def preprocess()

in src/embeddings_streamlit.py [0:0]

• 8 lines of code
• 5 McCabe index (conditional complexity)

def preprocess(text):
    delimiters = r"[|–-]"
    split_text = re.split(delimiters, text)
    enough_info_first = len(split_text) > 0 and len(text) - len(split_text[0]) > 5
    is_potential_domain_info = len(split_text) > 1 and len(split_text[-1]) < 20
    if enough_info_first and is_potential_domain_info:
        return ' '.join(filter(None, split_text[:-1])).strip().lower()
    return text.lower()