from gobbet import get_news, unicode_ranges

words, headlines, paragraphs = get_news("el") # ISO language code

# words is a Wordlist object; like Counter, but with a few more methods
words = (
    words.filter_popularity(2) # Only words which occur >=2 times
         .filter_length(5)     # ... of five letters or more ...
         .filter_unicodes(unicode_ranges["Greek"]) # .. in Greek script
)

# Want to find some kerning pair words?
bigrams = words.bigrams()
# bigrams['οώ'] === {'προώθησε', 'προώθηση', 'προώθησης', 'προώρων'}

# headlines and paragraphs are just lists, but...
from gobbet import filter_length, filter_unicodes

renderable = filter_unicodes(headlines, [font.getBestCmap().keys()])