Note: There is an ongoing discussion about whether adding sense, form and syntactic dependency is also needed for a tool like this
Simple CLI-tool to combine lexemes easily on Wikidata
python = ">=3.10,<3.13"
On systems with a lower python version than 3.10, try updating your python installation first.
Clone the git repo:
$ git clone https://github.com/dpriskorn/lexeme-combinator.git && cd lexeme-combinator
We use pip and poetry to set everything up.
$ pip install poetry && poetry install
Copy config.py.sample -> config.py
$ cp config.py.sample config.py
Then enter your botpassword credentials in config.py using any text editor. E.g. user_name: "test" and bot_password: "q62noap7251t8o3nwgqov0c0h8gvqt20"
Run:
poetry run python main.py
This will promp you for each lexeme where 2 parts was successfully found.
It defaults to fetching 10 lexemes with a minimum length from the working language specified in the config.py. It has been tested with Danish and Swedish
Big thanks to Nikki and Mahir for helping with the SPARQL query that makes this possible and Finn Nielsen and Jan Ainali for feedback on the program and documentation.
GPLv3+