Vladimir Kuzmenkov quicksilver0

After checking multiple tutorials I had to take pieces from each of the following to get this to work on my Win10 system:

Basically it uses the steps from the first article, but under the background path from the second article. Neither article on its own worked for me.

Steps:

	# Поддерживает русский язык, находит грамматические, стилистические и орфографические ошибки.
	pip install language-tool-python

	import language_tool_python

	# Создаём инструмент для русского языка
	tool = language_tool_python.LanguageTool('ru-RU')

	text = "Он пошол в школу. Я будеть дома."

	# find abbreviations and their positions (start and end indexes) in the text

	import re

	text = "США и ООН обсуждали события на G20 и 2FA. Также упоминался NASA."

	# Regular expression for abbreviations
	pattern = r'\b[A-ZА-ЯЁ0-9]{2,}\b'

	# Use re.finditer to get match objects with position info

	pip install rapidfuzz

	from rapidfuzz import process
	import re

	def suggest_names_fuzzy(text, name_database, score_cutoff=80):
	suggestions = []
	# Here re.finditer(r'\b\w+\b', text) change to the list of tokens found with spacy.
	# Example how to create a list of tokens from text with spacy: https://habr.com/ru/articles/531940/
	for match in re.finditer(r'\b\w+\b', text):

	import difflib

	#Finding identical matches in text with number of characters >=5. Min length should be adjusted for real case.
	def find_overlaps(text1, text2, min_length=5):
	matcher = difflib.SequenceMatcher(None, text1, text2)
	matches = []
	for match in matcher.get_matching_blocks():
	if match.size >= min_length:
	overlap = text1[match.a: match.a + match.size]
	matches.append(overlap)