add hyperlink support

.github/workflows/ci.yml

@@ -193,6 +193,10 @@ jobs:
       shell: bash
       run: ci/test-complete
 
+    - name: Print hostname detected by grep-cli crate
+      shell: bash
+      run: ${{ env.CARGO }} test --manifest-path crates/cli/Cargo.toml ${{ env.TARGET_FLAGS }} --lib print_hostname -- --nocapture
+
   rustfmt:
     runs-on: ubuntu-latest
     steps:

.gitignore

@@ -7,6 +7,7 @@ target
 /termcolor/Cargo.lock
 /wincolor/Cargo.lock
 /deployment
+/.idea
 
 # Snapcraft files
 stage

complete/_rg

@@ -305,6 +305,8 @@ _rg() {
     '--debug[show debug messages]'
     '--field-context-separator[set string to delimit fields in context lines]'
     '--field-match-separator[set string to delimit fields in matching lines]'
+    '--hostname-bin=[executable for getting system hostname]:hostname executable:_command_names -e'
+    '--hyperlink-format=[specify pattern for hyperlinks]:pattern'
     '--trace[show more verbose debug messages]'
     '--dfa-size-limit=[specify upper size limit of generated DFA]:DFA size (bytes)'
     "(1 stats)--files[show each file that would be searched (but don't search)]"

crates/cli/Cargo.toml

@@ -11,16 +11,16 @@ repository = "/~https://github.com/BurntSushi/ripgrep/tree/master/crates/cli"
 readme = "README.md"
 keywords = ["regex", "grep", "cli", "utility", "util"]
 license = "Unlicense OR MIT"
-edition = "2018"
+edition = "2021"
 
 [dependencies]
-bstr = "1.6.0"
+bstr = { version = "1.6.2", features = ["std"] }
 globset = { version = "0.4.10", path = "../globset" }
-lazy_static = "1.1.0"
-log = "0.4.5"
-regex = "1.1"
-same-file = "1.0.4"
-termcolor = "1.0.4"
+log = "0.4.20"
+termcolor = "1.3.0"
 
 [target.'cfg(windows)'.dependencies.winapi-util]
-version = "0.1.1"
+version = "0.1.6"
+
+[target.'cfg(unix)'.dependencies.libc]
+version = "0.2.148"

crates/cli/src/decompress.rs

@@ -1,8 +1,10 @@
-use std::ffi::{OsStr, OsString};
-use std::fs::File;
-use std::io;
-use std::path::{Path, PathBuf};
-use std::process::Command;
+use std::{
+    ffi::{OsStr, OsString},
+    fs::File,
+    io,
+    path::{Path, PathBuf},
+    process::Command,
+};
 
 use globset::{Glob, GlobSet, GlobSetBuilder};
 
@@ -161,7 +163,7 @@ impl DecompressionMatcher {
     /// Create a new matcher with default rules.
     ///
     /// To add more matching rules, build a matcher with
-    /// [`DecompressionMatcherBuilder`](struct.DecompressionMatcherBuilder.html).
+    /// [`DecompressionMatcherBuilder`].
     pub fn new() -> DecompressionMatcher {
         DecompressionMatcherBuilder::new()
             .build()
@@ -221,9 +223,8 @@ impl DecompressionReaderBuilder {
         path: P,
     ) -> Result<DecompressionReader, CommandError> {
         let path = path.as_ref();
-        let mut cmd = match self.matcher.command(path) {
-            None => return DecompressionReader::new_passthru(path),
-            Some(cmd) => cmd,
+        let Some(mut cmd) = self.matcher.command(path) else {
+            return DecompressionReader::new_passthru(path);
         };
         cmd.arg(path);
 
@@ -302,9 +303,7 @@ impl DecompressionReaderBuilder {
 /// The default matching rules are probably good enough for most cases, and if
 /// they require revision, pull requests are welcome. In cases where they must
 /// be changed or extended, they can be customized through the use of
-/// [`DecompressionMatcherBuilder`](struct.DecompressionMatcherBuilder.html)
-/// and
-/// [`DecompressionReaderBuilder`](struct.DecompressionReaderBuilder.html).
+/// [`DecompressionMatcherBuilder`] and [`DecompressionReaderBuilder`].
 ///
 /// By default, this reader will asynchronously read the processes' stderr.
 /// This prevents subtle deadlocking bugs for noisy processes that write a lot
@@ -320,15 +319,14 @@ impl DecompressionReaderBuilder {
 /// matcher.
 ///
 /// ```no_run
-/// use std::io::Read;
-/// use std::process::Command;
+/// use std::{io::Read, process::Command};
+///
 /// use grep_cli::DecompressionReader;
 ///
-/// # fn example() -> Result<(), Box<::std::error::Error>> {
 /// let mut rdr = DecompressionReader::new("/usr/share/man/man1/ls.1.gz")?;
 /// let mut contents = vec![];
 /// rdr.read_to_end(&mut contents)?;
-/// # Ok(()) }
+/// # Ok::<(), Box<dyn std::error::Error>>(())
 /// ```
 #[derive(Debug)]
 pub struct DecompressionReader {
@@ -347,9 +345,7 @@ impl DecompressionReader {
     ///
     /// This uses the default matching rules for determining how to decompress
     /// the given file. To change those matching rules, use
-    /// [`DecompressionReaderBuilder`](struct.DecompressionReaderBuilder.html)
-    /// and
-    /// [`DecompressionMatcherBuilder`](struct.DecompressionMatcherBuilder.html).
+    /// [`DecompressionReaderBuilder`] and [`DecompressionMatcherBuilder`].
     ///
     /// When creating readers for many paths. it is better to use the builder
     /// since it will amortize the cost of constructing the matcher.
@@ -453,26 +449,20 @@ fn try_resolve_binary<P: AsRef<Path>>(
     use std::env;
 
     fn is_exe(path: &Path) -> bool {
-        let md = match path.metadata() {
-            Err(_) => return false,
-            Ok(md) => md,
-        };
+        let Ok(md) = path.metadata() else { return false };
         !md.is_dir()
     }
 
     let prog = prog.as_ref();
     if prog.is_absolute() {
         return Ok(prog.to_path_buf());
     }
-    let syspaths = match env::var_os("PATH") {
-        Some(syspaths) => syspaths,
-        None => {
-            let msg = "system PATH environment variable not found";
-            return Err(CommandError::io(io::Error::new(
-                io::ErrorKind::Other,
-                msg,
-            )));
-        }
+    let Some(syspaths) = env::var_os("PATH") else {
+        let msg = "system PATH environment variable not found";
+        return Err(CommandError::io(io::Error::new(
+            io::ErrorKind::Other,
+            msg,
+        )));
     };
     for syspath in env::split_paths(&syspaths) {
         if syspath.as_os_str().is_empty() {

crates/cli/src/escape.rs

@@ -1,21 +1,7 @@
 use std::ffi::OsStr;
-use std::str;
 
 use bstr::{ByteSlice, ByteVec};
 
-/// A single state in the state machine used by `unescape`.
-#[derive(Clone, Copy, Eq, PartialEq)]
-enum State {
-    /// The state after seeing a `\`.
-    Escape,
-    /// The state after seeing a `\x`.
-    HexFirst,
-    /// The state after seeing a `\x[0-9A-Fa-f]`.
-    HexSecond(char),
-    /// Default state.
-    Literal,
-}
-
 /// Escapes arbitrary bytes into a human readable string.
 ///
 /// This converts `\t`, `\r` and `\n` into their escaped forms. It also
@@ -38,17 +24,7 @@ enum State {
 /// assert_eq!(r"foo\nbar\xFFbaz", escape(b"foo\nbar\xFFbaz"));
 /// ```
 pub fn escape(bytes: &[u8]) -> String {
-    let mut escaped = String::new();
-    for (s, e, ch) in bytes.char_indices() {
-        if ch == '\u{FFFD}' {
-            for b in bytes[s..e].bytes() {
-                escape_byte(b, &mut escaped);
-            }
-        } else {
-            escape_char(ch, &mut escaped);
-        }
-    }
-    escaped
+    bytes.escape_bytes().to_string()
 }
 
 /// Escapes an OS string into a human readable string.
@@ -89,76 +65,7 @@ pub fn escape_os(string: &OsStr) -> String {
 /// assert_eq!(&b"foo\nbar\xFFbaz"[..], &*unescape(r"foo\nbar\xFFbaz"));
 /// ```
 pub fn unescape(s: &str) -> Vec<u8> {
-    use self::State::*;
-
-    let mut bytes = vec![];
-    let mut state = Literal;
-    for c in s.chars() {
-        match state {
-            Escape => match c {
-                '\\' => {
-                    bytes.push(b'\\');
-                    state = Literal;
-                }
-                'n' => {
-                    bytes.push(b'\n');
-                    state = Literal;
-                }
-                'r' => {
-                    bytes.push(b'\r');
-                    state = Literal;
-                }
-                't' => {
-                    bytes.push(b'\t');
-                    state = Literal;
-                }
-                'x' => {
-                    state = HexFirst;
-                }
-                c => {
-                    bytes.extend(format!(r"\{}", c).into_bytes());
-                    state = Literal;
-                }
-            },
-            HexFirst => match c {
-                '0'..='9' | 'A'..='F' | 'a'..='f' => {
-                    state = HexSecond(c);
-                }
-                c => {
-                    bytes.extend(format!(r"\x{}", c).into_bytes());
-                    state = Literal;
-                }
-            },
-            HexSecond(first) => match c {
-                '0'..='9' | 'A'..='F' | 'a'..='f' => {
-                    let ordinal = format!("{}{}", first, c);
-                    let byte = u8::from_str_radix(&ordinal, 16).unwrap();
-                    bytes.push(byte);
-                    state = Literal;
-                }
-                c => {
-                    let original = format!(r"\x{}{}", first, c);
-                    bytes.extend(original.into_bytes());
-                    state = Literal;
-                }
-            },
-            Literal => match c {
-                '\\' => {
-                    state = Escape;
-                }
-                c => {
-                    bytes.extend(c.to_string().as_bytes());
-                }
-            },
-        }
-    }
-    match state {
-        Escape => bytes.push(b'\\'),
-        HexFirst => bytes.extend(b"\\x"),
-        HexSecond(c) => bytes.extend(format!("\\x{}", c).into_bytes()),
-        Literal => {}
-    }
-    bytes
+    Vec::unescape_bytes(s)
 }
 
 /// Unescapes an OS string.
@@ -171,27 +78,6 @@ pub fn unescape_os(string: &OsStr) -> Vec<u8> {
     unescape(&string.to_string_lossy())
 }
 
-/// Adds the given codepoint to the given string, escaping it if necessary.
-fn escape_char(cp: char, into: &mut String) {
-    if cp.is_ascii() {
-        escape_byte(cp as u8, into);
-    } else {
-        into.push(cp);
-    }
-}
-
-/// Adds the given byte to the given string, escaping it if necessary.
-fn escape_byte(byte: u8, into: &mut String) {
-    match byte {
-        0x21..=0x5B | 0x5D..=0x7D => into.push(byte as char),
-        b'\n' => into.push_str(r"\n"),
-        b'\r' => into.push_str(r"\r"),
-        b'\t' => into.push_str(r"\t"),
-        b'\\' => into.push_str(r"\\"),
-        _ => into.push_str(&format!(r"\x{:02X}", byte)),
-    }
-}
-
 #[cfg(test)]
 mod tests {
     use super::{escape, unescape};
@@ -215,7 +101,8 @@ mod tests {
     #[test]
     fn nul() {
         assert_eq!(b(b"\x00"), unescape(r"\x00"));
-        assert_eq!(r"\x00", escape(b"\x00"));
+        assert_eq!(b(b"\x00"), unescape(r"\0"));
+        assert_eq!(r"\0", escape(b"\x00"));
     }
 
     #[test]