feature: improve admin directory and crawl progress

This commit is contained in:
Anton
2026-04-28 17:24:10 +03:00
parent 51d83d7062
commit 4bd5f26469
19 changed files with 1082 additions and 58 deletions

98
tests/test_admin_data.py Normal file
View File

@@ -0,0 +1,98 @@
from datetime import datetime, timezone
from app.models import CrawlRun, Employee
from app.services.admin_data import employee_display_payload, list_employees_page, run_payload, stats_payload
def test_employee_display_payload_extracts_common_fields(db_session):
employee = Employee(
profile_key="staff:person",
canonical_url="https://www.hse.ru/staff/person",
full_name="Person Name",
status="active",
first_seen_at=datetime.now(timezone.utc),
last_seen_at=datetime.now(timezone.utc),
current_data={
"positions": ["Professor"],
"hse_start_year": 2024,
"contacts": {"emails": ["person@hse.ru"], "phones": ["+79990000000"], "address": "Moscow"},
"sections": [
{"type": "publications", "publications": [{"title": "Paper"}]},
{"type": "courses_by_year", "courses": [{"title": "Course"}]},
],
},
)
payload = employee_display_payload(employee)
assert payload["positions_text"] == "Professor"
assert payload["email_text"] == "person@hse.ru"
assert payload["publications_count"] == 1
assert payload["courses_count"] == 1
def test_list_employees_page_filters_sorts_and_paginates(db_session):
db_session.add(
Employee(
profile_key="staff:b",
canonical_url="https://www.hse.ru/staff/b",
full_name="Beta",
status="dismissed",
first_seen_at=datetime.now(timezone.utc),
last_seen_at=datetime.now(timezone.utc),
current_data={"contacts": {"emails": []}},
)
)
db_session.add(
Employee(
profile_key="staff:a",
canonical_url="https://www.hse.ru/staff/a",
full_name="Alpha",
status="active",
first_seen_at=datetime.now(timezone.utc),
last_seen_at=datetime.now(timezone.utc),
current_data={"contacts": {"emails": ["alpha@hse.ru"]}},
)
)
db_session.commit()
page = list_employees_page(db_session, status="active", sort="full_name", direction="asc", limit=10)
assert page["total"] == 1
assert page["items"][0]["full_name"] == "Alpha"
def test_stats_payload_uses_latest_run_new_count(db_session):
db_session.add(
Employee(
profile_key="staff:a",
canonical_url="https://www.hse.ru/staff/a",
full_name="Alpha",
status="active",
first_seen_at=datetime.now(timezone.utc),
last_seen_at=datetime.now(timezone.utc),
)
)
db_session.add(CrawlRun(source_url="https://miem.hse.ru/persons", status="completed", new_count=3))
db_session.commit()
payload = stats_payload(db_session)
assert payload["total"] == 1
assert payload["active"] == 1
assert payload["new_in_last_run"] == 3
def test_run_payload_calculates_progress():
run = CrawlRun(
source_url="https://miem.hse.ru/persons",
status="running",
found_count=10,
parsed_count=4,
error_count=1,
)
payload = run_payload(run)
assert payload["processed_count"] == 5
assert payload["progress_percent"] == 50.0