Fix: coreutils now allows options with incomplete names

src/bin/coreutils.rs

@@ -38,107 +38,205 @@ fn usage<T>(utils: &UtilityMap<T>, name: &str) {
     );
 }
 
+/// all defined coreutils options
+const COREUTILS_OPTIONS: [&'static str; 5] = ["--list", "-V", "--version", "-h", "--help"];
+
+/// Entry into Coreutils
+///
+/// # Arguments
+/// * first arg needs to be the binary/executable. \
+///   This is usually coreutils, but can be the util name itself, e.g. 'ls'. \
+///   The util name will be checked against the list of enabled utils, where
+///   * the name exactly matches the name of an applet/util or
+///   * the name matches <PREFIX><UTIL_NAME> pattern, e.g.
+///   'my_own_directory_service_ls' as long as the last letters match the utility.
+/// * coreutils arg: --list, --version, -V, --help, -h (or shortened long versions): \
+///   Output information about coreutils itself. \
+///   Multiple of these arguments, output limited to one, with help > version > list.
+/// * util name and any number of arguments: \
+///   Will get passed on to the selected utility. \
+///   Error if util name is not recognized.
+/// * --help or -h and a following util name: \
+///   Output help for that specific utility. \
+///   So 'coreutils sum --help' is the same as 'coreutils --help sum'.
 #[allow(clippy::cognitive_complexity)]
 fn main() {
     uucore::panic::mute_sigpipe_panic();
 
     let utils = util_map();
     let mut args = uucore::args_os();
 
+    // get binary which is always the first argument and remove it from args
     let binary = validation::binary_path(&mut args);
     let binary_as_util = validation::name(&binary).unwrap_or_else(|| {
+        // non UTF-8 name
         usage(&utils, "<unknown binary name>");
         process::exit(0);
     });
 
-    // binary name ends with util name?
-    let is_coreutils = binary_as_util.ends_with("utils");
-    let matched_util = utils
-        .keys()
-        .filter(|&&u| binary_as_util.ends_with(u) && !is_coreutils)
-        .max_by_key(|u| u.len()); //Prefer stty more than tty. *utils is not ls
-
-    let util_name = if let Some(&util) = matched_util {
-        Some(OsString::from(util))
-    } else if is_coreutils || binary_as_util.ends_with("box") {
-        // todo: Remove support of "*box" from binary
+    // get the called util
+    let util_os = if binary_as_util.ends_with("utils") {
+        // coreutils
         uucore::set_utility_is_second_arg();
-        args.next()
+        match args.next() {
+            Some(u) => u,
+            None => {
+                // no arguments provided
+                usage(&utils, binary_as_util);
+                process::exit(0);
+            }
+        }
     } else {
-        validation::not_found(&OsString::from(binary_as_util));
+        // Is the binary name a prefixed util name?
+        // Prefer stty more than tty. *utils is not ls
+        let name = if let Some(matched_util) = utils
+            .keys()
+            .filter(|&&util_name| binary_as_util.ends_with(util_name))
+            .max_by_key(|u| u.len())
+        {
+            *matched_util
+        } else {
+            binary_as_util
+        };
+
+        OsString::from(name)
     };
 
-    // 0th argument equals util name?
-    if let Some(util_os) = util_name {
-        let Some(util) = util_os.to_str() else {
-            validation::not_found(&util_os)
-        };
+    let Some(util) = util_os.to_str() else {
+        // non-UTF-8 name
+        validation::not_found(&util_os)
+    };
 
-        match util {
-            "--list" => {
-                // If --help is also present, show usage instead of list
-                if args.any(|arg| arg == "--help" || arg == "-h") {
-                    usage(&utils, binary_as_util);
-                    process::exit(0);
+    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"))]
+            validation::setup_localization_or_exit(util);
+            process::exit(uumain(vec![util_os].into_iter().chain(args)));
+        }
+        None => {
+            let (option, help_util) = find_dominant_option(&util_os, &mut args);
+            match option {
+                SelectedOption::Help => match help_util {
+                    // see if they want help on a specific util and if it is valid
+                    Some(u_os) => match utils.get(&u_os.to_string_lossy()) {
+                        Some(&(uumain, _)) => {
+                            let code = uumain(
+                                vec![u_os, OsString::from("--help")]
+                                    .into_iter()
+                                    // Function requires a chain like in the Some case, but
+                                    // the args are discarded as clap returns help immediately.
+                                    .chain(args),
+                            );
+                            io::stdout().flush().expect("could not flush stdout");
+                            process::exit(code);
+                        }
+                        None => validation::not_found(&u_os),
+                    },
+                    // show coreutils help
+                    None => usage(&utils, binary_as_util),
+                },
+                SelectedOption::Version => {
+                    println!("{binary_as_util} {VERSION} (multi-call binary)");
                 }
-                let utils: Vec<_> = utils.keys().collect();
-                for util in utils {
-                    println!("{util}");
+                SelectedOption::List => {
+                    let utils: Vec<_> = utils.keys().collect();
+                    for util in utils {
+                        println!("{util}");
+                    }
+                }
+                SelectedOption::Unrecognized(arg) => {
+                    // Argument looks like an option but wasn't recognized
+                    validation::unrecognized_option(binary_as_util, &arg);
                 }
-                process::exit(0);
-            }
-            "--version" | "-V" => {
-                println!("{binary_as_util} {VERSION} (multi-call binary)");
-                process::exit(0);
             }
-            // Not a special command: fallthrough to calling a util
-            _ => {}
+            // process::exit(0);
         }
+    }
+}
+
+/// The dominant selected option.
+#[derive(Debug, Clone, PartialEq)]
+enum SelectedOption {
+    Help,
+    Version,
+    List,
+    Unrecognized(OsString),
+}
 
-        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"))]
-                validation::setup_localization_or_exit(util);
-                process::exit(uumain(vec![util_os].into_iter().chain(args)));
+/// Coreutils only accepts one single option,
+/// if multiple are given, use the most dominant one.
+///
+/// Help > Version > List (e.g. 'coreutils --list --version' will return version)
+/// Unrecognized will return immediately.
+///
+/// # Returns
+/// (SelectedOption, Util for help request, if any)
+fn find_dominant_option(
+    first_arg: &OsString,
+    args: &mut impl Iterator<Item = OsString>,
+) -> (SelectedOption, Option<OsString>) {
+    let mut sel = identify_option_from_partial_text(first_arg);
+    match sel {
+        SelectedOption::Help => return (SelectedOption::Help, args.next()),
+        SelectedOption::Unrecognized(_) => {
+            return (sel, None);
+        }
+        _ => {}
+    };
+    // check remaining options, allows multiple
+    while let Some(arg) = args.next() {
+        let so = identify_option_from_partial_text(&arg);
+        match so {
+            // most dominant, return directly
+            SelectedOption::Help => {
+                // if help is wanted, check if a tool was named
+                return (so, args.next());
             }
-            None => {
-                if util == "--help" || util == "-h" {
-                    // see if they want help on a specific util
-                    if let Some(util_os) = args.next() {
-                        let Some(util) = util_os.to_str() else {
-                            validation::not_found(&util_os)
-                        };
-
-                        match utils.get(util) {
-                            Some(&(uumain, _)) => {
-                                let code = uumain(
-                                    vec![util_os, OsString::from("--help")]
-                                        .into_iter()
-                                        .chain(args),
-                                );
-                                io::stdout().flush().expect("could not flush stdout");
-                                process::exit(code);
-                            }
-                            None => validation::not_found(&util_os),
-                        }
-                    }
-                    usage(&utils, binary_as_util);
-                    process::exit(0);
-                } else if util.starts_with('-') {
-                    // Argument looks like an option but wasn't recognized
-                    validation::unrecognized_option(binary_as_util, &util_os);
-                } else {
-                    validation::not_found(&util_os);
+            // best after help, can be set directly
+            SelectedOption::Version => sel = SelectedOption::Version,
+            SelectedOption::List => {
+                if sel != SelectedOption::Version {
+                    sel = SelectedOption::List
                 }
             }
+            // unrecognized is not allowed
+            SelectedOption::Unrecognized(_) => {
+                return (so, None);
+            }
         }
-    } else {
-        // no arguments provided
-        usage(&utils, binary_as_util);
-        process::exit(0);
+    }
+
+    (sel, None)
+}
+
+// Will identify one, SelectedOption::None cannot be returned.
+fn identify_option_from_partial_text(arg: &OsString) -> SelectedOption {
+    let mut option = &arg.to_string_lossy()[..];
+    if let Some(p) = option.find('=') {
+        option = &option[0..p];
+    }
+    // // don't care about hyphens, -h and --h(elp) are identical
+    // let option = option.replace("-", "");
+    let l = option.len();
+    let possible_opts: Vec<usize> = COREUTILS_OPTIONS
+        .iter()
+        .enumerate()
+        .filter(|(_, it)| it.len() >= l && &it[0..l] == option)
+        .map(|(id, _)| id)
+        .collect();
+
+    match possible_opts.len() {
+        // exactly one hit
+        1 => match &possible_opts[0] {
+            0 => SelectedOption::List,
+            1 | 2 => SelectedOption::Version,
+            _ => SelectedOption::Help,
+        },
+        // None or more hits. The latter can not happen with the allowed options.
+        _ => SelectedOption::Unrecognized(arg.to_os_string()),
     }
 }

tests/test_util_name.rs

@@ -192,7 +192,7 @@ fn util_version() {
         println!("Skipping test: Binary not found at {:?}", scenario.bin_path);
         return;
     }
-    for arg in ["-V", "--version"] {
+    for arg in ["-V", "--version", "--ver"] {
         let child = Command::new(&scenario.bin_path)
             .arg(arg)
             .stdin(Stdio::piped())
@@ -209,6 +209,56 @@ fn util_version() {
     }
 }
 
+#[test]
+fn util_help() {
+    use std::process::{Command, Stdio};
+
+    let scenario = TestScenario::new("--version");
+    if !scenario.bin_path.exists() {
+        println!("Skipping test: Binary not found at {:?}", scenario.bin_path);
+        return;
+    }
+    for arg in ["-h", "--help", "--he"] {
+        let child = Command::new(&scenario.bin_path)
+            .arg(arg)
+            .stdin(Stdio::piped())
+            .stdout(Stdio::piped())
+            .stderr(Stdio::piped())
+            .spawn()
+            .unwrap();
+        let output = child.wait_with_output().unwrap();
+        assert_eq!(output.status.code(), Some(0));
+        assert_eq!(output.stderr, b"");
+        let output_str = String::from_utf8(output.stdout).unwrap();
+        assert!(output_str.contains("Usage: coreutils"));
+        assert!(output_str.contains("lists all defined functions"));
+    }
+}
+
+#[test]
+fn util_arg_priority() {
+    use std::process::{Command, Stdio};
+
+    let scenario = TestScenario::new("--version");
+    if !scenario.bin_path.exists() {
+        println!("Skipping test: Binary not found at {:?}", scenario.bin_path);
+        return;
+    }
+    let child = Command::new(&scenario.bin_path)
+        .arg("--list")
+        .arg("--version")
+        .stdout(Stdio::piped())
+        .stderr(Stdio::piped())
+        .spawn()
+        .unwrap();
+    let output = child.wait_with_output().unwrap();
+    assert_eq!(output.status.code(), Some(0));
+    assert_eq!(output.stderr, b"");
+    let output_str = String::from_utf8(output.stdout).unwrap();
+    let ver = env::var("CARGO_PKG_VERSION").unwrap();
+    assert_eq!(format!("coreutils {ver} (multi-call binary)\n"), output_str);
+}
+
 #[test]
 #[cfg(target_env = "musl")]
 fn test_musl_no_dynamic_deps() {