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

View File

@@ -1,12 +1,15 @@
from datetime import date
from fastapi import APIRouter, BackgroundTasks, Depends, Request
from sqlalchemy import desc, or_, select
from sqlalchemy import desc, select
from sqlalchemy.orm import Session
from app.config import Settings, get_settings
from app.db import SessionLocal, get_db
from app.models import CrawlRun, Employee
from app.security import require_admin
from app.services.crawler import run_crawl
from app.services.admin_data import employee_display_payload, list_employees_page, run_payload, stats_payload
from app.services.crawl_control import get_running_run, run_crawl_if_idle
from app.version import BACKEND_VERSION, FRONTEND_VERSION
router = APIRouter(prefix="/api")
@@ -22,20 +25,29 @@ def list_employees(
request: Request,
status: str | None = None,
q: str | None = None,
started_from: date | None = None,
started_to: date | None = None,
has_email: bool | None = None,
sort: str = "full_name",
direction: str = "asc",
limit: int = 50,
offset: int = 0,
db: Session = Depends(get_db),
settings: Settings = Depends(get_settings),
) -> dict:
require_admin(request, settings)
stmt = select(Employee)
if status:
stmt = stmt.where(Employee.status == status)
if q:
pattern = f"%{q}%"
stmt = stmt.where(or_(Employee.full_name.ilike(pattern), Employee.canonical_url.ilike(pattern)))
employees = db.scalars(stmt.order_by(Employee.full_name).limit(limit).offset(offset)).all()
return {"items": [_employee_summary(item) for item in employees], "limit": limit, "offset": offset}
return list_employees_page(
db,
status=status,
q=q,
started_from=started_from,
started_to=started_to,
has_email=has_email,
sort=sort,
direction=direction,
limit=limit,
offset=offset,
)
@router.get("/employees/{employee_id}")
@@ -61,34 +73,53 @@ def list_crawl_runs(
) -> dict:
require_admin(request, settings)
runs = db.scalars(select(CrawlRun).order_by(desc(CrawlRun.started_at)).limit(limit)).all()
return {"items": [_run_summary(run) for run in runs]}
return {"items": [run_payload(run) for run in runs]}
@router.get("/crawl-runs/latest")
def latest_crawl_run(
request: Request,
db: Session = Depends(get_db),
settings: Settings = Depends(get_settings),
) -> dict:
require_admin(request, settings)
running = get_running_run(db)
latest = db.scalar(select(CrawlRun).order_by(desc(CrawlRun.started_at)).limit(1))
return {"running": run_payload(running), "latest": run_payload(latest)}
@router.post("/crawl-runs")
def trigger_crawl(
request: Request,
background_tasks: BackgroundTasks,
db: Session = Depends(get_db),
settings: Settings = Depends(get_settings),
) -> dict:
require_admin(request, settings)
running = get_running_run(db)
if running:
return {"status": "already_running", "run": run_payload(running)}
def _crawl() -> None:
with SessionLocal() as db:
run_crawl(db, settings)
run_crawl_if_idle(db, settings)
background_tasks.add_task(_crawl)
return {"status": "scheduled"}
@router.get("/stats")
def stats(
request: Request,
db: Session = Depends(get_db),
settings: Settings = Depends(get_settings),
) -> dict:
require_admin(request, settings)
return stats_payload(db)
def _employee_summary(employee: Employee) -> dict:
return {
"id": employee.id,
"full_name": employee.full_name,
"status": employee.status,
"canonical_url": employee.canonical_url,
"last_seen_at": employee.last_seen_at.isoformat() if employee.last_seen_at else None,
"dismissed_at": employee.dismissed_at.isoformat() if employee.dismissed_at else None,
}
return employee_display_payload(employee)
def _employee_detail(employee: Employee) -> dict:
@@ -99,15 +130,4 @@ def _employee_detail(employee: Employee) -> dict:
def _run_summary(run: CrawlRun) -> dict:
return {
"id": run.id,
"source_url": run.source_url,
"status": run.status,
"started_at": run.started_at.isoformat() if run.started_at else None,
"finished_at": run.finished_at.isoformat() if run.finished_at else None,
"found_count": run.found_count,
"parsed_count": run.parsed_count,
"error_count": run.error_count,
"dismissed_count": run.dismissed_count,
"message": run.message,
}
return run_payload(run) or {}