██ ██ ██ ██ ██████ ██ ██████ ██ ██ ███████ ██████ ██ ██ ██ ██ ██ ██ ██ ██ ██ ██ ██ ██ ██ ██ ██ ██ ██ ████ ██ ██ ██ ███ ██ ██ ██ ██ ██ █████ ██████ ████ ██ ██ ██ ██ ██ ██ ██ ▄▄ ██ ██ ██ ██ ██ ██ ██ ██ ██████ ██████ ██ ██████ ██████ ███████ ██ ██ ██ ▀▀
YugiQuery is a Python package to query and display Yu-Gi-Oh! data extracted from the yugipedia database. It is entirely built on Jupyter notebooks and Git. The notebooks are rendered as HTML reports and can be displayed as an "always up to date" static web page by laveraging on GitHub pages. The raw data is kept as CSV files with timestamps and changelogs for a thorough record of the game's history. Every operation is recorded on git with a descriptive commit message.
Below are listed all the available reports and their execution timestamps.
| Report | Last execution |
|---|---|
| Bandai | 09/11/2024 10:23 UTC |
| Cards | 09/11/2024 10:27 UTC |
| Rush | 09/11/2024 10:29 UTC |
| Sets | 09/11/2024 10:37 UTC |
| Speed | 09/11/2024 10:38 UTC |
| Timeline | 09/11/2024 10:43 UTC |
The full YugiQuery flow was last executed at 09/11/2024 10:43 UTC
The full YugiQuery workflow can be run directly with
>> yugiquery
All commands and options can be displayed with the command
>> yugiquery -h
Any Jupyter notebook in the notebooks directory will be assumed to be a report and will be executed and saved as HTML in the reports directory. The index.md and README.md files will be updated, using their respective template files in the assets directory, to include a table with all the reports available and their timestamps. The source notebooks will then be cleared of their outputs and all changes will be commited to Git.
Template notebooks are included in the notebooks/templates folder.
Further user input can be made through the command
>> yugiquery run
To use the optional Discord bot, run
>> yugiquery bot SUBCLASS
Where SUBCLASS can be either telegram or discord.
Both the yugiquery.py and bot.py modules within the yugiquery package accept command line arguments. Using -h or --help will print an useful help message listing the parameters that can be passed and their usage.
YugiQuery is meant to be user friendly to users without much coding experience. It can be used "as is" from the repository, or installed via pip.
The post_install.py script in the assets directory has options to install:
1 - A nbconvert template which adds dynamic light and dark modes to the exported html report. This is the default template used by YugiQuery.
2 - The TQDM fork needed to run the discord bot subclass.
3 - A jupyter kernel for the current envyronment.
It can be run from the main yugiquery CLI with the command
>> yugiquery install
Further details can be found the documentation.
The repository is structured such that its root contains the web page index files, while the package files are kept in the yugiquery directory. Any template files (markdown, nbconvert, notebook, etc) and files used for reference such as dictionaries are kept in the assets directory. The raw data used by the reports is saved in the data directory. The Read The Docs source files are kept in the docs directory. The HTML reports are generated from the notebooks in the notebooks directory and saved in the reports directory. Below is an skeleton of the directory structure.
yugiquery/
├─ assets/
│ ├─ json/
│ │ ├─ colors.json
│ │ ├─ dates.json
│ │ ├─ headers.json
│ │ ├─ rarities.json
│ │ └─ regions.json
│ ├─ markdown/
│ │ ├─ footer.md
│ │ ├─ header.md
│ │ ├─ index.md
│ │ └─ REAMDME.md
│ ├─ scripts/
│ │ ├─ git_filters.sh
│ │ ├─ post_install.py
│ │ └─ unlock_git.sh
│ ├─ nbconvert/
| | └─ labdynamic/
| │ ├─ conf.json
| │ ├─ dynamic.css
| │ └─ index.html.j2
│ ├─ templates/
│ │ ├─ Collection.ipynb
│ │ └─ Template.ipynb
│ ├─ Gateway.html
│ └─ secrets.env
├─ data/
│ ├─ benchmark.json
│ ├─ report_data.bz2
│ └─ report_changelog.bz2
├─ docs/
│ ├─ Makefile
│ ├─ make.bat
│ ├─ conf.py
│ ├─ utils.rst
│ ├─ index.rst
│ ├─ bot.rst
│ └─ yugiquery.rst
├─ notebooks/
│ ├─ Bandai.ipynb
│ ├─ Cards.ipynb
│ ├─ Rush.ipynb
│ ├─ Sets.ipynb
│ ├─ Speed.ipynb
│ └─ Timeline.ipynb
├─ reports/
│ └─ report.html
├─ yugiquery/
│ ├─ utils
│ | ├─ __init__.py
│ | ├─ api.py
│ | ├─ dirs.py
│ | ├─ git.py
│ | ├─ helpers.py
│ | ├─ plot.py
│ | └─ progress_handler.py
│ ├─ bot
│ | ├─ __init__.py
│ | ├─ __main__.py
│ | ├─ base.py
│ | ├─ discord.py
│ | └─ telegram.py
│ ├─ __init__.py
│ ├─ __main__.py
│ ├─ metadata.py
│ └─ yugiquery.py
├─ _config.yml
├─ .devcontainer.json
├─ .readthedocs.yaml
├─ index.md
├─ LICENSE.md
├─ pyproject.toml
├─ README.md
└─ requirements.txt
Ideally, files in the assets directory should not be edited unless you know what you are doing. Files in the data directory are read and write files for the generation of the reports. The root of the repository should only contain files intended for the web page generation by GitHub pages or files that cannot be in another location.
The documentation can be found at ReadTheDocs
At present, TQDM relies on the deprecated disco-py package which won't build. To circunvent this problem until the officialTQDM release drops the disco-py dependency, we install TQDM from this fork, which uses pure REST API and/or discord.py.
Recent updates to IPython broke HALO in Jupyter notebooks. Until HALO conforms to the new IPython API, we install it from this fork.