2973cc7966
M-2: README и docstring normalize_candidate — при отсутствии документа candidate.warnings хост обрабатывает сам. M-3: тесты сопоставления по дате (переставленные даты) и по порядку (дубль даты). M-4: «Косарев» заменён на «Иванов» в тестах и docstring. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BJKMLHdxMwSUeYcdkwqX4W
75 lines
3.5 KiB
Python
75 lines
3.5 KiB
Python
import pytest
|
||
|
||
from arm_statements.codes import is_code_cell, majority_section, parse_code, sheet_joins
|
||
from arm_statements.loader import CodeSection, Rules
|
||
from arm_statements.model import DocumentType
|
||
from arm_statements.text import fix_lookalikes, is_disguised_digit, normalize_caption, normalize_label
|
||
|
||
|
||
@pytest.mark.parametrize(
|
||
("raw", "code", "fixed"),
|
||
[
|
||
("1220 ,", "1220", False),
|
||
("| 140", "1140", True),
|
||
("I1О0", "1100", True),
|
||
("1110.0", "1110", False),
|
||
("12301", "12301", False),
|
||
("3450", "3450", False),
|
||
("2024", None, False),
|
||
("1050", None, False),
|
||
("7450", None, False),
|
||
("ИЧ,250", None, False),
|
||
("изо :", None, False),
|
||
("НИ", None, False),
|
||
("1500\n1700", None, False),
|
||
("2", None, False),
|
||
],
|
||
)
|
||
def test_parse_code(rules: Rules, raw: str, code: str | None, fixed: bool) -> None:
|
||
assert parse_code(raw, rules) == (code, fixed)
|
||
|
||
|
||
def test_multiline_code_cell(rules: Rules) -> None:
|
||
assert is_code_cell("1500\n1700", rules)
|
||
assert not is_code_cell("44 30\nI |", rules)
|
||
|
||
|
||
def test_lookalikes_only_near_digit(rules: Rules) -> None:
|
||
assert fix_lookalikes("2ОО5", rules.lookalikes) == "2005"
|
||
assert fix_lookalikes("2|0|2|4|", rules.lookalikes).split() == ["2", "0", "2", "4"]
|
||
assert fix_lookalikes("Основные средства", rules.lookalikes) == "Основные средства"
|
||
assert fix_lookalikes("З", rules.lookalikes) == "З"
|
||
assert fix_lookalikes("З", rules.lookalikes, anywhere=True) == "3"
|
||
|
||
|
||
def test_disguised_digit(rules: Rules) -> None:
|
||
assert is_disguised_digit("I", rules.lookalikes) and is_disguised_digit("| О", rules.lookalikes)
|
||
assert not is_disguised_digit("Иванов", rules.lookalikes)
|
||
assert not is_disguised_digit("(расшифровка подписи)", rules.lookalikes)
|
||
|
||
|
||
def test_caption_normalization() -> None:
|
||
assert normalize_caption("Код строки") == "код строки"
|
||
assert normalize_caption("Изменение, %") == "изменение %"
|
||
assert normalize_caption("Наименование\nпоказателя") == "наименование показателя"
|
||
|
||
|
||
def test_label_normalization_keeps_punctuation(rules: Rules) -> None:
|
||
assert normalize_label("Местонахождение (адрес)") == "местонахождение (адрес)"
|
||
assert normalize_label("по ОКОПФ / ОКФС") == "по окопф / окфс"
|
||
assert normalize_label("Форма 0710001 с.2") == "форма 0710001 с.2"
|
||
assert normalize_label("Отчёт о финансовых\nрезультатах") == "отчет о финансовых результатах"
|
||
assert normalize_label("задолжен—ность") == "задолжен ность"
|
||
assert normalize_label(None) == ""
|
||
for key, patterns in rules.headers.labels.items():
|
||
assert patterns, key
|
||
assert any(p.search(normalize_label("Местонахождение (адрес)")) for p in rules.headers.labels["address"])
|
||
assert any(p.search(normalize_label("по ОКОПФ / ОКФС")) for p in rules.headers.labels["okopf_okfs"])
|
||
|
||
|
||
def test_majority_section_and_join(rules: Rules) -> None:
|
||
assert majority_section(["1110", "1150", "2110"], rules) is CodeSection.BALANCE
|
||
assert sheet_joins(DocumentType.INCOME, ["2400"], 1, 3, rules)
|
||
assert not sheet_joins(DocumentType.BALANCE, ["1110", "1150", "1100"], 4, 3, rules)
|
||
assert not sheet_joins(DocumentType.BALANCE, ["2110"], 3, 3, rules)
|