Check differences with degree tokenization changes

check.py

# script to test degree tokenization related changes.
# https://github.com/explosion/spaCy/pull/9155
import spacy

langs = ("af am ar az bg bn ca cs da de el en es et eu fa fi fr ga grc gu he hi "
         "hr hu hy id is it ja kn ko ky lb lij lt lv mk ml mr nb ne nl pl pt ro "
         "ru sa si sk sl sq sr sv ta te th ti tl tn tr tt uk ur vi xx yo zh").split()
check = ("°c °f °k °C °F °K °c. °f. °k. °C. °F. °K. 1°c 1°f 1°k 1°C 1°F 1°K 1°c. "
        "1°f. 1°k. 1°C. 1°F. 1°K.").split()

for lang in langs:
    try:
        nlp = spacy.blank(lang)
        for ex in check:
            toks = "|".join([tok.text for tok in nlp(ex)])
            print(lang.upper(), toks)
    except:
        # this can happen if we don't have a dependency
        print("Skipping {lang}")