You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: README.md
+6-56Lines changed: 6 additions & 56 deletions
Original file line number
Diff line number
Diff line change
@@ -2,6 +2,10 @@
2
2
3
3
This is a [Jupyter](http://jupyter.org/) kernel for [Dyalog APL](https://www.dyalog.com/). The [wiki](https://github.com/Dyalog/dyalog-jupyter-kernel/wiki) has more details.
4
4
5
+
For complete installation instructions from scratch on Windows, see [WINDOWS.md](WINDOWS.md) in this repository.
6
+
7
+
If you're a kernel developer, see [DEVELOPMENT.md](DEVELOPMENT.md) in this repository.
8
+
5
9
## Do you have any pre-made notebook documents?
6
10
7
11
A collection of ready-made notebooks is available [here](https://github.com/Dyalog/dyalog-jupyter-notebooks).
@@ -20,7 +24,7 @@ This kernel is installable via Python's package installer, [pip](https://pip.pyp
20
24
In your terminal, run the follwing command:
21
25
22
26
```sh
23
-
pip install dyalog-jupyter-kernel[==2.0.0]
27
+
pip install dyalog-jupyter-kernel[==2.0.1]
24
28
```
25
29
26
30
After successfully installing the kernel module itself, you need to register it with your Jupyter installation, using the following terminal command:
@@ -43,58 +47,4 @@ Start Jupyter:
43
47
```sh
44
48
jupyter lab # or jupyter notebook
45
49
```
46
-
and you should now see Dyalog APL amongst your kernels. Click on the Dyalog APL icon.
47
-
48
-
## Build notes (kernel developer interest only)
49
-
50
-
Build release assets with
51
-
52
-
```sh
53
-
python setup.py sdist bdist_wheel
54
-
```
55
-
56
-
Built assets will end up in `dist/`, and you can instruct `pip` to install from file for testing:
Installing, operating and maintaining a Python installation can be a bit of a chore, especially if you're not a Python developer. If you're coming to this wanting only to use Jupyter for Dyalog APL, here are some step by step instructions for how to set up the whole tool chain (on Windows) from scratch.
4
+
5
+
If you're a Python developer, or if you're already using a "canned" Python distribution, like [conda](https://docs.conda.io/en/latest/), you probably know what you're doing and can skip this document.
6
+
7
+
The below instructions install Jupyter and the Dyalog kernel inside a Python "virtual environment". This means that you can keep your Jupyter work separate from any other Python work, current or future. This is usually a good idea when working with Python.
8
+
9
+
1. Download and install the [Dyalog interpreter](https://www.dyalog.com/download-zone.htm)
10
+
2. Download, and run the official [Python installer](https://www.python.org/downloads/).
11
+
3. Open a PowerShell terminal. Type the following steps into the terminal window.
12
+
4. Create a virtual Python environment with a name of your choosing, for example
13
+
```
14
+
python -m venv jupy312
15
+
```
16
+
5. Activate the virtual environment -- this is important:
17
+
```
18
+
jupy312\Scripts\activate
19
+
```
20
+
Your terminal prompt should change to have a `(jupy312)` prefix.
21
+
6. Upgrade `pip` to the latest version:
22
+
```
23
+
python.exe -m pip installl --upgrade pip
24
+
```
25
+
7. Install `jupyterlab`:
26
+
```
27
+
pip install jupyterlab
28
+
```
29
+
8. Install the Dyalog kernel Python module:
30
+
```
31
+
pip install dyalog-jupyter-kernel
32
+
```
33
+
9. Register the kernel with Jupyter:
34
+
```
35
+
python -m 'dyalog_kernel' install
36
+
```
37
+
10. Start jupyterlab:
38
+
```
39
+
jupyter lab
40
+
```
41
+
42
+
A browser window should open up with the Jupyter Lab application. Click the "Dyalog APL" symbol to open a new notebook document with the Dyalog kernel.
43
+
44
+
Once you're done working with Jupyter Lab, deactivate the Python environment:
45
+
46
+
```
47
+
deactivate
48
+
```
49
+
50
+
To start up Jupyter Lab again, you need to run `jupy312\Scripts\activate` and `jupyter lab` again.
0 commit comments