feat(statements): колонка кодов, заголовок таблицы и колонки-периоды по содержимому
Refs #14 Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
This commit is contained in:
Andrey Epifancev
parent
6a3661c41b
commit
c3df9ff32c
4
api/src/parser/statements/__init__.py
Normal file
4
api/src/parser/statements/__init__.py
Normal file
@ -0,0 +1,4 @@
|
||||
"""Конвейер форм ББ/ОФР (Срез 2b): columns → detect → header → lines → checks → merge/statement.
|
||||
|
||||
Как и в ОСВ, структура ищется по содержимому: синонимы и соотношения — `rules/statement_*.yaml`.
|
||||
"""
|
||||
253
api/src/parser/statements/columns.py
Normal file
253
api/src/parser/statements/columns.py
Normal file
@ -0,0 +1,253 @@
|
||||
"""Заголовок таблицы формы, колонка кодов, колонка наименований, колонки-периоды (design D-4).
|
||||
|
||||
Колонка кодов — та, где не меньше `line_codes.min_share` непустых значений блока — коды строк форм; строка заголовка
|
||||
— с подписью «Код» и ячейками-периодами правее; заголовок в две строки склеивается; строка нумерации «1 2 3 4 5»
|
||||
пропускается. Ячейки заголовка без периода («Изменение, %») в колонки не попадают.
|
||||
"""
|
||||
|
||||
import re
|
||||
from dataclasses import dataclass
|
||||
from datetime import date
|
||||
|
||||
from src.parser.grid import Cell, Sheet
|
||||
from src.parser.rules import Rules
|
||||
|
||||
__all__ = [
|
||||
"LayoutNotFoundError",
|
||||
"PeriodColumn",
|
||||
"TableLayout",
|
||||
"cell_label",
|
||||
"find_code_column",
|
||||
"find_layout",
|
||||
"is_numbering_row",
|
||||
"line_code",
|
||||
]
|
||||
|
||||
_CODE = re.compile(r"^(\d{4,5})(\.0+)?$")
|
||||
_SHORT_TEXT = 3 # значения колонки «Пояснения»: 1–2 символа/цифры
|
||||
_MIN_NUMBERING_CELLS = 3
|
||||
NO_CODE_COLUMN = "колонка кодов не найдена"
|
||||
|
||||
|
||||
class LayoutNotFoundError(Exception):
|
||||
"""Таблица формы не найдена: `reason` — что именно (колонка кодов)."""
|
||||
|
||||
def __init__(self, reason: str) -> None:
|
||||
super().__init__(reason)
|
||||
self.reason = reason
|
||||
|
||||
|
||||
@dataclass(frozen=True)
|
||||
class PeriodColumn:
|
||||
col: int
|
||||
label: str # как в заголовке (переносы → пробел); "" у продолжения без заголовка
|
||||
period_date: date | None
|
||||
|
||||
|
||||
@dataclass(frozen=True)
|
||||
class TableLayout:
|
||||
header_row: int | None # None — заголовка таблицы нет (лист-продолжение)
|
||||
header_last_row: int | None # вторая строка заголовка или строка нумерации
|
||||
code_col: int
|
||||
name_col: int | None
|
||||
periods: tuple[PeriodColumn, ...]
|
||||
skipped_labels: tuple[str, ...] # заголовки правее кодов без периода
|
||||
first_data_row: int
|
||||
|
||||
|
||||
def cell_label(sheet: Sheet, row: int, col: int) -> str:
|
||||
return " ".join(sheet.text_at(row, col).split())
|
||||
|
||||
|
||||
def line_code(cell: Cell, rules: Rules) -> str | None:
|
||||
"""Код строки формы: 4–5 цифр, первые четыре — в диапазонах словаря; число `1110.0` читается как «1110»."""
|
||||
found = _CODE.match(cell.text.replace(" ", ""))
|
||||
if found is None:
|
||||
return None
|
||||
code = found.group(1)
|
||||
head = int(code[:4])
|
||||
ranges = rules.statement_headers.code_ranges.values()
|
||||
return code if any(low <= head <= high for low, high in ranges) else None
|
||||
|
||||
|
||||
def find_code_column(sheet: Sheet, rules: Rules, first_row: int, last_row: int) -> int | None:
|
||||
"""Колонка с наибольшим числом кодов, если кодов ≥ `min_codes` и их доля среди непустых ≥ `min_share`."""
|
||||
counts: dict[int, list[int]] = {}
|
||||
for row in range(first_row, last_row + 1):
|
||||
for col, cell in sheet.row_cells(row):
|
||||
tally = counts.setdefault(col, [0, 0])
|
||||
tally[1] += 1
|
||||
if line_code(cell, rules) is not None:
|
||||
tally[0] += 1
|
||||
headers = rules.statement_headers
|
||||
best: int | None = None
|
||||
for col in sorted(counts):
|
||||
codes, nonempty = counts[col]
|
||||
if codes < headers.min_codes or codes / nonempty < headers.code_min_share:
|
||||
continue
|
||||
if best is None or codes > counts[best][0]:
|
||||
best = col
|
||||
return best
|
||||
|
||||
|
||||
def is_numbering_row(sheet: Sheet, row: int, rules: Rules) -> bool:
|
||||
cells = sheet.row_cells(row)
|
||||
pattern = rules.statement_headers.column_numbering
|
||||
return len(cells) >= _MIN_NUMBERING_CELLS and all(pattern.search(cell.text) for _, cell in cells)
|
||||
|
||||
|
||||
def _first_code_row(sheet: Sheet, rules: Rules, code_col: int, first_row: int, last_row: int) -> int:
|
||||
for row in range(first_row, last_row + 1):
|
||||
if line_code(sheet.cell(row, code_col), rules) is not None:
|
||||
return row
|
||||
return last_row + 1
|
||||
|
||||
|
||||
def _is_period_label(label: str, rules: Rules, *, allow_bare_year: bool) -> bool:
|
||||
if rules.statement_periods.period_end(label) is None:
|
||||
return False
|
||||
patterns = (
|
||||
rules.statement_periods.on_date,
|
||||
rules.statement_periods.numeric_date,
|
||||
rules.statement_periods.for_months,
|
||||
rules.statement_periods.for_range,
|
||||
)
|
||||
return (
|
||||
allow_bare_year
|
||||
or not rules.statement_periods.for_year.search(label)
|
||||
or any(pattern.search(label) for pattern in patterns)
|
||||
)
|
||||
|
||||
|
||||
def _find_header_row(sheet: Sheet, rules: Rules, code_col: int, first_row: int, first_code_row: int) -> int | None:
|
||||
"""Последняя над первым кодом строка с подписью «Код» в колонке кодов; без подписи — последняя строка с двумя и
|
||||
более ячейками-периодами правее кодов либо с одной, но не голым годом (титульный блок НПЗ содержит «2025»)."""
|
||||
captioned = [
|
||||
row
|
||||
for row in range(first_row, first_code_row)
|
||||
if rules.statement_headers.is_code_caption(sheet.text_at(row, code_col))
|
||||
]
|
||||
if captioned:
|
||||
return captioned[-1]
|
||||
for row in range(first_code_row - 1, first_row - 1, -1):
|
||||
labels = [cell_label(sheet, row, col) for col, _ in sheet.row_cells(row) if col > code_col]
|
||||
strong = [label for label in labels if _is_period_label(label, rules, allow_bare_year=False)]
|
||||
weak = [label for label in labels if _is_period_label(label, rules, allow_bare_year=True)]
|
||||
if strong or len(weak) >= 2: # noqa: PLR2004 - две даты подряд — заголовок, одна «2025» — титул
|
||||
return row
|
||||
return None
|
||||
|
||||
|
||||
def _header_last_row(sheet: Sheet, rules: Rules, header_row: int, code_col: int, first_code_row: int) -> int:
|
||||
"""Вторая строка заголовка (продолжение подписей без кодов) и строка нумерации колонок."""
|
||||
last = header_row
|
||||
following = header_row + 1
|
||||
is_following_numbering = is_numbering_row(sheet, following, rules)
|
||||
is_following_empty = sheet.is_row_empty(following)
|
||||
if following < first_code_row and not is_following_numbering and not is_following_empty:
|
||||
# Проверяем, является ли следующая строка продолжением заголовка:
|
||||
# 1. Все непустые ячейки находятся правее кодов, ИЛИ
|
||||
# 2. В колонке кодов пусто, нет чисел, и есть ячейки правее кодов
|
||||
all_cells_right = all(col > code_col for col, _ in sheet.row_cells(following))
|
||||
code_col_empty = sheet.text_at(following, code_col) == ""
|
||||
no_numbers = not any(cell.kind == "number" for _, cell in sheet.row_cells(following))
|
||||
has_cells_right = any(col > code_col for col, _ in sheet.row_cells(following))
|
||||
|
||||
if all_cells_right or (code_col_empty and no_numbers and has_cells_right):
|
||||
last = following
|
||||
if last + 1 < first_code_row and is_numbering_row(sheet, last + 1, rules):
|
||||
last += 1
|
||||
return last
|
||||
|
||||
|
||||
def _joined_label(sheet: Sheet, header_row: int, last_row: int, col: int, rules: Rules) -> str:
|
||||
parts = [cell_label(sheet, header_row, col)]
|
||||
for row in range(header_row + 1, last_row + 1):
|
||||
if is_numbering_row(sheet, row, rules):
|
||||
continue
|
||||
origin_row, _ = sheet.merged_origin(row, col)
|
||||
if origin_row == header_row: # merged-продолжение той же ячейки
|
||||
continue
|
||||
parts.append(cell_label(sheet, row, col))
|
||||
return " ".join(part for part in parts if part)
|
||||
|
||||
|
||||
def _period_columns(
|
||||
sheet: Sheet, rules: Rules, header_row: int, header_last_row: int, code_col: int
|
||||
) -> tuple[tuple[PeriodColumn, ...], tuple[str, ...]]:
|
||||
periods: list[PeriodColumn] = []
|
||||
skipped: list[str] = []
|
||||
seen: set[int] = set()
|
||||
for row in range(header_row, header_last_row + 1):
|
||||
for col, _ in sheet.row_cells(row):
|
||||
origin_row, origin_col = sheet.merged_origin(row, col)
|
||||
if origin_col <= code_col or origin_col in seen or is_numbering_row(sheet, row, rules):
|
||||
continue
|
||||
seen.add(origin_col)
|
||||
label = _joined_label(sheet, origin_row, header_last_row, origin_col, rules)
|
||||
end = rules.statement_periods.period_end(label)
|
||||
if end is None and not label:
|
||||
continue
|
||||
if end is None:
|
||||
skipped.append(label)
|
||||
else:
|
||||
periods.append(PeriodColumn(origin_col, label, end))
|
||||
periods.sort(key=lambda column: column.col)
|
||||
return tuple(periods), tuple(skipped)
|
||||
|
||||
|
||||
def _value_columns(
|
||||
sheet: Sheet, rules: Rules, code_col: int, first_row: int, last_row: int
|
||||
) -> tuple[PeriodColumn, ...]:
|
||||
"""Продолжение без заголовка: колонки правее кодов, где в строках с кодами есть хотя бы одно число."""
|
||||
found: set[int] = set()
|
||||
for row in range(first_row, last_row + 1):
|
||||
if line_code(sheet.cell(row, code_col), rules) is None:
|
||||
continue
|
||||
found.update(col for col, cell in sheet.row_cells(row) if col > code_col and cell.kind == "number")
|
||||
return tuple(PeriodColumn(col, "", None) for col in sorted(found))
|
||||
|
||||
|
||||
def _name_column( # noqa: PLR0913, PLR0917 - сигнатура соответствует вызовам
|
||||
sheet: Sheet, rules: Rules, code_col: int, header_row: int | None, first_row: int, last_row: int
|
||||
) -> int | None:
|
||||
"""Колонка левее кодов с наибольшим числом текстовых значений в строках с кодами; «Пояснения» (по подписи или по
|
||||
коротким значениям) исключается; при равенстве — ближайшая к кодам."""
|
||||
texts: dict[int, int] = {}
|
||||
long_texts: dict[int, int] = {}
|
||||
for row in range(first_row, last_row + 1):
|
||||
if line_code(sheet.cell(row, code_col), rules) is None:
|
||||
continue
|
||||
for col, cell in sheet.row_cells(row):
|
||||
if col >= code_col or cell.kind != "text":
|
||||
continue
|
||||
texts[col] = texts.get(col, 0) + 1
|
||||
if len(cell.text) > _SHORT_TEXT:
|
||||
long_texts[col] = long_texts.get(col, 0) + 1
|
||||
candidates = [
|
||||
col
|
||||
for col in texts
|
||||
if long_texts.get(col, 0) > 0
|
||||
and not (
|
||||
header_row is not None and rules.statement_headers.is_explanations_caption(sheet.text_at(header_row, col))
|
||||
)
|
||||
]
|
||||
if not candidates:
|
||||
return None
|
||||
return max(candidates, key=lambda col: (texts[col], col))
|
||||
|
||||
|
||||
def find_layout(sheet: Sheet, rules: Rules, first_row: int, last_row: int) -> TableLayout:
|
||||
code_col = find_code_column(sheet, rules, first_row, last_row)
|
||||
if code_col is None:
|
||||
raise LayoutNotFoundError(NO_CODE_COLUMN)
|
||||
first_code_row = _first_code_row(sheet, rules, code_col, first_row, last_row)
|
||||
header_row = _find_header_row(sheet, rules, code_col, first_row, first_code_row)
|
||||
if header_row is None:
|
||||
periods = _value_columns(sheet, rules, code_col, first_row, last_row)
|
||||
name_col = _name_column(sheet, rules, code_col, None, first_row, last_row)
|
||||
return TableLayout(None, None, code_col, name_col, periods, (), first_row)
|
||||
header_last_row = _header_last_row(sheet, rules, header_row, code_col, first_code_row)
|
||||
periods, skipped = _period_columns(sheet, rules, header_row, header_last_row, code_col)
|
||||
name_col = _name_column(sheet, rules, code_col, header_row, header_last_row + 1, last_row)
|
||||
return TableLayout(header_row, header_last_row, code_col, name_col, periods, skipped, header_last_row + 1)
|
||||
0
api/tests/unit/parser/statements/__init__.py
Normal file
0
api/tests/unit/parser/statements/__init__.py
Normal file
241
api/tests/unit/parser/statements/test_columns.py
Normal file
241
api/tests/unit/parser/statements/test_columns.py
Normal file
@ -0,0 +1,241 @@
|
||||
"""Заголовок таблицы формы, колонка кодов, наименований и периодов (design D-4; spec «Заголовок таблицы, колонка кодов
|
||||
и колонки-периоды определяются по тексту»)."""
|
||||
|
||||
import datetime
|
||||
from decimal import Decimal
|
||||
|
||||
import pytest
|
||||
from gridbuilder import at, sheet_from_rows
|
||||
|
||||
from src.parser.grid import Cell
|
||||
from src.parser.rules import Rules, load_rules
|
||||
from src.parser.statements.columns import (
|
||||
LayoutNotFoundError,
|
||||
find_code_column,
|
||||
find_layout,
|
||||
line_code,
|
||||
)
|
||||
|
||||
|
||||
@pytest.fixture(scope="module")
|
||||
def rules() -> Rules:
|
||||
return load_rules()
|
||||
|
||||
|
||||
# Test constants
|
||||
CODE_COL_ONE_C = 4
|
||||
NAME_COL_ONE_C = 2
|
||||
CODE_COL_TDSHEET = 2
|
||||
NAME_COL_TDSHEET = 1
|
||||
HEADER_ROW_ONE_C = 4
|
||||
FIRST_CODE_ROW_ONE_C = 5
|
||||
PERIOD_COL_ONE_C_1 = 5
|
||||
PERIOD_COL_ONE_C_2 = 6
|
||||
HEADER_ROW_TDSHEET = 2
|
||||
HEADER_LAST_ROW_TDSHEET = 3
|
||||
FIRST_CODE_ROW_TDSHEET = 4
|
||||
HEADER_ROW_TWO_ROW = 1
|
||||
HEADER_LAST_ROW_TWO_ROW = 2
|
||||
FIRST_CODE_ROW_TWO_ROW = 3
|
||||
HEADER_ROW_NO_CAPTION = 1
|
||||
CODE_COL_NO_CAPTION = 2
|
||||
NAME_COL_NO_CAPTION = 1
|
||||
HEADER_ROW_CONTINUATION = None
|
||||
FIRST_DATA_ROW_CONTINUATION = 1
|
||||
CODE_COL_CONTINUATION = 6
|
||||
PERIOD_COL_CONT_1 = 7
|
||||
PERIOD_COL_CONT_2 = 8
|
||||
PERIOD_COL_CONT_3 = 9
|
||||
HEADER_ROW_BARE_YEAR = 2
|
||||
PERIOD_COL_BARE_YEAR_1 = 15
|
||||
PERIOD_COL_BARE_YEAR_2 = 20
|
||||
|
||||
|
||||
def test_line_code_from_text_and_numeric_cells(rules: Rules) -> None:
|
||||
assert line_code(Cell("1110", "text"), rules) == "1110"
|
||||
assert line_code(Cell(Decimal("1110"), "number"), rules) == "1110"
|
||||
assert line_code(Cell(Decimal("1110.0"), "number"), rules) == "1110" # Review Focus 1: xlrd даёт float
|
||||
assert line_code(Cell("11101", "text"), rules) == "11101"
|
||||
assert line_code(Cell("2110", "text"), rules) == "2110"
|
||||
assert line_code(Cell("5200", "text"), rules) == "5200"
|
||||
assert line_code(Cell("2023", "text"), rules) is None # год, не код
|
||||
assert line_code(Cell("0710001", "text"), rules) is None
|
||||
assert line_code(Cell("", "empty"), rules) is None
|
||||
|
||||
|
||||
# Test data: ONE_C_BALANCE sheet
|
||||
_HEADER_ROW_ONE_C = [
|
||||
"Пояснения",
|
||||
"Наименование показателя",
|
||||
None,
|
||||
"Код",
|
||||
"На 31 декабря 2024 г.",
|
||||
"На 31 декабря 2023 г.",
|
||||
"Изменение, %",
|
||||
]
|
||||
|
||||
ONE_C_BALANCE = [
|
||||
["Бухгалтерский баланс"],
|
||||
["на 31 декабря 2024 г."],
|
||||
[],
|
||||
_HEADER_ROW_ONE_C,
|
||||
[None, "АКТИВ"],
|
||||
[None, "Нематериальные активы", None, "1110", 5971264, 27405634, 12],
|
||||
[None, "Нематериальные активы в организации", None, "11101", 5852464, 27214614, 0],
|
||||
[None, None, None, None, 0, 0, 0],
|
||||
["1", "Основные средства", None, "1150", 2039077663, 1809528568, 7],
|
||||
[None, "Итого по разделу I", None, "1100", 2045048927, 1836934202, 7],
|
||||
]
|
||||
|
||||
|
||||
def test_code_column_by_share(rules: Rules) -> None:
|
||||
sheet = sheet_from_rows(ONE_C_BALANCE)
|
||||
|
||||
assert find_code_column(sheet, rules, 1, sheet.n_rows) == CODE_COL_ONE_C
|
||||
|
||||
|
||||
def test_code_column_needs_three_codes_and_sixty_percent(rules: Rules) -> None:
|
||||
sheet = sheet_from_rows([["x", "1110"], ["y", "abc"], ["z", "def"], ["w", "ghi"], ["v", "1150"]])
|
||||
|
||||
assert find_code_column(sheet, rules, 1, 5) is None
|
||||
|
||||
|
||||
def test_layout_one_c(rules: Rules) -> None:
|
||||
sheet = sheet_from_rows(ONE_C_BALANCE)
|
||||
|
||||
layout = find_layout(sheet, rules, 1, sheet.n_rows)
|
||||
|
||||
assert (layout.header_row, layout.header_last_row, layout.first_data_row) == (
|
||||
HEADER_ROW_ONE_C,
|
||||
HEADER_ROW_ONE_C,
|
||||
FIRST_CODE_ROW_ONE_C,
|
||||
)
|
||||
assert layout.code_col == CODE_COL_ONE_C and layout.name_col == NAME_COL_ONE_C
|
||||
expected_periods = [
|
||||
(PERIOD_COL_ONE_C_1, "На 31 декабря 2024 г.", datetime.date(2024, 12, 31)),
|
||||
(PERIOD_COL_ONE_C_2, "На 31 декабря 2023 г.", datetime.date(2023, 12, 31)),
|
||||
]
|
||||
assert [(column.col, column.label, column.period_date) for column in layout.periods] == expected_periods
|
||||
assert layout.skipped_labels == ("Изменение, %",)
|
||||
|
||||
|
||||
def test_layout_tdsheet_with_numbering_row_and_line_breaks(rules: Rules) -> None:
|
||||
sheet = sheet_from_rows(
|
||||
[
|
||||
["БУХГАЛТЕРСКИЙ БАЛАНС"],
|
||||
["Наименование показателя", "Коды", "За 9 месяцев\n2025 года", "На 31 декабря \n2024 года"],
|
||||
["1", "2", "3", "4"],
|
||||
["Нематериальные активы", "1110", 0, 0],
|
||||
["Основные средства", "1150", Decimal("898852.21795"), 966270],
|
||||
["ИТОГО по разделу I", "1100", Decimal("898852.21795"), 966270],
|
||||
]
|
||||
)
|
||||
|
||||
layout = find_layout(sheet, rules, 1, sheet.n_rows)
|
||||
|
||||
assert (layout.header_row, layout.header_last_row, layout.first_data_row) == (
|
||||
HEADER_ROW_TDSHEET,
|
||||
HEADER_LAST_ROW_TDSHEET,
|
||||
FIRST_CODE_ROW_TDSHEET,
|
||||
)
|
||||
assert layout.name_col == NAME_COL_TDSHEET and layout.code_col == CODE_COL_TDSHEET
|
||||
expected_labels = ["За 9 месяцев 2025 года", "На 31 декабря 2024 года"]
|
||||
assert [column.label for column in layout.periods] == expected_labels
|
||||
assert layout.periods[0].period_date == datetime.date(2025, 9, 30)
|
||||
|
||||
|
||||
def test_layout_two_row_header(rules: Rules) -> None:
|
||||
sheet = sheet_from_rows(
|
||||
[
|
||||
["Наименование показателя", "Код", "На 31 декабря", "На 31 декабря"],
|
||||
[None, None, "2024 г.", "2023 г."],
|
||||
["Запасы", "1210", 10, 20],
|
||||
["Дебиторская задолженность", "1230", 30, 40],
|
||||
["Итого по разделу II", "1200", 40, 60],
|
||||
]
|
||||
)
|
||||
|
||||
layout = find_layout(sheet, rules, 1, sheet.n_rows)
|
||||
|
||||
assert (layout.header_row, layout.header_last_row, layout.first_data_row) == (
|
||||
HEADER_ROW_TWO_ROW,
|
||||
HEADER_LAST_ROW_TWO_ROW,
|
||||
FIRST_CODE_ROW_TWO_ROW,
|
||||
)
|
||||
expected_dates = [datetime.date(2024, 12, 31), datetime.date(2023, 12, 31)]
|
||||
assert [column.period_date for column in layout.periods] == expected_dates
|
||||
|
||||
|
||||
def test_layout_without_code_caption(rules: Rules) -> None:
|
||||
sheet = sheet_from_rows(
|
||||
[
|
||||
["Показатель", None, "На 31.12.2024", "На 31.12.2023"],
|
||||
["Основные средства", "1150", 100, 90],
|
||||
["Запасы", "1210", 50, 40],
|
||||
["БАЛАНС", "1600", 150, 130],
|
||||
]
|
||||
)
|
||||
|
||||
layout = find_layout(sheet, rules, 1, sheet.n_rows)
|
||||
|
||||
assert layout.header_row == HEADER_ROW_NO_CAPTION
|
||||
assert layout.code_col == CODE_COL_NO_CAPTION
|
||||
assert layout.name_col == NAME_COL_NO_CAPTION
|
||||
expected_dates = [datetime.date(2024, 12, 31), datetime.date(2023, 12, 31)]
|
||||
assert [column.period_date for column in layout.periods] == expected_dates
|
||||
|
||||
|
||||
def test_layout_continuation_without_header(rules: Rules) -> None:
|
||||
sheet = sheet_from_rows(
|
||||
[
|
||||
[None, None, None, None, "в том числе:"],
|
||||
[None, None, None, None, "Биологические активы", "11901", 0, 0, 0],
|
||||
[None, None, None, None, "Итого по разделу I", "1100", 11853698344, 11570097340, 9391157985],
|
||||
[None, None, None, "II. Оборотные активы"],
|
||||
[None, None, None, "Запасы", None, "1210", 1001705104, 1426756617, 3393013092],
|
||||
]
|
||||
)
|
||||
|
||||
layout = find_layout(sheet, rules, 1, sheet.n_rows)
|
||||
|
||||
assert layout.header_row is HEADER_ROW_CONTINUATION
|
||||
assert layout.first_data_row == FIRST_DATA_ROW_CONTINUATION
|
||||
assert layout.code_col == CODE_COL_CONTINUATION
|
||||
expected_cols = [
|
||||
(PERIOD_COL_CONT_1, "", None),
|
||||
(PERIOD_COL_CONT_2, "", None),
|
||||
(PERIOD_COL_CONT_3, "", None),
|
||||
]
|
||||
assert [(column.col, column.label, column.period_date) for column in layout.periods] == expected_cols
|
||||
|
||||
|
||||
def test_explanations_column_is_not_name(rules: Rules) -> None:
|
||||
sheet = sheet_from_rows(ONE_C_BALANCE)
|
||||
|
||||
assert find_layout(sheet, rules, 1, sheet.n_rows).name_col == NAME_COL_ONE_C
|
||||
|
||||
|
||||
def test_no_code_column_raises(rules: Rules) -> None:
|
||||
sheet = sheet_from_rows([["Показатель", "Сумма"], ["Основные средства", 100], ["Запасы", 50]])
|
||||
|
||||
with pytest.raises(LayoutNotFoundError, match="колонка кодов не найдена"):
|
||||
find_layout(sheet, rules, 1, 3)
|
||||
|
||||
|
||||
def test_bare_year_in_title_block_is_not_a_header(rules: Rules) -> None:
|
||||
"""У НПЗ над таблицей стоит «Отчетная дата | 30 | 09 | 2025»: строка с подписью «Код» побеждает."""
|
||||
sheet = sheet_from_rows(
|
||||
[
|
||||
[*at({6: "Отчетная дата", 22: "30", 24: "09", 25: "2025"})],
|
||||
[*at({4: "Наименование показателя", 12: "Код", 15: "На 30 сентября 2025 г.", 20: "На 31 декабря 2024 г."})],
|
||||
[*at({4: "Гудвил", 12: "1105", 15: 0, 20: 0})],
|
||||
[*at({4: "Нематериальные активы", 12: "1110", 15: 586260, 20: 905786})],
|
||||
[*at({4: "Итого по разделу I", 12: "1100", 15: 586260, 20: 905786})],
|
||||
]
|
||||
)
|
||||
|
||||
layout = find_layout(sheet, rules, 1, sheet.n_rows)
|
||||
|
||||
period_cols = [PERIOD_COL_BARE_YEAR_1, PERIOD_COL_BARE_YEAR_2]
|
||||
assert layout.header_row == HEADER_ROW_BARE_YEAR
|
||||
assert [column.col for column in layout.periods] == period_cols
|
||||
Loading…
x
Reference in New Issue
Block a user