C++ implementation of the Jupyter Kernel protocol
Introduction
xeus is a library meant to facilitate the implementation of kernels for Jupyter. It takes the
burden of implementing the Jupyter Kernel protocol so developers can focus on implementing the
interpreter part of the kernel.
Several Jupyter kernels are built upon xeus, such as xeus-cling,
a kernel for the C++ programming language, and xeus-python, an alternative Python kernel for Jupyter.
Installation
xeus has been packaged on all platforms for the mamba (or conda) package manager.
mamba install xeus -c conda-forge
Documentation
To get started with using xeus, check out the full documentation
http://xeus.readthedocs.io/
Usage
xeus enables custom kernel authors to implement Jupyter kernels more easily. It takes the burden of implementing the Jupyter Kernel protocol so developers can focus on implementing the interpreter part of the Kernel.
The easiest way to get started with a new kernel is to inherit from the base interpreter class xeus::xinterpreter and implement the private virtual methods:
configure_implexecute_request_implcomplete_request_implinspect_request_implis_complete_request_implkernel_info_request_implshutdown_request_impl
as seen in the documentation.
#include "xeus/xinterpreter.hpp"
#include "nlohmann/json.hpp"
using xeus::xinterpreter;
namespace nl = nlohmann;
namespace custom
{
class custom_interpreter : public xinterpreter
{
public:
custom_interpreter() = default;
virtual ~custom_interpreter() = default;
private:
void configure_impl() override;
void execute_request_impl(send_reply_callback cb,
int execution_counter,
const std::string& code,
execute_request_config config,
nl::json user_expressions) override;
nl::json complete_request_impl(const std::string& code,
int cursor_pos) override;
nl::json inspect_request_impl(const std::string& code,
int cursor_pos,
int detail_level) override;
nl::json is_complete_request_impl(const std::string& code) override;
nl::json kernel_info_request_impl() override;
nl::json shutdown_request_impl(bool restart) override;
nl::json interrupt_request_impl() override;
};
}
Kernel authors can then rebind to the native APIs of the interpreter that is being interfaced, providing richer information than with the classical approach of a wrapper kernel capturing textual output.
Building from Source
xeus depends on nlohmann_json.
| xeus |
nlohmann json |
| master |
^3.12.0 |
| 6.x |
^3.12.0 |
| 5.x |
^3.11.0 |
Versions prior to version 5 depend on the following libraries: nlohmann_json and xtl.
| xeus |
xtl |
nlohmann json |
| master |
>=0.7.0,<0.8.0 |
^3.11.0 |
| 4.x |
>=0.7.0,<0.8.0 |
^3.11.0 |
| 3.x |
>=0.7.0,<0.8.0 |
^3.2.0 |
Versions prior to version 3 also depend on the following libraries: ZeroMQ,
cppzmq, and OpenSSL.
| xeus |
ZeroMQ |
cppzmq |
xtl |
nlohmann json |
OpenSSL |
| 2.4.1 |
^4.2.5 |
^4.7.1 |
>=0.7.0,<0.8.0 |
^3.2.0 |
^1.0.1 |
| 2.4.0 |
^4.2.5 |
^4.7.1 |
>=0.7.0,<0.8.0 |
^3.2.0 |
^1.0.1 |
| 2.3.1 |
^4.2.5 |
^4.7.1 |
>=0.7.0,<0.8.0 |
^3.2.0 |
^1.0.1 |
| 2.3.0 |
^4.2.5 |
^4.7.1 |
>=0.7.0,<0.8.0 |
^3.2.0 |
^1.0.1 |
| 2.2.0 |
^4.2.5 |
^4.7.1 |
>=0.7.0,<0.8.0 |
^3.2.0 |
^1.0.1 |
| 2.1.1 |
^4.2.5 |
^4.7.1 |
>=0.7.0,<0.8.0 |
^3.2.0 |
^1.0.1 |
| 2.1.0 |
^4.2.5 |
^4.7.1 |
>=0.7.0,<0.8.0 |
^3.2.0 |
^1.0.1 |
| 2.0.0 |
^4.2.5 |
^4.7.1 |
>=0.7.0,<0.8.0 |
^3.2.0 |
^1.0.1 |
| 1.0.4 |
^4.2.5 |
^4.7.1 |
>=0.7.0,<0.8.0 |
^3.2.0 |
^1.0.1 |
| 1.0.3 |
^4.2.5 |
^4.7.1 |
>=0.7.0,<0.8.0 |
^3.2.0 |
^1.0.1 |
| 1.0.2 |
^4.2.5 |
^4.7.1 |
>=0.7.0,<0.8.0 |
^3.2.0 |
^1.0.1 |
| 1.0.1 |
^4.2.5 |
^4.7.1 |
>=0.7.0,<0.8.0 |
^3.2.0 |
^1.0.1 |
| 1.0.0 |
^4.2.5 |
^4.7.1 |
>=0.7.0,<0.8.0 |
^3.2.0 |
^1.0.1 |
On Linux platforms, xeus also requires libuuid, which is available in all linux distributions (uuid-dev on Debian).
We have packaged all these dependencies on conda-forge. The simplest way to install them is to run:
mamba install cmake pkg-config xtl nlohmann_json -c conda-forge
On Linux platforms, you will also need libuuid:
mamba install libuuid -c conda-forge
Once you have installed the dependencies, you can build and install xeus.
From the build directory, run:
cmake -D CMAKE_BUILD_TYPE=Release ..
make
make install
Installing the Dependencies from Source
The dependencies can also be installed from source. Simply clone the directories and run the following cmake (cmake >= 3.8) and make instructions.
ZeroMQ
ZeroMQ is the messaging library underlying the Jupyter kernel protocol.
cmake -D WITH_PERF_TOOL=OFF -D ZMQ_BUILD_TESTS=OFF -D ENABLE_CPACK=OFF
-D CMAKE_BUILD_TYPE=Release
make
make install
OpenSSL
OpenSSL is packaged for most package managers (apt-get, rpm, mamba).
We recommend making use of an off-the-shelf build of OpenSSL for your system.
For more information on building OpenSSL, check out the official OpenSSL wiki.
cppzmq
cppzmq is a header only library:
cmake -D CMAKE_BUILD_TYPE=Release
make install
json for modern cpp
nlohmann_json is a header only library
cmake
make install
xtl
xtl is a header only library:
cmake -D CMAKE_BUILD_TYPE
make install
Contributing
See CONTRIBUTING.md to know how to contribute and set up a development environment.
License
We use a shared copyright model that enables all contributors to maintain the
copyright on their contributions.
This software is licensed under the BSD-3-Clause license. See the LICENSE file for details.
评论