feat: help, utf8 supports

Author: Charliechen114514

CMakeLists.txt

@@ -15,12 +15,22 @@ include(cmake/third_party/CPM.cmake)
 # ── Compiler flags ────────────────────────────────────────────
 include(cmake/compile/CompilerFlag.cmake)
 
-# ── Applet sources (shared between main exe and tests) ────────
-file(GLOB_RECURSE CFBOX_APPLET_SOURCES CONFIGURE_DEPENDS src/applets/*.cpp)
+# ── Applet configuration ─────────────────────────────────────
+include(cmake/Config.cmake)
+
+# ── Applet sources (conditional on config) ───────────────────
+set(CFBOX_APPLET_SOURCES)
+foreach(applet IN LISTS CFBOX_APPLETS)
+    string(TOUPPER "${applet}" APPLET_UPPER)
+    if(CFBOX_ENABLE_${APPLET_UPPER})
+        list(APPEND CFBOX_APPLET_SOURCES ${CMAKE_CURRENT_SOURCE_DIR}/src/applets/${applet}.cpp)
+    endif()
+endforeach()
 
 # ── Main executable ───────────────────────────────────────────
 add_executable(cfbox src/main.cpp ${CFBOX_APPLET_SOURCES})
 target_include_directories(cfbox PUBLIC include)
+target_include_directories(cfbox PUBLIC ${CMAKE_CURRENT_BINARY_DIR}/include)
 target_link_libraries(cfbox PRIVATE cfbox_compiler_flags)
 
 # ── GTest via CPM (FetchContent) ──────────────────────────────
@@ -39,6 +49,7 @@ if(GTest_ADDED)
     if(CFBOX_TEST_SOURCES)
         add_executable(cfbox_tests ${CFBOX_TEST_SOURCES} ${CFBOX_APPLET_SOURCES})
         target_include_directories(cfbox_tests PUBLIC include)
+        target_include_directories(cfbox_tests PUBLIC ${CMAKE_CURRENT_BINARY_DIR}/include)
         target_link_libraries(cfbox_tests PRIVATE
             cfbox_compiler_flags
             GTest::gtest_main

CONTRIBUTING.md

@@ -31,10 +31,10 @@ Release builds use `-O2` by default and enable LTO. For size-optimized builds, a
 ## Running Tests
 
 ```bash
-# Unit tests (108 GTest cases)
+# Unit tests (149 GTest cases)
 ctest --test-dir build --output-on-failure
 
-# Integration tests (16 shell scripts comparing against GNU coreutils)
+# Integration tests (17 shell scripts comparing against GNU coreutils)
 bash tests/integration/run_all.sh
 ```
 
@@ -138,12 +138,36 @@ The CI pipeline ([ci.yml](.github/workflows/ci.yml)) runs on every push/PR to `m
 
 1. Create `src/applets/<name>.cpp` with signature `auto <name>_main(int argc, char* argv[]) -> int`.
    - Add a comment header listing supported flags and known differences from GNU.
-2. Declare the function in [applets.hpp](include/cfbox/applets.hpp).
-3. Add one entry to `APPLET_REGISTRY` in [applets.hpp](include/cfbox/applets.hpp).
-4. Add GTest unit tests in `tests/unit/test_<name>.cpp` (see [test_capture.hpp](tests/unit/test_capture.hpp) for stdout capture and `TempDir` utilities).
-5. Add shell integration tests in `tests/integration/test_<name>.sh` following the pattern in existing scripts.
+   - Add a `constexpr cfbox::help::HelpEntry HELP` constant in the anonymous namespace.
+   - Handle `--help` / `--version` right after `args::parse()`:
+     ```cpp
+     if (parsed.has_long("help"))    { cfbox::help::print_help(HELP); return 0; }
+     if (parsed.has_long("version")) { cfbox::help::print_version(HELP); return 0; }
+     ```
+2. Declare the function in [applets.hpp](include/cfbox/applets.hpp), guarded by `#if CFBOX_ENABLE_<UPPER>`.
+3. Add one entry to `APPLET_REGISTRY` in [applets.hpp](include/cfbox/applets.hpp), also guarded by `#if CFBOX_ENABLE_<UPPER>`.
+4. Add the applet name to the `CFBOX_APPLETS` list in [cmake/Config.cmake](cmake/Config.cmake).
+5. Add a `#cmakedefine01 CFBOX_ENABLE_<UPPER>` line to [include/cfbox/applet_config.hpp.in](include/cfbox/applet_config.hpp.in).
+6. Add GTest unit tests in `tests/unit/test_<name>.cpp` (see [test_capture.hpp](tests/unit/test_capture.hpp) for stdout capture and `TempDir` utilities). Guard the test file with `#if CFBOX_ENABLE_<UPPER>`.
+7. Add shell integration tests in `tests/integration/test_<name>.sh` following the pattern in existing scripts.
 
-> **Note:** The `init` applet is special — it runs as PID 1 in QEMU system-mode tests. Regular applets should not need special PID 1 handling.
+> **Note:** The `init` applet is special — it runs as PID 1 in QEMU system-mode tests and uses manual `argv` scanning instead of `args::parse()`. Regular applets should not need special PID 1 handling.
+
+## Build Configuration
+
+CFBox supports per-applet configuration via CMake options:
+
+```bash
+# Disable individual applets
+cmake -DCFBOX_ENABLE_GREP=OFF -DCFBOX_ENABLE_SED=OFF ..
+
+# Use preset profiles
+cmake -DCFBOX_PROFILE=minimal ..   # Only core file operations
+cmake -DCFBOX_PROFILE=embedded ..  # Everything except optional text processing
+cmake -DCFBOX_PROFILE=desktop ..   # All applets enabled (default)
+```
+
+Available profiles: `minimal` (echo, cat, ls, cp, mv, rm, mkdir, grep), `embedded` (all except sort/uniq/sed), `desktop`/`full` (all enabled).
 
 ## Submitting Changes
 

README.en.md

@@ -8,12 +8,12 @@ A minimalist BusyBox alternative written in modern C++23.
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 [![C++23](https://img.shields.io/badge/C++23-00599C?logo=cplusplus)](https://en.cppreference.com/w/cpp/23)
 [![CMake](https://img.shields.io/badge/CMake-3.26+-064F8C?logo=cmake)](https://cmake.org/)
-[![Tests](https://img.shields.io/badge/Tests-124_passing-brightgreen)](tests/)
+[![Tests](https://img.shields.io/badge/Tests-149_passing-brightgreen)](tests/)
 [![Applets](https://img.shields.io/badge/Applets-17-brightgreen)](src/applets/)
 
 ## Overview
 
-CFBox is a single-executable Unix utility collection distributed via symbolic links. All development is complete — 17 applets implemented and tested, with a CI pipeline covering native builds, cross-compilation, and QEMU user/system-mode testing across 5 stages.
+CFBox is a single-executable Unix utility collection distributed via symbolic links. 17 applets implemented and tested, with a CI pipeline covering native builds, cross-compilation, and QEMU user/system-mode testing across 5 stages. Features configurable CMake builds (per-applet toggles), GNU-style long options, and colored help output.
 
 **Design philosophy:** Simplicity first — Modern C++ (`std::expected`) — Embedded-friendly (cross-compilation, static linking)
 
@@ -25,8 +25,8 @@ cmake -B build
 cmake --build build
 
 # Test
-ctest --test-dir build --output-on-failure   # 108 GTest unit tests
-bash tests/integration/run_all.sh            # 16 integration test scripts
+ctest --test-dir build --output-on-failure   # 149 GTest unit tests
+bash tests/integration/run_all.sh            # 17 integration test scripts
 
 # Run via subcommand
 ./build/cfbox echo "Hello, World!"
@@ -42,7 +42,7 @@ echo "Hello, World!"   # now calls cfbox via symlink
 
 | Applet | Supported Flags / Features |
 |--------|----------------------------|
-| `echo` | `-n` (no trailing newline), `-e` (interpret escape sequences) |
+| `echo` | `-n` (no trailing newline), `-e` (interpret escape sequences), all applets support `--help` / `--version` |
 | `printf` | Format strings (`%s` `%d` `%f` `%c` `%%`), format reuse |
 | `cat` | `-n` (number lines), `-b` (number non-blank), `-A` (show non-printing), stdin passthrough |
 | `head` | `-n N` (first N lines), `-c N` (first N bytes), multi-file headers |
@@ -57,16 +57,16 @@ echo "Hello, World!"   # now calls cfbox via symlink
 
 | Applet | Supported Flags / Features |
 |--------|----------------------------|
-| `mkdir` | `-p` (create parents), `-m MODE` (permissions) |
-| `rm` | `-r` (recursive), `-f` (force), `-i` (interactive), `/` safety check |
-| `cp` | `-r` (recursive), `-p` (preserve permissions), multi-file to directory |
+| `mkdir` | `-p`/`--parents` (create parents), `-m`/`--mode MODE` (permissions) |
+| `rm` | `-r`/`--recursive` (recursive), `-f`/`--force` (force), `-i` (interactive), `/` safety check |
+| `cp` | `-r`/`--recursive` (recursive), `-p`/`--preserve` (preserve permissions), multi-file to directory |
 | `mv` | `-f` (force overwrite), cross-filesystem fallback (copy + remove) |
 
 ### Directory & Search
 
 | Applet | Supported Flags / Features |
 |--------|----------------------------|
-| `ls` | `-a` (show hidden), `-l` (long format), `-h` (human-readable sizes) |
+| `ls` | `-a`/`--all` (show hidden), `-l`/`--long` (long format), `-h`/`--human-readable` (human-readable sizes) |
 | `find` | `-name PATTERN` (glob), `-type [f\|d\|l]`, `-maxdepth N`, `-exec CMD {} ;` |
 
 ### System
@@ -97,22 +97,30 @@ echo "Hello, World!"   # now calls cfbox via symlink
 cfbox/
 ├── CMakeLists.txt
 ├── cmake/
-│   ├── compile/CompilerFlag.cmake     # Compiler warnings & optimization flags
-│   ├── third_party/CPM.cmake         # CPM dependency manager
-│   └── toolchain/                    # Cross-compilation toolchains
+│   ├── Config.cmake                  # Per-applet configuration (CFBOX_ENABLE_xxx options)
+│   ├── compile/CompilerFlag.cmake    # Compiler warnings & optimization flags
+│   ├── third_party/CPM.cmake        # CPM dependency manager
+│   └── toolchain/                   # Cross-compilation toolchains
 ├── configs/
-│   └── qemu-virt-aarch64.config      # Minimal QEMU aarch64 kernel config
-├── document/                          # Detailed documentation
-├── include/cfbox/                     # Public headers
+│   └── qemu-virt-aarch64.config     # Minimal QEMU aarch64 kernel config
+├── document/                         # Detailed documentation
+├── include/cfbox/
+│   ├── applet_config.hpp.in         # CMake-generated config (version + enable flags)
+│   ├── applet.hpp / applets.hpp     # Registry & dispatch
+│   ├── args.hpp                     # Short + long option argument parser
+│   ├── help.hpp                     # --help / --version help system
+│   ├── term.hpp                     # ANSI colored output (NO_COLOR support)
+│   ├── utf8.hpp                     # Unicode-aware width/count utilities
+│   └── ...                          # error.hpp, io.hpp, fs_util.hpp, escape.hpp
 ├── src/
-│   ├── main.cpp                      # Dispatch entry
-│   └── applets/                      # 17 command implementations
+│   ├── main.cpp                     # Dispatch entry
+│   └── applets/                     # 17 command implementations
 ├── tests/
-│   ├── unit/                         # GTest unit tests (108 cases)
-│   └── integration/                  # Shell integration tests (16 scripts)
-├── scripts/                          # Build, test, install scripts
-├── .github/workflows/ci.yml          # CI pipeline
-└── CONTRIBUTING.md                    # Contributing guide
+│   ├── unit/                        # GTest unit tests (149 cases)
+│   └── integration/                 # Shell integration tests (17 scripts)
+├── scripts/                         # Build, test, install scripts
+├── .github/workflows/ci.yml         # CI pipeline
+└── CONTRIBUTING.md                   # Contributing guide
 ```
 
 ## Contributing

README.md

@@ -8,12 +8,12 @@
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 [![C++23](https://img.shields.io/badge/C++23-00599C?logo=cplusplus)](https://en.cppreference.com/w/cpp/23)
 [![CMake](https://img.shields.io/badge/CMake-3.26+-064F8C?logo=cmake)](https://cmake.org/)
-[![Tests](https://img.shields.io/badge/Tests-124_passing-brightgreen)](tests/)
+[![Tests](https://img.shields.io/badge/Tests-149_passing-brightgreen)](tests/)
 [![Applets](https://img.shields.io/badge/Applets-17-brightgreen)](src/applets/)
 
 ## 概述
 
-CFBox 是一个单一可执行文件的 Unix 工具集，通过符号链接分发。全部开发已完成，17 个 applet 已实现并通过测试，CI 流水线覆盖原生构建、交叉编译、QEMU 用户/系统模式 5 种测试场景。
+CFBox 是一个单一可执行文件的 Unix 工具集，通过符号链接分发。17 个 applet 已实现并通过测试，CI 流水线覆盖原生构建、交叉编译、QEMU 用户/系统模式 5 种测试场景。支持 CMake 配置化构建（per-applet 开关）、GNU 风格长选项、彩色帮助输出。
 
 **设计理念：** 简洁优先 — 现代C++（`std::expected`） — 嵌入式友好（交叉编译、静态链接）
 
@@ -25,8 +25,8 @@ cmake -B build
 cmake --build build
 
 # 测试
-ctest --test-dir build --output-on-failure   # 108 个 GTest 单元测试
-bash tests/integration/run_all.sh            # 16 套集成测试脚本
+ctest --test-dir build --output-on-failure   # 149 个 GTest 单元测试
+bash tests/integration/run_all.sh            # 17 套集成测试脚本
 
 # 通过子命令运行
 ./build/cfbox echo "Hello, World!"
@@ -42,7 +42,7 @@ echo "Hello, World!"   # 通过符号链接调用 cfbox
 
 | 命令 | 支持的标志 / 功能 |
 |------|-------------------|
-| `echo` | `-n`（不换行），`-e`（解释转义序列） |
+| `echo` | `-n`（不换行），`-e`（解释转义序列），所有 applet 支持 `--help` / `--version` |
 | `printf` | 格式字符串（`%s` `%d` `%f` `%c` `%%`），格式重用 |
 | `cat` | `-n`（显示行号），`-b`（非空行编号），`-A`（显示不可打印字符），stdin 透传 |
 | `head` | `-n N`（前 N 行），`-c N`（前 N 字节），多文件头部 |
@@ -57,16 +57,16 @@ echo "Hello, World!"   # 通过符号链接调用 cfbox
 
 | 命令 | 支持的标志 / 功能 |
 |------|-------------------|
-| `mkdir` | `-p`（递归创建父目录），`-m MODE`（设置权限） |
-| `rm` | `-r`（递归删除），`-f`（强制），`-i`（交互确认），`/` 安全检查 |
-| `cp` | `-r`（递归复制），`-p`（保留权限），多文件到目录 |
+| `mkdir` | `-p`/`--parents`（递归创建父目录），`-m`/`--mode MODE`（设置权限） |
+| `rm` | `-r`/`--recursive`（递归删除），`-f`/`--force`（强制），`-i`（交互确认），`/` 安全检查 |
+| `cp` | `-r`/`--recursive`（递归复制），`-p`/`--preserve`（保留权限），多文件到目录 |
 | `mv` | `-f`（强制覆盖），跨文件系统回退（复制 + 删除） |
 
 ### 目录与搜索
 
 | 命令 | 支持的标志 / 功能 |
 |------|-------------------|
-| `ls` | `-a`（显示隐藏文件），`-l`（长格式），`-h`（人类可读大小） |
+| `ls` | `-a`/`--all`（显示隐藏文件），`-l`/`--long`（长格式），`-h`/`--human-readable`（人类可读大小） |
 | `find` | `-name 模式`（glob 匹配），`-type [f\|d\|l]`，`-maxdepth N`，`-exec 命令 {} ;` |
 
 ### 系统
@@ -97,22 +97,30 @@ echo "Hello, World!"   # 通过符号链接调用 cfbox
 cfbox/
 ├── CMakeLists.txt
 ├── cmake/
-│   ├── compile/CompilerFlag.cmake     # 编译器警告与优化标志
-│   ├── third_party/CPM.cmake         # CPM 依赖管理
-│   └── toolchain/                    # 交叉编译工具链
+│   ├── Config.cmake                  # Per-applet 配置（CFBOX_ENABLE_xxx 选项）
+│   ├── compile/CompilerFlag.cmake    # 编译器警告与优化标志
+│   ├── third_party/CPM.cmake        # CPM 依赖管理
+│   └── toolchain/                   # 交叉编译工具链
 ├── configs/
-│   └── qemu-virt-aarch64.config      # QEMU aarch64 最小内核配置
-├── document/                          # 详细文档
-├── include/cfbox/                     # 公共头文件
+│   └── qemu-virt-aarch64.config     # QEMU aarch64 最小内核配置
+├── document/                         # 详细文档
+├── include/cfbox/
+│   ├── applet_config.hpp.in         # CMake 生成的配置（版本号 + 启用开关）
+│   ├── applet.hpp / applets.hpp     # 注册表与分发
+│   ├── args.hpp                     # 短选项 + 长选项参数解析器
+│   ├── help.hpp                     # --help / --version 帮助系统
+│   ├── term.hpp                     # ANSI 彩色输出（NO_COLOR 支持）
+│   ├── utf8.hpp                     # Unicode 感知的宽度/计数工具
+│   └── ...                          # error.hpp, io.hpp, fs_util.hpp, escape.hpp
 ├── src/
-│   ├── main.cpp                      # 分发入口
-│   └── applets/                      # 17 个命令实现
+│   ├── main.cpp                     # 分发入口
+│   └── applets/                     # 17 个命令实现
 ├── tests/
-│   ├── unit/                         # GTest 单元测试（108 个用例）
-│   └── integration/                  # Shell 集成测试（16 个脚本）
-├── scripts/                          # 构建、测试、安装脚本
-├── .github/workflows/ci.yml          # CI 流水线
-└── CONTRIBUTING.md                    # 贡献指南
+│   ├── unit/                        # GTest 单元测试（149 个用例）
+│   └── integration/                 # Shell 集成测试（17 个脚本）
+├── scripts/                         # 构建、测试、安装脚本
+├── .github/workflows/ci.yml         # CI 流水线
+└── CONTRIBUTING.md                   # 贡献指南
 ```
 
 ## 贡献