在spacy中，我如何确保一个特定的字符总是被认为是完整的标记？

import spacy

nlp = spacy.blank("en")

from spacy.tokens import Doc

def add_custom_rule(doc, token_match):
    with doc.retokenize() as retokenizer:
        for match in token_match:
            start, end, label = match
            span = doc[start:end]
            retokenizer.merge(span)

import re

def add_hyphen_rule(doc):
    hyphen_match = [(m.start(), m.end(), "HYPHEN") for m in re.finditer(r"\-", doc.text)]
    add_custom_rule(doc, hyphen_match)

# 调用示例
text = "I like to eat apples and oranges."
doc = nlp(text)
add_hyphen_rule(doc)