.webp "Arena-Hard-Auto : Advancing LLM Evaluation With Style Control Integration")
.webp "How To Install/Run – Streamlining OSINT Workflows For Enhanced Capabilities")
Homoglyphs is used to get similar letters, convert to ASCII, detect possible languages and UTF-8 group. Also can say python library for getting it and converting to ASCII.
Features
It’s smarter version of confusable_homoglyphs:
- Autodect or manual choosing category (aliases from ISO 15924).
- Auto or manual load only needed alphabets in memory.
- Converting to ASCII.
- More configurable.
- More stable.
Also Read Whatsapp_Automation : Collection Of APIs Interact With WhatsApp Running In An Android Emulator
Installation
sudo pip install homoglyphs
Usage
Importing:
import homoglyphs as hg
Languages
#detect
hg.Languages.detect('w')
# {'pl', 'da', 'nl', 'fi', 'cz', 'sr', 'pt', 'it', 'en', 'es', 'sk', 'de', 'fr', 'ro'}
hg.Languages.detect('т')
# {'mk', 'ru', 'be', 'bg', 'sr'}
hg.Languages.detect('.')
# set()
# get alphabet for languages
hg.Languages.get_alphabet(['ru'])
# {'в', 'Ё', 'К', 'Т', ..., 'Р', 'З', 'Э'}
Categories
Categories — (aliases from ISO 15924).
#detect
hg.Categories.detect('w')
# 'LATIN'
hg.Categories.detect('т')
# 'CYRILLIC'
hg.Categories.detect('.')
# 'COMMON'
# get alphabet for categories
hg.Categories.get_alphabet(['CYRILLIC'])
# {'ӗ', 'Ԍ', 'Ґ', 'Я', ..., 'Э', 'ԕ', 'ӻ'}
Homoglyphs
Get it:
# get homoglyphs (latin alphabet initialized by default)
hg.Homoglyphs().get_combinations('q')
# ['q', '𝐪', '𝑞', '𝒒', '𝓆', '𝓺', '𝔮', '𝕢', '𝖖', '𝗊', '𝗾', '𝘲', '𝙦', '𝚚']
Alphabet loading:
# load alphabet on init by categories
homoglyphs = hg.Homoglyphs(categories=('LATIN', 'COMMON', 'CYRILLIC')) # alphabet loaded here
homoglyphs.get_combinations('гы')
# ['rы', 'гы', 'ꭇы', 'ꭈы', '𝐫ы', '𝑟ы', '𝒓ы', '𝓇ы', '𝓻ы', '𝔯ы', '𝕣ы', '𝖗ы', '𝗋ы', '𝗿ы', '𝘳ы', '𝙧ы', '𝚛ы']
# load alphabet on init by languages
homoglyphs = hg.Homoglyphs(languages={'ru', 'en'}) # alphabet will be loaded here
homoglyphs.get_combinations('гы')
# ['rы', 'гы']
# manual set alphabet on init # eng rus
homoglyphs = hg.Homoglyphs(alphabet='abc абс')
homoglyphs.get_combinations('с')
# ['c', 'с']
# load alphabet on demand
homoglyphs = hg.Homoglyphs(languages={'en'}, strategy=hg.STRATEGY_LOAD)
# ^ alphabet will be loaded here for "en" language
homoglyphs.get_combinations('гы')
# ^ alphabet will be loaded here for "ru" language
# ['rы', 'гы']
You can combine categories, languages, alphabet and any strategies as you want.
Converting glyphs to ASCII chars
homoglyphs = hg.Homoglyphs(languages={'en'}, strategy=hg.STRATEGY_LOAD)
# convert
homoglyphs.to_ascii('тест')
# ['tect']
homoglyphs.to_ascii('ХР123.') # this is cyrillic "х" and "р"
# ['XP123.', 'XPI23.', 'XPl23.']
# string with chars which can't be converted by default will be ignored
homoglyphs.to_ascii('лол')
# []
# you can set strategy for removing not converted non-ASCII chars from result
homoglyphs = hg.Homoglyphs(
languages={'en'},
strategy=hg.STRATEGY_LOAD,
ascii_strategy=hg.STRATEGY_REMOVE,
)
homoglyphs.to_ascii('лол')
# ['o']
