def _find_approximate_matching_sequence()

in utils/span_heuristic.py [0:0]

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

def _find_approximate_matching_sequence(context: str, target: str) -> Tuple[str, float]:
    """Find some substring in the context which closely matches the target, returning this substring with a score."""
    if target in context:
        return target, 1.0

    target_length = len(target.split())
    max_sim_val = 0
    max_sim_string = ''
    seq_matcher = SM()
    seq_matcher.set_seq2(target)
    for ngram in ngrams(context.split(), target_length + int(0.05 * target_length)):
        candidate_ngram = ' '.join(ngram)
        seq_matcher.set_seq1(candidate_ngram)
        similarity = seq_matcher.quick_ratio()
        if similarity > max_sim_val:
            max_sim_val = similarity
            max_sim_string = candidate_ngram
        if similarity == 1.0:
            # early exiting
            break

    return max_sim_string, max_sim_val