Simplify is_printable check, closes #78

This is technically a breaking change because it halves the upper range
of accepted values for byte counts. I thought that fine under the
assumption it will be rare for users to ever want to specify something
in the upper half of `u64`'s value range, especially considering that
this can pave the way for having negative byte offsets (instead of just
counts).

.github/FUNDING.yml

@@ -1 +0,0 @@
-github: sharkdp

.travis.yml

@@ -14,11 +14,11 @@ matrix:
       env: TARGET=x86_64-apple-darwin
     - os: windows
       rust: stable
-      env: TARGET=x86_64-pc-windows-msvc
+      env: TARGET=x86_64-pc-windows-gnu
 
     # Minimum Rust supported channel.
     - os: linux
-      rust: 1.31.0
+      rust: 1.36.0
       env: TARGET=x86_64-unknown-linux-gnu
 
 
@@ -44,6 +44,7 @@ script:
   # Incorporate TARGET env var to the build and test process
   - cargo build --target $TARGET --verbose
   - cargo test --target $TARGET --verbose
+  - cargo run --target $TARGET --example=simple
 
 before_deploy:
   - bash ci/before_deploy.bash
@@ -72,7 +73,7 @@ deploy:
     # deploy only if we push a tag
     tags: true
     # deploy only on stable channel that has TARGET env variable sets
-    condition: $TRAVIS_RUST_VERSION = stable && $TARGET != "x86_64-pc-windows-msvc"
+    condition: $TRAVIS_RUST_VERSION = stable && $TARGET != "x86_64-pc-windows-gnu"
 
 notifications:
   email:

CHANGELOG.md

@@ -0,0 +1,106 @@
+# v0.8.0
+
+## Features
+
+- A new `--skip <N>` / `-s <N>` option can be used to skip the first `N` bytes of the input, see #16, #88 (@Tarnadas, @MaxJohansen, @ErichDonGubler)
+- The `--length`/`--bytes`/`--skip`/`--display-offset` options can now take units for their value argument, for example:
+  ``` bash
+  hexyl /dev/random --length=1KiB
+  hexyl $(which hexyl) --skip=1MiB --length=10KiB
+  ```
+  Both decimal SI prefixes (kB, MB, …) as well as binary IEC prefixes (KiB, MiB, …) are supported.
+  In addition, there is a new `--block-size <SIZE>` option that can be used to control the size of the `block`
+  unit:
+  ``` bash
+  hexyl /dev/random --block-size=4kB --length=2block
+  ```
+  See: #44 (@ErichDonGubler and @aswild)
+
+## Other
+
+- Various improvements throughout the code base by @ErichDonGubler
+
+## Packaging
+
+- `hexyl` is now available on Void Linux, see #91 (@notramo)
+
+# v0.7.0
+
+## Bugfixes
+
+- hexyl can now be closed with `Ctrl-C` when reading input from STDIN, see #84 
+
+## Changes
+
+- Breaking change (library): [`Printer::print_all`](https://docs.rs/hexyl/latest/hexyl/struct.Printer.html#method.print_all) does not take a second argument anymore.
+- Added an example on how to use `hexyl` as a library: https://github.com/sharkdp/hexyl/blob/v0.7.0/examples/simple.rs
+
+# v0.6.0
+
+## Features
+
+- `hexyl` can now be used as a library, see #67 (@tommilligan)
+
+- Added a new `-o`/`--display-offset` option to add a certain offset to the
+  reported file positions, see #57 (@tommilligan)
+
+## Bugfixes
+
+- Remove additional space on short input, see #69 (@nalshihabi)
+
+## Other
+
+- Performance improvements, see #73 and #66
+
+# v0.5.1
+
+## Bugfixes
+
+- A bug in the squeezing logic caused a wrong hexdump, see #62 (@awidegreen)
+- Some colors are printed even if they're disabled, see #64 (@awidegreen)
+- Fixed build failure on OpenBSD 6.5, see #61
+
+# v0.5.0
+
+## Features
+
+- Added support for squeezing where reoccuring lines are squashed together and visualized with an asterisk. A new `-v`/`--no-squeezing` option can be used to disable the feature. For details, see #59 (@awidegreen)
+- Added a new `--border` option with support for various styles (Unicode, ASCII, None), see #54 (@dmke)
+- The `--length`/`-n` argument can be passed as a hexadecimal number (`hexyl -n 0xff /dev/urandom`), see #45 (@Qyriad)
+- Added `--bytes`/`-c` as an alias for `--length`/`-n`, see #48 (@selfup)
+
+## Changes
+
+- Print header immediately before the first line, see #51 (@mziter)
+
+
+# v0.4.0
+
+## Features
+
+- Added a new `--color=always/auto/never` option which can be used
+  to control `hexyl`s color output, see #30 (@bennetthardwick)
+- Use 16 colors instead of 256, see #38
+
+## Changes
+
+- Various speed improvements, see #33 (@kballard)
+
+## Bugfixes
+
+- Proper Ctrl-C handling, see #35
+- Proper handling of broken pipes (`hexyl … | head`)
+
+# v0.3.1
+
+- Various (huge) performance improvements, see #23 and #24 (@kballard)
+- Replaced 24-bit truecolor ANSI codes by 8-bit codes to support
+  more terminal emulators, fixes #9
+
+# v0.3.0
+
+Windows support
+
+# v0.2.0
+
+Initial release

Cargo.lock

@@ -18,88 +18,57 @@ dependencies = [
 
 [[package]]
 name = "atty"
-version = "0.2.13"
+version = "0.2.14"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 dependencies = [
- "libc 0.2.62 (registry+https://github.com/rust-lang/crates.io-index)",
+ "hermit-abi 0.1.13 (registry+https://github.com/rust-lang/crates.io-index)",
+ "libc 0.2.71 (registry+https://github.com/rust-lang/crates.io-index)",
  "winapi 0.3.8 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
 [[package]]
 name = "bitflags"
-version = "1.2.0"
+version = "1.2.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-
-[[package]]
-name = "cc"
-version = "1.0.45"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-
-[[package]]
-name = "cfg-if"
-version = "0.1.10"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 
 [[package]]
 name = "clap"
-version = "2.33.0"
+version = "2.33.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 dependencies = [
  "ansi_term 0.11.0 (registry+https://github.com/rust-lang/crates.io-index)",
- "atty 0.2.13 (registry+https://github.com/rust-lang/crates.io-index)",
+ "atty 0.2.14 (registry+https://github.com/rust-lang/crates.io-index)",
- "bitflags 1.2.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "bitflags 1.2.1 (registry+https://github.com/rust-lang/crates.io-index)",
  "strsim 0.8.0 (registry+https://github.com/rust-lang/crates.io-index)",
- "term_size 0.3.1 (registry+https://github.com/rust-lang/crates.io-index)",
+ "term_size 0.3.2 (registry+https://github.com/rust-lang/crates.io-index)",
  "textwrap 0.11.0 (registry+https://github.com/rust-lang/crates.io-index)",
- "unicode-width 0.1.6 (registry+https://github.com/rust-lang/crates.io-index)",
+ "unicode-width 0.1.7 (registry+https://github.com/rust-lang/crates.io-index)",
- "vec_map 0.8.1 (registry+https://github.com/rust-lang/crates.io-index)",
+ "vec_map 0.8.2 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
 [[package]]
-name = "ctrlc"
+name = "hermit-abi"
-version = "3.1.3"
+version = "0.1.13"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 dependencies = [
- "nix 0.14.1 (registry+https://github.com/rust-lang/crates.io-index)",
+ "libc 0.2.71 (registry+https://github.com/rust-lang/crates.io-index)",
- "winapi 0.3.8 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
 [[package]]
 name = "hexyl"
-version = "0.6.0"
+version = "0.8.0"
 dependencies = [
  "ansi_term 0.12.1 (registry+https://github.com/rust-lang/crates.io-index)",
- "atty 0.2.13 (registry+https://github.com/rust-lang/crates.io-index)",
+ "atty 0.2.14 (registry+https://github.com/rust-lang/crates.io-index)",
- "clap 2.33.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "clap 2.33.1 (registry+https://github.com/rust-lang/crates.io-index)",
- "ctrlc 3.1.3 (registry+https://github.com/rust-lang/crates.io-index)",
+ "libc 0.2.71 (registry+https://github.com/rust-lang/crates.io-index)",
-]
-
-[[package]]
-name = "kernel32-sys"
-version = "0.2.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-dependencies = [
- "winapi 0.2.8 (registry+https://github.com/rust-lang/crates.io-index)",
- "winapi-build 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
 [[package]]
 name = "libc"
-version = "0.2.62"
+version = "0.2.71"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 
-[[package]]
-name = "nix"
-version = "0.14.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-dependencies = [
- "bitflags 1.2.0 (registry+https://github.com/rust-lang/crates.io-index)",
- "cc 1.0.45 (registry+https://github.com/rust-lang/crates.io-index)",
- "cfg-if 0.1.10 (registry+https://github.com/rust-lang/crates.io-index)",
- "libc 0.2.62 (registry+https://github.com/rust-lang/crates.io-index)",
- "void 1.0.2 (registry+https://github.com/rust-lang/crates.io-index)",
-]
-
 [[package]]
 name = "strsim"
 version = "0.8.0"
@@ -107,12 +76,11 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 
 [[package]]
 name = "term_size"
-version = "0.3.1"
+version = "0.3.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 dependencies = [
- "kernel32-sys 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)",
+ "libc 0.2.71 (registry+https://github.com/rust-lang/crates.io-index)",
- "libc 0.2.62 (registry+https://github.com/rust-lang/crates.io-index)",
+ "winapi 0.3.8 (registry+https://github.com/rust-lang/crates.io-index)",
- "winapi 0.2.8 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
 [[package]]
@@ -120,28 +88,18 @@ name = "textwrap"
 version = "0.11.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 dependencies = [
- "term_size 0.3.1 (registry+https://github.com/rust-lang/crates.io-index)",
+ "term_size 0.3.2 (registry+https://github.com/rust-lang/crates.io-index)",
- "unicode-width 0.1.6 (registry+https://github.com/rust-lang/crates.io-index)",
+ "unicode-width 0.1.7 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
 [[package]]
 name = "unicode-width"
-version = "0.1.6"
+version = "0.1.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 
 [[package]]
 name = "vec_map"
-version = "0.8.1"
+version = "0.8.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-
-[[package]]
-name = "void"
-version = "1.0.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-
-[[package]]
-name = "winapi"
-version = "0.2.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 
 [[package]]
@@ -153,11 +111,6 @@ dependencies = [
  "winapi-x86_64-pc-windows-gnu 0.4.0 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
-[[package]]
-name = "winapi-build"
-version = "0.1.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-
 [[package]]
 name = "winapi-i686-pc-windows-gnu"
 version = "0.4.0"
@@ -171,23 +124,16 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 [metadata]
 "checksum ansi_term 0.11.0 (registry+https://github.com/rust-lang/crates.io-index)" = "ee49baf6cb617b853aa8d93bf420db2383fab46d314482ca2803b40d5fde979b"
 "checksum ansi_term 0.12.1 (registry+https://github.com/rust-lang/crates.io-index)" = "d52a9bb7ec0cf484c551830a7ce27bd20d67eac647e1befb56b0be4ee39a55d2"
-"checksum atty 0.2.13 (registry+https://github.com/rust-lang/crates.io-index)" = "1803c647a3ec87095e7ae7acfca019e98de5ec9a7d01343f611cf3152ed71a90"
+"checksum atty 0.2.14 (registry+https://github.com/rust-lang/crates.io-index)" = "d9b39be18770d11421cdb1b9947a45dd3f37e93092cbf377614828a319d5fee8"
-"checksum bitflags 1.2.0 (registry+https://github.com/rust-lang/crates.io-index)" = "8a606a02debe2813760609f57a64a2ffd27d9fdf5b2f133eaca0b248dd92cdd2"
+"checksum bitflags 1.2.1 (registry+https://github.com/rust-lang/crates.io-index)" = "cf1de2fe8c75bc145a2f577add951f8134889b4795d47466a54a5c846d691693"
-"checksum cc 1.0.45 (registry+https://github.com/rust-lang/crates.io-index)" = "4fc9a35e1f4290eb9e5fc54ba6cf40671ed2a2514c3eeb2b2a908dda2ea5a1be"
+"checksum clap 2.33.1 (registry+https://github.com/rust-lang/crates.io-index)" = "bdfa80d47f954d53a35a64987ca1422f495b8d6483c0fe9f7117b36c2a792129"
-"checksum cfg-if 0.1.10 (registry+https://github.com/rust-lang/crates.io-index)" = "4785bdd1c96b2a846b2bd7cc02e86b6b3dbf14e7e53446c4f54c92a361040822"
+"checksum hermit-abi 0.1.13 (registry+https://github.com/rust-lang/crates.io-index)" = "91780f809e750b0a89f5544be56617ff6b1227ee485bcb06ebe10cdf89bd3b71"
-"checksum clap 2.33.0 (registry+https://github.com/rust-lang/crates.io-index)" = "5067f5bb2d80ef5d68b4c87db81601f0b75bca627bc2ef76b141d7b846a3c6d9"
+"checksum libc 0.2.71 (registry+https://github.com/rust-lang/crates.io-index)" = "9457b06509d27052635f90d6466700c65095fdf75409b3fbdd903e988b886f49"
-"checksum ctrlc 3.1.3 (registry+https://github.com/rust-lang/crates.io-index)" = "c7dfd2d8b4c82121dfdff120f818e09fc4380b0b7e17a742081a89b94853e87f"
-"checksum kernel32-sys 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)" = "7507624b29483431c0ba2d82aece8ca6cdba9382bff4ddd0f7490560c056098d"
-"checksum libc 0.2.62 (registry+https://github.com/rust-lang/crates.io-index)" = "34fcd2c08d2f832f376f4173a231990fa5aef4e99fb569867318a227ef4c06ba"
-"checksum nix 0.14.1 (registry+https://github.com/rust-lang/crates.io-index)" = "6c722bee1037d430d0f8e687bbdbf222f27cc6e4e68d5caf630857bb2b6dbdce"
 "checksum strsim 0.8.0 (registry+https://github.com/rust-lang/crates.io-index)" = "8ea5119cdb4c55b55d432abb513a0429384878c15dde60cc77b1c99de1a95a6a"
-"checksum term_size 0.3.1 (registry+https://github.com/rust-lang/crates.io-index)" = "9e5b9a66db815dcfd2da92db471106457082577c3c278d4138ab3e3b4e189327"
+"checksum term_size 0.3.2 (registry+https://github.com/rust-lang/crates.io-index)" = "1e4129646ca0ed8f45d09b929036bafad5377103edd06e50bf574b353d2b08d9"
 "checksum textwrap 0.11.0 (registry+https://github.com/rust-lang/crates.io-index)" = "d326610f408c7a4eb6f51c37c330e496b08506c9457c9d34287ecc38809fb060"
-"checksum unicode-width 0.1.6 (registry+https://github.com/rust-lang/crates.io-index)" = "7007dbd421b92cc6e28410fe7362e2e0a2503394908f417b68ec8d1c364c4e20"
+"checksum unicode-width 0.1.7 (registry+https://github.com/rust-lang/crates.io-index)" = "caaa9d531767d1ff2150b9332433f32a24622147e5ebb1f26409d5da67afd479"
-"checksum vec_map 0.8.1 (registry+https://github.com/rust-lang/crates.io-index)" = "05c78687fb1a80548ae3250346c3db86a80a7cdd77bda190189f2d0a0987c81a"
+"checksum vec_map 0.8.2 (registry+https://github.com/rust-lang/crates.io-index)" = "f1bddf1187be692e79c5ffeab891132dfb0f236ed36a43c7ed39f1165ee20191"
-"checksum void 1.0.2 (registry+https://github.com/rust-lang/crates.io-index)" = "6a02e4885ed3bc0f2de90ea6dd45ebcbb66dacffe03547fadbb0eeae2770887d"
-"checksum winapi 0.2.8 (registry+https://github.com/rust-lang/crates.io-index)" = "167dc9d6949a9b857f3451275e911c3f44255842c1f7a76f33c55103a909087a"
 "checksum winapi 0.3.8 (registry+https://github.com/rust-lang/crates.io-index)" = "8093091eeb260906a183e6ae1abdba2ef5ef2257a21801128899c3fc699229c6"
-"checksum winapi-build 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "2d315eee3b34aca4797b2da6b13ed88266e6d612562a0c46390af8299fc699bc"
 "checksum winapi-i686-pc-windows-gnu 0.4.0 (registry+https://github.com/rust-lang/crates.io-index)" = "ac3b87c63620426dd9b991e5ce0329eff545bccbbb34f3be09ff6fb6ab51b7b6"
 "checksum winapi-x86_64-pc-windows-gnu 0.4.0 (registry+https://github.com/rust-lang/crates.io-index)" = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f"

Cargo.toml

@@ -7,13 +7,13 @@ license = "MIT/Apache-2.0"
 name = "hexyl"
 readme = "README.md"
 repository = "https://github.com/sharkdp/hexyl"
-version = "0.6.0"
+version = "0.8.0"
 edition = "2018"
 
 [dependencies]
 ansi_term = "0.12"
 atty = "0.2"
-ctrlc = "3.1"
+libc = "0.2"
 
 [dependencies.clap]
 version = "2"

README.md

@@ -19,13 +19,31 @@ of bytes (NULL bytes, printable ASCII characters, ASCII whitespace characters, o
 
 ## Installation
 
-### On Debian-based systems
+### On Ubuntu
+
+*... and other Debian-based Linux distributions.*
+
+If you run Ubuntu 19.10 (Eoan Ermine) or newer, you can install the [officially maintained package](https://packages.ubuntu.com/eoan/hexyl):
+```bash
+sudo apt install hexyl
+```
+If you use an older version of Ubuntu, you can download
+the latest `.deb` package from the release page and install it via:
 
 ``` bash
-wget "https://github.com/sharkdp/hexyl/releases/download/v0.6.0/hexyl_0.6.0_amd64.deb"
+sudo dpkg -i hexyl_0.7.0_amd64.deb  # adapt version number and architecture
-sudo dpkg -i hexyl_0.6.0_amd64.deb
 ```
 
+### On Debian
+
+If you run Debian Buster or newer, you can install the [officially maintained Debian package](https://packages.debian.org/buster/hexyl):
+```bash
+sudo apt-get install hexyl
+```
+
+If you run an older version of Debian, see above for instructions on how to
+manually install `hexyl`.
+
 ### On Arch Linux
 
 You can install `hexyl` from [the official package repository](https://www.archlinux.org/packages/community/x86_64/hexyl/):
@@ -34,6 +52,11 @@ You can install `hexyl` from [the official package repository](https://www.archl
 pacman -S hexyl
 ```
 
+### On Void Linux
+```
+xbps-install hexyl
+```
+
 ### On macOS
 
 ```
@@ -58,12 +81,13 @@ Check out the [release page](https://github.com/sharkdp/hexyl/releases) for bina
 
 ### On Windows
 
-For now, you will have to install from source via `cargo` (see below). Make sure that
+For now, you will have to install from source via `cargo` (see below). Make sure that you
-you use a terminal that supports ANSI escape sequences (like PowerShell on Windows 10).
+use a terminal that supports ANSI escape sequences (like ConHost v2 since Windows 10 1703
+or Windows Terminal since Windows 10 1903).
 
 ### Via cargo
 
-If you have Rust 1.31 or higher, you can install `hexyl` from source via `cargo`:
+If you have Rust 1.36 or higher, you can install `hexyl` from source via `cargo`:
 ```
 cargo install hexyl
 ```

examples/simple.rs

@@ -0,0 +1,20 @@
+use std::io;
+
+use hexyl::{BorderStyle, Printer};
+
+fn main() {
+    let input = vec![
+        0x89, 0x50, 0x4e, 0x47, 0x0d, 0x0a, 0x1a, 0x0a, 0x00, 0x00, 0x00, 0x0d, 0x49, 0x48, 0x44,
+        0x52, 0x00, 0x00, 0x00, 0x80, 0x00, 0x00, 0x00, 0x44, 0x08, 0x02, 0x00, 0x00, 0x00,
+    ];
+
+    let stdout = io::stdout();
+    let mut handle = stdout.lock();
+
+    let show_color = true;
+    let use_squeezing = false;
+    let border_style = BorderStyle::Unicode;
+
+    let mut printer = Printer::new(&mut handle, show_color, border_style, use_squeezing);
+    printer.print_all(&input[..]).unwrap();
+}

src/bin/hexyl.rs

@@ -1,19 +1,15 @@
 #[macro_use]
 extern crate clap;
 
-use atty;
+use std::convert::TryFrom;
-use ctrlc;
-
 use std::fs::File;
-use std::io::{self, prelude::*};
+use std::io::{self, prelude::*, SeekFrom};
-use std::sync::atomic::{AtomicBool, Ordering};
-use std::sync::Arc;
 
 use clap::{App, AppSettings, Arg};
 
 use atty::Stream;
 
-use hexyl::{BorderStyle, Printer};
+use hexyl::{BorderStyle, Input, Printer};
 
 fn run() -> Result<(), Box<dyn std::error::Error>> {
     let app = App::new(crate_name!())
@@ -23,14 +19,21 @@ fn run() -> Result<(), Box<dyn std::error::Error>> {
         .setting(AppSettings::UnifiedHelpMessage)
         .version(crate_version!())
         .about(crate_description!())
-        .arg(Arg::with_name("file").help("File to display"))
+        .arg(
+            Arg::with_name("FILE")
+                .help("The file to display. If no FILE argument is given, read from STDIN."),
+        )
         .arg(
             Arg::with_name("length")
                 .short("n")
                 .long("length")
                 .takes_value(true)
                 .value_name("N")
-                .help("Read only N bytes from the input"),
+                .help(
+                    "Only read N bytes from the input. The N argument can also include a \
+                      unit with a decimal prefix (kB, MB, ..) or binary prefix (kiB, MiB, ..).\n\
+                      Examples: --length=64, --length=4KiB",
+                ),
         )
         .arg(
             Arg::with_name("bytes")
@@ -40,6 +43,27 @@ fn run() -> Result<(), Box<dyn std::error::Error>> {
                 .value_name("N")
                 .help("An alias for -n/--length"),
         )
+        .arg(
+            Arg::with_name("skip")
+                .short("s")
+                .long("skip")
+                .takes_value(true)
+                .value_name("N")
+                .help(
+                    "Skip the first N bytes of the input. The N argument can also include \
+                      a unit (see `--length` for details)",
+                ),
+        )
+        .arg(
+            Arg::with_name("block_size")
+                .long("block-size")
+                .takes_value(true)
+                .value_name("SIZE")
+                .help(
+                    "Sets the size of the `block` unit to SIZE.\n\
+                    Examples: --block-size=1024, --block-size=4kB",
+                ),
+        )
         .arg(
             Arg::with_name("nosqueezing")
                 .short("v")
@@ -54,7 +78,7 @@ fn run() -> Result<(), Box<dyn std::error::Error>> {
             Arg::with_name("color")
                 .long("color")
                 .takes_value(true)
-                .value_name("when")
+                .value_name("WHEN")
                 .possible_values(&["always", "auto", "never"])
                 .default_value("always")
                 .help(
@@ -66,34 +90,54 @@ fn run() -> Result<(), Box<dyn std::error::Error>> {
             Arg::with_name("border")
                 .long("border")
                 .takes_value(true)
+                .value_name("STYLE")
                 .possible_values(&["unicode", "ascii", "none"])
                 .default_value("unicode")
-                .help("Whether to draw a border with unicode or ASCII characters, or none at all"),
+                .help("Whether to draw a border with Unicode characters, ASCII characters, or none at all"),
         )
         .arg(
             Arg::with_name("display_offset")
                 .short("o")
                 .long("display-offset")
                 .takes_value(true)
-                .value_name("OFFSET")
+                .value_name("N")
-                .help("Add OFFSET to the displayed file position."),
+                .help("Add N bytes to the displayed file position. The N argument can also include \
+                a unit (see `--length` for details)"),
         );
 
     let matches = app.get_matches_safe()?;
 
     let stdin = io::stdin();
 
-    let mut reader: Box<dyn Read> = match matches.value_of("file") {
+    let mut reader: Input = match matches.value_of("FILE") {
-        Some(filename) => Box::new(File::open(filename)?),
+        Some(filename) => Input::File(File::open(filename)?),
-        None => Box::new(stdin.lock()),
+        None => Input::Stdin(stdin.lock()),
     };
 
-    let length_arg = matches.value_of("length").or(matches.value_of("bytes"));
+    let block_size = matches
+        .value_of("block_size")
+        .and_then(|bs| bs.parse().ok().and_then(PositiveI64::new))
+        .unwrap_or_else(|| PositiveI64::new(512).unwrap());
 
-    if let Some(length) = length_arg.and_then(parse_hex_or_int) {
+    let skip_arg = matches
-        reader = Box::new(reader.take(length));
+        .value_of("skip")
+        .and_then(|s| parse_byte_count(s, block_size));
+
+    if let Some(skip) = skip_arg {
+        reader.seek(SeekFrom::Current(skip.into_inner()))?;
     }
 
+    let length_arg = matches
+        .value_of("length")
+        .or_else(|| matches.value_of("bytes"));
+
+    let mut reader = if let Some(length) = length_arg.and_then(|s| parse_byte_count(s, block_size))
+    {
+        Box::new(reader.take(length.into()))
+    } else {
+        reader.into_inner()
+    };
+
     let show_color = match matches.value_of("color") {
         Some("never") => false,
         Some("auto") => atty::is(Stream::Stdout),
@@ -110,24 +154,17 @@ fn run() -> Result<(), Box<dyn std::error::Error>> {
 
     let display_offset = matches
         .value_of("display_offset")
-        .and_then(parse_hex_or_int)
+        .and_then(|s| parse_byte_count(s, block_size))
-        .unwrap_or(0);
+        .or(skip_arg)
-
+        .unwrap_or_default()
-    // Set up Ctrl-C handler
+        .into();
-    let cancelled = Arc::new(AtomicBool::new(false));
-    let c = cancelled.clone();
-
-    ctrlc::set_handler(move || {
-        c.store(true, Ordering::SeqCst);
-    })
-    .expect("Error setting Ctrl-C handler");
 
     let stdout = io::stdout();
     let mut stdout_lock = stdout.lock();
 
     let mut printer = Printer::new(&mut stdout_lock, show_color, border_style, squeeze);
-    printer.display_offset(display_offset as usize);
+    printer.display_offset(display_offset);
-    printer.print_all(&mut reader, Some(cancelled))?;
+    printer.print_all(&mut reader)?;
 
     Ok(())
 }
@@ -157,10 +194,159 @@ fn main() {
     }
 }
 
-fn parse_hex_or_int(n: &str) -> Option<u64> {
+#[derive(Clone, Copy, Debug, Default, Hash, Eq, Ord, PartialEq, PartialOrd)]
-    if n.starts_with("0x") {
+pub struct PositiveI64(i64);
-        u64::from_str_radix(n.trim_start_matches("0x"), 16).ok()
+
+impl PositiveI64 {
+    pub fn new(x: i64) -> Option<Self> {
+        if x.is_negative() {
+            None
         } else {
-        n.parse::<u64>().ok()
+            Some(Self(x))
+        }
+    }
+
+    pub fn into_inner(self) -> i64 {
+        self.0
     }
 }
+
+impl Into<u64> for PositiveI64 {
+    fn into(self) -> u64 {
+        u64::try_from(self.0)
+            .expect("invariant broken: PositiveI64 should contain a positive i64 value")
+    }
+}
+
+fn parse_byte_count(n: &str, block_size: PositiveI64) -> Option<PositiveI64> {
+    const HEX_PREFIX: &str = "0x";
+
+    let n = {
+        let mut chars = n.chars();
+        match chars.next()? {
+            '+' => chars.as_str(),
+            '-' => return None,
+            _ => n,
+        }
+    };
+
+    if n.starts_with(HEX_PREFIX) {
+        let n = &n[HEX_PREFIX.len()..];
+        match n.chars().next() {
+            Some('+') | Some('-') => return None,
+            _ => (),
+        }
+        return i64::from_str_radix(n, 16).ok().and_then(PositiveI64::new);
+    }
+
+    let (n, unit_multiplier) = match n.chars().position(|c| !c.is_ascii_digit()) {
+        Some(unit_begin_idx) => {
+            let (n, raw_unit) = n.split_at(unit_begin_idx);
+            let raw_unit = raw_unit.to_lowercase();
+            (
+                n,
+                [
+                    ("b", 1),
+                    ("kb", 1000i64.pow(1)),
+                    ("mb", 1000i64.pow(2)),
+                    ("gb", 1000i64.pow(3)),
+                    ("tb", 1000i64.pow(4)),
+                    ("kib", 1024i64.pow(1)),
+                    ("mib", 1024i64.pow(2)),
+                    ("gib", 1024i64.pow(3)),
+                    ("tib", 1024i64.pow(4)),
+                    ("block", block_size.into_inner()),
+                ]
+                .iter()
+                .cloned()
+                .find_map(|(unit, multiplier)| {
+                    if unit == raw_unit {
+                        Some(multiplier)
+                    } else {
+                        None
+                    }
+                })?,
+            )
+        }
+        None => (n, 1),
+    };
+
+    n.parse::<i64>()
+        .ok()?
+        .checked_mul(unit_multiplier)
+        .and_then(PositiveI64::new)
+}
+
+#[test]
+fn test_parse_byte_count() {
+    macro_rules! success {
+        ($input: expr, $expected: expr) => {
+            success!($input, 512, $expected)
+        };
+        ($input: expr, $block_size: expr, $expected: expr) => {
+            assert_eq!(
+                parse_byte_count($input, PositiveI64::new($block_size).unwrap()),
+                Some(PositiveI64::new($expected).unwrap())
+            );
+        };
+    }
+
+    macro_rules! error {
+        ($input: expr) => {
+            assert_eq!(
+                parse_byte_count($input, PositiveI64::new(512).unwrap()),
+                None
+            );
+        };
+    }
+
+    success!("0", 0);
+    success!("1", 1);
+    success!("1", 1);
+    success!("100", 100);
+    success!("+100", 100);
+
+    success!("1KB", 1000);
+    success!("2MB", 2000000);
+    success!("3GB", 3000000000);
+    success!("4TB", 4000000000000);
+    success!("+4TB", 4000000000000);
+
+    success!("1GiB", 1073741824);
+    success!("2TiB", 2199023255552);
+    success!("+2TiB", 2199023255552);
+
+    success!("0xff", 255);
+    success!("0xEE", 238);
+    success!("+0xFF", 255);
+
+    success!("1block", 512, 512);
+    success!("2block", 512, 1024);
+    success!("1block", 4, 4);
+    success!("2block", 4, 8);
+
+    // empty string is invalid
+    error!("");
+    // These are also bad.
+    error!("+");
+    error!("-");
+    error!("K");
+    error!("k");
+    error!("m");
+    error!("block");
+    // leading/trailing space is invalid
+    error!(" 0");
+    error!("0 ");
+    // Negatives make no sense for byte counts
+    error!("-1");
+    error!("0x-12");
+    // This was previously accepted but shouldn't be.
+    error!("0x+12");
+    // invalid suffix
+    error!("1234asdf");
+    // bad numbers
+    error!("asdf1234");
+    error!("a1s2d3f4");
+    // multiplication overflows u64
+    error!("20000000TiB");
+}

src/input.rs

@@ -0,0 +1,65 @@
+use std::convert::TryFrom;
+use std::fs;
+use std::io::{self, copy, sink, Read, Seek, SeekFrom};
+
+pub enum Input<'a> {
+    File(fs::File),
+    Stdin(io::StdinLock<'a>),
+}
+
+impl<'a> Read for Input<'a> {
+    fn read(&mut self, buf: &mut [u8]) -> io::Result<usize> {
+        match *self {
+            Input::File(ref mut file) => file.read(buf),
+            Input::Stdin(ref mut stdin) => stdin.read(buf),
+        }
+    }
+}
+
+impl<'a> Seek for Input<'a> {
+    fn seek(&mut self, pos: SeekFrom) -> io::Result<u64> {
+        fn try_skip<R>(reader: R, pos: SeekFrom, err_desc: &'static str) -> io::Result<u64>
+        where
+            R: Read,
+        {
+            let cant_seek_abs_err = || Err(io::Error::new(io::ErrorKind::Other, err_desc));
+
+            let offset = match pos {
+                SeekFrom::Current(o) => u64::try_from(o).or_else(|_e| cant_seek_abs_err())?,
+                SeekFrom::Start(_) | SeekFrom::End(_) => cant_seek_abs_err()?,
+            };
+
+            copy(&mut reader.take(offset), &mut sink())
+        }
+
+        match *self {
+            Input::File(ref mut file) => {
+                let seek_res = file.seek(pos);
+                if let Err(Some(libc::ESPIPE)) = seek_res.as_ref().map_err(|err| err.raw_os_error())
+                {
+                    try_skip(
+                        file,
+                        pos,
+                        "Pipes only support seeking forward with a relative offset",
+                    )
+                } else {
+                    seek_res
+                }
+            }
+            Input::Stdin(ref mut stdin) => try_skip(
+                stdin,
+                pos,
+                "STDIN only supports seeking forward with a relative offset",
+            ),
+        }
+    }
+}
+
+impl<'a> Input<'a> {
+    pub fn into_inner(self) -> Box<dyn Read + 'a> {
+        match self {
+            Input::File(file) => Box::new(file),
+            Input::Stdin(stdin) => Box::new(stdin),
+        }
+    }
+}

src/lib.rs

@@ -1,16 +1,15 @@
+pub(crate) mod input;
 pub mod squeezer;
 
+pub use input::*;
+
 use std::io::{self, Read, Write};
-use std::sync::atomic::{AtomicBool, Ordering};
-use std::sync::Arc;
 
 use ansi_term::Color;
 use ansi_term::Color::Fixed;
 
 use crate::squeezer::{SqueezeAction, Squeezer};
 
-type Cancelled = Arc<AtomicBool>;
-
 const BUFFER_SIZE: usize = 256;
 
 const COLOR_NULL: Color = Fixed(242); // grey
@@ -35,10 +34,7 @@ impl Byte {
     fn category(self) -> ByteCategory {
         if self.0 == 0x00 {
             ByteCategory::Null
-        } else if self.0.is_ascii_alphanumeric()
+        } else if self.0.is_ascii_graphic() {
-            || self.0.is_ascii_punctuation()
-            || self.0.is_ascii_graphic()
-        {
             ByteCategory::AsciiPrintable
         } else if self.0.is_ascii_whitespace() {
             ByteCategory::AsciiWhitespace
@@ -143,7 +139,7 @@ impl BorderStyle {
 }
 
 pub struct Printer<'a, Writer: Write> {
-    idx: usize,
+    idx: u64,
     /// The raw bytes used as input for the current line.
     raw_line: Vec<u8>,
     /// The buffered line built with each byte, ready to print to writer.
@@ -155,7 +151,7 @@ pub struct Printer<'a, Writer: Write> {
     byte_hex_table: Vec<String>,
     byte_char_table: Vec<String>,
     squeezer: Squeezer,
-    display_offset: usize,
+    display_offset: u64,
 }
 
 impl<'a, Writer: Write> Printer<'a, Writer> {
@@ -198,7 +194,7 @@ impl<'a, Writer: Write> Printer<'a, Writer> {
         }
     }
 
-    pub fn display_offset(&mut self, display_offset: usize) -> &mut Self {
+    pub fn display_offset(&mut self, display_offset: u64) -> &mut Self {
         self.display_offset = display_offset;
         self
     }
@@ -248,7 +244,7 @@ impl<'a, Writer: Write> Printer<'a, Writer> {
         }
 
         let style = COLOR_OFFSET.normal();
-        let byte_index = format!("{:08x}", (self.idx - 1) + self.display_offset);
+        let byte_index = format!("{:08x}", self.idx - 1 + self.display_offset);
         let formatted_string = if self.show_color {
             format!("{}", style.paint(byte_index))
         } else {
@@ -408,11 +404,10 @@ impl<'a, Writer: Write> Printer<'a, Writer> {
     }
 
     /// Loop through the given `Reader`, printing until the `Reader` buffer
-    /// is exhausted, or the optional `cancelled` bool is set to true.
+    /// is exhausted.
     pub fn print_all<Reader: Read>(
         &mut self,
         mut reader: Reader,
-        canceller: Option<Cancelled>,
     ) -> Result<(), Box<dyn std::error::Error>> {
         let mut buffer = [0; BUFFER_SIZE];
         'mainloop: loop {
@@ -421,13 +416,6 @@ impl<'a, Writer: Write> Printer<'a, Writer> {
                 break;
             }
 
-            if let Some(cancelled) = &canceller {
-                if cancelled.load(Ordering::SeqCst) {
-                    eprintln!("hexyl has been cancelled.");
-                    std::process::exit(130); // Set exit code to 128 + SIGINT
-                }
-            }
-
             for b in &buffer[..size] {
                 let res = self.print_byte(*b);
 
@@ -460,7 +448,7 @@ mod tests {
         let mut output = vec![];
         let mut printer = Printer::new(&mut output, false, BorderStyle::Unicode, true);
 
-        printer.print_all(input, None).unwrap();
+        printer.print_all(input).unwrap();
 
         let actual_string: &str = str::from_utf8(&output).unwrap();
         assert_eq!(actual_string, expected_string)
@@ -469,8 +457,8 @@ mod tests {
     #[test]
     fn empty_file_passes() {
         let input = io::empty();
-        let expected_string =
+        let expected_string = "\
-            "┌────────┬─────────────────────────┬─────────────────────────┬────────┬────────┐
+┌────────┬─────────────────────────┬─────────────────────────┬────────┬────────┐
 └────────┴─────────────────────────┴─────────────────────────┴────────┴────────┘
 "
         .to_owned();
@@ -480,8 +468,8 @@ mod tests {
     #[test]
     fn short_input_passes() {
         let input = io::Cursor::new(b"spam");
-        let expected_string =
+        let expected_string = "\
-            "┌────────┬─────────────────────────┬─────────────────────────┬────────┬────────┐
+┌────────┬─────────────────────────┬─────────────────────────┬────────┬────────┐
 │00000000│ 73 70 61 6d             ┊                         │spam    ┊        │
 └────────┴─────────────────────────┴─────────────────────────┴────────┴────────┘
 "
@@ -492,8 +480,8 @@ mod tests {
     #[test]
     fn display_offset() {
         let input = io::Cursor::new(b"spamspamspamspamspam");
-        let expected_string =
+        let expected_string = "\
-            "┌────────┬─────────────────────────┬─────────────────────────┬────────┬────────┐
+┌────────┬─────────────────────────┬─────────────────────────┬────────┬────────┐
 │deadbeef│ 73 70 61 6d 73 70 61 6d ┊ 73 70 61 6d 73 70 61 6d │spamspam┊spamspam│
 │deadbeff│ 73 70 61 6d             ┊                         │spam    ┊        │
 └────────┴─────────────────────────┴─────────────────────────┴────────┴────────┘
@@ -505,7 +493,7 @@ mod tests {
             Printer::new(&mut output, false, BorderStyle::Unicode, true);
         printer.display_offset(0xdeadbeef);
 
-        printer.print_all(input, None).unwrap();
+        printer.print_all(input).unwrap();
 
         let actual_string: &str = str::from_utf8(&output).unwrap();
         assert_eq!(actual_string, expected_string)

src/squeezer.rs

@@ -32,7 +32,7 @@ pub enum SqueezeAction {
 }
 
 /// line size
-const LSIZE: usize = 16;
+const LSIZE: u64 = 16;
 
 impl Squeezer {
     pub fn new(enabled: bool) -> Squeezer {
@@ -46,7 +46,7 @@ impl Squeezer {
         }
     }
 
-    pub fn process(&mut self, b: u8, i: usize) {
+    pub fn process(&mut self, b: u8, i: u64) {
         use self::SqueezeState::*;
         if self.state == Disabled {
             return;
@@ -111,10 +111,12 @@ impl Squeezer {
 mod tests {
     use super::*;
 
+    const LSIZE_USIZE: usize = LSIZE as usize;
+
     #[test]
     fn three_same_lines() {
         const LINES: usize = 3;
-        let v = vec![0u8; LINES * LSIZE];
+        let v = vec![0u8; LINES * LSIZE_USIZE];
         let mut s = Squeezer::new(true);
         // just initialized
         assert_eq!(s.action(), SqueezeAction::Ignore);
@@ -128,7 +130,7 @@ mod tests {
 
         let mut line = 0;
         let mut idx = 1;
-        for z in v.chunks(LSIZE) {
+        for z in v.chunks(LSIZE_USIZE) {
             for i in z {
                 s.process(*i, idx);
                 idx += 1;
@@ -143,7 +145,7 @@ mod tests {
     #[test]
     fn incomplete_while_squeeze() {
         // fourth line only has 12 bytes and should be printed
-        let v = vec![0u8; 3 * LSIZE + 12];
+        let v = vec![0u8; 3 * LSIZE_USIZE + 12];
         let mut s = Squeezer::new(true);
         // just initialized
         assert_eq!(s.action(), SqueezeAction::Ignore);
@@ -158,7 +160,7 @@ mod tests {
 
         let mut line = 0;
         let mut idx = 1;
-        for z in v.chunks(LSIZE) {
+        for z in v.chunks(LSIZE_USIZE) {
             for i in z {
                 s.process(*i, idx);
                 idx += 1;
@@ -190,7 +192,7 @@ mod tests {
 
         let mut line = 0;
         let mut idx = 1;
-        for z in v.chunks(LSIZE) {
+        for z in v.chunks(LSIZE_USIZE) {
             for i in z {
                 s.process(*i, idx);
                 idx += 1;
@@ -207,7 +209,7 @@ mod tests {
     /// print
     fn one_squeeze_no_delete() {
         const LINES: usize = 3;
-        let mut v = vec![0u8; (LINES - 1) * LSIZE];
+        let mut v = vec![0u8; (LINES - 1) * LSIZE_USIZE];
         v.extend(vec![1u8; 16]);
 
         let mut s = Squeezer::new(true);
@@ -223,7 +225,7 @@ mod tests {
 
         let mut line = 0;
         let mut idx = 1;
-        for z in v.chunks(LSIZE) {
+        for z in v.chunks(LSIZE_USIZE) {
             for i in z {
                 s.process(*i, idx);
                 idx += 1;
@@ -239,7 +241,7 @@ mod tests {
     /// First line all eq, 2nd half eq with first line, then change
     fn second_line_different() {
         const LINES: usize = 2;
-        let mut v = vec![0u8; (LINES - 1) * LSIZE];
+        let mut v = vec![0u8; (LINES - 1) * LSIZE_USIZE];
         v.extend(vec![0u8; 8]);
         v.extend(vec![1u8; 8]);
 
@@ -255,7 +257,7 @@ mod tests {
 
         let mut line = 0;
         let mut idx = 1;
-        for z in v.chunks(LSIZE) {
+        for z in v.chunks(LSIZE_USIZE) {
             for i in z {
                 s.process(*i, idx);
                 idx += 1;
@@ -291,7 +293,7 @@ mod tests {
 
         let mut line = 0;
         let mut idx = 1;
-        for z in v.chunks(LSIZE) {
+        for z in v.chunks(LSIZE_USIZE) {
             for i in z {
                 s.process(*i, idx);
                 idx += 1;
@@ -341,7 +343,7 @@ mod tests {
 
         let mut line = 0;
         let mut idx = 1;
-        for z in v.chunks(LSIZE) {
+        for z in v.chunks(LSIZE_USIZE) {
             for i in z {
                 s.process(*i, idx);
                 idx += 1;
@@ -377,7 +379,7 @@ mod tests {
 
         let mut line = 0;
         let mut idx = 1;
-        for z in v.chunks(LSIZE) {
+        for z in v.chunks(LSIZE_USIZE) {
             for i in z {
                 s.process(*i, idx);
                 idx += 1;
@@ -410,7 +412,7 @@ mod tests {
 
         let mut line = 0;
         let mut idx = 1;
-        for z in v.chunks(LSIZE) {
+        for z in v.chunks(LSIZE_USIZE) {
             for i in z {
                 s.process(*i, idx);
                 idx += 1;