rz-c / macaulay2-jupyter-kernel Goto Github PK

View Code? Open in Web Editor NEW

5.0 2.0 6.0 4.4 MB

Jupyter kernel for Macaulay2

License: MIT License

Python 87.49% JavaScript 7.43% Dockerfile 3.85% Makefile 1.23%

macaulay2-jupyter-kernel's Introduction

M2JK — Jupyter kernel for Macaulay2

Beta Testing!

You can now use Jupyter (Notebook or Lab) as a front-end for Macaulay2.

See the demo for sample use and an outline of the kernel-specific features. For bugs or requests, open an issue. For recent changes, see the changelog.

Requirements

You need a recent version of Python and pip. Python 3 is recommended for build installs and necessary for source installs. You can install Jupyter directly from PyPI by

pip3 install --upgrade pip
pip3 install jupyter

Macaulay2 needs to be installed and on your path. If you are using Emacs as your front-end, it already is, but you can test it by which M2. Otherwise, you can achieve that by running setup() from within an M2 session. Alternatively, you can configure M2JK to use a specific binary.

Installation

You can install the latest release version directly from PyPI by

$ pip3 install macaulay2-jupyter-kernel
$ python3 -m m2_kernel.install

Alternatively, you can install the latest development version from source by

$ git clone https://github.com/radoslavraynov/macaulay2-jupyter-kernel.git
$ cd macaulay2-jupyter-kernel
$ pip3 install .
$ python3 -m m2_kernel.install

Docker

A docker image packing v0.6.7-beta and Macaulay2 version 1.13 is available as rzlatev/m2jk. To run locally, you need to map port 8890.

$ docker run -p 8890:8890 rzlatev/m2jk &

Running the notebook

Once the installation is complete, you need to start (or restart) Jupyter by

$ jupyter notebook &

This shoud fire up a browser for you. If not, copy the output URL into one. Then select File ⇨ New Notebook ⇨ M2 from the main menu.

License

This software is not part of Macaulay2 and is released under the MIT License.

macaulay2-jupyter-kernel's People

Contributors

Stargazers

Watchers

Forkers

mwhybrow92 8d1h edgarcosta ahoefn coneill-math

macaulay2-jupyter-kernel's Issues

texmacs mode output

The output didn't save as tex or pdf when we changed mode as texmacs with --%mode=texmacs.

test issue

another test

Fix reoccurring issue when reading a long input line

The code below works fine but outputs garbage due to M2 breaking the input line (when outputting it back). Don't just break the line by printWidth... Instead, use better sentinels.

--%mode=original
bar = (op1,op2) -> ({op1,op2,op1|op2} / (op->print(degree op, isHomogeneous(op), degrees op)); null)

i1 : mode(false); << "[magic succeeded] mode = original";
[magic succeeded] mode = original
(op), degrees op)); null){op1,op2,op1|op2} / (op->print(degree op, isHomogeneous 

o2 = bar

o2 : FunctionClosure

Use subprocess instead of pexpect

There's no need for pexepect aside from the timeout guard. In fact, we can send shorter sentinel messages using subprocess and escape characters.

M2 + WSL + Jupyter Notebook = kernel timeout

Hello!

I installed Ubuntu 18.04 on Windows 10 using the Windows Subsystem for Linux, then I installed M2 1.15 in Ubuntu. Next I installed Jupyter Notebook and the M2 Jupyter kernel (via pip3). I can run M2 within a Jupyter notebook from a browser (Chrome and Edge). I ran the following:

A=QQ[x_(1,1)..x_(4,6)]
G=genericMatrix(A,6,4)
I=minors(3,G);
betti res I

The first three lines ran okay, but the last one produced the following message:

[KERNEL ENFORCED TIMEOUT]

after about 5 seconds. If I run the same commands in a terminal, the last line returns a Betti table after about 8 seconds.

Any idea how I might try to fix this?

Thank you,
Fred.

install script's --prefix option isn't used by install_nbextension

Currently, python3 -m m2_kernel.install --prefix /foo will only install the kernel spec under /foo and not the nbextension.

I'm getting around this by using the following patch, but it would only work if it's known that notebook is installed under /usr.

--- a/m2_kernel/install.py
+++ b/m2_kernel/install.py
@@ -35,7 +35,8 @@
 
     print("Installing nbextension for syntax highlighting ...")
     install_nbextension('{}/m2-mode'.format(assets_dir),
-            nbextensions_dir='{}/static/components/codemirror/mode/'.format(notebook_dir[0]),
+            nbextensions_dir='{}/static/components/codemirror/mode/'.format(
+                notebook_dir[0].replace('/usr', prefix)),
             destination='macaulay2', overwrite=True, symlink=False)

"KERNEL ENFORCED TIMEOUT" message

Do you know how to resolve the following message?

o8 = [KERNEL ENFORCED TIMEOUT]

It only happens in the JupyterLab/Notebook, but not in the terminal.