Bump serde from 1.0.197 to 1.0.199 (#2953)

Bumps [serde](https://github.com/serde-rs/serde) from 1.0.197 to 1.0.199.
- [Release notes](https://github.com/serde-rs/serde/releases)
- [Commits](https://github.com/serde-rs/serde/compare/v1.0.197...v1.0.199)

---
updated-dependencies:
- dependency-name: serde
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

.github/ISSUE_TEMPLATE/syntax_request.md

@@ -26,4 +26,4 @@ guidelines for adding new syntaxes:
 [Name or description of the syntax/language here]
 
 **Guideline Criteria:**
-[packagecontro.io link here]
+[packagecontrol.io link here]

.github/workflows/Auto-merge-dependabot-PRs.yml

@@ -0,0 +1,23 @@
+# This workflow triggers auto-merge of any PR that dependabot creates so that
+# PRs will be merged automatically without maintainer intervention if CI passes
+name: Auto-merge dependabot PRs
+
+on:
+  pull_request_target:
+    types: [opened]
+
+jobs:
+  auto-merge:
+    if: github.repository == 'sharkdp/bat' && startsWith(github.head_ref, 'dependabot/')
+    runs-on: ubuntu-latest
+    environment:
+      name: auto-merge
+      url: https://github.com/sharkdp/bat/blob/main/.github/workflows/Auto-merge-dependabot-PRs.yml
+    env:
+      GITHUB_TOKEN: ${{ secrets.AUTO_MERGE_GITHUB_TOKEN }}
+    steps:
+      - uses: actions/checkout@v4
+      - run: |
+          gh pr review ${{ github.event.pull_request.number }} --comment --body "If CI passes, this dependabot PR will be [auto-merged](https://github.com/sharkdp/bat/blob/main/.github/workflows/Auto-merge-dependabot-PRs.yml) 🚀"
+      - run: |
+          gh pr merge --auto --squash ${{ github.event.pull_request.number }}

.github/workflows/CICD.yml

@@ -2,6 +2,7 @@ name: CICD
 
 env:
   CICD_INTERMEDIATES_DIR: "_cicd-intermediates"
+  MSRV_FEATURES: --no-default-features --features minimal-application,bugreport,build-assets
 
 on:
   workflow_dispatch:
@@ -13,6 +14,43 @@ on:
       - '*'
 
 jobs:
+  all-jobs:
+    if: always() # Otherwise this job is skipped if the matrix job fails
+    name: all-jobs
+    runs-on: ubuntu-latest
+    needs:
+      - crate_metadata
+      - ensure_cargo_fmt
+      - min_version
+      - license_checks
+      - test_with_new_syntaxes_and_themes
+      - test_with_system_config
+      - documentation
+      - cargo-audit
+      - build
+    steps:
+      - run: jq --exit-status 'all(.result == "success")' <<< '${{ toJson(needs) }}'
+
+  crate_metadata:
+    name: Extract crate metadata
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+    - name: Extract crate information
+      id: crate_metadata
+      run: |
+        cargo metadata --no-deps --format-version 1 | jq -r '"name=" + .packages[0].name' | tee -a $GITHUB_OUTPUT
+        cargo metadata --no-deps --format-version 1 | jq -r '"version=" + .packages[0].version' | tee -a $GITHUB_OUTPUT
+        cargo metadata --no-deps --format-version 1 | jq -r '"maintainer=" + .packages[0].authors[0]' | tee -a $GITHUB_OUTPUT
+        cargo metadata --no-deps --format-version 1 | jq -r '"homepage=" + .packages[0].homepage' | tee -a $GITHUB_OUTPUT
+        cargo metadata --no-deps --format-version 1 | jq -r '"msrv=" + .packages[0].rust_version' | tee -a $GITHUB_OUTPUT
+    outputs:
+      name: ${{ steps.crate_metadata.outputs.name }}
+      version: ${{ steps.crate_metadata.outputs.version }}
+      maintainer: ${{ steps.crate_metadata.outputs.maintainer }}
+      homepage: ${{ steps.crate_metadata.outputs.homepage }}
+      msrv: ${{ steps.crate_metadata.outputs.msrv }}
+
   ensure_cargo_fmt:
     name: Ensure 'cargo fmt' has been run
     runs-on: ubuntu-20.04
@@ -20,46 +58,42 @@ jobs:
     - uses: dtolnay/rust-toolchain@stable
       with:
         components: rustfmt
-    - uses: actions/checkout@v3
+    - uses: actions/checkout@v4
     - run: cargo fmt -- --check
 
-  license_checks:
-    name: License checks
-    runs-on: ubuntu-20.04
-    steps:
-    - uses: actions/checkout@v3
-      with:
-        submodules: true # we especially want to perform license checks on submodules
-    - run: tests/scripts/license-checks.sh
-
   min_version:
     name: Minimum supported rust version
     runs-on: ubuntu-20.04
-    env:
-      MSRV_FEATURES: --no-default-features --features minimal-application,bugreport,build-assets
+    needs: crate_metadata
     steps:
     - name: Checkout source code
-      uses: actions/checkout@v3
+      uses: actions/checkout@v4
 
-    - name: Get the MSRV from the package metadata
-      id: msrv
-      run: cargo metadata --no-deps --format-version 1 | jq -r '"version=" + (.packages[] | select(.name == "bat")).rust_version' >> $GITHUB_OUTPUT
-    - name: Install rust toolchain (v${{ steps.msrv.outputs.version }})
+    - name: Install rust toolchain (v${{ needs.crate_metadata.outputs.msrv }})
       uses: dtolnay/rust-toolchain@master
       with:
-        toolchain: ${{ steps.msrv.outputs.version }}
+        toolchain: ${{ needs.crate_metadata.outputs.msrv }}
         components: clippy
     - name: Run clippy (on minimum supported rust version to prevent warnings we can't fix)
       run: cargo clippy --locked --all-targets ${{ env.MSRV_FEATURES }}
     - name: Run tests
       run: cargo test --locked ${{ env.MSRV_FEATURES }}
 
+  license_checks:
+    name: License checks
+    runs-on: ubuntu-20.04
+    steps:
+    - uses: actions/checkout@v4
+      with:
+        submodules: true # we especially want to perform license checks on submodules
+    - run: tests/scripts/license-checks.sh
+
   test_with_new_syntaxes_and_themes:
     name: Run tests with updated syntaxes and themes
     runs-on: ubuntu-20.04
     steps:
     - name: Git checkout
-      uses: actions/checkout@v3
+      uses: actions/checkout@v4
       with:
         submodules: true # we need all syntax and theme submodules
     - name: Install Rust toolchain
@@ -88,7 +122,7 @@ jobs:
     runs-on: ubuntu-20.04
     steps:
     - name: Git checkout
-      uses: actions/checkout@v3
+      uses: actions/checkout@v4
     - name: Prepare environment variables
       run: |
         echo "BAT_SYSTEM_CONFIG_PREFIX=$GITHUB_WORKSPACE/tests/examples/system_config" >> $GITHUB_ENV
@@ -104,7 +138,7 @@ jobs:
     runs-on: ubuntu-20.04
     steps:
     - name: Git checkout
-      uses: actions/checkout@v3
+      uses: actions/checkout@v4
     - name: Install Rust toolchain
       uses: dtolnay/rust-toolchain@stable
     - name: Check documentation
@@ -114,29 +148,37 @@ jobs:
     - name: Show man page
       run: man $(find . -name bat.1)
 
+  cargo-audit:
+    name: cargo audit
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - run: cargo audit
+
   build:
     name: ${{ matrix.job.target }} (${{ matrix.job.os }})
     runs-on: ${{ matrix.job.os }}
+    needs: crate_metadata
     strategy:
       fail-fast: false
       matrix:
         job:
-          - { target: aarch64-unknown-linux-gnu   , os: ubuntu-20.04, use-cross: true }
-          - { target: arm-unknown-linux-gnueabihf , os: ubuntu-20.04, use-cross: true }
-          - { target: arm-unknown-linux-musleabihf, os: ubuntu-20.04, use-cross: true }
-          - { target: i686-pc-windows-msvc        , os: windows-2019                  }
-          - { target: i686-unknown-linux-gnu      , os: ubuntu-20.04, use-cross: true }
-          - { target: i686-unknown-linux-musl     , os: ubuntu-20.04, use-cross: true }
-          - { target: x86_64-apple-darwin         , os: macos-10.15                   }
-          - { target: x86_64-pc-windows-gnu       , os: windows-2019                  }
-          - { target: x86_64-pc-windows-msvc      , os: windows-2019                  }
-          - { target: x86_64-unknown-linux-gnu    , os: ubuntu-20.04, use-cross: true }
-          - { target: x86_64-unknown-linux-musl   , os: ubuntu-20.04, use-cross: true }
+          - { target: aarch64-unknown-linux-gnu   , os: ubuntu-20.04, dpkg_arch: arm64,            use-cross: true }
+          - { target: arm-unknown-linux-gnueabihf , os: ubuntu-20.04, dpkg_arch: armhf,            use-cross: true }
+          - { target: arm-unknown-linux-musleabihf, os: ubuntu-20.04, dpkg_arch: musl-linux-armhf, use-cross: true }
+          - { target: i686-pc-windows-msvc        , os: windows-2019,                                              }
+          - { target: i686-unknown-linux-gnu      , os: ubuntu-20.04, dpkg_arch: i686,             use-cross: true }
+          - { target: i686-unknown-linux-musl     , os: ubuntu-20.04, dpkg_arch: musl-linux-i686,  use-cross: true }
+          - { target: x86_64-apple-darwin         , os: macos-12,                                                  }
+          - { target: x86_64-pc-windows-gnu       , os: windows-2019,                                              }
+          - { target: x86_64-pc-windows-msvc      , os: windows-2019,                                              }
+          - { target: x86_64-unknown-linux-gnu    , os: ubuntu-20.04, dpkg_arch: amd64,            use-cross: true }
+          - { target: x86_64-unknown-linux-musl   , os: ubuntu-20.04, dpkg_arch: musl-linux-amd64, use-cross: true }
     env:
       BUILD_CMD: cargo
     steps:
     - name: Checkout source code
-      uses: actions/checkout@v3
+      uses: actions/checkout@v4
 
     - name: Install prerequisites
       shell: bash
@@ -146,14 +188,6 @@ jobs:
           aarch64-unknown-linux-gnu) sudo apt-get -y update ; sudo apt-get -y install gcc-aarch64-linux-gnu ;;
         esac
 
-    - name: Extract crate information
-      shell: bash
-      run: |
-        echo "PROJECT_NAME=$(sed -n 's/^name = "\(.*\)"/\1/p' Cargo.toml | head -n1)" >> $GITHUB_ENV
-        echo "PROJECT_VERSION=$(sed -n 's/^version = "\(.*\)"/\1/p' Cargo.toml | head -n1)" >> $GITHUB_ENV
-        echo "PROJECT_MAINTAINER=$(sed -n 's/^authors = \["\(.*\)"\]/\1/p' Cargo.toml)" >> $GITHUB_ENV
-        echo "PROJECT_HOMEPAGE=$(sed -n 's/^homepage = "\(.*\)"/\1/p' Cargo.toml)" >> $GITHUB_ENV
-
     - name: Install Rust toolchain
       uses: dtolnay/rust-toolchain@stable
       with:
@@ -184,7 +218,7 @@ jobs:
       shell: bash
       run: $BUILD_CMD build --locked --release --target=${{ matrix.job.target }}
 
-    - name: Set bin name & path
+    - name: Set binary name & path
       id: bin
       shell: bash
       run: |
@@ -195,10 +229,10 @@ jobs:
         esac;
 
         # Setup paths
-        BIN_NAME="${{ env.PROJECT_NAME }}${EXE_suffix}"
+        BIN_NAME="${{ needs.crate_metadata.outputs.name }}${EXE_suffix}"
         BIN_PATH="target/${{ matrix.job.target }}/release/${BIN_NAME}"
 
-        # Let subsequent steps know where to find the bin
+        # Let subsequent steps know where to find the binary
         echo "BIN_PATH=${BIN_PATH}" >> $GITHUB_OUTPUT
         echo "BIN_NAME=${BIN_NAME}" >> $GITHUB_OUTPUT
 
@@ -208,12 +242,18 @@ jobs:
       run: |
         # test only library unit tests and binary for arm-type targets
         unset CARGO_TEST_OPTIONS
-        unset CARGO_TEST_OPTIONS ; case ${{ matrix.job.target }} in arm-* | aarch64-*) CARGO_TEST_OPTIONS="--lib --bin ${PROJECT_NAME}" ;; esac;
+        unset CARGO_TEST_OPTIONS ; case ${{ matrix.job.target }} in arm-* | aarch64-*) CARGO_TEST_OPTIONS="--lib --bin ${{ needs.crate_metadata.outputs.name }}" ;; esac;
         echo "CARGO_TEST_OPTIONS=${CARGO_TEST_OPTIONS}" >> $GITHUB_OUTPUT
 
     - name: Run tests
       shell: bash
-      run: $BUILD_CMD test --locked --target=${{ matrix.job.target }} ${{ steps.test-options.outputs.CARGO_TEST_OPTIONS}}
+      run: |
+        if [[ ${{ matrix.job.os }} = windows-* ]]
+        then
+          powershell.exe -command "$BUILD_CMD test --locked --target=${{ matrix.job.target }} ${{ steps.test-options.outputs.CARGO_TEST_OPTIONS}}"
+        else
+          $BUILD_CMD test --locked --target=${{ matrix.job.target }} ${{ steps.test-options.outputs.CARGO_TEST_OPTIONS}}
+        fi
 
     - name: Run bat
       shell: bash
@@ -248,7 +288,7 @@ jobs:
       shell: bash
       run: |
         PKG_suffix=".tar.gz" ; case ${{ matrix.job.target }} in *-pc-windows-*) PKG_suffix=".zip" ;; esac;
-        PKG_BASENAME=${PROJECT_NAME}-v${PROJECT_VERSION}-${{ matrix.job.target }}
+        PKG_BASENAME=${{ needs.crate_metadata.outputs.name }}-v${{ needs.crate_metadata.outputs.version }}-${{ matrix.job.target }}
         PKG_NAME=${PKG_BASENAME}${PKG_suffix}
         echo "PKG_NAME=${PKG_NAME}" >> $GITHUB_OUTPUT
 
@@ -260,17 +300,17 @@ jobs:
         # Binary
         cp "${{ steps.bin.outputs.BIN_PATH }}" "$ARCHIVE_DIR"
 
-        # Man page
-        cp 'target/${{ matrix.job.target }}/release/build/${{ env.PROJECT_NAME }}'-*/out/assets/manual/bat.1 "$ARCHIVE_DIR"
-
         # README, LICENSE and CHANGELOG files
         cp "README.md" "LICENSE-MIT" "LICENSE-APACHE" "CHANGELOG.md" "$ARCHIVE_DIR"
 
+        # Man page
+        cp 'target/${{ matrix.job.target }}/release/build/${{ needs.crate_metadata.outputs.name }}'-*/out/assets/manual/bat.1 "$ARCHIVE_DIR"
+
         # Autocompletion files
-        cp 'target/${{ matrix.job.target }}/release/build/${{ env.PROJECT_NAME }}'-*/out/assets/completions/bat.bash "$ARCHIVE_DIR/autocomplete/${{ env.PROJECT_NAME }}.bash"
-        cp 'target/${{ matrix.job.target }}/release/build/${{ env.PROJECT_NAME }}'-*/out/assets/completions/bat.fish "$ARCHIVE_DIR/autocomplete/${{ env.PROJECT_NAME }}.fish"
-        cp 'target/${{ matrix.job.target }}/release/build/${{ env.PROJECT_NAME }}'-*/out/assets/completions/_bat.ps1 "$ARCHIVE_DIR/autocomplete/_${{ env.PROJECT_NAME }}.ps1"
-        cp 'target/${{ matrix.job.target }}/release/build/${{ env.PROJECT_NAME }}'-*/out/assets/completions/bat.zsh "$ARCHIVE_DIR/autocomplete/${{ env.PROJECT_NAME }}.zsh"
+        cp 'target/${{ matrix.job.target }}/release/build/${{ needs.crate_metadata.outputs.name }}'-*/out/assets/completions/bat.bash "$ARCHIVE_DIR/autocomplete/${{ needs.crate_metadata.outputs.name }}.bash"
+        cp 'target/${{ matrix.job.target }}/release/build/${{ needs.crate_metadata.outputs.name }}'-*/out/assets/completions/bat.fish "$ARCHIVE_DIR/autocomplete/${{ needs.crate_metadata.outputs.name }}.fish"
+        cp 'target/${{ matrix.job.target }}/release/build/${{ needs.crate_metadata.outputs.name }}'-*/out/assets/completions/_bat.ps1 "$ARCHIVE_DIR/autocomplete/_${{ needs.crate_metadata.outputs.name }}.ps1"
+        cp 'target/${{ matrix.job.target }}/release/build/${{ needs.crate_metadata.outputs.name }}'-*/out/assets/completions/bat.zsh "$ARCHIVE_DIR/autocomplete/${{ needs.crate_metadata.outputs.name }}.zsh"
 
         # base compressed package
         pushd "${PKG_STAGING}/" >/dev/null
@@ -293,20 +333,11 @@ jobs:
         DPKG_DIR="${DPKG_STAGING}/dpkg"
         mkdir -p "${DPKG_DIR}"
 
-        DPKG_BASENAME=${PROJECT_NAME}
-        DPKG_CONFLICTS=${PROJECT_NAME}-musl
-        case ${{ matrix.job.target }} in *-musl) DPKG_BASENAME=${PROJECT_NAME}-musl ; DPKG_CONFLICTS=${PROJECT_NAME} ;; esac;
-        DPKG_VERSION=${PROJECT_VERSION}
-
-        unset DPKG_ARCH
-        case ${{ matrix.job.target }} in
-          aarch64-*-linux-*) DPKG_ARCH=arm64 ;;
-          arm-*-linux-*hf) DPKG_ARCH=armhf ;;
-          i686-*-linux-*) DPKG_ARCH=i686 ;;
-          x86_64-*-linux-*) DPKG_ARCH=amd64 ;;
-          *) DPKG_ARCH=notset ;;
-        esac;
-
+        DPKG_BASENAME=${{ needs.crate_metadata.outputs.name }}
+        DPKG_CONFLICTS=${{ needs.crate_metadata.outputs.name }}-musl
+        case ${{ matrix.job.target }} in *-musl) DPKG_BASENAME=${{ needs.crate_metadata.outputs.name }}-musl ; DPKG_CONFLICTS=${{ needs.crate_metadata.outputs.name }} ;; esac;
+        DPKG_VERSION=${{ needs.crate_metadata.outputs.version }}
+        DPKG_ARCH="${{ matrix.job.dpkg_arch }}"
         DPKG_NAME="${DPKG_BASENAME}_${DPKG_VERSION}_${DPKG_ARCH}.deb"
         echo "DPKG_NAME=${DPKG_NAME}" >> $GITHUB_OUTPUT
 
@@ -314,13 +345,13 @@ jobs:
         install -Dm755 "${{ steps.bin.outputs.BIN_PATH }}" "${DPKG_DIR}/usr/bin/${{ steps.bin.outputs.BIN_NAME }}"
 
         # Man page
-        install -Dm644 'target/${{ matrix.job.target }}/release/build/${{ env.PROJECT_NAME }}'-*/out/assets/manual/bat.1 "${DPKG_DIR}/usr/share/man/man1/${{ env.PROJECT_NAME }}.1"
-        gzip -n --best "${DPKG_DIR}/usr/share/man/man1/${{ env.PROJECT_NAME }}.1"
+        install -Dm644 'target/${{ matrix.job.target }}/release/build/${{ needs.crate_metadata.outputs.name }}'-*/out/assets/manual/bat.1 "${DPKG_DIR}/usr/share/man/man1/${{ needs.crate_metadata.outputs.name }}.1"
+        gzip -n --best "${DPKG_DIR}/usr/share/man/man1/${{ needs.crate_metadata.outputs.name }}.1"
 
         # Autocompletion files
-        install -Dm644 'target/${{ matrix.job.target }}/release/build/${{ env.PROJECT_NAME }}'-*/out/assets/completions/bat.bash "${DPKG_DIR}/usr/share/bash-completion/completions/${{ env.PROJECT_NAME }}"
-        install -Dm644 'target/${{ matrix.job.target }}/release/build/${{ env.PROJECT_NAME }}'-*/out/assets/completions/bat.fish "${DPKG_DIR}/usr/share/fish/vendor_completions.d/${{ env.PROJECT_NAME }}.fish"
-        install -Dm644 'target/${{ matrix.job.target }}/release/build/${{ env.PROJECT_NAME }}'-*/out/assets/completions/bat.zsh "${DPKG_DIR}/usr/share/zsh/vendor-completions/_${{ env.PROJECT_NAME }}"
+        install -Dm644 'target/${{ matrix.job.target }}/release/build/${{ needs.crate_metadata.outputs.name }}'-*/out/assets/completions/bat.bash "${DPKG_DIR}/usr/share/bash-completion/completions/${{ needs.crate_metadata.outputs.name }}"
+        install -Dm644 'target/${{ matrix.job.target }}/release/build/${{ needs.crate_metadata.outputs.name }}'-*/out/assets/completions/bat.fish "${DPKG_DIR}/usr/share/fish/vendor_completions.d/${{ needs.crate_metadata.outputs.name }}.fish"
+        install -Dm644 'target/${{ matrix.job.target }}/release/build/${{ needs.crate_metadata.outputs.name }}'-*/out/assets/completions/bat.zsh "${DPKG_DIR}/usr/share/zsh/vendor-completions/_${{ needs.crate_metadata.outputs.name }}"
 
         # README and LICENSE
         install -Dm644 "README.md" "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/README.md"
@@ -331,12 +362,12 @@ jobs:
 
         cat > "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/copyright" <<EOF
         Format: http://www.debian.org/doc/packaging-manuals/copyright-format/1.0/
-        Upstream-Name: ${{ env.PROJECT_NAME }}
-        Source: ${{ env.PROJECT_HOMEPAGE }}
+        Upstream-Name: ${{ needs.crate_metadata.outputs.name }}
+        Source: ${{ needs.crate_metadata.outputs.homepage }}
 
         Files: *
-        Copyright: ${{ env.PROJECT_MAINTAINER }}
-        Copyright: $COPYRIGHT_YEARS ${{ env.PROJECT_MAINTAINER }}
+        Copyright: ${{ needs.crate_metadata.outputs.maintainer }}
+        Copyright: $COPYRIGHT_YEARS ${{ needs.crate_metadata.outputs.maintainer }}
         License: Apache-2.0 or MIT
 
         License: Apache-2.0
@@ -377,10 +408,10 @@ jobs:
         Version: ${DPKG_VERSION}
         Section: utils
         Priority: optional
-        Maintainer: ${{ env.PROJECT_MAINTAINER }}
-        Homepage: ${{ env.PROJECT_HOMEPAGE }}
+        Maintainer: ${{ needs.crate_metadata.outputs.maintainer }}
+        Homepage: ${{ needs.crate_metadata.outputs.homepage }}
         Architecture: ${DPKG_ARCH}
-        Provides: ${{ env.PROJECT_NAME }}
+        Provides: ${{ needs.crate_metadata.outputs.name }}
         Conflicts: ${DPKG_CONFLICTS}
         Description: cat(1) clone with wings.
           A cat(1) clone with syntax highlighting and Git integration.
@@ -413,7 +444,7 @@ jobs:
         echo "IS_RELEASE=${IS_RELEASE}" >> $GITHUB_OUTPUT
 
     - name: Publish archives and packages
-      uses: softprops/action-gh-release@v1
+      uses: softprops/action-gh-release@v2
       if: steps.is-release.outputs.IS_RELEASE
       with:
         files: |
@@ -421,3 +452,15 @@ jobs:
           ${{ steps.debian-package.outputs.DPKG_PATH }}
       env:
         GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+
+  winget:
+    name: Publish to Winget
+    runs-on: ubuntu-latest
+    needs: build
+    if: startsWith(github.ref, 'refs/tags/v')
+    steps:
+      - uses: vedantmgoyal2009/winget-releaser@v2
+        with:
+          identifier: sharkdp.bat
+          installers-regex: '-pc-windows-msvc\.zip$'
+          token: ${{ secrets.WINGET_TOKEN }}

.github/workflows/require-changelog-for-PRs.yml

@@ -0,0 +1,33 @@
+name: Changelog
+
+on:
+  pull_request:
+
+jobs:
+  check-changelog:
+    name: Check for changelog entry
+    runs-on: ubuntu-latest
+    # dependabot PRs are automerged if CI passes; we shouldn't block these
+    if: github.actor != 'dependabot[bot]'
+    env:
+      PR_NUMBER: ${{ github.event.number }}
+      PR_BASE: ${{ github.base_ref }}
+    steps:
+      - uses: actions/checkout@v4
+      - name: Fetch PR base
+        run: git fetch --no-tags --prune --depth=1 origin
+
+      # cannot use `github.actor`: the triggering commit may be authored by a maintainer
+      - name: Get PR submitter
+        id: get-submitter
+        run: curl -sSfL https://api.github.com/repos/sharkdp/bat/pulls/${PR_NUMBER} | jq -r '"submitter=" + .user.login' | tee -a $GITHUB_OUTPUT
+
+      - name: Search for added line in changelog
+        env:
+          PR_SUBMITTER: ${{ steps.get-submitter.outputs.submitter }}
+        run: |
+          ADDED=$(git diff -U0 "origin/${PR_BASE}" HEAD -- CHANGELOG.md | grep -P '^\+[^\+].+$')
+          echo "Added lines in CHANGELOG.md:"
+          echo "$ADDED"
+          echo "Grepping for PR info (see CONTRIBUTING.md):"
+          grep "#${PR_NUMBER}\\b.*@${PR_SUBMITTER}\\b" <<< "$ADDED"

.gitmodules

@@ -254,3 +254,9 @@
 [submodule "assets/syntaxes/02_Extra/Crontab"]
 	path = assets/syntaxes/02_Extra/Crontab
 	url = https://github.com/michaelblyons/SublimeSyntax-Crontab
+[submodule "assets/syntaxes/02_Extra/NSIS"]
+	path = assets/syntaxes/02_Extra/NSIS
+	url = https://github.com/SublimeText/NSIS
+[submodule "assets/syntaxes/02_Extra/vscode-wgsl"]
+	path = assets/syntaxes/02_Extra/vscode-wgsl
+	url = https://github.com/PolyMeilex/vscode-wgsl.git

CHANGELOG.md

@@ -1,3 +1,107 @@
+# unreleased
+
+## Features
+
+- Set terminal title to file names when Paging is not Paging::Never #2807 (@Oliver-Looney)
+- `bat --squeeze-blank`/`bat -s` will now squeeze consecutive empty lines, see #1441 (@eth-p) and #2665 (@einfachIrgendwer0815)
+- `bat --squeeze-limit` to set the maximum number of empty consecutive when using `--squeeze-blank`, see #1441 (@eth-p) and #2665 (@einfachIrgendwer0815)
+- `PrettyPrinter::squeeze_empty_lines` to support line squeezing for bat as a library, see #1441 (@eth-p) and #2665 (@einfachIrgendwer0815)
+- Syntax highlighting for JavaScript files that start with `#!/usr/bin/env bun` #2913 (@sharunkumar)
+
+## Bugfixes
+
+- Fix long file name wrapping in header, see #2835 (@FilipRazek)
+- Fix `NO_COLOR` support, see #2767 (@acuteenvy)
+- Fix handling of inputs with OSC ANSI escape sequences, see #2541 and #2544 (@eth-p)
+- Fix handling of inputs with combined ANSI color and attribute sequences, see #2185 and #2856 (@eth-p)
+- Fix panel width when line 10000 wraps, see #2854 (@eth-p)
+
+## Other
+
+- Upgrade to Rust 2021 edition #2748 (@cyqsimon)
+- Refactor and cleanup build script #2756 (@cyqsimon)
+- Checks changelog has been written to for PRs in CI #2766 (@cyqsimon)
+  - Use GitHub API to get correct PR submitter #2791 (@cyqsimon)
+- Minor benchmark script improvements #2768 (@cyqsimon)
+- Update Arch Linux package URL in README files #2779 (@brunobell)
+- Update and improve `zsh` completion, see #2772 (@okapia)
+- More extensible syntax mapping mechanism #2755 (@cyqsimon)
+- Use proper Architecture for Debian packages built for musl, see #2811 (@Enselic)
+- Pull in fix for unsafe-libyaml security advisory, see #2812 (@dtolnay)
+- Update git-version dependency to use Syn v2, see #2816 (@dtolnay)
+- Update git2 dependency to v0.18.2, see #2852 (@eth-p)
+- Improve performance when color output disabled, see #2397 and #2857 (@eth-p)
+- Relax syntax mapping rule restrictions to allow brace expansion #2865 (@cyqsimon)
+- Apply clippy fixes #2864 (@cyqsimon)
+- Faster startup by offloading glob matcher building to a worker thread #2868 (@cyqsimon)
+- Display which theme is the default one in basic output (no colors), see #2937 (@sblondon)
+- Display which theme is the default one in colored output, see #2838 (@sblondon)
+
+## Syntaxes
+
+- `cmd-help`: scope subcommands followed by other terms, and other misc improvements, see #2819 (@victor-gp)
+- Upgrade JQ syntax, see #2820 (@dependabot[bot])
+- Add syntax mapping for quadman quadlets #2866 (@cyqsimon)
+- Map containers .conf files to TOML syntax #2867 (@cyqsimon)
+- Associate `xsh` files with `xonsh` syntax that is Python, see #2840 (@anki-code).
+- Added auto detect syntax for `.jsonc` #2795 (@mxaddict)
+- Added auto detect syntax for `.aws/{config,credentials}` #2795 (@mxaddict)
+- Add syntax mapping for Wireguard config #2874 (@cyqsimon)
+
+## Themes
+
+## `bat` as a library
+
+- Changes to `syntax_mapping::SyntaxMapping` #2755 (@cyqsimon)
+  - `SyntaxMapping::get_syntax_for` is now correctly public
+  - [BREAKING] `SyntaxMapping::{empty,builtin}` are removed; use `SyntaxMapping::new` instead
+  - [BREAKING] `SyntaxMapping::mappings` is replaced by `SyntaxMapping::{builtin,custom,all}_mappings`
+- Make `Controller::run_with_error_handler`'s error handler `FnMut`, see #2831 (@rhysd)
+- Improve compile time by 20%, see #2815 (@dtolnay)
+
+# v0.24.0
+
+## Features
+
+- Add environment variable `BAT_PAGING`, see #2629 (@einfachIrgendwer0815)
+- Add opt-in (`--features lessopen`) support for `LESSOPEN` and `LESSCLOSE`. See #1597, #1739, #2444, #2602, and #2662 (@Anomalocaridid)
+
+## Bugfixes
+
+- Fix `more` not being found on Windows when provided via `BAT_PAGER`, see #2570, #2580, and #2651 (@mataha)
+- Switched default behavior of `--map-syntax` to be case insensitive #2520
+- Updated version of `serde_yaml` to `0.9`. See #2627 (@Raghav-Bell)
+- Fix arithmetic overflow in `LineRange::from` and `LineRange::parse_range`, see #2674, #2698 (@skoriop)
+- Fix paging not happening when stdout is interactive but stdin is not, see #2574 (@Nigecat)
+- Make `-pp` override `--paging` and vice versa when passed as a later argument, see #2660 (@J-Kappes)
+
+## Other
+
+- Output directory for generated assets (completion, manual) can be customized, see #2515 (@tranzystorek-io)
+- Use the `is-terminal` crate instead of `atty`, see #2530 (@nickelc)
+- Add Winget Releaser workflow, see #2519 (@sitiom)
+- Bump MSRV to 1.70, see #2651 (@mataha)
+
+## Syntaxes
+
+- Associate `os-release` with `bash` syntax, see #2587 (@cyqsimon)
+- Associate `Containerfile` with `Dockerfile` syntax, see #2606 (@einfachIrgendwer0815)
+- Replaced quotes with double quotes so fzf integration example script works on windows and linux. see #2095 (@johnmatthiggins)
+- Associate `ksh` files with `bash` syntax, see #2633 (@johnmatthiggins)
+- Associate `sarif` files with `JSON` syntax, see #2695 (@rhysd)
+- Associate `ron` files with `rust` syntax, see #2427 (@YeungOnion)
+- Add support for [WebGPU Shader Language](https://www.w3.org/TR/WGSL/), see #2692 (@rhysd)
+- Add `.dpkg-new` and `.dpkg-tmp` to ignored suffixe, see #2595 (@scop)
+- fix: Add syntax mapping `*.jsonl` => `json`, see #2539 (@WinterCore)
+- Update `Julia` syntax, see #2553 (@dependabot)
+- add `NSIS` support, see #2577 (@idleberg)
+- Update `ssh-config`, see #2697 (@mrmeszaros)
+
+## `bat` as a library
+
+- Add optional output_buffer arg to `Controller::run()` and `Controller::run_with_error_handler()`, see #2618 (@Piturnah)
+
+
 # v0.23.0
 
 ## Features

CONTRIBUTING.md

@@ -6,21 +6,42 @@ Thank you for considering to contribute to `bat`!
 
 ## Add an entry to the changelog
 
-If your contribution changes the behavior of `bat` (as opposed to a typo-fix
-in the documentation), please update the [`CHANGELOG.md`](CHANGELOG.md) file
-and describe your changes. This makes the release process much easier and
-therefore helps to get your changes into a new `bat` release faster.
+Keeping the [`CHANGELOG.md`](CHANGELOG.md) file up-to-date makes the release
+process much easier and therefore helps to get your changes into a new `bat`
+release faster. However, not every change to the repository requires a
+changelog entry. Below are a few examples of that.
+
+Please update the changelog if your contribution contains changes regarding
+any of the following:
+  - the behavior of `bat`
+  - syntax mappings
+  - syntax definitions
+  - themes
+  - the build system, linting, or CI workflows
+
+A changelog entry is not necessary when:
+  - updating documentation
+  - fixing typos
+
+>[!NOTE]
+> For PRs, a CI workflow verifies that a suitable changelog entry is
+> added. If such an entry is missing, the workflow will fail. If your
+> changes do not need an entry to the changelog (see above), that
+> workflow failure can be disregarded.
+
+
+### Changelog entry format
 
 The top of the `CHANGELOG` contains a *"unreleased"* section with a few
 subsections (Features, Bugfixes, …). Please add your entry to the subsection
 that best describes your change.
 
-Entries follow this format:
+Entries must follow this format:
 ```
 - Short description of what has been changed, see #123 (@user)
 ```
-Here, `#123` is the number of the original issue and/or your pull request.
-Please replace `@user` by your GitHub username.
+Please replace `#123` with the number of your pull request (not issue) and
+`@user` by your GitHub username.
 
 
 ## Development

Cargo.toml

@@ -3,14 +3,14 @@ authors = ["David Peter <mail@david-peter.de>"]
 categories = ["command-line-utilities"]
 description = "A cat(1) clone with wings."
 homepage = "https://github.com/sharkdp/bat"
-license = "MIT/Apache-2.0"
+license = "MIT OR Apache-2.0"
 name = "bat"
 repository = "https://github.com/sharkdp/bat"
-version = "0.23.0"
+version = "0.24.0"
 exclude = ["assets/syntaxes/*", "assets/themes/*"]
-build = "build.rs"
-edition = '2018'
-rust-version = "1.64"
+build = "build/main.rs"
+edition = '2021'
+rust-version = "1.70"
 
 [features]
 default = ["application"]
@@ -25,15 +25,15 @@ application = [
 # Mainly for developers that want to iterate quickly
 # Be aware that the included features might change in the future
 minimal-application = [
-    "atty",
     "clap",
-    "dirs",
+    "etcetera",
     "paging",
     "regex-onig",
     "wild",
 ]
 git = ["git2"] # Support indicating git modifications
 paging = ["shell-words", "grep-cli"] # Support applying a pager on the output
+lessopen = ["run_script", "os_str_bytes"] # Support $LESSOPEN preprocessor
 build-assets = ["syntect/yaml-load", "syntect/plist-load", "regex", "walkdir"]
 
 # You need to use one of these if you depend on bat as a library:
@@ -41,64 +41,79 @@ regex-onig = ["syntect/regex-onig"] # Use the "oniguruma" regex engine
 regex-fancy = ["syntect/regex-fancy"] # Use the rust-only "fancy-regex" engine
 
 [dependencies]
-atty = { version = "0.2.14", optional = true }
-nu-ansi-term = "0.47.0"
+nu-ansi-term = "0.50.0"
 ansi_colours = "^1.2"
 bincode = "1.0"
-console = "0.15.5"
+console = "0.15.8"
 flate2 = "1.0"
-once_cell = "1.17"
+once_cell = "1.19"
 thiserror = "1.0"
-wild = { version = "2.1", optional = true }
+wild = { version = "2.2", optional = true }
 content_inspector = "0.2.4"
-encoding = "0.2"
 shell-words = { version = "1.1.0", optional = true }
-unicode-width = "0.1.10"
+unicode-width = "0.1.11"
 globset = "0.4"
-serde = { version = "1.0", features = ["derive"] }
-serde_yaml = "0.8"
+serde = "1.0"
+serde_derive = "1.0"
+serde_yaml = "0.9.28"
 semver = "1.0"
 path_abs = { version = "0.5", default-features = false }
-clircle = "0.3"
+clircle = "0.5"
 bugreport = { version = "0.5.0", optional = true }
-dirs = { version = "5.0.0", optional = true }
-grep-cli = { version = "0.1.7", optional = true }
-regex = { version = "1.7.0", optional = true }
-walkdir = { version = "2.0", optional = true }
-bytesize = { version = "1.1.0" }
+etcetera = { version = "0.8.0", optional = true }
+grep-cli = { version = "0.1.10", optional = true }
+regex = { version = "1.10.2", optional = true }
+walkdir = { version = "2.4", optional = true }
+bytesize = { version = "1.3.0" }
+encoding_rs = "0.8.33"
+os_str_bytes = { version = "~7.0", optional = true }
+run_script = { version = "^0.10.1", optional = true}
 
 [dependencies.git2]
-version = "0.16"
+version = "0.18"
 optional = true
 default-features = false
 
 [dependencies.syntect]
-version = "5.0.0"
+version = "5.2.0"
 default-features = false
 features = ["parsing"]
 
 [dependencies.clap]
-version = "4.1.8"
+version = "4.4.12"
 optional = true
 features = ["wrap_help", "cargo"]
 
 [target.'cfg(target_os = "macos")'.dependencies]
-dirs = "5.0.0"
-plist = "1.3"
+home = "0.5.9"
+plist = "1.6.0"
 
 [dev-dependencies]
-assert_cmd = "2.0.8"
-expect-test = "1.4.0"
-serial_test = "0.6.0"
-predicates = "2.1.5"
+assert_cmd = "2.0.12"
+expect-test = "1.4.1"
+serial_test = { version = "2.0.0", default-features = false }
+predicates = "3.1.0"
 wait-timeout = "0.2.0"
-tempfile = "3.3.0"
+tempfile = "3.8.1"
+serde = { version = "1.0", features = ["derive"] }
 
 [target.'cfg(unix)'.dev-dependencies]
-nix = { version = "0.26.2", default-features = false, features = ["term"] }
+nix = { version = "0.26.4", default-features = false, features = ["term"] }
+
+[build-dependencies]
+anyhow = "1.0.78"
+indexmap = { version = "2.2.6", features = ["serde"] }
+itertools = "0.12.1"
+once_cell = "1.18"
+regex = "1.10.2"
+serde = "1.0"
+serde_derive = "1.0"
+serde_with = { version = "3.8.1", default-features = false, features = ["macros"] }
+toml = { version = "0.8.9", features = ["preserve_order"] }
+walkdir = "2.4"
 
 [build-dependencies.clap]
-version = "4.1.8"
+version = "4.4.12"
 optional = true
 features = ["wrap_help", "cargo"]
 

README.md

@@ -32,6 +32,16 @@ A special *thank you* goes to our biggest <a href="doc/sponsors.md">sponsors</a>
   <sup>Add Single Sign-On (and more) in minutes instead of months.</sup>
 </a>
 
+<a href="https://www.warp.dev/?utm_source=github&utm_medium=referral&utm_campaign=bat_20231001">
+  <img src="doc/sponsors/warp-logo.png" width="200" alt="Warp">
+  <br>
+  <strong>Warp is a modern, Rust-based terminal with AI built in<br>so you and your team can build great software, faster.</strong>
+  <br>
+  <sub>Feel more productive on the command line with parameterized commands,</sub>
+  <br>
+  <sup>autosuggestions, and an IDE-like text editor.</sup>
+</a>
+
 ### Syntax highlighting
 
 `bat` supports syntax highlighting for a large number of programming and markup
@@ -118,7 +128,7 @@ use `bat`s `--color=always` option to force colorized output. You can also use `
 option to restrict the load times for long files:
 
 ```bash
-fzf --preview 'bat --color=always --style=numbers --line-range=:500 {}'
+fzf --preview "bat --color=always --style=numbers --line-range=:500 {}"
 ```
 
 For more information, see [`fzf`'s `README`](https://github.com/junegunn/fzf#preview-window).
@@ -228,12 +238,23 @@ help() {
 
 Then you can do `$ help cp` or `$ help git commit`.
 
+When you are using `zsh`, you can also use global aliases to override `-h` and `--help` entirely:
+
+```bash
+alias -g -- -h='-h 2>&1 | bat --language=help --style=plain'
+alias -g -- --help='--help 2>&1 | bat --language=help --style=plain'
+```
+
+This way, you can keep on using `cp --help`, but get colorized help pages.
+
+Be aware that in some cases, `-h` may not be a shorthand of `--help` (for example with `ls`).
+
 Please report any issues with the help syntax in [this repository](https://github.com/victor-gp/cmd-help-sublime-syntax).
 
 
 ## Installation
 
-[![Packaging status](https://repology.org/badge/vertical-allrepos/bat-cat.svg)](https://repology.org/project/bat-cat/versions)
+[![Packaging status](https://repology.org/badge/vertical-allrepos/bat-cat.svg?columns=3&exclude_unsupported=1)](https://repology.org/project/bat-cat/versions)
 
 ### On Ubuntu (using `apt`)
 *... and other Debian-based Linux distributions.*
@@ -275,7 +296,7 @@ apk add bat
 
 ### On Arch Linux
 
-You can install [the `bat` package](https://www.archlinux.org/packages/community/x86_64/bat/)
+You can install [the `bat` package](https://www.archlinux.org/packages/extra/x86_64/bat/)
 from the official sources:
 
 ```bash
@@ -390,6 +411,14 @@ take a look at the ["Using `bat` on Windows"](#using-bat-on-windows) section.
 
 You will need to install the [Visual C++ Redistributable](https://support.microsoft.com/en-us/help/2977003/the-latest-supported-visual-c-downloads) package.
 
+#### With WinGet
+
+You can install `bat` via [WinGet](https://learn.microsoft.com/en-us/windows/package-manager/winget):
+
+```bash
+winget install sharkdp.bat
+```
+
 #### With Chocolatey
 
 You can install `bat` via [Chocolatey](https://chocolatey.org/packages/Bat):
@@ -418,7 +447,7 @@ binaries are also available: look for archives with `musl` in the file name.
 
 ### From source
 
-If you want to build `bat` from source, you need Rust 1.64.0 or
+If you want to build `bat` from source, you need Rust 1.70.0 or
 higher. You can then use `cargo` to build everything:
 
 ```bash
@@ -573,7 +602,8 @@ set, `less` is used by default. If you want to use a different pager, you can ei
 `PAGER` variable or set the `BAT_PAGER` environment variable to override what is specified in
 `PAGER`.
 
-**Note**: If `PAGER` is `more` or `most`, `bat` will silently use `less` instead to ensure support for colors.
+>[!NOTE]
+> If `PAGER` is `more` or `most`, `bat` will silently use `less` instead to ensure support for colors.
 
 If you want to pass command-line arguments to the pager, you can also set them via the
 `PAGER`/`BAT_PAGER` variables:
@@ -584,20 +614,37 @@ export BAT_PAGER="less -RF"
 
 Instead of using environment variables, you can also use `bat`s [configuration file](https://github.com/sharkdp/bat#configuration-file) to configure the pager (`--pager` option).
 
-**Note**: By default, if the pager is set to `less` (and no command-line options are specified),
-`bat` will pass the following command line options to the pager: `-R`/`--RAW-CONTROL-CHARS`,
-`-F`/`--quit-if-one-screen` and `-X`/`--no-init`. The last option (`-X`) is only used for `less`
-versions older than 530.
 
-The `-R` option is needed to interpret ANSI colors correctly. The second option (`-F`) instructs
-less to exit immediately if the output size is smaller than the vertical size of the terminal.
-This is convenient for small files because you do not have to press `q` to quit the pager. The
-third option (`-X`) is needed to fix a bug with the `--quit-if-one-screen` feature in old versions
-of `less`. Unfortunately, it also breaks mouse-wheel support in `less`.
+### Using `less` as a pager
 
-If you want to enable mouse-wheel scrolling on older versions of `less`, you can pass just `-R` (as
-in the example above, this will disable the quit-if-one-screen feature). For less 530 or newer,
-it should work out of the box.
+When using `less` as a pager, `bat` will automatically pass extra options along to `less`
+to improve the experience. Specifically, `-R`/`--RAW-CONTROL-CHARS`, `-F`/`--quit-if-one-screen`,
+and under certain conditions, `-X`/`--no-init` and/or `-S`/`--chop-long-lines`.
+
+>[!IMPORTANT]
+> These options will not be added if:
+> - The pager is not named `less`.
+> - The `--pager` argument contains any command-line arguments (e.g. `--pager="less -R"`).
+> - The `BAT_PAGER` environment variable contains any command-line arguments (e.g. `export BAT_PAGER="less -R"`)
+>
+> The `--quit-if-one-screen` option will not be added when:
+> - The `--paging=always` argument is used.
+> - The `BAT_PAGING` environment is set to `always`.
+
+The `-R` option is needed to interpret ANSI colors correctly.
+
+The `-F` option instructs `less` to exit immediately if the output size is smaller than
+the vertical size of the terminal. This is convenient for small files because you do not
+have to press `q` to quit the pager.
+
+The `-X` option is needed to fix a bug with the `--quit-if-one-screen` feature in versions
+of `less` older than version 530. Unfortunately, it also breaks mouse-wheel support in `less`.
+If you want to enable mouse-wheel scrolling on older versions of `less` and do not mind losing
+the quit-if-one-screen feature, you can set the pager (via `--pager` or `BAT_PAGER`) to `less -R`.
+For `less` 530 or newer, it should work out of the box.
+
+The `-S` option is added when `bat`'s `-S`/`--chop-long-lines` option is used. This tells `less`
+to truncate any lines larger than the terminal width.
 
 ### Indentation
 
@@ -684,9 +731,7 @@ Windows 10 natively supports colors in both `conhost.exe` (Command Prompt) and P
 well as in newer versions of bash. On earlier versions of Windows, you can use
 [Cmder](http://cmder.net/), which includes [ConEmu](https://conemu.github.io/).
 
-**Note:** The Git and MSYS versions of `less` do not correctly interpret colors on Windows. If you
-don’t have any other pagers installed, you can disable paging entirely by passing `--paging=never`
-or by setting `BAT_PAGER` to an empty string.
+**Note:** Old versions of `less` do not correctly interpret colors on Windows. To fix this, you can add the optional Unix tools to your PATH when installing Git. If you don’t have any other pagers installed, you can disable paging entirely by passing `--paging=never` or by setting `BAT_PAGER` to an empty string.
 
 ### Cygwin
 

assets/completions/_bat.ps1.in

@@ -59,6 +59,8 @@ Register-ArgumentCompleter -Native -CommandName '{{PROJECT_EXECUTABLE}}' -Script
             [CompletionResult]::new('--unbuffered', 'unbuffered', [CompletionResultType]::ParameterName, 'unbuffered')
             [CompletionResult]::new('--no-config', 'no-config', [CompletionResultType]::ParameterName, 'Do not use the configuration file')
             [CompletionResult]::new('--no-custom-assets', 'no-custom-assets', [CompletionResultType]::ParameterName, 'Do not load custom assets')
+            [CompletionResult]::new('--lessopen', 'lessopen', [CompletionResultType]::ParameterName, 'Enable the $LESSOPEN preprocessor')
+            [CompletionResult]::new('--no-lessopen', 'no-lessopen', [CompletionResultType]::ParameterName, 'Disable the $LESSOPEN preprocessor if enabled (overrides --lessopen)')
             [CompletionResult]::new('--config-file', 'config-file', [CompletionResultType]::ParameterName, 'Show path to the configuration file.')
             [CompletionResult]::new('--generate-config-file', 'generate-config-file', [CompletionResultType]::ParameterName, 'Generates a default configuration file.')
             [CompletionResult]::new('--config-dir', 'config-dir', [CompletionResultType]::ParameterName, 'Show bat''s configuration directory.')

assets/completions/bat.bash.in

@@ -32,7 +32,7 @@ __bat_escape_completions()
 }
 
 _bat() {
-	local cur prev words cword split=false
+	local cur prev words split=false
 	if declare -F _init_completion >/dev/null 2>&1; then
 		_init_completion -s || return 0
 	else
@@ -78,6 +78,7 @@ _bat() {
 	--list-themes | \
 	--line-range | \
 	-L | --list-languages | \
+	--lessopen | \
 	--diagnostic | \
 	--acknowledgements | \
 	-h | --help | \
@@ -171,6 +172,7 @@ _bat() {
 			--style
 			--line-range
 			--list-languages
+			--lessopen
 			--diagnostic
 			--acknowledgements
 			--help

assets/completions/bat.fish.in

@@ -147,6 +147,8 @@ complete -c $bat -s d -l diff -d "Only show lines with Git changes" -n __bat_no_
 
 complete -c $bat -l diff-context -x -d "Show N context lines around Git changes" -n "__fish_seen_argument -s d -l diff"
 
+complete -c $bat -l generate-config-file -f -d "Generates a default configuration file" -n __fish_is_first_arg
+
 complete -c $bat -l file-name -x -d "Specify the display name" -n __bat_no_excl_args
 
 complete -c $bat -s f -l force-colorization -d "Force color and decorations" -n __bat_no_excl_args
@@ -163,6 +165,8 @@ complete -c $bat -l italic-text -x -a "$italic_text_opts" -d "When to use italic
 
 complete -c $bat -s l -l language -x -k -a "(__bat_complete_list_languages)" -d "Set the syntax highlighting language" -n __bat_no_excl_args
 
+complete -c $bat -l lessopen -d "Enable the $LESSOPEN preprocessor" -n __fish_is_first_arg
+
 complete -c $bat -s r -l line-range -x -d "Only print lines [M]:[N] (either optional)" -n __bat_no_excl_args
 
 complete -c $bat -l list-languages -f -d "List syntax highlighting languages" -n __fish_is_first_arg
@@ -171,6 +175,12 @@ complete -c $bat -l list-themes -f -d "List syntax highlighting themes" -n __fis
 
 complete -c $bat -s m -l map-syntax -x -a "(__bat_complete_map_syntax)" -d "Map <glob pattern>:<language syntax>" -n __bat_no_excl_args
 
+complete -c $bat -l no-config -d "Do not use the configuration file"
+
+complete -c $bat -l no-custom-assets -d "Do not load custom assets"
+
+complete -c $bat -l no-lessopen -d "Disable the $LESSOPEN preprocessor if enabled (overrides --lessopen)"
+
 complete -c $bat -s n -l number -d "Only show line numbers, no other decorations" -n __bat_no_excl_args
 
 complete -c $bat -l pager -x -a "$pager_opts" -d "Which pager to use" -n __bat_no_excl_args

assets/completions/bat.zsh.in

@@ -1,19 +1,20 @@
 #compdef {{PROJECT_EXECUTABLE}}
 
-local context state state_descr line
+local curcontext="$curcontext" ret=1
+local -a state state_descr line
 typeset -A opt_args
 
 (( $+functions[_{{PROJECT_EXECUTABLE}}_cache_subcommand] )) ||
 _{{PROJECT_EXECUTABLE}}_cache_subcommand() {
     local -a args
     args=(
-        '(-b --build -c --clear)'{-b,--build}'[Initialize or update the syntax/theme cache]'
-        '(-b --build -c --clear)'{-c,--clear}'[Remove the cached syntax definitions and themes]'
-        '(--source)'--source='[Use a different directory to load syntaxes and themes from]:directory:_files -/'
-        '(--target)'--target='[Use a different directory to store the cached syntax and theme set]:directory:_files -/'
-        '(--blank)'--blank'[Create completely new syntax and theme sets]'
-        '(: -)'{-h,--help}'[Prints help information]'
-        '*: :'
+        '(-b --build -c --clear)'{-b,--build}'[initialize or update the syntax/theme cache]'
+        '(-b --build -c --clear)'{-c,--clear}'[remove the cached syntax definitions and themes]'
+        --source='[specify directory to load syntaxes and themes from]:directory:_files -/'
+        --target='[specify directory to store the cached syntax and theme set in]:directory:_files -/'
+        --blank'[create completely new syntax and theme sets]'
+        --acknowledgements'[build acknowledgements.bin]'
+        '(: -)'{-h,--help}'[show help information]'
     )
 
     _arguments -S -s $args
@@ -23,67 +24,79 @@ _{{PROJECT_EXECUTABLE}}_cache_subcommand() {
 _{{PROJECT_EXECUTABLE}}_main() {
     local -a args
     args=(
-        '(-A --show-all)'{-A,--show-all}'[Show non-printable characters (space, tab, newline, ..)]'
-        '*'{-p,--plain}'[Show plain style (alias for `--style=plain`), repeat twice to disable disable automatic paging (alias for `--paging=never`)]'
-        '(-l --language)'{-l+,--language=}'[Set the language for syntax highlighting]:<language>:->language'
-        '(-H --highlight-line)'{-H,--highlight-line}'[Highlight lines N through M]:<N\:M>...'
-        '(--file-name)'--file-name'[Specify the name to display for a file]:<name>...:_files'
-        '(-d --diff)'--diff'[Only show lines that have been added/removed/modified]'
-        '(--diff-context)'--diff-context'[Include N lines of context around added/removed/modified lines when using `--diff`]:<N> (lines):()'
-        '(--tabs)'--tabs'[Set the tab width to T spaces]:<T> (tab width):()'
-        '(--wrap)'--wrap='[Specify the text-wrapping mode]:<when>:(auto never character)'
-        '(--terminal-width)'--terminal-width'[Explicitly set the width of the terminal instead of determining it automatically]:<width>'
-        '(-n --number)'{-n,--number}'[Show line numbers]'
-        '(--color)'--color='[When to use colors]:<when>:(auto never always)'
-        '(--italic-text)'--italic-text='[Use italics in output]:<when>:(always never)'
-        '(--decorations)'--decorations='[When to show the decorations]:<when>:(auto never always)'
-        '(--paging)'--paging='[Specify when to use the pager]:<when>:(auto never always)'
-        '(-m --map-syntax)'{-m+,--map-syntax=}'[Use the specified syntax for files matching the glob pattern]:<glob\:syntax>...'
-        '(--theme)'--theme='[Set the color theme for syntax highlighting]:<theme>:->theme'
-        '(: --list-themes --list-languages -L)'--list-themes'[Display all supported highlighting themes]'
-        '(--style)'--style='[Comma-separated list of style elements to display]:<components>:->style'
-        '(-r --line-range)'{-r+,--line-range=}'[Only print the lines from N to M]:<N\:M>...'
-        '(: --list-themes --list-languages -L)'{-L,--list-languages}'[Display all supported languages]'
-        '(: --no-config)'--no-config'[Do not use the configuration file]'
-        '(: --no-custom-assets)'--no-custom-assets'[Do not load custom assets]'
-        '(: --config-dir)'--config-dir'[Show bat'"'"'s configuration directory]'
-        '(: --config-file)'--config-file'[Show path to the configuration file]'
-        '(: --generate-config-file)'--generate-config-file'[Generates a default configuration file]'
-        '(: --cache-dir)'--cache-dir'[Show bat'"'"'s cache directory]'
-        '(: -)'{-h,--help}'[Print this help message]'
-        '(: -)'{-V,--version}'[Show version information]'
-        '*: :_files'
+        '(-A --show-all)'{-A,--show-all}'[show non-printable characters (space, tab, newline, ..)]'
+        --nonprintable-notation='[specify how to display non-printable characters when using --show-all]:notation:(caret unicode)'
+        \*{-p,--plain}'[show plain style (alias for `--style=plain`), repeat twice to disable disable automatic paging (alias for `--paging=never`)]'
+        '(-l --language)'{-l+,--language=}'[set the language for syntax highlighting]:language:->languages'
+        \*{-H+,--highlight-line=}'[highlight specified block of lines]:start\:end'
+        \*--file-name='[specify the name to display for a file]:name:_files'
+        '(-d --diff)'--diff'[only show lines that have been added/removed/modified]'
+        --diff-context='[specify lines of context around added/removed/modified lines when using `--diff`]:lines'
+        --tabs='[set the tab width]:tab width [4]'
+        --wrap='[specify the text-wrapping mode]:mode [auto]:(auto never character)'
+        '!(--wrap)'{-S,--chop-long-lines}
+        --terminal-width='[explicitly set the width of the terminal instead of determining it automatically]:width'
+        '(-n --number --diff --diff-context)'{-n,--number}'[show line numbers]'
+        --color='[specify when to use colors]:when:(auto never always)'
+        --italic-text='[use italics in output]:when:(always never)'
+        --decorations='[specify when to show the decorations]:when:(auto never always)'
+        --paging='[specify when to use the pager]:when:(auto never always)'
+        '(-m --map-syntax)'{-m+,--map-syntax=}'[map a glob pattern to an existing syntax name]: :->syntax-maps'
+        '(--theme)'--theme='[set the color theme for syntax highlighting]:theme:->themes'
+        '(: --list-themes --list-languages -L)'--list-themes'[show all supported highlighting themes]'
+        --style='[comma-separated list of style elements to display]: : _values "style [default]"
+            default auto full plain changes header header-filename header-filesize grid rule numbers snip'
+        \*{-r+,--line-range=}'[only print the specified line range]:start\:end'
+        '(* -)'{-L,--list-languages}'[display all supported languages]'
+        "--no-config[don't use the configuration file]"
+        "--no-custom-assets[don't load custom assets]"
+        '(--no-lessopen)'--lessopen'[enable the $LESSOPEN preprocessor]'
+        '(--lessopen)'--no-lessopen'[disable the $LESSOPEN preprocessor if enabled (overrides --lessopen)]'
+        '(* -)'--config-dir"[show bat's configuration directory]"
+        '(* -)'--config-file'[show path to the configuration file]'
+        '(* -)'--generate-config-file'[generate a default configuration file]'
+        '(* -)'--cache-dir"[show bat's cache directory]"
+        '(* -)'{-h,--help}'[show help information]'
+        '(* -)'{-V,--version}'[show version information]'
+        '*: :{ _files || compadd cache }'
     )
 
-    _arguments -S -s $args
+    _arguments -S -s $args && ret=0
 
     case "$state" in
-        language)
+        syntax-maps)
+          if ! compset -P '*:'; then
+            _message -e patterns 'glob pattern:language'
+            return
+          fi
+        ;& # fall-through
+
+        languages)
             local IFS=$'\n'
             local -a languages
             languages=( $({{PROJECT_EXECUTABLE}} --list-languages | awk -F':|,' '{ for (i = 1; i <= NF; ++i) printf("%s:%s\n", $i, $1) }') )
 
-            _describe 'language' languages
+            _describe 'language' languages && ret=0
         ;;
 
-        theme)
-            local IFS=$'\n'
-            local -a themes
-            themes=( $({{PROJECT_EXECUTABLE}} --list-themes | sort) )
+        themes)
+            local -a themes expl
+            themes=( ${(f)"$(_call_program themes {{PROJECT_EXECUTABLE}} --list-themes)"} )
 
-            _values 'theme' $themes
-        ;;
-
-        style)
-            _values -s , 'style' default auto full plain changes header header-filename header-filesize grid rule numbers snip
+            _wanted themes expl 'theme' compadd -a themes && ret=0
         ;;
     esac
+
+    return ret
 }
 
 case $words[2] in
     cache)
         ## Completion of the 'cache' command itself is removed for better UX
         ## See https://github.com/sharkdp/bat/issues/2085#issuecomment-1271646802
+        shift words
+        (( CURRENT-- ))
+        curcontext="${curcontext%:*}-${words[1]}:"
         _{{PROJECT_EXECUTABLE}}_cache_subcommand
     ;;
 

assets/manual/bat.1.in

@@ -25,7 +25,7 @@ either '--language value', '--language=value', '-l value' or '-lvalue'.
 Show non\-printable characters like space, tab or newline. Use '\-\-tabs' to
 control the width of the tab\-placeholders.
 .HP
-\fB\-\-nonprintable\-notation\fR
+\fB\-\-nonprintable\-notation\fR <notation>
 .IP
 Specify how to display non-printable characters when using \-\-show\-all.
 
@@ -243,6 +243,23 @@ If you ever want to remove the custom languages, you can clear the cache with `\
 Similarly to custom languages, {{PROJECT_EXECUTABLE}} supports Sublime Text \fB.tmTheme\fR themes.
 These can be installed to `\fB$({{PROJECT_EXECUTABLE}} --config-dir)/themes\fR`, and are added to the cache with
 `\fB{{PROJECT_EXECUTABLE}} cache --build`.
+
+.SH "INPUT PREPROCESSOR"
+Much like less(1) does, {{PROJECT_EXECUTABLE}} supports input preprocessors via the LESSOPEN and LESSCLOSE environment variables.
+In addition, {{PROJECT_EXECUTABLE}} attempts to be as compatible with less's preprocessor implementation as possible.
+
+To use the preprocessor, call:
+
+\fB{{PROJECT_EXECUTABLE}} --lessopen\fR
+
+Alternatively, the preprocessor may be enabled by default by adding the '\-\-lessopen' option to the configuration file. 
+
+To temporarily disable the preprocessor if it is enabled by default, call:
+
+\fB{{PROJECT_EXECUTABLE}} --no-lessopen\fR
+
+For more information, see the "INPUT PREPROCESSOR" section of less(1).  
+
 .SH "MORE INFORMATION"
 
 For more information and up-to-date documentation, visit the {{PROJECT_EXECUTABLE}} repo:

assets/patches/JavaScript.sublime-syntax.patch

@@ -0,0 +1,14 @@
+Submodule assets/syntaxes/01_Packages contains modified content
+diff --git syntaxes/01_Packages/JavaScript/JavaScript.sublime-syntax syntaxes/01_Packages/JavaScript/JavaScript.sublime-syntax
+index 05a4fed6..78a7bf55 100644
+--- syntaxes/01_Packages/JavaScript/JavaScript.sublime-syntax
++++ syntaxes/01_Packages/JavaScript/JavaScript.sublime-syntax
+@@ -5,7 +5,7 @@ name: JavaScript
+ file_extensions:
+   - js
+   - htc
+-first_line_match: ^#!\s*/.*\b(node|js)\b
++first_line_match: ^#!\s*/.*\b(node|bun|js)\b
+ scope: source.js
+ variables:
+   bin_digit: '[01_]'

assets/syntaxes/02_Extra/Docker

@@ -1 +1 @@
-Subproject commit 9e9a518aed93031042c54710f8f02c839301de26
+Subproject commit 0f6b7bc87acf684f7b0790fd480731ffb4615b87

assets/syntaxes/02_Extra/Julia

@@ -1 +1 @@
-Subproject commit 4fde0fdeddb3ca8486d3f490a2f051cba39a0a48
+Subproject commit 3366b10be91aaab7a61ae0bc0a5af5cc375e58d1

assets/syntaxes/02_Extra/NSIS

@@ -0,0 +1 @@
+Subproject commit 619a65a04efbf343bdfcde875700675b9b273368

assets/syntaxes/02_Extra/SublimeJQ

@@ -1 +1 @@
-Subproject commit 687058289c1a888e0895378432d66b41609a84d8
+Subproject commit b7e53e5d86814f04a48d2e441bcf5f9fdf07e9c1

assets/syntaxes/02_Extra/cmd-help

@@ -1 +1 @@
-Subproject commit f41e5fc8381fe200a4e18c410bb41e911110a6e9
+Subproject commit 209559b72f7e8848c988828088231b3a4d8b6838

assets/syntaxes/02_Extra/ssh-config

@@ -1 +1 @@
-Subproject commit e1012e9f13c6073f559b14206df2ede35720e884
+Subproject commit bf49e9181c6bf992a86beb133144d2651e826ddc

assets/syntaxes/02_Extra/vscode-wgsl

@@ -0,0 +1 @@
+Subproject commit acf26718d7a327377641e31d8f9a9dab376efa84

assets/syntaxes/02_Extra/wgsl.sublime-syntax

@@ -0,0 +1,198 @@
+%YAML 1.2
+---
+# http://www.sublimetext.com/docs/syntax.html
+name: WGSL
+file_extensions:
+  - wgsl
+scope: source.wgsl
+contexts:
+  main:
+    - include: line_comments
+    - include: block_comments
+    - include: keywords
+    - include: attributes
+    - include: functions
+    - include: function_calls
+    - include: constants
+    - include: types
+    - include: variables
+    - include: punctuation
+  attributes:
+    - match: '(@)([A-Za-z_]+)'
+      comment: attribute declaration
+      scope: meta.attribute.wgsl
+      captures:
+        1: keyword.operator.attribute.at
+        2: entity.name.attribute.wgsl
+  block_comments:
+    - match: /\*\*/
+      comment: empty block comments
+      scope: comment.block.wgsl
+    - match: /\*\*
+      comment: block documentation comments
+      push:
+        - meta_scope: comment.block.documentation.wgsl
+        - match: \*/
+          pop: true
+        - include: block_comments
+    - match: /\*(?!\*)
+      comment: block comments
+      push:
+        - meta_scope: comment.block.wgsl
+        - match: \*/
+          pop: true
+        - include: block_comments
+  constants:
+    - match: '(-?\b[0-9][0-9]*\.[0-9][0-9]*)([eE][+-]?[0-9]+)?\b'
+      comment: decimal float literal
+      scope: constant.numeric.float.wgsl
+    - match: '-?\b0x[0-9a-fA-F]+\b|\b0\b|-?\b[1-9][0-9]*\b'
+      comment: int literal
+      scope: constant.numeric.decimal.wgsl
+    - match: '\b0x[0-9a-fA-F]+u\b|\b0u\b|\b[1-9][0-9]*u\b'
+      comment: uint literal
+      scope: constant.numeric.decimal.wgsl
+    - match: \b(true|false)\b
+      comment: boolean constant
+      scope: constant.language.boolean.wgsl
+  function_calls:
+    - match: '([A-Za-z0-9_]+)(\()'
+      comment: function/method calls
+      captures:
+        1: entity.name.function.wgsl
+        2: punctuation.brackets.round.wgsl
+      push:
+        - meta_scope: meta.function.call.wgsl
+        - match: \)
+          captures:
+            0: punctuation.brackets.round.wgsl
+          pop: true
+        - include: line_comments
+        - include: block_comments
+        - include: keywords
+        - include: attributes
+        - include: function_calls
+        - include: constants
+        - include: types
+        - include: variables
+        - include: punctuation
+  functions:
+    - match: '\b(fn)\s+([A-Za-z0-9_]+)((\()|(<))'
+      comment: function definition
+      captures:
+        1: keyword.other.fn.wgsl
+        2: entity.name.function.wgsl
+        4: punctuation.brackets.round.wgsl
+      push:
+        - meta_scope: meta.function.definition.wgsl
+        - match: '\{'
+          captures:
+            0: punctuation.brackets.curly.wgsl
+          pop: true
+        - include: line_comments
+        - include: block_comments
+        - include: keywords
+        - include: attributes
+        - include: function_calls
+        - include: constants
+        - include: types
+        - include: variables
+        - include: punctuation
+  keywords:
+    - match: \b(bitcast|block|break|case|continue|continuing|default|discard|else|elseif|enable|fallthrough|for|function|if|loop|private|read|read_write|return|storage|switch|uniform|while|workgroup|write)\b
+      comment: other keywords
+      scope: keyword.control.wgsl
+    - match: \b(asm|const|do|enum|handle|mat|premerge|regardless|typedef|unless|using|vec|void)\b
+      comment: reserved keywords
+      scope: keyword.control.wgsl
+    - match: \b(let|var)\b
+      comment: storage keywords
+      scope: keyword.other.wgsl storage.type.wgsl
+    - match: \b(type)\b
+      comment: type keyword
+      scope: keyword.declaration.type.wgsl storage.type.wgsl
+    - match: \b(enum)\b
+      comment: enum keyword
+      scope: keyword.declaration.enum.wgsl storage.type.wgsl
+    - match: \b(struct)\b
+      comment: struct keyword
+      scope: keyword.declaration.struct.wgsl storage.type.wgsl
+    - match: \bfn\b
+      comment: fn
+      scope: keyword.other.fn.wgsl
+    - match: (\^|\||\|\||&&|<<|>>|!)(?!=)
+      comment: logical operators
+      scope: keyword.operator.logical.wgsl
+    - match: '&(?![&=])'
+      comment: logical AND, borrow references
+      scope: keyword.operator.borrow.and.wgsl
+    - match: (\+=|-=|\*=|/=|%=|\^=|&=|\|=|<<=|>>=)
+      comment: assignment operators
+      scope: keyword.operator.assignment.wgsl
+    - match: '(?<![<>])=(?!=|>)'
+      comment: single equal
+      scope: keyword.operator.assignment.equal.wgsl
+    - match: (=(=)?(?!>)|!=|<=|(?<!=)>=)
+      comment: comparison operators
+      scope: keyword.operator.comparison.wgsl
+    - match: '(([+%]|(\*(?!\w)))(?!=))|(-(?!>))|(/(?!/))'
+      comment: math operators
+      scope: keyword.operator.math.wgsl
+    - match: \.(?!\.)
+      comment: dot access
+      scope: keyword.operator.access.dot.wgsl
+    - match: '->'
+      comment: dashrocket, skinny arrow
+      scope: keyword.operator.arrow.skinny.wgsl
+  line_comments:
+    - match: \s*//.*
+      comment: single line comment
+      scope: comment.line.double-slash.wgsl
+  punctuation:
+    - match: ','
+      comment: comma
+      scope: punctuation.comma.wgsl
+    - match: '[{}]'
+      comment: curly braces
+      scope: punctuation.brackets.curly.wgsl
+    - match: '[()]'
+      comment: parentheses, round brackets
+      scope: punctuation.brackets.round.wgsl
+    - match: ;
+      comment: semicolon
+      scope: punctuation.semi.wgsl
+    - match: '[\[\]]'
+      comment: square brackets
+      scope: punctuation.brackets.square.wgsl
+    - match: '(?<![=-])[<>]'
+      comment: angle brackets
+      scope: punctuation.brackets.angle.wgsl
+  types:
+    - match: \b(bool|i32|u32|f32)\b
+      comment: scalar Types
+      scope: storage.type.wgsl
+    - match: \b(i64|u64|f64)\b
+      comment: reserved scalar Types
+      scope: storage.type.wgsl
+    - match: \b(vec2i|vec3i|vec4i|vec2u|vec3u|vec4u|vec2f|vec3f|vec4f|vec2h|vec3h|vec4h)\b
+      comment: vector type aliasses
+      scope: storage.type.wgsl
+    - match: \b(mat2x2f|mat2x3f|mat2x4f|mat3x2f|mat3x3f|mat3x4f|mat4x2f|mat4x3f|mat4x4f|mat2x2h|mat2x3h|mat2x4h|mat3x2h|mat3x3h|mat3x4h|mat4x2h|mat4x3h|mat4x4h)\b
+      comment: matrix type aliasses
+      scope: storage.type.wgsl
+    - match: '\b(vec[2-4]|mat[2-4]x[2-4])\b'
+      comment: vector/matrix types
+      scope: storage.type.wgsl
+    - match: \b(atomic)\b
+      comment: atomic types
+      scope: storage.type.wgsl
+    - match: \b(array)\b
+      comment: array types
+      scope: storage.type.wgsl
+    - match: '\b([A-Z][A-Za-z0-9]*)\b'
+      comment: Custom type
+      scope: entity.name.type.wgsl
+  variables:
+    - match: '\b(?<!(?<!\.)\.)(?:r#(?!(crate|[Ss]elf|super)))?[a-z0-9_]+\b'
+      comment: variables
+      scope: variable.other.wgsl

assets/themes/zenburn

@@ -1 +1 @@
-Subproject commit 43dc527731731666d6d2b1311e86951a8ce07fec
+Subproject commit 86d4ee7a1f884851a1d21d66249687f527fced32

build.rs

@@ -1,106 +0,0 @@
-// TODO: Re-enable generation of shell completion files (below) when clap 3 is out.
-// For more details, see https://github.com/sharkdp/bat/issues/372
-
-// For bat-as-a-library, no build script is required. The build script is for
-// the manpage and completions, which are only relevant to the bat application.
-#[cfg(not(feature = "application"))]
-fn main() {}
-
-#[cfg(feature = "application")]
-fn main() -> Result<(), Box<dyn std::error::Error>> {
-    use std::collections::HashMap;
-    use std::error::Error;
-    use std::fs;
-    use std::path::Path;
-
-    // Read environment variables.
-    let project_name = option_env!("PROJECT_NAME").unwrap_or("bat");
-    let executable_name = option_env!("PROJECT_EXECUTABLE").unwrap_or(project_name);
-    let executable_name_uppercase = executable_name.to_uppercase();
-    static PROJECT_VERSION: &str = env!("CARGO_PKG_VERSION");
-
-    /// Generates a file from a template.
-    fn template(
-        variables: &HashMap<&str, &str>,
-        in_file: &str,
-        out_file: impl AsRef<Path>,
-    ) -> Result<(), Box<dyn Error>> {
-        let mut content = fs::read_to_string(in_file)?;
-
-        for (variable_name, value) in variables {
-            // Replace {{variable_name}} by the value
-            let pattern = format!("{{{{{variable_name}}}}}", variable_name = variable_name);
-            content = content.replace(&pattern, value);
-        }
-
-        fs::write(out_file, content)?;
-        Ok(())
-    }
-
-    let mut variables = HashMap::new();
-    variables.insert("PROJECT_NAME", project_name);
-    variables.insert("PROJECT_EXECUTABLE", executable_name);
-    variables.insert("PROJECT_EXECUTABLE_UPPERCASE", &executable_name_uppercase);
-    variables.insert("PROJECT_VERSION", PROJECT_VERSION);
-
-    let out_dir_env = std::env::var_os("OUT_DIR").expect("OUT_DIR to be set in build.rs");
-    let out_dir = Path::new(&out_dir_env);
-
-    fs::create_dir_all(out_dir.join("assets/manual")).unwrap();
-    fs::create_dir_all(out_dir.join("assets/completions")).unwrap();
-
-    template(
-        &variables,
-        "assets/manual/bat.1.in",
-        out_dir.join("assets/manual/bat.1"),
-    )?;
-    template(
-        &variables,
-        "assets/completions/bat.bash.in",
-        out_dir.join("assets/completions/bat.bash"),
-    )?;
-    template(
-        &variables,
-        "assets/completions/bat.fish.in",
-        out_dir.join("assets/completions/bat.fish"),
-    )?;
-    template(
-        &variables,
-        "assets/completions/_bat.ps1.in",
-        out_dir.join("assets/completions/_bat.ps1"),
-    )?;
-    template(
-        &variables,
-        "assets/completions/bat.zsh.in",
-        out_dir.join("assets/completions/bat.zsh"),
-    )?;
-
-    Ok(())
-}
-
-// #[macro_use]
-// extern crate clap;
-
-// use clap::Shell;
-// use std::fs;
-
-// include!("src/clap_app.rs");
-
-// const BIN_NAME: &str = "bat";
-
-// fn main() {
-//     let outdir = std::env::var_os("SHELL_COMPLETIONS_DIR").or(std::env::var_os("OUT_DIR"));
-
-//     let outdir = match outdir {
-//         None => return,
-//         Some(outdir) => outdir,
-//     };
-
-//     fs::create_dir_all(&outdir).unwrap();
-
-//     let mut app = build_app(true);
-//     app.gen_completions(BIN_NAME, Shell::Bash, &outdir);
-//     app.gen_completions(BIN_NAME, Shell::Fish, &outdir);
-//     app.gen_completions(BIN_NAME, Shell::Zsh, &outdir);
-//     app.gen_completions(BIN_NAME, Shell::PowerShell, &outdir);
-// }

build/application.rs

@@ -0,0 +1,67 @@
+use std::{env, fs, path::PathBuf};
+
+use crate::util::render_template;
+
+/// Generate manpage and shell completions for the bat application.
+pub fn gen_man_and_comp() -> anyhow::Result<()> {
+    println!("cargo:rerun-if-changed=assets/manual/");
+    println!("cargo:rerun-if-changed=assets/completions/");
+
+    println!("cargo:rerun-if-env-changed=PROJECT_NAME");
+    println!("cargo:rerun-if-env-changed=PROJECT_EXECUTABLE");
+    println!("cargo:rerun-if-env-changed=CARGO_PKG_VERSION");
+    println!("cargo:rerun-if-env-changed=BAT_ASSETS_GEN_DIR");
+
+    // Read environment variables.
+    let project_name = env::var("PROJECT_NAME").unwrap_or("bat".into());
+    let executable_name = env::var("PROJECT_EXECUTABLE").unwrap_or(project_name.clone());
+    let executable_name_uppercase = executable_name.to_uppercase();
+    let project_version = env::var("CARGO_PKG_VERSION")?;
+
+    let variables = [
+        ("PROJECT_NAME", project_name),
+        ("PROJECT_EXECUTABLE", executable_name),
+        ("PROJECT_EXECUTABLE_UPPERCASE", executable_name_uppercase),
+        ("PROJECT_VERSION", project_version),
+    ]
+    .into_iter()
+    .collect();
+
+    let Some(out_dir) = env::var_os("BAT_ASSETS_GEN_DIR")
+        .or_else(|| env::var_os("OUT_DIR"))
+        .map(PathBuf::from)
+    else {
+        anyhow::bail!("BAT_ASSETS_GEN_DIR or OUT_DIR should be set for build.rs");
+    };
+
+    fs::create_dir_all(out_dir.join("assets/manual")).unwrap();
+    fs::create_dir_all(out_dir.join("assets/completions")).unwrap();
+
+    render_template(
+        &variables,
+        "assets/manual/bat.1.in",
+        out_dir.join("assets/manual/bat.1"),
+    )?;
+    render_template(
+        &variables,
+        "assets/completions/bat.bash.in",
+        out_dir.join("assets/completions/bat.bash"),
+    )?;
+    render_template(
+        &variables,
+        "assets/completions/bat.fish.in",
+        out_dir.join("assets/completions/bat.fish"),
+    )?;
+    render_template(
+        &variables,
+        "assets/completions/_bat.ps1.in",
+        out_dir.join("assets/completions/_bat.ps1"),
+    )?;
+    render_template(
+        &variables,
+        "assets/completions/bat.zsh.in",
+        out_dir.join("assets/completions/bat.zsh"),
+    )?;
+
+    Ok(())
+}

build/main.rs

@@ -0,0 +1,17 @@
+#[cfg(feature = "application")]
+mod application;
+mod syntax_mapping;
+mod util;
+
+fn main() -> anyhow::Result<()> {
+    // only watch manually-designated files
+    // see: https://doc.rust-lang.org/cargo/reference/build-scripts.html#rerun-if-changed
+    println!("cargo:rerun-if-changed=build/");
+
+    syntax_mapping::build_static_mappings()?;
+
+    #[cfg(feature = "application")]
+    application::gen_man_and_comp()?;
+
+    Ok(())
+}

build/syntax_mapping.rs

@@ -0,0 +1,294 @@
+use std::{
+    convert::Infallible,
+    env, fs,
+    path::{Path, PathBuf},
+    str::FromStr,
+};
+
+use anyhow::{anyhow, bail};
+use indexmap::IndexMap;
+use itertools::Itertools;
+use once_cell::sync::Lazy;
+use regex::Regex;
+use serde_derive::Deserialize;
+use serde_with::DeserializeFromStr;
+use walkdir::WalkDir;
+
+/// Known mapping targets.
+///
+/// Corresponds to `syntax_mapping::MappingTarget`.
+#[allow(clippy::enum_variant_names)]
+#[derive(Clone, Debug, Eq, PartialEq, Hash, DeserializeFromStr)]
+pub enum MappingTarget {
+    MapTo(String),
+    MapToUnknown,
+    MapExtensionToUnknown,
+}
+impl FromStr for MappingTarget {
+    type Err = Infallible;
+    fn from_str(s: &str) -> Result<Self, Self::Err> {
+        match s {
+            "MappingTarget::MapToUnknown" => Ok(Self::MapToUnknown),
+            "MappingTarget::MapExtensionToUnknown" => Ok(Self::MapExtensionToUnknown),
+            syntax => Ok(Self::MapTo(syntax.into())),
+        }
+    }
+}
+impl MappingTarget {
+    fn codegen(&self) -> String {
+        match self {
+            Self::MapTo(syntax) => format!(r###"MappingTarget::MapTo(r#"{syntax}"#)"###),
+            Self::MapToUnknown => "MappingTarget::MapToUnknown".into(),
+            Self::MapExtensionToUnknown => "MappingTarget::MapExtensionToUnknown".into(),
+        }
+    }
+}
+
+#[derive(Clone, Debug, PartialEq, Eq, Hash, DeserializeFromStr)]
+/// A single matcher.
+///
+/// Codegen converts this into a `Lazy<Option<GlobMatcher>>`.
+struct Matcher(Vec<MatcherSegment>);
+/// Parse a matcher.
+///
+/// Note that this implementation is rather strict: it will greedily interpret
+/// every valid environment variable replacement as such, then immediately
+/// hard-error if it finds a '$' anywhere in the remaining text segments.
+///
+/// The reason for this strictness is I currently cannot think of a valid reason
+/// why you would ever need '$' as plaintext in a glob pattern. Therefore any
+/// such occurrences are likely human errors.
+///
+/// If we later discover some edge cases, it's okay to make it more permissive.
+///
+/// Revision history:
+/// - 2024-02-20: allow `{` and `}` (glob brace expansion)
+impl FromStr for Matcher {
+    type Err = anyhow::Error;
+    fn from_str(s: &str) -> Result<Self, Self::Err> {
+        use MatcherSegment as Seg;
+        static VAR_REGEX: Lazy<Regex> = Lazy::new(|| Regex::new(r"\$\{([\w\d_]+)\}").unwrap());
+
+        let mut segments = vec![];
+        let mut text_start = 0;
+        for capture in VAR_REGEX.captures_iter(s) {
+            let match_0 = capture.get(0).unwrap();
+
+            // text before this var
+            let text_end = match_0.start();
+            segments.push(Seg::Text(s[text_start..text_end].into()));
+            text_start = match_0.end();
+
+            // this var
+            segments.push(Seg::Env(capture.get(1).unwrap().as_str().into()));
+        }
+        // possible trailing text
+        segments.push(Seg::Text(s[text_start..].into()));
+
+        // cleanup empty text segments
+        let non_empty_segments = segments
+            .into_iter()
+            .filter(|seg| seg.text().map(|t| !t.is_empty()).unwrap_or(true))
+            .collect_vec();
+
+        // sanity check
+        if non_empty_segments
+            .windows(2)
+            .any(|segs| segs[0].is_text() && segs[1].is_text())
+        {
+            unreachable!("Parsed into consecutive text segments: {non_empty_segments:?}");
+        }
+
+        // guard empty case
+        if non_empty_segments.is_empty() {
+            bail!(r#"Parsed an empty matcher: "{s}""#);
+        }
+
+        // guard variable syntax leftover fragments
+        if non_empty_segments
+            .iter()
+            .filter_map(Seg::text)
+            .any(|t| t.contains('$'))
+        {
+            bail!(r#"Invalid matcher: "{s}""#);
+        }
+
+        Ok(Self(non_empty_segments))
+    }
+}
+impl Matcher {
+    fn codegen(&self) -> String {
+        match self.0.len() {
+            0 => unreachable!("0-length matcher should never be created"),
+            // if-let guard would be ideal here
+            // see: https://github.com/rust-lang/rust/issues/51114
+            1 if self.0[0].is_text() => {
+                let s = self.0[0].text().unwrap();
+                format!(r###"Lazy::new(|| Some(build_matcher_fixed(r#"{s}"#)))"###)
+            }
+            // parser logic ensures that this case can only happen when there are dynamic segments
+            _ => {
+                let segs = self.0.iter().map(MatcherSegment::codegen).join(", ");
+                format!(r###"Lazy::new(|| build_matcher_dynamic(&[{segs}]))"###)
+            }
+        }
+    }
+}
+
+/// A segment in a matcher.
+///
+/// Corresponds to `syntax_mapping::MatcherSegment`.
+#[derive(Debug, Clone, PartialEq, Eq, Hash)]
+enum MatcherSegment {
+    Text(String),
+    Env(String),
+}
+#[allow(dead_code)]
+impl MatcherSegment {
+    fn is_text(&self) -> bool {
+        matches!(self, Self::Text(_))
+    }
+    fn is_env(&self) -> bool {
+        matches!(self, Self::Env(_))
+    }
+    fn text(&self) -> Option<&str> {
+        match self {
+            Self::Text(t) => Some(t),
+            Self::Env(_) => None,
+        }
+    }
+    fn env(&self) -> Option<&str> {
+        match self {
+            Self::Text(_) => None,
+            Self::Env(t) => Some(t),
+        }
+    }
+    fn codegen(&self) -> String {
+        match self {
+            Self::Text(s) => format!(r###"MatcherSegment::Text(r#"{s}"#)"###),
+            Self::Env(s) => format!(r###"MatcherSegment::Env(r#"{s}"#)"###),
+        }
+    }
+}
+
+/// A struct that models a single .toml file in /src/syntax_mapping/builtins/.
+#[derive(Clone, Debug, Deserialize)]
+struct MappingDefModel {
+    mappings: IndexMap<MappingTarget, Vec<Matcher>>,
+}
+impl MappingDefModel {
+    fn into_mapping_list(self) -> MappingList {
+        let list = self
+            .mappings
+            .into_iter()
+            .flat_map(|(target, matchers)| {
+                matchers
+                    .into_iter()
+                    .map(|matcher| (matcher, target.clone()))
+                    .collect::<Vec<_>>()
+            })
+            .collect();
+        MappingList(list)
+    }
+}
+
+#[derive(Clone, Debug)]
+struct MappingList(Vec<(Matcher, MappingTarget)>);
+impl MappingList {
+    fn codegen(&self) -> String {
+        let array_items: Vec<_> = self
+            .0
+            .iter()
+            .map(|(matcher, target)| {
+                format!("({m}, {t})", m = matcher.codegen(), t = target.codegen())
+            })
+            .collect();
+        let len = array_items.len();
+
+        format!(
+            "/// Generated by build script from /src/syntax_mapping/builtins/.\n\
+            pub(crate) static BUILTIN_MAPPINGS: [(Lazy<Option<GlobMatcher>>, MappingTarget); {len}] = [\n{items}\n];",
+            items = array_items.join(",\n")
+        )
+    }
+}
+
+/// Get the list of paths to all mapping definition files that should be
+/// included for the current target platform.
+fn get_def_paths() -> anyhow::Result<Vec<PathBuf>> {
+    let source_subdirs = [
+        "common",
+        #[cfg(target_family = "unix")]
+        "unix-family",
+        #[cfg(any(
+            target_os = "freebsd",
+            target_os = "netbsd",
+            target_os = "openbsd",
+            target_os = "macos"
+        ))]
+        "bsd-family",
+        #[cfg(target_os = "linux")]
+        "linux",
+        #[cfg(target_os = "macos")]
+        "macos",
+        #[cfg(target_os = "windows")]
+        "windows",
+    ];
+
+    let mut toml_paths = vec![];
+    for subdir in source_subdirs {
+        let wd = WalkDir::new(Path::new("src/syntax_mapping/builtins").join(subdir));
+        let paths = wd
+            .into_iter()
+            .filter_map_ok(|entry| {
+                let path = entry.path();
+                (path.is_file() && path.extension().map(|ext| ext == "toml").unwrap_or(false))
+                    .then(|| path.to_owned())
+            })
+            .collect::<Result<Vec<_>, _>>()?;
+        toml_paths.extend(paths);
+    }
+
+    toml_paths.sort_by_key(|path| {
+        path.file_name()
+            .expect("file name should not terminate in ..")
+            .to_owned()
+    });
+
+    Ok(toml_paths)
+}
+
+fn read_all_mappings() -> anyhow::Result<MappingList> {
+    let mut all_mappings = vec![];
+
+    for path in get_def_paths()? {
+        let toml_string = fs::read_to_string(path)?;
+        let mappings = toml::from_str::<MappingDefModel>(&toml_string)?.into_mapping_list();
+        all_mappings.extend(mappings.0);
+    }
+
+    let duplicates = all_mappings
+        .iter()
+        .duplicates_by(|(matcher, _)| matcher)
+        .collect_vec();
+    if !duplicates.is_empty() {
+        bail!("Rules with duplicate matchers found: {duplicates:?}");
+    }
+
+    Ok(MappingList(all_mappings))
+}
+
+/// Build the static syntax mappings defined in /src/syntax_mapping/builtins/
+/// into a .rs source file, which is to be inserted with `include!`.
+pub fn build_static_mappings() -> anyhow::Result<()> {
+    println!("cargo:rerun-if-changed=src/syntax_mapping/builtins/");
+
+    let mappings = read_all_mappings()?;
+
+    let codegen_path = Path::new(&env::var_os("OUT_DIR").ok_or(anyhow!("OUT_DIR is unset"))?)
+        .join("codegen_static_syntax_mappings.rs");
+
+    fs::write(codegen_path, mappings.codegen())?;
+
+    Ok(())
+}

build/util.rs

@@ -0,0 +1,21 @@
+#![allow(dead_code)]
+
+use std::{collections::HashMap, fs, path::Path};
+
+/// Generates a file from a template.
+pub fn render_template(
+    variables: &HashMap<&str, String>,
+    in_file: &str,
+    out_file: impl AsRef<Path>,
+) -> anyhow::Result<()> {
+    let mut content = fs::read_to_string(in_file)?;
+
+    for (variable_name, value) in variables {
+        // Replace {{variable_name}} by the value
+        let pattern = format!("{{{{{variable_name}}}}}");
+        content = content.replace(&pattern, value);
+    }
+
+    fs::write(out_file, content)?;
+    Ok(())
+}

doc/README-ja.md

@@ -181,7 +181,7 @@ man 2 select
 
 ## インストール
 
-[![Packaging status](https://repology.org/badge/vertical-allrepos/bat-cat.svg)](https://repology.org/project/bat-cat/versions)
+[![Packaging status](https://repology.org/badge/vertical-allrepos/bat-cat.svg?columns=3&exclude_unsupported=1)](https://repology.org/project/bat-cat/versions)
 
 ###  On Ubuntu (`apt` を使用)
 *... や他のDebianベースのLinuxディストリビューション*
@@ -219,7 +219,7 @@ apk add bat
 
 ###  On Arch Linux
 
-[Arch Linuxの公式リソース](https://www.archlinux.org/packages/community/x86_64/bat/)
+[Arch Linuxの公式リソース](https://www.archlinux.org/packages/extra/x86_64/bat/)
 からインストールできます。
 
 ```bash
@@ -366,7 +366,7 @@ ansible-galaxy install aeimer.install_bat
 ### From source
 
 
-`bat` をソースからビルドしたいならば、Rust 1.64.0 以上の環境が必要です。
+`bat` をソースからビルドしたいならば、Rust 1.70.0 以上の環境が必要です。
 `cargo` を使用してビルドすることができます:
 
 ```bash

doc/README-ko.md

@@ -214,7 +214,7 @@ man 2 select
 
 ## 설치
 
-[![Packaging status](https://repology.org/badge/vertical-allrepos/bat-cat.svg)](https://repology.org/project/bat-cat/versions)
+[![Packaging status](https://repology.org/badge/vertical-allrepos/bat-cat.svg?columns=3&exclude_unsupported=1)](https://repology.org/project/bat-cat/versions)
 
 ### Ubuntu에서 (`apt` 사용)
 *... 그리고 다른 Debian 기반의 Linux 배포판들에서.*
@@ -264,7 +264,7 @@ apk add bat
 ### Arch Linux에서
 
 공식 소스를 통해
-[`bat` 패키지](https://www.archlinux.org/packages/community/x86_64/bat/)를
+[`bat` 패키지](https://www.archlinux.org/packages/extra/x86_64/bat/)를
 설치할 수 있습니다:
 
 ```bash
@@ -416,7 +416,7 @@ scoop install bat
 
 ### 소스에서
 
-`bat`의 소스를 빌드하기 위해서는, Rust 1.64.0 이상이 필요합니다.
+`bat`의 소스를 빌드하기 위해서는, Rust 1.70.0 이상이 필요합니다.
 `cargo`를 이용해 전부 빌드할 수 있습니다:
 
 ```bash

doc/README-ru.md

@@ -12,7 +12,7 @@
   <a href="#установка">Установка</a> •
   <a href="#кастомизация">Кастомизация</a> •
   <a href="#цели-и-альтернативы">Цели и альтернативы </a><br>
-  [<a href="../README.md">English]
+  [<a href="../README.md">English</a>]
   [<a href="README-zh.md">中文</a>]
   [<a href="README-ja.md">日本語</a>]
   [<a href="README-ko.md">한국어</a>]
@@ -130,8 +130,8 @@ git show v0.6.0:src/main.rs | bat -l rs
 
 #### `xclip`
 
-Нумерация стро и отображение изменений затрудняет копирование содержимого файлов в буфер обмена.
-Чтобы спроваиться с этим, используйте флаг `-p`/`--plain` или просто перенаправьте стандартный вывод в `xclip`:
+Нумерация строк и отображение изменений затрудняет копирование содержимого файлов в буфер обмена.
+Чтобы справиться с этим, используйте флаг `-p`/`--plain` или просто перенаправьте стандартный вывод в `xclip`:
 ```bash
 bat main.cpp | xclip
 ```
@@ -160,7 +160,7 @@ man 2 select
 
 ## Установка
 
-[![Packaging status](https://repology.org/badge/vertical-allrepos/bat-cat.svg)](https://repology.org/project/bat-cat/versions)
+[![Packaging status](https://repology.org/badge/vertical-allrepos/bat-cat.svg?columns=3&exclude_unsupported=1)](https://repology.org/project/bat-cat/versions)
 
 ### Ubuntu (с помощью `apt`)
 *... и другие дистрибутивы основанные на Debian.*
@@ -201,7 +201,7 @@ apk add bat
 
 ### Arch Linux
 
-Вы можете установить [`bat`](https://www.archlinux.org/packages/community/x86_64/bat/) из официального источника:
+Вы можете установить [`bat`](https://www.archlinux.org/packages/extra/x86_64/bat/) из официального источника:
 
 ```bash
 pacman -S bat
@@ -344,7 +344,7 @@ ansible-galaxy install aeimer.install_bat
 
 ### Из исходников
 
-Если вы желаете установить `bat` из исходников, вам понадобится Rust 1.64.0 или выше. После этого используйте `cargo`, чтобы все скомпилировать:
+Если вы желаете установить `bat` из исходников, вам понадобится Rust 1.70.0 или выше. После этого используйте `cargo`, чтобы все скомпилировать:
 
 ```bash
 cargo install --locked bat
@@ -487,7 +487,7 @@ bat --generate-config-file
 # Использовать синтаксис C++ для всех Arduino .ino файлов
 --map-syntax "*.ino:C++"
 
-# Использовать синтаксик Git Ignore для всех файлов .ignore
+# Использовать синтаксис Git Ignore для всех файлов .ignore
 --map-syntax ".ignore:Git Ignore"
 ```
 
@@ -535,7 +535,7 @@ bat() {
 `bat` поддерживает терминалы *с* и *без* поддержки truecolor. Однако подсветка синтаксиса не оптимизирована для терминалов с 8-битными цветами, и рекомендуется использовать терминалы с поддержкой 24-битных цветов (`terminator`, `konsole`, `iTerm2`, ...).
 Смотрите [эту статью](https://gist.github.com/XVilka/8346728) для полного списка терминалов.
 
-Удостовертесь, что переменная `COLORTERM` равна `truecolor` или
+Удостоверьтесь, что переменная `COLORTERM` равна `truecolor` или
 `24bit`. Иначе `bat` не сможет определить поддержку 24-битных цветов (и будет использовать 8-битные).
 
 ### Текст и номера строк плохо видны
@@ -550,7 +550,7 @@ bat() {
 ``` bash
 iconv -f ISO-8859-1 -t UTF-8 my-file.php | bat
 ```
-Внимание: вам может понадобится флаг `-l`/`--language`, если `bat` не сможет автоматически определить синтаксис.
+Внимание: вам может понадобиться флаг `-l`/`--language`, если `bat` не сможет автоматически определить синтаксис.
 
 ## Разработка
 
@@ -568,7 +568,7 @@ cargo test
 # Установка (релизная версия)
 cargo install --locked
 
-# Компилирование исполняего файла bat с другим синтаксисом и темами
+# Компилирование исполняемого файла bat с другим синтаксисом и темами
 bash assets/create.sh
 cargo install --locked --force
 ```
@@ -592,6 +592,6 @@ cargo install --locked --force
 ## Лицензия
 Copyright (c) 2018-2021 [Разработчики bat](https://github.com/sharkdp/bat).
 
-`bat` распостраняется под лицензями MIT License и Apache License 2.0 (на выбор пользователя).
+`bat` распространяется под лицензиями MIT License и Apache License 2.0 (на выбор пользователя).
 
 Смотрите [LICENSE-APACHE](LICENSE-APACHE) и [LICENSE-MIT](LICENSE-MIT) для более подробного ознакомления.

doc/README-zh.md

@@ -191,7 +191,7 @@ man 2 select
 
 ## 安装
 
-[![Packaging status](https://repology.org/badge/vertical-allrepos/bat-cat.svg)](https://repology.org/project/bat-cat/versions)
+[![Packaging status](https://repology.org/badge/vertical-allrepos/bat-cat.svg?columns=3&exclude_unsupported=1)](https://repology.org/project/bat-cat/versions)
 
 ### Ubuntu (使用 `apt`)
 
@@ -232,7 +232,7 @@ apk add bat
 
 ### Arch Linux
 
-你可以用下面下列命令从官方源中安装[`bat`包](https://www.archlinux.org/packages/community/x86_64/bat/)：
+你可以用下面下列命令从官方源中安装[`bat`包](https://www.archlinux.org/packages/extra/x86_64/bat/)：
 
 ```bash
 pacman -S bat
@@ -372,7 +372,7 @@ scoop install bat
 
 ### 从源码编译
 
-如果你想要自己构建`bat`，那么你需要安装有高于1.64.0版本的 Rust。
+如果你想要自己构建`bat`，那么你需要安装有高于1.70.0版本的 Rust。
 
 使用以下命令编译。
 
@@ -412,7 +412,7 @@ bat --list-themes | fzf --preview="bat --theme={} --color=always /path/to/file"
 
 ### 输出样式
 
-你可以用`--style`参数来控制`bat`输出的样式。使用`--style=numbers,chanegs`可以只开启 Git 修改和行号显示而不添加其他内容。`BAT_STYLE`环境变量具有相同功能。
+你可以用`--style`参数来控制`bat`输出的样式。使用`--style=numbers,changes`可以只开启 Git 修改和行号显示而不添加其他内容。`BAT_STYLE`环境变量具有相同功能。
 
 ### 添加新的语言和语法
 

doc/assets.md

@@ -16,6 +16,9 @@ in the `.sublime-syntax` format.
 2. If the Sublime Text syntax is only available as a `.tmLanguage` file, open the file in
    Sublime Text and convert it to a `.sublime-syntax` file via *Tools* -> *Developer* ->
    *New Syntax from XXX.tmLanguage...*. Save the new file in the `assets/syntaxes` folder.
+   If only `.tmLanguage.json` or `.tmLanguage.yml` file is available, use
+   [PackageDev](https://packagecontrol.io/packages/PackageDev) to convert it to `.tmLanguage.plist`
+   format and then rename the converted file to `.tmLanguage` file.
 
 3. Run the `assets/create.sh` script. It calls `bat cache --build` to parse all available
    `.sublime-syntax` files and serialize them to a `syntaxes.bin` file.
@@ -84,6 +87,7 @@ The following files have been manually modified after converting from a `.tmLang
 * `Org mode.sublime-syntax` => removed `task` file type.
 * `Robot.sublime_syntax` => changed name to "Robot Framework", added `.resource` extension.
 * `SML.sublime_syntax` => removed `ml` file type.
+* `wgsl.sublime-syntax` => added `wgsl` file extension.
 
 ### Non-submodule additions
 

doc/long-help.txt

@@ -89,8 +89,8 @@ Options:
 
       --paging <when>
           Specify when to use the pager. To disable the pager, use --paging=never' or its
-          alias,'-P'. To disable the pager permanently, set BAT_PAGER to an empty string. To control
-          which pager is used, see the '--pager' option. Possible values: *auto*, never, always.
+          alias,'-P'. To disable the pager permanently, set BAT_PAGING to 'never'. To control which
+          pager is used, see the '--pager' option. Possible values: *auto*, never, always.
 
       --pager <command>
           Determine which pager is used. This option will override the PAGER and BAT_PAGER
@@ -116,6 +116,12 @@ Options:
       --list-themes
           Display a list of supported themes for syntax highlighting.
 
+  -s, --squeeze-blank
+          Squeeze consecutive empty lines into a single empty line.
+
+      --squeeze-limit <squeeze-limit>
+          Set the maximum number of consecutive empty lines to be printed.
+
       --style <components>
           Configure which elements (line numbers, file headers, grid borders, Git modifications, ..)
           to display in addition to the file contents. The argument is a comma-separated list of
@@ -123,6 +129,9 @@ Options:
           set a default style, add the '--style=".."' option to the configuration file or export the
           BAT_STYLE environment variable (e.g.: export BAT_STYLE="..").
           
+          By default, the following components are enabled:
+            changes, grid, header-filename, numbers, snip
+          
           Possible values:
           
             * default: enables recommended style components (default).
@@ -160,6 +169,9 @@ Options:
       --acknowledgements
           Show acknowledgements.
 
+      --set-terminal-title
+          Sets terminal title to filenames when using a pager.
+
   -h, --help
           Print help (see a summary with '-h')
 

doc/release-checklist.md

@@ -9,7 +9,16 @@
 - [ ] Update the version and the min. supported Rust version in `README.md` and
       `doc/README-*.md`. Check with
       `git grep -i -e 'rust.*1\.' -e '1\..*rust' | grep README | grep -v tests/`.
-- [ ] Update `CHANGELOG.md`. Introduce a section for the new release.
+
+## CHANGELOG.md updates
+
+- [ ] Go to https://github.com/sharkdp/bat/releases/new, click "Choose a tag",
+  type the name of the tag that will be created later, click "Generate release
+  notes". DO NOT ACTUALLY CREATE ANY RELEASE IN THIS STEP.
+- [ ] Compare current `CHANGELOG.md` with auto-generated release notes and add
+  missing entries. Expect in particular dependabot PRs to not be in
+  `CHANGELOG.md` since they are [auto-merged] if CI passes.
+- [ ] Introduce a section for the new release and perform final touch-ups.
 
 ## Update syntaxes and themes (build assets)
 
@@ -71,3 +80,5 @@
 
 
 ```
+
+[auto-merged]: https://github.com/sharkdp/bat/blob/master/.github/workflows/Auto-merge-dependabot-PRs.yml

doc/short-help.txt

@@ -43,6 +43,8 @@ Options:
           Set the color theme for syntax highlighting.
       --list-themes
           Display all supported highlighting themes.
+  -s, --squeeze-blank
+          Squeeze consecutive empty lines.
       --style <components>
           Comma-separated list of style elements to display (*default*, auto, full, plain, changes,
           header, header-filename, header-filesize, grid, rule, numbers, snip).

examples/buffer.rs

@@ -0,0 +1,17 @@
+use bat::{assets::HighlightingAssets, config::Config, controller::Controller, Input};
+
+fn main() {
+    let mut buffer = String::new();
+    let config = Config {
+        colored_output: true,
+        ..Default::default()
+    };
+    let assets = HighlightingAssets::from_binary();
+    let controller = Controller::new(&config, &assets);
+    let input = Input::from_file(file!());
+    controller
+        .run(vec![input.into()], Some(&mut buffer))
+        .unwrap();
+
+    println!("{buffer}");
+}

examples/list_syntaxes_and_themes.rs

@@ -13,6 +13,6 @@ fn main() {
 
     println!("Themes:");
     for theme in printer.themes() {
-        println!("- {}", theme);
+        println!("- {theme}");
     }
 }

examples/yaml.rs

@@ -23,7 +23,8 @@ fn main() {
         }],
     };
 
-    let bytes = serde_yaml::to_vec(&person).unwrap();
+    let mut bytes = Vec::with_capacity(128);
+    serde_yaml::to_writer(&mut bytes, &person).unwrap();
     PrettyPrinter::new()
         .language("yaml")
         .line_numbers(true)

src/assets.rs

@@ -380,7 +380,7 @@ fn asset_from_contents<T: serde::de::DeserializeOwned>(
     } else {
         bincode::deserialize_from(contents)
     }
-    .map_err(|_| format!("Could not parse {}", description).into())
+    .map_err(|_| format!("Could not parse {description}").into())
 }
 
 fn asset_from_cache<T: serde::de::DeserializeOwned>(
@@ -396,7 +396,7 @@ fn asset_from_cache<T: serde::de::DeserializeOwned>(
         )
     })?;
     asset_from_contents(&contents[..], description, compressed)
-        .map_err(|_| format!("Could not parse cached {}", description).into())
+        .map_err(|_| format!("Could not parse cached {description}").into())
 }
 
 #[cfg(target_os = "macos")]
@@ -404,7 +404,7 @@ fn macos_dark_mode_active() -> bool {
     const PREFERENCES_FILE: &str = "Library/Preferences/.GlobalPreferences.plist";
     const STYLE_KEY: &str = "AppleInterfaceStyle";
 
-    let preferences_file = dirs::home_dir()
+    let preferences_file = home::home_dir()
         .map(|home| home.join(PREFERENCES_FILE))
         .expect("Could not get home directory");
 
@@ -441,7 +441,7 @@ mod tests {
         fn new() -> Self {
             SyntaxDetectionTest {
                 assets: HighlightingAssets::from_binary(),
-                syntax_mapping: SyntaxMapping::builtin(),
+                syntax_mapping: SyntaxMapping::new(),
                 temp_dir: TempDir::new().expect("creation of temporary directory"),
             }
         }
@@ -466,7 +466,7 @@ mod tests {
             let file_path = self.temp_dir.path().join(file_name);
             {
                 let mut temp_file = File::create(&file_path).unwrap();
-                writeln!(temp_file, "{}", first_line).unwrap();
+                writeln!(temp_file, "{first_line}").unwrap();
             }
 
             let input = Input::ordinary_file(&file_path);
@@ -514,8 +514,7 @@ mod tests {
 
             if !consistent {
                 eprintln!(
-                    "Inconsistent syntax detection:\nFor File: {}\nFor Reader: {}",
-                    as_file, as_reader
+                    "Inconsistent syntax detection:\nFor File: {as_file}\nFor Reader: {as_reader}"
                 )
             }
 

src/assets/assets_metadata.rs

@@ -3,7 +3,7 @@ use std::path::Path;
 use std::time::SystemTime;
 
 use semver::Version;
-use serde::{Deserialize, Serialize};
+use serde_derive::{Deserialize, Serialize};
 
 use crate::error::*;
 

src/assets/build_assets.rs

@@ -93,7 +93,7 @@ fn print_unlinked_contexts(syntax_set: &SyntaxSet) {
     if !missing_contexts.is_empty() {
         println!("Some referenced contexts could not be found!");
         for context in missing_contexts {
-            println!("- {}", context);
+            println!("- {context}");
         }
     }
 }
@@ -152,7 +152,7 @@ pub(crate) fn asset_to_contents<T: serde::Serialize>(
     } else {
         bincode::serialize_into(&mut contents, asset)
     }
-    .map_err(|_| format!("Could not serialize {}", description))?;
+    .map_err(|_| format!("Could not serialize {description}"))?;
     Ok(contents)
 }
 

src/assets/build_assets/acknowledgements.rs

@@ -80,7 +80,7 @@ fn handle_license(path: &Path) -> Result<Option<String>> {
     } else if license_not_needed_in_acknowledgements(&license_text) {
         Ok(None)
     } else {
-        Err(format!("ERROR: License is of unknown type: {:?}", path).into())
+        Err(format!("ERROR: License is of unknown type: {path:?}").into())
     }
 }
 
@@ -125,7 +125,7 @@ fn append_to_acknowledgements(
     relative_path: &str,
     license_text: &str,
 ) {
-    write!(acknowledgements, "## {}\n\n{}", relative_path, license_text).ok();
+    write!(acknowledgements, "## {relative_path}\n\n{license_text}").ok();
 
     // Make sure the last char is a newline to not mess up formatting later
     if acknowledgements

src/assets/lazy_theme_set.rs

@@ -3,8 +3,7 @@ use super::*;
 use std::collections::BTreeMap;
 use std::convert::TryFrom;
 
-use serde::Deserialize;
-use serde::Serialize;
+use serde_derive::{Deserialize, Serialize};
 
 use once_cell::unsync::OnceCell;
 
@@ -89,7 +88,7 @@ impl TryFrom<ThemeSet> for LazyThemeSet {
             let lazy_theme = LazyTheme {
                 serialized: crate::assets::build_assets::asset_to_contents(
                     &theme,
-                    &format!("theme {}", name),
+                    &format!("theme {name}"),
                     COMPRESS_LAZY_THEMES,
                 )?,
                 deserialized: OnceCell::new(),

src/bin/bat/app.rs

@@ -1,9 +1,8 @@
 use std::collections::HashSet;
 use std::env;
+use std::io::IsTerminal;
 use std::path::{Path, PathBuf};
 
-use atty::{self, Stream};
-
 use crate::{
     clap_app,
     config::{get_args_from_config_file, get_args_from_env_opts_var, get_args_from_env_vars},
@@ -30,6 +29,10 @@ fn is_truecolor_terminal() -> bool {
         .unwrap_or(false)
 }
 
+pub fn env_no_color() -> bool {
+    env::var_os("NO_COLOR").is_some_and(|x| !x.is_empty())
+}
+
 pub struct App {
     pub matches: ArgMatches,
     interactive_output: bool,
@@ -40,7 +43,7 @@ impl App {
         #[cfg(windows)]
         let _ = nu_ansi_term::enable_ansi_support();
 
-        let interactive_output = atty::is(Stream::Stdout);
+        let interactive_output = std::io::stdout().is_terminal();
 
         Ok(App {
             matches: Self::matches(interactive_output)?,
@@ -104,7 +107,7 @@ impl App {
                     // If we are reading from stdin, only enable paging if we write to an
                     // interactive terminal and if we do not *read* from an interactive
                     // terminal.
-                    if self.interactive_output && !atty::is(Stream::Stdin) {
+                    if self.interactive_output && !std::io::stdin().is_terminal() {
                         PagingMode::QuitIfOneScreen
                     } else {
                         PagingMode::Never
@@ -118,7 +121,11 @@ impl App {
             _ => unreachable!("other values for --paging are not allowed"),
         };
 
-        let mut syntax_mapping = SyntaxMapping::builtin();
+        let mut syntax_mapping = SyntaxMapping::new();
+        // start building glob matchers for builtin mappings immediately
+        // this is an appropriate approach because it's statistically likely that
+        // all the custom mappings need to be checked
+        syntax_mapping.start_offload_build_all();
 
         if let Some(values) = self.matches.get_many::<String>("ignored-suffix") {
             for suffix in values {
@@ -127,7 +134,9 @@ impl App {
         }
 
         if let Some(values) = self.matches.get_many::<String>("map-syntax") {
-            for from_to in values {
+            // later args take precedence over earlier ones, hence `.rev()`
+            // see: https://github.com/sharkdp/bat/pull/2755#discussion_r1456416875
+            for from_to in values.rev() {
                 let parts: Vec<_> = from_to.split(':').collect();
 
                 if parts.len() != 2 {
@@ -208,7 +217,7 @@ impl App {
                 || match self.matches.get_one::<String>("color").map(|s| s.as_str()) {
                     Some("always") => true,
                     Some("never") => false,
-                    Some("auto") => env::var_os("NO_COLOR").is_none() && self.interactive_output,
+                    Some("auto") => !env_no_color() && self.interactive_output,
                     _ => unreachable!("other values for --color are not allowed"),
                 },
             paging_mode,
@@ -282,6 +291,19 @@ impl App {
                 .map(HighlightedLineRanges)
                 .unwrap_or_default(),
             use_custom_assets: !self.matches.get_flag("no-custom-assets"),
+            #[cfg(feature = "lessopen")]
+            use_lessopen: self.matches.get_flag("lessopen"),
+            set_terminal_title: self.matches.get_flag("set-terminal-title"),
+            squeeze_lines: if self.matches.get_flag("squeeze-blank") {
+                Some(
+                    self.matches
+                        .get_one::<usize>("squeeze-limit")
+                        .map(|limit| limit.to_owned())
+                        .unwrap_or(1),
+                )
+            } else {
+                None
+            },
         })
     }
 

src/bin/bat/assets.rs

@@ -44,7 +44,7 @@ pub fn assets_from_cache_or_binary(
 }
 
 fn clear_asset(path: PathBuf, description: &str) {
-    print!("Clearing {} ... ", description);
+    print!("Clearing {description} ... ");
     match fs::remove_file(&path) {
         Err(err) if err.kind() == io::ErrorKind::NotFound => {
             println!("skipped (not present)");

src/bin/bat/clap_app.rs

@@ -19,7 +19,7 @@ static VERSION: Lazy<String> = Lazy::new(|| {
 });
 
 pub fn build_app(interactive_output: bool) -> Command {
-    let color_when = if interactive_output && env::var_os("NO_COLOR").is_none() {
+    let color_when = if interactive_output && !crate::app::env_no_color() {
         ColorChoice::Auto
     } else {
         ColorChoice::Never
@@ -79,6 +79,7 @@ pub fn build_app(interactive_output: bool) -> Command {
             Arg::new("plain")
                 .overrides_with("plain")
                 .overrides_with("number")
+                .overrides_with("paging")
                 .short('p')
                 .long("plain")
                 .action(ArgAction::Count)
@@ -303,6 +304,7 @@ pub fn build_app(interactive_output: bool) -> Command {
                 .long("paging")
                 .overrides_with("paging")
                 .overrides_with("no-paging")
+                .overrides_with("plain")
                 .value_name("when")
                 .value_parser(["auto", "never", "always"])
                 .default_value("auto")
@@ -311,7 +313,7 @@ pub fn build_app(interactive_output: bool) -> Command {
                 .long_help(
                     "Specify when to use the pager. To disable the pager, use \
                     --paging=never' or its alias,'-P'. To disable the pager permanently, \
-                    set BAT_PAGER to an empty string. To control which pager is used, see the \
+                    set BAT_PAGING to 'never'. To control which pager is used, see the \
                     '--pager' option. Possible values: *auto*, never, always."
                 ),
         )
@@ -385,6 +387,21 @@ pub fn build_app(interactive_output: bool) -> Command {
                 .help("Display all supported highlighting themes.")
                 .long_help("Display a list of supported themes for syntax highlighting."),
         )
+        .arg(
+            Arg::new("squeeze-blank")
+                .long("squeeze-blank")
+                .short('s')
+                .action(ArgAction::SetTrue)
+                .help("Squeeze consecutive empty lines.")
+                .long_help("Squeeze consecutive empty lines into a single empty line.")
+        )
+        .arg(
+            Arg::new("squeeze-limit")
+                .long("squeeze-limit")
+                .value_parser(|s: &str| s.parse::<usize>().map_err(|_| "Requires a non-negative number".to_owned()))
+                .long_help("Set the maximum number of consecutive empty lines to be printed.")
+                .hide_short_help(true)
+        )
         .arg(
             Arg::new("style")
                 .long("style")
@@ -413,7 +430,7 @@ pub fn build_app(interactive_output: bool) -> Command {
                     });
 
                     if let Some(invalid) = invalid_vals.next() {
-                        Err(format!("Unknown style, '{}'", invalid))
+                        Err(format!("Unknown style, '{invalid}'"))
                     } else {
                         Ok(val.to_owned())
                     }
@@ -430,6 +447,8 @@ pub fn build_app(interactive_output: bool) -> Command {
                      pre-defined style ('full'). To set a default style, add the \
                      '--style=\"..\"' option to the configuration file or export the \
                      BAT_STYLE environment variable (e.g.: export BAT_STYLE=\"..\").\n\n\
+                     By default, the following components are enabled:\n  \
+                        changes, grid, header-filename, numbers, snip\n\n\
                      Possible values:\n\n  \
                      * default: enables recommended style components (default).\n  \
                      * full: enables all available components.\n  \
@@ -497,7 +516,28 @@ pub fn build_app(interactive_output: bool) -> Command {
                 .action(ArgAction::SetTrue)
                 .hide(true)
                 .help("Do not load custom assets"),
-        )
+        );
+
+    #[cfg(feature = "lessopen")]
+    {
+        app = app
+            .arg(
+                Arg::new("lessopen")
+                    .long("lessopen")
+                    .action(ArgAction::SetTrue)
+                    .help("Enable the $LESSOPEN preprocessor"),
+            )
+            .arg(
+                Arg::new("no-lessopen")
+                    .long("no-lessopen")
+                    .action(ArgAction::SetTrue)
+                    .overrides_with("lessopen")
+                    .hide(true)
+                    .help("Disable the $LESSOPEN preprocessor if enabled (overrides --lessopen)"),
+            )
+    }
+
+    app = app
         .arg(
             Arg::new("config-file")
                 .long("config-file")
@@ -536,7 +576,7 @@ pub fn build_app(interactive_output: bool) -> Command {
                 .alias("diagnostics")
                 .action(ArgAction::SetTrue)
                 .hide_short_help(true)
-                .help("Show diagnostic information for bug reports.")
+                .help("Show diagnostic information for bug reports."),
         )
         .arg(
             Arg::new("acknowledgements")
@@ -544,6 +584,13 @@ pub fn build_app(interactive_output: bool) -> Command {
                 .action(ArgAction::SetTrue)
                 .hide_short_help(true)
                 .help("Show acknowledgements."),
+        )
+        .arg(
+            Arg::new("set-terminal-title")
+                .long("set-terminal-title")
+                .action(ArgAction::SetTrue)
+                .hide_short_help(true)
+                .help("Sets terminal title to filenames when using a pager."),
         );
 
     // Check if the current directory contains a file name cache. Otherwise,

src/bin/bat/config.rs

@@ -142,6 +142,7 @@ pub fn get_args_from_env_vars() -> Vec<OsString> {
         ("--tabs", "BAT_TABS"),
         ("--theme", "BAT_THEME"),
         ("--pager", "BAT_PAGER"),
+        ("--paging", "BAT_PAGING"),
         ("--style", "BAT_STYLE"),
     ]
     .iter()

src/bin/bat/directories.rs

@@ -1,12 +1,11 @@
 use std::env;
 use std::path::{Path, PathBuf};
 
+use etcetera::BaseStrategy;
 use once_cell::sync::Lazy;
 
-/// Wrapper for 'dirs' that treats MacOS more like Linux, by following the XDG specification.
-/// The `XDG_CACHE_HOME` environment variable is checked first. `BAT_CONFIG_DIR`
-///  is then checked before the `XDG_CONFIG_HOME` environment variable.
-///  The fallback directories are `~/.cache/bat` and `~/.config/bat`, respectively.
+/// Wrapper for 'etcetera' that checks BAT_CACHE_PATH and BAT_CONFIG_DIR and falls back to the
+/// Windows known folder locations on Windows & the XDG Base Directory Specification everywhere else.
 pub struct BatProjectDirs {
     cache_dir: PathBuf,
     config_dir: PathBuf,
@@ -14,25 +13,24 @@ pub struct BatProjectDirs {
 
 impl BatProjectDirs {
     fn new() -> Option<BatProjectDirs> {
-        let cache_dir = BatProjectDirs::get_cache_dir()?;
+        let basedirs = etcetera::choose_base_strategy().ok()?;
 
-        // Checks whether or not $BAT_CONFIG_DIR exists. If it doesn't, set our config dir
-        // to our system's default configuration home.
-        let config_dir =
-            if let Some(config_dir_op) = env::var_os("BAT_CONFIG_DIR").map(PathBuf::from) {
-                config_dir_op
-            } else {
-                #[cfg(target_os = "macos")]
-                let config_dir_op = env::var_os("XDG_CONFIG_HOME")
-                    .map(PathBuf::from)
-                    .filter(|p| p.is_absolute())
-                    .or_else(|| dirs::home_dir().map(|d| d.join(".config")));
+        // Checks whether or not `$BAT_CACHE_PATH` exists. If it doesn't, set the cache dir to our
+        // system's default cache home.
+        let cache_dir = if let Some(cache_dir) = env::var_os("BAT_CACHE_PATH").map(PathBuf::from) {
+            cache_dir
+        } else {
+            basedirs.cache_dir().join("bat")
+        };
 
-                #[cfg(not(target_os = "macos"))]
-                let config_dir_op = dirs::config_dir();
-
-                config_dir_op.map(|d| d.join("bat"))?
-            };
+        // Checks whether or not `$BAT_CONFIG_DIR` exists. If it doesn't, set the config dir to our
+        // system's default configuration home.
+        let config_dir = if let Some(config_dir) = env::var_os("BAT_CONFIG_DIR").map(PathBuf::from)
+        {
+            config_dir
+        } else {
+            basedirs.config_dir().join("bat")
+        };
 
         Some(BatProjectDirs {
             cache_dir,
@@ -40,25 +38,6 @@ impl BatProjectDirs {
         })
     }
 
-    fn get_cache_dir() -> Option<PathBuf> {
-        // on all OS prefer BAT_CACHE_PATH if set
-        let cache_dir_op = env::var_os("BAT_CACHE_PATH").map(PathBuf::from);
-        if cache_dir_op.is_some() {
-            return cache_dir_op;
-        }
-
-        #[cfg(target_os = "macos")]
-        let cache_dir_op = env::var_os("XDG_CACHE_HOME")
-            .map(PathBuf::from)
-            .filter(|p| p.is_absolute())
-            .or_else(|| dirs::home_dir().map(|d| d.join(".cache")));
-
-        #[cfg(not(target_os = "macos"))]
-        let cache_dir_op = dirs::cache_dir();
-
-        cache_dir_op.map(|d| d.join("bat"))
-    }
-
     pub fn cache_dir(&self) -> &Path {
         &self.cache_dir
     }

src/bin/bat/main.rs

@@ -30,6 +30,7 @@ use directories::PROJECT_DIRS;
 use globset::GlobMatcher;
 
 use bat::{
+    assets::HighlightingAssets,
     config::Config,
     controller::Controller,
     error::*,
@@ -78,9 +79,11 @@ fn run_cache_subcommand(
     Ok(())
 }
 
-fn get_syntax_mapping_to_paths<'a>(
-    mappings: &[(GlobMatcher, MappingTarget<'a>)],
-) -> HashMap<&'a str, Vec<String>> {
+fn get_syntax_mapping_to_paths<'r, 't, I>(mappings: I) -> HashMap<&'t str, Vec<String>>
+where
+    I: IntoIterator<Item = (&'r GlobMatcher, &'r MappingTarget<'t>)>,
+    't: 'r, // target text outlives rule
+{
     let mut map = HashMap::new();
     for mapping in mappings {
         if let (matcher, MappingTarget::MapTo(s)) = mapping {
@@ -123,7 +126,7 @@ pub fn get_languages(config: &Config, cache_dir: &Path) -> Result<String> {
 
     languages.sort_by_key(|lang| lang.name.to_uppercase());
 
-    let configured_languages = get_syntax_mapping_to_paths(config.syntax_mapping.mappings());
+    let configured_languages = get_syntax_mapping_to_paths(config.syntax_mapping.all_mappings());
 
     for lang in &mut languages {
         if let Some(additional_paths) = configured_languages.get(lang.name.as_str()) {
@@ -197,19 +200,31 @@ pub fn list_themes(cfg: &Config, config_dir: &Path, cache_dir: &Path) -> Result<
     let stdout = io::stdout();
     let mut stdout = stdout.lock();
 
-    if config.colored_output {
-        for theme in assets.themes() {
+    let default_theme = HighlightingAssets::default_theme();
+    for theme in assets.themes() {
+        let default_theme_info = if default_theme == theme {
+            " (default)"
+        } else {
+            ""
+        };
+        if config.colored_output {
             writeln!(
                 stdout,
-                "Theme: {}\n",
-                Style::new().bold().paint(theme.to_string())
+                "Theme: {}{}\n",
+                Style::new().bold().paint(theme.to_string()),
+                default_theme_info
             )?;
             config.theme = theme.to_string();
             Controller::new(&config, &assets)
-                .run(vec![theme_preview_file()])
+                .run(vec![theme_preview_file()], None)
                 .ok();
             writeln!(stdout)?;
+        } else {
+            writeln!(stdout, "{theme}{default_theme_info}")?;
         }
+    }
+
+    if config.colored_output {
         writeln!(
             stdout,
             "Further themes can be installed to '{}', \
@@ -218,19 +233,36 @@ pub fn list_themes(cfg: &Config, config_dir: &Path, cache_dir: &Path) -> Result<
             https://github.com/sharkdp/bat#adding-new-themes",
             config_dir.join("themes").to_string_lossy()
         )?;
-    } else {
-        for theme in assets.themes() {
-            writeln!(stdout, "{}", theme)?;
-        }
     }
 
     Ok(())
 }
 
+fn set_terminal_title_to(new_terminal_title: String) {
+    let osc_command_for_setting_terminal_title = "\x1b]0;";
+    let osc_end_command = "\x07";
+    print!("{osc_command_for_setting_terminal_title}{new_terminal_title}{osc_end_command}");
+    io::stdout().flush().unwrap();
+}
+
+fn get_new_terminal_title(inputs: &Vec<Input>) -> String {
+    let mut new_terminal_title = "bat: ".to_string();
+    for (index, input) in inputs.iter().enumerate() {
+        new_terminal_title += input.description().title();
+        if index < inputs.len() - 1 {
+            new_terminal_title += ", ";
+        }
+    }
+    new_terminal_title
+}
+
 fn run_controller(inputs: Vec<Input>, config: &Config, cache_dir: &Path) -> Result<bool> {
     let assets = assets_from_cache_or_binary(config.use_custom_assets, cache_dir)?;
     let controller = Controller::new(config, &assets);
-    controller.run(inputs)
+    if config.paging_mode != PagingMode::Never && config.set_terminal_title {
+        set_terminal_title_to(get_new_terminal_title(&inputs));
+    }
+    controller.run(inputs, None)
 }
 
 #[cfg(feature = "bugreport")]
@@ -249,23 +281,25 @@ fn invoke_bugreport(app: &App, cache_dir: &Path) {
         .info(OperatingSystem::default())
         .info(CommandLine::default())
         .info(EnvironmentVariables::list(&[
-            "SHELL",
-            "PAGER",
-            "LESS",
-            "LANG",
-            "LC_ALL",
-            "BAT_PAGER",
             "BAT_CACHE_PATH",
             "BAT_CONFIG_PATH",
             "BAT_OPTS",
+            "BAT_PAGER",
+            "BAT_PAGING",
             "BAT_STYLE",
             "BAT_TABS",
             "BAT_THEME",
-            "XDG_CONFIG_HOME",
-            "XDG_CACHE_HOME",
             "COLORTERM",
-            "NO_COLOR",
+            "LANG",
+            "LC_ALL",
+            "LESS",
             "MANPAGER",
+            "NO_COLOR",
+            "PAGER",
+            "SHELL",
+            "TERM",
+            "XDG_CACHE_HOME",
+            "XDG_CONFIG_HOME",
         ]))
         .info(FileContent::new("System Config file", system_config_file()))
         .info(FileContent::new("Config file", config_file()))

src/config.rs

@@ -90,6 +90,16 @@ pub struct Config<'a> {
     /// Whether or not to allow custom assets. If this is false or if custom assets (a.k.a.
     /// cached assets) are not available, assets from the binary will be used instead.
     pub use_custom_assets: bool,
+
+    // Whether or not to use $LESSOPEN if set
+    #[cfg(feature = "lessopen")]
+    pub use_lessopen: bool,
+
+    // Weather or not to set terminal title when using a pager
+    pub set_terminal_title: bool,
+
+    /// The maximum number of consecutive empty lines to display
+    pub squeeze_lines: Option<usize>,
 }
 
 #[cfg(all(feature = "minimal-application", feature = "paging"))]

src/controller.rs

@@ -6,34 +6,48 @@ use crate::config::{Config, VisibleLines};
 use crate::diff::{get_git_diff, LineChanges};
 use crate::error::*;
 use crate::input::{Input, InputReader, OpenedInput};
+#[cfg(feature = "lessopen")]
+use crate::lessopen::LessOpenPreprocessor;
 #[cfg(feature = "git")]
 use crate::line_range::LineRange;
 use crate::line_range::{LineRanges, RangeCheckResult};
 use crate::output::OutputType;
 #[cfg(feature = "paging")]
 use crate::paging::PagingMode;
-use crate::printer::{InteractivePrinter, Printer, SimplePrinter};
+use crate::printer::{InteractivePrinter, OutputHandle, Printer, SimplePrinter};
 
 use clircle::{Clircle, Identifier};
 
 pub struct Controller<'a> {
     config: &'a Config<'a>,
     assets: &'a HighlightingAssets,
+    #[cfg(feature = "lessopen")]
+    preprocessor: Option<LessOpenPreprocessor>,
 }
 
 impl<'b> Controller<'b> {
     pub fn new<'a>(config: &'a Config, assets: &'a HighlightingAssets) -> Controller<'a> {
-        Controller { config, assets }
+        Controller {
+            config,
+            assets,
+            #[cfg(feature = "lessopen")]
+            preprocessor: LessOpenPreprocessor::new().ok(),
+        }
     }
 
-    pub fn run(&self, inputs: Vec<Input>) -> Result<bool> {
-        self.run_with_error_handler(inputs, default_error_handler)
+    pub fn run(
+        &self,
+        inputs: Vec<Input>,
+        output_buffer: Option<&mut dyn std::fmt::Write>,
+    ) -> Result<bool> {
+        self.run_with_error_handler(inputs, output_buffer, default_error_handler)
     }
 
     pub fn run_with_error_handler(
         &self,
         inputs: Vec<Input>,
-        handle_error: impl Fn(&Error, &mut dyn Write),
+        output_buffer: Option<&mut dyn std::fmt::Write>,
+        mut handle_error: impl FnMut(&Error, &mut dyn Write),
     ) -> Result<bool> {
         let mut output_type;
 
@@ -74,7 +88,10 @@ impl<'b> Controller<'b> {
             clircle::Identifier::stdout()
         };
 
-        let writer = output_type.handle()?;
+        let mut writer = match output_buffer {
+            Some(buf) => OutputHandle::FmtWrite(buf),
+            None => OutputHandle::IoWrite(output_type.handle()?),
+        };
         let mut no_errors: bool = true;
         let stderr = io::stderr();
 
@@ -82,16 +99,23 @@ impl<'b> Controller<'b> {
             let identifier = stdout_identifier.as_ref();
             let is_first = index == 0;
             let result = if input.is_stdin() {
-                self.print_input(input, writer, io::stdin().lock(), identifier, is_first)
+                self.print_input(input, &mut writer, io::stdin().lock(), identifier, is_first)
             } else {
                 // Use dummy stdin since stdin is actually not used (#1902)
-                self.print_input(input, writer, io::empty(), identifier, is_first)
+                self.print_input(input, &mut writer, io::empty(), identifier, is_first)
             };
             if let Err(error) = result {
-                if attached_to_pager {
-                    handle_error(&error, writer);
-                } else {
-                    handle_error(&error, &mut stderr.lock());
+                match writer {
+                    // It doesn't make much sense to send errors straight to stderr if the user
+                    // provided their own buffer, so we just return it.
+                    OutputHandle::FmtWrite(_) => return Err(error),
+                    OutputHandle::IoWrite(ref mut writer) => {
+                        if attached_to_pager {
+                            handle_error(&error, writer);
+                        } else {
+                            handle_error(&error, &mut stderr.lock());
+                        }
+                    }
                 }
                 no_errors = false;
             }
@@ -103,12 +127,23 @@ impl<'b> Controller<'b> {
     fn print_input<R: BufRead>(
         &self,
         input: Input,
-        writer: &mut dyn Write,
+        writer: &mut OutputHandle,
         stdin: R,
         stdout_identifier: Option<&Identifier>,
         is_first: bool,
     ) -> Result<()> {
-        let mut opened_input = input.open(stdin, stdout_identifier)?;
+        let mut opened_input = {
+            #[cfg(feature = "lessopen")]
+            match self.preprocessor {
+                Some(ref preprocessor) if self.config.use_lessopen => {
+                    preprocessor.open(input, stdin, stdout_identifier)?
+                }
+                _ => input.open(stdin, stdout_identifier)?,
+            }
+
+            #[cfg(not(feature = "lessopen"))]
+            input.open(stdin, stdout_identifier)?
+        };
         #[cfg(feature = "git")]
         let line_changes = if self.config.visible_lines.diff_mode()
             || (!self.config.loop_through && self.config.style_components.changes())
@@ -164,7 +199,7 @@ impl<'b> Controller<'b> {
     fn print_file(
         &self,
         printer: &mut dyn Printer,
-        writer: &mut dyn Write,
+        writer: &mut OutputHandle,
         input: &mut OpenedInput,
         add_header_padding: bool,
         #[cfg(feature = "git")] line_changes: &Option<LineChanges>,
@@ -202,7 +237,7 @@ impl<'b> Controller<'b> {
     fn print_file_ranges(
         &self,
         printer: &mut dyn Printer,
-        writer: &mut dyn Write,
+        writer: &mut OutputHandle,
         reader: &mut InputReader,
         line_ranges: &LineRanges,
     ) -> Result<()> {

src/decorations.rs

@@ -46,7 +46,7 @@ impl Decoration for LineNumberDecoration {
         _printer: &InteractivePrinter,
     ) -> DecorationText {
         if continuation {
-            if line_number > self.cached_wrap_invalid_at {
+            if line_number >= self.cached_wrap_invalid_at {
                 let new_width = self.cached_wrap.width + 1;
                 return DecorationText {
                     text: self.color.paint(" ".repeat(new_width)).to_string(),
@@ -56,7 +56,7 @@ impl Decoration for LineNumberDecoration {
 
             self.cached_wrap.clone()
         } else {
-            let plain: String = format!("{:4}", line_number);
+            let plain: String = format!("{line_number:4}");
             DecorationText {
                 width: plain.len(),
                 text: self.color.paint(plain).to_string(),

src/error.rs

@@ -7,6 +7,8 @@ pub enum Error {
     #[error(transparent)]
     Io(#[from] ::std::io::Error),
     #[error(transparent)]
+    Fmt(#[from] ::std::fmt::Error),
+    #[error(transparent)]
     SyntectError(#[from] ::syntect::Error),
     #[error(transparent)]
     SyntectLoadingError(#[from] ::syntect::LoadingError),
@@ -26,6 +28,12 @@ pub enum Error {
     InvalidPagerValueBat,
     #[error("{0}")]
     Msg(String),
+    #[cfg(feature = "lessopen")]
+    #[error(transparent)]
+    VarError(#[from] ::std::env::VarError),
+    #[cfg(feature = "lessopen")]
+    #[error(transparent)]
+    CommandParseError(#[from] ::shell_words::ParseError),
 }
 
 impl From<&'static str> for Error {

src/input.rs

@@ -197,7 +197,7 @@ impl<'a> Input<'a> {
             InputKind::StdIn => {
                 if let Some(stdout) = stdout_identifier {
                     let input_identifier = Identifier::try_from(clircle::Stdio::Stdin)
-                        .map_err(|e| format!("Stdin: Error identifying file: {}", e))?;
+                        .map_err(|e| format!("Stdin: Error identifying file: {e}"))?;
                     if stdout.surely_conflicts_with(&input_identifier) {
                         return Err("IO circle detected. The input from stdin is also an output. Aborting to avoid infinite loop.".into());
                     }
@@ -256,7 +256,7 @@ pub(crate) struct InputReader<'a> {
 }
 
 impl<'a> InputReader<'a> {
-    fn new<R: BufRead + 'a>(mut reader: R) -> InputReader<'a> {
+    pub(crate) fn new<R: BufRead + 'a>(mut reader: R) -> InputReader<'a> {
         let mut first_line = vec![];
         reader.read_until(b'\n', &mut first_line).ok();
 

src/lessopen.rs

@@ -0,0 +1,401 @@
+#![cfg(feature = "lessopen")]
+
+use std::convert::TryFrom;
+use std::env;
+use std::fs::File;
+use std::io::{BufRead, BufReader, Cursor, Read, Write};
+use std::path::PathBuf;
+use std::str;
+
+use clircle::{Clircle, Identifier};
+use os_str_bytes::RawOsString;
+use run_script::{IoOptions, ScriptOptions};
+
+use crate::error::Result;
+use crate::{
+    bat_warning,
+    input::{Input, InputKind, InputReader, OpenedInput, OpenedInputKind},
+};
+
+/// Preprocess files and/or stdin using $LESSOPEN and $LESSCLOSE
+pub(crate) struct LessOpenPreprocessor {
+    lessopen: String,
+    lessclose: Option<String>,
+    command_options: ScriptOptions,
+    kind: LessOpenKind,
+    /// Whether or not data piped via stdin is to be preprocessed
+    preprocess_stdin: bool,
+}
+
+enum LessOpenKind {
+    Piped,
+    PipedIgnoreExitCode,
+    TempFile,
+}
+
+impl LessOpenPreprocessor {
+    /// Create a new instance of LessOpenPreprocessor
+    /// Will return Ok if and only if $LESSOPEN is set and contains exactly one %s
+    pub(crate) fn new() -> Result<LessOpenPreprocessor> {
+        let lessopen = env::var("LESSOPEN")?;
+
+        // Ignore $LESSOPEN if it does not contains exactly one %s
+        // Note that $LESSCLOSE has no such requirement
+        if lessopen.match_indices("%s").count() != 1 {
+            let error_msg = "LESSOPEN ignored: must contain exactly one %s";
+            bat_warning!("{}", error_msg);
+            return Err(error_msg.into());
+        }
+
+        // "||" means pipe directly to bat without making a temporary file
+        // Also, if preprocessor output is empty and exit code is zero, use the empty output
+        // Otherwise, if output is empty and exit code is nonzero, use original file contents
+        let (kind, lessopen) = if lessopen.starts_with("||") {
+            (LessOpenKind::Piped, lessopen.chars().skip(2).collect())
+        // "|" means pipe, but ignore exit code, always using preprocessor output
+        } else if lessopen.starts_with('|') {
+            (
+                LessOpenKind::PipedIgnoreExitCode,
+                lessopen.chars().skip(1).collect(),
+            )
+        // If neither appear, write output to a temporary file and read from that
+        } else {
+            (LessOpenKind::TempFile, lessopen)
+        };
+
+        // "-" means that stdin is preprocessed along with files and may appear alongside "|" and "||"
+        let (stdin, lessopen) = if lessopen.starts_with('-') {
+            (true, lessopen.chars().skip(1).collect())
+        } else {
+            (false, lessopen)
+        };
+
+        let mut command_options = ScriptOptions::new();
+        command_options.runner = env::var("SHELL").ok();
+        command_options.input_redirection = IoOptions::Pipe;
+
+        Ok(Self {
+            lessopen: lessopen.replacen("%s", "$1", 1),
+            lessclose: env::var("LESSCLOSE")
+                .ok()
+                .map(|str| str.replacen("%s", "$1", 1).replacen("%s", "$2", 1)),
+            command_options,
+            kind,
+            preprocess_stdin: stdin,
+        })
+    }
+
+    pub(crate) fn open<'a, R: BufRead + 'a>(
+        &self,
+        input: Input<'a>,
+        mut stdin: R,
+        stdout_identifier: Option<&Identifier>,
+    ) -> Result<OpenedInput<'a>> {
+        let (lessopen_stdout, path_str, kind) = match input.kind {
+            InputKind::OrdinaryFile(ref path) => {
+                let path_str = match path.to_str() {
+                    Some(str) => str,
+                    None => return input.open(stdin, stdout_identifier),
+                };
+
+                let (exit_code, lessopen_stdout, _) = match run_script::run(
+                    &self.lessopen,
+                    &vec![path_str.to_string()],
+                    &self.command_options,
+                ) {
+                    Ok(output) => output,
+                    Err(_) => return input.open(stdin, stdout_identifier),
+                };
+
+                if self.fall_back_to_original_file(&lessopen_stdout, exit_code) {
+                    return input.open(stdin, stdout_identifier);
+                }
+
+                (
+                    RawOsString::from_string(lessopen_stdout),
+                    path_str.to_string(),
+                    OpenedInputKind::OrdinaryFile(path.to_path_buf()),
+                )
+            }
+            InputKind::StdIn => {
+                if self.preprocess_stdin {
+                    if let Some(stdout) = stdout_identifier {
+                        let input_identifier = Identifier::try_from(clircle::Stdio::Stdin)
+                            .map_err(|e| format!("Stdin: Error identifying file: {}", e))?;
+                        if stdout.surely_conflicts_with(&input_identifier) {
+                            return Err("IO circle detected. The input from stdin is also an output. Aborting to avoid infinite loop.".into());
+                        }
+                    }
+
+                    // stdin isn't Clone, so copy it to a cloneable buffer
+                    let mut stdin_buffer = Vec::new();
+                    stdin.read_to_end(&mut stdin_buffer).unwrap();
+
+                    let mut lessopen_handle = match run_script::spawn(
+                        &self.lessopen,
+                        &vec!["-".to_string()],
+                        &self.command_options,
+                    ) {
+                        Ok(handle) => handle,
+                        Err(_) => {
+                            return input.open(stdin, stdout_identifier);
+                        }
+                    };
+
+                    if lessopen_handle
+                        .stdin
+                        .as_mut()
+                        .unwrap()
+                        .write_all(&stdin_buffer.clone())
+                        .is_err()
+                    {
+                        return input.open(stdin, stdout_identifier);
+                    }
+
+                    let lessopen_output = match lessopen_handle.wait_with_output() {
+                        Ok(output) => output,
+                        Err(_) => {
+                            return input.open(Cursor::new(stdin_buffer), stdout_identifier);
+                        }
+                    };
+
+                    if lessopen_output.stdout.is_empty()
+                        && (!lessopen_output.status.success()
+                            || matches!(self.kind, LessOpenKind::PipedIgnoreExitCode))
+                    {
+                        return input.open(Cursor::new(stdin_buffer), stdout_identifier);
+                    }
+
+                    (
+                        RawOsString::assert_from_raw_vec(lessopen_output.stdout),
+                        "-".to_string(),
+                        OpenedInputKind::StdIn,
+                    )
+                } else {
+                    return input.open(stdin, stdout_identifier);
+                }
+            }
+            InputKind::CustomReader(_) => {
+                return input.open(stdin, stdout_identifier);
+            }
+        };
+
+        Ok(OpenedInput {
+            kind,
+            reader: InputReader::new(BufReader::new(
+                if matches!(self.kind, LessOpenKind::TempFile) {
+                    // Remove newline at end of temporary file path returned by $LESSOPEN
+                    let stdout = match lessopen_stdout.strip_suffix("\n") {
+                        Some(stripped) => stripped.to_owned(),
+                        None => lessopen_stdout,
+                    };
+
+                    let stdout = stdout.into_os_string();
+
+                    let file = match File::open(PathBuf::from(&stdout)) {
+                        Ok(file) => file,
+                        Err(_) => {
+                            return input.open(stdin, stdout_identifier);
+                        }
+                    };
+
+                    Preprocessed {
+                        kind: PreprocessedKind::TempFile(file),
+                        lessclose: self.lessclose.clone(),
+                        command_args: vec![path_str, stdout.to_str().unwrap().to_string()],
+                        command_options: self.command_options.clone(),
+                    }
+                } else {
+                    Preprocessed {
+                        kind: PreprocessedKind::Piped(Cursor::new(lessopen_stdout.into_raw_vec())),
+                        lessclose: self.lessclose.clone(),
+                        command_args: vec![path_str, "-".to_string()],
+                        command_options: self.command_options.clone(),
+                    }
+                },
+            )),
+            metadata: input.metadata,
+            description: input.description,
+        })
+    }
+
+    fn fall_back_to_original_file(&self, lessopen_output: &str, exit_code: i32) -> bool {
+        lessopen_output.is_empty()
+            && (exit_code != 0 || matches!(self.kind, LessOpenKind::PipedIgnoreExitCode))
+    }
+
+    #[cfg(test)]
+    /// For testing purposes only
+    /// Create an instance of LessOpenPreprocessor with specified valued for $LESSOPEN and $LESSCLOSE
+    fn mock_new(lessopen: Option<&str>, lessclose: Option<&str>) -> Result<LessOpenPreprocessor> {
+        if let Some(command) = lessopen {
+            env::set_var("LESSOPEN", command)
+        } else {
+            env::remove_var("LESSOPEN")
+        }
+
+        if let Some(command) = lessclose {
+            env::set_var("LESSCLOSE", command)
+        } else {
+            env::remove_var("LESSCLOSE")
+        }
+
+        Self::new()
+    }
+}
+
+enum PreprocessedKind {
+    Piped(Cursor<Vec<u8>>),
+    TempFile(File),
+}
+
+impl Read for PreprocessedKind {
+    fn read(&mut self, buf: &mut [u8]) -> std::result::Result<usize, std::io::Error> {
+        match self {
+            PreprocessedKind::Piped(data) => data.read(buf),
+            PreprocessedKind::TempFile(data) => data.read(buf),
+        }
+    }
+}
+
+pub struct Preprocessed {
+    kind: PreprocessedKind,
+    lessclose: Option<String>,
+    command_args: Vec<String>,
+    command_options: ScriptOptions,
+}
+
+impl Read for Preprocessed {
+    fn read(&mut self, buf: &mut [u8]) -> std::result::Result<usize, std::io::Error> {
+        self.kind.read(buf)
+    }
+}
+
+impl Drop for Preprocessed {
+    fn drop(&mut self) {
+        if let Some(ref command) = self.lessclose {
+            self.command_options.output_redirection = IoOptions::Inherit;
+
+            run_script::run(command, &self.command_args, &self.command_options)
+                .expect("failed to run $LESSCLOSE to clean up file");
+        }
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    // All tests here are serial because they all involve reading and writing environment variables
+    // Running them in parallel causes these tests and some others to randomly fail
+    use serial_test::serial;
+
+    use super::*;
+
+    /// Reset environment variables after each test as a precaution
+    fn reset_env_vars() {
+        env::remove_var("LESSOPEN");
+        env::remove_var("LESSCLOSE");
+    }
+
+    #[test]
+    #[serial]
+    fn test_just_lessopen() -> Result<()> {
+        let preprocessor = LessOpenPreprocessor::mock_new(Some("|batpipe %s"), None)?;
+
+        assert_eq!(preprocessor.lessopen, "batpipe $1");
+        assert!(preprocessor.lessclose.is_none());
+
+        reset_env_vars();
+
+        Ok(())
+    }
+
+    #[test]
+    #[serial]
+    fn test_just_lessclose() -> Result<()> {
+        let preprocessor = LessOpenPreprocessor::mock_new(None, Some("lessclose.sh %s %s"));
+
+        assert!(preprocessor.is_err());
+
+        reset_env_vars();
+
+        Ok(())
+    }
+
+    #[test]
+    #[serial]
+    fn test_both_lessopen_and_lessclose() -> Result<()> {
+        let preprocessor =
+            LessOpenPreprocessor::mock_new(Some("lessopen.sh %s"), Some("lessclose.sh %s %s"))?;
+
+        assert_eq!(preprocessor.lessopen, "lessopen.sh $1");
+        assert_eq!(preprocessor.lessclose.unwrap(), "lessclose.sh $1 $2");
+
+        reset_env_vars();
+
+        Ok(())
+    }
+
+    #[test]
+    #[serial]
+    fn test_lessopen_prefixes() -> Result<()> {
+        let preprocessor = LessOpenPreprocessor::mock_new(Some("batpipe %s"), None)?;
+
+        assert_eq!(preprocessor.lessopen, "batpipe $1");
+        assert!(matches!(preprocessor.kind, LessOpenKind::TempFile));
+        assert!(!preprocessor.preprocess_stdin);
+
+        let preprocessor = LessOpenPreprocessor::mock_new(Some("|batpipe %s"), None)?;
+
+        assert_eq!(preprocessor.lessopen, "batpipe $1");
+        assert!(matches!(
+            preprocessor.kind,
+            LessOpenKind::PipedIgnoreExitCode
+        ));
+        assert!(!preprocessor.preprocess_stdin);
+
+        let preprocessor = LessOpenPreprocessor::mock_new(Some("||batpipe %s"), None)?;
+
+        assert_eq!(preprocessor.lessopen, "batpipe $1");
+        assert!(matches!(preprocessor.kind, LessOpenKind::Piped));
+        assert!(!preprocessor.preprocess_stdin);
+
+        let preprocessor = LessOpenPreprocessor::mock_new(Some("-batpipe %s"), None)?;
+
+        assert_eq!(preprocessor.lessopen, "batpipe $1");
+        assert!(matches!(preprocessor.kind, LessOpenKind::TempFile));
+        assert!(preprocessor.preprocess_stdin);
+
+        let preprocessor = LessOpenPreprocessor::mock_new(Some("|-batpipe %s"), None)?;
+
+        assert_eq!(preprocessor.lessopen, "batpipe $1");
+        assert!(matches!(
+            preprocessor.kind,
+            LessOpenKind::PipedIgnoreExitCode
+        ));
+        assert!(preprocessor.preprocess_stdin);
+
+        let preprocessor = LessOpenPreprocessor::mock_new(Some("||-batpipe %s"), None)?;
+
+        assert_eq!(preprocessor.lessopen, "batpipe $1");
+        assert!(matches!(preprocessor.kind, LessOpenKind::Piped));
+        assert!(preprocessor.preprocess_stdin);
+
+        reset_env_vars();
+
+        Ok(())
+    }
+
+    #[test]
+    #[serial]
+    fn replace_part_of_argument() -> Result<()> {
+        let preprocessor =
+            LessOpenPreprocessor::mock_new(Some("|echo File:%s"), Some("echo File:%s Temp:%s"))?;
+
+        assert_eq!(preprocessor.lessopen, "echo File:$1");
+        assert_eq!(preprocessor.lessclose.unwrap(), "echo File:$1 Temp:$2");
+
+        reset_env_vars();
+
+        Ok(())
+    }
+}

src/lib.rs

@@ -34,6 +34,8 @@ mod diff;
 pub mod error;
 pub mod input;
 mod less;
+#[cfg(feature = "lessopen")]
+mod lessopen;
 pub mod line_range;
 pub(crate) mod nonprintable_notation;
 mod output;

src/line_range.rs

@@ -53,7 +53,7 @@ impl LineRange {
                     let more_lines = &line_numbers[1][1..]
                         .parse()
                         .map_err(|_| "Invalid character after +")?;
-                    new_range.lower + more_lines
+                    new_range.lower.saturating_add(*more_lines)
                 } else if first_byte == Some(b'-') {
                     // this will prevent values like "-+5" even though "+5" is valid integer
                     if line_numbers[1][1..].bytes().next() == Some(b'+') {
@@ -128,6 +128,13 @@ fn test_parse_plus() {
     assert_eq!(50, range.upper);
 }
 
+#[test]
+fn test_parse_plus_overflow() {
+    let range = LineRange::from(&format!("{}:+1", usize::MAX)).expect("Shouldn't fail on test!");
+    assert_eq!(usize::MAX, range.lower);
+    assert_eq!(usize::MAX, range.upper);
+}
+
 #[test]
 fn test_parse_plus_fail() {
     let range = LineRange::from("40:+z");

src/output.rs

@@ -114,6 +114,10 @@ impl OutputType {
                 p.args(args);
             }
             p.env("LESSCHARSET", "UTF-8");
+
+            #[cfg(feature = "lessopen")]
+            // Ensures that 'less' does not preprocess input again if '$LESSOPEN' is set.
+            p.arg("--no-lessopen");
         } else {
             p.args(args);
         };

src/preprocessor.rs

@@ -91,31 +91,27 @@ pub fn replace_nonprintable(
                     });
                     line_idx = 0;
                 }
-                // carriage return
-                '\x0D' => output.push_str(match nonprintable_notation {
-                    NonprintableNotation::Caret => "^M",
-                    NonprintableNotation::Unicode => "␍",
-                }),
-                // null
-                '\x00' => output.push_str(match nonprintable_notation {
-                    NonprintableNotation::Caret => "^@",
-                    NonprintableNotation::Unicode => "␀",
-                }),
-                // bell
-                '\x07' => output.push_str(match nonprintable_notation {
-                    NonprintableNotation::Caret => "^G",
-                    NonprintableNotation::Unicode => "␇",
-                }),
-                // backspace
-                '\x08' => output.push_str(match nonprintable_notation {
-                    NonprintableNotation::Caret => "^H",
-                    NonprintableNotation::Unicode => "␈",
-                }),
-                // escape
-                '\x1B' => output.push_str(match nonprintable_notation {
-                    NonprintableNotation::Caret => "^[",
-                    NonprintableNotation::Unicode => "␛",
-                }),
+                // ASCII control characters
+                '\x00'..='\x1F' => {
+                    let c = u32::from(chr);
+
+                    match nonprintable_notation {
+                        NonprintableNotation::Caret => {
+                            let caret_character = char::from_u32(0x40 + c).unwrap();
+                            write!(output, "^{caret_character}").ok();
+                        }
+
+                        NonprintableNotation::Unicode => {
+                            let replacement_symbol = char::from_u32(0x2400 + c).unwrap();
+                            output.push(replacement_symbol)
+                        }
+                    }
+                }
+                // delete
+                '\x7F' => match nonprintable_notation {
+                    NonprintableNotation::Caret => output.push_str("^?"),
+                    NonprintableNotation::Unicode => output.push('\u{2421}'),
+                },
                 // printable ASCII
                 c if c.is_ascii_alphanumeric()
                     || c.is_ascii_punctuation()

src/pretty_printer.rs

@@ -230,6 +230,12 @@ impl<'a> PrettyPrinter<'a> {
         self
     }
 
+    /// Specify the maximum number of consecutive empty lines to print.
+    pub fn squeeze_empty_lines(&mut self, maximum: Option<usize>) -> &mut Self {
+        self.config.squeeze_lines = maximum;
+        self
+    }
+
     /// Specify the highlighting theme
     pub fn theme(&mut self, theme: impl AsRef<str>) -> &mut Self {
         self.config.theme = theme.as_ref().to_owned();
@@ -300,7 +306,7 @@ impl<'a> PrettyPrinter<'a> {
 
         // Run the controller
         let controller = Controller::new(&self.config, &self.assets);
-        controller.run(inputs.into_iter().map(|i| i.into()).collect())
+        controller.run(inputs.into_iter().map(|i| i.into()).collect(), None)
     }
 }
 

src/printer.rs

@@ -1,4 +1,5 @@
-use std::io::Write;
+use std::fmt;
+use std::io;
 use std::vec::Vec;
 
 use nu_ansi_term::Color::{Fixed, Green, Red, Yellow};
@@ -6,17 +7,15 @@ use nu_ansi_term::Style;
 
 use bytesize::ByteSize;
 
-use console::AnsiCodeIterator;
-
 use syntect::easy::HighlightLines;
 use syntect::highlighting::Color;
+use syntect::highlighting::FontStyle;
 use syntect::highlighting::Theme;
 use syntect::parsing::SyntaxSet;
 
 use content_inspector::ContentType;
 
-use encoding::all::{UTF_16BE, UTF_16LE};
-use encoding::{DecoderTrap, Encoding};
+use encoding_rs::{UTF_16BE, UTF_16LE};
 
 use unicode_width::UnicodeWidthChar;
 
@@ -33,24 +32,68 @@ use crate::line_range::RangeCheckResult;
 use crate::preprocessor::{expand_tabs, replace_nonprintable};
 use crate::style::StyleComponent;
 use crate::terminal::{as_terminal_escaped, to_ansi_color};
-use crate::vscreen::AnsiStyle;
+use crate::vscreen::{AnsiStyle, EscapeSequence, EscapeSequenceIterator};
 use crate::wrapping::WrappingMode;
 
+const ANSI_UNDERLINE_ENABLE: EscapeSequence = EscapeSequence::CSI {
+    raw_sequence: "\x1B[4m",
+    parameters: "4",
+    intermediates: "",
+    final_byte: "m",
+};
+
+const ANSI_UNDERLINE_DISABLE: EscapeSequence = EscapeSequence::CSI {
+    raw_sequence: "\x1B[24m",
+    parameters: "24",
+    intermediates: "",
+    final_byte: "m",
+};
+
+const EMPTY_SYNTECT_STYLE: syntect::highlighting::Style = syntect::highlighting::Style {
+    foreground: Color {
+        r: 127,
+        g: 127,
+        b: 127,
+        a: 255,
+    },
+    background: Color {
+        r: 127,
+        g: 127,
+        b: 127,
+        a: 255,
+    },
+    font_style: FontStyle::empty(),
+};
+
+pub enum OutputHandle<'a> {
+    IoWrite(&'a mut dyn io::Write),
+    FmtWrite(&'a mut dyn fmt::Write),
+}
+
+impl<'a> OutputHandle<'a> {
+    fn write_fmt(&mut self, args: fmt::Arguments<'_>) -> Result<()> {
+        match self {
+            Self::IoWrite(handle) => handle.write_fmt(args).map_err(Into::into),
+            Self::FmtWrite(handle) => handle.write_fmt(args).map_err(Into::into),
+        }
+    }
+}
+
 pub(crate) trait Printer {
     fn print_header(
         &mut self,
-        handle: &mut dyn Write,
+        handle: &mut OutputHandle,
         input: &OpenedInput,
         add_header_padding: bool,
     ) -> Result<()>;
-    fn print_footer(&mut self, handle: &mut dyn Write, input: &OpenedInput) -> Result<()>;
+    fn print_footer(&mut self, handle: &mut OutputHandle, input: &OpenedInput) -> Result<()>;
 
-    fn print_snip(&mut self, handle: &mut dyn Write) -> Result<()>;
+    fn print_snip(&mut self, handle: &mut OutputHandle) -> Result<()>;
 
     fn print_line(
         &mut self,
         out_of_range: bool,
-        handle: &mut dyn Write,
+        handle: &mut OutputHandle,
         line_number: usize,
         line_buffer: &[u8],
     ) -> Result<()>;
@@ -58,39 +101,58 @@ pub(crate) trait Printer {
 
 pub struct SimplePrinter<'a> {
     config: &'a Config<'a>,
+    consecutive_empty_lines: usize,
 }
 
 impl<'a> SimplePrinter<'a> {
     pub fn new(config: &'a Config) -> Self {
-        SimplePrinter { config }
+        SimplePrinter {
+            config,
+            consecutive_empty_lines: 0,
+        }
     }
 }
 
 impl<'a> Printer for SimplePrinter<'a> {
     fn print_header(
         &mut self,
-        _handle: &mut dyn Write,
+        _handle: &mut OutputHandle,
         _input: &OpenedInput,
         _add_header_padding: bool,
     ) -> Result<()> {
         Ok(())
     }
 
-    fn print_footer(&mut self, _handle: &mut dyn Write, _input: &OpenedInput) -> Result<()> {
+    fn print_footer(&mut self, _handle: &mut OutputHandle, _input: &OpenedInput) -> Result<()> {
         Ok(())
     }
 
-    fn print_snip(&mut self, _handle: &mut dyn Write) -> Result<()> {
+    fn print_snip(&mut self, _handle: &mut OutputHandle) -> Result<()> {
         Ok(())
     }
 
     fn print_line(
         &mut self,
         out_of_range: bool,
-        handle: &mut dyn Write,
+        handle: &mut OutputHandle,
         _line_number: usize,
         line_buffer: &[u8],
     ) -> Result<()> {
+        // Skip squeezed lines.
+        if let Some(squeeze_limit) = self.config.squeeze_lines {
+            if String::from_utf8_lossy(line_buffer)
+                .trim_end_matches(|c| c == '\r' || c == '\n')
+                .is_empty()
+            {
+                self.consecutive_empty_lines += 1;
+                if self.consecutive_empty_lines > squeeze_limit {
+                    return Ok(());
+                }
+            } else {
+                self.consecutive_empty_lines = 0;
+            }
+        }
+
         if !out_of_range {
             if self.config.show_nonprintable {
                 let line = replace_nonprintable(
@@ -98,9 +160,21 @@ impl<'a> Printer for SimplePrinter<'a> {
                     self.config.tab_width,
                     self.config.nonprintable_notation,
                 );
-                write!(handle, "{}", line)?;
+                write!(handle, "{line}")?;
             } else {
-                handle.write_all(line_buffer)?
+                match handle {
+                    OutputHandle::IoWrite(handle) => handle.write_all(line_buffer)?,
+                    OutputHandle::FmtWrite(handle) => {
+                        write!(
+                            handle,
+                            "{}",
+                            std::str::from_utf8(line_buffer).map_err(|_| Error::Msg(
+                                "encountered invalid utf8 while printing to non-io buffer"
+                                    .to_string()
+                            ))?
+                        )?;
+                    }
+                }
             };
         }
         Ok(())
@@ -132,6 +206,7 @@ pub(crate) struct InteractivePrinter<'a> {
     pub line_changes: &'a Option<LineChanges>,
     highlighter_from_set: Option<HighlighterFromSet<'a>>,
     background_color_highlight: Option<Color>,
+    consecutive_empty_lines: usize,
 }
 
 impl<'a> InteractivePrinter<'a> {
@@ -184,11 +259,13 @@ impl<'a> InteractivePrinter<'a> {
             panel_width = 0;
         }
 
-        let highlighter_from_set = if input
+        // Get the highlighter for the output.
+        let is_printing_binary = input
             .reader
             .content_type
-            .map_or(false, |c| c.is_binary() && !config.show_nonprintable)
-        {
+            .map_or(false, |c| c.is_binary() && !config.show_nonprintable);
+
+        let highlighter_from_set = if is_printing_binary || !config.colored_output {
             None
         } else {
             // Determine the type of syntax for highlighting
@@ -215,10 +292,15 @@ impl<'a> InteractivePrinter<'a> {
             line_changes,
             highlighter_from_set,
             background_color_highlight,
+            consecutive_empty_lines: 0,
         })
     }
 
-    fn print_horizontal_line_term(&mut self, handle: &mut dyn Write, style: Style) -> Result<()> {
+    fn print_horizontal_line_term(
+        &mut self,
+        handle: &mut OutputHandle,
+        style: Style,
+    ) -> Result<()> {
         writeln!(
             handle,
             "{}",
@@ -227,7 +309,7 @@ impl<'a> InteractivePrinter<'a> {
         Ok(())
     }
 
-    fn print_horizontal_line(&mut self, handle: &mut dyn Write, grid_char: char) -> Result<()> {
+    fn print_horizontal_line(&mut self, handle: &mut OutputHandle, grid_char: char) -> Result<()> {
         if self.panel_width == 0 {
             self.print_horizontal_line_term(handle, self.colors.grid)?;
         } else {
@@ -251,13 +333,21 @@ impl<'a> InteractivePrinter<'a> {
             " ".repeat(self.panel_width - 1 - text_truncated.len())
         );
         if self.config.style_components.grid() {
-            format!("{} │ ", text_filled)
+            format!("{text_filled} │ ")
         } else {
             text_filled
         }
     }
 
-    fn print_header_component_indent(&mut self, handle: &mut dyn Write) -> std::io::Result<()> {
+    fn get_header_component_indent_length(&self) -> usize {
+        if self.config.style_components.grid() && self.panel_width > 0 {
+            self.panel_width + 2
+        } else {
+            self.panel_width
+        }
+    }
+
+    fn print_header_component_indent(&mut self, handle: &mut OutputHandle) -> Result<()> {
         if self.config.style_components.grid() {
             write!(
                 handle,
@@ -272,6 +362,55 @@ impl<'a> InteractivePrinter<'a> {
         }
     }
 
+    fn print_header_component_with_indent(
+        &mut self,
+        handle: &mut OutputHandle,
+        content: &str,
+    ) -> Result<()> {
+        self.print_header_component_indent(handle)?;
+        writeln!(handle, "{content}")
+    }
+
+    fn print_header_multiline_component(
+        &mut self,
+        handle: &mut OutputHandle,
+        content: &str,
+    ) -> Result<()> {
+        let mut content = content;
+        let content_width = self.config.term_width - self.get_header_component_indent_length();
+        while content.len() > content_width {
+            let (content_line, remaining) = content.split_at(content_width);
+            self.print_header_component_with_indent(handle, content_line)?;
+            content = remaining;
+        }
+        self.print_header_component_with_indent(handle, content)
+    }
+
+    fn highlight_regions_for_line<'b>(
+        &mut self,
+        line: &'b str,
+    ) -> Result<Vec<(syntect::highlighting::Style, &'b str)>> {
+        let highlighter_from_set = match self.highlighter_from_set {
+            Some(ref mut highlighter_from_set) => highlighter_from_set,
+            _ => return Ok(vec![(EMPTY_SYNTECT_STYLE, line)]),
+        };
+
+        // skip syntax highlighting on long lines
+        let too_long = line.len() > 1024 * 16;
+
+        let for_highlighting: &str = if too_long { "\n" } else { line };
+
+        let mut highlighted_line = highlighter_from_set
+            .highlighter
+            .highlight_line(for_highlighting, highlighter_from_set.syntax_set)?;
+
+        if too_long {
+            highlighted_line[0].1 = &line;
+        }
+
+        Ok(highlighted_line)
+    }
+
     fn preprocess(&self, text: &str, cursor: &mut usize) -> String {
         if self.config.tab_width > 0 {
             return expand_tabs(text, self.config.tab_width, cursor);
@@ -285,7 +424,7 @@ impl<'a> InteractivePrinter<'a> {
 impl<'a> Printer for InteractivePrinter<'a> {
     fn print_header(
         &mut self,
-        handle: &mut dyn Write,
+        handle: &mut OutputHandle,
         input: &OpenedInput,
         add_header_padding: bool,
     ) -> Result<()> {
@@ -347,31 +486,32 @@ impl<'a> Printer for InteractivePrinter<'a> {
             }
         }
 
-        header_components.iter().try_for_each(|component| {
-            self.print_header_component_indent(handle)?;
-
-            match component {
-                StyleComponent::HeaderFilename => writeln!(
-                    handle,
-                    "{}{}{}",
-                    description
-                        .kind()
-                        .map(|kind| format!("{}: ", kind))
-                        .unwrap_or_else(|| "".into()),
-                    self.colors.header_value.paint(description.title()),
-                    mode
-                ),
-
+        header_components
+            .iter()
+            .try_for_each(|component| match component {
+                StyleComponent::HeaderFilename => {
+                    let header_filename = format!(
+                        "{}{}{}",
+                        description
+                            .kind()
+                            .map(|kind| format!("{kind}: "))
+                            .unwrap_or_else(|| "".into()),
+                        self.colors.header_value.paint(description.title()),
+                        mode
+                    );
+                    self.print_header_multiline_component(handle, &header_filename)
+                }
                 StyleComponent::HeaderFilesize => {
                     let bsize = metadata
                         .size
                         .map(|s| format!("{}", ByteSize(s)))
                         .unwrap_or_else(|| "-".into());
-                    writeln!(handle, "Size: {}", self.colors.header_value.paint(bsize))
+                    let header_filesize =
+                        format!("Size: {}", self.colors.header_value.paint(bsize));
+                    self.print_header_multiline_component(handle, &header_filesize)
                 }
                 _ => Ok(()),
-            }
-        })?;
+            })?;
 
         if self.config.style_components.grid() {
             if self.content_type.map_or(false, |c| c.is_text()) || self.config.show_nonprintable {
@@ -384,7 +524,7 @@ impl<'a> Printer for InteractivePrinter<'a> {
         Ok(())
     }
 
-    fn print_footer(&mut self, handle: &mut dyn Write, _input: &OpenedInput) -> Result<()> {
+    fn print_footer(&mut self, handle: &mut OutputHandle, _input: &OpenedInput) -> Result<()> {
         if self.config.style_components.grid()
             && (self.content_type.map_or(false, |c| c.is_text()) || self.config.show_nonprintable)
         {
@@ -394,7 +534,7 @@ impl<'a> Printer for InteractivePrinter<'a> {
         }
     }
 
-    fn print_snip(&mut self, handle: &mut dyn Write) -> Result<()> {
+    fn print_snip(&mut self, handle: &mut OutputHandle) -> Result<()> {
         let panel = self.create_fake_panel(" ...");
         let panel_count = panel.chars().count();
 
@@ -412,7 +552,7 @@ impl<'a> Printer for InteractivePrinter<'a> {
             "{}",
             self.colors
                 .grid
-                .paint(format!("{}{}{}{}", panel, snip_left, title, snip_right))
+                .paint(format!("{panel}{snip_left}{title}{snip_right}"))
         )?;
 
         Ok(())
@@ -421,7 +561,7 @@ impl<'a> Printer for InteractivePrinter<'a> {
     fn print_line(
         &mut self,
         out_of_range: bool,
-        handle: &mut dyn Write,
+        handle: &mut OutputHandle,
         line_number: usize,
         line_buffer: &[u8],
     ) -> Result<()> {
@@ -431,58 +571,45 @@ impl<'a> Printer for InteractivePrinter<'a> {
                 self.config.tab_width,
                 self.config.nonprintable_notation,
             )
+            .into()
         } else {
-            let line = match self.content_type {
+            match self.content_type {
                 Some(ContentType::BINARY) | None => {
                     return Ok(());
                 }
-                Some(ContentType::UTF_16LE) => UTF_16LE
-                    .decode(line_buffer, DecoderTrap::Replace)
-                    .map_err(|_| "Invalid UTF-16LE")?,
-                Some(ContentType::UTF_16BE) => UTF_16BE
-                    .decode(line_buffer, DecoderTrap::Replace)
-                    .map_err(|_| "Invalid UTF-16BE")?,
-                _ => String::from_utf8_lossy(line_buffer).to_string(),
-            };
-            // Remove byte order mark from the first line if it exists
-            if line_number == 1 {
-                match line.strip_prefix('\u{feff}') {
-                    Some(stripped) => stripped.to_string(),
-                    None => line,
-                }
-            } else {
-                line
-            }
-        };
-
-        let regions = {
-            let highlighter_from_set = match self.highlighter_from_set {
-                Some(ref mut highlighter_from_set) => highlighter_from_set,
+                Some(ContentType::UTF_16LE) => UTF_16LE.decode_with_bom_removal(line_buffer).0,
+                Some(ContentType::UTF_16BE) => UTF_16BE.decode_with_bom_removal(line_buffer).0,
                 _ => {
-                    return Ok(());
+                    let line = String::from_utf8_lossy(line_buffer);
+                    if line_number == 1 {
+                        match line.strip_prefix('\u{feff}') {
+                            Some(stripped) => stripped.to_string().into(),
+                            None => line,
+                        }
+                    } else {
+                        line
+                    }
                 }
-            };
-
-            // skip syntax highlighting on long lines
-            let too_long = line.len() > 1024 * 16;
-
-            let for_highlighting: &str = if too_long { "\n" } else { &line };
-
-            let mut highlighted_line = highlighter_from_set
-                .highlighter
-                .highlight_line(for_highlighting, highlighter_from_set.syntax_set)?;
-
-            if too_long {
-                highlighted_line[0].1 = &line;
             }
-
-            highlighted_line
         };
 
+        let regions = self.highlight_regions_for_line(&line)?;
         if out_of_range {
             return Ok(());
         }
 
+        // Skip squeezed lines.
+        if let Some(squeeze_limit) = self.config.squeeze_lines {
+            if line.trim_end_matches(|c| c == '\r' || c == '\n').is_empty() {
+                self.consecutive_empty_lines += 1;
+                if self.consecutive_empty_lines > squeeze_limit {
+                    return Ok(());
+                }
+            } else {
+                self.consecutive_empty_lines = 0;
+            }
+        }
+
         let mut cursor: usize = 0;
         let mut cursor_max: usize = self.config.term_width;
         let mut cursor_total: usize = 0;
@@ -493,7 +620,7 @@ impl<'a> Printer for InteractivePrinter<'a> {
             self.config.highlighted_lines.0.check(line_number) == RangeCheckResult::InRange;
 
         if highlight_this_line && self.config.theme == "ansi" {
-            self.ansi_style.update("^[4m");
+            self.ansi_style.update(ANSI_UNDERLINE_ENABLE);
         }
 
         let background_color = self
@@ -520,23 +647,17 @@ impl<'a> Printer for InteractivePrinter<'a> {
             let italics = self.config.use_italic_text;
 
             for &(style, region) in &regions {
-                let ansi_iterator = AnsiCodeIterator::new(region);
+                let ansi_iterator = EscapeSequenceIterator::new(region);
                 for chunk in ansi_iterator {
                     match chunk {
-                        // ANSI escape passthrough.
-                        (ansi, true) => {
-                            self.ansi_style.update(ansi);
-                            write!(handle, "{}", ansi)?;
-                        }
-
                         // Regular text.
-                        (text, false) => {
-                            let text = &*self.preprocess(text, &mut cursor_total);
+                        EscapeSequence::Text(text) => {
+                            let text = self.preprocess(text, &mut cursor_total);
                             let text_trimmed = text.trim_end_matches(|c| c == '\r' || c == '\n');
 
                             write!(
                                 handle,
-                                "{}",
+                                "{}{}",
                                 as_terminal_escaped(
                                     style,
                                     &format!("{}{}", self.ansi_style, text_trimmed),
@@ -544,9 +665,11 @@ impl<'a> Printer for InteractivePrinter<'a> {
                                     colored_output,
                                     italics,
                                     background_color
-                                )
+                                ),
+                                self.ansi_style.to_reset_sequence(),
                             )?;
 
+                            // Pad the rest of the line.
                             if text.len() != text_trimmed.len() {
                                 if let Some(background_color) = background_color {
                                     let ansi_style = Style {
@@ -564,6 +687,12 @@ impl<'a> Printer for InteractivePrinter<'a> {
                                 write!(handle, "{}", &text[text_trimmed.len()..])?;
                             }
                         }
+
+                        // ANSI escape passthrough.
+                        _ => {
+                            write!(handle, "{}", chunk.raw())?;
+                            self.ansi_style.update(chunk);
+                        }
                     }
                 }
             }
@@ -573,17 +702,11 @@ impl<'a> Printer for InteractivePrinter<'a> {
             }
         } else {
             for &(style, region) in &regions {
-                let ansi_iterator = AnsiCodeIterator::new(region);
+                let ansi_iterator = EscapeSequenceIterator::new(region);
                 for chunk in ansi_iterator {
                     match chunk {
-                        // ANSI escape passthrough.
-                        (ansi, true) => {
-                            self.ansi_style.update(ansi);
-                            write!(handle, "{}", ansi)?;
-                        }
-
                         // Regular text.
-                        (text, false) => {
+                        EscapeSequence::Text(text) => {
                             let text = self.preprocess(
                                 text.trim_end_matches(|c| c == '\r' || c == '\n'),
                                 &mut cursor_total,
@@ -626,7 +749,7 @@ impl<'a> Printer for InteractivePrinter<'a> {
                                     // It wraps.
                                     write!(
                                         handle,
-                                        "{}\n{}",
+                                        "{}{}\n{}",
                                         as_terminal_escaped(
                                             style,
                                             &format!("{}{}", self.ansi_style, line_buf),
@@ -635,6 +758,7 @@ impl<'a> Printer for InteractivePrinter<'a> {
                                             self.config.use_italic_text,
                                             background_color
                                         ),
+                                        self.ansi_style.to_reset_sequence(),
                                         panel_wrap.clone().unwrap()
                                     )?;
 
@@ -663,6 +787,12 @@ impl<'a> Printer for InteractivePrinter<'a> {
                                 )
                             )?;
                         }
+
+                        // ANSI escape passthrough.
+                        _ => {
+                            write!(handle, "{}", chunk.raw())?;
+                            self.ansi_style.update(chunk);
+                        }
                     }
                 }
             }
@@ -683,8 +813,8 @@ impl<'a> Printer for InteractivePrinter<'a> {
         }
 
         if highlight_this_line && self.config.theme == "ansi" {
-            self.ansi_style.update("^[24m");
-            write!(handle, "\x1B[24m")?;
+            write!(handle, "{}", ANSI_UNDERLINE_DISABLE.raw())?;
+            self.ansi_style.update(ANSI_UNDERLINE_DISABLE);
         }
 
         Ok(())

src/style.rs

@@ -80,7 +80,7 @@ impl FromStr for StyleComponent {
             "full" => Ok(StyleComponent::Full),
             "default" => Ok(StyleComponent::Default),
             "plain" => Ok(StyleComponent::Plain),
-            _ => Err(format!("Unknown style '{}'", s).into()),
+            _ => Err(format!("Unknown style '{s}'").into()),
         }
     }
 }

src/syntax_mapping.rs

@@ -1,12 +1,31 @@
-use std::path::Path;
-
-use crate::error::Result;
-use ignored_suffixes::IgnoredSuffixes;
+use std::{
+    path::Path,
+    sync::{
+        atomic::{AtomicBool, Ordering},
+        Arc,
+    },
+    thread,
+};
 
 use globset::{Candidate, GlobBuilder, GlobMatcher};
+use once_cell::sync::Lazy;
 
+use crate::error::Result;
+use builtin::BUILTIN_MAPPINGS;
+use ignored_suffixes::IgnoredSuffixes;
+
+mod builtin;
 pub mod ignored_suffixes;
 
+fn make_glob_matcher(from: &str) -> Result<GlobMatcher> {
+    let matcher = GlobBuilder::new(from)
+        .case_insensitive(true)
+        .literal_separator(true)
+        .build()?
+        .compile_matcher();
+    Ok(matcher)
+}
+
 #[derive(Debug, Clone, Copy, PartialEq, Eq)]
 #[non_exhaustive]
 pub enum MappingTarget<'a> {
@@ -29,175 +48,108 @@ pub enum MappingTarget<'a> {
 
 #[derive(Debug, Clone, Default)]
 pub struct SyntaxMapping<'a> {
-    mappings: Vec<(GlobMatcher, MappingTarget<'a>)>,
+    /// User-defined mappings at run time.
+    ///
+    /// Rules in front have precedence.
+    custom_mappings: Vec<(GlobMatcher, MappingTarget<'a>)>,
+
     pub(crate) ignored_suffixes: IgnoredSuffixes<'a>,
+
+    /// A flag to halt glob matcher building, which is offloaded to another thread.
+    ///
+    /// We have this so that we can signal the thread to halt early when appropriate.
+    halt_glob_build: Arc<AtomicBool>,
+}
+
+impl<'a> Drop for SyntaxMapping<'a> {
+    fn drop(&mut self) {
+        // signal the offload thread to halt early
+        self.halt_glob_build.store(true, Ordering::Relaxed);
+    }
 }
 
 impl<'a> SyntaxMapping<'a> {
-    pub fn empty() -> SyntaxMapping<'a> {
+    pub fn new() -> SyntaxMapping<'a> {
         Default::default()
     }
 
-    pub fn builtin() -> SyntaxMapping<'a> {
-        let mut mapping = Self::empty();
-        mapping.insert("*.h", MappingTarget::MapTo("C++")).unwrap();
-        mapping
-            .insert(".clang-format", MappingTarget::MapTo("YAML"))
-            .unwrap();
-        mapping.insert("*.fs", MappingTarget::MapTo("F#")).unwrap();
-        mapping
-            .insert("build", MappingTarget::MapToUnknown)
-            .unwrap();
-        mapping
-            .insert("**/.ssh/config", MappingTarget::MapTo("SSH Config"))
-            .unwrap();
-        mapping
-            .insert(
-                "**/bat/config",
-                MappingTarget::MapTo("Bourne Again Shell (bash)"),
-            )
-            .unwrap();
-        mapping
-            .insert(
-                "/etc/profile",
-                MappingTarget::MapTo("Bourne Again Shell (bash)"),
-            )
-            .unwrap();
-        mapping
-            .insert("*.pac", MappingTarget::MapTo("JavaScript (Babel)"))
-            .unwrap();
-        mapping
-            .insert("fish_history", MappingTarget::MapTo("YAML"))
-            .unwrap();
-
-        // See #2151, https://nmap.org/book/nse-language.html
-        mapping
-            .insert("*.nse", MappingTarget::MapTo("Lua"))
-            .unwrap();
-
-        // See #1008
-        mapping
-            .insert("rails", MappingTarget::MapToUnknown)
-            .unwrap();
-
-        // Nginx and Apache syntax files both want to style all ".conf" files
-        // see #1131 and #1137
-        mapping
-            .insert("*.conf", MappingTarget::MapExtensionToUnknown)
-            .unwrap();
-
-        for glob in &[
-            "/etc/nginx/**/*.conf",
-            "/etc/nginx/sites-*/**/*",
-            "nginx.conf",
-            "mime.types",
-        ] {
-            mapping.insert(glob, MappingTarget::MapTo("nginx")).unwrap();
-        }
-
-        for glob in &[
-            "/etc/apache2/**/*.conf",
-            "/etc/apache2/sites-*/**/*",
-            "httpd.conf",
-        ] {
-            mapping
-                .insert(glob, MappingTarget::MapTo("Apache Conf"))
-                .unwrap();
-        }
-
-        for glob in &[
-            "**/systemd/**/*.conf",
-            "**/systemd/**/*.example",
-            "*.automount",
-            "*.device",
-            "*.dnssd",
-            "*.link",
-            "*.mount",
-            "*.netdev",
-            "*.network",
-            "*.nspawn",
-            "*.path",
-            "*.service",
-            "*.scope",
-            "*.slice",
-            "*.socket",
-            "*.swap",
-            "*.target",
-            "*.timer",
-        ] {
-            mapping.insert(glob, MappingTarget::MapTo("INI")).unwrap();
-        }
-
-        // unix mail spool
-        for glob in &["/var/spool/mail/*", "/var/mail/*"] {
-            mapping.insert(glob, MappingTarget::MapTo("Email")).unwrap()
-        }
-
-        // pacman hooks
-        mapping
-            .insert("*.hook", MappingTarget::MapTo("INI"))
-            .unwrap();
-
-        // Global git config files rooted in `$XDG_CONFIG_HOME/git/` or `$HOME/.config/git/`
-        // See e.g. https://git-scm.com/docs/git-config#FILES
-        if let Some(xdg_config_home) =
-            std::env::var_os("XDG_CONFIG_HOME").filter(|val| !val.is_empty())
-        {
-            insert_git_config_global(&mut mapping, &xdg_config_home);
-        }
-        if let Some(default_config_home) = std::env::var_os("HOME")
-            .filter(|val| !val.is_empty())
-            .map(|home| Path::new(&home).join(".config"))
-        {
-            insert_git_config_global(&mut mapping, &default_config_home);
-        }
-
-        fn insert_git_config_global(mapping: &mut SyntaxMapping, config_home: impl AsRef<Path>) {
-            let git_config_path = config_home.as_ref().join("git");
-
-            mapping
-                .insert(
-                    &git_config_path.join("config").to_string_lossy(),
-                    MappingTarget::MapTo("Git Config"),
-                )
-                .ok();
-
-            mapping
-                .insert(
-                    &git_config_path.join("ignore").to_string_lossy(),
-                    MappingTarget::MapTo("Git Ignore"),
-                )
-                .ok();
-
-            mapping
-                .insert(
-                    &git_config_path.join("attributes").to_string_lossy(),
-                    MappingTarget::MapTo("Git Attributes"),
-                )
-                .ok();
-        }
-
-        mapping
+    /// Start a thread to build the glob matchers for all builtin mappings.
+    ///
+    /// The use of this function while not necessary, is useful to speed up startup
+    /// times by starting this work early in parallel.
+    ///
+    /// The thread halts if/when `halt_glob_build` is set to true.
+    pub fn start_offload_build_all(&self) {
+        let halt = Arc::clone(&self.halt_glob_build);
+        thread::spawn(move || {
+            for (matcher, _) in BUILTIN_MAPPINGS.iter() {
+                if halt.load(Ordering::Relaxed) {
+                    break;
+                }
+                Lazy::force(matcher);
+            }
+        });
+        // Note that this thread is not joined upon completion because there's
+        // no shared resources that need synchronization to be safely dropped.
+        // If we later add code into this thread that requires interesting
+        // resources (e.g. IO), it would be a good idea to store the handle
+        // and join it on drop.
     }
 
     pub fn insert(&mut self, from: &str, to: MappingTarget<'a>) -> Result<()> {
-        let glob = GlobBuilder::new(from)
-            .case_insensitive(false)
-            .literal_separator(true)
-            .build()?;
-        self.mappings.push((glob.compile_matcher(), to));
+        let matcher = make_glob_matcher(from)?;
+        self.custom_mappings.push((matcher, to));
         Ok(())
     }
 
-    pub fn mappings(&self) -> &[(GlobMatcher, MappingTarget<'a>)] {
-        &self.mappings
+    /// Returns an iterator over all mappings. User-defined mappings are listed
+    /// before builtin mappings; mappings in front have higher precedence.
+    ///
+    /// Builtin mappings' `GlobMatcher`s are lazily compiled.
+    ///
+    /// Note that this function only returns mappings that are valid under the
+    /// current environment. For details see [`Self::builtin_mappings`].
+    pub fn all_mappings(&self) -> impl Iterator<Item = (&GlobMatcher, &MappingTarget<'a>)> {
+        self.custom_mappings()
+            .iter()
+            .map(|(matcher, target)| (matcher, target)) // as_ref
+            .chain(
+                // we need a map with a closure to "do" the lifetime variance
+                // see: https://discord.com/channels/273534239310479360/1120124565591425034/1170543402870382653
+                // also, clippy false positive:
+                // see: https://github.com/rust-lang/rust-clippy/issues/9280
+                #[allow(clippy::map_identity)]
+                self.builtin_mappings().map(|rule| rule),
+            )
     }
 
-    pub(crate) fn get_syntax_for(&self, path: impl AsRef<Path>) -> Option<MappingTarget<'a>> {
+    /// Returns an iterator over all valid builtin mappings. Mappings in front
+    /// have higher precedence.
+    ///
+    /// The `GlabMatcher`s are lazily compiled.
+    ///
+    /// Mappings that are invalid under the current environment (i.e. rule
+    /// requires environment variable(s) that is unset, or the joined string
+    /// after variable(s) replacement is not a valid glob expression) are
+    /// ignored.
+    pub fn builtin_mappings(
+        &self,
+    ) -> impl Iterator<Item = (&'static GlobMatcher, &'static MappingTarget<'static>)> {
+        BUILTIN_MAPPINGS
+            .iter()
+            .filter_map(|(matcher, target)| matcher.as_ref().map(|glob| (glob, target)))
+    }
+
+    /// Returns all user-defined mappings.
+    pub fn custom_mappings(&self) -> &[(GlobMatcher, MappingTarget<'a>)] {
+        &self.custom_mappings
+    }
+
+    pub fn get_syntax_for(&self, path: impl AsRef<Path>) -> Option<MappingTarget<'a>> {
         // Try matching on the file name as-is.
         let candidate = Candidate::new(&path);
         let candidate_filename = path.as_ref().file_name().map(Candidate::new);
-        for (ref glob, ref syntax) in self.mappings.iter().rev() {
+        for (glob, syntax) in self.all_mappings() {
             if glob.is_match_candidate(&candidate)
                 || candidate_filename
                     .as_ref()
@@ -220,48 +172,93 @@ impl<'a> SyntaxMapping<'a> {
     }
 }
 
-#[test]
-fn basic() {
-    let mut map = SyntaxMapping::empty();
-    map.insert("/path/to/Cargo.lock", MappingTarget::MapTo("TOML"))
-        .ok();
-    map.insert("/path/to/.ignore", MappingTarget::MapTo("Git Ignore"))
-        .ok();
+#[cfg(test)]
+mod tests {
+    use super::*;
 
-    assert_eq!(
-        map.get_syntax_for("/path/to/Cargo.lock"),
-        Some(MappingTarget::MapTo("TOML"))
-    );
-    assert_eq!(map.get_syntax_for("/path/to/other.lock"), None);
+    #[test]
+    fn builtin_mappings_work() {
+        let map = SyntaxMapping::new();
 
-    assert_eq!(
-        map.get_syntax_for("/path/to/.ignore"),
-        Some(MappingTarget::MapTo("Git Ignore"))
-    );
-}
-
-#[test]
-fn user_can_override_builtin_mappings() {
-    let mut map = SyntaxMapping::builtin();
-
-    assert_eq!(
-        map.get_syntax_for("/etc/profile"),
-        Some(MappingTarget::MapTo("Bourne Again Shell (bash)"))
-    );
-    map.insert("/etc/profile", MappingTarget::MapTo("My Syntax"))
-        .ok();
-    assert_eq!(
-        map.get_syntax_for("/etc/profile"),
-        Some(MappingTarget::MapTo("My Syntax"))
-    );
-}
-
-#[test]
-fn builtin_mappings() {
-    let map = SyntaxMapping::builtin();
-
-    assert_eq!(
-        map.get_syntax_for("/path/to/build"),
-        Some(MappingTarget::MapToUnknown)
-    );
+        assert_eq!(
+            map.get_syntax_for("/path/to/build"),
+            Some(MappingTarget::MapToUnknown)
+        );
+    }
+
+    #[test]
+    fn all_fixed_builtin_mappings_can_compile() {
+        let map = SyntaxMapping::new();
+
+        // collect call evaluates all lazy closures
+        // fixed builtin mappings will panic if they fail to compile
+        let _mappings = map.builtin_mappings().collect::<Vec<_>>();
+    }
+
+    #[test]
+    fn builtin_mappings_matcher_only_compile_once() {
+        let map = SyntaxMapping::new();
+
+        let two_iterations: Vec<_> = (0..2)
+            .map(|_| {
+                // addresses of every matcher
+                map.builtin_mappings()
+                    .map(|(matcher, _)| matcher as *const _ as usize)
+                    .collect::<Vec<_>>()
+            })
+            .collect();
+
+        // if the matchers are only compiled once, their address should remain the same
+        assert_eq!(two_iterations[0], two_iterations[1]);
+    }
+
+    #[test]
+    fn custom_mappings_work() {
+        let mut map = SyntaxMapping::new();
+        map.insert("/path/to/Cargo.lock", MappingTarget::MapTo("TOML"))
+            .ok();
+        map.insert("/path/to/.ignore", MappingTarget::MapTo("Git Ignore"))
+            .ok();
+
+        assert_eq!(
+            map.get_syntax_for("/path/to/Cargo.lock"),
+            Some(MappingTarget::MapTo("TOML"))
+        );
+        assert_eq!(map.get_syntax_for("/path/to/other.lock"), None);
+
+        assert_eq!(
+            map.get_syntax_for("/path/to/.ignore"),
+            Some(MappingTarget::MapTo("Git Ignore"))
+        );
+    }
+
+    #[test]
+    fn custom_mappings_override_builtin() {
+        let mut map = SyntaxMapping::new();
+
+        assert_eq!(
+            map.get_syntax_for("/path/to/httpd.conf"),
+            Some(MappingTarget::MapTo("Apache Conf"))
+        );
+        map.insert("httpd.conf", MappingTarget::MapTo("My Syntax"))
+            .ok();
+        assert_eq!(
+            map.get_syntax_for("/path/to/httpd.conf"),
+            Some(MappingTarget::MapTo("My Syntax"))
+        );
+    }
+
+    #[test]
+    fn custom_mappings_precedence() {
+        let mut map = SyntaxMapping::new();
+
+        map.insert("/path/to/foo", MappingTarget::MapTo("alpha"))
+            .ok();
+        map.insert("/path/to/foo", MappingTarget::MapTo("bravo"))
+            .ok();
+        assert_eq!(
+            map.get_syntax_for("/path/to/foo"),
+            Some(MappingTarget::MapTo("alpha"))
+        );
+    }
 }

src/syntax_mapping/builtin.rs

@@ -0,0 +1,91 @@
+use std::env;
+
+use globset::GlobMatcher;
+use once_cell::sync::Lazy;
+
+use crate::syntax_mapping::{make_glob_matcher, MappingTarget};
+
+// Static syntax mappings generated from /src/syntax_mapping/builtins/ by the
+// build script (/build/syntax_mapping.rs).
+include!(concat!(
+    env!("OUT_DIR"),
+    "/codegen_static_syntax_mappings.rs"
+));
+
+// The defined matcher strings are analysed at compile time and converted into
+// lazily-compiled `GlobMatcher`s. This is so that the string searches are moved
+// from run time to compile time, thus improving startup performance.
+//
+// To any future maintainer (including possibly myself) wondering why there is
+// not a `BuiltinMatcher` enum that looks like this:
+//
+// ```
+// enum BuiltinMatcher {
+//     Fixed(&'static str),
+//     Dynamic(Lazy<Option<String>>),
+// }
+// ```
+//
+// Because there was. I tried it and threw it out.
+//
+// Naively looking at the problem from a distance, this may seem like a good
+// design (strongly typed etc. etc.). It would also save on compiled size by
+// extracting out common behaviour into functions. But while actually
+// implementing the lazy matcher compilation logic, I realised that it's most
+// convenient for `BUILTIN_MAPPINGS` to have the following type:
+//
+// `[(Lazy<Option<GlobMatcher>>, MappingTarget); N]`
+//
+// The benefit for this is that operations like listing all builtin mappings
+// would be effectively memoised. The caller would not have to compile another
+// `GlobMatcher` for rules that they have previously visited.
+//
+// Unfortunately, this means we are going to have to store a distinct closure
+// for each rule anyway, which makes a `BuiltinMatcher` enum a pointless layer
+// of indirection.
+//
+// In the current implementation, the closure within each generated rule simply
+// calls either `build_matcher_fixed` or `build_matcher_dynamic`, depending on
+// whether the defined matcher contains dynamic segments or not.
+
+/// Compile a fixed glob string into a glob matcher.
+///
+/// A failure to compile is a fatal error.
+///
+/// Used internally by `Lazy<Option<GlobMatcher>>`'s lazy evaluation closure.
+fn build_matcher_fixed(from: &str) -> GlobMatcher {
+    make_glob_matcher(from).expect("A builtin fixed glob matcher failed to compile")
+}
+
+/// Join a list of matcher segments to create a glob string, replacing all
+/// environment variables, then compile to a glob matcher.
+///
+/// Returns `None` if any replacement fails, or if the joined glob string fails
+/// to compile.
+///
+/// Used internally by `Lazy<Option<GlobMatcher>>`'s lazy evaluation closure.
+fn build_matcher_dynamic(segs: &[MatcherSegment]) -> Option<GlobMatcher> {
+    // join segments
+    let mut buf = String::new();
+    for seg in segs {
+        match seg {
+            MatcherSegment::Text(s) => buf.push_str(s),
+            MatcherSegment::Env(var) => {
+                let replaced = env::var(var).ok()?;
+                buf.push_str(&replaced);
+            }
+        }
+    }
+    // compile glob matcher
+    let matcher = make_glob_matcher(&buf).ok()?;
+    Some(matcher)
+}
+
+/// A segment of a dynamic builtin matcher.
+///
+/// Used internally by `Lazy<Option<GlobMatcher>>`'s lazy evaluation closure.
+#[derive(Clone, Debug)]
+enum MatcherSegment {
+    Text(&'static str),
+    Env(&'static str),
+}

src/syntax_mapping/builtins/README.md

@@ -0,0 +1,116 @@
+# `/src/syntax_mapping/builtins`
+
+The files in this directory define path/name-based syntax mappings, which amend
+and take precedence over the extension/content-based syntax mappings provided by
+[syntect](https://github.com/trishume/syntect).
+
+## File organisation
+
+Each TOML file should describe the syntax mappings of a single application, or
+otherwise a set of logically-related rules.
+
+What defines "a single application" here is deliberately vague, since the
+file-splitting is purely for maintainability reasons. (Technically, we could
+just as well use a single TOML file.) So just use common sense.
+
+TOML files should reside in the corresponding subdirectory of the platform(s)
+that they intend to target. At compile time, the build script will go through
+each subdirectory that is applicable to the compilation target, collect the
+syntax mappings defined by all TOML files, and embed them into the binary.
+
+## File syntax
+
+Each TOML file should contain a single section named `mappings`, with each of
+its keys being a language identifier (first column of `bat -L`; also referred to
+as "target").
+
+The value of each key should be an array of strings, with each item being a glob
+matcher. We will call each of these items a "rule".
+
+For example, if `foo-application` uses both TOML and YAML configuration files,
+we could write something like this:
+
+```toml
+# 30-foo-application.toml
+[mappings]
+"TOML" = [
+    # rules for TOML syntax go here
+    "/usr/share/foo-application/toml-config/*.conf",
+    "/etc/foo-application/toml-config/*.conf",
+]
+"YAML" = [
+    # rules for YAML syntax go here
+    # ...
+]
+```
+
+### Dynamic environment variable replacement
+
+In additional to the standard glob matcher syntax, rules also support dynamic
+replacement of environment variables at runtime. This allows us to concisely
+handle things like [XDG](https://specifications.freedesktop.org/basedir-spec/latest/).
+
+All environment variables intended to be replaced at runtime must be enclosed in
+`${}`, for example `"/foo/*/${YOUR_ENV}-suffix/*.log"`. Note that this is the
+**only** admissible syntax; other variable substitution syntaxes are not
+supported and will either cause a compile time error, or be treated as plain
+text.
+
+For example, if `foo-application` also supports per-user configuration files, we
+could write something like this:
+
+```toml
+# 30-foo-application.toml
+[mappings]
+"TOML" = [
+    # rules for TOML syntax go here
+    "/usr/share/foo-application/toml-config/*.conf",
+    "/etc/foo-application/toml-config/*.conf",
+    "${XDG_CONFIG_HOME}/foo-application/toml-config/*.conf",
+    "${HOME}/.config/foo-application/toml-config/*.conf",
+]
+"YAML" = [
+    # rules for YAML syntax go here
+    # ...
+]
+```
+
+If any environment variable replacement in a rule fails (for example when a
+variable is unset), or if the glob string after replacements is invalid, the
+entire rule will be ignored.
+
+### Explicitly mapping to unknown
+
+Sometimes it may be necessary to "unset" a particular syntect mapping - perhaps
+a syntax's matching rules are "too greedy", and is claiming files that it should
+not. In this case, there are two special identifiers:
+`MappingTarget::MapToUnknown` and `MappingTarget::MapExtensionToUnknown`
+(corresponding to the two variants of the `syntax_mapping::MappingTarget` enum).
+
+An example of this would be `*.conf` files in general. So we may write something
+like this:
+
+```toml
+# 99-unset-ambiguous-extensions.toml
+[mappings]
+"MappingTarget::MapExtensionToUnknown" = [
+    "*.conf",
+]
+```
+
+## Ordering
+
+At compile time, all TOML files applicable to the target are processed in
+lexicographical filename order. So `00-foo.toml` takes precedence over
+`10-bar.toml`, which takes precedence over `20-baz.toml`, and so on. Note that
+**only** the filenames of the TOML files are taken into account; the
+subdirectories they are placed in have no influence on ordering.
+
+This behaviour can be occasionally useful for creating high/low priority rules,
+such as in the aforementioned example of explicitly mapping `*.conf` files to
+unknown. Generally this should not be much of a concern though, since rules
+should be written as specifically as possible for each application.
+
+Rules within each TOML file are processed (and therefore matched) in the order
+in which they are defined. At runtime, the syntax selection algorithm will
+short-circuit and return the target of the first matching rule.

src/syntax_mapping/builtins/bsd-family/50-os-release.toml

@@ -0,0 +1,2 @@
+[mappings]
+"Bourne Again Shell (bash)" = ["/etc/os-release", "/var/run/os-release"]

src/syntax_mapping/builtins/common/50-apache.toml

@@ -0,0 +1,2 @@
+[mappings]
+"Apache Conf" = ["httpd.conf"]

src/syntax_mapping/builtins/common/50-aws-credentials.toml

@@ -0,0 +1,2 @@
+[mappings]
+"INI" = ["**/.aws/credentials", "**/.aws/config"]

src/syntax_mapping/builtins/common/50-bat.toml

@@ -0,0 +1,2 @@
+[mappings]
+"Bourne Again Shell (bash)" = ["**/bat/config"]

src/syntax_mapping/builtins/common/50-container.toml

@@ -0,0 +1,2 @@
+[mappings]
+"Dockerfile" = ["Containerfile"]

src/syntax_mapping/builtins/common/50-cpp.toml

@@ -0,0 +1,6 @@
+[mappings]
+"C++" = [
+    # probably better than the default Objective C mapping #877
+    "*.h",
+]
+"YAML" = [".clang-format"]

src/syntax_mapping/builtins/common/50-f-sharp.toml

@@ -0,0 +1,2 @@
+[mappings]
+"F#" = ["*.fs"]

src/syntax_mapping/builtins/common/50-git.toml

@@ -0,0 +1,10 @@
+# Global git config files rooted in `$XDG_CONFIG_HOME/git/` or `$HOME/.config/git/`
+# See e.g. https://git-scm.com/docs/git-config#FILES
+
+[mappings]
+"Git Config" = ["${XDG_CONFIG_HOME}/git/config", "${HOME}/.config/git/config"]
+"Git Ignore" = ["${XDG_CONFIG_HOME}/git/ignore", "${HOME}/.config/git/ignore"]
+"Git Attributes" = [
+    "${XDG_CONFIG_HOME}/git/attributes",
+    "${HOME}/.config/git/attributes",
+]

src/syntax_mapping/builtins/common/50-json.toml

@@ -0,0 +1,3 @@
+# JSON Lines is a simple variation of JSON #2535
+[mappings]
+"JSON" = ["*.jsonl", "*.jsonc"]

src/syntax_mapping/builtins/common/50-nginx.toml

@@ -0,0 +1,2 @@
+[mappings]
+"nginx" = ["nginx.conf", "mime.types"]

src/syntax_mapping/builtins/common/50-nmap.toml

@@ -0,0 +1,3 @@
+[mappings]
+# See #2151, https://nmap.org/book/nse-language.html
+"Lua" = ["*.nse"]

src/syntax_mapping/builtins/common/50-proxy-auto-config.toml

@@ -0,0 +1,3 @@
+# 1515
+[mappings]
+"JavaScript (Babel)" = ["*.pac"]

src/syntax_mapping/builtins/common/50-ron.toml

@@ -0,0 +1,3 @@
+# Rusty Object Notation #2427
+[mappings]
+"Rust" = ["*.ron"]

src/syntax_mapping/builtins/common/50-sarif.toml

@@ -0,0 +1,3 @@
+# SARIF is a format for reporting static analysis results #2695
+[mappings]
+"JSON" = ["*.sarif"]

src/syntax_mapping/builtins/common/50-ssh.toml

@@ -0,0 +1,2 @@
+[mappings]
+"SSH Config" = ["**/.ssh/config"]

src/syntax_mapping/builtins/common/99-unset-ambiguous-extensions.toml

@@ -0,0 +1,5 @@
+[mappings]
+"MappingTarget::MapExtensionToUnknown" = [
+    # common extension used for all kinds of formats
+    "*.conf",
+]

src/syntax_mapping/builtins/common/99-unset-ambiguous-filenames.toml

@@ -0,0 +1,7 @@
+[mappings]
+"MappingTarget::MapToUnknown" = [
+    # "NAnt Build File" should only match *.build files, not files named "build"
+    "build",
+    # "bin/rails" scripts in a Ruby project misidentified as HTML (Rails) #1008
+    "rails",
+]

src/syntax_mapping/builtins/common/xonsh.toml

@@ -0,0 +1,3 @@
+# Xonsh shell (https://xon.sh/)
+[mappings]
+"Python" = ["*.xsh", "*.xonshrc"]

src/syntax_mapping/builtins/linux/50-containers.toml

@@ -0,0 +1,8 @@
+# see https://github.com/containers/image/tree/main/docs
+[mappings]
+"TOML" = [
+    "/usr/share/containers/**/*.conf",
+    "/etc/containers/**/*.conf",
+    "${HOME}/.config/containers/**/*.conf",
+    "${XDG_CONFIG_HOME}/containers/**/*.conf",
+]

src/syntax_mapping/builtins/linux/50-os-release.toml

@@ -0,0 +1,7 @@
+[mappings]
+"Bourne Again Shell (bash)" = [
+    "/etc/os-release",
+    "/usr/lib/os-release",
+    "/etc/initrd-release",
+    "/usr/lib/extension-release.d/extension-release.*",
+]

src/syntax_mapping/builtins/linux/50-pacman.toml

@@ -0,0 +1,3 @@
+[mappings]
+# pacman hooks
+"INI" = ["/usr/share/libalpm/hooks/*.hook", "/etc/pacman.d/hooks/*.hook"]

src/syntax_mapping/builtins/linux/50-podman-quadlet.toml

@@ -0,0 +1,7 @@
+# see `man quadlet`
+[mappings]
+"INI" = [
+    "**/containers/systemd/*.{container,volume,network,kube,image}",
+    "**/containers/systemd/users/*.{container,volume,network,kube,image}",
+    "**/containers/systemd/users/*/*.{container,volume,network,kube,image}",
+]

src/syntax_mapping/builtins/linux/50-systemd.toml

@@ -0,0 +1,21 @@
+[mappings]
+"INI" = [
+    "**/systemd/**/*.conf",
+    "**/systemd/**/*.example",
+    "*.automount",
+    "*.device",
+    "*.dnssd",
+    "*.link",
+    "*.mount",
+    "*.netdev",
+    "*.network",
+    "*.nspawn",
+    "*.path",
+    "*.service",
+    "*.scope",
+    "*.slice",
+    "*.socket",
+    "*.swap",
+    "*.target",
+    "*.timer",
+]

src/syntax_mapping/builtins/unix-family/50-apache.toml

@@ -0,0 +1,2 @@
+[mappings]
+"Apache Conf" = ["/etc/apache2/**/*.conf", "/etc/apache2/sites-*/**/*"]