo-tarpaulin-0.9.3-travis.tar.gz
+  #       tar xf cargo-tarpaulin-0.9.3-travis.tar.gz
+  #       cp cargo-tarpaulin "$(dirname -- "$(which cargo)")"/
+  #       popd >/dev/null
+  #   - name: Generate coverage
+  #     run: |
+  #       cargo tarpaulin --out Xml
+  #   - name: Upload coverage results (CodeCov.io)
+  #     # CODECOV_TOKEN (aka, "Repository Upload Token" for REPO from CodeCov.io) ## set via REPO/Settings/Secrets
+  #     # if: secrets.CODECOV_TOKEN (not supported {yet?}; see <https://github.community/t5/GitHub-Actions/jobs-lt-job-id-gt-if-does-not-work-with-env-secrets/m-p/38549>)
+  #     if: steps.vars.outputs.HAS_CODECOV_TOKEN
+  #     run: |
+  #       # CodeCov.io
+  #       cargo tarpaulin --out Xml
+  #       bash <(curl -s https://codecov.io/bash)
+  #     env:
+  #       CODECOV_TOKEN: "${{ secrets.CODECOV_TOKEN }}"

.gitignore

@@ -2,10 +2,8 @@
 # will have compiled files and executables
 /target/
 
-# Remove Cargo.lock from gitignore if creating an executable, leave it for libraries
-# More information here http://doc.crates.io/guide.html#cargotoml-vs-cargolock
-Cargo.lock
-
 # These are backup files generated by rustfmt
 **/*.rs.bk
 *.swp
+.vscode/*
+*.idea/*

Cargo.lock

@@ -0,0 +1,657 @@
+# This file is automatically @generated by Cargo.
+# It is not intended for manual editing.
+version = 3
+
+[[package]]
+name = "aho-corasick"
+version = "0.7.18"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1e37cfd5e7657ada45f742d6e99ca5788580b5c529dc78faf11ece6dc702656f"
+dependencies = [
+ "memchr",
+]
+
+[[package]]
+name = "ansi_term"
+version = "0.12.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d52a9bb7ec0cf484c551830a7ce27bd20d67eac647e1befb56b0be4ee39a55d2"
+dependencies = [
+ "winapi",
+]
+
+[[package]]
+name = "assert_cmd"
+version = "1.0.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c98233c6673d8601ab23e77eb38f999c51100d46c5703b17288c57fddf3a1ffe"
+dependencies = [
+ "bstr",
+ "doc-comment",
+ "predicates",
+ "predicates-core",
+ "predicates-tree",
+ "wait-timeout",
+]
+
+[[package]]
+name = "atty"
+version = "0.2.14"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d9b39be18770d11421cdb1b9947a45dd3f37e93092cbf377614828a319d5fee8"
+dependencies = [
+ "hermit-abi",
+ "libc",
+ "winapi",
+]
+
+[[package]]
+name = "autocfg"
+version = "1.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d468802bab17cbc0cc575e9b053f41e72aa36bfa6b7f55e3529ffa43161b97fa"
+
+[[package]]
+name = "bitflags"
+version = "1.3.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bef38d45163c2f1dde094a7dfd33ccf595c92905c8f8f4fdc18d06fb1037718a"
+
+[[package]]
+name = "bstr"
+version = "0.2.17"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ba3569f383e8f1598449f1a423e72e99569137b47740b1da11ef19af3d5c3223"
+dependencies = [
+ "lazy_static",
+ "memchr",
+ "regex-automata",
+]
+
+[[package]]
+name = "cfg-if"
+version = "1.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"
+
+[[package]]
+name = "clap"
+version = "3.2.17"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "29e724a68d9319343bb3328c9cc2dfde263f4b3142ee1059a9980580171c954b"
+dependencies = [
+ "atty",
+ "bitflags",
+ "clap_lex",
+ "indexmap",
+ "strsim",
+ "termcolor",
+ "textwrap",
+]
+
+[[package]]
+name = "clap_complete"
+version = "3.2.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e4179da71abd56c26b54dd0c248cc081c1f43b0a1a7e8448e28e57a29baa993d"
+dependencies = [
+ "clap",
+]
+
+[[package]]
+name = "clap_lex"
+version = "0.2.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "2850f2f5a82cbf437dd5af4d49848fbdfc27c157c3d010345776f952765261c5"
+dependencies = [
+ "os_str_bytes",
+]
+
+[[package]]
+name = "config-file"
+version = "0.2.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "df51e72c150781d2c7d4cbcb0b803277caaa80476786994a62961a8f1010dafb"
+dependencies = [
+ "serde",
+ "thiserror",
+ "toml",
+]
+
+[[package]]
+name = "crossbeam-channel"
+version = "0.5.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c2dd04ddaf88237dc3b8d8f9a3c1004b506b54b3313403944054d23c0870c521"
+dependencies = [
+ "cfg-if",
+ "crossbeam-utils",
+]
+
+[[package]]
+name = "crossbeam-deque"
+version = "0.8.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "715e8152b692bba2d374b53d4875445368fdf21a94751410af607a5ac677d1fc"
+dependencies = [
+ "cfg-if",
+ "crossbeam-epoch",
+ "crossbeam-utils",
+]
+
+[[package]]
+name = "crossbeam-epoch"
+version = "0.9.10"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "045ebe27666471bb549370b4b0b3e51b07f56325befa4284db65fc89c02511b1"
+dependencies = [
+ "autocfg",
+ "cfg-if",
+ "crossbeam-utils",
+ "memoffset",
+ "once_cell",
+ "scopeguard",
+]
+
+[[package]]
+name = "crossbeam-utils"
+version = "0.8.11"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "51887d4adc7b564537b15adcfb307936f8075dfcd5f00dde9a9f1d29383682bc"
+dependencies = [
+ "cfg-if",
+ "once_cell",
+]
+
+[[package]]
+name = "difflib"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6184e33543162437515c2e2b48714794e37845ec9851711914eec9d308f6ebe8"
+
+[[package]]
+name = "directories"
+version = "4.0.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f51c5d4ddabd36886dd3e1438cb358cdcb0d7c499cb99cb4ac2e38e18b5cb210"
+dependencies = [
+ "dirs-sys",
+]
+
+[[package]]
+name = "dirs-sys"
+version = "0.3.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1b1d1d91c932ef41c0f2663aa8b0ca0342d444d842c06914aa0a7e352d0bada6"
+dependencies = [
+ "libc",
+ "redox_users",
+ "winapi",
+]
+
+[[package]]
+name = "doc-comment"
+version = "0.3.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fea41bba32d969b513997752735605054bc0dfa92b4c56bf1189f2e174be7a10"
+
+[[package]]
+name = "du-dust"
+version = "0.8.3"
+dependencies = [
+ "ansi_term",
+ "assert_cmd",
+ "clap",
+ "clap_complete",
+ "config-file",
+ "directories",
+ "lscolors",
+ "rayon",
+ "regex",
+ "serde",
+ "stfu8",
+ "tempfile",
+ "terminal_size",
+ "thousands",
+ "unicode-width",
+ "winapi-util",
+]
+
+[[package]]
+name = "either"
+version = "1.8.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "90e5c1c8368803113bf0c9584fc495a58b86dc8a29edbf8fe877d21d9507e797"
+
+[[package]]
+name = "fastrand"
+version = "1.8.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a7a407cfaa3385c4ae6b23e84623d48c2798d06e3e6a1878f7f59f17b3f86499"
+dependencies = [
+ "instant",
+]
+
+[[package]]
+name = "getrandom"
+version = "0.2.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4eb1a864a501629691edf6c15a593b7a51eebaa1e8468e9ddc623de7c9b58ec6"
+dependencies = [
+ "cfg-if",
+ "libc",
+ "wasi",
+]
+
+[[package]]
+name = "hashbrown"
+version = "0.12.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8a9ee70c43aaf417c914396645a0fa852624801b24ebb7ae78fe8272889ac888"
+
+[[package]]
+name = "hermit-abi"
+version = "0.1.19"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "62b467343b94ba476dcb2500d242dadbb39557df889310ac77c5d99100aaac33"
+dependencies = [
+ "libc",
+]
+
+[[package]]
+name = "indexmap"
+version = "1.9.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "10a35a97730320ffe8e2d410b5d3b69279b98d2c14bdb8b70ea89ecf7888d41e"
+dependencies = [
+ "autocfg",
+ "hashbrown",
+]
+
+[[package]]
+name = "instant"
+version = "0.1.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7a5bbe824c507c5da5956355e86a746d82e0e1464f65d862cc5e71da70e94b2c"
+dependencies = [
+ "cfg-if",
+]
+
+[[package]]
+name = "itertools"
+version = "0.10.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a9a9d19fa1e79b6215ff29b9d6880b706147f16e9b1dbb1e4e5947b5b02bc5e3"
+dependencies = [
+ "either",
+]
+
+[[package]]
+name = "lazy_static"
+version = "1.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e2abad23fbc42b3700f2f279844dc832adb2b2eb069b2df918f455c4e18cc646"
+
+[[package]]
+name = "libc"
+version = "0.2.132"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8371e4e5341c3a96db127eb2465ac681ced4c433e01dd0e938adbef26ba93ba5"
+
+[[package]]
+name = "lscolors"
+version = "0.7.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d24b894c45c9da468621cdd615a5a79ee5e5523dd4f75c76ebc03d458940c16e"
+dependencies = [
+ "ansi_term",
+]
+
+[[package]]
+name = "memchr"
+version = "2.5.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "2dffe52ecf27772e601905b7522cb4ef790d2cc203488bbd0e2fe85fcb74566d"
+
+[[package]]
+name = "memoffset"
+version = "0.6.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5aa361d4faea93603064a027415f07bd8e1d5c88c9fbf68bf56a285428fd79ce"
+dependencies = [
+ "autocfg",
+]
+
+[[package]]
+name = "num_cpus"
+version = "1.13.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "19e64526ebdee182341572e50e9ad03965aa510cd94427a4549448f285e957a1"
+dependencies = [
+ "hermit-abi",
+ "libc",
+]
+
+[[package]]
+name = "once_cell"
+version = "1.13.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "074864da206b4973b84eb91683020dbefd6a8c3f0f38e054d93954e891935e4e"
+
+[[package]]
+name = "os_str_bytes"
+version = "6.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9ff7415e9ae3fff1225851df9e0d9e4e5479f947619774677a63572e55e80eff"
+
+[[package]]
+name = "predicates"
+version = "2.1.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a5aab5be6e4732b473071984b3164dbbfb7a3674d30ea5ff44410b6bcd960c3c"
+dependencies = [
+ "difflib",
+ "itertools",
+ "predicates-core",
+]
+
+[[package]]
+name = "predicates-core"
+version = "1.0.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "da1c2388b1513e1b605fcec39a95e0a9e8ef088f71443ef37099fa9ae6673fcb"
+
+[[package]]
+name = "predicates-tree"
+version = "1.0.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4d86de6de25020a36c6d3643a86d9a6a9f552107c0559c60ea03551b5e16c032"
+dependencies = [
+ "predicates-core",
+ "termtree",
+]
+
+[[package]]
+name = "proc-macro2"
+version = "1.0.43"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0a2ca2c61bc9f3d74d2886294ab7b9853abd9c1ad903a3ac7815c58989bb7bab"
+dependencies = [
+ "unicode-ident",
+]
+
+[[package]]
+name = "quote"
+version = "1.0.21"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bbe448f377a7d6961e30f5955f9b8d106c3f5e449d493ee1b125c1d43c2b5179"
+dependencies = [
+ "proc-macro2",
+]
+
+[[package]]
+name = "rayon"
+version = "1.5.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bd99e5772ead8baa5215278c9b15bf92087709e9c1b2d1f97cdb5a183c933a7d"
+dependencies = [
+ "autocfg",
+ "crossbeam-deque",
+ "either",
+ "rayon-core",
+]
+
+[[package]]
+name = "rayon-core"
+version = "1.9.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "258bcdb5ac6dad48491bb2992db6b7cf74878b0384908af124823d118c99683f"
+dependencies = [
+ "crossbeam-channel",
+ "crossbeam-deque",
+ "crossbeam-utils",
+ "num_cpus",
+]
+
+[[package]]
+name = "redox_syscall"
+version = "0.2.16"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fb5a58c1855b4b6819d59012155603f0b22ad30cad752600aadfcb695265519a"
+dependencies = [
+ "bitflags",
+]
+
+[[package]]
+name = "redox_users"
+version = "0.4.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b033d837a7cf162d7993aded9304e30a83213c648b6e389db233191f891e5c2b"
+dependencies = [
+ "getrandom",
+ "redox_syscall",
+ "thiserror",
+]
+
+[[package]]
+name = "regex"
+version = "1.6.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4c4eb3267174b8c6c2f654116623910a0fef09c4753f8dd83db29c48a0df988b"
+dependencies = [
+ "aho-corasick",
+ "memchr",
+ "regex-syntax",
+]
+
+[[package]]
+name = "regex-automata"
+version = "0.1.10"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6c230d73fb8d8c1b9c0b3135c5142a8acee3a0558fb8db5cf1cb65f8d7862132"
+
+[[package]]
+name = "regex-syntax"
+version = "0.6.27"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a3f87b73ce11b1619a3c6332f45341e0047173771e8b8b73f87bfeefb7b56244"
+
+[[package]]
+name = "remove_dir_all"
+version = "0.5.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3acd125665422973a33ac9d3dd2df85edad0f4ae9b00dafb1a05e43a9f5ef8e7"
+dependencies = [
+ "winapi",
+]
+
+[[package]]
+name = "scopeguard"
+version = "1.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d29ab0c6d3fc0ee92fe66e2d99f700eab17a8d57d1c1d3b748380fb20baa78cd"
+
+[[package]]
+name = "serde"
+version = "1.0.143"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "53e8e5d5b70924f74ff5c6d64d9a5acd91422117c60f48c4e07855238a254553"
+dependencies = [
+ "serde_derive",
+]
+
+[[package]]
+name = "serde_derive"
+version = "1.0.143"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d3d8e8de557aee63c26b85b947f5e59b690d0454c753f3adeb5cd7835ab88391"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
+[[package]]
+name = "stfu8"
+version = "0.2.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "019f0c664fd85d5a87dcfb62b40b691055392a35a6e59f4df83d4b770db7e876"
+dependencies = [
+ "lazy_static",
+ "regex",
+]
+
+[[package]]
+name = "strsim"
+version = "0.10.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "73473c0e59e6d5812c5dfe2a064a6444949f089e20eec9a2e5506596494e4623"
+
+[[package]]
+name = "syn"
+version = "1.0.99"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "58dbef6ec655055e20b86b15a8cc6d439cca19b667537ac6a1369572d151ab13"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "unicode-ident",
+]
+
+[[package]]
+name = "tempfile"
+version = "3.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5cdb1ef4eaeeaddc8fbd371e5017057064af0911902ef36b39801f67cc6d79e4"
+dependencies = [
+ "cfg-if",
+ "fastrand",
+ "libc",
+ "redox_syscall",
+ "remove_dir_all",
+ "winapi",
+]
+
+[[package]]
+name = "termcolor"
+version = "1.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bab24d30b911b2376f3a13cc2cd443142f0c81dda04c118693e35b3835757755"
+dependencies = [
+ "winapi-util",
+]
+
+[[package]]
+name = "terminal_size"
+version = "0.1.17"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "633c1a546cee861a1a6d0dc69ebeca693bf4296661ba7852b9d21d159e0506df"
+dependencies = [
+ "libc",
+ "winapi",
+]
+
+[[package]]
+name = "termtree"
+version = "0.2.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "507e9898683b6c43a9aa55b64259b721b52ba226e0f3779137e50ad114a4c90b"
+
+[[package]]
+name = "textwrap"
+version = "0.15.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b1141d4d61095b28419e22cb0bbf02755f5e54e0526f97f1e3d1d160e60885fb"
+
+[[package]]
+name = "thiserror"
+version = "1.0.32"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f5f6586b7f764adc0231f4c79be7b920e766bb2f3e51b3661cdb263828f19994"
+dependencies = [
+ "thiserror-impl",
+]
+
+[[package]]
+name = "thiserror-impl"
+version = "1.0.32"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "12bafc5b54507e0149cdf1b145a5d80ab80a90bcd9275df43d4fff68460f6c21"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
+[[package]]
+name = "thousands"
+version = "0.2.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3bf63baf9f5039dadc247375c29eb13706706cfde997d0330d05aa63a77d8820"
+
+[[package]]
+name = "toml"
+version = "0.5.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8d82e1a7758622a465f8cee077614c73484dac5b836c02ff6a40d5d1010324d7"
+dependencies = [
+ "serde",
+]
+
+[[package]]
+name = "unicode-ident"
+version = "1.0.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c4f5b37a154999a8f3f98cc23a628d850e154479cd94decf3414696e12e31aaf"
+
+[[package]]
+name = "unicode-width"
+version = "0.1.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3ed742d4ea2bd1176e236172c8429aaf54486e7ac098db29ffe6529e0ce50973"
+
+[[package]]
+name = "wait-timeout"
+version = "0.2.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9f200f5b12eb75f8c1ed65abd4b2db8a6e1b138a20de009dacee265a2498f3f6"
+dependencies = [
+ "libc",
+]
+
+[[package]]
+name = "wasi"
+version = "0.11.0+wasi-snapshot-preview1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9c8d87e72b64a3b4db28d11ce29237c246188f4f51057d65a7eab63b7987e423"
+
+[[package]]
+name = "winapi"
+version = "0.3.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5c839a674fcd7a98952e593242ea400abe93992746761e38641405d28b00f419"
+dependencies = [
+ "winapi-i686-pc-windows-gnu",
+ "winapi-x86_64-pc-windows-gnu",
+]
+
+[[package]]
+name = "winapi-i686-pc-windows-gnu"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ac3b87c63620426dd9b991e5ce0329eff545bccbbb34f3be09ff6fb6ab51b7b6"
+
+[[package]]
+name = "winapi-util"
+version = "0.1.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "70ec6ce85bb158151cae5e5c87f95a8e97d2c0c4b001223f33a334e3ce5de178"
+dependencies = [
+ "winapi",
+]
+
+[[package]]
+name = "winapi-x86_64-pc-windows-gnu"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f"

Cargo.toml

@@ -1,8 +1,81 @@
 [package]
+name = "du-dust"
+description = "A more intuitive version of du"
+version = "0.8.3"
+authors = ["bootandy <bootandy@gmail.com>", "nebkor <code@ardent.nebcorp.com>"]
+edition = "2021"
+readme = "README.md"
+
+documentation = "https://github.com/bootandy/dust"
+homepage = "https://github.com/bootandy/dust"
+repository = "https://github.com/bootandy/dust"
+
+keywords = ["du", "command-line", "disk", "disk-usage"]
+categories = ["command-line-utilities"]
+license = "Apache-2.0"
+
+[badges]
+travis-ci = { repository = "https://travis-ci.org/bootandy/dust" }
+
+[[bin]]
 name = "dust"
-version = "0.1.0"
-authors = ["bootandy <bootandy@gmail.com>"]
+path = "src/main.rs"
+
+[profile.release]
+codegen-units = 1
+lto = true
+strip = true
 
 [dependencies]
-ansi_term = "0.11"
-clap = "2.31"
+ansi_term = "0.12"
+clap = "3.2.17"
+lscolors = "0.7"
+terminal_size = "0.1"
+unicode-width = "0.1"
+rayon = "1"
+thousands = "0.2"
+stfu8 = "0.2"
+regex = "1"
+config-file = "0.2"
+serde = { version = "1.0", features = ["derive"] }
+directories = "4"
+
+[target.'cfg(windows)'.dependencies]
+winapi-util = "0.1"
+
+[dev-dependencies]
+assert_cmd = "1"
+tempfile = "=3"
+
+[build-dependencies]
+clap = "3.2.17"
+clap_complete = "3.2.4"
+
+[[test]]
+name = "integration"
+path = "tests/tests.rs"
+
+[package.metadata.deb]
+section = "utils"
+assets = [
+  [
+    "target/release/dust",
+    "usr/bin/",
+    "755",
+  ],
+  [
+    "LICENSE",
+    "usr/share/doc/du-dust/",
+    "644",
+  ],
+  [
+    "README.md",
+    "usr/share/doc/du-dust/README",
+    "644",
+  ],
+]
+extended-description = """\
+Dust is meant to give you an instant overview of which directories are using
+disk space without requiring sort or head. Dust will print a maximum of one
+'Did not have permissions message'.
+"""

README.md

@@ -1,39 +1,86 @@
+[![Build Status](https://travis-ci.org/bootandy/dust.svg?branch=master)](https://travis-ci.org/bootandy/dust)
+
 # Dust
-du + rust = dust. A rust alternative to du
 
-Unlike du, dust is meant to give you an instant overview of which directories are using disk space without requiring sort or head. Dust does not count file system blocks; it uses file sizes instead. Dust will print a maximum of 1 'Did not have permissions message'.
+du + rust = dust. Like du but more intuitive.
 
+# Why
 
-Dust will list the 15 biggest sub directories and will smartly recurse down the tree to find the larger ones. There is no need for a '-d' flag or a '-h' flag. The largest sub directory will have its size shown in red
+Because I want an easy way to see where my disk is being used.
+
+# Demo
+
+![Example](media/snap.png)
+
+## Install
+
+#### Cargo <a href="https://repology.org/project/du-dust/versions"><img src="https://repology.org/badge/vertical-allrepos/du-dust.svg" alt="Packaging status" align="right"></a>
+
+- `cargo install du-dust`
+
+#### 🍺 Homebrew (Mac OS)
+
+- `brew install dust`
+
+#### 🍺 Homebrew (Linux)
+
+- `brew tap tgotwig/linux-dust && brew install dust`
+
+#### [Pacstall](https://github.com/pacstall/pacstall) (Debian/Ubuntu)
+
+- `pacstall -I dust-bin`
+
+#### Windows:
+
+- Windows GNU version - works
+- Windows MSVC - requires: [VCRUNTIME140.dll](https://docs.microsoft.com/en-gb/cpp/windows/latest-supported-vc-redist?view=msvc-170)
+
+#### Download
+
+- Download Linux/Mac binary from [Releases](https://github.com/bootandy/dust/releases)
+- unzip file: `tar -xvf _downloaded_file.tar.gz`
+- move file to executable path: `sudo mv dust /usr/local/bin/`
+
+## Overview
+
+Dust is meant to give you an instant overview of which directories are using disk space without requiring sort or head. Dust will print a maximum of one 'Did not have permissions message'.
+
+Dust will list a slightly-less-than-the-terminal-height number of the biggest subdirectories or files and will smartly recurse down the tree to find the larger ones. There is no need for a '-d' flag or a '-h' flag. The largest subdirectories will be colored.
+
+The different colors on the bars: These represent the combined tree hierarchy & disk usage. The shades of grey are used to indicate which parent folder a subfolder belongs to. For instance, look at the above screenshot. `.steam` is a folder taking 44% of the space. From the `.steam` bar is a light grey line that goes up. All these folders are inside `.steam` so if you delete `.steam` all that stuff will be gone too.
+
+## Usage
 
 ```
+Usage: dust
 Usage: dust <dir>
-Usage: dust -n 30  <dir>  (Shows 30 directories not 15)
-```
-
+Usage: dust <dir>  <another_dir> <and_more>
+Usage: dust -p (full-path - Show fullpath of the subdirectories)
+Usage: dust -s (apparent-size - shows the length of the file as opposed to the amount of disk space it uses)
+Usage: dust -n 30  (Shows 30 directories instead of the default [default is terminal height])
+Usage: dust -d 3  (Shows 3 levels of subdirectories)
+Usage: dust -r (reverse order of output)
+Usage: dust -H (si print sizes in powers of 1000 instead of 1024)
+Usage: dust -X ignore  (ignore all files and directories with the name 'ignore')
+Usage: dust -x (Only show directories on the same filesystem)
+Usage: dust -b (Do not show percentages or draw ASCII bars)
+Usage: dust -i (Do not show hidden files)
+Usage: dust -c (No colors [monochrome])
+Usage: dust -f (Count files instead of diskspace)
+Usage: dust -t (Group by filetype)
+Usage: dust -z 10M (min-size, Only include files larger than 10M)
+Usage: dust -e regex (Only include files matching this regex (eg dust -e "\.png$" would match png files))
+Usage: dust -v regex (Exclude files matching this regex (eg dust -v "\.png$" would ignore png files))
 
-```
-dust .
- 161M  .
- 160M └── ./target
- 123M    ├── ./target/debug
-  83M    │  ├── ./target/debug/deps
-  16M    │  │  ├── ./target/debug/deps/libclap-82e6176feef5d4b7.rlib
- 8.6M    │  │  └── ./target/debug/deps/dust-993f7d919d92f0f8.dSYM
- 8.6M    │  │     └── ./target/debug/deps/dust-993f7d919d92f0f8.dSYM/Contents
- 8.6M    │  │        └── ./target/debug/deps/dust-993f7d919d92f0f8.dSYM/Contents/Resources
-  27M    │  ├── ./target/debug/incremental
-  12M    │  └── ./target/debug/build
-  20M    ├── ./target/x86_64-apple-darwin
-  20M    │  └── ./target/x86_64-apple-darwin/debug
-  20M    │     └── ./target/x86_64-apple-darwin/debug/deps
-  16M    │        └── ./target/x86_64-apple-darwin/debug/deps/libclap-7e3f8513c52cd558.rlib
-  16M    └── ./target/release
-  13M       └── ./target/release/deps
 ```
 
-Performance: dust is currently about 4 times slower than du.
-
-
+## Alternatives
 
+- [NCDU](https://dev.yorhel.nl/ncdu)
+- [dutree](https://github.com/nachoparker/dutree)
+- [dua](https://github.com/Byron/dua-cli/)
+- [pdu](https://github.com/KSXGitHub/parallel-disk-usage)
+- [dirstat-rs](https://github.com/scullionw/dirstat-rs)
+- du -d 1 -h | sort -h
 
+Note: Apparent-size is calculated slightly differently in dust to gdu. In dust each hard link is counted as using file_length space. In gdu only the first entry is counted.

build.rs

@@ -0,0 +1,18 @@
+use clap_complete::{generate_to, shells::*};
+use std::io::Error;
+
+include!("src/cli.rs");
+
+fn main() -> Result<(), Error> {
+    let outdir = "completions";
+    let app_name = "dust";
+    let mut cmd = build_cli();
+
+    generate_to(Bash, &mut cmd, app_name, outdir)?;
+    generate_to(Zsh, &mut cmd, app_name, outdir)?;
+    generate_to(Fish, &mut cmd, app_name, outdir)?;
+    generate_to(PowerShell, &mut cmd, app_name, outdir)?;
+    generate_to(Elvish, &mut cmd, app_name, outdir)?;
+
+    Ok(())
+}

ci/before_deploy.ps1

@@ -0,0 +1,23 @@
+# This script takes care of packaging the build artifacts that will go in the
+# release zipfile
+
+$SRC_DIR = $PWD.Path
+$STAGE = [System.Guid]::NewGuid().ToString()
+
+Set-Location $ENV:Temp
+New-Item -Type Directory -Name $STAGE
+Set-Location $STAGE
+
+$ZIP = "$SRC_DIR\$($Env:CRATE_NAME)-$($Env:APPVEYOR_REPO_TAG_NAME)-$($Env:TARGET).zip"
+
+# TODO Update this to package the right artifacts
+Copy-Item "$SRC_DIR\target\$($Env:TARGET)\release\dust" '.\'
+
+7z a "$ZIP" *
+
+Push-AppveyorArtifact "$ZIP"
+
+Remove-Item *.* -Force
+Set-Location ..
+Remove-Item $STAGE
+Set-Location $SRC_DIR

ci/before_deploy.sh

@@ -0,0 +1,34 @@
+#!/usr/bin/env bash
+# This script takes care of building your crate and packaging it for release
+
+set -ex
+
+main() {
+    local src=$(pwd) \
+          stage=
+
+    case $TRAVIS_OS_NAME in
+        linux)
+            stage=$(mktemp -d)
+            ;;
+        osx)
+            stage=$(mktemp -d -t tmp)
+            ;;
+    esac
+
+    test -f Cargo.lock || cargo generate-lockfile
+
+    # TODO Update this to build the artifacts that matter to you
+    cross rustc --bin dust --target $TARGET --release -- -C lto
+
+    # TODO Update this to package the right artifacts
+    cp target/$TARGET/release/dust $stage/
+
+    cd $stage
+    tar czf $src/$CRATE_NAME-$TRAVIS_TAG-$TARGET.tar.gz *
+    cd $src
+
+    rm -rf $stage
+}
+
+main

ci/how2publish.txt

@@ -0,0 +1,10 @@
+# ----------- To do a release ---------
+# edit version in cargo.toml
+# tag a commit and push (increment version in Cargo.toml first):
+#   git tag v0.4.5
+#   git push origin v0.4.5
+
+# cargo publish to put it in crates.io
+
+# To install locally [Do before pushing it]
+#cargo install --path .

ci/install.sh

@@ -0,0 +1,28 @@
+#!/usr/bin/env bash
+set -ex
+
+main() {
+    local target=
+    if [ $TRAVIS_OS_NAME = linux ]; then
+        target=x86_64-unknown-linux-musl
+        sort=sort
+    else
+        target=x86_64-apple-darwin
+        sort=gsort  # for `sort --sort-version`, from brew's coreutils.
+    fi
+
+    # This fetches latest stable release
+    local tag=$(git ls-remote --tags --refs --exit-code https://github.com/japaric/cross \
+                       | cut -d/ -f3 \
+                       | grep -E '^v[0.1.0-9.]+$' \
+                       | $sort --version-sort \
+                       | tail -n1)
+    curl -LSfs https://japaric.github.io/trust/install.sh | \
+        sh -s -- \
+           --force \
+           --git japaric/cross \
+           --tag $tag \
+           --target $target
+}
+
+main

ci/script.sh

@@ -0,0 +1,25 @@
+#!/usr/bin/env bash
+# This script takes care of testing your crate
+
+set -ex
+
+# TODO This is the "test phase", tweak it as you see fit
+main() {
+    cross build --target $TARGET
+    cross build --target $TARGET --release
+
+    if [ ! -z $DISABLE_TESTS ]; then
+        return
+    fi
+
+    cross test --target $TARGET
+    cross test --target $TARGET --release
+
+    cross run --target $TARGET
+    cross run --target $TARGET --release
+}
+
+# we don't run the "test phase" when doing deploys
+if [ -z $TRAVIS_TAG ]; then
+    main
+fi

completions/_dust

@@ -0,0 +1,67 @@
+#compdef dust
+
+autoload -U is-at-least
+
+_dust() {
+    typeset -A opt_args
+    typeset -a _arguments_options
+    local ret=1
+
+    if is-at-least 5.2; then
+        _arguments_options=(-s -S -C)
+    else
+        _arguments_options=(-s -C)
+    fi
+
+    local context curcontext="$curcontext" state line
+    _arguments "${_arguments_options[@]}" \
+'-d+[Depth to show]: : ' \
+'--depth=[Depth to show]: : ' \
+'-n+[Number of lines of output to show. (Default is terminal_height - 10)]: : ' \
+'--number-of-lines=[Number of lines of output to show. (Default is terminal_height - 10)]: : ' \
+'*-X+[Exclude any file or directory with this name]: : ' \
+'*--ignore-directory=[Exclude any file or directory with this name]: : ' \
+'-z+[Minimum size file to include in output]: : ' \
+'--min-size=[Minimum size file to include in output]: : ' \
+'(-e --filter -t --file_types)*-v+[Exclude filepaths matching this regex. To ignore png files type: -v "\\.png$" ]: : ' \
+'(-e --filter -t --file_types)*--invert-filter=[Exclude filepaths matching this regex. To ignore png files type: -v "\\.png$" ]: : ' \
+'(-t --file_types)*-e+[Only include filepaths matching this regex. For png files type: -e "\\.png$" ]: : ' \
+'(-t --file_types)*--filter=[Only include filepaths matching this regex. For png files type: -e "\\.png$" ]: : ' \
+'-w+[Specify width of output overriding the auto detection of terminal width]: : ' \
+'--terminal_width=[Specify width of output overriding the auto detection of terminal width]: : ' \
+'-h[Print help information]' \
+'--help[Print help information]' \
+'-V[Print version information]' \
+'--version[Print version information]' \
+'-p[Subdirectories will not have their path shortened]' \
+'--full-paths[Subdirectories will not have their path shortened]' \
+'-x[Only count the files and directories on the same filesystem as the supplied directory]' \
+'--limit-filesystem[Only count the files and directories on the same filesystem as the supplied directory]' \
+'-s[Use file length instead of blocks]' \
+'--apparent-size[Use file length instead of blocks]' \
+'-r[Print tree upside down (biggest highest)]' \
+'--reverse[Print tree upside down (biggest highest)]' \
+'-c[No colors will be printed (Useful for commands like: watch)]' \
+'--no-colors[No colors will be printed (Useful for commands like: watch)]' \
+'-b[No percent bars or percentages will be displayed]' \
+'--no-percent-bars[No percent bars or percentages will be displayed]' \
+'--skip-total[No total row will be displayed]' \
+'-f[Directory '\''size'\'' is number of child files/dirs not disk size]' \
+'--filecount[Directory '\''size'\'' is number of child files/dirs not disk size]' \
+'-i[Do not display hidden files]' \
+'--ignore_hidden[Do not display hidden files]' \
+'(-d --depth)-t[show only these file types]' \
+'(-d --depth)--file_types[show only these file types]' \
+'-H[print sizes in powers of 1000 (e.g., 1.1G)]' \
+'--si[print sizes in powers of 1000 (e.g., 1.1G)]' \
+'*::inputs:' \
+&& ret=0
+}
+
+(( $+functions[_dust_commands] )) ||
+_dust_commands() {
+    local commands; commands=()
+    _describe -t commands 'dust commands' commands "$@"
+}
+
+_dust "$@"

completions/_dust.ps1

@@ -0,0 +1,69 @@
+
+using namespace System.Management.Automation
+using namespace System.Management.Automation.Language
+
+Register-ArgumentCompleter -Native -CommandName 'dust' -ScriptBlock {
+    param($wordToComplete, $commandAst, $cursorPosition)
+
+    $commandElements = $commandAst.CommandElements
+    $command = @(
+        'dust'
+        for ($i = 1; $i -lt $commandElements.Count; $i++) {
+            $element = $commandElements[$i]
+            if ($element -isnot [StringConstantExpressionAst] -or
+                $element.StringConstantType -ne [StringConstantType]::BareWord -or
+                $element.Value.StartsWith('-') -or
+                $element.Value -eq $wordToComplete) {
+                break
+        }
+        $element.Value
+    }) -join ';'
+
+    $completions = @(switch ($command) {
+        'dust' {
+            [CompletionResult]::new('-d', 'd', [CompletionResultType]::ParameterName, 'Depth to show')
+            [CompletionResult]::new('--depth', 'depth', [CompletionResultType]::ParameterName, 'Depth to show')
+            [CompletionResult]::new('-n', 'n', [CompletionResultType]::ParameterName, 'Number of lines of output to show. (Default is terminal_height - 10)')
+            [CompletionResult]::new('--number-of-lines', 'number-of-lines', [CompletionResultType]::ParameterName, 'Number of lines of output to show. (Default is terminal_height - 10)')
+            [CompletionResult]::new('-X', 'X', [CompletionResultType]::ParameterName, 'Exclude any file or directory with this name')
+            [CompletionResult]::new('--ignore-directory', 'ignore-directory', [CompletionResultType]::ParameterName, 'Exclude any file or directory with this name')
+            [CompletionResult]::new('-z', 'z', [CompletionResultType]::ParameterName, 'Minimum size file to include in output')
+            [CompletionResult]::new('--min-size', 'min-size', [CompletionResultType]::ParameterName, 'Minimum size file to include in output')
+            [CompletionResult]::new('-v', 'v', [CompletionResultType]::ParameterName, 'Exclude filepaths matching this regex. To ignore png files type: -v "\.png$" ')
+            [CompletionResult]::new('--invert-filter', 'invert-filter', [CompletionResultType]::ParameterName, 'Exclude filepaths matching this regex. To ignore png files type: -v "\.png$" ')
+            [CompletionResult]::new('-e', 'e', [CompletionResultType]::ParameterName, 'Only include filepaths matching this regex. For png files type: -e "\.png$" ')
+            [CompletionResult]::new('--filter', 'filter', [CompletionResultType]::ParameterName, 'Only include filepaths matching this regex. For png files type: -e "\.png$" ')
+            [CompletionResult]::new('-w', 'w', [CompletionResultType]::ParameterName, 'Specify width of output overriding the auto detection of terminal width')
+            [CompletionResult]::new('--terminal_width', 'terminal_width', [CompletionResultType]::ParameterName, 'Specify width of output overriding the auto detection of terminal width')
+            [CompletionResult]::new('-h', 'h', [CompletionResultType]::ParameterName, 'Print help information')
+            [CompletionResult]::new('--help', 'help', [CompletionResultType]::ParameterName, 'Print help information')
+            [CompletionResult]::new('-V', 'V', [CompletionResultType]::ParameterName, 'Print version information')
+            [CompletionResult]::new('--version', 'version', [CompletionResultType]::ParameterName, 'Print version information')
+            [CompletionResult]::new('-p', 'p', [CompletionResultType]::ParameterName, 'Subdirectories will not have their path shortened')
+            [CompletionResult]::new('--full-paths', 'full-paths', [CompletionResultType]::ParameterName, 'Subdirectories will not have their path shortened')
+            [CompletionResult]::new('-x', 'x', [CompletionResultType]::ParameterName, 'Only count the files and directories on the same filesystem as the supplied directory')
+            [CompletionResult]::new('--limit-filesystem', 'limit-filesystem', [CompletionResultType]::ParameterName, 'Only count the files and directories on the same filesystem as the supplied directory')
+            [CompletionResult]::new('-s', 's', [CompletionResultType]::ParameterName, 'Use file length instead of blocks')
+            [CompletionResult]::new('--apparent-size', 'apparent-size', [CompletionResultType]::ParameterName, 'Use file length instead of blocks')
+            [CompletionResult]::new('-r', 'r', [CompletionResultType]::ParameterName, 'Print tree upside down (biggest highest)')
+            [CompletionResult]::new('--reverse', 'reverse', [CompletionResultType]::ParameterName, 'Print tree upside down (biggest highest)')
+            [CompletionResult]::new('-c', 'c', [CompletionResultType]::ParameterName, 'No colors will be printed (Useful for commands like: watch)')
+            [CompletionResult]::new('--no-colors', 'no-colors', [CompletionResultType]::ParameterName, 'No colors will be printed (Useful for commands like: watch)')
+            [CompletionResult]::new('-b', 'b', [CompletionResultType]::ParameterName, 'No percent bars or percentages will be displayed')
+            [CompletionResult]::new('--no-percent-bars', 'no-percent-bars', [CompletionResultType]::ParameterName, 'No percent bars or percentages will be displayed')
+            [CompletionResult]::new('--skip-total', 'skip-total', [CompletionResultType]::ParameterName, 'No total row will be displayed')
+            [CompletionResult]::new('-f', 'f', [CompletionResultType]::ParameterName, 'Directory ''size'' is number of child files/dirs not disk size')
+            [CompletionResult]::new('--filecount', 'filecount', [CompletionResultType]::ParameterName, 'Directory ''size'' is number of child files/dirs not disk size')
+            [CompletionResult]::new('-i', 'i', [CompletionResultType]::ParameterName, 'Do not display hidden files')
+            [CompletionResult]::new('--ignore_hidden', 'ignore_hidden', [CompletionResultType]::ParameterName, 'Do not display hidden files')
+            [CompletionResult]::new('-t', 't', [CompletionResultType]::ParameterName, 'show only these file types')
+            [CompletionResult]::new('--file_types', 'file_types', [CompletionResultType]::ParameterName, 'show only these file types')
+            [CompletionResult]::new('-H', 'H', [CompletionResultType]::ParameterName, 'print sizes in powers of 1000 (e.g., 1.1G)')
+            [CompletionResult]::new('--si', 'si', [CompletionResultType]::ParameterName, 'print sizes in powers of 1000 (e.g., 1.1G)')
+            break
+        }
+    })
+
+    $completions.Where{ $_.CompletionText -like "$wordToComplete*" } |
+        Sort-Object -Property ListItemText
+}

completions/dust.bash

@@ -0,0 +1,94 @@
+_dust() {
+    local i cur prev opts cmds
+    COMPREPLY=()
+    cur="${COMP_WORDS[COMP_CWORD]}"
+    prev="${COMP_WORDS[COMP_CWORD-1]}"
+    cmd=""
+    opts=""
+
+    for i in ${COMP_WORDS[@]}
+    do
+        case "${i}" in
+            "$1")
+                cmd="dust"
+                ;;
+            *)
+                ;;
+        esac
+    done
+
+    case "${cmd}" in
+        dust)
+            opts="-h -V -d -n -p -X -x -s -r -c -b -z -f -i -v -e -t -w -H --help --version --depth --number-of-lines --full-paths --ignore-directory --limit-filesystem --apparent-size --reverse --no-colors --no-percent-bars --min-size --skip-total --filecount --ignore_hidden --invert-filter --filter --file_types --terminal_width --si <inputs>..."
+            if [[ ${cur} == -* || ${COMP_CWORD} -eq 1 ]] ; then
+                COMPREPLY=( $(compgen -W "${opts}" -- "${cur}") )
+                return 0
+            fi
+            case "${prev}" in
+                --depth)
+                    COMPREPLY=($(compgen -f "${cur}"))
+                    return 0
+                    ;;
+                -d)
+                    COMPREPLY=($(compgen -f "${cur}"))
+                    return 0
+                    ;;
+                --number-of-lines)
+                    COMPREPLY=($(compgen -f "${cur}"))
+                    return 0
+                    ;;
+                -n)
+                    COMPREPLY=($(compgen -f "${cur}"))
+                    return 0
+                    ;;
+                --ignore-directory)
+                    COMPREPLY=($(compgen -f "${cur}"))
+                    return 0
+                    ;;
+                -X)
+                    COMPREPLY=($(compgen -f "${cur}"))
+                    return 0
+                    ;;
+                --min-size)
+                    COMPREPLY=($(compgen -f "${cur}"))
+                    return 0
+                    ;;
+                -z)
+                    COMPREPLY=($(compgen -f "${cur}"))
+                    return 0
+                    ;;
+                --invert-filter)
+                    COMPREPLY=($(compgen -f "${cur}"))
+                    return 0
+                    ;;
+                -v)
+                    COMPREPLY=($(compgen -f "${cur}"))
+                    return 0
+                    ;;
+                --filter)
+                    COMPREPLY=($(compgen -f "${cur}"))
+                    return 0
+                    ;;
+                -e)
+                    COMPREPLY=($(compgen -f "${cur}"))
+                    return 0
+                    ;;
+                --terminal_width)
+                    COMPREPLY=($(compgen -f "${cur}"))
+                    return 0
+                    ;;
+                -w)
+                    COMPREPLY=($(compgen -f "${cur}"))
+                    return 0
+                    ;;
+                *)
+                    COMPREPLY=()
+                    ;;
+            esac
+            COMPREPLY=( $(compgen -W "${opts}" -- "${cur}") )
+            return 0
+            ;;
+    esac
+}
+
+complete -F _dust -o bashdefault -o default dust

completions/dust.elv

@@ -0,0 +1,63 @@
+
+use builtin;
+use str;
+
+set edit:completion:arg-completer[dust] = {|@words|
+    fn spaces {|n|
+        builtin:repeat $n ' ' | str:join ''
+    }
+    fn cand {|text desc|
+        edit:complex-candidate $text &display=$text' '(spaces (- 14 (wcswidth $text)))$desc
+    }
+    var command = 'dust'
+    for word $words[1..-1] {
+        if (str:has-prefix $word '-') {
+            break
+        }
+        set command = $command';'$word
+    }
+    var completions = [
+        &'dust'= {
+            cand -d 'Depth to show'
+            cand --depth 'Depth to show'
+            cand -n 'Number of lines of output to show. (Default is terminal_height - 10)'
+            cand --number-of-lines 'Number of lines of output to show. (Default is terminal_height - 10)'
+            cand -X 'Exclude any file or directory with this name'
+            cand --ignore-directory 'Exclude any file or directory with this name'
+            cand -z 'Minimum size file to include in output'
+            cand --min-size 'Minimum size file to include in output'
+            cand -v 'Exclude filepaths matching this regex. To ignore png files type: -v "\.png$" '
+            cand --invert-filter 'Exclude filepaths matching this regex. To ignore png files type: -v "\.png$" '
+            cand -e 'Only include filepaths matching this regex. For png files type: -e "\.png$" '
+            cand --filter 'Only include filepaths matching this regex. For png files type: -e "\.png$" '
+            cand -w 'Specify width of output overriding the auto detection of terminal width'
+            cand --terminal_width 'Specify width of output overriding the auto detection of terminal width'
+            cand -h 'Print help information'
+            cand --help 'Print help information'
+            cand -V 'Print version information'
+            cand --version 'Print version information'
+            cand -p 'Subdirectories will not have their path shortened'
+            cand --full-paths 'Subdirectories will not have their path shortened'
+            cand -x 'Only count the files and directories on the same filesystem as the supplied directory'
+            cand --limit-filesystem 'Only count the files and directories on the same filesystem as the supplied directory'
+            cand -s 'Use file length instead of blocks'
+            cand --apparent-size 'Use file length instead of blocks'
+            cand -r 'Print tree upside down (biggest highest)'
+            cand --reverse 'Print tree upside down (biggest highest)'
+            cand -c 'No colors will be printed (Useful for commands like: watch)'
+            cand --no-colors 'No colors will be printed (Useful for commands like: watch)'
+            cand -b 'No percent bars or percentages will be displayed'
+            cand --no-percent-bars 'No percent bars or percentages will be displayed'
+            cand --skip-total 'No total row will be displayed'
+            cand -f 'Directory ''size'' is number of child files/dirs not disk size'
+            cand --filecount 'Directory ''size'' is number of child files/dirs not disk size'
+            cand -i 'Do not display hidden files'
+            cand --ignore_hidden 'Do not display hidden files'
+            cand -t 'show only these file types'
+            cand --file_types 'show only these file types'
+            cand -H 'print sizes in powers of 1000 (e.g., 1.1G)'
+            cand --si 'print sizes in powers of 1000 (e.g., 1.1G)'
+        }
+    ]
+    $completions[$command]
+}

completions/dust.fish

@@ -0,0 +1,20 @@
+complete -c dust -s d -l depth -d 'Depth to show' -r
+complete -c dust -s n -l number-of-lines -d 'Number of lines of output to show. (Default is terminal_height - 10)' -r
+complete -c dust -s X -l ignore-directory -d 'Exclude any file or directory with this name' -r
+complete -c dust -s z -l min-size -d 'Minimum size file to include in output' -r
+complete -c dust -s v -l invert-filter -d 'Exclude filepaths matching this regex. To ignore png files type: -v "\\.png$" ' -r
+complete -c dust -s e -l filter -d 'Only include filepaths matching this regex. For png files type: -e "\\.png$" ' -r
+complete -c dust -s w -l terminal_width -d 'Specify width of output overriding the auto detection of terminal width' -r
+complete -c dust -s h -l help -d 'Print help information'
+complete -c dust -s V -l version -d 'Print version information'
+complete -c dust -s p -l full-paths -d 'Subdirectories will not have their path shortened'
+complete -c dust -s x -l limit-filesystem -d 'Only count the files and directories on the same filesystem as the supplied directory'
+complete -c dust -s s -l apparent-size -d 'Use file length instead of blocks'
+complete -c dust -s r -l reverse -d 'Print tree upside down (biggest highest)'
+complete -c dust -s c -l no-colors -d 'No colors will be printed (Useful for commands like: watch)'
+complete -c dust -s b -l no-percent-bars -d 'No percent bars or percentages will be displayed'
+complete -c dust -l skip-total -d 'No total row will be displayed'
+complete -c dust -s f -l filecount -d 'Directory \'size\' is number of child files/dirs not disk size'
+complete -c dust -s i -l ignore_hidden -d 'Do not display hidden files'
+complete -c dust -s t -l file_types -d 'show only these file types'
+complete -c dust -s H -l si -d 'print sizes in powers of 1000 (e.g., 1.1G)'

config/config.toml

@@ -0,0 +1,13 @@
+# Sample Config file, works with toml and yaml
+# Place in either:
+#   ~/.config/dust/config.toml
+#   ~/.dust.toml
+
+reverse=true
+display-full-paths=true
+display-apparent-size=true
+no-colors=true
+no-bars=true
+skip-total=true
+ignore-hidden=true
+iso=true

src/cli.rs

@@ -0,0 +1,135 @@
+use clap::{Arg, Command};
+
+pub fn build_cli() -> Command<'static> {
+    Command::new("Dust")
+        .about("Like du but more intuitive")
+        .version(env!("CARGO_PKG_VERSION"))
+        .trailing_var_arg(true)
+        .arg(
+            Arg::new("depth")
+                .short('d')
+                .long("depth")
+                .help("Depth to show")
+                .takes_value(true)
+        )
+        .arg(
+            Arg::new("number_of_lines")
+                .short('n')
+                .long("number-of-lines")
+                .help("Number of lines of output to show. (Default is terminal_height - 10)")
+                .takes_value(true)
+        )
+        .arg(
+            Arg::new("display_full_paths")
+                .short('p')
+                .long("full-paths")
+                .help("Subdirectories will not have their path shortened"),
+        )
+        .arg(
+            Arg::new("ignore_directory")
+                .short('X')
+                .long("ignore-directory")
+                .takes_value(true)
+                .number_of_values(1)
+                .multiple_occurrences(true)
+                .help("Exclude any file or directory with this name"),
+        )
+        .arg(
+            Arg::new("limit_filesystem")
+                .short('x')
+                .long("limit-filesystem")
+                .help("Only count the files and directories on the same filesystem as the supplied directory"),
+        )
+        .arg(
+            Arg::new("display_apparent_size")
+                .short('s')
+                .long("apparent-size")
+                .help("Use file length instead of blocks"),
+        )
+        .arg(
+            Arg::new("reverse")
+                .short('r')
+                .long("reverse")
+                .help("Print tree upside down (biggest highest)"),
+        )
+        .arg(
+            Arg::new("no_colors")
+                .short('c')
+                .long("no-colors")
+                .help("No colors will be printed (Useful for commands like: watch)"),
+        )
+        .arg(
+            Arg::new("no_bars")
+                .short('b')
+                .long("no-percent-bars")
+                .help("No percent bars or percentages will be displayed"),
+        )
+        .arg(
+            Arg::new("min_size")
+                .short('z')
+                .long("min-size")
+                .takes_value(true)
+                .number_of_values(1)
+                .help("Minimum size file to include in output"),
+        )
+        .arg(
+            Arg::new("skip_total")
+                .long("skip-total")
+                .help("No total row will be displayed"),
+        )
+        .arg(
+            Arg::new("by_filecount")
+                .short('f')
+                .long("filecount")
+                .help("Directory 'size' is number of child files/dirs not disk size"),
+        )
+        .arg(
+            Arg::new("ignore_hidden")
+                .short('i') // Do not use 'h' this is used by 'help'
+                .long("ignore_hidden")
+                .help("Do not display hidden files"),
+        )
+        .arg(
+            Arg::new("invert_filter")
+                .short('v')
+                .long("invert-filter")
+                .takes_value(true)
+                .number_of_values(1)
+                .multiple_occurrences(true)
+                .conflicts_with("filter")
+                .conflicts_with("types")
+                .help("Exclude filepaths matching this regex. To ignore png files type: -v \"\\.png$\" "),
+        )
+        .arg(
+            Arg::new("filter")
+                .short('e')
+                .long("filter")
+                .takes_value(true)
+                .number_of_values(1)
+                .multiple_occurrences(true)
+                .conflicts_with("types")
+                .help("Only include filepaths matching this regex. For png files type: -e \"\\.png$\" "),
+        )
+        .arg(
+            Arg::new("types")
+                .short('t')
+                .long("file_types")
+                .conflicts_with("depth")
+                .help("show only these file types"),
+        )
+        .arg(
+            Arg::new("width")
+                .short('w')
+                .long("terminal_width")
+                .takes_value(true)
+                .number_of_values(1)
+                .help("Specify width of output overriding the auto detection of terminal width"),
+        )
+        .arg(
+            Arg::new("iso")
+                .short('H')
+                .long("si")
+                .help("print sizes in powers of 1000 (e.g., 1.1G)")
+        )
+        .arg(Arg::new("inputs").multiple_occurrences(true).default_value("."))
+}

src/config.rs

@@ -0,0 +1,148 @@
+use clap::ArgMatches;
+use config_file::FromConfigFile;
+use serde::Deserialize;
+use std::path::Path;
+use std::path::PathBuf;
+
+use crate::display::UNITS;
+
+#[derive(Deserialize, Default)]
+#[serde(rename_all = "kebab-case")]
+#[serde(deny_unknown_fields)]
+pub struct Config {
+    pub display_full_paths: Option<bool>,
+    pub display_apparent_size: Option<bool>,
+    pub reverse: Option<bool>,
+    pub no_colors: Option<bool>,
+    pub no_bars: Option<bool>,
+    pub skip_total: Option<bool>,
+    pub ignore_hidden: Option<bool>,
+    pub iso: Option<bool>,
+    pub min_size: Option<String>,
+}
+
+impl Config {
+    pub fn get_no_colors(&self, options: &ArgMatches) -> bool {
+        Some(true) == self.no_colors || options.is_present("no_colors")
+    }
+    pub fn get_apparent_size(&self, options: &ArgMatches) -> bool {
+        Some(true) == self.display_apparent_size || options.is_present("display_apparent_size")
+    }
+    pub fn get_ignore_hidden(&self, options: &ArgMatches) -> bool {
+        Some(true) == self.ignore_hidden || options.is_present("ignore_hidden")
+    }
+    pub fn get_full_paths(&self, options: &ArgMatches) -> bool {
+        Some(true) == self.display_full_paths || options.is_present("display_full_paths")
+    }
+    pub fn get_reverse(&self, options: &ArgMatches) -> bool {
+        Some(true) == self.reverse || options.is_present("reverse")
+    }
+    pub fn get_no_bars(&self, options: &ArgMatches) -> bool {
+        Some(true) == self.no_bars || options.is_present("no_bars")
+    }
+    pub fn get_iso(&self, options: &ArgMatches) -> bool {
+        Some(true) == self.iso || options.is_present("iso")
+    }
+    pub fn get_skip_total(&self, options: &ArgMatches) -> bool {
+        Some(true) == self.skip_total || options.is_present("skip_total")
+    }
+    pub fn get_min_size(&self, options: &ArgMatches, iso: bool) -> Option<usize> {
+        let size_from_param = options.value_of("min_size");
+        self._get_min_size(size_from_param, iso)
+    }
+    fn _get_min_size(&self, min_size: Option<&str>, iso: bool) -> Option<usize> {
+        let size_from_param = min_size.and_then(|a| convert_min_size(a, iso));
+
+        if size_from_param.is_none() {
+            self.min_size
+                .as_ref()
+                .and_then(|a| convert_min_size(a.as_ref(), iso))
+        } else {
+            size_from_param
+        }
+    }
+}
+
+fn convert_min_size(input: &str, iso: bool) -> Option<usize> {
+    let chars_as_vec: Vec<char> = input.chars().collect();
+    match chars_as_vec.split_last() {
+        Some((last, start)) => {
+            let mut starts: String = start.iter().collect::<String>();
+
+            for (i, u) in UNITS.iter().rev().enumerate() {
+                if Some(*u) == last.to_uppercase().next() {
+                    return match starts.parse::<usize>() {
+                        Ok(pure) => {
+                            let num: usize = if iso { 1000 } else { 1024 };
+                            let marker = pure * num.pow((i + 1) as u32);
+                            Some(marker)
+                        }
+                        Err(_) => {
+                            eprintln!("Ignoring invalid min-size: {}", input);
+                            None
+                        }
+                    };
+                }
+            }
+            starts.push(*last);
+            starts
+                .parse()
+                .map_err(|_| {
+                    eprintln!("Ignoring invalid min-size: {}", input);
+                })
+                .ok()
+        }
+        None => None,
+    }
+}
+
+fn get_config_locations(base: &Path) -> Vec<PathBuf> {
+    vec![
+        base.join(".dust.toml"),
+        base.join(".config").join("dust").join("config.toml"),
+    ]
+}
+
+pub fn get_config() -> Config {
+    if let Some(home) = directories::BaseDirs::new() {
+        for path in get_config_locations(home.home_dir()) {
+            if path.exists() {
+                if let Ok(config) = Config::from_config_file(path) {
+                    return config;
+                }
+            }
+        }
+    }
+    Config {
+        ..Default::default()
+    }
+}
+
+mod tests {
+    #[allow(unused_imports)]
+    use super::*;
+
+    #[test]
+    fn test_conversion() {
+        assert_eq!(convert_min_size("55", false), Some(55));
+        assert_eq!(convert_min_size("12344321", false), Some(12344321));
+        assert_eq!(convert_min_size("95RUBBISH", false), None);
+        assert_eq!(convert_min_size("10K", false), Some(10 * 1024));
+        assert_eq!(convert_min_size("10M", false), Some(10 * 1024usize.pow(2)));
+        assert_eq!(convert_min_size("10M", true), Some(10 * 1000usize.pow(2)));
+        assert_eq!(convert_min_size("2G", false), Some(2 * 1024usize.pow(3)));
+    }
+
+    #[test]
+    fn test_min_size_from_config_applied_or_overridden() {
+        let c = Config {
+            min_size: Some("1K".to_owned()),
+            ..Default::default()
+        };
+        assert_eq!(c._get_min_size(None, false), Some(1024));
+        assert_eq!(c._get_min_size(Some("2K"), false), Some(2048));
+
+        assert_eq!(c._get_min_size(None, true), Some(1000));
+        assert_eq!(c._get_min_size(Some("2K"), true), Some(2000));
+    }
+}

src/dir_walker.rs

@@ -0,0 +1,226 @@
+use std::fs;
+
+use crate::node::Node;
+use crate::utils::is_filtered_out_due_to_invert_regex;
+use crate::utils::is_filtered_out_due_to_regex;
+use rayon::iter::ParallelBridge;
+use rayon::prelude::ParallelIterator;
+use regex::Regex;
+use std::path::PathBuf;
+
+use std::sync::atomic;
+use std::sync::atomic::AtomicBool;
+
+use std::collections::HashSet;
+
+use crate::node::build_node;
+use std::fs::DirEntry;
+
+use crate::platform::get_metadata;
+
+pub struct WalkData<'a> {
+    pub ignore_directories: HashSet<PathBuf>,
+    pub filter_regex: &'a [Regex],
+    pub invert_filter_regex: &'a [Regex],
+    pub allowed_filesystems: HashSet<u64>,
+    pub use_apparent_size: bool,
+    pub by_filecount: bool,
+    pub ignore_hidden: bool,
+}
+
+pub fn walk_it(dirs: HashSet<PathBuf>, walk_data: WalkData) -> (Vec<Node>, bool) {
+    let permissions_flag = AtomicBool::new(false);
+
+    let top_level_nodes: Vec<_> = dirs
+        .into_iter()
+        .filter_map(|d| {
+            clean_inodes(
+                walk(d, &permissions_flag, &walk_data, 0)?,
+                &mut HashSet::new(),
+                walk_data.use_apparent_size,
+            )
+        })
+        .collect();
+    (top_level_nodes, permissions_flag.into_inner())
+}
+
+// Remove files which have the same inode, we don't want to double count them.
+fn clean_inodes(
+    x: Node,
+    inodes: &mut HashSet<(u64, u64)>,
+    use_apparent_size: bool,
+) -> Option<Node> {
+    if !use_apparent_size {
+        if let Some(id) = x.inode_device {
+            if !inodes.insert(id) {
+                return None;
+            }
+        }
+    }
+
+    // Sort Nodes so iteration order is predictable
+    let mut tmp: Vec<_> = x.children;
+    tmp.sort_by(sort_by_inode);
+    let new_children: Vec<_> = tmp
+        .into_iter()
+        .filter_map(|c| clean_inodes(c, inodes, use_apparent_size))
+        .collect();
+
+    Some(Node {
+        name: x.name,
+        size: x.size + new_children.iter().map(|c| c.size).sum::<u64>(),
+        children: new_children,
+        inode_device: x.inode_device,
+        depth: x.depth,
+    })
+}
+
+fn sort_by_inode(a: &Node, b: &Node) -> std::cmp::Ordering {
+    // Sorting by inode is quicker than by sorting by name/size
+    if let Some(x) = a.inode_device {
+        if let Some(y) = b.inode_device {
+            if x.0 != y.0 {
+                return x.0.cmp(&y.0);
+            } else if x.1 != y.1 {
+                return x.1.cmp(&y.1);
+            }
+        }
+    }
+    a.name.cmp(&b.name)
+}
+
+fn ignore_file(entry: &DirEntry, walk_data: &WalkData) -> bool {
+    let is_dot_file = entry.file_name().to_str().unwrap_or("").starts_with('.');
+    let is_ignored_path = walk_data.ignore_directories.contains(&entry.path());
+
+    if !walk_data.allowed_filesystems.is_empty() {
+        let size_inode_device = get_metadata(&entry.path(), false);
+
+        if let Some((_size, Some((_id, dev)))) = size_inode_device {
+            if !walk_data.allowed_filesystems.contains(&dev) {
+                return true;
+            }
+        }
+    }
+
+    // Keeping `walk_data.filter_regex.is_empty()` is important for performance reasons, it stops unnecessary work
+    if !walk_data.filter_regex.is_empty()
+        && entry.path().is_file()
+        && is_filtered_out_due_to_regex(walk_data.filter_regex, &entry.path())
+    {
+        return true;
+    }
+
+    if !walk_data.invert_filter_regex.is_empty()
+        && entry.path().is_file()
+        && is_filtered_out_due_to_invert_regex(walk_data.invert_filter_regex, &entry.path())
+    {
+        return true;
+    }
+
+    (is_dot_file && walk_data.ignore_hidden) || is_ignored_path
+}
+
+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) {
+        children = entries
+            .into_iter()
+            .par_bridge()
+            .filter_map(|entry| {
+                if let Ok(ref entry) = entry {
+                    // uncommenting the below line gives simpler code but
+                    // rayon doesn't parallelize as well giving a 3X performance drop
+                    // hence we unravel the recursion a bit
+
+                    // return walk(entry.path(), permissions_flag, ignore_directories, allowed_filesystems, use_apparent_size, by_filecount, ignore_hidden);
+
+                    if !ignore_file(entry, walk_data) {
+                        if let Ok(data) = entry.file_type() {
+                            return if data.is_dir() && !data.is_symlink() {
+                                walk(entry.path(), permissions_flag, walk_data, depth + 1)
+                            } else {
+                                build_node(
+                                    entry.path(),
+                                    vec![],
+                                    walk_data.filter_regex,
+                                    walk_data.invert_filter_regex,
+                                    walk_data.use_apparent_size,
+                                    data.is_symlink(),
+                                    data.is_file(),
+                                    walk_data.by_filecount,
+                                    depth,
+                                )
+                            };
+                        }
+                    }
+                } else {
+                    permissions_flag.store(true, atomic::Ordering::Relaxed);
+                }
+                None
+            })
+            .collect();
+    } else {
+        // Handle edge case where dust is called with a file instead of a directory
+        if !dir.exists() {
+            permissions_flag.store(true, atomic::Ordering::Relaxed);
+        }
+    }
+    build_node(
+        dir,
+        children,
+        walk_data.filter_regex,
+        walk_data.invert_filter_regex,
+        walk_data.use_apparent_size,
+        false,
+        false,
+        walk_data.by_filecount,
+        depth,
+    )
+}
+
+mod tests {
+    #[allow(unused_imports)]
+    use super::*;
+
+    #[cfg(test)]
+    fn create_node() -> Node {
+        Node {
+            name: PathBuf::new(),
+            size: 10,
+            children: vec![],
+            inode_device: Some((5, 6)),
+            depth: 0,
+        }
+    }
+
+    #[test]
+    #[allow(clippy::redundant_clone)]
+    fn test_should_ignore_file() {
+        let mut inodes = HashSet::new();
+        let n = create_node();
+
+        // First time we insert the node
+        assert_eq!(clean_inodes(n.clone(), &mut inodes, false), Some(n.clone()));
+
+        // Second time is a duplicate - we ignore it
+        assert_eq!(clean_inodes(n.clone(), &mut inodes, false), None);
+    }
+
+    #[test]
+    #[allow(clippy::redundant_clone)]
+    fn test_should_not_ignore_files_if_using_apparent_size() {
+        let mut inodes = HashSet::new();
+        let n = create_node();
+
+        // If using apparent size we include Nodes, even if duplicate inodes
+        assert_eq!(clean_inodes(n.clone(), &mut inodes, true), Some(n.clone()));
+        assert_eq!(clean_inodes(n.clone(), &mut inodes, true), Some(n.clone()));
+    }
+}

src/display.rs

@@ -0,0 +1,476 @@
+use crate::display_node::DisplayNode;
+
+use ansi_term::Colour::Red;
+use lscolors::{LsColors, Style};
+
+use unicode_width::UnicodeWidthStr;
+
+use stfu8::encode_u8;
+
+use std::cmp::max;
+use std::cmp::min;
+use std::fs;
+use std::iter::repeat;
+use std::path::Path;
+use thousands::Separable;
+
+pub static UNITS: [char; 4] = ['T', 'G', 'M', 'K'];
+static BLOCKS: [char; 5] = ['█', '▓', '▒', '░', ' '];
+
+pub struct DisplayData {
+    pub short_paths: bool,
+    pub is_reversed: bool,
+    pub colors_on: bool,
+    pub by_filecount: bool,
+    pub num_chars_needed_on_left_most: usize,
+    pub base_size: u64,
+    pub longest_string_length: usize,
+    pub ls_colors: LsColors,
+    pub iso: bool,
+}
+
+impl DisplayData {
+    fn get_tree_chars(&self, was_i_last: bool, has_children: bool) -> &'static str {
+        match (self.is_reversed, was_i_last, has_children) {
+            (true, true, true) => "┌─┴",
+            (true, true, false) => "┌──",
+            (true, false, true) => "├─┴",
+            (true, false, false) => "├──",
+            (false, true, true) => "└─┬",
+            (false, true, false) => "└──",
+            (false, false, true) => "├─┬",
+            (false, false, false) => "├──",
+        }
+    }
+
+    fn is_biggest(&self, num_siblings: usize, max_siblings: u64) -> bool {
+        if self.is_reversed {
+            num_siblings == (max_siblings - 1) as usize
+        } else {
+            num_siblings == 0
+        }
+    }
+
+    fn is_last(&self, num_siblings: usize, max_siblings: u64) -> bool {
+        if self.is_reversed {
+            num_siblings == 0
+        } else {
+            num_siblings == (max_siblings - 1) as usize
+        }
+    }
+
+    fn percent_size(&self, node: &DisplayNode) -> f32 {
+        let result = node.size as f32 / self.base_size as f32;
+        if result.is_normal() {
+            result
+        } else {
+            0.0
+        }
+    }
+}
+
+struct DrawData<'a> {
+    indent: String,
+    percent_bar: String,
+    display_data: &'a DisplayData,
+}
+
+impl DrawData<'_> {
+    fn get_new_indent(&self, has_children: bool, was_i_last: bool) -> String {
+        let chars = self.display_data.get_tree_chars(was_i_last, has_children);
+        self.indent.to_string() + chars
+    }
+
+    // TODO: can we test this?
+    fn generate_bar(&self, node: &DisplayNode, level: usize) -> String {
+        let chars_in_bar = self.percent_bar.chars().count();
+        let num_bars = chars_in_bar as f32 * self.display_data.percent_size(node);
+        let mut num_not_my_bar = (chars_in_bar as i32) - num_bars as i32;
+
+        let mut new_bar = "".to_string();
+        let idx = 5 - min(4, max(1, level));
+
+        for c in self.percent_bar.chars() {
+            num_not_my_bar -= 1;
+            if num_not_my_bar <= 0 {
+                new_bar.push(BLOCKS[0]);
+            } else if c == BLOCKS[0] {
+                new_bar.push(BLOCKS[idx]);
+            } else {
+                new_bar.push(c);
+            }
+        }
+        new_bar
+    }
+}
+
+#[allow(clippy::too_many_arguments)]
+pub fn draw_it(
+    use_full_path: bool,
+    is_reversed: bool,
+    no_colors: bool,
+    no_percent_bars: bool,
+    terminal_width: usize,
+    by_filecount: bool,
+    root_node: &DisplayNode,
+    iso: bool,
+    skip_total: bool,
+) {
+    let biggest = match skip_total {
+        false => root_node,
+        true => root_node
+            .get_children_from_node(false)
+            .next()
+            .unwrap_or(root_node),
+    };
+
+    let num_chars_needed_on_left_most = if by_filecount {
+        let max_size = biggest.size;
+        max_size.separate_with_commas().chars().count()
+    } else {
+        find_biggest_size_str(root_node, iso)
+    };
+
+    assert!(
+        terminal_width > num_chars_needed_on_left_most + 2,
+        "Not enough terminal width"
+    );
+
+    let allowed_width = terminal_width - num_chars_needed_on_left_most - 2;
+    let num_indent_chars = 3;
+    let longest_string_length =
+        find_longest_dir_name(root_node, num_indent_chars, allowed_width, !use_full_path);
+
+    let max_bar_length = if no_percent_bars || longest_string_length + 7 >= allowed_width as usize {
+        0
+    } else {
+        allowed_width as usize - longest_string_length - 7
+    };
+
+    let first_size_bar = repeat(BLOCKS[0]).take(max_bar_length).collect();
+
+    let display_data = DisplayData {
+        short_paths: !use_full_path,
+        is_reversed,
+        colors_on: !no_colors,
+        by_filecount,
+        num_chars_needed_on_left_most,
+        base_size: biggest.size,
+        longest_string_length,
+        ls_colors: LsColors::from_env().unwrap_or_default(),
+        iso,
+    };
+    let draw_data = DrawData {
+        indent: "".to_string(),
+        percent_bar: first_size_bar,
+        display_data: &display_data,
+    };
+
+    if !skip_total {
+        display_node(root_node, &draw_data, true, true);
+    } else {
+        for (count, c) in root_node
+            .get_children_from_node(draw_data.display_data.is_reversed)
+            .enumerate()
+        {
+            let is_biggest = display_data.is_biggest(count, root_node.num_siblings());
+            let was_i_last = display_data.is_last(count, root_node.num_siblings());
+            display_node(c, &draw_data, is_biggest, was_i_last);
+        }
+    }
+}
+
+fn find_biggest_size_str(node: &DisplayNode, iso: bool) -> usize {
+    let mut mx = human_readable_number(node.size, iso).chars().count();
+    for n in node.children.iter() {
+        mx = max(mx, find_biggest_size_str(n, iso));
+    }
+    mx
+}
+
+fn find_longest_dir_name(
+    node: &DisplayNode,
+    indent: usize,
+    terminal: usize,
+    long_paths: bool,
+) -> usize {
+    let printable_name = get_printable_name(&node.name, long_paths);
+    let longest = min(
+        UnicodeWidthStr::width(&*printable_name) + 1 + indent,
+        terminal,
+    );
+
+    // each none root tree drawing is 2 more chars, hence we increment indent by 2
+    node.children
+        .iter()
+        .map(|c| find_longest_dir_name(c, indent + 2, terminal, long_paths))
+        .fold(longest, max)
+}
+
+fn display_node(node: &DisplayNode, draw_data: &DrawData, is_biggest: bool, is_last: bool) {
+    // hacky way of working out how deep we are in the tree
+    let indent = draw_data.get_new_indent(!node.children.is_empty(), is_last);
+    let level = ((indent.chars().count() - 1) / 2) - 1;
+    let bar_text = draw_data.generate_bar(node, level);
+
+    let to_print = format_string(node, &indent, &bar_text, is_biggest, draw_data.display_data);
+
+    if !draw_data.display_data.is_reversed {
+        println!("{}", to_print)
+    }
+
+    let dd = DrawData {
+        indent: clean_indentation_string(&indent),
+        percent_bar: bar_text,
+        display_data: draw_data.display_data,
+    };
+
+    let num_siblings = node.num_siblings();
+
+    for (count, c) in node
+        .get_children_from_node(draw_data.display_data.is_reversed)
+        .enumerate()
+    {
+        let is_biggest = dd.display_data.is_biggest(count, num_siblings);
+        let was_i_last = dd.display_data.is_last(count, num_siblings);
+        display_node(c, &dd, is_biggest, was_i_last);
+    }
+
+    if draw_data.display_data.is_reversed {
+        println!("{}", to_print)
+    }
+}
+
+fn clean_indentation_string(s: &str) -> String {
+    let mut is: String = s.into();
+    // For reversed:
+    is = is.replace("┌─┴", "  ");
+    is = is.replace("┌──", "  ");
+    is = is.replace("├─┴", "│ ");
+    is = is.replace("─┴", " ");
+    // For normal
+    is = is.replace("└─┬", "  ");
+    is = is.replace("└──", "  ");
+    is = is.replace("├─┬", "│ ");
+    is = is.replace("─┬", " ");
+    // For both
+    is = is.replace("├──", "│ ");
+    is
+}
+
+fn get_printable_name<P: AsRef<Path>>(dir_name: &P, long_paths: bool) -> String {
+    let dir_name = dir_name.as_ref();
+    let printable_name = {
+        if long_paths {
+            match dir_name.parent() {
+                Some(prefix) => match dir_name.strip_prefix(prefix) {
+                    Ok(base) => base,
+                    Err(_) => dir_name,
+                },
+                None => dir_name,
+            }
+        } else {
+            dir_name
+        }
+    };
+    encode_u8(printable_name.display().to_string().as_bytes())
+}
+
+fn pad_or_trim_filename(node: &DisplayNode, indent: &str, display_data: &DisplayData) -> String {
+    let name = get_printable_name(&node.name, display_data.short_paths);
+    let indent_and_name = format!("{} {}", indent, name);
+    let width = UnicodeWidthStr::width(&*indent_and_name);
+
+    assert!(
+        display_data.longest_string_length >= width,
+        "Terminal width not wide enough to draw directory tree"
+    );
+
+    // Add spaces after the filename so we can draw the % used bar chart.
+    let name_and_padding = name
+        + " "
+            .repeat(display_data.longest_string_length - width)
+            .as_str();
+
+    name_and_padding
+}
+
+fn maybe_trim_filename(name_in: String, indent: &str, display_data: &DisplayData) -> String {
+    let indent_length = UnicodeWidthStr::width(indent);
+    assert!(
+        display_data.longest_string_length >= indent_length + 2,
+        "Terminal width not wide enough to draw directory tree"
+    );
+
+    let max_size = display_data.longest_string_length - indent_length;
+    if UnicodeWidthStr::width(&*name_in) > max_size {
+        let name = name_in.chars().take(max_size - 2).collect::<String>();
+        name + ".."
+    } else {
+        name_in
+    }
+}
+
+pub fn format_string(
+    node: &DisplayNode,
+    indent: &str,
+    percent_bar: &str,
+    is_biggest: bool,
+    display_data: &DisplayData,
+) -> String {
+    let (percents, name_and_padding) = get_name_percent(node, indent, percent_bar, display_data);
+    let pretty_size = get_pretty_size(node, is_biggest, display_data);
+    let pretty_name = get_pretty_name(node, name_and_padding, display_data);
+    format!("{} {} {}{}", pretty_size, indent, pretty_name, percents)
+}
+
+fn get_name_percent(
+    node: &DisplayNode,
+    indent: &str,
+    bar_chart: &str,
+    display_data: &DisplayData,
+) -> (String, String) {
+    if !bar_chart.is_empty() {
+        let percent_size_str = format!("{:.0}%", display_data.percent_size(node) * 100.0);
+        let percents = format!("│{} │ {:>4}", bar_chart, percent_size_str);
+        let name_and_padding = pad_or_trim_filename(node, indent, display_data);
+        (percents, name_and_padding)
+    } else {
+        let n = get_printable_name(&node.name, display_data.short_paths);
+        let name = maybe_trim_filename(n, indent, display_data);
+        ("".into(), name)
+    }
+}
+
+fn get_pretty_size(node: &DisplayNode, is_biggest: bool, display_data: &DisplayData) -> String {
+    let output = if display_data.by_filecount {
+        node.size.separate_with_commas()
+    } else {
+        human_readable_number(node.size, display_data.iso)
+    };
+    let spaces_to_add = display_data.num_chars_needed_on_left_most - output.chars().count();
+    let output = " ".repeat(spaces_to_add) + output.as_str();
+
+    if is_biggest && display_data.colors_on {
+        format!("{}", Red.paint(output))
+    } else {
+        output
+    }
+}
+
+fn get_pretty_name(
+    node: &DisplayNode,
+    name_and_padding: String,
+    display_data: &DisplayData,
+) -> String {
+    if display_data.colors_on {
+        let meta_result = fs::metadata(&node.name);
+        let directory_color = display_data
+            .ls_colors
+            .style_for_path_with_metadata(&node.name, meta_result.as_ref().ok());
+        let ansi_style = directory_color
+            .map(Style::to_ansi_term_style)
+            .unwrap_or_default();
+        format!("{}", ansi_style.paint(name_and_padding))
+    } else {
+        name_and_padding
+    }
+}
+
+fn human_readable_number(size: u64, iso: bool) -> String {
+    for (i, u) in UNITS.iter().enumerate() {
+        let num: u64 = if iso { 1000 } else { 1024 };
+        let marker = num.pow((UNITS.len() - i) as u32);
+        if size >= marker {
+            if size / marker < 10 {
+                return format!("{:.1}{}", (size as f32 / marker as f32), u);
+            } else {
+                return format!("{}{}", (size / marker), u);
+            }
+        }
+    }
+    format!("{}B", size)
+}
+
+mod tests {
+    #[allow(unused_imports)]
+    use super::*;
+    #[allow(unused_imports)]
+    use std::path::PathBuf;
+
+    #[cfg(test)]
+    fn get_fake_display_data(longest_string_length: usize) -> DisplayData {
+        DisplayData {
+            short_paths: true,
+            is_reversed: false,
+            colors_on: false,
+            by_filecount: false,
+            num_chars_needed_on_left_most: 5,
+            base_size: 1,
+            longest_string_length,
+            ls_colors: LsColors::from_env().unwrap_or_default(),
+            iso: false,
+        }
+    }
+
+    #[test]
+    fn test_format_str() {
+        let n = DisplayNode {
+            name: PathBuf::from("/short"),
+            size: 2_u64.pow(12), // This is 4.0K
+            children: vec![],
+        };
+        let indent = "┌─┴";
+        let percent_bar = "";
+        let is_biggest = false;
+
+        let s = format_string(
+            &n,
+            indent,
+            percent_bar,
+            is_biggest,
+            &get_fake_display_data(20),
+        );
+        assert_eq!(s, " 4.0K ┌─┴ short");
+    }
+
+    #[test]
+    fn test_format_str_long_name() {
+        let name = "very_long_name_longer_than_the_eighty_character_limit_very_long_name_this_bit_will_truncate";
+        let n = DisplayNode {
+            name: PathBuf::from(name),
+            size: 2_u64.pow(12), // This is 4.0K
+            children: vec![],
+        };
+        let indent = "┌─┴";
+        let percent_bar = "";
+        let is_biggest = false;
+
+        let dd = get_fake_display_data(64);
+        let s = format_string(&n, indent, percent_bar, is_biggest, &dd);
+        assert_eq!(
+            s,
+            " 4.0K ┌─┴ very_long_name_longer_than_the_eighty_character_limit_very_.."
+        );
+    }
+
+    #[test]
+    fn test_human_readable_number() {
+        assert_eq!(human_readable_number(1, false), "1B");
+        assert_eq!(human_readable_number(956, false), "956B");
+        assert_eq!(human_readable_number(1004, false), "1004B");
+        assert_eq!(human_readable_number(1024, false), "1.0K");
+        assert_eq!(human_readable_number(1536, false), "1.5K");
+        assert_eq!(human_readable_number(1024 * 512, false), "512K");
+        assert_eq!(human_readable_number(1024 * 1024, false), "1.0M");
+        assert_eq!(
+            human_readable_number(1024 * 1024 * 1024 - 1, false),
+            "1023M"
+        );
+        assert_eq!(human_readable_number(1024 * 1024 * 1024 * 20, false), "20G");
+        assert_eq!(
+            human_readable_number(1024 * 1024 * 1024 * 1024, false),
+            "1.0T"
+        );
+    }
+}

src/display_node.rs

@@ -0,0 +1,25 @@
+use std::path::PathBuf;
+
+#[derive(Debug, PartialEq, Eq, PartialOrd, Ord, Clone)]
+pub struct DisplayNode {
+    // Note: the order of fields in important here, for PartialEq and PartialOrd
+    pub size: u64,
+    pub name: PathBuf, //todo: consider moving to a string?
+    pub children: Vec<DisplayNode>,
+}
+
+impl DisplayNode {
+    pub fn num_siblings(&self) -> u64 {
+        self.children.len() as u64
+    }
+
+    pub fn get_children_from_node(&self, is_reversed: bool) -> impl Iterator<Item = &DisplayNode> {
+        // we box to avoid the clippy lint warning
+        let out: Box<dyn Iterator<Item = &DisplayNode>> = if is_reversed {
+            Box::new(self.children.iter().rev())
+        } else {
+            Box::new(self.children.iter())
+        };
+        out
+    }
+}

src/filter.rs

@@ -0,0 +1,87 @@
+use crate::display_node::DisplayNode;
+use crate::node::Node;
+use std::collections::BinaryHeap;
+use std::collections::HashSet;
+use std::path::Path;
+use std::path::PathBuf;
+
+pub fn get_biggest(
+    top_level_nodes: Vec<Node>,
+    min_size: Option<usize>,
+    n: usize,
+    depth: usize,
+    using_a_filter: bool,
+) -> Option<DisplayNode> {
+    if top_level_nodes.is_empty() {
+        // perhaps change this, bring back Error object?
+        return None;
+    }
+
+    let mut heap = BinaryHeap::new();
+    let number_top_level_nodes = top_level_nodes.len();
+    let root = get_new_root(top_level_nodes);
+    let mut allowed_nodes = HashSet::new();
+
+    allowed_nodes.insert(root.name.as_path());
+    heap = add_children(using_a_filter, min_size, &root, depth, heap);
+
+    for _ in number_top_level_nodes..n {
+        let line = heap.pop();
+        match line {
+            Some(line) => {
+                allowed_nodes.insert(line.name.as_path());
+                heap = add_children(using_a_filter, min_size, line, depth, heap);
+            }
+            None => break,
+        }
+    }
+    recursive_rebuilder(&allowed_nodes, &root)
+}
+
+fn add_children<'a>(
+    using_a_filter: bool,
+    min_size: Option<usize>,
+    file_or_folder: &'a Node,
+    depth: usize,
+    mut heap: BinaryHeap<&'a Node>,
+) -> BinaryHeap<&'a Node> {
+    if depth > file_or_folder.depth {
+        heap.extend(file_or_folder.children.iter().filter(|c| match min_size {
+            Some(ms) => c.size > ms as u64,
+            None => !using_a_filter || c.name.is_file() || c.size > 0,
+        }))
+    }
+    heap
+}
+
+fn get_new_root(top_level_nodes: Vec<Node>) -> Node {
+    if top_level_nodes.len() != 1 {
+        let size = top_level_nodes.iter().map(|node| node.size).sum();
+        Node {
+            name: PathBuf::from("(total)"),
+            size,
+            children: top_level_nodes,
+            inode_device: None,
+            depth: 0,
+        }
+    } else {
+        top_level_nodes.into_iter().next().unwrap()
+    }
+}
+
+fn recursive_rebuilder(allowed_nodes: &HashSet<&Path>, current: &Node) -> Option<DisplayNode> {
+    let mut new_children: Vec<_> = current
+        .children
+        .iter()
+        .filter(|c| allowed_nodes.contains(c.name.as_path()))
+        .filter_map(|c| recursive_rebuilder(allowed_nodes, c))
+        .collect();
+
+    new_children.sort_by(|lhs, rhs| lhs.cmp(rhs).reverse());
+
+    Some(DisplayNode {
+        name: current.name.clone(),
+        size: current.size,
+        children: new_children,
+    })
+}

src/filter_type.rs

@@ -0,0 +1,75 @@
+use crate::display_node::DisplayNode;
+use crate::node::Node;
+use std::collections::HashMap;
+use std::ffi::OsStr;
+use std::path::PathBuf;
+
+#[derive(PartialEq, Eq, PartialOrd, Ord)]
+struct ExtensionNode<'a> {
+    size: u64,
+    extension: Option<&'a OsStr>,
+}
+
+pub fn get_all_file_types(top_level_nodes: &[Node], n: usize) -> Option<DisplayNode> {
+    let ext_nodes = {
+        let mut extension_cumulative_sizes = HashMap::new();
+        build_by_all_file_types(top_level_nodes, &mut extension_cumulative_sizes);
+
+        let mut extension_cumulative_sizes: Vec<ExtensionNode<'_>> = extension_cumulative_sizes
+            .iter()
+            .map(|(&extension, &size)| ExtensionNode { extension, size })
+            .collect();
+
+        extension_cumulative_sizes.sort_by(|lhs, rhs| lhs.cmp(rhs).reverse());
+
+        extension_cumulative_sizes
+    };
+
+    let mut ext_nodes_iter = ext_nodes.iter();
+
+    // First, collect the first N - 1 nodes...
+    let mut displayed: Vec<DisplayNode> = ext_nodes_iter
+        .by_ref()
+        .take(if n > 1 { n - 1 } else { 1 })
+        .map(|node| DisplayNode {
+            name: PathBuf::from(
+                node.extension
+                    .map(|ext| format!(".{}", ext.to_string_lossy()))
+                    .unwrap_or_else(|| "(no extension)".to_owned()),
+            ),
+            size: node.size,
+            children: vec![],
+        })
+        .collect();
+
+    // ...then, aggregate the remaining nodes (if any) into a single  "(others)" node
+    if ext_nodes_iter.len() > 0 {
+        displayed.push(DisplayNode {
+            name: PathBuf::from("(others)"),
+            size: ext_nodes_iter.map(|node| node.size).sum(),
+            children: vec![],
+        });
+    }
+
+    let result = DisplayNode {
+        name: PathBuf::from("(total)"),
+        size: displayed.iter().map(|node| node.size).sum(),
+        children: displayed,
+    };
+
+    Some(result)
+}
+
+fn build_by_all_file_types<'a>(
+    top_level_nodes: &'a [Node],
+    counter: &mut HashMap<Option<&'a OsStr>, u64>,
+) {
+    for node in top_level_nodes {
+        if node.name.is_file() {
+            let ext = node.name.extension();
+            let cumulative_size = counter.entry(ext).or_default();
+            *cumulative_size += node.size;
+        }
+        build_by_all_file_types(&node.children, counter)
+    }
+}

src/main.rs

@@ -1,364 +1,201 @@
-// test:
-// recursive dirs that link to each other.
-// Pass in bad dir name
-// num to search for is less than num available
-// admin files.
-//
-extern crate ansi_term;
-#[macro_use]
-extern crate clap;
+mod cli;
+mod config;
+mod dir_walker;
+mod display;
+mod display_node;
+mod filter;
+mod filter_type;
+mod node;
+mod platform;
+mod utils;
 
+use crate::cli::build_cli;
 use std::collections::HashSet;
+use std::panic;
+use std::process;
 
-use ansi_term::Colour::Fixed;
-use clap::{App, AppSettings, Arg};
+use self::display::draw_it;
+use clap::Values;
+use config::get_config;
+use dir_walker::{walk_it, WalkData};
+use filter::get_biggest;
+use filter_type::get_all_file_types;
+use rayon::ThreadPoolBuildError;
+use regex::Regex;
+use std::cmp::max;
+use std::path::PathBuf;
+use terminal_size::{terminal_size, Height, Width};
+use utils::get_filesystem_devices;
+use utils::simplify_dir_names;
 
-use std::cmp;
-use std::cmp::Ordering;
-use std::fs;
-use std::fs::ReadDir;
-use std::io;
+static DEFAULT_NUMBER_OF_LINES: usize = 30;
+static DEFAULT_TERMINAL_WIDTH: usize = 80;
 
-#[derive(Clone, Debug)]
-struct Node {
-    dir: Dir,
-    children: Vec<Node>,
-}
-impl Ord for Node {
-    fn cmp(&self, other: &Self) -> Ordering {
-        if self.dir.size > other.dir.size {
-            Ordering::Less
-        } else if self.dir.size < other.dir.size {
-            Ordering::Greater
+fn init_color(no_color: bool) -> bool {
+    #[cfg(windows)]
+    {
+        // If no color is already set do not print a warning message
+        if no_color {
+            true
         } else {
-            let my_slashes = self.dir.name.matches("/").count();
-            let other_slashes = other.dir.name.matches("/").count();
-
-            if my_slashes > other_slashes {
-                Ordering::Greater
-            } else if my_slashes < other_slashes {
-                Ordering::Less
-            } else {
-                if self.dir.name < other.dir.name {
-                    Ordering::Less
-                } else if self.dir.name > other.dir.name {
-                    Ordering::Greater
-                } else {
-                    Ordering::Equal
+            // Required for windows 10
+            // Fails to resolve for windows 8 so disable color
+            match ansi_term::enable_ansi_support() {
+                Ok(_) => no_color,
+                Err(_) => {
+                    eprintln!(
+                    "This version of Windows does not support ANSI colors, setting no_color flag"
+                );
+                    true
                 }
             }
         }
     }
-}
-impl PartialOrd for Node {
-    fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
-        Some(self.cmp(other))
+    #[cfg(not(windows))]
+    {
+        no_color
     }
 }
-impl PartialEq for Node {
-    fn eq(&self, other: &Self) -> bool {
-        (&self.dir.name, self.dir.size) == (&other.dir.name, other.dir.size)
-    }
-}
-impl Eq for Node {}
 
-#[derive(Clone, Debug)]
-struct Dir {
-    name: String,
-    size: u64,
+fn get_height_of_terminal() -> usize {
+    // Simplify once https://github.com/eminence/terminal-size/pull/41 is
+    // merged
+    terminal_size()
+        // Windows CI runners detect a terminal height of 0
+        .map(|(_, Height(h))| max(h as usize, DEFAULT_NUMBER_OF_LINES))
+        .unwrap_or(DEFAULT_NUMBER_OF_LINES)
+        - 10
 }
 
-static DEFAULT_NUMBER_OF_LINES: &'static str = &"15";
+fn get_width_of_terminal() -> usize {
+    // Simplify once https://github.com/eminence/terminal-size/pull/41 is
+    // merged
+    terminal_size()
+        .map(|(Width(w), _)| match cfg!(windows) {
+            // Windows CI runners detect a very low terminal width
+            true => max(w as usize, DEFAULT_TERMINAL_WIDTH),
+            false => w as usize,
+        })
+        .unwrap_or(DEFAULT_TERMINAL_WIDTH)
+}
+
+fn get_regex_value(maybe_value: Option<Values>) -> Vec<Regex> {
+    maybe_value
+        .unwrap_or_default()
+        .map(|reg| {
+            Regex::new(reg).unwrap_or_else(|err| {
+                eprintln!("Ignoring bad value for regex {:?}", err);
+                process::exit(1)
+            })
+        })
+        .collect()
+}
 
 fn main() {
-    let options = App::new("Trailing args example")
-        .setting(AppSettings::TrailingVarArg)
-        .arg(
-            Arg::with_name("number_of_lines")
-                .short("n")
-                .help("Number of lines of output to show")
-                .takes_value(true)
-                .default_value(DEFAULT_NUMBER_OF_LINES),
-        )
-        .arg(Arg::with_name("inputs").multiple(true))
-        .get_matches();
+    let options = build_cli().get_matches();
+    let config = get_config();
 
-    let filenames = {
-        match options.values_of("inputs") {
-            None => vec!["."],
-            Some(r) => r.collect(),
-        }
+    let target_dirs = options
+        .values_of("inputs")
+        .expect("Should be a default value here")
+        .collect();
+
+    let summarize_file_types = options.is_present("types");
+
+    let filter_regexs = get_regex_value(options.values_of("filter"));
+    let invert_filter_regexs = get_regex_value(options.values_of("invert_filter"));
+
+    let terminal_width = options
+        .value_of_t("width")
+        .unwrap_or_else(|_| get_width_of_terminal());
+
+    let depth = options.value_of_t("depth").unwrap_or(usize::MAX);
+
+    // If depth is set, then 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 = value_t!(options.value_of("number_of_lines"), usize).unwrap();
 
-    let (permissions, results) = get_dir_tree(filenames);
-    let slice_it = find_big_ones(&results, number_of_lines);
-    display(permissions, slice_it);
-}
+    let number_of_lines = options
+        .value_of("number_of_lines")
+        .and_then(|v| {
+            v.parse()
+                .map_err(|_| eprintln!("Ignoring bad value for number_of_lines"))
+                .ok()
+        })
+        .unwrap_or(default_height);
 
-fn get_dir_tree(filenames: Vec<&str>) -> (bool, Vec<Node>) {
-    let mut permissions = true;
-    let mut results = vec![];
-    for b in filenames {
-        let mut new_name = String::from(b);
-        while new_name.chars().last() == Some('/') && new_name.len() != 1 {
-            new_name.pop();
-        }
-        let (hp, data) = examine_dir_str(new_name);
-        permissions = permissions && hp;
-        results.push(data);
+    let no_colors = init_color(config.get_no_colors(&options));
+
+    let ignore_directories = options
+        .values_of("ignore_directory")
+        .unwrap_or_default()
+        .map(PathBuf::from);
+
+    let by_filecount = options.is_present("by_filecount");
+    let limit_filesystem = options.is_present("limit_filesystem");
+
+    let simplified_dirs = simplify_dir_names(target_dirs);
+    let allowed_filesystems = limit_filesystem
+        .then(|| get_filesystem_devices(simplified_dirs.iter()))
+        .unwrap_or_default();
+
+    let ignored_full_path: HashSet<PathBuf> = ignore_directories
+        .flat_map(|x| simplified_dirs.iter().map(move |d| d.join(&x)))
+        .collect();
+
+    let walk_data = WalkData {
+        ignore_directories: ignored_full_path,
+        filter_regex: &filter_regexs,
+        invert_filter_regex: &invert_filter_regexs,
+        allowed_filesystems,
+        use_apparent_size: config.get_apparent_size(&options),
+        by_filecount,
+        ignore_hidden: config.get_ignore_hidden(&options),
+    };
+    let pool = panic::catch_unwind(init_rayon);
+    if pool.is_err() {
+        eprintln!("Warning: Could not configure threads {:?}", pool.err());
     }
-    (permissions, results)
-}
 
-fn examine_dir_str(loc: String) -> (bool, Node) {
-    let mut inodes: HashSet<u64> = HashSet::new();
-    let (hp, result) = examine_dir(fs::read_dir(&loc), &mut inodes);
+    let iso = config.get_iso(&options);
+    let (top_level_nodes, has_errors) = walk_it(simplified_dirs, walk_data);
 
-    // This needs to be folded into the below recursive call somehow
-    let new_size = result.iter().fold(0, |a, b| a + b.dir.size);
-    (
-        hp,
-        Node {
-            dir: Dir {
-                name: loc,
-                size: new_size,
-            },
-            children: result,
-        },
-    )
-}
+    let tree = match summarize_file_types {
+        true => get_all_file_types(&top_level_nodes, number_of_lines),
+        false => get_biggest(
+            top_level_nodes,
+            config.get_min_size(&options, iso),
+            number_of_lines,
+            depth,
+            options.values_of("filter").is_some() || options.value_of("invert_filter").is_some(),
+        ),
+    };
 
-#[cfg(target_os = "linux")]
-fn get_metadata_blocks_and_inode(d: &std::fs::DirEntry) -> Option<(u64, u64)> {
-    use std::os::linux::fs::MetadataExt;
-    match d.metadata().ok() {
-        Some(md) => Some((md.len(), md.st_ino())),
-        None => None,
-    }
-}
-
-#[cfg(target_os = "unix")]
-fn get_metadata_blocks_and_inode(d: &std::fs::DirEntry) -> Option<(u64, u64)> {
-    use std::os::unix::fs::MetadataExt;
-    match d.metadata().ok() {
-        Some(md) => Some((md.len(), md.ino())),
-        None => None,
-    }
-}
-
-#[cfg(target_os = "macos")]
-fn get_metadata_blocks_and_inode(d: &std::fs::DirEntry) -> Option<(u64, u64)> {
-    use std::os::macos::fs::MetadataExt;
-    match d.metadata().ok() {
-        Some(md) => Some((md.len(), md.st_ino())),
-        None => None,
-    }
-}
-
-#[cfg(not(any(target_os = "linux", target_os = "unix", target_os = "macos")))]
-fn get_metadata_blocks_and_inode(_d: &std::fs::DirEntry) -> Option<(u64, u64)> {
-    match d.metadata().ok() {
-        Some(md) => Some((md.len(), 0)), //move to option not 0
-        None => None,
-    }
-}
-
-fn examine_dir(a_dir: io::Result<ReadDir>, inodes: &mut HashSet<u64>) -> (bool, Vec<Node>) {
-    let mut result = vec![];
-    let mut have_permission = true;
-
-    if a_dir.is_ok() {
-        let paths = a_dir.unwrap();
-        for dd in paths {
-            match dd {
-                Ok(d) => {
-                    let file_type = d.file_type().ok();
-                    let maybe_size_and_inode = get_metadata_blocks_and_inode(&d);
-
-                    match (file_type, maybe_size_and_inode) {
-                        (Some(file_type), Some((size, inode))) => {
-                            let s = d.path().to_string_lossy().to_string();
-                            if inodes.contains(&inode) {
-                                continue;
-                            }
-                            inodes.insert(inode);
-
-                            if d.path().is_dir() && !file_type.is_symlink() {
-                                let (hp, recursive) = examine_dir(fs::read_dir(d.path()), inodes);
-                                have_permission = have_permission && hp;
-                                let new_size = recursive.iter().fold(size, |a, b| a + b.dir.size);
-                                result.push(Node {
-                                    dir: Dir {
-                                        name: s,
-                                        size: new_size,
-                                    },
-                                    children: recursive,
-                                })
-                            } else {
-                                result.push(Node {
-                                    dir: Dir {
-                                        name: s,
-                                        size: size,
-                                    },
-                                    children: vec![],
-                                })
-                            }
-                        }
-                        (_, None) => have_permission = false,
-                        (_, _) => (),
-                    }
-                }
-                Err(_) => (),
-            }
-        }
-    } else {
-        have_permission = false;
-    }
-    (have_permission, result)
-}
-
-// We start with a list of root directories - these must be the biggest folders
-// We then repeadedly merge in the children of the biggest directory - Each iteration
-// the next biggest directory's children are merged in.
-fn find_big_ones<'a>(l: &'a Vec<Node>, max_to_show: usize) -> Vec<&Node> {
-    let mut new_l: Vec<&Node> = l.iter().map(|a| a).collect();
-    new_l.sort();
-
-    for processed_pointer in 0..max_to_show {
-        if new_l.len() == processed_pointer {
-            break;
-        }
-        // Must be a list of pointers into new_l otherwise b_list will go out of scope
-        // when it is deallocated
-        let mut b_list: Vec<&Node> = new_l[processed_pointer]
-            .children
-            .iter()
-            .map(|a| a)
-            .collect();
-        new_l.extend(b_list);
-        new_l.sort();
-        /*println!(
-            "{:?} -------------------",
-            new_l
-                .iter()
-                .map(|a| a.dir.size.to_string() + ": " + &a.dir.name)
-                .collect::<Vec<String>>()
-        );*/
-    }
-    if new_l.len() > max_to_show {
-        new_l[0..max_to_show + 1].to_vec()
-    } else {
-        new_l
-    }
-}
-
-fn display(permissions: bool, to_display: Vec<&Node>) -> () {
-    if !permissions {
+    if has_errors {
         eprintln!("Did not have permissions for all directories");
     }
-
-    display_node(to_display[0], &to_display, true, 1, "")
-}
-
-fn display_node<S: Into<String>>(
-    node_to_print: &Node,
-    to_display: &Vec<&Node>,
-    is_first: bool,
-    depth: u8,
-    indentation_str: S,
-) {
-    let mut is = indentation_str.into();
-    print_this_node(node_to_print, is_first, depth, is.as_ref());
-
-    is = is.replace("└──", "   ");
-    is = is.replace("├──", "│  ");
-
-    let printable_node_slashes = node_to_print.dir.name.matches("/").count();
-
-    let mut num_sibblings = to_display.iter().fold(0, |a, b| {
-        if node_to_print.children.contains(b)
-            && b.dir.name.matches("/").count() == printable_node_slashes + 1
-        {
-            a + 1
-        } else {
-            a
-        }
-    });
-
-    let mut is_biggest = true;
-    for node in to_display {
-        if node_to_print.children.contains(node) {
-            if node.dir.name.matches("/").count() == printable_node_slashes + 1 {
-                num_sibblings -= 1;
-                let tree_chars = {
-                    if num_sibblings == 0 {
-                        "└──"
-                    } else {
-                        "├──"
-                    }
-                };
-                display_node(
-                    &node,
-                    to_display,
-                    is_biggest,
-                    depth + 1,
-                    is.to_string() + tree_chars,
-                );
-                is_biggest = false;
-            }
-        }
+    if let Some(root_node) = tree {
+        draw_it(
+            config.get_full_paths(&options),
+            !config.get_reverse(&options),
+            no_colors,
+            config.get_no_bars(&options),
+            terminal_width,
+            by_filecount,
+            &root_node,
+            iso,
+            config.get_skip_total(&options),
+        )
     }
 }
 
-fn print_this_node(node_to_print: &Node, is_biggest: bool, depth: u8, indentation_str: &str) {
-    let padded_size = format!("{:>5}", human_readable_number(node_to_print.dir.size),);
-    println!(
-        "{} {} {}",
-        if is_biggest {
-            Fixed(196).paint(padded_size)
-        } else {
-            Fixed(7).paint(padded_size)
-        },
-        indentation_str,
-        Fixed(7)
-            .on(Fixed(cmp::min(8, (depth) as u8) + 231))
-            .paint(node_to_print.dir.name.to_string())
-    );
-}
-
-fn human_readable_number(size: u64) -> (String) {
-    let units = vec!["T", "G", "M", "K"]; //make static
-
-    //return format!("{}B", size);
-
-    for (i, u) in units.iter().enumerate() {
-        let marker = 1024u64.pow((units.len() - i) as u32);
-        if size >= marker {
-            if size / marker < 10 {
-                return format!("{:.1}{}", (size as f32 / marker as f32), u);
-            } else {
-                return format!("{}{}", (size / marker), u);
-            }
-        }
-    }
-    return format!("{}B", size);
-}
-
-mod tests {
-    use super::*;
-
-    #[test]
-    fn test_human_readable_number() {
-        assert_eq!(human_readable_number(1), "1B");
-        assert_eq!(human_readable_number(956), "956B");
-        assert_eq!(human_readable_number(1004), "1004B");
-        assert_eq!(human_readable_number(1024), "1.0K");
-        assert_eq!(human_readable_number(1536), "1.5K");
-        assert_eq!(human_readable_number(1024 * 512), "512K");
-        assert_eq!(human_readable_number(1024 * 1024), "1.0M");
-        assert_eq!(human_readable_number(1024 * 1024 * 1024 - 1), "1023M");
-        assert_eq!(human_readable_number(1024 * 1024 * 1024 * 20), "20G");
-        assert_eq!(human_readable_number(1024 * 1024 * 1024 * 1024), "1.0T");
-    }
+fn init_rayon() -> Result<(), ThreadPoolBuildError> {
+    // Larger stack size to handle cases with lots of nested directories
+    rayon::ThreadPoolBuilder::new()
+        .stack_size(usize::pow(1024, 3))
+        .build_global()
 }

src/node.rs

@@ -0,0 +1,78 @@
+use crate::platform::get_metadata;
+use crate::utils::is_filtered_out_due_to_invert_regex;
+use crate::utils::is_filtered_out_due_to_regex;
+
+use regex::Regex;
+use std::cmp::Ordering;
+use std::path::PathBuf;
+
+#[derive(Debug, Eq, Clone)]
+pub struct Node {
+    pub name: PathBuf,
+    pub size: u64,
+    pub children: Vec<Node>,
+    pub inode_device: Option<(u64, u64)>,
+    pub depth: usize,
+}
+
+#[allow(clippy::too_many_arguments)]
+pub fn build_node(
+    dir: PathBuf,
+    children: Vec<Node>,
+    filter_regex: &[Regex],
+    invert_filter_regex: &[Regex],
+    use_apparent_size: bool,
+    is_symlink: bool,
+    is_file: bool,
+    by_filecount: bool,
+    depth: usize,
+) -> Option<Node> {
+    get_metadata(&dir, use_apparent_size).map(|data| {
+        let inode_device = if is_symlink && !use_apparent_size {
+            None
+        } else {
+            data.1
+        };
+
+        let size = if is_filtered_out_due_to_regex(filter_regex, &dir)
+            || is_filtered_out_due_to_invert_regex(invert_filter_regex, &dir)
+            || (is_symlink && !use_apparent_size)
+            || by_filecount && !is_file
+        {
+            0
+        } else if by_filecount {
+            1
+        } else {
+            data.0
+        };
+
+        Node {
+            name: dir,
+            size,
+            children,
+            inode_device,
+            depth,
+        }
+    })
+}
+
+impl PartialEq for Node {
+    fn eq(&self, other: &Self) -> bool {
+        self.name == other.name && self.size == other.size && self.children == other.children
+    }
+}
+
+impl Ord for Node {
+    fn cmp(&self, other: &Self) -> Ordering {
+        self.size
+            .cmp(&other.size)
+            .then_with(|| self.name.cmp(&other.name))
+            .then_with(|| self.children.cmp(&other.children))
+    }
+}
+
+impl PartialOrd for Node {
+    fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
+        Some(self.cmp(other))
+    }
+}

src/platform.rs

@@ -0,0 +1,128 @@
+#[allow(unused_imports)]
+use std::fs;
+
+use std::path::Path;
+
+#[cfg(target_family = "unix")]
+fn get_block_size() -> u64 {
+    // All os specific implementations of MetadataExt seem to define a block as 512 bytes
+    // https://doc.rust-lang.org/std/os/linux/fs/trait.MetadataExt.html#tymethod.st_blocks
+    512
+}
+
+#[cfg(target_family = "unix")]
+pub fn get_metadata(d: &Path, use_apparent_size: bool) -> Option<(u64, Option<(u64, u64)>)> {
+    use std::os::unix::fs::MetadataExt;
+    match d.metadata() {
+        Ok(md) => {
+            if use_apparent_size {
+                Some((md.len(), Some((md.ino(), md.dev()))))
+            } else {
+                Some((md.blocks() * get_block_size(), Some((md.ino(), md.dev()))))
+            }
+        }
+        Err(_e) => None,
+    }
+}
+
+#[cfg(target_family = "windows")]
+pub fn get_metadata(d: &Path, _use_apparent_size: bool) -> Option<(u64, Option<(u64, u64)>)> {
+    // On windows opening the file to get size, file ID and volume can be very
+    // expensive because 1) it causes a few system calls, and more importantly 2) it can cause
+    // windows defender to scan the file.
+    // Therefore we try to avoid doing that for common cases, mainly those of
+    // plain files:
+
+    // The idea is to make do with the file size that we get from the OS for
+    // free as part of iterating a folder. Therefore we want to make sure that
+    // it makes sense to use that free size information:
+
+    // Volume boundaries:
+    // The user can ask us not to cross volume boundaries. If the DirEntry is a
+    // plain file and not a reparse point or other non-trivial stuff, we assume
+    // that the file is located on the same volume as the directory that
+    // contains it.
+
+    // File ID:
+    // This optimization does deprive us of access to a file ID. As a
+    // workaround, we just make one up that hopefully does not collide with real
+    // file IDs.
+    // Hard links: Unresolved. We don't get inode/file index, so hard links
+    // count once for each link. Hopefully they are not too commonly in use on
+    // windows.
+
+    // Size:
+    // We assume (naively?) that for the common cases the free size info is the
+    // same as one would get by doing the expensive thing. Sparse, encrypted and
+    // compressed files are not included in the common cases, as one can image
+    // there being more than view on their size.
+
+    // Savings in orders of magnitude in terms of time, io and cpu have been
+    // observed on hdd, windows 10, some 100Ks files taking up some hundreds of
+    // GBs:
+    // Consistently opening the file: 30 minutes.
+    // With this optimization:         8 sec.
+
+    use std::io;
+    use winapi_util::Handle;
+    fn handle_from_path_limited<P: AsRef<Path>>(path: P) -> io::Result<Handle> {
+        use std::fs::OpenOptions;
+        use std::os::windows::fs::OpenOptionsExt;
+        const FILE_READ_ATTRIBUTES: u32 = 0x0080;
+
+        // So, it seems that it does does have to be that expensive to open
+        // files to get their info: Avoiding opening the file with the full
+        // GENERIC_READ is key:
+
+        // https://docs.microsoft.com/en-us/windows/win32/secauthz/generic-access-rights:
+        // "For example, a Windows file object maps the GENERIC_READ bit to the
+        // READ_CONTROL and SYNCHRONIZE standard access rights and to the
+        // FILE_READ_DATA, FILE_READ_EA, and FILE_READ_ATTRIBUTES
+        // object-specific access rights"
+
+        // The flag FILE_READ_DATA seems to be the expensive one, so we'll avoid
+        // that, and a most of the other ones. Simply because it seems that we
+        // don't need them.
+
+        let file = OpenOptions::new()
+            .access_mode(FILE_READ_ATTRIBUTES)
+            .open(path)?;
+        Ok(Handle::from_file(file))
+    }
+
+    fn get_metadata_expensive(d: &Path) -> Option<(u64, Option<(u64, u64)>)> {
+        use winapi_util::file::information;
+
+        let h = handle_from_path_limited(d).ok()?;
+        let info = information(&h).ok()?;
+
+        Some((
+            info.file_size(),
+            Some((info.file_index(), info.volume_serial_number())),
+        ))
+    }
+
+    use std::os::windows::fs::MetadataExt;
+    match d.metadata() {
+        Ok(ref md) => {
+            const FILE_ATTRIBUTE_ARCHIVE: u32 = 0x20;
+            const FILE_ATTRIBUTE_READONLY: u32 = 0x01;
+            const FILE_ATTRIBUTE_HIDDEN: u32 = 0x02;
+            const FILE_ATTRIBUTE_SYSTEM: u32 = 0x04;
+            const FILE_ATTRIBUTE_NORMAL: u32 = 0x80;
+            const FILE_ATTRIBUTE_DIRECTORY: u32 = 0x10;
+
+            let attr_filtered = md.file_attributes()
+                & !(FILE_ATTRIBUTE_HIDDEN | FILE_ATTRIBUTE_READONLY | FILE_ATTRIBUTE_SYSTEM);
+            if (attr_filtered & FILE_ATTRIBUTE_ARCHIVE) != 0
+                || (attr_filtered & FILE_ATTRIBUTE_DIRECTORY) != 0
+                || md.file_attributes() == FILE_ATTRIBUTE_NORMAL
+            {
+                Some((md.len(), None))
+            } else {
+                get_metadata_expensive(d)
+            }
+        }
+        _ => get_metadata_expensive(d),
+    }
+}

src/utils.rs

@@ -0,0 +1,158 @@
+use platform::get_metadata;
+use std::collections::HashSet;
+use std::path::{Path, PathBuf};
+
+use crate::platform;
+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());
+
+    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) {
+                to_remove.push(tt.to_path_buf());
+            } else if is_a_parent_of(tt, &top_level_name) {
+                can_add = false;
+            }
+        }
+        for r in to_remove {
+            top_level_names.remove(&r);
+        }
+        if can_add {
+            top_level_names.insert(top_level_name);
+        }
+    }
+
+    top_level_names
+}
+
+pub fn get_filesystem_devices<'a, P: IntoIterator<Item = &'a PathBuf>>(paths: P) -> HashSet<u64> {
+    // Gets the device ids for the filesystems which are used by the argument paths
+    paths
+        .into_iter()
+        .filter_map(|p| match get_metadata(p, false) {
+            Some((_size, Some((_id, dev)))) => Some(dev),
+            _ => None,
+        })
+        .collect()
+}
+
+pub fn normalize_path<P: AsRef<Path>>(path: P) -> PathBuf {
+    // normalize path ...
+    // 1. removing repeated separators
+    // 2. removing interior '.' ("current directory") path segments
+    // 3. removing trailing extra separators and '.' ("current directory") path segments
+    // * `Path.components()` does all the above work; ref: <https://doc.rust-lang.org/std/path/struct.Path.html#method.components>
+    // 4. changing to os preferred separator (automatically done by recollecting components back into a PathBuf)
+    path.as_ref().components().collect()
+}
+
+pub fn is_filtered_out_due_to_regex(filter_regex: &[Regex], dir: &Path) -> bool {
+    if filter_regex.is_empty() {
+        false
+    } else {
+        filter_regex
+            .iter()
+            .all(|f| !f.is_match(&dir.as_os_str().to_string_lossy()))
+    }
+}
+
+pub fn is_filtered_out_due_to_invert_regex(filter_regex: &[Regex], dir: &Path) -> bool {
+    filter_regex
+        .iter()
+        .any(|f| f.is_match(&dir.as_os_str().to_string_lossy()))
+}
+
+fn is_a_parent_of<P: AsRef<Path>>(parent: P, child: P) -> bool {
+    let parent = parent.as_ref();
+    let child = child.as_ref();
+    child.starts_with(parent) && !parent.starts_with(child)
+}
+
+mod tests {
+    #[allow(unused_imports)]
+    use super::*;
+
+    #[test]
+    fn test_simplify_dir() {
+        let mut correct = HashSet::new();
+        correct.insert(PathBuf::from("a"));
+        assert_eq!(simplify_dir_names(vec!["a"]), correct);
+    }
+
+    #[test]
+    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);
+    }
+
+    #[test]
+    fn test_simplify_dir_duplicates() {
+        let mut correct = HashSet::new();
+        correct.insert(["a", "b"].iter().collect::<PathBuf>());
+        correct.insert(PathBuf::from("c"));
+        assert_eq!(
+            simplify_dir_names(vec![
+                "a/b",
+                "a/b//",
+                "a/././b///",
+                "c",
+                "c/",
+                "c/.",
+                "c/././",
+                "c/././."
+            ]),
+            correct
+        );
+    }
+    #[test]
+    fn test_simplify_dir_rm_subdir_and_not_substrings() {
+        let mut correct = HashSet::new();
+        correct.insert(PathBuf::from("b"));
+        correct.insert(["c", "a", "b"].iter().collect::<PathBuf>());
+        correct.insert(["a", "b"].iter().collect::<PathBuf>());
+        assert_eq!(simplify_dir_names(vec!["a/b", "c/a/b/", "b"]), correct);
+    }
+
+    #[test]
+    fn test_simplify_dir_dots() {
+        let mut correct = HashSet::new();
+        correct.insert(PathBuf::from("src"));
+        assert_eq!(simplify_dir_names(vec!["src/."]), correct);
+    }
+
+    #[test]
+    fn test_simplify_dir_substring_names() {
+        let mut correct = HashSet::new();
+        correct.insert(PathBuf::from("src"));
+        correct.insert(PathBuf::from("src_v2"));
+        assert_eq!(simplify_dir_names(vec!["src/", "src_v2"]), correct);
+    }
+
+    #[test]
+    fn test_is_a_parent_of() {
+        assert!(is_a_parent_of("/usr", "/usr/andy"));
+        assert!(is_a_parent_of("/usr", "/usr/andy/i/am/descendant"));
+        assert!(!is_a_parent_of("/usr", "/usr/."));
+        assert!(!is_a_parent_of("/usr", "/usr/"));
+        assert!(!is_a_parent_of("/usr", "/usr"));
+        assert!(!is_a_parent_of("/usr/", "/usr"));
+        assert!(!is_a_parent_of("/usr/andy", "/usr"));
+        assert!(!is_a_parent_of("/usr/andy", "/usr/sibling"));
+        assert!(!is_a_parent_of("/usr/folder", "/usr/folder_not_a_child"));
+    }
+
+    #[test]
+    fn test_is_a_parent_of_root() {
+        assert!(is_a_parent_of("/", "/usr/andy"));
+        assert!(is_a_parent_of("/", "/usr"));
+        assert!(!is_a_parent_of("/", "/"));
+    }
+}

tests/test_dir/many/hello_file

@@ -0,0 +1 @@
+hello

tests/test_dir2/dir/hello

@@ -0,0 +1 @@
+hello

tests/test_dir2/dir_name_clash

@@ -0,0 +1 @@
+hello

tests/test_dir2/dir_substring/hello

@@ -0,0 +1 @@
+hello

tests/test_dir_hidden_entries/.hidden_file

@@ -0,0 +1 @@
+hi

tests/test_exact_output.rs

@@ -0,0 +1,208 @@
+use assert_cmd::Command;
+use std::ffi::OsStr;
+use std::str;
+use std::sync::Once;
+
+static INIT: Once = Once::new();
+
+/**
+ * This file contains tests that verify the exact output of the command.
+ * This output differs on Linux / Mac so the tests are harder to write and debug
+ * Windows is ignored here because the results vary by host making exact testing impractical
+ *
+ * Despite the above problems, these tests are good as they are the closest to 'the real thing'.
+ */
+
+//  Warning: File sizes differ on both platform and on the format of the disk.
+/// Copy to /tmp dir - we assume that the formatting of the /tmp partition
+/// is consistent. If the tests fail your /tmp filesystem probably differs
+fn copy_test_data(dir: &str) {
+    // First remove the existing directory - just in case it is there and has incorrect data
+    let last_slash = dir.rfind('/').unwrap();
+    let last_part_of_dir = dir.chars().skip(last_slash).collect::<String>();
+    let _ = Command::new("rm")
+        .arg("-rf")
+        .arg("/tmp/".to_owned() + &*last_part_of_dir)
+        .ok();
+
+    let _ = Command::new("cp")
+        .arg("-r")
+        .arg(dir)
+        .arg("/tmp/")
+        .ok()
+        .map_err(|err| eprintln!("Error copying directory for test setup\n{:?}", err));
+}
+
+fn initialize() {
+    INIT.call_once(|| {
+        copy_test_data("tests/test_dir");
+        copy_test_data("tests/test_dir2");
+        copy_test_data("tests/test_dir_unicode");
+    });
+}
+
+fn exact_output_test<T: AsRef<OsStr>>(valid_outputs: Vec<String>, command_args: Vec<T>) {
+    initialize();
+
+    let mut a = &mut Command::cargo_bin("dust").unwrap();
+
+    for p in command_args {
+        a = a.arg(p);
+    }
+
+    let output = str::from_utf8(&a.unwrap().stdout).unwrap().to_owned();
+
+    assert!(valid_outputs.iter().any(|i| output.contains(i)));
+}
+
+// "windows" result data can vary by host (size seems to be variable by one byte); fix code vs test and re-enable
+#[cfg_attr(target_os = "windows", ignore)]
+#[test]
+pub fn test_main_basic() {
+    // -c is no color mode - This makes testing much simpler
+    exact_output_test(main_output(), vec!["-c", "/tmp/test_dir/"])
+}
+
+#[cfg_attr(target_os = "windows", ignore)]
+#[test]
+pub fn test_main_multi_arg() {
+    let command_args = vec![
+        "-c",
+        "/tmp/test_dir/many/",
+        "/tmp/test_dir",
+        "/tmp/test_dir",
+    ];
+    exact_output_test(main_output(), command_args);
+}
+
+fn main_output() -> Vec<String> {
+    // Some linux currently thought to be Manjaro, Arch
+    // Although probably depends on how drive is formatted
+    let mac_and_some_linux = r#"
+  0B     ┌── a_file    │░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░█ │   0%
+4.0K     ├── hello_file│█████████████████████████████████████████████████ │ 100%
+4.0K   ┌─┴ many        │█████████████████████████████████████████████████ │ 100%
+4.0K ┌─┴ test_dir      │█████████████████████████████████████████████████ │ 100%
+"#
+    .trim()
+    .to_string();
+
+    let ubuntu = r#"
+  0B     ┌── a_file    │                ░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░█ │   0%
+4.0K     ├── hello_file│                ░░░░░░░░░░░░░░░░█████████████████ │  33%
+8.0K   ┌─┴ many        │                █████████████████████████████████ │  67%
+ 12K ┌─┴ test_dir      │█████████████████████████████████████████████████ │ 100%
+  "#
+    .trim()
+    .to_string();
+
+    vec![mac_and_some_linux, ubuntu]
+}
+
+#[cfg_attr(target_os = "windows", ignore)]
+#[test]
+pub fn test_main_long_paths() {
+    let command_args = vec!["-c", "-p", "/tmp/test_dir/"];
+    exact_output_test(main_output_long_paths(), command_args);
+}
+
+fn main_output_long_paths() -> Vec<String> {
+    let mac_and_some_linux = r#"
+  0B     ┌── /tmp/test_dir/many/a_file    │░░░░░░░░░░░░░░░░░░░░░░░░░░░░░█ │   0%
+4.0K     ├── /tmp/test_dir/many/hello_file│██████████████████████████████ │ 100%
+4.0K   ┌─┴ /tmp/test_dir/many             │██████████████████████████████ │ 100%
+4.0K ┌─┴ /tmp/test_dir                    │██████████████████████████████ │ 100%
+"#
+    .trim()
+    .to_string();
+    let ubuntu = r#"
+  0B     ┌── /tmp/test_dir/many/a_file    │         ░░░░░░░░░░░░░░░░░░░░█ │   0%
+4.0K     ├── /tmp/test_dir/many/hello_file│         ░░░░░░░░░░███████████ │  33%
+8.0K   ┌─┴ /tmp/test_dir/many             │         █████████████████████ │  67%
+ 12K ┌─┴ /tmp/test_dir                    │██████████████████████████████ │ 100%
+"#
+    .trim()
+    .to_string();
+    vec![mac_and_some_linux, ubuntu]
+}
+
+// Check against directories and files whose names are substrings of each other
+#[cfg_attr(target_os = "windows", ignore)]
+#[test]
+pub fn test_substring_of_names_and_long_names() {
+    let command_args = vec!["-c", "/tmp/test_dir2"];
+    exact_output_test(no_substring_of_names_output(), command_args);
+}
+
+fn no_substring_of_names_output() -> Vec<String> {
+    let ubuntu = "
+  0B   ┌── long_dir_name_what_a_very_long_dir_name_what_happens_when_this_goes..
+4.0K   ├── dir_name_clash
+4.0K   │ ┌── hello
+8.0K   ├─┴ dir
+4.0K   │ ┌── hello
+8.0K   ├─┴ dir_substring
+ 24K ┌─┴ test_dir2
+    "
+    .trim()
+    .into();
+
+    let mac_and_some_linux = "
+  0B   ┌── long_dir_name_what_a_very_long_dir_name_what_happens_when_this_goes..
+4.0K   │ ┌── hello
+4.0K   ├─┴ dir
+4.0K   ├── dir_name_clash
+4.0K   │ ┌── hello
+4.0K   ├─┴ dir_substring
+ 12K ┌─┴ test_dir2
+  "
+    .trim()
+    .into();
+    vec![mac_and_some_linux, ubuntu]
+}
+
+#[cfg_attr(target_os = "windows", ignore)]
+#[test]
+pub fn test_unicode_directories() {
+    let command_args = vec!["-c", "/tmp/test_dir_unicode"];
+    exact_output_test(unicode_dir(), command_args);
+}
+
+fn unicode_dir() -> Vec<String> {
+    // The way unicode & asian characters are rendered on the terminal should make this line up
+    let ubuntu = "
+  0B   ┌── ラウトは難しいです！.japan│                                  █ │   0%
+  0B   ├── 👩.unicode                │                                  █ │   0%
+4.0K ┌─┴ test_dir_unicode            │███████████████████████████████████ │ 100%
+    "
+    .trim()
+    .into();
+
+    let mac_and_some_linux = "
+0B   ┌── ラウトは難しいです！.japan│                                    █ │   0%
+0B   ├── 👩.unicode                │                                    █ │   0%
+0B ┌─┴ test_dir_unicode            │                                    █ │   0%
+    "
+    .trim()
+    .into();
+    vec![mac_and_some_linux, ubuntu]
+}
+
+#[cfg_attr(target_os = "windows", ignore)]
+#[test]
+pub fn test_apparent_size() {
+    let command_args = vec!["-c", "-s", "-b", "/tmp/test_dir"];
+    exact_output_test(apparent_size_output(), command_args);
+}
+
+fn apparent_size_output() -> Vec<String> {
+    // The apparent directory sizes are too unpredictable and system dependent to try and match
+    let files = r#"
+  0B     ┌── a_file
+  6B     ├── hello_file
+ "#
+    .trim()
+    .to_string();
+
+    vec![files]
+}

tests/test_flags.rs

@@ -0,0 +1,195 @@
+use assert_cmd::Command;
+use std::ffi::OsStr;
+use std::str;
+
+/**
+ * This file contains tests that test a substring of the output using '.contains'
+ *
+ * These tests should be the same cross platform
+ */
+
+fn build_command<T: AsRef<OsStr>>(command_args: Vec<T>) -> String {
+    let mut cmd = &mut Command::cargo_bin("dust").unwrap();
+    for p in command_args {
+        cmd = cmd.arg(p);
+    }
+    let finished = &cmd.unwrap();
+    let stderr = str::from_utf8(&finished.stderr).unwrap();
+    assert_eq!(stderr, "");
+
+    str::from_utf8(&finished.stdout).unwrap().into()
+}
+
+// We can at least test the file names are there
+#[test]
+pub fn test_basic_output() {
+    let output = build_command(vec!["tests/test_dir/"]);
+
+    assert!(output.contains(" ┌─┴ "));
+    assert!(output.contains("test_dir "));
+    assert!(output.contains("  ┌─┴ "));
+    assert!(output.contains("many "));
+    assert!(output.contains("    ├── "));
+    assert!(output.contains("hello_file"));
+    assert!(output.contains("     ┌── "));
+    assert!(output.contains("a_file "));
+}
+
+#[test]
+pub fn test_output_no_bars_means_no_excess_spaces() {
+    let output = build_command(vec!["-b", "tests/test_dir/"]);
+    // If bars are not being shown we don't need to pad the output with spaces
+    assert!(output.contains("many"));
+    assert!(!output.contains("many    "));
+}
+
+#[test]
+pub fn test_reverse_flag() {
+    let output = build_command(vec!["-r", "-c", "tests/test_dir/"]);
+    assert!(output.contains(" └─┬ test_dir "));
+    assert!(output.contains("  └─┬ many "));
+    assert!(output.contains("    ├── hello_file"));
+    assert!(output.contains("    └── a_file "));
+}
+
+#[test]
+pub fn test_d_flag_works() {
+    // We should see the top level directory but not the sub dirs / files:
+    let output = build_command(vec!["-d", "1", "tests/test_dir/"]);
+    assert!(!output.contains("hello_file"));
+}
+
+#[test]
+pub fn test_d_flag_works_and_still_recurses_down() {
+    // We had a bug where running with '-d 1' would stop at the first directory and the code
+    // would fail to recurse down
+    let output = build_command(vec!["-d", "1", "-f", "-c", "tests/test_dir2/"]);
+    assert!(output.contains("4 ┌─┴ test_dir2"));
+}
+
+// Check against directories and files whose names are substrings of each other
+#[test]
+pub fn test_ignore_dir() {
+    let output = build_command(vec!["-c", "-X", "dir_substring", "tests/test_dir2/"]);
+    assert!(!output.contains("dir_substring"));
+}
+
+#[test]
+pub fn test_with_bad_param() {
+    let mut cmd = Command::cargo_bin("dust").unwrap();
+    let result = cmd.arg("bad_place").unwrap();
+    let stderr = str::from_utf8(&result.stderr).unwrap();
+    assert!(stderr.contains("Did not have permissions for all directories"));
+}
+
+#[test]
+pub fn test_hidden_flag() {
+    // Check we can see the hidden file normally
+    let output = build_command(vec!["-c", "tests/test_dir_hidden_entries/"]);
+    assert!(output.contains(".hidden_file"));
+    assert!(output.contains("┌─┴ test_dir_hidden_entries"));
+
+    // Check that adding the '-h' flag causes us to not see hidden files
+    let output = build_command(vec!["-c", "-i", "tests/test_dir_hidden_entries/"]);
+    assert!(!output.contains(".hidden_file"));
+    assert!(output.contains("┌── test_dir_hidden_entries"));
+}
+
+#[test]
+pub fn test_number_of_files() {
+    // Check we can see the hidden file normally
+    let output = build_command(vec!["-c", "-f", "tests/test_dir"]);
+    assert!(output.contains("1     ┌── a_file "));
+    assert!(output.contains("1     ├── hello_file"));
+    assert!(output.contains("2   ┌─┴ many"));
+    assert!(output.contains("2 ┌─┴ test_dir"));
+}
+
+#[test]
+pub fn test_show_files_by_type() {
+    // Check we can list files by type
+    let output = build_command(vec!["-c", "-t", "tests"]);
+    assert!(output.contains(" .unicode"));
+    assert!(output.contains(" .japan"));
+    assert!(output.contains(" .rs"));
+    assert!(output.contains(" (no extension)"));
+    assert!(output.contains("┌─┴ (total)"));
+}
+
+#[test]
+pub fn test_output_skip_total() {
+    let output = build_command(vec![
+        "--skip-total",
+        "tests/test_dir/many/hello_file",
+        "tests/test_dir/many/a_file",
+    ]);
+    assert!(output.contains("hello_file"));
+    assert!(!output.contains("(total)"));
+}
+
+#[test]
+pub fn test_show_files_by_regex_match_lots() {
+    // Check we can see '.rs' files in the tests directory
+    let output = build_command(vec!["-c", "-e", "\\.rs$", "tests"]);
+    assert!(output.contains(" ┌─┴ tests"));
+    assert!(!output.contains("0B ┌── tests"));
+    assert!(!output.contains("0B ┌─┴ tests"));
+}
+
+#[test]
+pub fn test_show_files_by_regex_match_nothing() {
+    // Check there are no files named: '.match_nothing' in the tests directory
+    let output = build_command(vec!["-c", "-e", "match_nothing$", "tests"]);
+    assert!(output.contains("0B ┌── tests"));
+}
+
+#[test]
+pub fn test_show_files_by_regex_match_multiple() {
+    let output = build_command(vec![
+        "-c",
+        "-e",
+        "test_dir_hidden",
+        "-e",
+        "test_dir2",
+        "-n",
+        "100",
+        "tests",
+    ]);
+    assert!(output.contains("test_dir2"));
+    assert!(output.contains("test_dir_hidden"));
+    assert!(!output.contains("many")); // We do not find the 'many' folder in the 'test_dir' folder
+}
+
+#[test]
+pub fn test_show_files_by_invert_regex() {
+    let output = build_command(vec!["-c", "-f", "-v", "e", "tests/test_dir2"]);
+    // There are 0 files without 'e' in the name
+    assert!(output.contains("0 ┌── test_dir2"));
+
+    let output = build_command(vec!["-c", "-f", "-v", "a", "tests/test_dir2"]);
+    // There are 2 files without 'a' in the name
+    assert!(output.contains("2 ┌─┴ test_dir2"));
+
+    // There are 4 files in the test_dir2 hierarchy
+    let output = build_command(vec!["-c", "-f", "-v", "match_nothing$", "tests/test_dir2"]);
+    assert!(output.contains("4 ┌─┴ test_dir2"));
+}
+
+#[test]
+pub fn test_show_files_by_invert_regex_match_multiple() {
+    // We ignore test_dir2 & test_dir_unicode, leaving the test_dir folder
+    // which has the 'many' folder inside
+    let output = build_command(vec![
+        "-c",
+        "-v",
+        "test_dir2",
+        "-v",
+        "test_dir_unicode",
+        "-n",
+        "100",
+        "tests",
+    ]);
+    assert!(!output.contains("test_dir2"));
+    assert!(!output.contains("test_dir_unicode"));
+    assert!(output.contains("many"));
+}

tests/tests.rs

@@ -0,0 +1 @@
+

tests/tests_symlinks.rs

@@ -0,0 +1,118 @@
+use assert_cmd::Command;
+use std::fs::File;
+use std::io::Write;
+use std::path::PathBuf;
+use std::str;
+
+use tempfile::Builder;
+use tempfile::TempDir;
+
+// File sizes differ on both platform and on the format of the disk.
+// Windows: `ln` is not usually an available command; creation of symbolic links requires special enhanced permissions
+
+fn build_temp_file(dir: &TempDir) -> PathBuf {
+    let file_path = dir.path().join("notes.txt");
+    let mut file = File::create(&file_path).unwrap();
+    writeln!(file, "I am a temp file").unwrap();
+    file_path
+}
+
+#[cfg_attr(target_os = "windows", ignore)]
+#[test]
+pub fn test_soft_sym_link() {
+    let dir = Builder::new().tempdir().unwrap();
+    let file = build_temp_file(&dir);
+    let dir_s = dir.path().to_str().unwrap();
+    let file_path_s = file.to_str().unwrap();
+
+    let link_name = dir.path().join("the_link");
+    let link_name_s = link_name.to_str().unwrap();
+    let c = Command::new("ln")
+        .arg("-s")
+        .arg(file_path_s)
+        .arg(link_name_s)
+        .output();
+    assert!(c.is_ok());
+
+    let c = format!(" ├── {}", link_name_s);
+    let b = format!(" ┌── {}", file_path_s);
+    let a = format!("─┴ {}", dir_s);
+
+    let mut cmd = Command::cargo_bin("dust").unwrap();
+    // Mac test runners create long filenames in tmp directories
+    let output = cmd
+        .args(["-p", "-c", "-s", "-w 999", dir_s])
+        .unwrap()
+        .stdout;
+
+    let output = str::from_utf8(&output).unwrap();
+
+    assert!(output.contains(a.as_str()));
+    assert!(output.contains(b.as_str()));
+    assert!(output.contains(c.as_str()));
+}
+
+#[cfg_attr(target_os = "windows", ignore)]
+#[test]
+pub fn test_hard_sym_link() {
+    let dir = Builder::new().tempdir().unwrap();
+    let file = build_temp_file(&dir);
+    let dir_s = dir.path().to_str().unwrap();
+    let file_path_s = file.to_str().unwrap();
+
+    let link_name = dir.path().join("the_link");
+    let link_name_s = link_name.to_str().unwrap();
+    let c = Command::new("ln")
+        .arg(file_path_s)
+        .arg(link_name_s)
+        .output();
+    assert!(c.is_ok());
+
+    let file_output = format!(" ┌── {}", file_path_s);
+    let dirs_output = format!("─┴ {}", dir_s);
+
+    let mut cmd = Command::cargo_bin("dust").unwrap();
+    // Mac test runners create long filenames in tmp directories
+    let output = cmd.args(["-p", "-c", "-w 999", dir_s]).unwrap().stdout;
+
+    // The link should not appear in the output because multiple inodes are now ordered
+    // then filtered.
+    let output = str::from_utf8(&output).unwrap();
+    assert!(output.contains(dirs_output.as_str()));
+    assert!(output.contains(file_output.as_str()));
+}
+
+#[cfg_attr(target_os = "windows", ignore)]
+#[test]
+pub fn test_recursive_sym_link() {
+    let dir = Builder::new().tempdir().unwrap();
+    let dir_s = dir.path().to_str().unwrap();
+
+    let link_name = dir.path().join("the_link");
+    let link_name_s = link_name.to_str().unwrap();
+
+    let c = Command::new("ln")
+        .arg("-s")
+        .arg(dir_s)
+        .arg(link_name_s)
+        .output();
+    assert!(c.is_ok());
+
+    let a = format!("─┬ {}", dir_s);
+    let b = format!(" └── {}", link_name_s);
+
+    let mut cmd = Command::cargo_bin("dust").unwrap();
+    let output = cmd
+        .arg("-p")
+        .arg("-c")
+        .arg("-r")
+        .arg("-s")
+        .arg("-w 999")
+        .arg(dir_s)
+        .unwrap()
+        .stdout;
+    let output = str::from_utf8(&output).unwrap();
+
+    assert!(output.contains(a.as_str()));
+    assert!(output.contains(b.as_str()));
+}