def has_answer()

in dpr/data/qa_validation.py [0:0]

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

def has_answer(answers, text, tokenizer, match_type) -> bool:
    """Check if a document contains an answer string.
    If `match_type` is string, token matching is done between the text and answer.
    If `match_type` is regex, we search the whole text with the regex.
    """
    text = _normalize(text)

    if match_type == "string":
        # Answer is a list of possible strings
        text = tokenizer.tokenize(text).words(uncased=True)

        for single_answer in answers:
            single_answer = _normalize(single_answer)
            single_answer = tokenizer.tokenize(single_answer)
            single_answer = single_answer.words(uncased=True)

            for i in range(0, len(text) - len(single_answer) + 1):
                if single_answer == text[i : i + len(single_answer)]:
                    return True

    elif match_type == "regex":
        # Answer is a regex
        for single_answer in answers:
            single_answer = _normalize(single_answer)
            if regex_match(text, single_answer):
                return True
    return False