prepare version 0.2.0 (#519 )

fix: identify uutils in cli output (#435 )
sed: fix clippy cast_possible_truncation errors (#518 )
2026-10-16 02:48:04 -07:00 · 2026-08-09 01:57:25 +02:00 · 2026-08-09 00:55:37 +02:00 · 2026-08-09 00:52:26 +02:00 · 2026-08-08 23:28:22 +02:00 · 2026-08-07 08:07:56 +02:00
23 changed files with 3388 additions and 1184 deletions
@@ -33,7 +33,7 @@ jobs:
      - uses: Swatinem/rust-cache@v2

      - name: Run sccache-cache
-        uses: mozilla-actions/sccache-action@v0.0.10
+        uses: mozilla-actions/sccache-action@v0.0.11

      - name: Install tools
        uses: taiki-e/install-action@v2
@@ -47,7 +47,7 @@ jobs:
          cargo codspeed build -p ${{ matrix.benchmark-target.package }}

      - name: Run benchmarks for ${{ matrix.benchmark-target.package }}
-        uses: CodSpeedHQ/action@v4
+        uses: CodSpeedHQ/action@v5
        env:
          CODSPEED_LOG: debug
        with:
@@ -51,7 +51,7 @@ jobs:
          components: clippy
      - uses: Swatinem/rust-cache@v2
      - name: Run sccache-cache
-        uses: mozilla-actions/sccache-action@v0.0.10
+        uses: mozilla-actions/sccache-action@v0.0.11
      - name: Initialize workflow variables
        id: vars
        shell: bash
@@ -5,8 +5,7 @@

 [package]
 name = "sed"
-version = "0.1.1"
-authors = ["uutils developers"]
+version = "0.2.0"
 license = "MIT"
 description = "sed ~ implemented as universal (cross-platform) utils, written in Rust"
 default-run = "sed"
@@ -29,7 +28,6 @@ sed = []

 [workspace.dependencies]
 assert_fs = "1.1.3"
-bytesize = "2.0.0"
 chrono = { version = "0.4.37", default-features = false, features = [
  "clock",
 ] }
@@ -48,11 +46,11 @@ predicates = "3.1.3"
 rand = { version = "0.10.0" }
 regex = "1.10.4"
 sha2 = "0.11"
-sysinfo = "0.38"
 tempfile = "3.10.1"
 terminal_size = "0.4.2"
 textwrap = { version = "0.16.1", features = ["terminal_size"] }
-uucore = { version = "0.9.0", features = ["libc"] }
+uucore = { version = "0.10.0", features = ["libc"] }
+rustix = "1.1.4"
 xattr = "1.3.1"


@@ -66,11 +64,11 @@ memmap2.workspace = true
 phf = { workspace = true }
 predicates = { workspace = true }
 regex = { workspace = true }
-sysinfo = { workspace = true }
 tempfile = { workspace = true }
 terminal_size = { workspace = true }
 textwrap = { workspace = true }
 uucore = { workspace = true }
+rustix = { workspace = true }

 [dev-dependencies]
 assert_fs = { workspace = true }
@@ -85,7 +83,7 @@ regex = { workspace = true }
 sha2 = { workspace = true }
 tempfile = { workspace = true }
 uucore = { workspace = true, features = ["entries", "process", "signals", "benchmark"] }
-uutests = "0.9.0"
+uutests = "0.10.0"

 [target.'cfg(unix)'.dev-dependencies]
 xattr = { workspace = true }
@@ -135,7 +133,7 @@ must_use_candidate = "allow"                 # 56
 needless_raw_string_hashes = "allow"         # 20
 needless_pass_by_value = "allow"             # 15
 missing_panics_doc = "allow"                 # 12
-cast_possible_truncation = "allow"           # 7
+# cast_possible_truncation = "allow"           # 7
 unnecessary_wraps = "allow"                  # 6
 match_wildcard_for_single_variants = "allow" # 4
 cast_sign_loss = "allow"                     # 4
@@ -26,7 +26,7 @@ Further work aims to:
 * Implement more GNU extensions,
 * Improve performance where possible.

-## Installation
+## Installation and Use

 We provide a Linux x86_64 binary archive from the main branch at
 https://github.com/uutils/sed/releases/tag/latest-commit .
@@ -44,6 +44,10 @@ cargo run --release

 The binary is named `sed` in `target/release/sed`.

+You can also try *sed* on the web
+through the [uutils Playground](https://uutils.org//playground/)
+by clicking on the `Load sed` button.
+
 ## Testing

 ### GNU sed Compatibility Testing
@@ -81,6 +85,7 @@ cargo test
 * Spaces can precede a regular expression modifier.
 * `I` can be used in as a synonym for the `i` (case insensitive) substitution
  flag.
+* `M` and `m` substitution flags allow multi-line matching.
 * In addition to `\n`, other escape sequences (octal, hex, C) are supported
  in the strings of the `y` command.
  Under POSIX these yield undefined behavior.
@@ -92,11 +97,13 @@ cargo test
 * A `Q` command (optionally followed by an exit code) quits immediately.
 * The `q` command can be optionally followed by an exit code.
 * The `l` command can be optionally followed by the output width.
-* The `--follow-symlinks` flag for in-place editing.
+* The `--follow-symlinks` option for in-place editing.
+* The `--sandbox` option that limits potentially destructive commands.
 * Address 0 can be used to specify an address range that is already
  active on line 1 and can finish with the specified regular expression.
 * Address steps can be specified in the form of start~step and start,~step
  ranges.
+* Address 0 can be used in the `r` command to prepend a file.

 ### Supported BSD and GNU extensions
 * The second address in a range can be specified as a relative address with +N.
@@ -109,12 +116,19 @@ cargo test
  sequences.

 ### Incompatibilities
-* The input is assumed to be valid UTF-8 (this includes 7-bit ASCII).
-  If the input is in another code page, consider converting it through UTF-8
-  in order to avoid errors on invalid UTF-8 sequences and for the correct
-  handling of regular expressions.
-  This _sed_ program can also handle arbitrary byte sequences if no part of the
-  input is treated as string.
+* Similarly to GNU _sed_, input is processed as raw bytes or as valid UTF-8
+  (this includes 7-bit ASCII) based on the locale as specified by the
+  `LC_ALL`, `LC_CTYPE`, and `LANG` environment variables,
+  with the default being byte processing.
+  However, in contrast with GNU _sed_, other locales (e.g. ISO-8859-1)
+  are not supported. If the input is in another code page or encoding
+  and requires locale-specific processing (e.g. ignore/map case,
+  character classes), consider converting it through UTF-8 to ensure
+  the correct handling of locale-specific regular expressions.
+  This _sed_ program can also handle arbitrary byte sequences
+  if no part of the input requires treating it as a Rust String.
+* Back-references aren't supported when input is processed as bytes
+  (`LC_ALL=C`).
 * The command will report an error and fail if duplicate labels are found
  in the script.
  This matches the BSD behavior. The GNU version accepts duplicate labels.
@@ -245,9 +245,9 @@ checksum = "1d07550c9036bf2ae0c684c4297d503f838287c83c53686d05370d0e139ae570"

 [[package]]
 name = "console"
-version = "0.16.3"
+version = "0.16.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d64e8af5551369d19cf50138de61f1c42074ab970f74e99be916646777f8fc87"
+checksum = "4fe5f465a4f6fee88fad41b85d990f84c835335e85b5d9e6e63e0d06d28cba7c"
 dependencies = [
 "encode_unicode",
 "libc",
@@ -571,9 +571,9 @@ checksum = "09edd9e8b54e49e587e4f6295a7d29c3ea94d469cb40ab8ca70b288248a81db2"

 [[package]]
 name = "libc"
-version = "0.2.186"
+version = "0.2.189"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "68ab91017fe16c622486840e4c83c9a37afeff978bd239b5293d61ece587de66"
+checksum = "3eaf3ede3fee6db1a4c2ee091bf8a8b4dccdc6d17f656fb07896ee72867612f2"

 [[package]]
 name = "libfuzzer-sys"
@@ -646,15 +646,6 @@ version = "0.3.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "61807f77802ff30975e01f4f071c8ba10c022052f98b3294119f3e615d13e5be"

-[[package]]
-name = "ntapi"
-version = "0.4.3"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c3b335231dfd352ffb0f8017f3b6027a4917f7df785ea2143d8af2adc66980ae"
-dependencies = [
- "winapi",
-]
-
 [[package]]
 name = "num-bigint"
 version = "0.4.6"
@@ -683,25 +674,6 @@ dependencies = [
 "autocfg",
 ]

-[[package]]
-name = "objc2-core-foundation"
-version = "0.3.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2a180dd8642fa45cdb7dd721cd4c11b1cadd4929ce112ebd8b9f5803cc79d536"
-dependencies = [
- "bitflags",
-]
-
-[[package]]
-name = "objc2-io-kit"
-version = "0.3.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "33fafba39597d6dc1fb709123dfa8289d39406734be322956a69f0931c73bb15"
-dependencies = [
- "libc",
- "objc2-core-foundation",
-]
-
 [[package]]
 name = "once_cell"
 version = "1.21.4"
@@ -725,9 +697,9 @@ dependencies = [

 [[package]]
 name = "phf"
-version = "0.13.1"
+version = "0.14.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c1562dc717473dbaa4c1f85a36410e03c047b2e7df7f45ee938fbef64ae7fadf"
+checksum = "010378780309880b08997fae13be7834dba947d36393bd372f2b1556deb2a2f6"
 dependencies = [
 "phf_shared",
 "serde",
@@ -735,9 +707,9 @@ dependencies = [

 [[package]]
 name = "phf_codegen"
-version = "0.13.1"
+version = "0.14.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "49aa7f9d80421bca176ca8dbfebe668cc7a2684708594ec9f3c0db0805d5d6e1"
+checksum = "41b585a510fb76fdebead6897982ef2a03a21d8e6cbcca904999742a4afc6ffe"
 dependencies = [
 "phf_generator",
 "phf_shared",
@@ -745,9 +717,9 @@ dependencies = [

 [[package]]
 name = "phf_generator"
-version = "0.13.1"
+version = "0.14.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "135ace3a761e564ec88c03a77317a7c6b80bb7f7135ef2544dbe054243b89737"
+checksum = "aeb62e0959d5a1bebc965f4d15d9e2b7cea002b6b0f5ba8cde6cc26738467100"
 dependencies = [
 "fastrand",
 "phf_shared",
@@ -755,9 +727,9 @@ dependencies = [

 [[package]]
 name = "phf_shared"
-version = "0.13.1"
+version = "0.14.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e57fef6bc5981e38c2ce2d63bfa546861309f875b8a75f092d1d54ae2d64f266"
+checksum = "c6fd9027e2d9319be6349febd1db4e8d02aa544921200c9b777720ac34a3aa89"
 dependencies = [
 "siphasher",
 ]
@@ -848,9 +820,9 @@ checksum = "f8dcc9c7d52a811697d2151c701e0d08956f92b0e24136cf4cf27b57a6a0d9bf"

 [[package]]
 name = "rand"
-version = "0.10.1"
+version = "0.10.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d2e8e8bcc7961af1fdac401278c6a831614941f6164ee3bf4ce61b7edb162207"
+checksum = "c7f5fa3a058cd35567ef9bfa5e75732bee0f9e4c55fa90477bef2dfcdbc4be80"
 dependencies = [
 "chacha20",
 "getrandom 0.4.2",
@@ -937,7 +909,7 @@ dependencies = [
 "phf_codegen",
 "predicates",
 "regex",
- "sysinfo",
+ "rustix",
 "tempfile",
 "terminal_size",
 "textwrap",
@@ -1013,9 +985,9 @@ checksum = "703d5c7ef118737c72f1af64ad2f6f8c5e1921f818cdcb97b8fe6fc69bf66214"

 [[package]]
 name = "similar"
-version = "3.1.1"
+version = "3.1.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e6505efef05804732ed8a3f2d4f279429eb485bd69d5b0cc6b19cc02005cda16"
+checksum = "85ee016af5d736b69fc89e19254540fa4b5f5492853fb5503920f084011c78b6"
 dependencies = [
 "bstr",
 ]
@@ -1055,20 +1027,6 @@ dependencies = [
 "unicode-ident",
 ]

-[[package]]
-name = "sysinfo"
-version = "0.38.4"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "92ab6a2f8bfe508deb3c6406578252e491d299cbbf3bc0529ecc3313aee4a52f"
-dependencies = [
- "libc",
- "memchr",
- "ntapi",
- "objc2-core-foundation",
- "objc2-io-kit",
- "windows",
-]
-
 [[package]]
 name = "tempfile"
 version = "3.27.0"
@@ -1194,9 +1152,9 @@ checksum = "06abde3611657adf66d383f00b093d7faecc7fa57071cce2578660c9f1010821"

 [[package]]
 name = "uucore"
-version = "0.9.0"
+version = "0.10.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "069b34217c27f611e1589f540f58118dbf226a9e407d38ab472052ff075a1dc2"
+checksum = "e97dd0fd8bb2ed12affba9f956c3b94b8ad3f9432746717475420e916ae0b8d4"
 dependencies = [
 "bigdecimal",
 "clap",
@@ -1212,6 +1170,7 @@ dependencies = [
 "rustix",
 "thiserror",
 "unic-langid",
+ "unicode-width",
 "uucore_procs",
 "wild",
 ]
@@ -1233,9 +1192,9 @@ dependencies = [

 [[package]]
 name = "uucore_procs"
-version = "0.9.0"
+version = "0.10.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "34337da211e7abfff7189b794afb3b5018fe356fb36474b73645458fc1201350"
+checksum = "907f1b828046ea42cdb4bd898f05b48ab873a41937bf75ebe12f09a5bf411f58"
 dependencies = [
 "proc-macro2",
 "quote",
@@ -1243,9 +1202,9 @@ dependencies = [

 [[package]]
 name = "uufuzz"
-version = "0.9.0"
+version = "0.10.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8e4192b67d5e7f0a3300db353ec3217beee89442f02c11d8ae108867eee392d0"
+checksum = "72d4c5bac0f8ccabaa8e8ccbf2724446f8b007c06ad343fabfe1a1525a4ff4f2"
 dependencies = [
 "console",
 "rand",
@@ -1361,49 +1320,6 @@ dependencies = [
 "glob",
 ]

-[[package]]
-name = "winapi"
-version = "0.3.9"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5c839a674fcd7a98952e593242ea400abe93992746761e38641405d28b00f419"
-dependencies = [
- "winapi-i686-pc-windows-gnu",
- "winapi-x86_64-pc-windows-gnu",
-]
-
-[[package]]
-name = "winapi-i686-pc-windows-gnu"
-version = "0.4.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ac3b87c63620426dd9b991e5ce0329eff545bccbbb34f3be09ff6fb6ab51b7b6"
-
-[[package]]
-name = "winapi-x86_64-pc-windows-gnu"
-version = "0.4.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f"
-
-[[package]]
-name = "windows"
-version = "0.62.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "527fadee13e0c05939a6a05d5bd6eec6cd2e3dbd648b9f8e447c6518133d8580"
-dependencies = [
- "windows-collections",
- "windows-core",
- "windows-future",
- "windows-numerics",
-]
-
-[[package]]
-name = "windows-collections"
-version = "0.3.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "23b2d95af1a8a14a3c7367e1ed4fc9c20e0a26e79551b1454d72583c97cc6610"
-dependencies = [
- "windows-core",
-]
-
 [[package]]
 name = "windows-core"
 version = "0.62.2"
@@ -1417,17 +1333,6 @@ dependencies = [
 "windows-strings",
 ]

-[[package]]
-name = "windows-future"
-version = "0.3.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e1d6f90251fe18a279739e78025bd6ddc52a7e22f921070ccdc67dde84c605cb"
-dependencies = [
- "windows-core",
- "windows-link",
- "windows-threading",
-]
-
 [[package]]
 name = "windows-implement"
 version = "0.60.2"
@@ -1456,16 +1361,6 @@ version = "0.2.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "f0805222e57f7521d6a62e36fa9163bc891acd422f971defe97d64e70d0a4fe5"

-[[package]]
-name = "windows-numerics"
-version = "0.3.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6e2e40844ac143cdb44aead537bbf727de9b044e107a0f1220392177d15b0f26"
-dependencies = [
- "windows-core",
- "windows-link",
-]
-
 [[package]]
 name = "windows-result"
 version = "0.4.1"
@@ -1493,15 +1388,6 @@ dependencies = [
 "windows-link",
 ]

-[[package]]
-name = "windows-threading"
-version = "0.2.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3949bd5b99cafdf1c7ca86b43ca564028dfe27d66958f2470940f73d86d75b37"
-dependencies = [
- "windows-link",
-]
-
 [[package]]
 name = "wit-bindgen"
 version = "0.51.0"
@@ -14,8 +14,8 @@ libc = "0.2.153"
 tempfile = "3.15.0"
 rand = { version = "0.10.0" }
 similar = "3.0.0"
-uucore = { version = "0.9.0", features = ["libc"] }
-uufuzz = "0.9.0"
+uucore = { version = "0.10.0", features = ["libc"] }
+uufuzz = "0.10.0"

 sed = { path = ".." }

@@ -52,6 +52,8 @@ pub struct ProcessingContext {
    pub last_file: bool,
    /// Stop processing further input.
    pub stop_processing: bool,
+    /// Whether sed operates on bytes or UTF-8 characters
+    pub character_mode: CharacterMode,
    /// Previously compiled RE, saved for reuse when specifying an empty RE
    pub saved_regex: Option<Regex>,
    /// Modification of input processing action
@@ -59,7 +61,7 @@ pub struct ProcessingContext {
    // command.
    pub input_action: Option<InputAction>,
    /// Hold space
-    pub hold: StringSpace,
+    pub hold: ByteSpace,
    /// Nesting of { } at compile time
    pub parsed_block_nesting: usize,
    /// Command associated with each label
@@ -75,14 +77,22 @@ pub struct ProcessingContext {
 #[derive(Clone, Debug)]
 /// Elements that shall be appended at the end of each command processing cycle
 pub enum AppendElement {
-    Text(Rc<str>), // The specified text string
-    Path(PathBuf), // The contents of the specified file path
+    Text(Rc<[u8]>), // The specified text bytes
+    Path(PathBuf),  // The contents of the specified file path
+}
+
+#[derive(Clone, Copy, Debug, Default, Eq, PartialEq)]
+/// Whether sed operates on bytes or UTF-8 characters
+pub enum CharacterMode {
+    Byte, // Interpret data as arbitrary bytes (C/POSIX locale).
+    #[default]
+    Utf8, // Interpret data as UTF-8 characters (UTF-8 locale).
 }

 #[derive(Clone, Debug, Default, PartialEq)]
-/// A space mirroring IOChunk, but only with a String
-pub struct StringSpace {
-    pub content: String,   // Line content without newline
+/// A space mirroring IOChunk without mmap-backed storage.
+pub struct ByteSpace {
+    pub content: Vec<u8>,  // Line content without newline
    pub has_newline: bool, // True if \n-terminated
 }

@@ -100,9 +110,9 @@ pub enum Address {
 #[derive(Debug)]
 /// A single part of an RE replacement
 pub enum ReplacementPart {
-    Literal(String), // Normal text
-    WholeMatch,      // &
-    Group(u32),      // \1 to \9
+    Literal(Vec<u8>), // Normal text
+    WholeMatch,       // &
+    Group(u32),       // \1 to \9
 }

 // The maximum value allowed in regex quantifier
@@ -152,8 +162,8 @@ impl ReplacementTemplate {
    /// let result = regex.replace_all(input, |caps: &Captures| {
    ///    template.apply_captures(&command, caps) });
    /// Returns an error if a backreference in the template was not matched by the RE.
-    pub fn apply_captures(&self, command: &Command, caps: &Captures) -> UResult<String> {
-        let mut result = String::new();
+    pub fn apply_captures(&self, command: &Command, caps: &Captures) -> UResult<Vec<u8>> {
+        let mut result = Vec::new();

        // Invalid group numbers may end here through (unkown at compile time)
        // reused REs.
@@ -169,15 +179,17 @@ impl ReplacementTemplate {

        for part in &self.parts {
            match part {
-                ReplacementPart::Literal(s) => result.push_str(s),
+                ReplacementPart::Literal(s) => result.extend_from_slice(s),

                ReplacementPart::WholeMatch => {
-                    result.push_str(caps.get(0)?.map(|m| m.as_str()).unwrap_or_default());
+                    result
+                        .extend_from_slice(caps.get(0)?.map(|m| m.as_bytes()).unwrap_or_default());
                }

                ReplacementPart::Group(n) => {
                    let i: usize = (*n).try_into().unwrap();
-                    result.push_str(caps.get(i)?.map(|m| m.as_str()).unwrap_or_default());
+                    result
+                        .extend_from_slice(caps.get(i)?.map(|m| m.as_bytes()).unwrap_or_default());
                }
            }
        }
@@ -186,14 +198,14 @@ impl ReplacementTemplate {
    }

    /// Apply the template to the given RE single match.
-    pub fn apply_match(&self, m: &Match) -> String {
-        let mut result = String::new();
+    pub fn apply_match(&self, m: &Match) -> Vec<u8> {
+        let mut result = Vec::new();

        for part in &self.parts {
            match part {
-                ReplacementPart::Literal(s) => result.push_str(s),
+                ReplacementPart::Literal(s) => result.extend_from_slice(s),

-                ReplacementPart::WholeMatch => result.push_str(m.as_str()),
+                ReplacementPart::WholeMatch => result.extend_from_slice(m.as_bytes()),

                ReplacementPart::Group(_) => {
                    panic!("unexpected Regex group replacement")
@@ -211,10 +223,20 @@ pub struct Substitution {
    pub replacement: ReplacementTemplate,             // Specified broken-down replacement
    pub occurrence: usize,                            // Which occurrence to substitute
    pub print_flag: bool,                             // True if 'p' flag
+    pub p_before_e: bool,                             // True if 'p' appears before 'e'
    pub ignore_case: bool,                            // True if 'I' flag
+    pub execute: bool,                                // True if 'e' flag (GNU extension)
+    pub multiline: bool,                              // True if 'm' or 'M' flag (GNU extension)
    pub write_file: Option<Rc<RefCell<NamedWriter>>>, // Writer to file if 'w' flag is used
 }

+#[derive(Debug, PartialEq, Eq)]
+/// Result of parsing a transliteration string in byte or character mode.
+pub enum ParsedTransliteration {
+    Bytes(Vec<u8>),
+    Text(String),
+}
+
 /// The block of the first and most common Unicode characters:
 /// ASCII, Latin Extended, Greek, Curillic, Coptic, Arabic, etc.
 /// It comprises all UCS-2 characters.  We use a fast lookup array for these.
@@ -223,20 +245,28 @@ const COMMON_UNICODE: usize = 2048;
 #[derive(Debug)]
 /// Transliteration command (y)
 pub struct Transliteration {
-    fast: [char; COMMON_UNICODE],
-    slow: HashMap<char, char>,
+    byte_fast: [u8; 256],
+    unicode_fast: [char; COMMON_UNICODE],
+    unicode_slow: HashMap<char, char>,
+    pub(crate) is_byte_identity: bool,
 }

 impl Default for Transliteration {
    /// Create a new Transliteration with identity mapping for the fast-path.
    fn default() -> Self {
-        let mut fast = ['\0'; COMMON_UNICODE];
-        for (i, slot) in fast.iter_mut().enumerate() {
-            *slot = char::from_u32(i as u32).unwrap_or('\0');
+        let mut fast = [0u8; 256];
+        for (slot, value) in fast.iter_mut().zip(0..=u8::MAX) {
+            *slot = value;
+        }
+        let mut unicode_fast = ['\0'; COMMON_UNICODE];
+        for (slot, cp) in unicode_fast.iter_mut().zip(0u32..) {
+            *slot = char::from_u32(cp).unwrap_or('\0');
        }
        Self {
-            fast,
-            slow: HashMap::new(),
+            byte_fast: fast,
+            unicode_fast,
+            unicode_slow: HashMap::new(),
+            is_byte_identity: true,
        }
    }
 }
@@ -251,25 +281,44 @@ impl Transliteration {
        result
    }

+    /// Create through byte mappings from `source` to `target`.
+    pub fn from_bytes(source: &[u8], target: &[u8]) -> Self {
+        let mut result = Self::default();
+        for (&from, &to) in source.iter().zip(target) {
+            result.byte_fast[from as usize] = to;
+        }
+        result
+    }
+
    /// Set a transliteration mapping from one character to another.
    fn insert(&mut self, from: char, to: char) {
        let cp = from as usize;
        if cp < COMMON_UNICODE {
-            self.fast[cp] = to;
+            self.unicode_fast[cp] = to;
        } else {
-            self.slow.insert(from, to);
+            self.unicode_slow.insert(from, to);
+        }
+        if from.is_ascii() && to.is_ascii() {
+            self.byte_fast[from as usize] = to as u8;
+        } else {
+            self.is_byte_identity = false;
        }
    }

    /// Look up a character transliteration.
-    pub fn lookup(&self, ch: char) -> char {
+    pub fn lookup_char(&self, ch: char) -> char {
        let cp = ch as usize;
        if cp < COMMON_UNICODE {
-            self.fast[cp]
+            self.unicode_fast[cp]
        } else {
-            self.slow.get(&ch).copied().unwrap_or(ch)
+            self.unicode_slow.get(&ch).copied().unwrap_or(ch)
        }
    }
+
+    /// Fast byte lookup for pure ASCII transliterations.
+    pub fn lookup_byte(&self, byte: u8) -> u8 {
+        self.byte_fast[byte as usize]
+    }
 }

 #[derive(Debug)]
@@ -312,16 +361,16 @@ impl Command {

 #[derive(Debug)]
 /// Command-specific data
-/// After parsing, t, b Label elements are converted into BranchTarget ones.
+/// After parsing, t, T, b Label elements are converted into BranchTarget ones.
 pub enum CommandData {
    None,
-    BranchTarget(Option<Rc<RefCell<Command>>>), // Commands for 'b', 't', '{'
-    Label(Option<String>),                      // Label name for 'b', 't', ':'
+    BranchTarget(Option<Rc<RefCell<Command>>>), // Commands for 'b', 't', 'T', '{'
+    Label(Option<String>),                      // Label name for 'b', 't', 'T', ':'
    Path(PathBuf),                              // File path for 'r'
    NamedWriter(Rc<RefCell<NamedWriter>>),      // File output for 'w'
    Number(usize),                              // Number for 'l', 'q', 'Q' (GNU)
    Substitution(Box<Substitution>),            // Substitute command 's'
-    Text(Rc<str>),                              // Text for 'a', 'c', 'i'
+    Text(Rc<[u8]>),                             // Text for 'a', 'c', 'i'
    Transliteration(Box<Transliteration>),      // Transliteration command 'y'
 }

@@ -338,7 +387,7 @@ pub struct InputAction {
    /// Next command to execute (rather than commands from start)
    pub next_command: Option<Rc<RefCell<Command>>>,
    /// Data to prepend to the read contents
-    pub prepend: String,
+    pub prepend: Vec<u8>,
 }

 #[cfg(test)]
@@ -348,7 +397,7 @@ mod tests {

    // Return the captures for the RE applied to the specified string
    fn caps_for<'a>(re: &str, chunk: &'a mut IOChunk) -> Captures<'a> {
-        Regex::new(re)
+        Regex::new(re, CharacterMode::Utf8)
            .unwrap()
            .captures(chunk)
            .unwrap()
@@ -364,26 +413,26 @@ mod tests {
        let cmd = Command::default();

        let result = template.apply_captures(&cmd, &caps).unwrap();
-        assert_eq!(result, "");
+        assert_eq!(result, b"");
    }

    #[test]
    // s/abc/hello/
    fn test_literal_only() {
-        let template = ReplacementTemplate::new(vec![ReplacementPart::Literal("hello".into())]);
+        let template = ReplacementTemplate::new(vec![ReplacementPart::Literal(b"hello".to_vec())]);
        let input = &mut IOChunk::new_from_str("abc");
        let caps = caps_for("abc", input);
        let cmd = Command::default();

        let result = template.apply_captures(&cmd, &caps).unwrap();
-        assert_eq!(result, "hello");
+        assert_eq!(result, b"hello");
    }

    #[test]
    // s/foo\d+/got: &/
    fn test_whole_match() {
        let template = ReplacementTemplate::new(vec![
-            ReplacementPart::Literal("got: ".into()),
+            ReplacementPart::Literal(b"got: ".to_vec()),
            ReplacementPart::WholeMatch,
        ]);
        let input = &mut IOChunk::new_from_str("foo42");
@@ -391,14 +440,26 @@ mod tests {
        let cmd = Command::default();

        let result = template.apply_captures(&cmd, &caps).unwrap();
-        assert_eq!(result, "got: foo42");
+        assert_eq!(result, b"got: foo42");
+    }
+
+    #[test]
+    fn test_apply_match_uses_matched_bytes() {
+        let template = ReplacementTemplate::new(vec![
+            ReplacementPart::Literal(b"<".to_vec()),
+            ReplacementPart::WholeMatch,
+            ReplacementPart::Literal(b">".to_vec()),
+        ]);
+        let m = Match::from_bytes(1, 3, b"\xE9x");
+
+        assert_eq!(template.apply_match(&m), b"<\xE9x>");
    }

    #[test]
    // s/foo(\d+)/number: \1/
    fn test_backreference() {
        let template = ReplacementTemplate::new(vec![
-            ReplacementPart::Literal("number: ".into()),
+            ReplacementPart::Literal(b"number: ".to_vec()),
            ReplacementPart::Group(1),
        ]);
        let input = &mut IOChunk::new_from_str("foo42");
@@ -406,16 +467,16 @@ mod tests {
        let cmd = Command::default();

        let result = template.apply_captures(&cmd, &caps).unwrap();
-        assert_eq!(result, "number: 42");
+        assert_eq!(result, b"number: 42");
    }

    #[test]
    // s/(\w+):(\d+)/key: \1, value: \2/
    fn test_multiple_parts() {
        let template = ReplacementTemplate::new(vec![
-            ReplacementPart::Literal("key: ".into()),
+            ReplacementPart::Literal(b"key: ".to_vec()),
            ReplacementPart::Group(1),
-            ReplacementPart::Literal(", value: ".into()),
+            ReplacementPart::Literal(b", value: ".to_vec()),
            ReplacementPart::Group(2),
        ]);
        let input = &mut IOChunk::new_from_str("x:123");
@@ -423,16 +484,16 @@ mod tests {
        let cmd = Command::default();

        let result = template.apply_captures(&cmd, &caps).unwrap();
-        assert_eq!(result, "key: x, value: 123");
+        assert_eq!(result, b"key: x, value: 123");
    }

    #[test]
    // s/(\w+):(\d+)/key: \1, value: \3/
    fn test_invalid_group() {
        let template = ReplacementTemplate::new(vec![
-            ReplacementPart::Literal("key: ".into()),
+            ReplacementPart::Literal(b"key: ".to_vec()),
            ReplacementPart::Group(1),
-            ReplacementPart::Literal(", value: ".into()),
+            ReplacementPart::Literal(b", value: ".to_vec()),
            ReplacementPart::Group(3),
        ]);
        let input = &mut IOChunk::new_from_str("x:123");
@@ -450,11 +511,11 @@ mod tests {
    #[test]
    fn test_max_group_number_with_groups() {
        let template = ReplacementTemplate::new(vec![
-            ReplacementPart::Literal("a".into()),
+            ReplacementPart::Literal(b"a".to_vec()),
            ReplacementPart::Group(2),
            ReplacementPart::WholeMatch,
            ReplacementPart::Group(5),
-            ReplacementPart::Literal("z".into()),
+            ReplacementPart::Literal(b"z".to_vec()),
        ]);
        assert_eq!(template.max_group_number, 5);
    }
@@ -462,9 +523,9 @@ mod tests {
    #[test]
    fn test_max_group_number_without_groups() {
        let template = ReplacementTemplate::new(vec![
-            ReplacementPart::Literal("no".into()),
+            ReplacementPart::Literal(b"no".to_vec()),
            ReplacementPart::WholeMatch,
-            ReplacementPart::Literal("groups".into()),
+            ReplacementPart::Literal(b"groups".to_vec()),
        ]);
        assert_eq!(template.max_group_number, 0);
    }
@@ -474,16 +535,31 @@ mod tests {
    #[test]
    fn test_identity_lookup_fast_path() {
        let t = Transliteration::default();
-        assert_eq!(t.lookup('A'), 'A');
-        assert_eq!(t.lookup('z'), 'z');
-        assert_eq!(t.lookup('\u{07FF}'), '\u{07FF}'); // highest 2-byte UTF-8 char
+        assert_eq!(t.lookup_char('A'), 'A');
+        assert_eq!(t.lookup_char('z'), 'z');
+        assert_eq!(t.lookup_char('\u{07FF}'), '\u{07FF}'); // highest 2-byte UTF-8 char
    }

    #[test]
    fn test_identity_lookup_slow_path() {
        let t = Transliteration::default();
-        assert_eq!(t.lookup('\u{0800}'), '\u{0800}'); // just outside fast path
-        assert_eq!(t.lookup('\u{1F600}'), '\u{1F600}'); // 😀
+        assert_eq!(t.lookup_char('\u{0800}'), '\u{0800}'); // just outside fast path
+        assert_eq!(t.lookup_char('\u{1F600}'), '\u{1F600}'); // 😀
+    }
+
+    #[test]
+    fn test_from_bytes_and_lookup_byte() {
+        let t = Transliteration::from_bytes(b"a\xE9", b"Z!");
+        assert_eq!(t.lookup_byte(b'a'), b'Z');
+        assert_eq!(t.lookup_byte(0xE9), b'!');
+        assert_eq!(t.lookup_byte(b'b'), b'b');
+    }
+
+    #[test]
+    fn test_is_byte_identity_tracks_non_ascii_character_mappings() {
+        assert!(Transliteration::from_strings("ab", "xy").is_byte_identity);
+        assert!(!Transliteration::from_strings("aé", "xy").is_byte_identity);
+        assert!(!Transliteration::from_strings("ab", "xé").is_byte_identity);
    }

    #[test]
@@ -491,38 +567,38 @@ mod tests {
        let mut t = Transliteration::default();
        t.insert('a', 'α');
        t.insert('b', 'β');
-        assert_eq!(t.lookup('a'), 'α');
-        assert_eq!(t.lookup('b'), 'β');
-        assert_eq!(t.lookup('c'), 'c'); // unchanged
+        assert_eq!(t.lookup_char('a'), 'α');
+        assert_eq!(t.lookup_char('b'), 'β');
+        assert_eq!(t.lookup_char('c'), 'c'); // unchanged
    }

    #[test]
    fn test_insert_and_lookup_slow_path() {
        let mut t = Transliteration::default();
        t.insert('🦀', 'c'); // U+1F980 Crab emoji -> 'c'
-        assert_eq!(t.lookup('🦀'), 'c');
-        assert_eq!(t.lookup('🦁'), '🦁'); // unchanged
+        assert_eq!(t.lookup_char('🦀'), 'c');
+        assert_eq!(t.lookup_char('🦁'), '🦁'); // unchanged
    }

    #[test]
    fn test_overwrite_mapping() {
        let mut t = Transliteration::default();
        t.insert('x', '1');
-        assert_eq!(t.lookup('x'), '1');
+        assert_eq!(t.lookup_char('x'), '1');
        t.insert('x', '2');
-        assert_eq!(t.lookup('x'), '2');
+        assert_eq!(t.lookup_char('x'), '2');
    }

    #[test]
    fn test_all_fast_path_mapped_to_space() {
        let mut t = Transliteration::default();
        for cp in 0..COMMON_UNICODE {
-            if let Some(ch) = char::from_u32(cp as u32) {
+            if let Some(ch) = u32::try_from(cp).ok().and_then(char::from_u32) {
                t.insert(ch, ' ');
            }
        }
-        assert_eq!(t.lookup('A'), ' ');
-        assert_eq!(t.lookup('\u{07FF}'), ' ');
+        assert_eq!(t.lookup_char('A'), ' ');
+        assert_eq!(t.lookup_char('\u{07FF}'), ' ');
    }

    // from_strings
@@ -530,11 +606,11 @@ mod tests {
    fn test_basic_transliteration() {
        let t = Transliteration::from_strings("abcδ", "1234");

-        assert_eq!(t.lookup('a'), '1');
-        assert_eq!(t.lookup('b'), '2');
-        assert_eq!(t.lookup('c'), '3');
-        assert_eq!(t.lookup('δ'), '4');
-        assert_eq!(t.lookup('e'), 'e'); // not mapped, fallback
+        assert_eq!(t.lookup_char('a'), '1');
+        assert_eq!(t.lookup_char('b'), '2');
+        assert_eq!(t.lookup_char('c'), '3');
+        assert_eq!(t.lookup_char('δ'), '4');
+        assert_eq!(t.lookup_char('e'), 'e'); // not mapped, fallback
    }

    #[test]
@@ -543,16 +619,16 @@ mod tests {
        let target = "e文c";
        let t = Transliteration::from_strings(source, target);

-        assert_eq!(t.lookup('é'), 'e');
-        assert_eq!(t.lookup('漢'), '文');
-        assert_eq!(t.lookup('🦀'), 'c');
-        assert_eq!(t.lookup('x'), 'x'); // fast fallback
-        assert_eq!(t.lookup('文'), '文'); // slow fallback
+        assert_eq!(t.lookup_char('é'), 'e');
+        assert_eq!(t.lookup_char('漢'), '文');
+        assert_eq!(t.lookup_char('🦀'), 'c');
+        assert_eq!(t.lookup_char('x'), 'x'); // fast fallback
+        assert_eq!(t.lookup_char('文'), '文'); // slow fallback
    }

    #[test]
    fn test_overwrite_fast_path() {
        let t = Transliteration::from_strings("aa", "12");
-        assert_eq!(t.lookup('a'), '2'); // last mapping wins
+        assert_eq!(t.lookup_char('a'), '2'); // last mapping wins
    }
 }
@@ -20,18 +20,20 @@ pub mod processor;
 pub mod script_char_provider;
 pub mod script_line_provider;

-use crate::sed::command::{ProcessingContext, StringSpace};
+use crate::sed::command::{ByteSpace, CharacterMode, ProcessingContext};
 use crate::sed::compiler::compile;
 use crate::sed::processor::process_all_files;
 use crate::sed::script_line_provider::ScriptValue;
-use clap::{Arg, ArgMatches, Command, arg, crate_version};
+use clap::{Arg, ArgMatches, Command, arg};
 use std::collections::HashMap;
+use std::env;
 use std::path::PathBuf;
-use uucore::error::{UResult, UUsageError};
+use uucore::error::{UResult, USimpleError, UUsageError};
 use uucore::format_usage;

-const ABOUT: &str = "Stream editor for filtering and transforming text";
+const ABOUT: &str = "Stream editor for filtering and transforming text (part of uutils)";
 const USAGE: &str = "sed [OPTION]... [script] [file]...";
+const VERSION: &str = concat!(env!("CARGO_PKG_VERSION"), " (uutils)");

 #[uucore::main]
 pub fn uumain(args: impl uucore::Args) -> UResult<()> {
@@ -46,7 +48,7 @@ pub fn uumain(args: impl uucore::Args) -> UResult<()> {
    }

    let (scripts, files) = get_scripts_files(&matches)?;
-    let mut context = build_context(&matches);
+    let mut context = build_context(&matches)?;

    let executable = compile(scripts, &mut context)?;
    process_all_files(executable, files, &mut context)?;
@@ -61,7 +63,7 @@ pub fn uu_app() -> Command {
    let util_name = uucore::util_name();

    Command::new(util_name)
-        .version(crate_version!())
+        .version(VERSION)
        .about(ABOUT)
        .override_usage(format_usage(USAGE))
        .args_override_self(true)
@@ -192,9 +194,40 @@ fn get_scripts_files(matches: &ArgMatches) -> UResult<(Vec<ScriptValue>, Vec<Pat
    Ok((scripts, files))
 }

-// Parse CLI flag arguments and return a ProcessingContext struct based on them
-fn build_context(matches: &ArgMatches) -> ProcessingContext {
-    ProcessingContext {
+/// Return the character interpretation mode implied by the process locale.
+///
+/// The mode is determined from the first non-empty of LC_ALL, LC_CTYPE, and
+/// LANG. The C and POSIX locales select byte mode. UTF-8 locales (including
+/// C.UTF-8) select UTF-8 mode. All other locales result in an error.
+pub fn character_mode_for_locale(locale: &str) -> UResult<CharacterMode> {
+    if locale == "C" || locale == "POSIX" {
+        Ok(CharacterMode::Byte)
+    } else if locale.eq_ignore_ascii_case("C.UTF-8")
+        || locale.to_ascii_lowercase().ends_with(".utf-8")
+        || locale.to_ascii_lowercase().ends_with(".utf8")
+    {
+        Ok(CharacterMode::Utf8)
+    } else {
+        Err(USimpleError::new(
+            1,
+            format!("unsupported locale: {locale}"),
+        ))
+    }
+}
+
+/// Return a ProcessingContext based on parsed CLI flags and environment.
+fn build_context(matches: &ArgMatches) -> UResult<ProcessingContext> {
+    let locale = ["LC_ALL", "LC_CTYPE", "LANG"]
+        .into_iter()
+        .find_map(|name| {
+            let value = env::var(name).ok()?;
+            (!value.is_empty()).then_some(value)
+        })
+        // Same default as GNU sed
+        .unwrap_or_else(|| "C".to_string());
+
+    Ok(ProcessingContext {
+        // CLI arguments
        all_output_files: matches.get_flag("all-output-files"),
        debug: matches.get_flag("debug"),
        regex_extended: matches.get_flag("regexp-extended"),
@@ -211,6 +244,9 @@ fn build_context(matches: &ArgMatches) -> ProcessingContext {
        unbuffered: matches.get_flag("unbuffered"),
        null_data: matches.get_flag("null-data"),

+        // Environment
+        character_mode: character_mode_for_locale(&locale)?,
+
        // Other context
        input_name: "<stdin>".to_string(),
        line_number: 0,
@@ -220,8 +256,8 @@ fn build_context(matches: &ArgMatches) -> ProcessingContext {
        stop_processing: false,
        saved_regex: None,
        input_action: None,
-        hold: StringSpace {
-            content: String::new(),
+        hold: ByteSpace {
+            content: Vec::new(),
            has_newline: true,
        },
        parsed_block_nesting: 0,
@@ -229,7 +265,7 @@ fn build_context(matches: &ArgMatches) -> ProcessingContext {
        range_commands: Vec::new(),
        substitution_made: false,
        append_elements: Vec::new(),
-    }
+    })
 }

 #[cfg(test)]
@@ -338,7 +374,7 @@ mod tests {
    #[test]
    fn test_defaults() {
        let matches = test_matches(&[]);
-        let ctx = build_context(&matches);
+        let ctx = build_context(&matches).unwrap();

        assert!(!ctx.all_output_files);
        assert!(!ctx.debug);
@@ -373,7 +409,7 @@ mod tests {
            "-z",
        ]);

-        let ctx = build_context(&matches);
+        let ctx = build_context(&matches).unwrap();

        assert!(ctx.all_output_files);
        assert!(ctx.debug);
@@ -393,7 +429,7 @@ mod tests {
    #[test]
    fn test_multiple_same_arguments() {
        let matches = test_matches(&["-E", "-r"]);
-        let ctx = build_context(&matches);
+        let ctx = build_context(&matches).unwrap();

        assert!(ctx.regex_extended);
    }
@@ -401,7 +437,7 @@ mod tests {
    #[test]
    fn test_in_place_with_suffix() {
        let matches = test_matches(&["-i", ".bak"]);
-        let ctx = build_context(&matches);
+        let ctx = build_context(&matches).unwrap();

        assert!(ctx.in_place);
        assert_eq!(ctx.in_place_suffix, Some(".bak".to_string()));
@@ -412,10 +448,59 @@ mod tests {
        let matches_default = test_matches(&[]);
        let matches_custom = test_matches(&["-l", "120"]);

-        let ctx_default = build_context(&matches_default);
-        let ctx_custom = build_context(&matches_custom);
+        let ctx_default = build_context(&matches_default).unwrap();
+        let ctx_custom = build_context(&matches_custom).unwrap();

        assert_eq!(ctx_default.length, 70);
        assert_eq!(ctx_custom.length, 120);
    }
+
+    #[test]
+    fn c_locale_selects_byte_mode() {
+        assert_eq!(character_mode_for_locale("C").unwrap(), CharacterMode::Byte);
+    }
+
+    #[test]
+    fn posix_locale_selects_byte_mode() {
+        assert_eq!(
+            character_mode_for_locale("POSIX").unwrap(),
+            CharacterMode::Byte
+        );
+    }
+
+    #[test]
+    fn c_utf8_locale_selects_utf8_mode() {
+        assert_eq!(
+            character_mode_for_locale("C.UTF-8").unwrap(),
+            CharacterMode::Utf8
+        );
+    }
+
+    #[test]
+    fn dot_utf8_locale_selects_utf8_mode() {
+        assert_eq!(
+            character_mode_for_locale("en_US.UTF-8").unwrap(),
+            CharacterMode::Utf8
+        );
+    }
+
+    #[test]
+    fn dot_utf8_locale_is_case_insensitive() {
+        assert_eq!(
+            character_mode_for_locale("el_GR.utf8").unwrap(),
+            CharacterMode::Utf8
+        );
+    }
+
+    #[test]
+    fn unsupported_locale_reports_locale_name() {
+        let err = character_mode_for_locale("el_GR.ISO-8859-7").unwrap_err();
+
+        assert_eq!(err.code(), 1);
+        assert!(
+            err.to_string()
+                .contains("unsupported locale: el_GR.ISO-8859-7"),
+            "{err}"
+        );
+    }
 }
@@ -57,13 +57,21 @@ impl NamedWriter {

    /// Write a line to the file with a newline, returning descriptive errors.
    pub fn write_line(&mut self, line: &str) -> UResult<()> {
-        writeln!(self.writer, "{line}").map_err(|e| {
-            runtime_error::<()>(
-                &self.location,
-                format!("writing to file {}: {e}", self.path.quote()),
-            )
-            .unwrap_err()
-        })
+        self.write_line_bytes(line.as_bytes())
+    }
+
+    /// Write bytes to the file with a newline, returning descriptive errors.
+    pub fn write_line_bytes(&mut self, line: &[u8]) -> UResult<()> {
+        self.writer
+            .write_all(line)
+            .and_then(|()| self.writer.write_all(b"\n"))
+            .map_err(|e| {
+                runtime_error::<()>(
+                    &self.location,
+                    format!("writing to file {}: {e}", self.path.quote()),
+                )
+                .unwrap_err()
+            })
    }

    /// Flush the writer, returning a descriptive error.
@@ -88,3 +96,22 @@ pub fn flush_all() -> UResult<()> {
        Ok(())
    })
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use std::fs;
+    use tempfile::NamedTempFile;
+
+    #[test]
+    fn test_write_line_bytes_appends_newline() {
+        let file = NamedTempFile::new().unwrap();
+        let path = file.path().to_path_buf();
+        let writer = NamedWriter::new(path.clone(), ScriptLocation::default()).unwrap();
+
+        writer.borrow_mut().write_line_bytes(b"a\xE9").unwrap();
+        writer.borrow_mut().flush().unwrap();
+
+        assert_eq!(fs::read(path).unwrap(), b"a\xE9\n");
+    }
+}
@@ -1,4 +1,4 @@
-// Provide the script contents character by character
+// Provide the script contents byte by byte
 //
 // SPDX-License-Identifier: MIT
 // Copyright (c) 2025 Diomidis Spinellis
@@ -10,19 +10,20 @@

 #[derive(Debug)]
 pub struct ScriptCharProvider {
-    line: Vec<char>,
+    line: Vec<u8>,
    pos: usize,
 }

 impl ScriptCharProvider {
-    pub fn new(line_string: &str) -> Self {
+    /// Build from raw script bytes.
+    pub fn new(line: impl AsRef<[u8]>) -> Self {
        Self {
-            line: line_string.chars().collect(),
+            line: line.as_ref().to_vec(),
            pos: 0,
        }
    }

-    /// Advances to the next character, if not at end of line.
+    /// Advances to the next byte, if not at end of line.
    pub fn advance(&mut self) {
        if self.pos < self.line.len() {
            self.pos += 1;
@@ -39,8 +40,13 @@ impl ScriptCharProvider {
        self.pos = pos;
    }

-    /// Returns the current character. Panics if out of bounds.
+    /// Returns the current byte as a character. Panics if out of bounds.
    pub fn current(&self) -> char {
+        char::from(self.line[self.pos])
+    }
+
+    /// Returns the current script byte. Panics if out of bounds.
+    pub fn current_byte(&self) -> u8 {
        self.line[self.pos]
    }

@@ -51,8 +57,8 @@ impl ScriptCharProvider {

    /// Advances the position past any whitespace characters.
    pub fn eat_spaces(&mut self) {
-        while self.pos < self.line.len() && self.line[self.pos].is_whitespace() {
-            self.pos += 1;
+        while self.pos < self.line.len() && self.current().is_whitespace() {
+            self.advance();
        }
    }

@@ -139,4 +145,28 @@ mod tests {
        assert_eq!(chars.get_pos(), 2);
        assert_eq!(chars.current(), 'c');
    }
+
+    #[test]
+    fn test_current_for_utf8_bytes() {
+        let mut chars = ScriptCharProvider::new("αb");
+
+        assert_eq!(chars.current(), char::from(0xCE));
+        assert_eq!(chars.current_byte(), 0xCE);
+        chars.advance();
+        assert_eq!(chars.current(), char::from(0xB1));
+        assert_eq!(chars.current_byte(), 0xB1);
+        chars.advance();
+        assert_eq!(chars.current(), 'b');
+    }
+
+    #[test]
+    fn test_current_for_invalid_utf8_byte() {
+        let mut chars = ScriptCharProvider::new(vec![0xC2, b'a']);
+
+        assert_eq!(chars.current(), char::from(0xC2));
+        assert_eq!(chars.current_byte(), 0xC2);
+        chars.advance();
+        assert_eq!(chars.current(), 'a');
+        assert_eq!(chars.current_byte(), b'a');
+    }
 }
@@ -69,8 +69,8 @@ impl ScriptLineProvider {
    }

    /// Return the next script line to process across all scripts.
-    pub fn next_line(&mut self) -> UResult<Option<String>> {
-        let mut line = String::new();
+    pub fn next_line(&mut self) -> UResult<Option<Vec<u8>>> {
+        let mut line = Vec::new();

        loop {
            let advance = match &mut self.state {
@@ -82,13 +82,13 @@ impl ScriptLineProvider {
                    ..
                } => {
                    line.clear();
-                    let bytes = reader.read_line(&mut line)?;
+                    let bytes = reader.read_until(b'\n', &mut line)?;
                    if bytes == 0 {
                        Some(*index + 1) // finished reading this source
                    } else {
                        *line_number += 1;
                        // Remove trailing newline
-                        if line.ends_with('\n') {
+                        if line.ends_with(b"\n") {
                            line.pop();
                        }
                        return Ok(Some(line));
@@ -114,7 +114,7 @@ impl ScriptLineProvider {

        match &self.sources[next_index] {
            ScriptValue::StringVal(s) => {
-                let cursor = std::io::Cursor::new(s.clone());
+                let cursor = std::io::Cursor::new(s.as_bytes().to_vec());
                self.state = State::Active {
                    index: next_index,
                    reader: Box::new(BufReader::new(cursor)),
@@ -199,7 +199,7 @@ mod tests {

        let mut lines = Vec::new();
        while let Some(line) = provider.next_line().unwrap() {
-            lines.push(line.trim_end().to_string());
+            lines.push(String::from_utf8(line).unwrap().trim_end().to_string());
        }

        assert_eq!(lines, vec!["line one", "line two", "line three"]);
@@ -216,7 +216,7 @@ mod tests {

        let mut lines = Vec::new();
        while let Some(line) = provider.next_line().unwrap() {
-            lines.push(line.trim_end().to_string());
+            lines.push(String::from_utf8(line).unwrap().trim_end().to_string());
        }

        assert_eq!(lines, vec!["file line 1", "file line 2"]);
@@ -241,7 +241,7 @@ mod tests {

        let mut lines = Vec::new();
        while let Some(line) = provider.next_line().unwrap() {
-            lines.push(line.trim_end().to_string());
+            lines.push(String::from_utf8(line).unwrap().trim_end().to_string());
        }

        assert_eq!(
@@ -266,7 +266,7 @@ mod tests {
        let mut provider = ScriptLineProvider::new(input);

        if let Some(line) = provider.next_line().unwrap() {
-            assert_eq!(line.trim(), "l1");
+            assert_eq!(String::from_utf8(line).unwrap().trim(), "l1");
            assert_eq!(provider.get_line_number(), 1);
            assert_eq!(provider.get_input_name(), "<script argument 1>");
        } else {
@@ -274,7 +274,7 @@ mod tests {
        }

        if let Some(line) = provider.next_line().unwrap() {
-            assert_eq!(line.trim(), "l2");
+            assert_eq!(String::from_utf8(line).unwrap().trim(), "l2");
            assert_eq!(provider.get_line_number(), 2);
            assert_eq!(provider.get_input_name(), "<script argument 1>");
        } else {
@@ -282,11 +282,22 @@ mod tests {
        }

        if let Some(line) = provider.next_line().unwrap() {
-            assert_eq!(line.trim(), "l3");
+            assert_eq!(String::from_utf8(line).unwrap().trim(), "l3");
            assert_eq!(provider.get_line_number(), 1);
            assert_eq!(provider.get_input_name(), "<script argument 2>");
        } else {
            panic!("Expected a line");
        }
    }
+
+    #[test]
+    fn test_file_source_preserves_invalid_utf8_bytes() {
+        let mut temp_file = NamedTempFile::new().unwrap();
+        temp_file.write_all(b"s/\xC2\xE7/X/\n").unwrap();
+
+        let input = vec![ScriptValue::PathVal(temp_file.path().to_path_buf())];
+        let mut provider = ScriptLineProvider::new(input);
+
+        assert_eq!(provider.next_line().unwrap().unwrap(), b"s/\xC2\xE7/X/");
+    }
 }
@@ -0,0 +1,14 @@
+CLEARED ^l1_1
+CLEARED ^l1_2
+CLEARED ^l1_3
+CLEARED ^l1_4
+CLEARED ^l1_5
+CLEARED ^l1_6
+CLEARED ^l1_7
+CLEARED ^l1_8
+CLEARED ^l1_9
+CLEARED ^l1_10
+CLEARED ^l1_11
+CLEARED ^l1_12
+CLEARED ^l1_13
+CLEARED ^l1_14
@@ -0,0 +1,14 @@
+l1_1
+SUB l1_X
+l1_3
+l1_4
+l1_5
+l1_6
+l1_7
+l1_8
+l1_9
+l1_10
+l1_11
+SUB l1_1X
+l1_13
+l1_14
Author	SHA1	Message	Date
Sylvestre Ledru	2ce633cb8d	prepare version 0.2.0 (#519 )	2026-08-09 01:57:25 +02:00
Mukunda Rao Katta	91465e566b	fix: identify uutils in cli output (#435 )	2026-08-09 00:55:37 +02:00
Sylvestre Ledru	d842aa305e	sed: fix clippy cast_possible_truncation errors (#518 )	2026-08-09 00:52:26 +02:00
xtqqczze	d28a1f8e73	chore(cargo): remove deprecated package authors field (#444 )	2026-08-08 23:28:22 +02:00
renovate[bot]	af5f8afdcb	Update Rust crate clap_complete to v4.6.9	2026-08-07 08:07:56 +02:00
renovate[bot]	8f3f126816	Update Rust crate clap_mangen to v0.3.2	2026-08-07 08:05:08 +02:00
renovate[bot]	c771318a7a	Update Rust crate clap to v4.6.6	2026-08-07 08:03:52 +02:00
renovate[bot]	36a1cf4954	Update Rust crate uucore to 0.10.0	2026-08-06 09:02:46 +02:00
renovate[bot]	13bb82ba13	Update Rust crate clap_mangen to v0.3.1	2026-08-06 08:32:10 +02:00
renovate[bot]	cb9eb1da42	Update Rust crate uufuzz to 0.10.0	2026-08-06 08:15:55 +02:00
renovate[bot]	b101462d1b	Update Rust crate uutests to 0.10.0	2026-08-06 08:10:31 +02:00
oech3	9f2e1a1ad4	remove sysinfo dep	2026-08-05 14:18:30 +02:00
Franklin-Qi	2b092d53b0	fast_io.rs: replace the unsafe libc::copy_file_range call by rustix (#458 ) Replace the unsafe `libc::copy_file_range` call with the safe `rustix::fs::copy_file_range`. Closes: #442	2026-08-05 13:32:45 +02:00
renovate[bot]	550d545d72	Update Rust crate similar to v3.1.2	2026-08-05 08:01:27 +02:00
Diomidis Spinellisandmukunda katta	a5d219764f	feat: add T command (branch on no-substitution) (#508 ) * feat: add T command (branch on no-substitution) Implements the GNU `T LABEL` command: branch to LABEL if no s/// substitution has succeeded since the last input line was read or the last t/T was taken. It is the inverse of `t` and shares the same address and label parsing. Wires 'T' into get_cmd_spec, resolve_branch_targets, and the processor branch logic, resetting the substitution flag on evaluation just like 't'. Adds integration tests (simple branch + flag-clear) and a cmd-spec unit test. Closes #395 Co-authored-by: Diomidis Spinellis <dds@aueb.gr> * ci: re-trigger checks (windows runner crates.io download flake) * Disable T in POSIX mode --------- Co-authored-by: mukunda katta <mukunda.vjcs6@gmail.com>	2026-08-02 17:47:18 +03:00
Develandpre-commit-ci[bot]	397ab86302	Implement `v` command #397 (#420 ) * implement version command * add tests for version option * fix patch comparsion * absurd extreme comparsion fix * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * accept missing minor or patch * guard 'v' with if posix * add a few tests * add test_default_version * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * check if minor is empty and push zero instead of shadowing --------- Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>	2026-08-01 18:18:05 +03:00
Mukunda Rao Katta	2288789f33	fix: allow closing brace as command terminator (#427 )	2026-08-01 14:03:34 +03:00
Pranav Ram Joshi	84ae1b07a7	Implement `e` (execute) command (#481 ) * chore: fix typo * add: standalone e command Implement the 'e' command. No argument implies execution of pattern space and any respective argument is to run with the output written to stream. Rejected under --posix and --sandbox. Also stop silently swallowing shell stderr for e and s///e. Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com> * add: e command tests Includes 17 new tests. Cover both forms of the new 'e' command: with-argument execution (including shell side effects, addressing, ';' not separating the argument, escape decoding, and backslash continuation) and the no-argument pattern-space form. Also cover --posix/--sandbox rejection and stderr surfacing on command failure. Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com> * chore: gate LF-related tests to UNIIX only * address e command review Validate shell output as utf-8 and error out instead of lossily converting it. Make CRLF strip conditionally compile to Windows only. Rename execute_pattern_as_shell to execute_pattern_as_shell_command. Add tests for non-utf-8 error paths. Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com> * add: Windows tests for e command with argument Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com> * fix: broken pipe race in sandbox e command test Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com> * Apply s/// p and e flags in written order GNU sed applies the 'p' and 'e' substitution flags in the order they appear. 'pe' prints the pre-execution text then executes, while 'ep' executes then prints the result. Record whether 'p' precedes 'e' and apply them accordingly. Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com> --------- Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com>	2026-08-01 13:29:42 +03:00
renovate[bot]	f32296fc22	Update Rust crate clap to v4.6.5	2026-08-01 07:12:40 +02:00
renovate[bot]	7239fb0e08	Update Rust crate ctor to v1.0.12	2026-07-30 07:57:26 +02:00
renovate[bot]	c7e1f78486	Update mozilla-actions/sccache-action action to v0.0.11	2026-07-29 15:19:54 +02:00
renovate[bot]	fb907703a0	Update CodSpeedHQ/action action to v5	2026-07-29 14:01:03 +02:00
renovate[bot]	3d52077d7f	Update Rust crate clap_complete to v4.6.8	2026-07-27 17:37:39 +02:00
Diomidis Spinellis	e9792a8ef9	Improve GNU compatibility through byte processing (#487 ) The original Rust implementation assummed a UTF-8 environment, as is common in modern systems. This commit improves GNU sed compatibility by changing most processing to run on bytes rather than Rust's Unicode characters and strings. This change also improves the handling of binary files. The commit adds a simple POSIX-compatible locale configuration to support UTF-8 processing when specified. While at it, also fix script-specified file handling to take place through OsString and add a few missing unit tests. The change fixes two failing GNU test suite cases.	2026-07-26 19:44:54 +02:00
renovate[bot]	c732b60cfc	Update Rust crate ctor to v1.0.11 (#501 ) Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>	2026-07-26 19:43:52 +02:00
renovate[bot]	c1e5d8ea08	Update Rust crate ctor to v1.0.10	2026-07-22 08:23:11 +02:00
renovate[bot]	6887201cce	Update Rust crate clap to v4.6.4	2026-07-22 08:09:28 +02:00
renovate[bot]	75d856e4c2	Update Rust crate libc to v0.2.189	2026-07-22 08:00:30 +02:00
renovate[bot]	d7a2106258	Update Rust crate libc to v0.2.188	2026-07-21 14:00:30 +02:00
renovate[bot]	a726ac6959	Update Rust crate clap to v4.6.3	2026-07-21 08:13:36 +02:00
renovate[bot]	49590e63be	Update Rust crate libc to v0.2.187	2026-07-21 08:03:39 +02:00
renovate[bot]	99a9957e89	Update Rust crate ctor to v1.0.9	2026-07-16 08:48:11 +02:00
renovate[bot]	c521271ca8	Update Rust crate regex to v1.13.1	2026-07-16 08:19:19 +02:00
renovate[bot]	7830c20bc1	Update Rust crate clap to v4.6.2	2026-07-16 08:08:20 +02:00
renovate[bot]	b37e23fa98	Update Rust crate regex to v1.13.0	2026-07-10 10:55:46 +02:00
Daniel Hofstetter	07eada8531	Bump crossbeam-epoch from 0.9.18 to 0.9.20 (#486 ) https://rustsec.org/advisories/RUSTSEC-2026-0204	2026-07-08 18:57:10 +03:00
renovate[bot]	15def35a2a	chore(deps): update rust crate memchr to v2.8.3	2026-07-08 07:50:55 +02:00
renovate[bot]	30607c4ddc	chore(deps): update rust crate ctor to v1.0.8	2026-07-06 08:07:50 +02:00
Diomidis Spinellis	01c8412327	Optimize multiline matching (#483 ) The multiline matching resulted in a large performance regression due to the indiscriminate use of the dotall (?s) flag. Change the regex selection process to use only the flag when probably needed.	2026-07-05 15:13:53 +02:00
Diomidis SpinellisandMukunda Katta	f3a79c5847	Support multiline substitution flag (#482 ) While at it fix . matching of embedded newlines. This commit increases GNU passed tests by 2. Co-authored-by: Mukunda Katta <mukunda.vjcs6@gmail.com>	2026-07-05 10:48:39 +02:00
Diomidis Spinellis	a9e5fe778c	Add sandbox checks (#477 ) Following the introduction of the first sandbox check in `b14272c1` complete the checks for the remaining commands.	2026-07-03 16:55:50 +02:00
renovate[bot]	02d0239447	chore(deps): update rust crate rand to v0.10.2	2026-07-02 16:50:42 +02:00
renovate[bot]	8b5dea703a	chore(deps): update rust crate console to v0.16.4	2026-07-02 08:05:54 +02:00
renovate[bot]	34dfbc2ef1	chore(deps): update rust crate clap_complete to v4.6.7	2026-07-02 08:05:18 +02:00
Develandpre-commit-ci[bot]	8cb7090bbc	--posix should reject GNU substitute flags i/I #401 (#419 ) * reject 'i', 'I', 'm', 'M' and 'E' when --posix flag is set * add test for issue 401 * formatting * add a test for 'm' flag in test_posix_reject_flags * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Update compiler.rs * cargo fmt * add a const variable with the err msg * Update compiler.rs * cargo fmt --------- Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>	2026-07-01 19:30:25 +03:00
Mukunda Rao Katta	2a5ab5a9a2	feat: add z command (#428 ) * feat: add z command * test: avoid stdin race in z POSIX rejection	2026-07-01 12:51:11 +03:00
Pranav Ram Joshi	b14272c1d7	Add e substitution flag * feat: introduce e substituion for sed When the s command is requested for sed, the e substitution flag enables execution of the result of substitution as a shell command. It matches the GNU sed behavior. On src/sed/command.rs, add 'execute' field to 'Substitution' structure. On src/sed/compiler.rs, handle 'e' in 'compile_subst_flags' and reject it if --posix or --sandbox flag is used in sed. On src/sed/processor.rs, spawn /bin/sh with -c flag for a successful match and replace the pattern space with the command's standard output stream. Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com> * chore: comply with rustfmt Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com> * chore: support for Windows command-line Instead of calling bourne shell, define a new function: shell_command() with two definitions; conditionally compiling to UNIX-like and Windows. On UNIX-like, executes the bourne shell, whereas command prompt is executed on Windows. Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com> * chore: check newline for Windows Unix-like system only uses the newline character whereas Windows uses the carriage return as well as newline. Hence, when working on Windows, we need to truncate the last two characters. Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com> * add: 'e' flag tests Nine new tests have been introduced in regards to new 'e' substitution option. 'test_subst_e_flag_multiline_outuput' is conditionally compiled as printf command is not guaranteed to be present on all Windows systems. Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com> * chore: fallback for shell_command Executing a shell or a command prompt requires appropriate platforms. WASM does not currently support it. Also fix a linting issue. Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com> * chore: linting cleanup Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com> --------- Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com>	2026-07-01 11:18:02 +03:00
Diomidis Spinellis	cd3b10db48	Document uutils Playground	2026-07-01 10:48:54 +03:00
Diomidis Spinellis	48c2f56358	Document r command extension	2026-07-01 10:42:22 +03:00
Diomidis SpinellisandPranavRJoshi	dcd2df848a	Implement zero-address r command to insert file before first line (#474 ) * feat: support zero-address file insertion Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com> * Polish the handling of 0r - Use common constant for error message. - Simplify command check at compile and runtime. - Separate runtime handling into a function. --------- Signed-off-by: PranavRJoshi <pranavrjoshi1@gmail.com> Co-authored-by: PranavRJoshi <pranavrjoshi1@gmail.com>	2026-07-01 08:36:03 +02:00
renovate[bot]	9efab796be	chore(deps): update rust crate clap_complete to v4.6.6	2026-07-01 07:05:51 +02:00