plywood-gallery - GitLite

KO | EN

gitlite — search

#agent #database #kubernetes #api #rust #nextjs #mysql #awesome #html #video

plywood-gallery

(★ 27)

Gallery like documentation, where code snippets are represented by images.

파일 탐색기

ZIP 다운로드

- - pypi-publish.yml
  - python-package.yml
- settings.json
- hello.json
- x.ipynb
- code_animation_idea.html
- idea_plywood_notebook.html
- idea_plywood_notebook.ipynb
- make_release.md
- webserver_ideas.ipynb
- - template_index.html
- - gallery_config.yaml
  - gallery_notebook.ipynb
  - html_build.ipynb
  - icon.png
  - preview_image.png
- __init__.py
- browser_utils.py
- generate_html.py
- plywood_tools.py
- quickstart.py
- test_ChapterConfig.py
- test_quickstart.py
.gitignore
changelog.md
howto.md
LICENSE
pyproject.toml
README.md

# CDN으로 사용하기

jsDelivr

jsDelivr는 공개 GitHub 리포지토리를 별도 설정 없이 CDN으로 즉시 서빙합니다. 버전과 파일을 고르면 웹페이지에 바로 붙일 수 있는 링크와 예시 코드가 만들어집니다.

버전

파일

링크

예시

명령어 용어집

이 문서에서 사용된 명령어를 모아봤습니다. 낯선 명령어가 있다면 펼쳐서 확인해보세요.

🔍

`black`

설명 보기 ▼

Python 자동 코드 formatter.
관련 항목: `ruff`.
더 많은 정보: <https://black.readthedocs.io/en/stable/usage_and_configuration/the_basics.html>.

black {{path/to/file_or_directory}}

파일 또는 전체 디렉토리의 자동 포맷:

black {{[-c|--code]}} "{{code}}"

전달된 코드를 문자열로 포맷:

black --check {{path/to/file_or_directory}}

파일을 다시 쓰지 않고 상태 반환:

🔍

`pip install`

설명 보기 ▼

Install Python packages.

pip install {{package1 package2 ...}}

Install one or more packages:

pip install {{package1 package2 ...}} {{[-U|--upgrade]}}

Upgrade all specified packages to the latest version, installing any that are not already present:

pip install {{package}}=={{version}}

Install a specific version of a package:

# 프로젝트 배지

PyPI version 바로가기

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

Plywood Gallery - a new framework to generate python documentation via notebooks.  
The special feature: code snippets are represented by images.  
And it’s possible to show code variations with example branching.

https://user-images.githubusercontent.com/44469195/197733527-cb7daf87-8770-44dc-8f4c-4879145813cf.mov

**NOTE: Still work in progress, there might be breaking changes**

# Installation

* Install with  
    ```
    pip install plywood-gallery
    ```

* Open a python console and type:
    ```py
    >>> from plywood_gallery import quickstart, generate_html
    >>> quickstart()
    >>> # now, set project details in gallery_config.yaml e.g. project name, description, etc. 
    >>> generate_html(yaml_file="gallery_config.yaml", html_file="index.html")
    ```
All needed files are now generated.  
* Open `index.html`.  
This can be done with `from plywood_gallery import open_webpage; open_webpage()`  
 or alternatively in VS Code by clicking 'Live Preview: Show Preview'.  
* Add your code to `gallery.ipynb`.

* Start with the gallery creation!

# Examples

* https://github.com/kolibril13/plywood-gallery-functions
* https://github.com/kolibril13/plywood-gallery-napari
* https://github.com/kolibril13/mobject-gallery
* https://github.com/kolibril13/plywood-gallery-pathlib
* https://github.com/kolibril13/plywood-gallery-f-string
* https://github.com/kolibril13/plywood-gallery-matplotlib-ImageGrid
* https://github.com/kolibril13/plywood-gallery-seaborn
* https://github.com/kolibril13/plywood-gallery-matplotlib-hatch
* https://github.com/kolibril13/plywood-gallery-matplotlib-scalebar
* https://github.com/kolibril13/plywood-gallery-minimal-example
* https://github.com/kolibril13/plywood-gallery-2D-array
* https://github.com/kolibril13/plywood-gallery-3D-array
  
# Gallery creation

First some background infos: The plywood gallery will turn your Jupyter cell output images into a gallery. Thereby, one can distinguish three cell types: "Header", "Normal" and "Dependent".
All information is phrased through the cell magic **`%%capture_png`** (more in section *User guide*).

Add the **`%%capture_png`** magic in Jupyter cells and the output images will automatically and in real-time be added to your gallery.   
There will be two phases of building the gallery, the *preparation phase* and the *deployment phase.*

## 1. Preparation phase

During the preparation phase, you can tinker around your cells and debug them and run them again and again.

Every single cell run will create a new entry in the plywood gallery, so running cell A and cell B  in the order AABA will also display the cells in the order AABA and no previous cells will be overwritten.  
This behavior can be used as a visual version control system to see how your plots evolve over the preparation process.  
*There is no "replace plywood gallery entry" yet, but might be implemented in the future, which would render AABA to AB.*

## 2. Publish phase

When you want to publish your gallery notebook e.g. on GitHub pages, then first clean all cells from the preparation phase with the ChapterConfig  
```py
from plywood_tools import ChapterConfig
ChapterConfig.clean_all() # cleans all cells from all chapters
#ChapterConfig.clean("Array 2D")  # cleans all cells from a specified chapter
```

And all previous cells will be deleted.
Next, restart the kernel and click "Run All".

**Why is the kernel restart necessary?**  
In the deployment phase, cells must be executed from top to button with a freshly started kernel to avoid unwanted artifacts from the preparation phase.

# User guide

**Chapters**

All gallery entries must be placed in chapters, chapter names are defined as follows:
```py
from plywood_gallery import ChapterConfig
ChapterConfig.set_chapter_name("Array 2D")
```
As soon as a new name is given, this new chapter will be added to the gallery below.
Chapters can be sorted by 
```py
from plywood_gallery import ChapterConfig
ChapterConfig.sort(["2D Array", "3D Arrays"])
```

**Cellmagic**  
One can run the `%%capture_png` with three arguments:
```
--path: (required)
    the path where the image will be saved to
--celltype: (optional)
    Cell can be of type 'Normal', 'Header', and 'Dependent'
--style: (optional)
    Add extra css style for the gallery entries
```
**--path**:  
A basic cell can look like this:
```py
%%capture_png --path MyExample.png
import matplotlib.pyplot as plt
plt.plot([1,2],[10,20])
```
This will generate the file `gallery_assets/2D_Arrays_001_MyExample.png`. Note that plywood makes sure that file names are unique and can be easily sorted in the file explorer.
Running this cell again would generate `gallery_assets/2D_Arrays_002_MyExample.png`.   
**--style**:  
Change the gallery entry's size, border width, or other CSS parameters.  
**--celltype**:    
This is a powerful feature that has to be used carefully.  
Here is its purpose:

* **Header:**  
Define imports and make definitions of functions

* **Normal:**  
Show usage of imports and functions that you've made in the header cell. WARNING: All normal cells should ALWAYS be independent of each other. Try to not change variables in these cells, and if you have to, make sure to not use them in other normal cells again. If you need to use them again, make sure to rest their value before adding another "Normal cell". This is crucial because a user should always be able to run the cell in the combination "Header+Normal"
* **Dependent**:  
The purpose of this cell is to show slight changes in function calls or post-processing of what happened in a "Normal cell". It should only have a few lines of code and only change as few parameters as possible. This makes it possible to see slight visual adjustments in the plots and easily associate the code changes to the visual changes.  
 "Dependent" Cells can be run in the combination "Header+Normal+Dependent", as well as "Header+Normal+Dependent+Dependent".

# Tips and Tricks

* Best you run the notebook in a Jupyter notebook instance and open the website separate (e.g. with a split-screen or second monitor) to keep track of the live updates.

* Multiple notebooks can be used for the same gallery

* When you want one line of your Jupyter cell to be executed, but not to be added to the gallery, simply add the comment `#NOT` at the end of this code line.

* In an open gallery, press Escape to clear the code block area.

# VS Code Extension
With the [Plywood-Gallery-For-VSCode](https://github.com/Rickaym/plywood-gallery-for-vscode/) Extension,  plywood galleries can be displayed in the VS Code side panel.  
 As soon as one gallery entry is clicked, the code is automatically copied to the last current cursor position in a python script or Jupyter notebook cell.  
*Add Screen recording Here*

# How it works
Jupyter writes the image to the path and adds information about image_path, style, cell type, and corresponding code into a JSON file.   
This JSON file is read by JavaScript and the elements are placed on an HTML page.

# Notebook formatting
Formatting with black can be done this way:
`black --python-cell-magics capture_png  docs/gallery.ipynb`

# Updating
Get the latest version of the plywood gallery with   
`pip install plywood_gallery --upgrade`
# Attribution

Thanks to @christopher-besch for discussing the project architecture with me.  
Thanks to [Marlene Knoche
](https://twitter.com/sanguinikde) for feedback on the design.  
Thanks to @krassowski for the help to capture the png output images.  
Thanks to @Rickaym for some inspiration about the JSON files and VSCode related things.  
Thanks to @behackl for providing the GitHub Actions auto-release script.