This repository aims to convert the German nouns present in BEOLINGUS, TU Chemnitz's website downloadable files to a .json file, in order to organize them by gender and respective article.
I'm doing this because later I have plans to create an API to find out a German noun's gender in the simplest way possible, and the generated .json file of this project will be the basis of my database.
The input files of this experiment are inside the input-files directory. There de-en.txt is the original file containing thousands of lines, and the de-en-sample.txt is a short version of it, in order to make my tests without having to wait for minutes.
It reads the input file and uses regex to find the nouns and their respective genders (based on {m}, {f} and {n} marks).
The regex is not 100% accurate, but for my personal purposes it works just fine. Feel fee to improve it!
I didn't search for any plural words because they are all feminine in German
It's very simple. You will simply run node src passing the following arguments:
- -i: the input file path
- -o: the output file path
You don't even have to run yarn install, since this project doesn't have any dependencies.
I've put an empty formatted-nouns.json file in the output/ directory, but feel free to specify your input and output file paths.
This is my first project using the GPL-03 license. I hope I didn't do anything wrong, but you can contact me in my email in any case.
Thanks for visiting this repository, and I hope it'll be useful to you somehow. See ya!