reproc

reproc

• What is reproc?
• Features
• Questions
• Installation
• Dependencies
• CMake options
• Documentation
• Error handling
• Multithreading
• Gotchas

What is reproc?

reproc (Redirected Process) is a cross-platform C/C++ library that simplifies starting, stopping and communicating with external programs. The main use case is executing command line applications directly from C or C++ code and retrieving their output.

reproc consists out of two libraries: reproc and reproc++. reproc is a C99 library that contains the actual code for working with external programs. reproc++ depends on reproc and adapts its API to an idiomatic C++11 API. It also adds a few extras that simplify working with external programs from C++.

Features

• Start any program directly from C or C++ code.
• Communicate with a program via its standard streams.
• Wait for a program to exit or forcefully stop it yourself. When forcefully stopping a process you can either allow the process to clean up its resources or stop it immediately.
• The core library (reproc) is written in C99. An optional C++11 wrapper library (reproc++) with extra features is available for use in C++ applications.
• Multiple installation methods. Either build reproc as part of your project or use a system installed version of reproc.

Usage

#include <reproc/run.h>

int main(void)
{
  const char *args[] = { "echo", "Hello, world!", NULL };
  return reproc_run(args, (reproc_options) { 0 });
}

Questions

If you have any questions after reading the readme and documentation you can either make an issue or ask questions directly in the reproc gitter channel.

Installation

Note: Building reproc requires CMake 3.12 or higher.

There are multiple ways to get reproc into your project. One way is to build reproc as part of your project using CMake. To do this, we first have to get the reproc source code into the project. This can be done using any of the following options:

• When using CMake 3.11 or later, you can use the CMake FetchContent API to download reproc when running CMake. See https://cliutils.gitlab.io/modern-cmake/chapters/projects/fetch.html for an example.
• Another option is to include reproc’s repository as a git submodule. https://cliutils.gitlab.io/modern-cmake/chapters/projects/submodule.html provides more information.
• A very simple solution is to just include reproc’s source code in your repository. You can download a zip of the source code without the git history and add it to your repository in a separate directory.

After including reproc’s source code in your project, it can be built from the root CMakeLists.txt file as follows:

add_subdirectory(<path-to-reproc>) # For example: add_subdirectory(external/reproc)

CMake options can be specified before calling add_subdirectory:

set(REPROC++ ON)
add_subdirectory(<path-to-reproc>)

Note: If the option has already been cached in a previous CMake run, you’ll have to clear CMake’s cache to apply the new default value.

For more information on configuring reproc’s build, see CMake options.

You can also depend on an installed version of reproc. You can either build and install reproc yourself or install reproc via a package manager. reproc is available in the following package repositories:

• Arch User Repository (https://aur.archlinux.org/packages/reproc)
• vcpkg (https://github.com/microsoft/vcpkg/tree/master/ports/reproc)

If using a package manager is not an option, you can build and install reproc from source (CMake 3.13+):

cmake -B build
cmake --build build
cmake --install build

Enable the REPROC_TEST option and build the test target to run the tests (CMake 3.13+):

cmake -B build -DREPROC_TEST=ON
cmake --build build
cmake --build build --target test

After installing reproc your build system will have to find it. reproc provides both CMake config files and pkg-config files to simplify finding a reproc installation using CMake and pkg-config respectively. Note that reproc and reproc++ are separate libraries and as a result have separate config files as well. Make sure to search for the one you want to use.

To find an installed version of reproc using CMake:

find_package(reproc) # Find reproc.
find_package(reproc++) # Find reproc++.

After building reproc as part of your project or finding a installed version of reproc, you can link against it from within your CMakeLists.txt file as follows:

target_link_libraries(myapp reproc) # Link against reproc.
target_link_libraries(myapp reproc++) # Link against reproc++.

From Meson 0.53.2 onwards, reproc can be included as a CMake subproject in Meson build scripts. See https://mesonbuild.com/CMake-module.html for more information.

Dependencies

By default, reproc has a dependency on pthreads on POSIX systems (-pthread) and a dependency on Winsock 2.2 on Windows systems (-lws2_32). CMake and pkg-config handle these dependencies automatically.

CMake options

reproc’s build can be configured using the following CMake options:

User

• REPROC++: Build reproc++ (default: ${REPROC_DEVELOP})

• REPROC_TEST: Build tests (default: ${REPROC_DEVELOP})

  Run the tests by running the test binary which can be found in the build directory after building reproc.

• REPROC_EXAMPLES: Build examples (default: ${REPROC_DEVELOP})

  The resulting binaries will be located in the examples folder of each project subdirectory in the build directory after building reproc.

Advanced

• REPROC_OBJECT_LIBRARIES: Build CMake object libraries (default: ${REPROC_DEVELOP})

  This is useful to directly include reproc in another library. When building reproc as a static or shared library, it has to be installed alongside the consuming library which makes distributing the consuming library harder. When using object libraries, reproc’s object files are included directly into the consuming library and no extra installation is necessary.

  Note: reproc’s object libraries will only link correctly from CMake 3.14 onwards.

  Note: This option overrides BUILD_SHARED_LIBS.

• REPROC_INSTALL: Generate installation rules (default: ON unless REPROC_OBJECT_LIBRARIES is enabled)

• REPROC_INSTALL_CMAKECONFIGDIR: CMake config files installation directory (default: ${CMAKE_INSTALL_LIBDIR}/cmake)

• REPROC_INSTALL_PKGCONFIG: Install pkg-config files (default: ON)

• REPROC_INSTALL_PKGCONFIGDIR: pkg-config files installation directory (default: ${CMAKE_INSTALL_LIBDIR}/pkgconfig)

• REPROC_MULTITHREADED: Use pthread_sigmask and link against the system’s thread library (default: ON)

Developer

• REPROC_DEVELOP: Configure option default values for development (default: OFF unless the REPROC_DEVELOP environment variable is set)
• REPROC_SANITIZERS: Build with sanitizers (default: ${REPROC_DEVELOP})
• REPROC_TIDY: Run clang-tidy when building (default: ${REPROC_DEVELOP})
• REPROC_WARNINGS: Enable compiler warnings (default: ${REPROC_DEVELOP})
• REPROC_WARNINGS_AS_ERRORS: Add -Werror or equivalent to the compile flags and clang-tidy (default: OFF)

Documentation

Each function and class is documented extensively in its header file. Examples can be found in the examples subdirectory of reproc and reproc++.

Error handling

On failure, Most functions in reproc’s API return a negative errno (POSIX) or GetLastError (Windows) style error code. For actionable errors, reproc provides constants (REPROC_ETIMEDOUT, REPROC_EPIPE, …) that can be used to match against the error without having to write platform-specific code. To get a string representation of an error, pass it to reproc_strerror.

reproc++'s API integrates with the C++ standard library error codes mechanism (std::error_code and std::error_condition). Most methods in reproc++'s API return std::error_code values that contain the actual system error that occurred. You can test against these error codes using values from the std::errc enum.

See the examples for more information on how to handle errors when using reproc.

Note:

Both reproc and reproc++ APIs take options argument that may define one or more stop actions such as terminate or kill. For that reason if the child process is being terminated or killed using a signal on POSIX, the error code will not reflect an error.

It’s up to the downstream project to interpret status codes reflecting unexpected behaviors alongside error codes (see this example).

Multithreading

Don’t call the same operation on the same child process from more than one thread at the same time. For example: reading and writing to a child process from different threads is fine but waiting on the same child process from two different threads at the same time will result in issues.

Gotchas

• (POSIX) It is strongly recommended to not call waitpid on pids of processes started by reproc.

  reproc uses waitpid to wait until a process has exited. Unfortunately, waitpid cannot be called twice on the same process. This means that reproc_wait won’t work correctly if waitpid has already been called on a child process beforehand outside of reproc.

• It is strongly recommended to make sure each child process actually exits using reproc_wait or reproc_stop.

  On POSIX, a child process that has exited is a zombie process until the parent process waits on it using waitpid. A zombie process takes up resources and can be seen as a resource leak so it is important to make sure all processes exit correctly in a timely fashion.

• It is strongly recommended to try terminating a child process by waiting for it to exit or by calling reproc_terminate before resorting to reproc_kill.

  When using reproc_kill the child process does not receive a chance to perform cleanup which could result in resources being leaked. Chief among these leaks is that the child process will not be able to stop its own child processes. Always try to let a child process exit normally by calling reproc_terminate before calling reproc_kill. reproc_stop is a handy helper function that can be used to perform multiple stop actions in a row with timeouts inbetween.

• (POSIX) It is strongly recommended to ignore the SIGPIPE signal in the parent process.

  On POSIX, writing to a closed stdin pipe of a child process will terminate the parent process with the SIGPIPE signal by default. To avoid this, the SIGPIPE signal has to be ignored in the parent process. If the SIGPIPE signal is ignored reproc_write will return REPROC_EPIPE as expected when writing to a closed stdin pipe.

• While reproc_terminate allows the child process to perform cleanup it is up to the child process to correctly clean up after itself. reproc only sends a termination signal to the child process. The child process itself is responsible for cleaning up its own child processes and other resources.

• (Windows) reproc_kill is not guaranteed to kill a child process immediately on Windows. For more information, read the Remarks section in the documentation of the Windows TerminateProcess function that reproc uses to kill child processes on Windows.

• Child processes spawned via reproc inherit a single extra file handle which is used to wait for the child process to exit. If the child process closes this file handle manually, reproc will wrongly detect the child process has exited. If this handle is further inherited by other processes that outlive the child process, reproc will detect the child process is still running even if it has exited. If data is written to this handle, reproc will also wrongly detect the child process has exited.

• (Windows) It’s not possible to detect if a child process closes its stdout or stderr stream before exiting. The parent process will only be notified that a child process output stream is closed once that child process exits.

• (Windows) reproc assumes that Windows creates sockets that are usable as file system objects. More specifically, the default sockets returned by WSASocket should have the XP1_IFS_HANDLES flag set. This might not be the case if there are external LSP providers installed on a Windows machine. If this is the case, we recommend removing the software that’s providing the extra service providers since they’re deprecated and should not be used anymore (see https://docs.microsoft.com/en-us/windows/win32/winsock/categorizing-layered-service-providers-and-applications).