Russian Declension 0.1

Extremely simple script, to help learn declensions of Russian Language

Inspirationn

Russian declension are relatively difficult, evefor a native greek speaker, the language of whom contains declensions. Repetition is the key, and this script is used by me to solidify basic declension. I have been using a similar android app but the letters on the phone keyboard became to small :-)

Usage

python3 declensions.py (declensions.csv must be on the same dir)

How

The script reads a csv file (here named declensions.csv) and then asks the user the correct form of a specific word. A full list of all declension is printed if the answer is not correct. Finally, a list of all answers is output on the screen.

Acknowledgement

The words used were handpicked from the russian edition of wikiktionary and wikipedia. The script was inspired by an android app named Russian Declension (from Marinokiba) ( I have no relation with the creators of the app) A LLM was used to make the script code look better

Disclaimer

I have no official training and limited knowledge of the Russian Language, and therefor, no guarantees on the correctness of the grammar - spelling in the file.

ToDo List

1. Use a trakcer of the errors, to focus learning on them
2. Fix bugs on the word comparison. A strip_accent function (copied from a [stack_overflow] (https://stackoverflow.com/questions/517923/what-is-the-best-way-to-remove-accents-normalize-in-a-python-unicode-string) answer is used to remove accents, but the problem remains that there are many unicode representations for specific letters (i.e. ё ) making comparisons difficult. A simple solution could be to parse the cvs (maybe with an iconv _to_ASCII -> iconv _to_UTF)
3. Generalise to other languages -relatively easy, change the csv file and add-remove cases
4. Use directly wikiktionary data for declension patterns. Most tools for extracting wiktionary data do not support declension boxes yet, and too lazy to parse html (wiktionary uses a [rather interesting way of generating inflections] (https://www.dfki.de/web/forschung/projekte-publikationen/publikation/10149] and the declensions are generated real time on the fly)
5. Combine a keyboard library (i like [blessed] (https://pypi.org/project/blessed/) ) with a transileration package to remove the necessity of an OS keyboard layout of the target language (like a custom keyboard layout for the application)
6. Expand functionality to combination of nouns and adjectives
7. Use wikiktionary to include example sentences for better memory retention of words
8. Use a simple GUI