def update_datapoint()

in python/json_duplicate_filter.py [0:0]

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

def update_datapoint(line, dataset):
    curr_object = dataset[line["URI"]]
    for new_question in line["Questions"]:
        new_question_text = get_full_question(new_question)
        if len(new_question_text) > 0:
            new_question_text = normalize_answer(new_question_text)
            if new_question_text in curr_object["Questions"].keys():
                for new_answer in new_question["Answers"]:
                    new_answer_text = get_full_answer(new_answer)
                    if len(new_answer_text) > 0:
                        new_answer_text = normalize_answer(new_answer_text)
                        if (
                            new_answer_text
                            not in curr_object["Questions"][new_question_text][
                                "Answers"
                            ]
                        ):
                            curr_object["Questions"][new_question_text]["Answers"][
                                new_answer_text
                            ] = new_answer
            else:
                condensed_question = copy.copy(new_question)
                condensed_question.pop("Answers")
                curr_object["Questions"][
                    normalize_answer(get_full_question(condensed_question))
                ] = condensed_question
                dataset[line["URI"]]["Questions"][
                    normalize_answer(get_full_question(condensed_question))
                ]["Answers"] = {}
                for answer in new_question["Answers"]:
                    curr_object["Questions"][
                        normalize_answer(get_full_question(condensed_question))
                    ]["Answers"][normalize_answer(get_full_answer(answer))] = answer
    return dataset