An open API service indexing awesome lists of open source software.

https://github.com/iwpnd/toponym

build grammatical cases for words in slavic languages
https://github.com/iwpnd/toponym

grammar-parser grammatical-cases nlp python slavic-grammar slavic-languages toponym

Last synced: over 1 year ago
JSON representation

build grammatical cases for words in slavic languages

Awesome Lists containing this project

README

          



Build Status


Coverage

# Toponym

Build grammatical cases for words in Slavic languages from pre-defined recipes.

**documentation**: [https://toponym.iwpnd.pw/](https://toponym.iwpnd.pw/)

## Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.

### Installing

for usage:
```
pip install toponym
```

for development:
```
git clone https://github.com/iwpnd/toponym.git
pip install flit
flit install toponym --symlink
```

# Description

## Problem
In Slavic languages a word can change, depending on how and where it is used within a sentence. The city Moscow (`Москва`) changes to `Москве` when used prepositional.
So when you want to eg. know if:

```python
"Москва" in "В Москве с начала года отремонтировали 3 тысячи подъездов"

>> False
```

## Solution
This is where Toponym comes in. Utilizing pre-defined recipes it naively creates grammatical cases depending on the ending of the input word that the user wants to create Toponyms from. The recipe looks as follows:

### Recipe
```python
recipe = {
"а": { # ending of the input-word
"nominative": [[""], 0],
"genitive": [ # case that we need
["ы","и"], # ending of the output-word
1 # chars to be deleted, before ending of output is added
],
"dative": [["е"], 1],
"accusative": [["у"], 1],
"instrumental": [...]
}
```

If multiple endings are given, multiple toponyms with that ending will be created. Some of those created toponyms do not make sense, or are not used in the wild. If you have an idea about how to remove those that are unreal please contact me.

With the built toponyms for you can now check:

```python
from toponym.recipes import Recipes
from toponym.toponym import Toponym

recipes_russian = Recipes()
recipes_russian.load_from_language(language='russian')

city = "Москва"

t = Toponym(input_word=city, recipes=recipes_russian)
t.build()

print(t.list_toponyms())
>> ['Москвой', 'Москвы', 'Москви', 'Москве', 'Москву', 'Москва']

any([word in "В Москве с начала года отремонтировали 3 тысячи подъездов" for word in tn.list_toponyms()])
>> True
```

### supported languages:

```
full name iso code
croatian hr
russian ru
ukrainian uk
romanian ro
latvian lv
hungarian hu
greek el
polish pl
```

## Running the tests

```
pytest toponym/tests/
```