l10n: convert the md files to fluent

Cargo.toml

@@ -364,6 +364,7 @@ digest = "0.10.7"
 fluent-bundle = "0.16.0"
 fluent = "0.17.0"
 unic-langid = "0.9.6"
+fluent-syntax = "0.12.0"
 
 uucore = { version = "0.1.0", package = "uucore", path = "src/uucore" }
 uucore_procs = { version = "0.1.0", package = "uucore_procs", path = "src/uucore_procs" }

src/bin/coreutils.rs

@@ -3,7 +3,7 @@
 // For the full copyright and license information, please view the LICENSE
 // file that was distributed with this source code.
 
-// spell-checker:ignore manpages mangen
+// spell-checker:ignore manpages mangen prefixcat testcat
 
 use clap::{Arg, Command};
 use clap_complete::Shell;
@@ -14,6 +14,7 @@
 use std::path::{Path, PathBuf};
 use std::process;
 use uucore::display::Quotable;
+use uucore::locale;
 
 const VERSION: &str = env!("CARGO_PKG_VERSION");
 
@@ -50,6 +51,48 @@
     binary_path.file_stem()?.to_str()
 }
 
+fn get_canonical_util_name(util_name: &str) -> &str {
+    match util_name {
+        // uu_test aliases - '[' is an alias for test
+        "[" => "test",
+
+        // hashsum aliases - all these hash commands are aliases for hashsum
+        "md5sum" | "sha1sum" | "sha224sum" | "sha256sum" | "sha384sum" | "sha512sum"
+        | "sha3sum" | "sha3-224sum" | "sha3-256sum" | "sha3-384sum" | "sha3-512sum"
+        | "shake128sum" | "shake256sum" | "b2sum" | "b3sum" => "hashsum",
+
+        "dir" => "ls", // dir is an alias for ls
+
+        // Default case - return the util name as is
+        _ => util_name,
+    }
+}
+
+fn find_prefixed_util<'a>(
+    binary_name: &str,
+    mut util_keys: impl Iterator<Item = &'a str>,
+) -> Option<&'a str> {
+    util_keys.find(|util| {
+        binary_name.ends_with(*util)
+            && binary_name.len() > util.len() // Ensure there's actually a prefix
+            && !binary_name[..binary_name.len() - (*util).len()]
+                .ends_with(char::is_alphanumeric)
+    })
+}
+
+fn setup_localization_or_exit(util_name: &str) {
+    locale::setup_localization(get_canonical_util_name(util_name)).unwrap_or_else(|err| {
+        match err {
+            uucore::locale::LocalizationError::ParseResource {
+                error: err_msg,
+                snippet,
+            } => eprintln!("Localization parse error at {snippet}: {err_msg}"),
+            other => eprintln!("Could not init the localization system: {other}"),
+        }
+        process::exit(99)
+    });
+}
+
 #[allow(clippy::cognitive_complexity)]
 fn main() {
     uucore::panic::mute_sigpipe_panic();
@@ -70,13 +113,10 @@
 
     // binary name equals prefixed util name?
     // * prefix/stem may be any string ending in a non-alphanumeric character
-    let util_name = if let Some(util) = utils.keys().find(|util| {
-        binary_as_util.ends_with(*util)
-            && !binary_as_util[..binary_as_util.len() - (*util).len()]
-                .ends_with(char::is_alphanumeric)
-    }) {
+    // For example, if the binary is named `uu_test`, it will match `test` as a utility.
+    let util_name = if let Some(util) = find_prefixed_util(binary_as_util, utils.keys().copied()) {
         // prefixed util => replace 0th (aka, executable name) argument
-        Some(OsString::from(*util))
+        Some(OsString::from(util))
     } else {
         // unmatched binary name => regard as multi-binary container and advance argument list
         uucore::set_utility_is_second_arg();
@@ -111,6 +151,12 @@
 
         match utils.get(util) {
             Some(&(uumain, _)) => {
+                // TODO: plug the deactivation of the translation
+                // and load the English strings directly at compilation time in the
+                // binary to avoid the load of the flt
+                // Could be something like:
+                // #[cfg(not(feature = "only_english"))]
+                setup_localization_or_exit(util);
                 process::exit(uumain(vec![util_os].into_iter().chain(args)));
             }
             None => {
@@ -213,6 +259,7 @@
     let command = if utility == "coreutils" {
         gen_coreutils_app(util_map)
     } else {
+        setup_localization_or_exit(utility);
         util_map.get(utility).unwrap().1()
     };
 
@@ -239,3 +286,70 @@
     }
     command
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use std::path::Path;
+
+    #[test]
+    fn test_get_canonical_util_name() {
+        // Test a few key aliases
+        assert_eq!(get_canonical_util_name("["), "test");
+        assert_eq!(get_canonical_util_name("md5sum"), "hashsum");
+        assert_eq!(get_canonical_util_name("dir"), "ls");
+
+        // Test passthrough case
+        assert_eq!(get_canonical_util_name("cat"), "cat");
+    }
+
+    #[test]
+    fn test_name() {
+        // Test normal executable name
+        assert_eq!(name(Path::new("/usr/bin/ls")), Some("ls"));
+        assert_eq!(name(Path::new("cat")), Some("cat"));
+        assert_eq!(
+            name(Path::new("./target/debug/coreutils")),
+            Some("coreutils")
+        );
+
+        // Test with extensions
+        assert_eq!(name(Path::new("program.exe")), Some("program"));
+        assert_eq!(name(Path::new("/path/to/utility.bin")), Some("utility"));
+
+        // Test edge cases
+        assert_eq!(name(Path::new("")), None);
+        assert_eq!(name(Path::new("/")), None);
+    }
+
+    #[test]
+    fn test_find_prefixed_util() {
+        let utils = ["test", "cat", "ls", "cp"];
+
+        // Test exact prefixed matches
+        assert_eq!(
+            find_prefixed_util("uu_test", utils.iter().copied()),
+            Some("test")
+        );
+        assert_eq!(
+            find_prefixed_util("my-cat", utils.iter().copied()),
+            Some("cat")
+        );
+        assert_eq!(
+            find_prefixed_util("prefix_ls", utils.iter().copied()),
+            Some("ls")
+        );
+
+        // Test non-alphanumeric separator requirement
+        assert_eq!(find_prefixed_util("prefixcat", utils.iter().copied()), None); // no separator
+        assert_eq!(find_prefixed_util("testcat", utils.iter().copied()), None); // no separator
+
+        // Test no match
+        assert_eq!(find_prefixed_util("unknown", utils.iter().copied()), None);
+        assert_eq!(find_prefixed_util("", utils.iter().copied()), None);
+
+        // Test exact util name (should not match as prefixed)
+        assert_eq!(find_prefixed_util("test", utils.iter().copied()), None);
+        assert_eq!(find_prefixed_util("cat", utils.iter().copied()), None);
+    }
+}

src/uu/arch/src/arch.rs

@@ -7,11 +7,10 @@ use platform_info::*;
 
 use clap::Command;
 use uucore::error::{UResult, USimpleError};
-use uucore::locale::{self, get_message};
+use uucore::locale::get_message;
 
 #[uucore::main]
 pub fn uumain(args: impl uucore::Args) -> UResult<()> {
-    locale::setup_localization(uucore::util_name())?;
     uu_app().try_get_matches_from(args)?;
 
     let uts =

src/uu/base32/locales/en-US.ftl

@@ -0,0 +1,9 @@
+base32-about = encode/decode data and print to standard output
+  With no FILE, or when FILE is -, read standard input.
+
+  The data are encoded as described for the base32 alphabet in RFC 4648.
+  When decoding, the input may contain newlines in addition
+  to the bytes of the formal base32 alphabet. Use --ignore-garbage
+  to attempt to recover from any other non-alphabet bytes in the
+  encoded stream.
+base32-usage = base32 [OPTION]... [FILE]

src/uu/base32/src/base32.rs

@@ -5,24 +5,25 @@
 
 pub mod base_common;
 
-use base_common::ReadSeek;
 use clap::Command;
-use uucore::{encoding::Format, error::UResult, help_about, help_usage};
-
-const ABOUT: &str = help_about!("base32.md");
-const USAGE: &str = help_usage!("base32.md");
+use uucore::{encoding::Format, error::UResult, locale::get_message};
 
 #[uucore::main]
 pub fn uumain(args: impl uucore::Args) -> UResult<()> {
     let format = Format::Base32;
-
-    let config = base_common::parse_base_cmd_args(args, ABOUT, USAGE)?;
-
-    let mut input: Box<dyn ReadSeek> = base_common::get_input(&config)?;
-
+    let (about, usage) = get_info();
+    let config = base_common::parse_base_cmd_args(args, about, usage)?;
+    let mut input = base_common::get_input(&config)?;
     base_common::handle_input(&mut input, format, config)
 }
 
 pub fn uu_app() -> Command {
-    base_common::base_app(ABOUT, USAGE)
+    let (about, usage) = get_info();
+    base_common::base_app(about, usage)
+}
+
+fn get_info() -> (&'static str, &'static str) {
+    let about: &'static str = Box::leak(get_message("base32-about").into_boxed_str());
+    let usage: &'static str = Box::leak(get_message("base32-usage").into_boxed_str());
+    (about, usage)
 }

src/uu/base64/locales/en-US.ftl

@@ -0,0 +1,9 @@
+base64-about = encode/decode data and print to standard output
+  With no FILE, or when FILE is -, read standard input.
+
+  The data are encoded as described for the base64 alphabet in RFC 3548.
+  When decoding, the input may contain newlines in addition
+  to the bytes of the formal base64 alphabet. Use --ignore-garbage
+  to attempt to recover from any other non-alphabet bytes in the
+  encoded stream.
+base64-usage = base64 [OPTION]... [FILE]

src/uu/base64/src/base64.rs

@@ -5,22 +5,24 @@
 
 use clap::Command;
 use uu_base32::base_common;
-use uucore::{encoding::Format, error::UResult, help_about, help_usage};
-
-const ABOUT: &str = help_about!("base64.md");
-const USAGE: &str = help_usage!("base64.md");
+use uucore::{encoding::Format, error::UResult, locale::get_message};
 
 #[uucore::main]
 pub fn uumain(args: impl uucore::Args) -> UResult<()> {
     let format = Format::Base64;
-
-    let config = base_common::parse_base_cmd_args(args, ABOUT, USAGE)?;
-
+    let (about, usage) = get_info();
+    let config = base_common::parse_base_cmd_args(args, about, usage)?;
     let mut input = base_common::get_input(&config)?;
-
     base_common::handle_input(&mut input, format, config)
 }
 
 pub fn uu_app() -> Command {
-    base_common::base_app(ABOUT, USAGE)
+    let (about, usage) = get_info();
+    base_common::base_app(about, usage)
+}
+
+fn get_info() -> (&'static str, &'static str) {
+    let about: &'static str = Box::leak(get_message("base64-about").into_boxed_str());
+    let usage: &'static str = Box::leak(get_message("base64-usage").into_boxed_str());
+    (about, usage)
 }

src/uu/basename/locales/en-US.ftl

@@ -0,0 +1,4 @@
+basename-about = Print NAME with any leading directory components removed
+  If specified, also remove a trailing SUFFIX
+basename-usage = basename [-z] NAME [SUFFIX]
+  basename OPTION... NAME...

src/uu/basename/src/basename.rs

@@ -9,12 +9,10 @@ use clap::{Arg, ArgAction, Command};
 use std::path::{PathBuf, is_separator};
 use uucore::display::Quotable;
 use uucore::error::{UResult, UUsageError};
+use uucore::format_usage;
 use uucore::line_ending::LineEnding;
-use uucore::{format_usage, help_about, help_usage};
 
-static ABOUT: &str = help_about!("basename.md");
-
-const USAGE: &str = help_usage!("basename.md");
+use uucore::locale::get_message;
 
 pub mod options {
     pub static MULTIPLE: &str = "multiple";
@@ -77,8 +75,8 @@ pub fn uumain(args: impl uucore::Args) -> UResult<()> {
 pub fn uu_app() -> Command {
     Command::new(uucore::util_name())
         .version(uucore::crate_version!())
-        .about(ABOUT)
-        .override_usage(format_usage(USAGE))
+        .about(get_message("basename-about"))
+        .override_usage(format_usage(&get_message("basename-usage")))
         .infer_long_args(true)
         .arg(
             Arg::new(options::MULTIPLE)

src/uu/basenc/locales/en-US.ftl

@@ -0,0 +1,7 @@
+basenc-about = Encode/decode data and print to standard output
+  With no FILE, or when FILE is -, read standard input.
+
+  When decoding, the input may contain newlines in addition to the bytes of
+  the formal alphabet. Use --ignore-garbage to attempt to recover
+  from any other non-alphabet bytes in the encoded stream.
+basenc-usage = basenc [OPTION]... [FILE]