1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
|
import pytest
import ssg.entities.common as common
def test_make_items_product_specific_not_overwriting_qualified_entries_with_unqualified():
a_dict_with_a_mix_of_qualified_and_unqualified_entries = {
"name": "Name: ?",
"name@foo": "Name: Foo",
}
qualifier = "foo"
with pytest.raises(ValueError):
common.make_items_product_specific(a_dict_with_a_mix_of_qualified_and_unqualified_entries,
qualifier, allow_overwrites=False)
def test_make_items_product_specific_for_unqualified_entry_order_dependent_behaviour_regression():
# With this order of elements in the dictionary a qualified element would be overwritten
# by an unqualified one if we don't take special precautions. This behaviour depends on
# the order in which elements are emitted. Since Python 3.6 it follows the order
# of initialization (guaranteed since 3.7). In Python 2.7 this particular set of values is kept
# in the required order, so the test behaviour is stable in all versions that we care about.
a_dict_with_a_mix_of_qualified_and_unqualified_entries = {
"name@foo": "Name: Foo",
"name": "Name: ?",
}
qualifier = "foo"
normal_dict = common.make_items_product_specific(
a_dict_with_a_mix_of_qualified_and_unqualified_entries, qualifier,
allow_overwrites=True)
assert normal_dict == {'name': 'Name: Foo'}
|
