fix(parsers): enrich EIS procurement identities from details

This commit is contained in:
2026-05-19 10:28:14 +02:00
parent 4ca2fa25d5
commit 1c7c7238be
2 changed files with 254 additions and 4 deletions

View File

@@ -0,0 +1,106 @@
from __future__ import annotations
from apps.parsers.clients.base import HTTPClientError
from apps.parsers.clients.common.structured import StructuredDataClient
from django.test import SimpleTestCase
class _FakeHTTPClient:
def __init__(self, responses: dict[str, bytes | Exception]) -> None:
self.responses = responses
self.downloaded_urls: list[str] = []
def download_file(self, endpoint: str, **_kwargs) -> bytes:
self.downloaded_urls.append(endpoint)
response = self.responses[endpoint]
if isinstance(response, Exception):
raise response
return response
class StructuredDataClientEisCardEnrichmentTest(SimpleTestCase):
def test_procurement_card_enriches_customer_identity_from_detail_page(self):
detail_url = (
"https://zakupki.gov.ru/epz/order/notice/ea20/view/"
"common-info.html?regNumber=0338100002026000022"
)
client = StructuredDataClient(source="procurements_44fz")
client._http_client = _FakeHTTPClient(
{
detail_url: """
<html>
<body>
<h2>Сведения о заказчике</h2>
<div>Полное наименование</div>
<div>ФЕДЕРАЛЬНОЕ ГБУ НАУКИ</div>
<div>ИНН / КПП</div>
<div>4101020011 / 410101001</div>
<div>ОГРН</div>
<div>1024101023456</div>
</body>
</html>
""".encode()
}
)
records = client.fetch_records(
content=f"""
<html>
<body>
<div class="search-registry-entry-block">
<a href="{detail_url}">№ 0338100002026000022</a>
<div>Работа комиссии</div>
<div>Объект закупки</div>
<div>Поставка оборудования</div>
<div>Заказчик</div>
<div>ФЕДЕРАЛЬНОЕ ГБУ НАУКИ</div>
<div>Начальная цена</div>
<div>1 000,00 ₽</div>
<div>Размещено</div>
<div>19.05.2026</div>
</div>
</body>
</html>
""".encode(),
file_name="results.html",
)
self.assertEqual(len(records), 1)
record = records[0]
self.assertEqual(record.inn, "4101020011")
self.assertEqual(record.ogrn, "1024101023456")
self.assertEqual(record.payload["kpp"], "410101001")
self.assertEqual(record.payload["detail_url"], detail_url)
self.assertEqual(client._http_client.downloaded_urls, [detail_url])
def test_procurement_card_keeps_record_when_detail_page_is_unavailable(self):
detail_url = (
"https://zakupki.gov.ru/epz/order/notice/ea20/view/"
"common-info.html?regNumber=0338100002026000022"
)
client = StructuredDataClient(source="procurements_44fz")
client._http_client = _FakeHTTPClient(
{detail_url: HTTPClientError("detail unavailable", url=detail_url)}
)
records = client.fetch_records(
content=f"""
<html>
<body>
<div class="search-registry-entry-block">
<a href="{detail_url}">№ 0338100002026000022</a>
<div>Объект закупки</div>
<div>Поставка оборудования</div>
<div>Заказчик</div>
<div>ФЕДЕРАЛЬНОЕ ГБУ НАУКИ</div>
</div>
</body>
</html>
""".encode(),
file_name="results.html",
)
self.assertEqual(len(records), 1)
self.assertEqual(records[0].inn, "")
self.assertEqual(records[0].ogrn, "")
self.assertEqual(records[0].organisation_name, "ФЕДЕРАЛЬНОЕ ГБУ НАУКИ")