Update to crossbeam-channel 0.5.15

To grab a double-free fix. Link: https://github.com/crossbeam-rs/crossbeam/pull/1187
Merge pull request #1711 from musicinmybrain/etcetera-0.10
2026-09-01 04:28:07 +08:00 · 2025-04-29 09:06:02 -04:00 · 2025-04-29 09:05:44 -04:00 · 2025-04-29 08:55:31 -04:00 · 2025-04-24 09:36:48 -06:00 · 2025-04-24 10:03:57 -04:00
31 changed files with 1557 additions and 992 deletions
@@ -36,7 +36,7 @@ jobs:

  ensure_cargo_fmt:
    name: Ensure 'cargo fmt' has been run
-    runs-on: ubuntu-20.04
+    runs-on: ubuntu-22.04
    steps:
    - uses: dtolnay/rust-toolchain@stable
      with:
@@ -44,9 +44,19 @@ jobs:
    - uses: actions/checkout@v4
    - run: cargo fmt -- --check

+  lint_check:
+    name: Ensure 'cargo clippy' has no warnings
+    runs-on: ubuntu-latest
+    steps:
+    - uses: dtolnay/rust-toolchain@stable
+      with:
+        components: clippy
+    - uses: actions/checkout@v4
+    - run: cargo clippy --all-targets --all-features -- -Dwarnings
+
  min_version:
    name: Minimum supported rust version
-    runs-on: ubuntu-20.04
+    runs-on: ubuntu-22.04
    needs: crate_metadata
    steps:
    - name: Checkout source code
@@ -70,17 +80,19 @@ jobs:
      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-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, use-cross: true }
-          - { target: x86_64-unknown-linux-musl   , os: ubuntu-20.04, use-cross: true }
+          - { target: aarch64-unknown-linux-gnu   , os: ubuntu-22.04, use-cross: true }
+          - { target: aarch64-unknown-linux-musl  , os: ubuntu-22.04, use-cross: true }
+          - { target: arm-unknown-linux-gnueabihf , os: ubuntu-22.04, use-cross: true }
+          - { target: arm-unknown-linux-musleabihf, os: ubuntu-22.04, use-cross: true }
+          - { target: i686-pc-windows-msvc        , os: windows-2022                  }
+          - { target: i686-unknown-linux-gnu      , os: ubuntu-22.04, use-cross: true }
+          - { target: i686-unknown-linux-musl     , os: ubuntu-22.04, use-cross: true }
+          - { target: x86_64-apple-darwin         , os: macos-13                      }
+          - { target: aarch64-apple-darwin        , os: macos-14                      }
+          - { target: x86_64-pc-windows-gnu       , os: windows-2022                  }
+          - { target: x86_64-pc-windows-msvc      , os: windows-2022                  }
+          - { target: x86_64-unknown-linux-gnu    , os: ubuntu-22.04, use-cross: true }
+          - { target: x86_64-unknown-linux-musl   , os: ubuntu-22.04, use-cross: true }
    env:
      BUILD_CMD: cargo
    steps:
@@ -99,6 +111,9 @@ jobs:
      uses: dtolnay/rust-toolchain@stable
      with:
        targets: ${{ matrix.job.target }}
+        # On windows, for now build with 1.77.2, so that it works on windows 7.
+        # When we update the MSRV again, we'll need to revisit this, and probably drop support for Win7
+        toolchain: "${{ contains(matrix.job.target, 'windows-') && '1.77.2' || 'stable' }}"

    - name: Install cross
      if: matrix.job.use-cross
@@ -201,113 +216,11 @@ jobs:
      id: debian-package
      shell: bash
      if: startsWith(matrix.job.os, 'ubuntu')
-      run: |
-        COPYRIGHT_YEARS="2018 - "$(date "+%Y")
-        DPKG_STAGING="${{ env.CICD_INTERMEDIATES_DIR }}/debian-package"
-        DPKG_DIR="${DPKG_STAGING}/dpkg"
-        mkdir -p "${DPKG_DIR}"
-
-        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 }}
-
-        unset DPKG_ARCH
-        case ${{ matrix.job.target }} in
-          aarch64-*-linux-*) DPKG_ARCH=arm64 ;;
-          arm-*-linux-*hf) DPKG_ARCH=armhf ;;
-          i686-*-linux-*) DPKG_ARCH=i686 ;;
-          x86_64-*-linux-*) DPKG_ARCH=amd64 ;;
-          *) DPKG_ARCH=notset ;;
-        esac;
-
-        DPKG_NAME="${DPKG_BASENAME}_${DPKG_VERSION}_${DPKG_ARCH}.deb"
-        echo "DPKG_NAME=${DPKG_NAME}" >> $GITHUB_OUTPUT
-
-        # Binary
-        install -Dm755 "${{ steps.bin.outputs.BIN_PATH }}" "${DPKG_DIR}/usr/bin/${{ steps.bin.outputs.BIN_NAME }}"
-
-        # Man page
-        install -Dm644 'doc/${{ needs.crate_metadata.outputs.name }}.1' "${DPKG_DIR}/usr/share/man/man1/${{ needs.crate_metadata.outputs.name }}.1"
-        gzip -n --best "${DPKG_DIR}/usr/share/man/man1/${{ needs.crate_metadata.outputs.name }}.1"
-
-        # Autocompletion files
-        install -Dm644 'autocomplete/fd.bash' "${DPKG_DIR}/usr/share/bash-completion/completions/${{ needs.crate_metadata.outputs.name }}"
-        install -Dm644 'autocomplete/fd.fish' "${DPKG_DIR}/usr/share/fish/vendor_completions.d/${{ needs.crate_metadata.outputs.name }}.fish"
-        install -Dm644 'autocomplete/_fd' "${DPKG_DIR}/usr/share/zsh/vendor-completions/_${{ needs.crate_metadata.outputs.name }}"
-
-        # README and LICENSE
-        install -Dm644 "README.md" "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/README.md"
-        install -Dm644 "LICENSE-MIT" "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/LICENSE-MIT"
-        install -Dm644 "LICENSE-APACHE" "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/LICENSE-APACHE"
-        install -Dm644 "CHANGELOG.md" "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/changelog"
-        gzip -n --best "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/changelog"
-
-        cat > "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/copyright" <<EOF
-        Format: http://www.debian.org/doc/packaging-manuals/copyright-format/1.0/
-        Upstream-Name: ${{ needs.crate_metadata.outputs.name }}
-        Source: ${{ needs.crate_metadata.outputs.homepage }}
-
-        Files: *
-        Copyright: ${{ needs.crate_metadata.outputs.maintainer }}
-        Copyright: $COPYRIGHT_YEARS ${{ needs.crate_metadata.outputs.maintainer }}
-        License: Apache-2.0 or MIT
-
-        License: Apache-2.0
-          On Debian systems, the complete text of the Apache-2.0 can be found in the
-          file /usr/share/common-licenses/Apache-2.0.
-
-        License: MIT
-          Permission is hereby granted, free of charge, to any
-          person obtaining a copy of this software and associated
-          documentation files (the "Software"), to deal in the
-          Software without restriction, including without
-          limitation the rights to use, copy, modify, merge,
-          publish, distribute, sublicense, and/or sell copies of
-          the Software, and to permit persons to whom the Software
-          is furnished to do so, subject to the following
-          conditions:
-          .
-          The above copyright notice and this permission notice
-          shall be included in all copies or substantial portions
-          of the Software.
-          .
-          THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF
-          ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED
-          TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A
-          PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
-          SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
-          CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
-          OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR
-          IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
-          DEALINGS IN THE SOFTWARE.
-        EOF
-          chmod 644 "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/copyright"
-
-          # control file
-          mkdir -p "${DPKG_DIR}/DEBIAN"
-          cat > "${DPKG_DIR}/DEBIAN/control" <<EOF
-        Package: ${DPKG_BASENAME}
-        Version: ${DPKG_VERSION}
-        Section: utils
-        Priority: optional
-        Maintainer: ${{ needs.crate_metadata.outputs.maintainer }}
-        Homepage: ${{ needs.crate_metadata.outputs.homepage }}
-        Architecture: ${DPKG_ARCH}
-        Provides: ${{ needs.crate_metadata.outputs.name }}
-        Conflicts: ${DPKG_CONFLICTS}
-        Description: simple, fast and user-friendly alternative to find
-          fd is a program to find entries in your filesystem.
-          It is a simple, fast and user-friendly alternative to find.
-          While it does not aim to support all of finds powerful functionality, it provides
-          sensible (opinionated) defaults for a majority of use cases.
-        EOF
-
-        DPKG_PATH="${DPKG_STAGING}/${DPKG_NAME}"
-        echo "DPKG_PATH=${DPKG_PATH}" >> $GITHUB_OUTPUT
-
-        # build dpkg
-        fakeroot dpkg-deb --build "${DPKG_DIR}" "${DPKG_PATH}"
+      run: bash scripts/create-deb.sh
+      env:
+        TARGET: ${{ matrix.job.target }}
+        DPKG_VERSION: ${{ needs.crate_metadata.version }}
+        BIN_PATH: ${{ steps.bin.outputs.BIN_PATH }}

    - name: "Artifact upload: tarball"
      uses: actions/upload-artifact@master
@@ -330,7 +243,7 @@ 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: |
@@ -1,9 +1,88 @@
+# Upcoming release
+
+## Features
+
+- Add a hidden `--mindepth` alias for `--min-depth`. (#1617)
+
+
+## Bugfixes
+
+
+## Changes
+
+- Replace `humantime` crate and `chrono` crate with `jiff` crate, see #1690 (@sorairolake). This has some small changes to the
+  way dates given to options such `--changed-within` and `--changed-before` including:
+  - 'M' no longer means "month", as that could be confusing with minutes. Use "mo", "mos", "month" or "months" instead.
+  - month and year now account for variability in the calander rather than being a hard-coded number of seconds. That is probably
+    what you would expect, but it is a slight change in behavior.
+
+## Other
+
+
+# 10.2.0
+
+## Features
+
+- Add --hyperlink option to add OSC 8 hyperlinks to output
+
+
+## Bugfixes
+
+
+## Changes
+
+- Build windows releases with rust 1.77 so windows 7 is still supported
+- Deb packages now include symlink for fdfind to be more consistent with official packages
+
+
+## Other
+
+# 10.1.0
+
+## Features
+
+- Allow passing an optional argument to `--strip-cwd-prefix` of "always", "never", or "auto". to force whether the cwd prefix is stripped or not.
+- Add a `--format` option which allows using a format template for direct ouput similar to the template used for `--exec`. (#1043)
+
+## Bugfixes
+- Fix aarch64 page size again. This time it should actually work. (#1085, #1549) (@tavianator)
+
+
+## Other
+
+- aarch64-apple-darwin target added to builds on the release page. Note that this is a tier 2 rust target.
+
+# v10.0.0
+
+## Features
+
+- Add `dir` as an alias to `directory` when using `-t` \ `--type`, see #1460 and #1464 (@Ato2207).
+- Add support for @%s date format in time filters similar to GNU date (seconds since Unix epoch for --older/--newer), see #1493 (@nabellows)
+- Breaking: No longer automatically ignore `.git` when using `--hidden` with vcs ignore enabled. This reverts the change in v9.0.0. While this feature
+  was often useful, it also broke some existing workflows, and there wasn't a good way to opt out of it. And there isn't really a good way for us to add
+  a way to opt out of it. And you can easily get similar behavior by adding `.git/` to your global fdignore file.
+    See #1457.
+
+## Bugfixes
+
+- Respect NO_COLOR environment variable with `--list-details` option. (#1455)
+- Fix bug that would cause hidden files to be included despite gitignore rules
+  if search path is "." (#1461, BurntSushi/ripgrep#2711).
+- aarch64 builds now use 64k page sizes with jemalloc. This fixes issues on some systems, such as ARM Macs that
+  have a larger system page size than the system that the binary was built on. (#1547)
+- Address [CVE-2024-24576](https://blog.rust-lang.org/2024/04/09/cve-2024-24576.html), by increasing minimum rust version.
+
+
+## Changes
+- Minimum supported rust version is now 1.77.2
+
+
 # v9.0.0

 ## Performance

 - Performance has been *significantly improved*, both due to optimizations in the underlying `ignore`
-  crate (#1429), and in `fd` itself (#1422, #1408, #13620) - @tavianator.
+  crate (#1429), and in `fd` itself (#1422, #1408, #1362) - @tavianator.
  [Benchmarks results](https://gist.github.com/tavianator/32edbe052f33ef60570cf5456b59de81) show gains
  of 6-8x for full traversals of smaller directories (100k files) and up to 13x for larger directories (1M files).

@@ -5,6 +5,7 @@ categories = ["command-line-utilities"]
 description = "fd is a simple, fast and user-friendly alternative to find."
 exclude = ["/benchmarks/*"]
 homepage = "https://github.com/sharkdp/fd"
+documentation = "https://docs.rs/fd-find"
 keywords = [
    "search",
    "find",
@@ -16,9 +17,9 @@ license = "MIT OR Apache-2.0"
 name = "fd-find"
 readme = "README.md"
 repository = "https://github.com/sharkdp/fd"
-version = "9.0.0"
+version = "10.2.0"
 edition= "2021"
-rust-version = "1.70.0"
+rust-version = "1.77.2"

 [badges.appveyor]
 repository = "sharkdp/fd"
@@ -34,38 +35,33 @@ path = "src/main.rs"
 version_check = "0.9"

 [dependencies]
-aho-corasick = "1.0"
-nu-ansi-term = "0.49"
+aho-corasick = "1.1"
+nu-ansi-term = "0.50"
 argmax = "0.3.1"
-ignore = "0.4.21"
-regex = "1.9.6"
+ignore = "0.4.23"
+regex = "1.11.1"
 regex-syntax = "0.8"
 ctrlc = "3.2"
-humantime = "2.1"
 globset = "0.4"
 anyhow = "1.0"
-etcetera = "0.8"
+etcetera = "0.10"
 normpath = "1.1.1"
-crossbeam-channel = "0.5.8"
-clap_complete = {version = "4.4.4", optional = true}
+crossbeam-channel = "0.5.15"
+clap_complete = {version = "4.5.47", optional = true}
 faccess = "0.2.4"
+jiff = "0.2.5"

 [dependencies.clap]
-version = "4.4.10"
+version = "4.5.35"
 features = ["suggestions", "color", "wrap_help", "cargo", "derive"]

-[dependencies.chrono]
-version = "0.4.31"
-default-features = false
-features = ["std", "clock"]
-
 [dependencies.lscolors]
-version = "0.16"
+version = "0.20"
 default-features = false
 features = ["nu-ansi-term"]

 [target.'cfg(unix)'.dependencies]
-nix = { version = "0.27.1", default-features = false, features = ["signal", "user"] }
+nix = { version = "0.29.0", default-features = false, features = ["signal", "user", "hostname"] }

 [target.'cfg(all(unix, not(target_os = "redox")))'.dependencies]
 libc = "0.2"
@@ -78,7 +74,7 @@ jemallocator = {version = "0.5.4", optional = true}

 [dev-dependencies]
 diff = "0.1"
-tempfile = "3.8"
+tempfile = "3.19"
 filetime = "0.2"
 test-case = "3.3"

@@ -0,0 +1,6 @@
+# https://github.com/sharkdp/fd/issues/1085
+[target.aarch64-unknown-linux-gnu.env]
+passthrough = ["JEMALLOC_SYS_WITH_LG_PAGE=16"]
+
+[target.aarch64-unknown-linux-musl.env]
+passthrough = ["JEMALLOC_SYS_WITH_LG_PAGE=16"]
@@ -6,7 +6,7 @@ datadir=$(prefix)/share
 exe_name=fd

 $(EXE): Cargo.toml src/**/*.rs
-	cargo build --profile $(PROFILE)
+	cargo build --profile $(PROFILE) --locked

 .PHONY: completions
 completions: autocomplete/fd.bash autocomplete/fd.fish autocomplete/fd.ps1 autocomplete/_fd
@@ -10,10 +10,7 @@ It is a simple, fast and user-friendly alternative to [`find`](https://www.gnu.o
 While it does not aim to support all of `find`'s powerful functionality, it provides sensible
 (opinionated) defaults for a majority of use cases.

-Quick links:
-* [How to use](#how-to-use)
-* [Installation](#installation)
-* [Troubleshooting](#troubleshooting)
+[Installation](#installation) • [How to use](#how-to-use) • [Troubleshooting](#troubleshooting)

 ## Features

@@ -63,7 +60,7 @@ X11/xinit/xinitrc
 X11/xinit/xserverrc
 ```

-The regular expression syntax used by `fd` is [documented here](https://docs.rs/regex/1.0.0/regex/#syntax).
+The regular expression syntax used by `fd` is [documented here](https://docs.rs/regex/latest/regex/#syntax).

 ### Specifying the root directory

@@ -261,12 +258,17 @@ To make exclude-patterns like these permanent, you can create a `.fdignore` file
 /mnt/external-drive
 *.bak
 ```
-Note: `fd` also supports `.ignore` files that are used by other programs such as `rg` or `ag`.
+
+> [!NOTE]
+> `fd` also supports `.ignore` files that are used by other programs such as `rg` or `ag`.

 If you want `fd` to ignore these patterns globally, you can put them in `fd`'s global ignore file.
 This is usually located in `~/.config/fd/ignore` in macOS or Linux, and `%APPDATA%\fd\ignore` in
 Windows.

+You may wish to include `.git/` in your `fd/ignore` file so that `.git` directories, and their contents
+are not included in output if you use the `--hidden` option.
+
 ### Deleting files

 You can use `fd` to remove all files and directories that are matched by your search pattern.
@@ -284,7 +286,8 @@ option:
 If you also want to remove a certain class of directories, you can use the same technique. You will
 have to use `rm`s `--recursive`/`-r` flag to remove directories.

-Note: there are scenarios where using `fd … -X rm -r` can cause race conditions: if you have a
+> [!NOTE]
+> There are scenarios where using `fd … -X rm -r` can cause race conditions: if you have a
 path like `…/foo/bar/foo/…` and want to remove all directories named `foo`, you can end up in a
 situation where the outer `foo` directory is removed first, leading to (harmless) *"'foo/bar/foo':
 No such file or directory"* errors in the `rm` call.
@@ -313,7 +316,7 @@ Options:
  -p, --full-path                  Search full abs. path (default: filename only)
  -d, --max-depth <depth>          Set maximum search depth (default: none)
  -E, --exclude <pattern>          Exclude entries that match the given glob pattern
-  -t, --type <filetype>            Filter by type: file (f), directory (d), symlink (l),
+  -t, --type <filetype>            Filter by type: file (f), directory (d/dir), symlink (l),
                                   executable (x), empty (e), socket (s), pipe (p), char-device
                                   (c), block-device (b)
  -e, --extension <ext>            Filter by file extension
@@ -321,10 +324,13 @@ Options:
      --changed-within <date|dur>  Filter by file modification time (newer than)
      --changed-before <date|dur>  Filter by file modification time (older than)
  -o, --owner <user:group>         Filter by owning user and/or group
+      --format <fmt>               Print results according to template
  -x, --exec <cmd>...              Execute a command for each search result
  -X, --exec-batch <cmd>...        Execute a command with all search results at once
  -c, --color <when>               When to use colors [default: auto] [possible values: auto,
                                   always, never]
+      --hyperlink[=<when>]         Add hyperlinks to output paths [default: never] [possible
+                                   values: auto, always, never]
  -h, --help                       Print help (see more with '--help')
  -V, --version                    Print version
 ```
@@ -351,7 +357,7 @@ Benchmark 2: find ~ -iname '*[0-9].jpg'
 ```

 Now let's try the same for `fd`. Note that `fd` performs a regular expression
-search by defautl. The options `-u`/`--unrestricted` option is needed here for
+search by default. The options `-u`/`--unrestricted` option is needed here for
 a fair comparison. Otherwise `fd` does not have to traverse hidden folders and
 ignored paths (see below):
 ```
@@ -373,6 +379,20 @@ also used in [ripgrep](https://github.com/BurntSushi/ripgrep) (check it out!).

 ## Troubleshooting

+### `fd` does not find my file!
+
+Remember that `fd` ignores hidden directories and files by default. It also ignores patterns
+from `.gitignore` files. If you want to make sure to find absolutely every possible file, always
+use the options `-u`/`--unrestricted` option (or `-HI` to enable hidden and ignored files):
+``` bash
+> fd -u …
+```
+
+Also remember that by default, `fd` only searches based on the filename and
+doesn't compare the pattern to the full path. If you want to search based on the
+full path (similar to the `-path` option of `find`) you need to use the `--full-path`
+(or `-p`) option.
+
 ### Colorized output

 `fd` can colorize files by extension, just like `ls`. In order for this to work, the environment
@@ -386,15 +406,6 @@ for alternative, more complete (or more colorful) variants, see [here](https://g

 `fd` also honors the [`NO_COLOR`](https://no-color.org/) environment variable.

-### `fd` does not find my file!
-
-Remember that `fd` ignores hidden directories and files by default. It also ignores patterns
-from `.gitignore` files. If you want to make sure to find absolutely every possible file, always
-use the options `-u`/`--unrestricted` option (or `-HI` to enable hidden and ignored files):
-``` bash
-> fd -u …
-```
-
 ### `fd` doesn't seem to interpret my regex pattern correctly

 A lot of special regex characters (like `[]`, `^`, `$`, ..) are also special characters in your
@@ -517,7 +528,7 @@ newlines). In the same way, the `-0` option of `xargs` tells it to read the inpu
 If you run Ubuntu 19.04 (Disco Dingo) or newer, you can install the
 [officially maintained package](https://packages.ubuntu.com/fd-find):
 ```
-sudo apt install fd-find
+apt install fd-find
 ```
 Note that the binary is called `fdfind` as the binary name `fd` is already used by another package.
 It is recommended that after installation, you add a link to `fd` by executing command
@@ -527,21 +538,25 @@ Make sure that `$HOME/.local/bin` is in your `$PATH`.
 If you use an older version of Ubuntu, you can download the latest `.deb` package from the
 [release page](https://github.com/sharkdp/fd/releases) and install it via:
 ``` bash
-sudo dpkg -i fd_9.0.0_amd64.deb # adapt version number and architecture
+dpkg -i fd_9.0.0_amd64.deb # adapt version number and architecture
 ```

+Note that the .deb packages on the release page for this project still name the executable `fd`.
+
 ### On Debian

 If you run Debian Buster or newer, you can install the
 [officially maintained Debian package](https://tracker.debian.org/pkg/rust-fd-find):
 ```
-sudo apt-get install fd-find
+apt-get install fd-find
 ```
 Note that the binary is called `fdfind` as the binary name `fd` is already used by another package.
 It is recommended that after installation, you add a link to `fd` by executing command
 `ln -s $(which fdfind) ~/.local/bin/fd`, in order to use `fd` in the same way as in this documentation.
 Make sure that `$HOME/.local/bin` is in your `$PATH`.

+Note that the .deb packages on the release page for this project still name the executable `fd`.
+
 ### On Fedora

 Starting with Fedora 28, you can install `fd` from the official package sources:
@@ -559,10 +574,12 @@ apk add fd

 ### On Arch Linux

-You can install [the fd package](https://www.archlinux.org/packages/community/x86_64/fd/) from the official repos:
+You can install [the fd package](https://www.archlinux.org/packages/extra/x86_64/fd/) from the official repos:
 ```
 pacman -S fd
 ```
+You can also install fd [from the AUR](https://aur.archlinux.org/packages/fd-git).
+
 ### On Gentoo Linux

 You can use [the fd ebuild](https://packages.gentoo.org/packages/sys-apps/fd) from the official repo:
@@ -584,6 +601,20 @@ You can install `fd` via xbps-install:
 xbps-install -S fd
 ```

+### On ALT Linux
+
+You can install [the fd package](https://packages.altlinux.org/en/sisyphus/srpms/fd/) from the official repo:
+```
+apt-get install fd
+```
+
+### On Solus
+
+You can install [the fd package](https://github.com/getsolus/packages/tree/main/packages/f/fd) from the official repo:
+```
+eopkg install fd
+```
+
 ### On RedHat Enterprise Linux 8/9 (RHEL8/9), Almalinux 8/9, EuroLinux 8/9 or Rocky Linux 8/9

 You can install [the `fd` package](https://copr.fedorainfracloud.org/coprs/tkbcopr/fd/) from Fedora Copr.
@@ -604,7 +635,7 @@ brew install fd

 … or with MacPorts:
 ```
-sudo port install fd
+port install fd
 ```

 ### On Windows
@@ -640,6 +671,13 @@ You can use the [Nix package manager](https://nixos.org/nix/) to install `fd`:
 nix-env -i fd
 ```

+### Via Flox
+
+You can use [Flox](https://flox.dev) to install `fd` into a Flox environment:
+```
+flox install fd
+```
+
 ### On FreeBSD

 You can install [the fd-find package](https://www.freshports.org/sysutils/fd) from the official repo:
@@ -649,7 +687,7 @@ pkg install fd-find

 ### From npm

-On linux and macOS, you can install the [fd-find](https://npm.im/fd-find) package:
+On Linux and macOS, you can install the [fd-find](https://npm.im/fd-find) package:

 ```
 npm install -g fd-find
@@ -661,7 +699,7 @@ With Rust's package manager [cargo](https://github.com/rust-lang/cargo), you can
 ```
 cargo install fd-find
 ```
-Note that rust version *1.70.0* or later is required.
+Note that rust version *1.77.2* or later is required.

 `make` is also needed for the build.

@@ -692,8 +730,6 @@ cargo install --path .

 ## License

-Copyright (c) 2017-2021 The fd developers
-
 `fd` is distributed under the terms of both the MIT License and the Apache License 2.0.

 See the [LICENSE-APACHE](LICENSE-APACHE) and [LICENSE-MIT](LICENSE-MIT) files for license details.
@@ -5,7 +5,7 @@ fn main() {
        Some(true) => {}
        // rustc version too small or can't figure it out
        _ => {
-            eprintln!("'fd' requires rustc >= {}", min_version);
+            eprintln!("'fd' requires rustc >= {min_version}");
            std::process::exit(1);
        }
    }
@@ -139,6 +139,8 @@ _fd() {
      always\:"always use colorized output"
    ))'

+    '--hyperlink=-[add hyperlinks to output paths]::when:(auto never always)'
+
    + '(threads)'
    {-j+,--threads=}'[set the number of threads for searching and executing]:number of threads'

@@ -162,7 +164,7 @@ _fd() {
    $no'(*)*--search-path=[set search path (instead of positional <path> arguments)]:directory:_files -/'

    + strip-cwd-prefix
-    $no'(strip-cwd-prefix exec-cmds)--strip-cwd-prefix[Strip ./ prefix when output is redirected]'
+    $no'(strip-cwd-prefix exec-cmds)--strip-cwd-prefix=-[When to strip ./]::when:(always never auto)'

    + and
    '--and=[additional required search path]:pattern'
@@ -33,16 +33,14 @@ with the '\-\-glob' option.
 By default
 .B fd
 will exclude hidden files and directories, as well as any files that match gitignore rules
-or ignore rules in .ignore or .fdignore files. For convenenience, '.git' is treated as if it
-was always included in gitignore rules. These files can be included with options such as
-'\-\-hidden' and '\-\-no\-ignore'.
+or ignore rules in .ignore or .fdignore files.
 .SH OPTIONS
 .TP
 .B \-H, \-\-hidden
 Include hidden files and directories in the search results
 (default: hidden files and directories are skipped). The flag can be overridden with '--no-hidden'.
 .IP
-Ignored files and .git/ are still excluded unless \-\-no\-ignore or \-\-no\-ignore\-vcs
+Ignored files are still excluded unless \-\-no\-ignore or \-\-no\-ignore\-vcs
 is also used.
 .TP
 .B \-I, \-\-no\-ignore
@@ -79,7 +77,6 @@ and the global gitignore configuration
 .RI ( core.excludesFile
 git setting, which defaults to
 .IR $HOME/.config/git/ignore ).
-The pattern ".git/" is automatically added to the list of VCS ignore rules.
 The flag can be overridden with '--ignore-vcs'.
 .TP
 .B \-\-no\-require\-git
@@ -159,9 +156,20 @@ can be used as an alias.
 Enable the display of filesystem errors for situations such as insufficient
 permissions or dead symlinks.
 .TP
-.B \-\-strip-cwd-prefix
-By default, relative paths are prefixed with './' when the output goes to a non interactive terminal
-(TTY). Use this flag to disable this behaviour.
+.B \-\-strip-cwd-prefix [when]
+By default, relative paths are prefixed with './' when -x/--exec,
+-X/--exec-batch, or -0/--print0 are given, to reduce the risk of a
+path starting with '-' being treated as a command line option. Use
+this flag to change this behavior. If this flag is used without a value,
+it is equivalent to passing "always". Possible values are:
+.RS
+.IP never
+Never strip the ./ at the beginning of paths
+.IP always
+Always strip the ./ at the beginning of paths
+.IP auto
+Only strip if used with --exec, --exec-batch, or --print0. That is, it resets to the default behavior.
+.RE
 .TP
 .B \-\-one\-file\-system, \-\-mount, \-\-xdev
 By default, fd will traverse the file system tree as far as other options dictate. With this flag, fd ensures that it does not descend into a different file system than the one it started in. Comparable to the -mount or -xdev filters of find(1).
@@ -191,7 +199,7 @@ Filter search by type:
 .RS
 .IP "f, file"
 regular files
-.IP "d, directory"
+.IP "d, dir, directory"
 directories
 .IP "l, symlink"
 symbolic links
@@ -268,6 +276,24 @@ Do not colorize output.
 Always colorize output.
 .RE
 .TP
+.B "\-\-hyperlink
+Specify whether the output should use terminal escape codes to indicate a hyperlink to a
+file url pointing to the path.
+
+The value can be auto, always, or never.
+
+Currently, the default is "never", and if the option is used without an argument "auto" is
+used. In the future this may be changed to "auto" and "always".
+.RS
+.IP auto
+Only output hyperlinks if color is also enabled, as a proxy for whether terminal escape
+codes are acceptable.
+.IP never
+Never output hyperlink escapes.
+.IP always
+Always output hyperlink escapes, regardless of color settings.
+.RE
+.TP
 .BI "\-j, \-\-threads " num
 Set number of threads to use for searching & executing (default: number of available CPU cores).
 .TP
@@ -310,10 +336,11 @@ tebibytes
 .TP
 .BI "\-\-changed-within " date|duration
 Filter results based on the file modification time.
-Files with modification times greater than the argument will be returned.
+Files with modification timestamps greater than the argument will be returned.
 The argument can be provided as a duration (\fI10h, 1d, 35min\fR) or as a specific point
-in time in either full RFC3339 format with time zone, or as a date or datetime in the
-local time zone (\fIYYYY-MM-DD\fR or \fIYYYY-MM-DD HH:MM:SS\fR).
+in time as full RFC3339 format with time zone, as a date or datetime in the
+local time zone (\fIYYYY-MM-DD\fR or \fIYYYY-MM-DD HH:MM:SS\fR), or as the prefix '@'
+followed by the number of seconds since the Unix epoch (@[0-9]+).
 \fB\-\-change-newer-than\fR,
 .B --newer
 or
@@ -324,13 +351,15 @@ Examples:
  \-\-changed-within 2weeks
  \-\-change-newer-than "2018-10-27 10:00:00"
  \-\-newer 2018-10-27
+  \-\-changed-after @1704067200
 .TP
 .BI "\-\-changed-before " date|duration
 Filter results based on the file modification time.
-Files with modification times less than the argument will be returned.
+Files with modification timestamps less than the argument will be returned.
 The argument can be provided as a duration (\fI10h, 1d, 35min\fR) or as a specific point
-in time in either full RFC3339 format with time zone, or as a date or datetime in the
-local time zone (\fIYYYY-MM-DD\fR or \fIYYYY-MM-DD HH:MM:SS\fR).
+in time as full RFC3339 format with time zone, as a date or datetime in the
+local time zone (\fIYYYY-MM-DD\fR or \fIYYYY-MM-DD HH:MM:SS\fR), or as the prefix '@'
+followed by the number of seconds since the Unix epoch (@[0-9]+).
 .B --change-older-than
 or
 .B --older
@@ -339,6 +368,7 @@ can be used as aliases.
 Examples:
  \-\-changed-before "2018-10-27 10:00:00"
  \-\-change-older-than 2weeks
+  \-\-older @1704067200
 .TP
 .BI "-o, \-\-owner " [user][:group]
 Filter files by their user and/or group. Format: [(user|uid)][:(group|gid)]. Either side
@@ -363,6 +393,30 @@ Set the path separator to use when printing file paths. The default is the OS-sp
 Provide paths to search as an alternative to the positional \fIpath\fR argument. Changes the usage to
 \'fd [FLAGS/OPTIONS] \-\-search\-path PATH \-\-search\-path PATH2 [PATTERN]\'
 .TP
+.BI "\-\-format " fmt
+Specify a template string that is used for printing a line for each file found.
+
+The following placeholders are substituted into the string for each file before printing:
+.RS
+.IP {}
+path (of the current search result)
+.IP {/}
+basename
+.IP {//}
+parent directory
+.IP {.}
+path without file extension
+.IP {/.}
+basename without file extension
+.IP {{
+literal '{' (an escape sequence)
+.IP }}
+literal '}' (an escape sequence)
+.P
+Notice that you can use "{{" and "}}" to escape "{" and "}" respectively, which is especially
+useful if you need to include the literal text of one of the above placeholders.
+.RE
+.TP
 .BI "\-x, \-\-exec " command
 .RS
 Execute
@@ -383,29 +437,12 @@ If parallelism is enabled, the order commands will be executed in is non-determi
 --threads=1, the order is determined by the operating system and may not be what you expect. Thus, it is
 recommended that you don't rely on any ordering of the results.

-The following placeholders are substituted before the command is executed:
-.RS
-.IP {}
-path (of the current search result)
-.IP {/}
-basename
-.IP {//}
-parent directory
-.IP {.}
-path without file extension
-.IP {/.}
-basename without file extension
-.IP {{
-literal '{' (an escape sequence)
-.IP }}
-literal '}' (an escape sequence)
-.RE
+Before executing the command, any placeholder patterns in the command are replaced with the
+corresponding values for the current file. The same placeholders are used as in the "\-\-format"
+option.

 If no placeholder is present, an implicit "{}" at the end is assumed.

-Notice that you can use "{{" and "}}" to escape "{" and "}" respectively, which is especially
-useful if you need to include the literal text of one of the above placeholders.
-
 Examples:

  - find all *.zip files and unzip them:
@@ -429,19 +466,9 @@ once, with all search results as arguments.

 The order of the arguments is non-deterministic and should not be relied upon.

-One of the following placeholders is substituted before the command is executed:
-.RS
-.IP {}
-path (of all search results)
-.IP {/}
-basename
-.IP {//}
-parent directory
-.IP {.}
-path without file extension
-.IP {/.}
-basename without file extension
-.RE
+This uses the same placeholders as "\-\-format" and "\-\-exec", but instead of expanding
+once per command invocation each argument containing a placeholder is expanding for every
+file in a batch and passed as separate arguments.

 If no placeholder is present, an implicit "{}" at the end is assumed.

@@ -490,6 +517,17 @@ is set, use
 .IR $XDG_CONFIG_HOME/fd/ignore .
 Otherwise, use
 .IR $HOME/.config/fd/ignore .
+.SH FILES
+.TP
+.B .fdignore
+This file works similarly to a .gitignore file anywhere in the searched tree and specifies patterns
+that should be excluded from the search. However, this file is specific to fd, and will be used even
+if the --no-ignore-vcs option is used.
+.TP
+.B $XDG_CONFIG_HOME/fd/ignore
+Global ignore file. Unless ignore mode is turned off (such as with --no-ignore)
+ignore entries in this file will be ignored, as if it was an .fdignore file in the
+current directory.
 .SH EXAMPLES
 .TP
 .RI "Find files and directories that match the pattern '" needle "':"
@@ -503,6 +541,16 @@ $ fd -e py
 .TP
 .RI "Open all search results with vim:"
 $ fd pattern -X vim
+.SH Tips and Tricks
+.IP \[bu]
+If you add ".git/" to your global ignore file ($XDG_CONFIG_HOME/fd/ignore), then
+".git" folders will be ignored by default, even when the --hidden option is used.
+.IP \[bu]
+You can use a shell alias or a wrapper script in order to pass desired flags to fd
+by default. For example if you do not like the default behavior of respecting gitignore,
+you can use
+`alias fd="/usr/bin/fd --no-ignore-vcs"`
+in your .bashrc to create an alias for fd that doesn't ignore git files by default.
 .SH BUGS
 Bugs can be reported on GitHub: https://github.com/sharkdp/fd/issues
 .SH SEE ALSO
@@ -0,0 +1,12 @@
+## Sponsors
+
+`fd` development is sponsored by many individuals and companies. Thank you very much!
+
+Please note, that being sponsored does not affect the individuality of the `fd`
+project or affect the maintainers' actions in any way.
+We remain impartial and continue to assess pull requests solely on merit - the
+features added, bugs solved, and effect on the overall complexity of the code.
+No issue will have a different priority based on sponsorship status of the
+reporter.
+
+Contributions from anybody are most welcomed, please see our [`CONTRIBUTING.md`](../CONTRIBUTING.md) guide.
@@ -0,0 +1,134 @@
+#!/bin/bash
+COPYRIGHT_YEARS="2018 - "$(date "+%Y")
+MAINTAINER="David Peter <mail@david-peter.de>"
+REPO="https://github.com/sharkdp/fd"
+DPKG_STAGING="${CICD_INTERMEDIATES_DIR:-.}/debian-package"
+DPKG_DIR="${DPKG_STAGING}/dpkg"
+mkdir -p "${DPKG_DIR}"
+
+if [[ -z "$TARGET" ]]; then
+  TARGET="$(rustc -vV | sed -n 's|host: \(.*\)|\1|p')"
+fi
+
+case "$TARGET" in
+  *-musl*)
+    DPKG_BASENAME=fd-musl
+    DPKG_CONFLICTS="fd, fd-find"
+    ;;
+  *)
+    DPKG_BASENAME=fd
+    DPKG_CONFLICTS="fd-musl, fd-find"
+    ;;
+esac
+
+if [[ -z "$DPKG_VERSION" ]]; then
+  DPKG_VERSION=$(cargo metadata --no-deps --format-version 1 | jq -r .packages[0].version)
+fi
+
+unset DPKG_ARCH
+case "${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_NAME="${DPKG_BASENAME}_${DPKG_VERSION}_${DPKG_ARCH}.deb"
+
+BIN_PATH=${BIN_PATH:-target/${TARGET}/release/fd}
+
+# Binary
+install -Dm755 "${BIN_PATH}" "${DPKG_DIR}/usr/bin/fd"
+
+# Man page
+install -Dm644 'doc/fd.1' "${DPKG_DIR}/usr/share/man/man1/fd.1"
+gzip -n --best "${DPKG_DIR}/usr/share/man/man1/fd.1"
+
+# Autocompletion files
+install -Dm644 'autocomplete/fd.bash' "${DPKG_DIR}/usr/share/bash-completion/completions/fd"
+install -Dm644 'autocomplete/fd.fish' "${DPKG_DIR}/usr/share/fish/vendor_completions.d/fd.fish"
+install -Dm644 'autocomplete/_fd' "${DPKG_DIR}/usr/share/zsh/vendor-completions/_fd"
+
+# README and LICENSE
+install -Dm644 "README.md" "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/README.md"
+install -Dm644 "LICENSE-MIT" "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/LICENSE-MIT"
+install -Dm644 "LICENSE-APACHE" "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/LICENSE-APACHE"
+install -Dm644 "CHANGELOG.md" "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/changelog"
+gzip -n --best "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/changelog"
+
+# Create symlinks so fdfind can be used as well:
+ln -s "/usr/bin/fd" "${DPKG_DIR}/usr/bin/fdfind"
+ln -s  './fd.bash' "${DPKG_DIR}/usr/share/bash-completion/completions/fdfind"
+ln -s  './fd.fish' "${DPKG_DIR}/usr/share/fish/vendor_completions.d/fdfind.fish"
+ln -s  './_fd' "${DPKG_DIR}/usr/share/zsh/vendor-completions/_fdfind"
+
+cat > "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/copyright" <<EOF
+Format: http://www.debian.org/doc/packaging-manuals/copyright-format/1.0/
+Upstream-Name: fd
+Source: ${REPO}
+
+Files: *
+Copyright: ${MAINTAINER}
+Copyright: $COPYRIGHT_YEARS ${MAINTAINER}
+License: Apache-2.0 or MIT
+
+License: Apache-2.0
+  On Debian systems, the complete text of the Apache-2.0 can be found in the
+  file /usr/share/common-licenses/Apache-2.0.
+
+License: MIT
+  Permission is hereby granted, free of charge, to any
+  person obtaining a copy of this software and associated
+  documentation files (the "Software"), to deal in the
+  Software without restriction, including without
+  limitation the rights to use, copy, modify, merge,
+  publish, distribute, sublicense, and/or sell copies of
+  the Software, and to permit persons to whom the Software
+  is furnished to do so, subject to the following
+  conditions:
+  .
+  The above copyright notice and this permission notice
+  shall be included in all copies or substantial portions
+  of the Software.
+  .
+  THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF
+  ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED
+  TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A
+  PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
+  SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
+  CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+  OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR
+  IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+  DEALINGS IN THE SOFTWARE.
+EOF
+  chmod 644 "${DPKG_DIR}/usr/share/doc/${DPKG_BASENAME}/copyright"
+
+  # control file
+  mkdir -p "${DPKG_DIR}/DEBIAN"
+  cat > "${DPKG_DIR}/DEBIAN/control" <<EOF
+Package: ${DPKG_BASENAME}
+Version: ${DPKG_VERSION}
+Section: utils
+Priority: optional
+Maintainer: ${MAINTAINER}
+Homepage: ${REPO}
+Architecture: ${DPKG_ARCH}
+Provides: fd
+Conflicts: ${DPKG_CONFLICTS}
+Description: simple, fast and user-friendly alternative to find
+  fd is a program to find entries in your filesystem.
+  It is a simple, fast and user-friendly alternative to find.
+  While it does not aim to support all of finds powerful functionality, it provides
+  sensible (opinionated) defaults for a majority of use cases.
+EOF
+
+DPKG_PATH="${DPKG_STAGING}/${DPKG_NAME}"
+
+if [[ -n $GITHUB_OUTPUT ]]; then
+  echo "DPKG_NAME=${DPKG_NAME}" >> "$GITHUB_OUTPUT"
+  echo "DPKG_PATH=${DPKG_PATH}" >> "$GITHUB_OUTPUT"
+fi
+
+# build dpkg
+fakeroot dpkg-deb --build "${DPKG_DIR}" "${DPKG_PATH}"
@@ -0,0 +1,22 @@
+#!/usr/bin/bash
+
+set -eu
+
+# This script automates the "Version bump" section
+
+version="$1"
+
+if [[ -z $version ]]; then
+  echo "Usage: must supply version as first argument" >&2
+  exit 1
+fi
+
+git switch -C "release-$version"
+sed -i -e "0,/^\[badges/{s/^version =.*/version = \"$version\"/}" Cargo.toml
+
+msrv="$(grep -F rust-version Cargo.toml | sed -e 's/^rust-version= "\(.*\)"/\1/')"
+
+sed -i -e "s/Note that rust version \*[0-9.]+\* or later/Note that rust version *$msrv* or later/" README.md
+
+sed -i -e "s/^# Upcoming release/# $version/" CHANGELOG.md
+
@@ -49,8 +49,7 @@ pub struct Opts {
    no_hidden: (),

    /// Show search results from files and directories that would otherwise be
-    /// ignored by '.gitignore', '.ignore', '.fdignore', the global ignore file,
-    /// or the default rule that excludes .git/.
+    /// ignored by '.gitignore', '.ignore', '.fdignore', or the global ignore file,
    /// The flag can be overridden with --ignore.
    #[arg(
        long,
@@ -64,7 +63,7 @@ pub struct Opts {
    #[arg(long, overrides_with = "no_ignore", hide = true, action = ArgAction::SetTrue)]
    ignore: (),

-    ///Show search results from '.git/' folders and files and directories that
+    ///Show search results from files and directories that
    ///would otherwise be ignored by '.gitignore' files.
    ///The flag can be overridden with --ignore-vcs.
    #[arg(
@@ -276,6 +275,7 @@ pub struct Opts {
        long,
        value_name = "depth",
        hide_short_help = true,
+        alias("mindepth"),
        help = "Only show search results starting at the given depth.",
        long_help
    )]
@@ -314,7 +314,7 @@ pub struct Opts {

    /// Filter the search by type:
    /// {n}  'f' or 'file':         regular files
-    /// {n}  'd' or 'directory':    directories
+    /// {n}  'd' or 'dir' or 'directory':    directories
    /// {n}  'l' or 'symlink':      symbolic links
    /// {n}  's' or 'socket':       socket
    /// {n}  'p' or 'pipe':         named pipe (FIFO)
@@ -352,7 +352,7 @@ pub struct Opts {
        value_name = "filetype",
        hide_possible_values = true,
        value_enum,
-        help = "Filter by type: file (f), directory (d), symlink (l), \
+        help = "Filter by type: file (f), directory (d/dir), symlink (l), \
                executable (x), empty (e), socket (s), pipe (p), \
                char-device (c), block-device (b)",
        long_help
@@ -399,7 +399,7 @@ pub struct Opts {

    /// Filter results based on the file modification time. Files with modification times
    /// greater than the argument are returned. The argument can be provided
-    /// as a specific point in time (YYYY-MM-DD HH:MM:SS) or as a duration (10h, 1d, 35min).
+    /// as a specific point in time (YYYY-MM-DD HH:MM:SS or @timestamp) or as a duration (10h, 1d, 35min).
    /// If the time is not specified, it defaults to 00:00:00.
    /// '--change-newer-than', '--newer', or '--changed-after' can be used as aliases.
    ///
@@ -421,7 +421,7 @@ pub struct Opts {

    /// Filter results based on the file modification time. Files with modification times
    /// less than the argument are returned. The argument can be provided
-    /// as a specific point in time (YYYY-MM-DD HH:MM:SS) or as a duration (10h, 1d, 35min).
+    /// as a specific point in time (YYYY-MM-DD HH:MM:SS or @timestamp) or as a duration (10h, 1d, 35min).
    /// '--change-older-than' or '--older' can be used as aliases.
    ///
    /// Examples:
@@ -453,6 +453,20 @@ pub struct Opts {
        )]
    pub owner: Option<OwnerFilter>,

+    /// Instead of printing the file normally, print the format string with the following placeholders replaced:
+    ///   '{}': path (of the current search result)
+    ///   '{/}': basename
+    ///   '{//}': parent directory
+    ///   '{.}': path without file extension
+    ///   '{/.}': basename without file extension
+    #[arg(
+        long,
+        value_name = "fmt",
+        help = "Print results according to template",
+        conflicts_with = "list_details"
+    )]
+    pub format: Option<String>,
+
    #[command(flatten)]
    pub exec: Exec,

@@ -496,6 +510,24 @@ pub struct Opts {
    )]
    pub color: ColorWhen,

+    /// Add a terminal hyperlink to a file:// url for each path in the output.
+    ///
+    /// Auto mode  is used if no argument is given to this option.
+    ///
+    /// This doesn't do anything for --exec and --exec-batch.
+    #[arg(
+        long,
+        alias = "hyper",
+        value_name = "when",
+        require_equals = true,
+        value_enum,
+        default_value_t = HyperlinkWhen::Never,
+        default_missing_value = "auto",
+        num_args = 0..=1,
+        help = "Add hyperlinks to output paths"
+    )]
+    pub hyperlink: HyperlinkWhen,
+
    /// Set number of threads to use for searching & executing (default: number
    /// of available CPU cores)
    #[arg(long, short = 'j', value_name = "num", hide_short_help = true, value_parser = str::parse::<NonZeroUsize>)]
@@ -618,9 +650,10 @@ pub struct Opts {
    /// By default, relative paths are prefixed with './' when -x/--exec,
    /// -X/--exec-batch, or -0/--print0 are given, to reduce the risk of a
    /// path starting with '-' being treated as a command line option. Use
-    /// this flag to disable this behaviour.
-    #[arg(long, conflicts_with_all(&["path", "search_path"]), hide_short_help = true, long_help)]
-    pub strip_cwd_prefix: bool,
+    /// this flag to change this behavior. If this flag is used without a value,
+    /// it is equivalent to passing "always".
+    #[arg(long, conflicts_with_all(&["path", "search_path"]), value_name = "when", hide_short_help = true, require_equals = true, long_help)]
+    strip_cwd_prefix: Option<Option<StripCwdWhen>>,

    /// By default, fd will traverse the file system tree as far as other options
    /// dictate. With this flag, fd ensures that it does not descend into a
@@ -643,7 +676,7 @@ impl Opts {
        } else if !self.search_path.is_empty() {
            &self.search_path
        } else {
-            let current_directory = Path::new(".");
+            let current_directory = Path::new("./");
            ensure_current_directory_exists(current_directory)?;
            return Ok(vec![self.normalize_path(current_directory)]);
        };
@@ -666,6 +699,9 @@ impl Opts {
    fn normalize_path(&self, path: &Path) -> PathBuf {
        if self.absolute_path {
            filesystem::absolute_path(path.normalize().unwrap().as_path()).unwrap()
+        } else if path == Path::new(".") {
+            // Change "." to "./" as a workaround for https://github.com/BurntSushi/ripgrep/pull/2711
+            PathBuf::from("./")
        } else {
            path.to_path_buf()
        }
@@ -698,6 +734,16 @@ impl Opts {
            .or_else(|| self.max_one_result.then_some(1))
    }

+    pub fn strip_cwd_prefix<P: FnOnce() -> bool>(&self, auto_pred: P) -> bool {
+        use self::StripCwdWhen::*;
+        self.no_search_paths()
+            && match self.strip_cwd_prefix.map_or(Auto, |o| o.unwrap_or(Always)) {
+                Auto => auto_pred(),
+                Always => true,
+                Never => false,
+            }
+    }
+
    #[cfg(feature = "completions")]
    pub fn gen_completions(&self) -> anyhow::Result<Option<Shell>> {
        self.gen_completions
@@ -729,7 +775,7 @@ fn default_num_threads() -> NonZeroUsize {
 pub enum FileType {
    #[value(alias = "f")]
    File,
-    #[value(alias = "d")]
+    #[value(alias = "d", alias = "dir")]
    Directory,
    #[value(alias = "l")]
    Symlink,
@@ -758,15 +804,24 @@ pub enum ColorWhen {
    Never,
 }

-impl ColorWhen {
-    pub fn as_str(&self) -> &'static str {
-        use ColorWhen::*;
-        match *self {
-            Auto => "auto",
-            Never => "never",
-            Always => "always",
-        }
-    }
+#[derive(Copy, Clone, PartialEq, Eq, Debug, ValueEnum)]
+pub enum StripCwdWhen {
+    /// Use the default behavior
+    Auto,
+    /// Always strip the ./ at the beginning of paths
+    Always,
+    /// Never strip the ./
+    Never,
+}
+
+#[derive(Copy, Clone, PartialEq, Eq, Debug, ValueEnum)]
+pub enum HyperlinkWhen {
+    /// Use hyperlinks only if color is enabled
+    Auto,
+    /// Always use hyperlinks when printing file paths
+    Always,
+    /// Never use hyperlinks
+    Never,
 }

 // there isn't a derive api for getting grouped values yet,
@@ -8,6 +8,7 @@ use crate::filetypes::FileTypes;
 #[cfg(unix)]
 use crate::filter::OwnerFilter;
 use crate::filter::{SizeFilter, TimeFilter};
+use crate::fmt::FormatTemplate;

 /// Configuration options for *fd*.
 pub struct Config {
@@ -74,6 +75,7 @@ pub struct Config {
    pub ls_colors: Option<LsColors>,

    /// Whether or not we are writing to an interactive terminal
+    #[cfg_attr(not(unix), allow(unused))]
    pub interactive_terminal: bool,

    /// The type of file to search for. If set to `None`, all file types are displayed. If
@@ -85,6 +87,9 @@ pub struct Config {
    /// The value (if present) will be a lowercase string without leading dots.
    pub extensions: Option<RegexSet>,

+    /// A format string to use to format results, similarly to exec
+    pub format: Option<FormatTemplate>,
+
    /// If a value is supplied, each item found will be used to generate and execute commands.
    pub command: Option<Arc<CommandSet>>,

@@ -122,6 +127,9 @@ pub struct Config {

    /// Whether or not to strip the './' prefix for search results
    pub strip_cwd_prefix: bool,
+
+    /// Whether or not to use hyperlinks on paths
+    pub hyperlink: bool,
 }

 impl Config {
@@ -113,7 +113,7 @@ impl Eq for DirEntry {}
 impl PartialOrd for DirEntry {
    #[inline]
    fn partial_cmp(&self, other: &Self) -> Option<std::cmp::Ordering> {
-        self.path().partial_cmp(other.path())
+        Some(self.cmp(other))
    }
 }

@@ -137,7 +137,7 @@ impl Colorable for DirEntry {
                // but we want it for all component types, so we open code it.
                // Copied from LsColors::style_for_path_with_metadata().
                path.components()
-                    .last()
+                    .next_back()
                    .map(|c| c.as_os_str())
                    .unwrap_or_else(|| path.as_os_str())
            }
@@ -103,7 +103,7 @@ pub fn handle_cmd_error(cmd: Option<&Command>, err: io::Error) -> ExitCode {
            ExitCode::GeneralError
        }
        (_, err) => {
-            print_error(format!("Problem while executing command: {}", err));
+            print_error(format!("Problem while executing command: {err}"));
            ExitCode::GeneralError
        }
    }
@@ -1,13 +1,10 @@
 mod command;
-mod input;
 mod job;
-mod token;

-use std::borrow::Cow;
-use std::ffi::{OsStr, OsString};
+use std::ffi::OsString;
 use std::io;
 use std::iter;
-use std::path::{Component, Path, PathBuf, Prefix};
+use std::path::{Path, PathBuf};
 use std::process::Stdio;
 use std::sync::Mutex;

@@ -15,11 +12,10 @@ use anyhow::{bail, Result};
 use argmax::Command;

 use crate::exit_codes::{merge_exitcodes, ExitCode};
+use crate::fmt::{FormatTemplate, Token};

 use self::command::{execute_commands, handle_cmd_error};
-use self::input::{basename, dirname, remove_extension};
 pub use self::job::{batch, job};
-use self::token::{tokenize, Token};

 /// Execution mode of the command
 #[derive(Debug, Clone, Copy, PartialEq, Eq)]
@@ -131,7 +127,7 @@ impl CommandSet {
 #[derive(Debug)]
 struct CommandBuilder {
    pre_args: Vec<OsString>,
-    path_arg: ArgumentTemplate,
+    path_arg: FormatTemplate,
    post_args: Vec<OsString>,
    cmd: Command,
    count: usize,
@@ -220,7 +216,7 @@ impl CommandBuilder {
 /// `generate_and_execute()` method will be used to generate a command and execute it.
 #[derive(Debug, Clone, PartialEq)]
 struct CommandTemplate {
-    args: Vec<ArgumentTemplate>,
+    args: Vec<FormatTemplate>,
 }

 impl CommandTemplate {
@@ -235,7 +231,7 @@ impl CommandTemplate {
        for arg in input {
            let arg = arg.as_ref();

-            let tmpl = tokenize(arg);
+            let tmpl = FormatTemplate::parse(arg);
            has_placeholder |= tmpl.has_tokens();
            args.push(tmpl);
        }
@@ -251,7 +247,7 @@ impl CommandTemplate {

        // If a placeholder token was not supplied, append one at the end of the command.
        if !has_placeholder {
-            args.push(ArgumentTemplate::Tokens(vec![Token::Placeholder]));
+            args.push(FormatTemplate::Tokens(vec![Token::Placeholder]));
        }

        Ok(CommandTemplate { args })
@@ -274,111 +270,6 @@ impl CommandTemplate {
    }
 }

-/// Represents a template for a single command argument.
-///
-/// The argument is either a collection of `Token`s including at least one placeholder variant, or
-/// a fixed text.
-#[derive(Clone, Debug, PartialEq)]
-enum ArgumentTemplate {
-    Tokens(Vec<Token>),
-    Text(String),
-}
-
-impl ArgumentTemplate {
-    pub fn has_tokens(&self) -> bool {
-        matches!(self, ArgumentTemplate::Tokens(_))
-    }
-
-    /// Generate an argument from this template. If path_separator is Some, then it will replace
-    /// the path separator in all placeholder tokens. Text arguments and tokens are not affected by
-    /// path separator substitution.
-    pub fn generate(&self, path: impl AsRef<Path>, path_separator: Option<&str>) -> OsString {
-        use self::Token::*;
-        let path = path.as_ref();
-
-        match *self {
-            ArgumentTemplate::Tokens(ref tokens) => {
-                let mut s = OsString::new();
-                for token in tokens {
-                    match *token {
-                        Basename => s.push(Self::replace_separator(basename(path), path_separator)),
-                        BasenameNoExt => s.push(Self::replace_separator(
-                            &remove_extension(basename(path).as_ref()),
-                            path_separator,
-                        )),
-                        NoExt => s.push(Self::replace_separator(
-                            &remove_extension(path),
-                            path_separator,
-                        )),
-                        Parent => s.push(Self::replace_separator(&dirname(path), path_separator)),
-                        Placeholder => {
-                            s.push(Self::replace_separator(path.as_ref(), path_separator))
-                        }
-                        Text(ref string) => s.push(string),
-                    }
-                }
-                s
-            }
-            ArgumentTemplate::Text(ref text) => OsString::from(text),
-        }
-    }
-
-    /// Replace the path separator in the input with the custom separator string. If path_separator
-    /// is None, simply return a borrowed Cow<OsStr> of the input. Otherwise, the input is
-    /// interpreted as a Path and its components are iterated through and re-joined into a new
-    /// OsString.
-    fn replace_separator<'a>(path: &'a OsStr, path_separator: Option<&str>) -> Cow<'a, OsStr> {
-        // fast-path - no replacement necessary
-        if path_separator.is_none() {
-            return Cow::Borrowed(path);
-        }
-
-        let path_separator = path_separator.unwrap();
-        let mut out = OsString::with_capacity(path.len());
-        let mut components = Path::new(path).components().peekable();
-
-        while let Some(comp) = components.next() {
-            match comp {
-                // Absolute paths on Windows are tricky.  A Prefix component is usually a drive
-                // letter or UNC path, and is usually followed by RootDir. There are also
-                // "verbatim" prefixes beginning with "\\?\" that skip normalization. We choose to
-                // ignore verbatim path prefixes here because they're very rare, might be
-                // impossible to reach here, and there's no good way to deal with them. If users
-                // are doing something advanced involving verbatim windows paths, they can do their
-                // own output filtering with a tool like sed.
-                Component::Prefix(prefix) => {
-                    if let Prefix::UNC(server, share) = prefix.kind() {
-                        // Prefix::UNC is a parsed version of '\\server\share'
-                        out.push(path_separator);
-                        out.push(path_separator);
-                        out.push(server);
-                        out.push(path_separator);
-                        out.push(share);
-                    } else {
-                        // All other Windows prefix types are rendered as-is. This results in e.g. "C:" for
-                        // drive letters. DeviceNS and Verbatim* prefixes won't have backslashes converted,
-                        // but they're not returned by directories fd can search anyway so we don't worry
-                        // about them.
-                        out.push(comp.as_os_str());
-                    }
-                }
-
-                // Root directory is always replaced with the custom separator.
-                Component::RootDir => out.push(path_separator),
-
-                // Everything else is joined normally, with a trailing separator if we're not last
-                _ => {
-                    out.push(comp.as_os_str());
-                    if components.peek().is_some() {
-                        out.push(path_separator);
-                    }
-                }
-            }
-        }
-        Cow::Owned(out)
-    }
-}
-
 #[cfg(test)]
 mod tests {
    use super::*;
@@ -398,9 +289,9 @@ mod tests {
            CommandSet {
                commands: vec![CommandTemplate {
                    args: vec![
-                        ArgumentTemplate::Text("echo".into()),
-                        ArgumentTemplate::Text("${SHELL}:".into()),
-                        ArgumentTemplate::Tokens(vec![Token::Placeholder]),
+                        FormatTemplate::Text("echo".into()),
+                        FormatTemplate::Text("${SHELL}:".into()),
+                        FormatTemplate::Tokens(vec![Token::Placeholder]),
                    ]
                }],
                mode: ExecutionMode::OneByOne,
@@ -415,8 +306,8 @@ mod tests {
            CommandSet {
                commands: vec![CommandTemplate {
                    args: vec![
-                        ArgumentTemplate::Text("echo".into()),
-                        ArgumentTemplate::Tokens(vec![Token::NoExt]),
+                        FormatTemplate::Text("echo".into()),
+                        FormatTemplate::Tokens(vec![Token::NoExt]),
                    ],
                }],
                mode: ExecutionMode::OneByOne,
@@ -431,8 +322,8 @@ mod tests {
            CommandSet {
                commands: vec![CommandTemplate {
                    args: vec![
-                        ArgumentTemplate::Text("echo".into()),
-                        ArgumentTemplate::Tokens(vec![Token::Basename]),
+                        FormatTemplate::Text("echo".into()),
+                        FormatTemplate::Tokens(vec![Token::Basename]),
                    ],
                }],
                mode: ExecutionMode::OneByOne,
@@ -447,8 +338,8 @@ mod tests {
            CommandSet {
                commands: vec![CommandTemplate {
                    args: vec![
-                        ArgumentTemplate::Text("echo".into()),
-                        ArgumentTemplate::Tokens(vec![Token::Parent]),
+                        FormatTemplate::Text("echo".into()),
+                        FormatTemplate::Tokens(vec![Token::Parent]),
                    ],
                }],
                mode: ExecutionMode::OneByOne,
@@ -463,8 +354,8 @@ mod tests {
            CommandSet {
                commands: vec![CommandTemplate {
                    args: vec![
-                        ArgumentTemplate::Text("echo".into()),
-                        ArgumentTemplate::Tokens(vec![Token::BasenameNoExt]),
+                        FormatTemplate::Text("echo".into()),
+                        FormatTemplate::Tokens(vec![Token::BasenameNoExt]),
                    ],
                }],
                mode: ExecutionMode::OneByOne,
@@ -494,9 +385,9 @@ mod tests {
            CommandSet {
                commands: vec![CommandTemplate {
                    args: vec![
-                        ArgumentTemplate::Text("cp".into()),
-                        ArgumentTemplate::Tokens(vec![Token::Placeholder]),
-                        ArgumentTemplate::Tokens(vec![
+                        FormatTemplate::Text("cp".into()),
+                        FormatTemplate::Tokens(vec![Token::Placeholder]),
+                        FormatTemplate::Tokens(vec![
                            Token::BasenameNoExt,
                            Token::Text(".ext".into())
                        ]),
@@ -514,8 +405,8 @@ mod tests {
            CommandSet {
                commands: vec![CommandTemplate {
                    args: vec![
-                        ArgumentTemplate::Text("echo".into()),
-                        ArgumentTemplate::Tokens(vec![Token::NoExt]),
+                        FormatTemplate::Text("echo".into()),
+                        FormatTemplate::Tokens(vec![Token::NoExt]),
                    ],
                }],
                mode: ExecutionMode::Batch,
@@ -540,7 +431,7 @@ mod tests {

    #[test]
    fn generate_custom_path_separator() {
-        let arg = ArgumentTemplate::Tokens(vec![Token::Placeholder]);
+        let arg = FormatTemplate::Tokens(vec![Token::Placeholder]);
        macro_rules! check {
            ($input:expr, $expected:expr) => {
                assert_eq!(arg.generate($input, Some("#")), OsString::from($expected));
@@ -555,7 +446,7 @@ mod tests {
    #[cfg(windows)]
    #[test]
    fn generate_custom_path_separator_windows() {
-        let arg = ArgumentTemplate::Tokens(vec![Token::Placeholder]);
+        let arg = FormatTemplate::Tokens(vec![Token::Placeholder]);
        macro_rules! check {
            ($input:expr, $expected:expr) => {
                assert_eq!(arg.generate($input, Some("#")), OsString::from($expected));
@@ -1,98 +0,0 @@
-use aho_corasick::AhoCorasick;
-use std::fmt::{self, Display, Formatter};
-use std::sync::OnceLock;
-
-use super::ArgumentTemplate;
-
-/// Designates what should be written to a buffer
-///
-/// Each `Token` contains either text, or a placeholder variant, which will be used to generate
-/// commands after all tokens for a given command template have been collected.
-#[derive(Clone, Debug, PartialEq, Eq)]
-pub enum Token {
-    Placeholder,
-    Basename,
-    Parent,
-    NoExt,
-    BasenameNoExt,
-    Text(String),
-}
-
-impl Display for Token {
-    fn fmt(&self, f: &mut Formatter) -> fmt::Result {
-        match *self {
-            Token::Placeholder => f.write_str("{}")?,
-            Token::Basename => f.write_str("{/}")?,
-            Token::Parent => f.write_str("{//}")?,
-            Token::NoExt => f.write_str("{.}")?,
-            Token::BasenameNoExt => f.write_str("{/.}")?,
-            Token::Text(ref string) => f.write_str(string)?,
-        }
-        Ok(())
-    }
-}
-
-static PLACEHOLDERS: OnceLock<AhoCorasick> = OnceLock::new();
-
-pub(super) fn tokenize(input: &str) -> ArgumentTemplate {
-    // NOTE: we assume that { and } have the same length
-    const BRACE_LEN: usize = '{'.len_utf8();
-    let mut tokens = Vec::new();
-    let mut remaining = input;
-    let mut buf = String::new();
-    let placeholders = PLACEHOLDERS.get_or_init(|| {
-        AhoCorasick::new(&["{{", "}}", "{}", "{/}", "{//}", "{.}", "{/.}"]).unwrap()
-    });
-    while let Some(m) = placeholders.find(remaining) {
-        match m.pattern().as_u32() {
-            0 | 1 => {
-                // we found an escaped {{ or }}, so add
-                // everything up to the first char to the buffer
-                // then skip the second one.
-                buf += &remaining[..m.start() + BRACE_LEN];
-                remaining = &remaining[m.end()..];
-            }
-            id if !remaining[m.end()..].starts_with('}') => {
-                buf += &remaining[..m.start()];
-                if !buf.is_empty() {
-                    tokens.push(Token::Text(std::mem::take(&mut buf)));
-                }
-                tokens.push(token_from_pattern_id(id));
-                remaining = &remaining[m.end()..];
-            }
-            _ => {
-                // We got a normal pattern, but the final "}"
-                // is escaped, so add up to that to the buffer, then
-                // skip the final }
-                buf += &remaining[..m.end()];
-                remaining = &remaining[m.end() + BRACE_LEN..];
-            }
-        }
-    }
-    // Add the rest of the string to the buffer, and add the final buffer to the tokens
-    if !remaining.is_empty() {
-        buf += remaining;
-    }
-    if tokens.is_empty() {
-        // No placeholders were found, so just return the text
-        return ArgumentTemplate::Text(buf);
-    }
-    // Add final text segment
-    if !buf.is_empty() {
-        tokens.push(Token::Text(buf));
-    }
-    debug_assert!(!tokens.is_empty());
-    ArgumentTemplate::Tokens(tokens)
-}
-
-fn token_from_pattern_id(id: u32) -> Token {
-    use Token::*;
-    match id {
-        2 => Placeholder,
-        3 => Basename,
-        4 => Parent,
-        5 => NoExt,
-        6 => BasenameNoExt,
-        _ => unreachable!(),
-    }
-}
@@ -128,13 +128,11 @@ pub fn strip_current_dir(path: &Path) -> &Path {
 pub fn default_path_separator() -> Option<String> {
    if cfg!(windows) {
        let msystem = env::var("MSYSTEM").ok()?;
-        match msystem.as_str() {
-            "MINGW64" | "MINGW32" | "MSYS" => Some("/".to_owned()),
-            _ => None,
+        if !msystem.is_empty() {
+            return Some("/".to_owned());
        }
-    } else {
-        None
    }
+    None
 }

 #[cfg(test)]
@@ -1,6 +1,6 @@
-use chrono::{DateTime, Local, NaiveDate, NaiveDateTime};
+use jiff::{civil::DateTime, tz::TimeZone, Span, Timestamp, Zoned};

-use std::time::SystemTime;
+use std::time::{Duration, SystemTime, UNIX_EPOCH};

 /// Filter based on time ranges.
 #[derive(Debug, PartialEq, Eq)]
@@ -9,38 +9,49 @@ pub enum TimeFilter {
    After(SystemTime),
 }

+#[cfg(not(test))]
+fn now() -> Zoned {
+    Zoned::now()
+}
+
+#[cfg(test)]
+thread_local! {
+    static TESTTIME: std::cell::RefCell<Option<Zoned>> = None.into();
+}
+
+/// This allows us to set a specific time when running tests
+#[cfg(test)]
+fn now() -> Zoned {
+    TESTTIME.with_borrow(|reftime| reftime.as_ref().cloned().unwrap_or_else(Zoned::now))
+}
+
 impl TimeFilter {
-    fn from_str(ref_time: &SystemTime, s: &str) -> Option<SystemTime> {
-        humantime::parse_duration(s)
-            .map(|duration| *ref_time - duration)
-            .ok()
-            .or_else(|| {
-                DateTime::parse_from_rfc3339(s)
-                    .map(|dt| dt.into())
-                    .ok()
-                    .or_else(|| {
-                        NaiveDate::parse_from_str(s, "%F")
-                            .ok()?
-                            .and_hms_opt(0, 0, 0)?
-                            .and_local_timezone(Local)
-                            .latest()
-                    })
-                    .or_else(|| {
-                        NaiveDateTime::parse_from_str(s, "%F %T")
-                            .ok()?
-                            .and_local_timezone(Local)
-                            .latest()
-                    })
-                    .map(|dt| dt.into())
-            })
+    fn from_str(s: &str) -> Option<SystemTime> {
+        if let Ok(span) = s.parse::<Span>() {
+            let datetime = now().checked_sub(span).ok()?;
+            Some(datetime.into())
+        } else if let Ok(timestamp) = s.parse::<Timestamp>() {
+            Some(timestamp.into())
+        } else if let Ok(datetime) = s.parse::<DateTime>() {
+            Some(
+                TimeZone::system()
+                    .to_ambiguous_zoned(datetime)
+                    .later()
+                    .ok()?
+                    .into(),
+            )
+        } else {
+            let timestamp_secs: u64 = s.strip_prefix('@')?.parse().ok()?;
+            Some(UNIX_EPOCH + Duration::from_secs(timestamp_secs))
+        }
    }

-    pub fn before(ref_time: &SystemTime, s: &str) -> Option<TimeFilter> {
-        TimeFilter::from_str(ref_time, s).map(TimeFilter::Before)
+    pub fn before(s: &str) -> Option<TimeFilter> {
+        TimeFilter::from_str(s).map(TimeFilter::Before)
    }

-    pub fn after(ref_time: &SystemTime, s: &str) -> Option<TimeFilter> {
-        TimeFilter::from_str(ref_time, s).map(TimeFilter::After)
+    pub fn after(s: &str) -> Option<TimeFilter> {
+        TimeFilter::from_str(s).map(TimeFilter::After)
    }

    pub fn applies_to(&self, t: &SystemTime) -> bool {
@@ -56,84 +67,116 @@ mod tests {
    use super::*;
    use std::time::Duration;

+    struct TestTime(SystemTime);
+
+    impl TestTime {
+        fn new(time: Zoned) -> Self {
+            TESTTIME.with_borrow_mut(|t| *t = Some(time.clone()));
+            TestTime(time.into())
+        }
+
+        fn set(&mut self, time: Zoned) {
+            TESTTIME.with_borrow_mut(|t| *t = Some(time.clone()));
+            self.0 = time.into();
+        }
+
+        fn timestamp(&self) -> SystemTime {
+            self.0
+        }
+    }
+
+    impl Drop for TestTime {
+        fn drop(&mut self) {
+            // Stop using manually set times
+            TESTTIME.with_borrow_mut(|t| *t = None);
+        }
+    }
+
    #[test]
    fn is_time_filter_applicable() {
-        let ref_time = NaiveDateTime::parse_from_str("2010-10-10 10:10:10", "%F %T")
-            .unwrap()
-            .and_local_timezone(Local)
-            .latest()
-            .unwrap()
-            .into();
+        let local_tz = TimeZone::system();
+        let mut test_time = TestTime::new(
+            local_tz
+                .to_ambiguous_zoned("2010-10-10 10:10:10".parse::<DateTime>().unwrap())
+                .later()
+                .unwrap(),
+        );
+        let mut ref_time = test_time.timestamp();

-        assert!(TimeFilter::after(&ref_time, "1min")
-            .unwrap()
-            .applies_to(&ref_time));
-        assert!(!TimeFilter::before(&ref_time, "1min")
-            .unwrap()
-            .applies_to(&ref_time));
+        assert!(TimeFilter::after("1min").unwrap().applies_to(&ref_time));
+        assert!(!TimeFilter::before("1min").unwrap().applies_to(&ref_time));

        let t1m_ago = ref_time - Duration::from_secs(60);
-        assert!(!TimeFilter::after(&ref_time, "30sec")
-            .unwrap()
-            .applies_to(&t1m_ago));
-        assert!(TimeFilter::after(&ref_time, "2min")
-            .unwrap()
-            .applies_to(&t1m_ago));
+        assert!(!TimeFilter::after("30sec").unwrap().applies_to(&t1m_ago));
+        assert!(TimeFilter::after("2min").unwrap().applies_to(&t1m_ago));

-        assert!(TimeFilter::before(&ref_time, "30sec")
-            .unwrap()
-            .applies_to(&t1m_ago));
-        assert!(!TimeFilter::before(&ref_time, "2min")
-            .unwrap()
-            .applies_to(&t1m_ago));
+        assert!(TimeFilter::before("30sec").unwrap().applies_to(&t1m_ago));
+        assert!(!TimeFilter::before("2min").unwrap().applies_to(&t1m_ago));

        let t10s_before = "2010-10-10 10:10:00";
-        assert!(!TimeFilter::before(&ref_time, t10s_before)
+        assert!(!TimeFilter::before(t10s_before)
            .unwrap()
            .applies_to(&ref_time));
-        assert!(TimeFilter::before(&ref_time, t10s_before)
+        assert!(TimeFilter::before(t10s_before)
            .unwrap()
            .applies_to(&t1m_ago));

-        assert!(TimeFilter::after(&ref_time, t10s_before)
+        assert!(TimeFilter::after(t10s_before)
            .unwrap()
            .applies_to(&ref_time));
-        assert!(!TimeFilter::after(&ref_time, t10s_before)
-            .unwrap()
-            .applies_to(&t1m_ago));
+        assert!(!TimeFilter::after(t10s_before).unwrap().applies_to(&t1m_ago));

        let same_day = "2010-10-10";
-        assert!(!TimeFilter::before(&ref_time, same_day)
-            .unwrap()
-            .applies_to(&ref_time));
-        assert!(!TimeFilter::before(&ref_time, same_day)
-            .unwrap()
-            .applies_to(&t1m_ago));
+        assert!(!TimeFilter::before(same_day).unwrap().applies_to(&ref_time));
+        assert!(!TimeFilter::before(same_day).unwrap().applies_to(&t1m_ago));

-        assert!(TimeFilter::after(&ref_time, same_day)
-            .unwrap()
-            .applies_to(&ref_time));
-        assert!(TimeFilter::after(&ref_time, same_day)
-            .unwrap()
-            .applies_to(&t1m_ago));
+        assert!(TimeFilter::after(same_day).unwrap().applies_to(&ref_time));
+        assert!(TimeFilter::after(same_day).unwrap().applies_to(&t1m_ago));

-        let ref_time = DateTime::parse_from_rfc3339("2010-10-10T10:10:10+00:00")
-            .unwrap()
-            .into();
+        test_time.set(
+            "2010-10-10T10:10:10+00:00"
+                .parse::<Timestamp>()
+                .unwrap()
+                .to_zoned(local_tz.clone()),
+        );
+        ref_time = test_time.timestamp();
        let t1m_ago = ref_time - Duration::from_secs(60);
        let t10s_before = "2010-10-10T10:10:00+00:00";
-        assert!(!TimeFilter::before(&ref_time, t10s_before)
+        assert!(!TimeFilter::before(t10s_before)
            .unwrap()
            .applies_to(&ref_time));
-        assert!(TimeFilter::before(&ref_time, t10s_before)
+        assert!(TimeFilter::before(t10s_before)
            .unwrap()
            .applies_to(&t1m_ago));

-        assert!(TimeFilter::after(&ref_time, t10s_before)
+        assert!(TimeFilter::after(t10s_before)
            .unwrap()
            .applies_to(&ref_time));
-        assert!(!TimeFilter::after(&ref_time, t10s_before)
+        assert!(!TimeFilter::after(t10s_before).unwrap().applies_to(&t1m_ago));
+
+        let ref_timestamp = 1707723412u64; // Mon Feb 12 07:36:52 UTC 2024
+        test_time.set(
+            "2024-02-12T07:36:52+00:00"
+                .parse::<Timestamp>()
+                .unwrap()
+                .to_zoned(local_tz),
+        );
+        ref_time = test_time.timestamp();
+        let t1m_ago = ref_time - Duration::from_secs(60);
+        let t1s_later = ref_time + Duration::from_secs(1);
+        // Timestamp only supported via '@' prefix
+        assert!(TimeFilter::before(&ref_timestamp.to_string()).is_none());
+        assert!(TimeFilter::before(&format!("@{ref_timestamp}"))
            .unwrap()
            .applies_to(&t1m_ago));
+        assert!(!TimeFilter::before(&format!("@{ref_timestamp}"))
+            .unwrap()
+            .applies_to(&t1s_later));
+        assert!(!TimeFilter::after(&format!("@{ref_timestamp}"))
+            .unwrap()
+            .applies_to(&t1m_ago));
+        assert!(TimeFilter::after(&format!("@{ref_timestamp}"))
+            .unwrap()
+            .applies_to(&t1s_later));
    }
 }
@@ -34,10 +34,10 @@ pub fn dirname(path: &Path) -> OsString {
 #[cfg(test)]
 mod path_tests {
    use super::*;
-    use std::path::MAIN_SEPARATOR;
+    use std::path::MAIN_SEPARATOR_STR;

    fn correct(input: &str) -> String {
-        input.replace('/', &MAIN_SEPARATOR.to_string())
+        input.replace('/', MAIN_SEPARATOR_STR)
    }

    macro_rules! func_tests {
@@ -0,0 +1,281 @@
+mod input;
+
+use std::borrow::Cow;
+use std::ffi::{OsStr, OsString};
+use std::fmt::{self, Display, Formatter};
+use std::path::{Component, Path, Prefix};
+use std::sync::OnceLock;
+
+use aho_corasick::AhoCorasick;
+
+use self::input::{basename, dirname, remove_extension};
+
+/// Designates what should be written to a buffer
+///
+/// Each `Token` contains either text, or a placeholder variant, which will be used to generate
+/// commands after all tokens for a given command template have been collected.
+#[derive(Clone, Debug, PartialEq, Eq)]
+pub enum Token {
+    Placeholder,
+    Basename,
+    Parent,
+    NoExt,
+    BasenameNoExt,
+    Text(String),
+}
+
+impl Display for Token {
+    fn fmt(&self, f: &mut Formatter) -> fmt::Result {
+        match *self {
+            Token::Placeholder => f.write_str("{}")?,
+            Token::Basename => f.write_str("{/}")?,
+            Token::Parent => f.write_str("{//}")?,
+            Token::NoExt => f.write_str("{.}")?,
+            Token::BasenameNoExt => f.write_str("{/.}")?,
+            Token::Text(ref string) => f.write_str(string)?,
+        }
+        Ok(())
+    }
+}
+
+/// A parsed format string
+///
+/// This is either a collection of `Token`s including at least one placeholder variant,
+/// or a fixed text.
+#[derive(Clone, Debug, PartialEq)]
+pub enum FormatTemplate {
+    Tokens(Vec<Token>),
+    Text(String),
+}
+
+static PLACEHOLDERS: OnceLock<AhoCorasick> = OnceLock::new();
+
+impl FormatTemplate {
+    pub fn has_tokens(&self) -> bool {
+        matches!(self, FormatTemplate::Tokens(_))
+    }
+
+    pub fn parse(fmt: &str) -> Self {
+        // NOTE: we assume that { and } have the same length
+        const BRACE_LEN: usize = '{'.len_utf8();
+        let mut tokens = Vec::new();
+        let mut remaining = fmt;
+        let mut buf = String::new();
+        let placeholders = PLACEHOLDERS.get_or_init(|| {
+            AhoCorasick::new(["{{", "}}", "{}", "{/}", "{//}", "{.}", "{/.}"]).unwrap()
+        });
+        while let Some(m) = placeholders.find(remaining) {
+            match m.pattern().as_u32() {
+                0 | 1 => {
+                    // we found an escaped {{ or }}, so add
+                    // everything up to the first char to the buffer
+                    // then skip the second one.
+                    buf += &remaining[..m.start() + BRACE_LEN];
+                    remaining = &remaining[m.end()..];
+                }
+                id if !remaining[m.end()..].starts_with('}') => {
+                    buf += &remaining[..m.start()];
+                    if !buf.is_empty() {
+                        tokens.push(Token::Text(std::mem::take(&mut buf)));
+                    }
+                    tokens.push(token_from_pattern_id(id));
+                    remaining = &remaining[m.end()..];
+                }
+                _ => {
+                    // We got a normal pattern, but the final "}"
+                    // is escaped, so add up to that to the buffer, then
+                    // skip the final }
+                    buf += &remaining[..m.end()];
+                    remaining = &remaining[m.end() + BRACE_LEN..];
+                }
+            }
+        }
+        // Add the rest of the string to the buffer, and add the final buffer to the tokens
+        if !remaining.is_empty() {
+            buf += remaining;
+        }
+        if tokens.is_empty() {
+            // No placeholders were found, so just return the text
+            return FormatTemplate::Text(buf);
+        }
+        // Add final text segment
+        if !buf.is_empty() {
+            tokens.push(Token::Text(buf));
+        }
+        debug_assert!(!tokens.is_empty());
+        FormatTemplate::Tokens(tokens)
+    }
+
+    /// Generate a result string from this template. If path_separator is Some, then it will replace
+    /// the path separator in all placeholder tokens. Fixed text and tokens are not affected by
+    /// path separator substitution.
+    pub fn generate(&self, path: impl AsRef<Path>, path_separator: Option<&str>) -> OsString {
+        use Token::*;
+        let path = path.as_ref();
+
+        match *self {
+            Self::Tokens(ref tokens) => {
+                let mut s = OsString::new();
+                for token in tokens {
+                    match token {
+                        Basename => s.push(Self::replace_separator(basename(path), path_separator)),
+                        BasenameNoExt => s.push(Self::replace_separator(
+                            &remove_extension(basename(path).as_ref()),
+                            path_separator,
+                        )),
+                        NoExt => s.push(Self::replace_separator(
+                            &remove_extension(path),
+                            path_separator,
+                        )),
+                        Parent => s.push(Self::replace_separator(&dirname(path), path_separator)),
+                        Placeholder => {
+                            s.push(Self::replace_separator(path.as_ref(), path_separator))
+                        }
+                        Text(ref string) => s.push(string),
+                    }
+                }
+                s
+            }
+            Self::Text(ref text) => OsString::from(text),
+        }
+    }
+
+    /// Replace the path separator in the input with the custom separator string. If path_separator
+    /// is None, simply return a borrowed Cow<OsStr> of the input. Otherwise, the input is
+    /// interpreted as a Path and its components are iterated through and re-joined into a new
+    /// OsString.
+    fn replace_separator<'a>(path: &'a OsStr, path_separator: Option<&str>) -> Cow<'a, OsStr> {
+        // fast-path - no replacement necessary
+        if path_separator.is_none() {
+            return Cow::Borrowed(path);
+        }
+
+        let path_separator = path_separator.unwrap();
+        let mut out = OsString::with_capacity(path.len());
+        let mut components = Path::new(path).components().peekable();
+
+        while let Some(comp) = components.next() {
+            match comp {
+                // Absolute paths on Windows are tricky.  A Prefix component is usually a drive
+                // letter or UNC path, and is usually followed by RootDir. There are also
+                // "verbatim" prefixes beginning with "\\?\" that skip normalization. We choose to
+                // ignore verbatim path prefixes here because they're very rare, might be
+                // impossible to reach here, and there's no good way to deal with them. If users
+                // are doing something advanced involving verbatim windows paths, they can do their
+                // own output filtering with a tool like sed.
+                Component::Prefix(prefix) => {
+                    if let Prefix::UNC(server, share) = prefix.kind() {
+                        // Prefix::UNC is a parsed version of '\\server\share'
+                        out.push(path_separator);
+                        out.push(path_separator);
+                        out.push(server);
+                        out.push(path_separator);
+                        out.push(share);
+                    } else {
+                        // All other Windows prefix types are rendered as-is. This results in e.g. "C:" for
+                        // drive letters. DeviceNS and Verbatim* prefixes won't have backslashes converted,
+                        // but they're not returned by directories fd can search anyway so we don't worry
+                        // about them.
+                        out.push(comp.as_os_str());
+                    }
+                }
+
+                // Root directory is always replaced with the custom separator.
+                Component::RootDir => out.push(path_separator),
+
+                // Everything else is joined normally, with a trailing separator if we're not last
+                _ => {
+                    out.push(comp.as_os_str());
+                    if components.peek().is_some() {
+                        out.push(path_separator);
+                    }
+                }
+            }
+        }
+        Cow::Owned(out)
+    }
+}
+
+// Convert the id from an aho-corasick match to the
+// appropriate token
+fn token_from_pattern_id(id: u32) -> Token {
+    use Token::*;
+    match id {
+        2 => Placeholder,
+        3 => Basename,
+        4 => Parent,
+        5 => NoExt,
+        6 => BasenameNoExt,
+        _ => unreachable!(),
+    }
+}
+
+#[cfg(test)]
+mod fmt_tests {
+    use super::*;
+    use std::path::PathBuf;
+
+    #[test]
+    fn parse_no_placeholders() {
+        let templ = FormatTemplate::parse("This string has no placeholders");
+        assert_eq!(
+            templ,
+            FormatTemplate::Text("This string has no placeholders".into())
+        );
+    }
+
+    #[test]
+    fn parse_only_brace_escapes() {
+        let templ = FormatTemplate::parse("This string only has escapes like {{ and }}");
+        assert_eq!(
+            templ,
+            FormatTemplate::Text("This string only has escapes like { and }".into())
+        );
+    }
+
+    #[test]
+    fn all_placeholders() {
+        use Token::*;
+
+        let templ = FormatTemplate::parse(
+            "{{path={} \
+            basename={/} \
+            parent={//} \
+            noExt={.} \
+            basenameNoExt={/.} \
+            }}",
+        );
+        assert_eq!(
+            templ,
+            FormatTemplate::Tokens(vec![
+                Text("{path=".into()),
+                Placeholder,
+                Text(" basename=".into()),
+                Basename,
+                Text(" parent=".into()),
+                Parent,
+                Text(" noExt=".into()),
+                NoExt,
+                Text(" basenameNoExt=".into()),
+                BasenameNoExt,
+                Text(" }".into()),
+            ])
+        );
+
+        let mut path = PathBuf::new();
+        path.push("a");
+        path.push("folder");
+        path.push("file.txt");
+
+        let expanded = templ.generate(&path, Some("/")).into_string().unwrap();
+
+        assert_eq!(
+            expanded,
+            "{path=a/folder/file.txt \
+            basename=file.txt \
+            parent=a/folder \
+            noExt=a/folder/file \
+            basenameNoExt=file }"
+        );
+    }
+}
@@ -0,0 +1,87 @@
+use crate::filesystem::absolute_path;
+use std::fmt::{self, Formatter, Write};
+use std::path::{Path, PathBuf};
+
+pub(crate) struct PathUrl(PathBuf);
+
+impl PathUrl {
+    pub(crate) fn new(path: &Path) -> Option<PathUrl> {
+        Some(PathUrl(absolute_path(path).ok()?))
+    }
+}
+
+impl fmt::Display for PathUrl {
+    fn fmt(&self, f: &mut Formatter<'_>) -> fmt::Result {
+        write!(f, "file://{}", host())?;
+        let bytes = self.0.as_os_str().as_encoded_bytes();
+        for &byte in bytes.iter() {
+            encode(f, byte)?;
+        }
+        Ok(())
+    }
+}
+
+fn encode(f: &mut Formatter, byte: u8) -> fmt::Result {
+    // NOTE:
+    // Most terminals can handle non-ascii unicode characters in a file url fine. But on some OSes (notably
+    // windows), the encoded bytes of the path may not be valid UTF-8. Since we don't know if a
+    // byte >= 128 is part of a valid UTF-8 encoding or not, we just percent encode any non-ascii
+    // byte.
+    // Percent encoding these bytes is probably safer anyway.
+    match byte {
+        b'0'..=b'9' | b'A'..=b'Z' | b'a'..=b'z' | b'/' | b':' | b'-' | b'.' | b'_' | b'~' => {
+            f.write_char(byte.into())
+        }
+        #[cfg(windows)]
+        b'\\' => f.write_char('/'),
+        _ => {
+            write!(f, "%{byte:02X}")
+        }
+    }
+}
+
+#[cfg(unix)]
+fn host() -> &'static str {
+    use std::sync::OnceLock;
+
+    static HOSTNAME: OnceLock<String> = OnceLock::new();
+
+    HOSTNAME
+        .get_or_init(|| {
+            nix::unistd::gethostname()
+                .ok()
+                .and_then(|h| h.into_string().ok())
+                .unwrap_or_default()
+        })
+        .as_ref()
+}
+
+#[cfg(not(unix))]
+const fn host() -> &'static str {
+    ""
+}
+
+#[cfg(test)]
+mod test {
+    use super::*;
+
+    // This allows us to test the encoding without having to worry about the host, or absolute path
+    struct Encoded(&'static str);
+
+    impl fmt::Display for Encoded {
+        fn fmt(&self, f: &mut Formatter<'_>) -> fmt::Result {
+            for byte in self.0.bytes() {
+                encode(f, byte)?;
+            }
+            Ok(())
+        }
+    }
+
+    #[test]
+    fn test_unicode_encoding() {
+        assert_eq!(
+            Encoded("$*\x1bßé/∫😃\x07").to_string(),
+            "%24%2A%1B%C3%9F%C3%A9/%E2%88%AB%F0%9F%98%83%07",
+        );
+    }
+}
@@ -7,6 +7,8 @@ mod exit_codes;
 mod filesystem;
 mod filetypes;
 mod filter;
+mod fmt;
+mod hyperlink;
 mod output;
 mod regex_helper;
 mod walk;
@@ -15,7 +17,6 @@ use std::env;
 use std::io::IsTerminal;
 use std::path::Path;
 use std::sync::Arc;
-use std::time;

 use anyhow::{anyhow, bail, Context, Result};
 use clap::{CommandFactory, Parser};
@@ -23,7 +24,7 @@ use globset::GlobBuilder;
 use lscolors::LsColors;
 use regex::bytes::{Regex, RegexBuilder, RegexSetBuilder};

-use crate::cli::{ColorWhen, Opts};
+use crate::cli::{ColorWhen, HyperlinkWhen, Opts};
 use crate::config::Config;
 use crate::exec::CommandSet;
 use crate::exit_codes::ExitCode;
@@ -61,7 +62,7 @@ fn main() {
            exit_code.exit();
        }
        Err(err) => {
-            eprintln!("[fd error]: {:#}", err);
+            eprintln!("[fd error]: {err:#}");
            ExitCode::GeneralError.exit();
        }
    }
@@ -233,6 +234,11 @@ fn construct_config(mut opts: Opts, pattern_regexps: &[String]) -> Result<Config
    } else {
        None
    };
+    let hyperlink = match opts.hyperlink {
+        HyperlinkWhen::Always => true,
+        HyperlinkWhen::Never => false,
+        HyperlinkWhen::Auto => colored_output,
+    };
    let command = extract_command(&mut opts, colored_output)?;
    let has_command = command.is_some();

@@ -257,6 +263,7 @@ fn construct_config(mut opts: Opts, pattern_regexps: &[String]) -> Result<Config
        threads: opts.threads().get(),
        max_buffer_time: opts.max_buffer_time,
        ls_colors,
+        hyperlink,
        interactive_terminal,
        file_types: opts.filetype.as_ref().map(|values| {
            use crate::cli::FileType::*;
@@ -299,6 +306,10 @@ fn construct_config(mut opts: Opts, pattern_regexps: &[String]) -> Result<Config
                    .build()
            })
            .transpose()?,
+        format: opts
+            .format
+            .as_deref()
+            .map(crate::fmt::FormatTemplate::parse),
        command: command.map(Arc::new),
        batch_size: opts.batch_size,
        exclude_patterns: opts.exclude.iter().map(|p| String::from("!") + p).collect(),
@@ -311,8 +322,7 @@ fn construct_config(mut opts: Opts, pattern_regexps: &[String]) -> Result<Config
        path_separator,
        actual_path_separator,
        max_results: opts.max_results(),
-        strip_cwd_prefix: (opts.no_search_paths()
-            && (opts.strip_cwd_prefix || !(opts.null_separator || has_command))),
+        strip_cwd_prefix: opts.strip_cwd_prefix(|| !(opts.null_separator || has_command)),
    })
 }

@@ -325,18 +335,22 @@ fn extract_command(opts: &mut Opts, colored_output: bool) -> Result<Option<Comma
            if !opts.list_details {
                return None;
            }
-            let color_arg = format!("--color={}", opts.color.as_str());

-            let res = determine_ls_command(&color_arg, colored_output)
+            let res = determine_ls_command(colored_output)
                .map(|cmd| CommandSet::new_batch([cmd]).unwrap());
            Some(res)
        })
        .transpose()
 }

-fn determine_ls_command(color_arg: &str, colored_output: bool) -> Result<Vec<&str>> {
+fn determine_ls_command(colored_output: bool) -> Result<Vec<&'static str>> {
    #[allow(unused)]
    let gnu_ls = |command_name| {
+        let color_arg = if colored_output {
+            "--color=always"
+        } else {
+            "--color=never"
+        };
        // Note: we use short options here (instead of --long-options) to support more
        // platforms (like BusyBox).
        vec![
@@ -414,10 +428,9 @@ fn determine_ls_command(color_arg: &str, colored_output: bool) -> Result<Vec<&st
 }

 fn extract_time_constraints(opts: &Opts) -> Result<Vec<TimeFilter>> {
-    let now = time::SystemTime::now();
    let mut time_constraints: Vec<TimeFilter> = Vec::new();
    if let Some(ref t) = opts.changed_within {
-        if let Some(f) = TimeFilter::after(&now, t) {
+        if let Some(f) = TimeFilter::after(t) {
            time_constraints.push(f);
        } else {
            return Err(anyhow!(
@@ -427,7 +440,7 @@ fn extract_time_constraints(opts: &Opts) -> Result<Vec<TimeFilter>> {
        }
    }
    if let Some(ref t) = opts.changed_before {
-        if let Some(f) = TimeFilter::before(&now, t) {
+        if let Some(f) = TimeFilter::before(t) {
            time_constraints.push(f);
        } else {
            return Err(anyhow!(
@@ -5,29 +5,39 @@ use lscolors::{Indicator, LsColors, Style};

 use crate::config::Config;
 use crate::dir_entry::DirEntry;
-use crate::error::print_error;
-use crate::exit_codes::ExitCode;
+use crate::fmt::FormatTemplate;
+use crate::hyperlink::PathUrl;

 fn replace_path_separator(path: &str, new_path_separator: &str) -> String {
    path.replace(std::path::MAIN_SEPARATOR, new_path_separator)
 }

 // TODO: this function is performance critical and can probably be optimized
-pub fn print_entry<W: Write>(stdout: &mut W, entry: &DirEntry, config: &Config) {
-    let r = if let Some(ref ls_colors) = config.ls_colors {
-        print_entry_colorized(stdout, entry, config, ls_colors)
+pub fn print_entry<W: Write>(stdout: &mut W, entry: &DirEntry, config: &Config) -> io::Result<()> {
+    let mut has_hyperlink = false;
+    if config.hyperlink {
+        if let Some(url) = PathUrl::new(entry.path()) {
+            write!(stdout, "\x1B]8;;{url}\x1B\\")?;
+            has_hyperlink = true;
+        }
+    }
+
+    if let Some(ref format) = config.format {
+        print_entry_format(stdout, entry, config, format)?;
+    } else if let Some(ref ls_colors) = config.ls_colors {
+        print_entry_colorized(stdout, entry, config, ls_colors)?;
    } else {
-        print_entry_uncolorized(stdout, entry, config)
+        print_entry_uncolorized(stdout, entry, config)?;
    };

-    if let Err(e) = r {
-        if e.kind() == ::std::io::ErrorKind::BrokenPipe {
-            // Exit gracefully in case of a broken pipe (e.g. 'fd ... | head -n 3').
-            ExitCode::Success.exit();
-        } else {
-            print_error(format!("Could not write to output: {}", e));
-            ExitCode::GeneralError.exit();
-        }
+    if has_hyperlink {
+        write!(stdout, "\x1B]8;;\x1B\\")?;
+    }
+
+    if config.null_separator {
+        write!(stdout, "\0")
+    } else {
+        writeln!(stdout)
    }
 }

@@ -41,7 +51,7 @@ fn print_trailing_slash<W: Write>(
    config: &Config,
    style: Option<&Style>,
 ) -> io::Result<()> {
-    if entry.file_type().map_or(false, |ft| ft.is_dir()) {
+    if entry.file_type().is_some_and(|ft| ft.is_dir()) {
        write!(
            stdout,
            "{}",
@@ -54,6 +64,21 @@ fn print_trailing_slash<W: Write>(
    Ok(())
 }

+// TODO: this function is performance critical and can probably be optimized
+fn print_entry_format<W: Write>(
+    stdout: &mut W,
+    entry: &DirEntry,
+    config: &Config,
+    format: &FormatTemplate,
+) -> io::Result<()> {
+    let output = format.generate(
+        entry.stripped_path(config),
+        config.path_separator.as_deref(),
+    );
+    // TODO: support writing raw bytes on unix?
+    write!(stdout, "{}", output.to_string_lossy())
+}
+
 // TODO: this function is performance critical and can probably be optimized
 fn print_entry_colorized<W: Write>(
    stdout: &mut W,
@@ -103,12 +128,6 @@ fn print_entry_colorized<W: Write>(
        ls_colors.style_for_indicator(Indicator::Directory),
    )?;

-    if config.null_separator {
-        write!(stdout, "\0")?;
-    } else {
-        writeln!(stdout)?;
-    }
-
    Ok(())
 }

@@ -118,16 +137,14 @@ fn print_entry_uncolorized_base<W: Write>(
    entry: &DirEntry,
    config: &Config,
 ) -> io::Result<()> {
-    let separator = if config.null_separator { "\0" } else { "\n" };
    let path = entry.stripped_path(config);

    let mut path_string = path.to_string_lossy();
    if let Some(ref separator) = config.path_separator {
        *path_string.to_mut() = replace_path_separator(&path_string, separator);
    }
-    write!(stdout, "{}", path_string)?;
-    print_trailing_slash(stdout, entry, config, None)?;
-    write!(stdout, "{}", separator)
+    write!(stdout, "{path_string}")?;
+    print_trailing_slash(stdout, entry, config, None)
 }

 #[cfg(not(unix))]
@@ -152,9 +169,7 @@ fn print_entry_uncolorized<W: Write>(
        print_entry_uncolorized_base(stdout, entry, config)
    } else {
        // Print path as raw bytes, allowing invalid UTF-8 filenames to be passed to other processes
-        let separator = if config.null_separator { b"\0" } else { b"\n" };
        stdout.write_all(entry.stripped_path(config).as_os_str().as_bytes())?;
-        print_trailing_slash(stdout, entry, config, None)?;
-        stdout.write_all(separator)
+        print_trailing_slash(stdout, entry, config, None)
    }
 }
@@ -16,7 +16,7 @@ fn hir_has_uppercase_char(hir: &Hir) -> bool {
    use regex_syntax::hir::*;

    match hir.kind() {
-        HirKind::Literal(Literal(bytes)) => match std::str::from_utf8(&bytes) {
+        HirKind::Literal(Literal(bytes)) => match std::str::from_utf8(bytes) {
            Ok(s) => s.chars().any(|c| c.is_uppercase()),
            Err(_) => bytes.iter().any(|b| char::from(*b).is_uppercase()),
        },
@@ -67,7 +67,7 @@ fn hir_matches_strings_with_leading_dot(hir: &Hir) -> bool {

            if let Some(hir) = hirs.next() {
                match hir.kind() {
-                    HirKind::Literal(Literal(bytes)) => bytes.starts_with(&[b'.']),
+                    HirKind::Literal(Literal(bytes)) => bytes.starts_with(b"."),
                    _ => false,
                }
            } else {
@@ -12,7 +12,7 @@ use anyhow::{anyhow, Result};
 use crossbeam_channel::{bounded, Receiver, RecvTimeoutError, SendError, Sender};
 use etcetera::BaseStrategy;
 use ignore::overrides::{Override, OverrideBuilder};
-use ignore::{self, WalkBuilder, WalkParallel, WalkState};
+use ignore::{WalkBuilder, WalkParallel, WalkState};
 use regex::bytes::Regex;

 use crate::config::Config;
@@ -250,7 +250,12 @@ impl<'a, W: Write> ReceiverBuffer<'a, W> {

    /// Output a path.
    fn print(&mut self, entry: &DirEntry) -> Result<(), ExitCode> {
-        output::print_entry(&mut self.stdout, entry, &self.config);
+        if let Err(e) = output::print_entry(&mut self.stdout, entry, self.config) {
+            if e.kind() != ::std::io::ErrorKind::BrokenPipe {
+                print_error(format!("Could not write to output: {e}"));
+                return Err(ExitCode::GeneralError);
+            }
+        }

        if self.interrupt_flag.load(Ordering::Relaxed) {
            // Ignore any errors on flush, because we're about to exit anyway
@@ -334,10 +339,6 @@ impl WorkerState {
                .map_err(|e| anyhow!("Malformed exclude pattern: {}", e))?;
        }

-        if config.read_vcsignore {
-            builder.add("!.git/").expect("Invalid exclude pattern");
-        }
-
        builder
            .build()
            .map_err(|_| anyhow!("Mismatch in exclude patterns"))
@@ -375,10 +376,7 @@ impl WorkerState {
                    match result {
                        Some(ignore::Error::Partial(_)) => (),
                        Some(err) => {
-                            print_error(format!(
-                                "Malformed pattern in global ignore file. {}.",
-                                err
-                            ));
+                            print_error(format!("Malformed pattern in global ignore file. {err}."));
                        }
                        None => (),
                    }
@@ -391,7 +389,7 @@ impl WorkerState {
            match result {
                Some(ignore::Error::Partial(_)) => (),
                Some(err) => {
-                    print_error(format!("Malformed pattern in custom ignore file. {}.", err));
+                    print_error(format!("Malformed pattern in custom ignore file. {err}."));
                }
                None => (),
            }
@@ -413,7 +411,7 @@ impl WorkerState {
        // This will be set to `Some` if the `--exec` argument was supplied.
        if let Some(ref cmd) = config.command {
            if cmd.in_batch_mode() {
-                exec::batch(rx.into_iter().flatten(), cmd, &config)
+                exec::batch(rx.into_iter().flatten(), cmd, config)
            } else {
                let out_perm = Mutex::new(());

@@ -426,7 +424,7 @@ impl WorkerState {

                        // Spawn a job thread that will listen for and execute inputs.
                        let handle = scope
-                            .spawn(|| exec::job(rx.into_iter().flatten(), cmd, &out_perm, &config));
+                            .spawn(|| exec::job(rx.into_iter().flatten(), cmd, &out_perm, config));

                        // Push the handle of the spawned thread into the vector for later joining.
                        handles.push(handle);
@@ -479,7 +477,7 @@ impl WorkerState {
                                && path
                                    .symlink_metadata()
                                    .ok()
-                                    .map_or(false, |m| m.file_type().is_symlink()) =>
+                                    .is_some_and(|m| m.file_type().is_symlink()) =>
                        {
                            DirEntry::broken_symlink(path)
                        }
@@ -104,9 +104,9 @@ fn format_output_error(args: &[&str], expected: &str, actual: &str) -> String {
    let diff_text = diff::lines(expected, actual)
        .into_iter()
        .map(|diff| match diff {
-            diff::Result::Left(l) => format!("-{}", l),
-            diff::Result::Both(l, _) => format!(" {}", l),
-            diff::Result::Right(r) => format!("+{}", r),
+            diff::Result::Left(l) => format!("-{l}"),
+            diff::Result::Both(l, _) => format!(" {l}"),
+            diff::Result::Right(r) => format!("+{r}"),
        })
        .collect::<Vec<_>>()
        .join("\n");
@@ -129,7 +129,7 @@ fn normalize_output(s: &str, trim_start: bool, normalize_line: bool) -> String {
        .lines()
        .map(|line| {
            let line = if trim_start { line.trim_start() } else { line };
-            let line = line.replace('/', &std::path::MAIN_SEPARATOR.to_string());
+            let line = line.replace('/', std::path::MAIN_SEPARATOR_STR);
            if normalize_line {
                let mut words: Vec<_> = line.split_whitespace().collect();
                words.sort_unstable();
@@ -290,7 +290,7 @@ impl TestEnv {
    pub fn assert_failure_with_error(&self, args: &[&str], expected: &str) {
        let status = self.assert_error_subdirectory(".", args, Some(expected));
        if status.success() {
-            panic!("error '{}' did not occur.", expected);
+            panic!("error '{expected}' did not occur.");
        }
    }

@@ -316,6 +316,9 @@ impl TestEnv {
        } else {
            cmd.arg("--no-global-ignore-file");
        }
+        // Make sure LS_COLORS is unset to ensure consistent
+        // color output
+        cmd.env("LS_COLORS", "");
        cmd.args(args);

        // Run *fd*.
@@ -8,6 +8,7 @@ use std::path::Path;
 use std::time::{Duration, SystemTime};
 use test_case::test_case;

+use jiff::Timestamp;
 use normpath::PathExt;
 use regex::escape;

@@ -595,10 +596,7 @@ fn test_full_path() {
    let prefix = escape(&root.to_string_lossy());

    te.assert_output(
-        &[
-            "--full-path",
-            &format!("^{prefix}.*three.*foo$", prefix = prefix),
-        ],
+        &["--full-path", &format!("^{prefix}.*three.*foo$")],
        "one/two/three/d.foo
        one/two/three/directory_foo/",
    );
@@ -1311,7 +1309,8 @@ fn test_type_executable() {
    let te = TestEnv::new(DEFAULT_DIRS, DEFAULT_FILES);

    fs::OpenOptions::new()
-        .create(true)
+        .create_new(true)
+        .truncate(true)
        .write(true)
        .mode(0o777)
        .open(te.test_root().join("executable-file.sh"))
@@ -1319,6 +1318,7 @@ fn test_type_executable() {

    fs::OpenOptions::new()
        .create(true)
+        .truncate(true)
        .write(true)
        .mode(0o645)
        .open(te.test_root().join("not-user-executable-file.sh"))
@@ -1516,7 +1516,7 @@ fn test_symlink_as_absolute_root() {
    let (te, abs_path) = get_test_env_with_abs_path(DEFAULT_DIRS, DEFAULT_FILES);

    te.assert_output(
-        &["", &format!("{abs_path}/symlink", abs_path = abs_path)],
+        &["", &format!("{abs_path}/symlink")],
        &format!(
            "{abs_path}/symlink/c.foo
            {abs_path}/symlink/C.Foo2
@@ -1541,7 +1541,7 @@ fn test_symlink_and_full_path() {
        &[
            "--absolute-path",
            "--full-path",
-            &format!("^{prefix}.*three", prefix = prefix),
+            &format!("^{prefix}.*three"),
        ],
        &format!(
            "{abs_path}/{expected_path}/three/
@@ -1561,8 +1561,8 @@ fn test_symlink_and_full_path_abs_path() {
    te.assert_output(
        &[
            "--full-path",
-            &format!("^{prefix}.*symlink.*three", prefix = prefix),
-            &format!("{abs_path}/symlink", abs_path = abs_path),
+            &format!("^{prefix}.*symlink.*three"),
+            &format!("{abs_path}/symlink"),
        ],
        &format!(
            "{abs_path}/symlink/three/
@@ -1622,6 +1622,66 @@ fn test_excludes() {
    );
 }

+#[test]
+fn format() {
+    let te = TestEnv::new(DEFAULT_DIRS, DEFAULT_FILES);
+
+    te.assert_output(
+        &["--format", "path={}", "--path-separator=/"],
+        "path=a.foo
+        path=e1 e2
+        path=one
+        path=one/b.foo
+        path=one/two
+        path=one/two/C.Foo2
+        path=one/two/c.foo
+        path=one/two/three
+        path=one/two/three/d.foo
+        path=one/two/three/directory_foo
+        path=symlink",
+    );
+
+    te.assert_output(
+        &["foo", "--format", "noExt={.}", "--path-separator=/"],
+        "noExt=a
+        noExt=one/b
+        noExt=one/two/C
+        noExt=one/two/c
+        noExt=one/two/three/d
+        noExt=one/two/three/directory_foo",
+    );
+
+    te.assert_output(
+        &["foo", "--format", "basename={/}", "--path-separator=/"],
+        "basename=a.foo
+        basename=b.foo
+        basename=C.Foo2
+        basename=c.foo
+        basename=d.foo
+        basename=directory_foo",
+    );
+
+    te.assert_output(
+        &["foo", "--format", "name={/.}", "--path-separator=/"],
+        "name=a
+        name=b
+        name=C
+        name=c
+        name=d
+        name=directory_foo",
+    );
+
+    te.assert_output(
+        &["foo", "--format", "parent={//}", "--path-separator=/"],
+        "parent=.
+        parent=one
+        parent=one/two
+        parent=one/two
+        parent=one/two/three
+        parent=one/two/three",
+    );
+}
+
 /// Shell script execution (--exec)
 #[test]
 fn test_exec() {
@@ -2229,7 +2289,10 @@ fn test_modified_relative() {

 #[cfg(test)]
 fn change_file_modified<P: AsRef<Path>>(path: P, iso_date: &str) {
-    let st = humantime::parse_rfc3339(iso_date).expect("invalid date");
+    let st = iso_date
+        .parse::<Timestamp>()
+        .map(SystemTime::from)
+        .expect("invalid date");
    let ft = filetime::FileTime::from_system_time(st);
    filetime::set_file_times(path, ft, ft).expect("time modification failde");
 }
@@ -2275,7 +2338,7 @@ fn test_owner_current_user() {
 fn test_owner_current_group() {
    let te = TestEnv::new(DEFAULT_DIRS, DEFAULT_FILES);
    let gid = Gid::current();
-    te.assert_output(&["--owner", &format!(":{}", gid), "a.foo"], "a.foo");
+    te.assert_output(&["--owner", &format!(":{gid}"), "a.foo"], "a.foo");
    if let Ok(Some(group)) = Group::from_gid(gid) {
        te.assert_output(&["--owner", &format!(":{}", group.name), "a.foo"], "a.foo");
    }
@@ -2465,6 +2528,7 @@ fn test_number_parsing_errors() {
    te.assert_failure(&["--threads=0"]);

    te.assert_failure(&["--min-depth=a"]);
+    te.assert_failure(&["--mindepth=a"]);
    te.assert_failure(&["--max-depth=a"]);
    te.assert_failure(&["--maxdepth=a"]);
    te.assert_failure(&["--exact-depth=a"]);
@@ -2553,7 +2617,7 @@ fn test_invalid_cwd() {
        .unwrap();

    if !output.status.success() {
-        panic!("{:?}", output);
+        panic!("{output:?}");
    }
 }

@@ -2571,7 +2635,14 @@ fn test_git_dir() {
        ],
    );

-    te.assert_output(&["--hidden", "foo"], "");
+    te.assert_output(
+        &["--hidden", "foo"],
+        ".git/one/foo.a
+        .git/.foo
+        .git/a.foo
+        other_dir/.git/foo1
+        nested/dir/.git/foo2",
+    );
    te.assert_output(&["--no-ignore", "foo"], "");
    te.assert_output(
        &["--hidden", "--no-ignore", "foo"],
@@ -2590,3 +2661,34 @@ fn test_git_dir() {
         nested/dir/.git/foo2",
    );
 }
+
+#[test]
+fn test_gitignore_parent() {
+    let te = TestEnv::new(&["sub"], &[".abc", "sub/.abc"]);
+
+    fs::File::create(te.test_root().join(".gitignore"))
+        .unwrap()
+        .write_all(b".abc\n")
+        .unwrap();
+
+    te.assert_output_subdirectory("sub", &["--hidden"], "");
+    te.assert_output_subdirectory("sub", &["--hidden", "--search-path", "."], "");
+}
+
+#[test]
+fn test_hyperlink() {
+    let te = TestEnv::new(DEFAULT_DIRS, DEFAULT_FILES);
+
+    #[cfg(unix)]
+    let hostname = nix::unistd::gethostname().unwrap().into_string().unwrap();
+    #[cfg(not(unix))]
+    let hostname = "";
+
+    let expected = format!(
+        "\x1b]8;;file://{}{}/a.foo\x1b\\a.foo\x1b]8;;\x1b\\",
+        hostname,
+        get_absolute_root_path(&te),
+    );
+
+    te.assert_output(&["--hyperlink=always", "a.foo"], &expected);
+}