def append_subfield_from_list_to_dict()

in anli/src/utils/list_dict_data_tool.py [0:0]

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

def append_subfield_from_list_to_dict(subf_list, d_dict, o_key_field_name, subfield_key_name,
                                      subfield_name='merged_field', check=False):
    # Often times, we will need to split the one data point to multiple items to be feeded into neural networks
    # and after we obtain the results we will need to map the results back to original data point with some keys.

    # This method is used for this purpose.
    # The method can be invoke multiple times, (in practice usually one batch per time.)
    """
    :param subf_list:               The forward list.
    :param d_dict:                  The dict that contain keys mapping to original data point.
    :param o_key_field_name:        The fieldname of original data point key. 'pid'
    :param subfield_key_name:       The fieldname of the sub item. 'fid'
    :param subfield_name:           The merge field name.       'merged_field'
    :param check:
    :return:
    """
    for key in d_dict.keys():
        d_dict[key][subfield_name] = dict()

    for item in subf_list:
        assert o_key_field_name in item
        assert subfield_key_name in item
        map_id = item[o_key_field_name]
        sub_filed_id = item[subfield_key_name]
        assert map_id in d_dict

        # if subfield_name not in d_dict[map_id]:
        #     d_dict[map_id][subfield_name] = dict()

        if sub_filed_id not in d_dict[map_id][subfield_name]:
            if check:
                assert item[o_key_field_name] == map_id
            d_dict[map_id][subfield_name][sub_filed_id] = item
        else:
            print("Duplicate forward item with key:", sub_filed_id)

    return d_dict