src.fairreckitlib.data.filter.filter_params

 1"""This module contains the parameter creation functions for filters.
 2
 3Functions:
 4
 5    create_params_numerical: TODO
 6    create_params_categorical: TODO
 7    create_params_count: TODO
 8
 9This program has been developed by students from the bachelor Computer Science at
10Utrecht University within the Software Project course.
11© Copyright Utrecht University (Department of Information and Computing Sciences)
12"""
13
14from ...core.config.config_parameters import ConfigParameters
15
16
17def create_params_numerical(**kwargs) -> ConfigParameters:
18    """Create the parameters of a numerical filter.
19
20    Keyword Args:
21        column_name(str): the name of the column that has numerical values.
22        dataset(Dataset): the dataset associated with the filter.
23        matrix_name(str): the matrix name related to the dataset.
24
25    Returns:
26        the configuration parameters of the numerical filter.
27    """
28    column_name = kwargs['column_name']
29    dataset = kwargs['dataset']
30    matrix_name = kwargs['matrix_name']
31
32    params = ConfigParameters()
33
34    for table_name, table_columns in dataset.get_available_columns(matrix_name).items():
35        if column_name in table_columns:
36            table = dataset.read_table(table_name, [column_name])
37            numerical_range = (int(table[column_name].min()), int(table[column_name].max()))
38            params.add_range('range', int, numerical_range, numerical_range)
39
40    return params
41
42
43def create_params_categorical(**kwargs) -> ConfigParameters:
44    """Create the parameters of a categorical filter.
45
46    Keyword Args:
47        column_name(str): the name of the column that has categorical values.
48        dataset(Dataset): the dataset associated with the filter.
49        matrix_name(str): the matrix name related to the dataset.
50
51    Returns:
52        the configuration parameters of the categorical filter.
53    """
54    column_name = kwargs['column_name']
55    dataset = kwargs['dataset']
56    matrix_name = kwargs['matrix_name']
57
58    params = ConfigParameters()
59
60    for table_name, table_columns in dataset.get_available_columns(matrix_name).items():
61        if column_name in table_columns:
62            table = dataset.read_table(table_name, [column_name])
63            categories = list(table[column_name].fillna('').unique())
64            categories = [None if len(c) == 0 else c for c in categories]
65            params.add_multi_option('values', categories, categories)
66
67    return params
68
69def create_params_count(**_) -> ConfigParameters:
70    """Create the parameters of a count filter.
71
72    Returns:
73        the configuration parameters of the count filter.
74    """
75    params = ConfigParameters()
76    params.add_number('threshold', int, 100, (1, 10000000000))
77    return params