-
-
Notifications
You must be signed in to change notification settings - Fork 28
Windows Installation
First download and install miniconda. Unless you have reason to do otherwise
check the option to add miniconda to the PATH
environment variable. Next, execute the following in a command shell.
conda install -c conda-forge jupyterlab jupyter_console m2w64-gcc m2w64-zeromq
Ensure that you have a functioning Lisp implementation with Quicklisp installed. Execute the equivalent with your Lisp implementation.
sbcl --non-interactive --eval "(ql:quickload :common-lisp-jupyter)" --eval "(clj:install)"
First install Python. After Python has been installed ensure that the path the appropriate paths to find python
and pip
are in the system environment variable Path
. If you installed Python as a single user installation the paths are probably like the following.
%USERPROFILE%\AppData\Local\Programs\Python\Python37
%USERPROFILE%\AppData\Local\Programs\Python\Python37\Scripts
Once Python has been installed then download the Roswell archive and extract into your user profile directory (home directory). Add the following paths to Path
.
%USERPROFILE%\roswell
%USERPROFILE%\.roswell\bin
Open a command shell and execute the following (use the implementation you prefer in place of sbcl-bin).
ros install sbcl-bin
ros shell
Once the shell opens execute the following.
pacman -Syu mingw64/mingw-w64-x86_64-gcc mingw64/mingw-w64-x86_64-zeromq mingw32/mingw-w64-i686-zeromq
ros install yitzchak/common-lisp-jupyter
Close the shell and add the appropriate path to Path
so that cl-jupyter
can find libzmq.dll
. It will be a folder inside the .roswell
directory that looks like the following.
%USERPROFILE%\.roswell\impls\x86-64\windows\msys2\20180531\mingw64\bin
Close any open command shells and then open a new command shell and execute the following to test the install.
jupyter-console --kernel=common-lisp