add interactive parser of наш.дом.рф ЖК+flats with stealth Playwright

- Discovered real endpoints via chrome-devtools: /сервисы/api/kn/object
  (offset/limit/place/objStatus) + /portal-kn/api/sales/portal/table.
  Server-side filter ignored — local filter by developer.companyGroup.
- Pure Playwright in-context fetch (httpx blocked by ServicePipe TLS-fp).
- Saved fingerprint state in data/playwright_state.json for server reuse.
- Celery beat (configurable cron + jitter) + admin trigger endpoint
  + /admin/scrape UI page.
- Schema migration: UNIQUE(id, snapshot_date) for versioned snapshots,
  kn_scrape_runs journal.
- Smoke-tested on PRINZIP: 28 ЖК / 826 квартир as expected.

backend/Dockerfile

@@ -56,6 +56,10 @@ COPY --from=builder --chown=app:app /app/app /app/app
 COPY --from=builder --chown=app:app /app/alembic.ini /app/alembic.ini
 COPY --from=builder --chown=app:app /app/alembic /app/alembic
 
+# Playwright Chromium for /сервисы/api/kn/* WAF-bootstrap (used only by worker).
+# --with-deps installs ~80MB of system libs needed by Chromium headless.
+RUN /app/.venv/bin/playwright install --with-deps chromium
+
 USER app
 
 EXPOSE 8000

backend/app/api/v1/admin_scrape.py

@@ -0,0 +1,101 @@
+"""Admin trigger for ad-hoc kn-API sweeps.
+
+POST /api/v1/admin/scrape/kn
+Body: {"region_code": 66, "developers": ["6208_0"]?, "fetch_flats": true?}
+Header: X-Admin-Token: <SCRAPE_ADMIN_TOKEN env var>
+
+Disabled if SCRAPE_ADMIN_TOKEN is empty (default).
+Returns the Celery task id; poll /api/v1/admin/scrape/runs/{run_id} to track DB-side progress.
+"""
+
+from __future__ import annotations
+
+from typing import Annotated, Any
+
+from fastapi import APIRouter, Depends, Header, HTTPException
+from pydantic import BaseModel, Field
+from sqlalchemy import text
+from sqlalchemy.orm import Session
+
+from app.core.config import settings
+from app.core.db import get_db
+
+router = APIRouter()
+
+
+class TriggerKnRequest(BaseModel):
+    region_code: int = Field(..., ge=1, le=99)
+    developers: list[str] | None = Field(default=None, max_length=20)
+    fetch_flats: bool = True
+
+
+def _check_token(x_admin_token: str | None) -> None:
+    if not settings.scrape_admin_token:
+        raise HTTPException(
+            status_code=503,
+            detail="admin scrape disabled — set SCRAPE_ADMIN_TOKEN",
+        )
+    if x_admin_token != settings.scrape_admin_token:
+        raise HTTPException(status_code=401, detail="invalid admin token")
+
+
+@router.post("/kn")
+def trigger_kn_sweep(
+    payload: TriggerKnRequest,
+    x_admin_token: Annotated[str | None, Header(alias="X-Admin-Token")] = None,
+) -> dict[str, Any]:
+    _check_token(x_admin_token)
+    # Defer Celery import so the API works without a broker in dev.
+    from app.workers.tasks.scrape_kn import scrape_kn_region
+
+    result = scrape_kn_region.apply_async(
+        args=[payload.region_code, payload.developers],
+        kwargs={"skip_jitter": True},
+    )
+    return {
+        "task_id": result.id,
+        "region_code": payload.region_code,
+        "developers": payload.developers,
+        "queued_at": result.date_done.isoformat() if result.date_done else None,
+    }
+
+
+@router.get("/runs")
+def list_runs(
+    db: Annotated[Session, Depends(get_db)],
+    x_admin_token: Annotated[str | None, Header(alias="X-Admin-Token")] = None,
+    limit: int = 20,
+) -> list[dict[str, Any]]:
+    _check_token(x_admin_token)
+    rows = (
+        db.execute(
+            text(
+                """
+            SELECT run_id, started_at, finished_at, region_codes, developer_ids,
+                   objects_count, flats_count, requests_count, status, error, snapshot_date
+              FROM kn_scrape_runs
+             ORDER BY started_at DESC
+             LIMIT :lim
+            """
+            ),
+            {"lim": min(limit, 100)},
+        )
+        .mappings()
+        .all()
+    )
+    return [
+        {
+            "run_id": r["run_id"],
+            "started_at": r["started_at"].isoformat() if r["started_at"] else None,
+            "finished_at": r["finished_at"].isoformat() if r["finished_at"] else None,
+            "region_codes": list(r["region_codes"]) if r["region_codes"] else [],
+            "developer_ids": list(r["developer_ids"]) if r["developer_ids"] else [],
+            "objects_count": r["objects_count"],
+            "flats_count": r["flats_count"],
+            "requests_count": r["requests_count"],
+            "status": r["status"],
+            "error": r["error"],
+            "snapshot_date": r["snapshot_date"].isoformat() if r["snapshot_date"] else None,
+        }
+        for r in rows
+    ]

backend/app/core/config.py

@@ -14,5 +14,21 @@ class Settings(BaseSettings):
     rosreestr_pkk_base_url: str = "https://pkk.rosreestr.ru/api/features/1"
     overpass_url: str = "https://overpass-api.de/api/interpreter"
 
+    # Scraper schedule (наш.дом.рф kn-API).
+    # Crontab format: "minute hour day_of_month month day_of_week" (Celery crontab).
+    # Default: каждый понедельник в окне 04:00–05:00 по МСК.
+    scrape_kn_cron: str = "15 4 * * mon"
+    # Random delay window in seconds added on top of scheduled start so the worker
+    # does not hit DOM.РФ at exactly the same minute every cycle. 0 = disabled.
+    scrape_kn_jitter_seconds: int = 1800
+    # Default region(s) for scheduled sweeps. Comma-separated.
+    scrape_kn_default_regions: str = "66"
+    # Path to a pre-captured Playwright storage_state.json (committed in repo,
+    # used by worker to skip cold-start WAF challenge).
+    scrape_kn_state_path: str = "data/playwright_state.json"
+    # Token to authorize ad-hoc /api/v1/admin/scrape/* trigger calls.
+    # Empty string = endpoint disabled.
+    scrape_admin_token: str = ""
+
 
 settings = Settings()

backend/app/main.py

@@ -5,7 +5,7 @@ import sentry_sdk
 from fastapi import FastAPI
 from fastapi.middleware.cors import CORSMiddleware
 
-from app.api.v1 import analytics, concepts, parcels
+from app.api.v1 import admin_scrape, analytics, concepts, parcels
 from app.core.config import settings
 
 
@@ -33,6 +33,7 @@ app.add_middleware(
 app.include_router(concepts.router, prefix="/api/v1/concepts", tags=["concepts"])
 app.include_router(parcels.router, prefix="/api/v1/parcels", tags=["parcels"])
 app.include_router(analytics.router, prefix="/api/v1/analytics", tags=["analytics"])
+app.include_router(admin_scrape.router, prefix="/api/v1/admin/scrape", tags=["admin"])
 
 
 @app.get("/health")

backend/app/services/scrapers/domrf_kn.py

@@ -0,0 +1,537 @@
+"""Scraper for наш.дом.рф kn-API (objects + flats).
+
+Endpoints discovered via chrome-devtools probe:
+    GET /сервисы/api/kn/object
+        params: offset, limit, place, objStatus, sortField=default, sortType=desc
+        place: "0-4" (city Екатеринбург) or "66" (region Свердл)
+        objStatus: 0=строящиеся, 1=архив-1, 2=сданные (sweep 0..2 for full coverage)
+        Returns {data: {list: [obj, ...], total: N}}.
+
+    GET /portal-kn/api/sales/portal/table?externalId={objId}
+        Per-object flat catalogue: nested entrances → floors → flats.
+
+Auth header: `Authorization: Basic MTpxd2U=` (decodes to `1:qwe`, public debug
+auth shipped in their frontend bundle).
+"""
+
+from __future__ import annotations
+
+import json
+import logging
+from datetime import date, datetime
+from typing import Any
+
+from sqlalchemy import text
+from sqlalchemy.orm import Session
+
+from app.core.db import SessionLocal
+from app.services.scrapers.stealth import BrowserSession
+
+logger = logging.getLogger(__name__)
+
+PATH_OBJECTS = "/сервисы/api/kn/object"
+PATH_FLATS_TABLE = "/portal-kn/api/sales/portal/table"
+RAW_SECTION = "kn_api"
+
+# Three statuses observed in production (Apr 2026):
+#   0 — строящиеся ЖК
+#   1 — небольшая выборка ~2 объекта на регион (legacy/archive flag)
+#   2 — сданные / завершённые (самый большой бакет)
+DEFAULT_STATUSES = (0, 1, 2)
+
+
+# ── helpers ──────────────────────────────────────────────────────────────────
+
+
+def _g(d: dict[str, Any], *keys: str, default: Any = None) -> Any:
+    for k in keys:
+        if k in d and d[k] is not None:
+            return d[k]
+    return default
+
+
+def _to_bool(v: Any) -> bool | None:
+    """Coerce DOM.РФ 0/1 ints into Python bool. Pass through actual bools/None."""
+    if v is None:
+        return None
+    if isinstance(v, bool):
+        return v
+    if isinstance(v, int | float):
+        return bool(v)
+    if isinstance(v, str):
+        s = v.strip().lower()
+        if s in ("true", "1", "y", "yes"):
+            return True
+        if s in ("false", "0", "n", "no", ""):
+            return False
+    return None
+
+
+def _to_date(v: Any) -> date | None:
+    """Coerce date string to date. Accept 'YYYY-MM-DD', 'YYYY-MM-DD HH:MM:SS',
+    'DD-MM-YYYY HH:MM:SS', or quarter-string like 'IV кв. 2028'.
+    Quarter strings get mapped to the last day of the quarter.
+    """
+    if v is None or v == "":
+        return None
+    if isinstance(v, date) and not isinstance(v, datetime):
+        return v
+    if isinstance(v, datetime):
+        return v.date()
+    if not isinstance(v, str):
+        return None
+    s = v.strip()
+    for fmt in ("%Y-%m-%d", "%Y-%m-%d %H:%M:%S", "%d-%m-%Y %H:%M:%S", "%d-%m-%Y", "%d.%m.%Y"):
+        try:
+            return datetime.strptime(s, fmt).date()
+        except ValueError:
+            pass
+    # Quarter-like 'IV кв. 2028' / '4 кв. 2028' / 'III квартал 2028'
+    quarter_end = {1: (3, 31), 2: (6, 30), 3: (9, 30), 4: (12, 31)}
+    roman = {"I": 1, "II": 2, "III": 3, "IV": 4}
+    parts = s.replace(".", " ").replace("кв", " ").replace("квартал", " ").split()
+    qq = year = None
+    for p in parts:
+        p = p.strip()
+        if p in roman and qq is None:
+            qq = roman[p]
+        elif p.isdigit():
+            n = int(p)
+            if 1 <= n <= 4 and qq is None:
+                qq = n
+            elif 2000 <= n <= 2050 and year is None:
+                year = n
+    if qq and year:
+        m, d = quarter_end[qq]
+        return date(year, m, d)
+    return None
+
+
+def _problem_text(v: Any) -> str | None:
+    """Coerce problem flag (sometimes int 0/1, sometimes text) to text or None."""
+    if v is None:
+        return None
+    if isinstance(v, bool):
+        return "1" if v else "0"
+    if isinstance(v, int | float):
+        return str(int(v))
+    if isinstance(v, str):
+        return v.strip() or None
+    return None
+
+
+def _extract_total(payload: dict[str, Any]) -> int | None:
+    data = payload.get("data")
+    if isinstance(data, dict):
+        for key in ("total", "totalCount", "count"):
+            v = data.get(key)
+            if isinstance(v, int):
+                return v
+    return None
+
+
+def _extract_list(payload: dict[str, Any]) -> list[dict[str, Any]]:
+    data = payload.get("data")
+    if isinstance(data, list):
+        return data
+    if isinstance(data, dict):
+        for key in ("list", "items", "objects", "content", "rows"):
+            v = data.get(key)
+            if isinstance(v, list):
+                return v
+    return []
+
+
+# ── normalisation ────────────────────────────────────────────────────────────
+
+
+def _norm_object(row: dict[str, Any], region_cd: int | None = None) -> dict[str, Any]:
+    """Map naш.дом.рф /kn/object row → domrf_kn_objects column dict.
+
+    Real API field names:
+      objId, hobjId, developer{devId, shortName, fullName, groupName, companyGroup, devInn},
+      rpdRegionCd, objAddr, shortAddr, objCommercNm, objFloorMin, objFloorMax,
+      objElemLivingCnt, objSquareLiving, objReady100PercDt, objClass, latitude, longitude,
+      objProblemFlg, problemFlag, siteStatus, objGreenHouseFlg, objGuarantyEscrowFlg,
+      objStatus, freeFlatsInfo{priceMin, numberFlats}.
+    """
+    dev = row.get("developer") if isinstance(row.get("developer"), dict) else {}
+    company_group = _g(dev, "companyGroup") if dev else None
+    # Our DB convention: dev_id="<companyGroup>_0" matches v_developer_full_metrics.
+    dev_id = f"{company_group}_0" if company_group else None
+    return {
+        "obj_id": _g(row, "objId", "obj_id", "id"),
+        "hobj_id": _g(row, "hobjId", "hobj_id"),
+        "comm_name": _g(row, "objCommercNm"),
+        "addr": _g(row, "objAddr"),
+        "short_addr": _g(row, "shortAddr"),
+        "region_cd": _g(row, "rpdRegionCd") or region_cd,
+        "dev_id": dev_id,
+        "dev_name": _g(dev, "groupName") or _g(dev, "shortName") or _g(dev, "fullName"),
+        "dev_inn": _g(dev, "devInn"),
+        "floor_min": _g(row, "objFloorMin"),
+        "floor_max": _g(row, "objFloorMax"),
+        "flat_count": _g(row, "objElemLivingCnt"),
+        "square_living": _g(row, "objSquareLiving"),
+        "ready_dt": _to_date(_g(row, "objReady100PercDt")),
+        "problem_flag": _problem_text(_g(row, "problemFlag", "objProblemFlg")),
+        "site_status": _g(row, "siteStatus"),
+        "green_house": _to_bool(_g(row, "objGreenHouseFlg")),
+        "escrow": _to_bool(_g(row, "objGuarantyEscrowFlg")),
+        "obj_class": _g(row, "objClass"),
+        "wall_type": _g(row, "wallType"),
+        "energy_eff": _g(row, "energyEff"),
+        "latitude": _g(row, "latitude"),
+        "longitude": _g(row, "longitude"),
+        "obj_status": _g(row, "objStatus"),
+    }
+
+
+def _flatten_table(table_payload: dict[str, Any]) -> list[dict[str, Any]]:
+    """Iterate nested entrances → floors → flats[] from /sales/portal/table response."""
+    out: list[dict[str, Any]] = []
+    ext_id = table_payload.get("externalId")
+    for entrance in table_payload.get("entrances") or []:
+        for floor in entrance.get("floors") or []:
+            for flat in floor.get("flats") or []:
+                # Inject obj_id and floor for normalisation
+                flat = {
+                    **flat,
+                    "_objId": ext_id,
+                    "_floor": floor.get("floorNumber"),
+                    "_entrance": entrance.get("entranceNumber"),
+                }
+                out.append(flat)
+    return out
+
+
+def _norm_flat(row: dict[str, Any], region_cd: int | None) -> dict[str, Any]:
+    """Map flat from /sales/portal/table → domrf_kn_flats columns.
+
+    Real fields: flatId|id (numeric, may be null), odsId, elemId (uuid hash),
+    type, number, isStudio, totalArea, livingArea, rooms, status (free|booked|sold),
+    price, pricePerSquareMeter, numberFloors. Plus injected _objId and _floor.
+    """
+    flat_id = _g(row, "flatId", "id")
+    if flat_id is None:
+        # Fallback: hash of elemId (stable, fits BIGINT). Without an id we cannot UPSERT.
+        elem = _g(row, "elemId")
+        if elem:
+            flat_id = abs(hash(elem)) % (2**63 - 1)
+    return {
+        "id": flat_id,
+        "ods_id": _g(row, "odsId"),
+        "flat_type": _g(row, "type", "flatType"),
+        "flat_number": _g(row, "number", "flatNumber"),
+        "is_studio": _to_bool(_g(row, "isStudio")),
+        "total_area": _g(row, "totalArea"),
+        "living_area": _g(row, "livingArea"),
+        "rooms": _g(row, "rooms"),
+        "status": _g(row, "status"),
+        "price_rub": _g(row, "price"),
+        "price_per_m2": _g(row, "pricePerSquareMeter"),
+        "floor": _g(row, "_floor", "floor"),
+        "num_floors": _g(row, "numberFloors"),
+        "obj_id": _g(row, "_objId", "objId"),
+        "city": _g(row, "city"),
+        "region_cd": region_cd,
+        "obj_name": _g(row, "objName") or _g(row.get("objInfo") or {}, "objCommercNm"),
+    }
+
+
+# ── DB writes ────────────────────────────────────────────────────────────────
+
+UPSERT_OBJECT_SQL = text(
+    """
+    INSERT INTO domrf_kn_objects (
+        obj_id, hobj_id, comm_name, addr, short_addr, region_cd,
+        dev_id, dev_name, dev_inn,
+        floor_min, floor_max, flat_count, square_living, ready_dt,
+        problem_flag, site_status, green_house, escrow,
+        obj_class, wall_type, energy_eff,
+        latitude, longitude, obj_status, snapshot_date
+    ) VALUES (
+        :obj_id, :hobj_id, :comm_name, :addr, :short_addr, :region_cd,
+        :dev_id, :dev_name, :dev_inn,
+        :floor_min, :floor_max, :flat_count, :square_living, :ready_dt,
+        :problem_flag, :site_status, :green_house, :escrow,
+        :obj_class, :wall_type, :energy_eff,
+        :latitude, :longitude, :obj_status, :snapshot_date
+    )
+    ON CONFLICT (obj_id, snapshot_date) DO UPDATE SET
+        comm_name = EXCLUDED.comm_name,
+        addr = EXCLUDED.addr,
+        dev_id = EXCLUDED.dev_id,
+        dev_name = EXCLUDED.dev_name,
+        flat_count = EXCLUDED.flat_count,
+        square_living = EXCLUDED.square_living,
+        ready_dt = EXCLUDED.ready_dt,
+        site_status = EXCLUDED.site_status,
+        escrow = EXCLUDED.escrow,
+        obj_class = EXCLUDED.obj_class,
+        latitude = EXCLUDED.latitude,
+        longitude = EXCLUDED.longitude,
+        obj_status = EXCLUDED.obj_status
+    """
+)
+
+UPSERT_FLAT_SQL = text(
+    """
+    INSERT INTO domrf_kn_flats (
+        id, ods_id, flat_type, flat_number, is_studio, total_area, living_area,
+        rooms, status, price_rub, price_per_m2, floor, num_floors, obj_id,
+        city, region_cd, obj_name, snapshot_date
+    ) VALUES (
+        :id, :ods_id, :flat_type, :flat_number, :is_studio, :total_area, :living_area,
+        :rooms, :status, :price_rub, :price_per_m2, :floor, :num_floors, :obj_id,
+        :city, :region_cd, :obj_name, :snapshot_date
+    )
+    ON CONFLICT (id, snapshot_date) DO UPDATE SET
+        status = EXCLUDED.status,
+        price_rub = EXCLUDED.price_rub,
+        price_per_m2 = EXCLUDED.price_per_m2,
+        obj_id = EXCLUDED.obj_id,
+        region_cd = EXCLUDED.region_cd,
+        obj_name = EXCLUDED.obj_name
+    """
+)
+
+INSERT_RAW_SQL = text(
+    """
+    INSERT INTO domrf_raw_endpoints (
+        snapshot_date, section, endpoint, source_url, payload, payload_size
+    ) VALUES (
+        :snapshot_date, :section, :endpoint, :source_url,
+        CAST(:payload AS JSONB), :payload_size
+    )
+    ON CONFLICT (snapshot_date, section, endpoint) DO UPDATE SET
+        payload = EXCLUDED.payload,
+        payload_size = EXCLUDED.payload_size,
+        source_url = EXCLUDED.source_url
+    """
+)
+
+
+def _insert_raw(db: Session, snapshot_date: date, endpoint_label: str, payload: Any) -> None:
+    body = json.dumps(payload, ensure_ascii=False)
+    db.execute(
+        INSERT_RAW_SQL,
+        {
+            "snapshot_date": snapshot_date,
+            "section": RAW_SECTION,
+            "endpoint": endpoint_label[:255],
+            "source_url": f"https://xn--80az8a.xn--d1aqf.xn--p1ai{endpoint_label}",
+            "payload": body,
+            "payload_size": len(body.encode("utf-8")),
+        },
+    )
+
+
+def upsert_objects(
+    db: Session, raw: list[dict[str, Any]], snapshot_date: date, region_cd: int
+) -> int:
+    inserted = 0
+    for row in raw:
+        norm = _norm_object(row, region_cd=region_cd)
+        if not norm["obj_id"]:
+            continue
+        norm["snapshot_date"] = snapshot_date
+        try:
+            db.execute(UPSERT_OBJECT_SQL, norm)
+            inserted += 1
+        except Exception as e:
+            logger.warning("upsert object %s failed: %s", norm.get("obj_id"), e)
+            db.rollback()
+    return inserted
+
+
+def upsert_flats(
+    db: Session, raw: list[dict[str, Any]], snapshot_date: date, region_cd: int
+) -> int:
+    inserted = 0
+    skipped_no_id = 0
+    for row in raw:
+        norm = _norm_flat(row, region_cd)
+        if norm["id"] is None:
+            skipped_no_id += 1
+            continue
+        norm["snapshot_date"] = snapshot_date
+        try:
+            db.execute(UPSERT_FLAT_SQL, norm)
+            inserted += 1
+        except Exception as e:
+            logger.warning("upsert flat %s failed: %s", norm.get("id"), e)
+            db.rollback()
+    if skipped_no_id:
+        logger.info("skipped %d flats without numeric id", skipped_no_id)
+    return inserted
+
+
+# ── orchestrator ─────────────────────────────────────────────────────────────
+
+
+def _place_str(region_code: int) -> str:
+    """Map region code → DOM.РФ `place` parameter. By default uses the regionCode itself
+    (e.g. 66=Свердл region). City-level codes (e.g. '0-4'=ЕКБ) can be passed as-is via
+    PLACE_OVERRIDE if more granularity is needed.
+    """
+    return str(region_code)
+
+
+async def fetch_objects_for_status(
+    sess: BrowserSession, place: str, status: int
+) -> list[dict[str, Any]]:
+    """Fetch ALL objects for a given (place, objStatus) — server returns up to limit=999999."""
+    payload = await sess.get_json(
+        PATH_OBJECTS,
+        {
+            "offset": 0,
+            "limit": 999999,
+            "sortField": "default",
+            "sortType": "desc",
+            "place": place,
+            "objStatus": status,
+        },
+    )
+    rows = _extract_list(payload)
+    total = _extract_total(payload)
+    logger.info("kn/object place=%s status=%d -> %d/%s rows", place, status, len(rows), total)
+    return rows
+
+
+async def fetch_flats_for_object(sess: BrowserSession, obj_id: int) -> list[dict[str, Any]]:
+    """Fetch flat-table for one object, return flat rows (entrance/floor flattened)."""
+    try:
+        payload = await sess.get_json(PATH_FLATS_TABLE, {"externalId": obj_id})
+    except Exception as e:
+        logger.warning("flats fetch obj=%s failed: %s", obj_id, e)
+        return []
+    # Body shape: {externalId, entrances: [{entranceNumber, floors:[{floorNumber, flats:[...]}]}]}
+    return _flatten_table(payload)
+
+
+async def run_region_sweep(
+    region_code: int,
+    developers: list[str] | None = None,
+    headed: bool = False,
+    snapshot_date: date | None = None,
+    load_state: str | None = None,
+    save_state: str | None = None,
+    statuses: tuple[int, ...] = DEFAULT_STATUSES,
+    fetch_flats: bool = True,
+    place_override: str | None = None,
+) -> dict[str, Any]:
+    """Full sweep for one region: bootstrap → fetch objects (each objStatus) → fetch flats.
+
+    `developers` filters the OBJECT set locally by `developer.companyGroup` matching the
+    "<group>_0" id pattern (since DOM.РФ doesn't honour developer filter on the API).
+    """
+    snapshot_date = snapshot_date or date.today()
+    db = SessionLocal()
+    run_id = db.execute(
+        text(
+            """
+            INSERT INTO kn_scrape_runs (region_codes, developer_ids, snapshot_date, status)
+            VALUES (ARRAY[:rc]::int[], CAST(:devs AS text[]), :snap, 'running')
+            RETURNING run_id
+            """
+        ),
+        {
+            "rc": region_code,
+            "devs": developers if developers else None,
+            "snap": snapshot_date,
+        },
+    ).scalar_one()
+    db.commit()
+
+    place = place_override or _place_str(region_code)
+
+    try:
+        async with BrowserSession(
+            region_code=region_code,
+            headed=headed,
+            load_state=load_state,
+            save_state=save_state,
+        ) as sess:
+            # --- objects across statuses -------------------------------------
+            all_objects: list[dict[str, Any]] = []
+            for status in statuses:
+                rows = await fetch_objects_for_status(sess, place, status)
+                all_objects.extend(rows)
+            logger.info("place=%s total objects across %s = %d", place, statuses, len(all_objects))
+
+            # local developer filter
+            if developers:
+                wanted_groups = {int(d.split("_")[0]) for d in developers if "_" in d}
+                all_objects = [
+                    o
+                    for o in all_objects
+                    if isinstance(o.get("developer"), dict)
+                    and o["developer"].get("companyGroup") in wanted_groups
+                ]
+                logger.info("after developers filter: %d objects", len(all_objects))
+
+            # --- flats per object --------------------------------------------
+            all_flats: list[dict[str, Any]] = []
+            if fetch_flats:
+                for i, obj in enumerate(all_objects):
+                    obj_id = obj.get("objId") or obj.get("obj_id")
+                    if not obj_id:
+                        continue
+                    flats = await fetch_flats_for_object(sess, obj_id)
+                    all_flats.extend(flats)
+                    if (i + 1) % 25 == 0:
+                        logger.info(
+                            "flats progress: %d/%d objs, %d flats so far",
+                            i + 1,
+                            len(all_objects),
+                            len(all_flats),
+                        )
+                logger.info("total flats: %d (across %d objs)", len(all_flats), len(all_objects))
+
+            request_count = sess.request_count
+
+        # --- write all in one transaction ------------------------------------
+        obj_count = upsert_objects(db, all_objects, snapshot_date, region_code)
+        flat_count = upsert_flats(db, all_flats, snapshot_date, region_code) if fetch_flats else 0
+        # Save raw payloads (one row per status; flats are huge so we store the count, not bodies).
+        _insert_raw(db, snapshot_date, f"kn_object_place_{place}", all_objects)
+
+        db.execute(
+            text(
+                """
+                UPDATE kn_scrape_runs
+                   SET finished_at=NOW(), objects_count=:oc, flats_count=:fc,
+                       requests_count=:rq, status='done'
+                 WHERE run_id=:rid
+                """
+            ),
+            {"oc": obj_count, "fc": flat_count, "rq": request_count, "rid": run_id},
+        )
+        db.commit()
+        return {
+            "run_id": run_id,
+            "region_code": region_code,
+            "place": place,
+            "objects": obj_count,
+            "flats": flat_count,
+            "requests": request_count,
+            "snapshot_date": snapshot_date.isoformat(),
+        }
+    except Exception as e:
+        logger.exception("kn sweep failed for region=%d", region_code)
+        db.rollback()
+        db.execute(
+            text(
+                """
+                UPDATE kn_scrape_runs
+                   SET finished_at=NOW(), status='failed', error=:err
+                 WHERE run_id=:rid
+                """
+            ),
+            {"err": str(e)[:2000], "rid": run_id},
+        )
+        db.commit()
+        raise
+    finally:
+        db.close()

backend/app/services/scrapers/stealth.py

@@ -0,0 +1,182 @@
+"""Stealth-fetch utilities for наш.дом.рф scraping.
+
+Pure-httpx hits ServicePipe WAF JS-challenge even with Playwright-exported
+cookies (different TLS-fingerprint = blocked). The proven pattern from
+data/sql/30_scrape_domrf.py is to issue `fetch()` calls *inside* the live
+Chromium page — same JA3 fingerprint as a real browser, cookies/CSP/CORS
+all handled natively.
+
+BrowserSession keeps the browser+page open for the lifetime of a sweep and
+exposes get_json() that calls page.evaluate(fetch_in_browser).
+"""
+
+from __future__ import annotations
+
+import asyncio
+import json
+import logging
+import random
+from typing import Any
+from urllib.parse import urlencode
+
+from playwright.async_api import Browser, BrowserContext, Page, async_playwright
+
+logger = logging.getLogger(__name__)
+
+BASE_URL = "https://xn--80az8a.xn--d1aqf.xn--p1ai"
+
+# Real Chrome 147 fingerprint (Win10 x64). Keep in sync with data/sql/30_scrape_domrf.py.
+USER_AGENT = (
+    "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 "
+    "(KHTML, like Gecko) Chrome/147.0.0.0 Safari/537.36"
+)
+
+# Маппинг region_code → URL-сегмент города для реалистичного Referer.
+# Не обязан быть исчерпывающим — fallback на /новостройки/строящиеся/.
+REGION_LANDING_PATH = {
+    66: "/новостройки/строящиеся/екатеринбург/",
+    77: "/новостройки/строящиеся/москва/",
+    78: "/новостройки/строящиеся/санкт-петербург/",
+    50: "/новостройки/строящиеся/московская-область/",
+    47: "/новостройки/строящиеся/ленинградская-область/",
+}
+
+
+def make_referer(region_code: int) -> str:
+    path = REGION_LANDING_PATH.get(region_code, "/новостройки/строящиеся/")
+    return BASE_URL + path
+
+
+async def jitter_sleep(min_ms: int = 600, max_ms: int = 1500) -> None:
+    delay = random.uniform(min_ms / 1000, max_ms / 1000)
+    await asyncio.sleep(delay)
+
+
+# JS executed inside the live page. Returns {ok, status, body, contentType}.
+# Auth header is hardcoded `Basic MTpxd2U=` (decodes to `1:qwe`, the public
+# debug auth shipped in the site's frontend bundle for the kn API).
+_FETCH_JS = """
+async ({url, auth}) => {
+  const headers = {'Accept': 'application/json, text/plain, */*'};
+  if (auth) headers['Authorization'] = auth;
+  try {
+    const r = await fetch(url, {credentials: 'include', headers});
+    const ctype = r.headers.get('content-type') || '';
+    const body = await r.text();
+    return {ok: r.ok, status: r.status, body, contentType: ctype};
+  } catch (e) {
+    return {ok: false, status: 0, body: String(e), contentType: ''};
+  }
+}
+"""
+
+
+class WafBlockedError(RuntimeError):
+    """Server returned non-JSON (likely ServicePipe JS-challenge)."""
+
+
+class BrowserSession:
+    """Keeps a live Playwright Chromium page; issues kn-API fetches inside it.
+
+    Use as async context manager. Concurrency is bounded by an asyncio.Semaphore
+    of size 3 (the same page can serve multiple concurrent fetches via JS-await).
+    """
+
+    def __init__(
+        self,
+        region_code: int = 66,
+        headed: bool = False,
+        auth: str | None = "Basic MTpxd2U=",
+        load_state: str | None = None,
+        save_state: str | None = None,
+    ) -> None:
+        self.region_code = region_code
+        self.headed = headed
+        self.auth = auth
+        # Path to a Playwright storage_state JSON (cookies + localStorage).
+        # If load_state is set, the context is created with these cookies preloaded —
+        # ServicePipe sees a returning user, no fresh JS-challenge.
+        # If save_state is set, the post-bootstrap state is dumped to that path
+        # so it can be committed and reused on a server.
+        self.load_state = load_state
+        self.save_state = save_state
+        self._pw: Any = None
+        self._browser: Browser | None = None
+        self._context: BrowserContext | None = None
+        self._page: Page | None = None
+        self._sem = asyncio.Semaphore(3)
+        self._request_count = 0
+
+    async def __aenter__(self) -> BrowserSession:
+        await self._bootstrap()
+        return self
+
+    async def __aexit__(self, *exc: Any) -> None:
+        if self._browser is not None:
+            await self._browser.close()
+        if self._pw is not None:
+            await self._pw.stop()
+
+    async def _bootstrap(self) -> None:
+        landing_url = make_referer(self.region_code)
+        logger.info("bootstrap: opening %s", landing_url)
+        self._pw = await async_playwright().start()
+        self._browser = await self._pw.chromium.launch(headless=not self.headed)
+        ctx_kwargs: dict[str, Any] = {
+            "user_agent": USER_AGENT,
+            "locale": "ru-RU",
+            "viewport": {"width": 1920, "height": 1080},
+        }
+        if self.load_state:
+            ctx_kwargs["storage_state"] = self.load_state
+            logger.info("bootstrap: loading saved storage_state from %s", self.load_state)
+        self._context = await self._browser.new_context(**ctx_kwargs)
+        self._page = await self._context.new_page()
+        await self._page.goto(landing_url, wait_until="domcontentloaded", timeout=30_000)
+        try:
+            await self._page.wait_for_load_state("networkidle", timeout=10_000)
+        except Exception:
+            pass
+        if self.save_state:
+            await self._context.storage_state(path=self.save_state)
+            logger.info("bootstrap: storage_state saved to %s", self.save_state)
+        logger.info("bootstrap: page ready, WAF challenge passed")
+
+    async def get_json(self, path: str, params: dict[str, Any]) -> dict[str, Any]:
+        """Fetch JSON via fetch() in browser. Retries with backoff on transient errors."""
+        if self._page is None:
+            raise RuntimeError("BrowserSession not bootstrapped")
+        url = BASE_URL + path + "?" + urlencode(params)
+        last_err: Exception | None = None
+        for attempt in range(5):
+            async with self._sem:
+                await jitter_sleep()
+                try:
+                    self._request_count += 1
+                    result = await self._page.evaluate(_FETCH_JS, {"url": url, "auth": self.auth})
+                except Exception as e:
+                    last_err = e
+                    logger.warning("evaluate err attempt=%d: %r", attempt, e)
+                    await asyncio.sleep(2**attempt)
+                    continue
+            status = result.get("status", 0)
+            body = result.get("body", "")
+            ctype = result.get("contentType", "")
+            if status in (429,) or status >= 500 or status == 0:
+                last_err = RuntimeError(f"transient status={status}")
+                logger.warning("transient status=%d attempt=%d, backing off", status, attempt)
+                await asyncio.sleep(2**attempt)
+                continue
+            if "application/json" not in ctype:
+                # WAF returned the JS challenge HTML; re-bootstrap could help, but for now fail.
+                raise WafBlockedError(
+                    f"non-JSON response: status={status} ctype={ctype} body[:120]={body[:120]!r}"
+                )
+            if status != 200:
+                raise RuntimeError(f"http {status}: {body[:200]}")
+            return json.loads(body)
+        raise RuntimeError(f"max retries exhausted: {last_err!r}")
+
+    @property
+    def request_count(self) -> int:
+        return self._request_count

backend/app/workers/celery_app.py

@@ -1,11 +1,46 @@
-"""Celery app + beat schedule.
+"""Celery app + beat schedule."""
-
-Stage 2a: daily PKK Rosreestr sync.
-"""
 
 from celery import Celery
+from celery.schedules import crontab
 
 from app.core.config import settings
 
-celery_app = Celery("gendesign", broker=settings.redis_url, backend=settings.redis_url)
+
+def _parse_cron(spec: str) -> crontab:
+    """Parse 'M H DoM Mon DoW' crontab string into Celery crontab. Empty fields default to '*'."""
+    parts = spec.strip().split()
+    if len(parts) != 5:
+        raise ValueError(f"crontab spec must have 5 fields, got: {spec!r}")
+    minute, hour, dom, month, dow = parts
+    return crontab(
+        minute=minute,
+        hour=hour,
+        day_of_month=dom,
+        month_of_year=month,
+        day_of_week=dow,
+    )
+
+
+def _default_regions() -> list[int]:
+    return [int(x.strip()) for x in settings.scrape_kn_default_regions.split(",") if x.strip()]
+
+
+celery_app = Celery(
+    "gendesign",
+    broker=settings.redis_url,
+    backend=settings.redis_url,
+    include=["app.workers.tasks.scrape_kn"],
+)
 celery_app.conf.timezone = "Europe/Moscow"
+
+# Расписание задаётся через SCRAPE_KN_CRON env var (см. app.core.config).
+# Каждый региональный sweep оборачивается случайной задержкой 0..SCRAPE_KN_JITTER_SECONDS
+# внутри самой задачи (см. tasks.scrape_kn) — чтобы не бить ровно в одну минуту.
+celery_app.conf.beat_schedule = {
+    f"kn-region-{rc}": {
+        "task": "tasks.scrape_kn.scrape_kn_region",
+        "schedule": _parse_cron(settings.scrape_kn_cron),
+        "args": [rc, None],
+    }
+    for rc in _default_regions()
+}

backend/app/workers/tasks/scrape_kn.py

@@ -0,0 +1,52 @@
+"""Celery task wrapper for naш.дом.рф kn-API sweeps."""
+
+from __future__ import annotations
+
+import asyncio
+import logging
+import random
+import time
+from pathlib import Path
+from typing import Any
+
+from app.core.config import settings
+from app.services.scrapers.domrf_kn import run_region_sweep
+from app.workers.celery_app import celery_app
+
+logger = logging.getLogger(__name__)
+
+
+@celery_app.task(bind=True, name="tasks.scrape_kn.scrape_kn_region", max_retries=2)
+def scrape_kn_region(
+    self: Any,
+    region_code: int,
+    developers: list[str] | None = None,
+    skip_jitter: bool = False,
+) -> dict[str, Any]:
+    """Run a full kn-API sweep for one region. Returns summary dict.
+
+    For scheduled runs (skip_jitter=False), sleeps a random 0..SCRAPE_KN_JITTER_SECONDS
+    interval before starting so multiple workers / repeated weekly cycles don't hit
+    DOM.РФ at exactly the same minute. Manual triggers pass skip_jitter=True for
+    immediate execution.
+    """
+    if not skip_jitter and settings.scrape_kn_jitter_seconds > 0:
+        delay = random.uniform(0, settings.scrape_kn_jitter_seconds)
+        logger.info("jitter delay %.0fs before scrape", delay)
+        time.sleep(delay)
+
+    state_path: str | None = settings.scrape_kn_state_path or None
+    if state_path and not Path(state_path).exists():
+        logger.warning("storage_state %s not found — bootstrap will run cold", state_path)
+        state_path = None
+
+    logger.info(
+        "scrape_kn_region start region=%s devs=%s state=%s", region_code, developers, state_path
+    )
+    return asyncio.run(
+        run_region_sweep(
+            region_code=region_code,
+            developers=developers,
+            load_state=state_path,
+        )
+    )

backend/pyproject.toml

@@ -18,6 +18,8 @@ dependencies = [
     "httpx>=0.27.0",
     "redis>=5.0.0",
     "celery>=5.4.0",
+    "playwright>=1.45.0",
+    "tenacity>=9.0.0",
     "weasyprint>=62.0",
     "ezdxf>=1.3.0",
     "openpyxl>=3.1.0",

data/sql/50_schema_kn_extensions.sql

@@ -0,0 +1,47 @@
+-- Расширения схемы для регулярного парсинга наш.дом.рф /сервисы/api/kn/{flats,objects}.
+-- См. план C:\Users\user\.claude\plans\crispy-waddling-swing.md.
+
+-- ── Версионирование строк через UNIQUE (id, snapshot_date) ────────────────
+-- Текущий PK на flats.id и objects.obj_id мешает сохранять несколько срезов
+-- одного и того же ЖК/квартиры. Меняем на UNIQUE по (id, snapshot_date)
+-- чтобы UPSERT идемпотентно записывал свежий snapshot.
+
+ALTER TABLE domrf_kn_flats DROP CONSTRAINT IF EXISTS domrf_kn_flats_pkey;
+ALTER TABLE domrf_kn_flats
+    ADD CONSTRAINT uq_kn_flats_id_snap UNIQUE (id, snapshot_date);
+
+ALTER TABLE domrf_kn_objects DROP CONSTRAINT IF EXISTS domrf_kn_objects_pkey;
+ALTER TABLE domrf_kn_objects
+    ADD CONSTRAINT uq_kn_objects_id_snap UNIQUE (obj_id, snapshot_date);
+
+-- ── Backfill region_cd для Антоновых flats где obj_id матчится с Свердл ──
+UPDATE domrf_kn_flats f
+   SET region_cd = 66
+  FROM domrf_kn_objects o
+ WHERE f.obj_id = o.obj_id
+   AND o.region_cd = 66
+   AND f.region_cd IS NULL;
+
+-- ── Композитные индексы для выборок последнего среза ──────────────────────
+CREATE INDEX IF NOT EXISTS idx_kn_flats_region_snap
+    ON domrf_kn_flats (region_cd, snapshot_date);
+CREATE INDEX IF NOT EXISTS idx_kn_objects_region_snap
+    ON domrf_kn_objects (region_cd, snapshot_date);
+
+-- ── Журнал прогонов скрейпера ─────────────────────────────────────────────
+CREATE TABLE IF NOT EXISTS kn_scrape_runs (
+    run_id           BIGSERIAL PRIMARY KEY,
+    started_at       TIMESTAMPTZ NOT NULL DEFAULT NOW(),
+    finished_at      TIMESTAMPTZ,
+    region_codes     INT[] NOT NULL,
+    developer_ids    TEXT[],
+    objects_count    INT,
+    flats_count      INT,
+    requests_count   INT,
+    status           TEXT NOT NULL DEFAULT 'running',  -- running | done | failed
+    error            TEXT,
+    snapshot_date    DATE NOT NULL DEFAULT CURRENT_DATE
+);
+CREATE INDEX IF NOT EXISTS idx_kn_runs_status ON kn_scrape_runs (status, started_at DESC);
+
+COMMENT ON TABLE kn_scrape_runs IS 'Журнал запусков парсера наш.дом.рф kn-API. Один прогон = один (regions, developers) sweep.';

data/sql/50_scrape_kn_flats.py

@@ -0,0 +1,94 @@
+"""CLI for ad-hoc naш.дом.рф kn-API sweeps.
+
+Usage:
+    PYTHONIOENCODING=utf-8 python data/sql/50_scrape_kn_flats.py --regions 66
+    python data/sql/50_scrape_kn_flats.py --regions 66 --developers 6208_0 --headed
+"""
+
+from __future__ import annotations
+
+import argparse
+import asyncio
+import logging
+import os
+import sys
+from pathlib import Path
+
+# allow `from app.services.scrapers.domrf_kn import ...` when run from repo root
+ROOT = Path(__file__).resolve().parents[2]
+sys.path.insert(0, str(ROOT / "backend"))
+
+from app.services.scrapers.domrf_kn import run_region_sweep  # noqa: E402
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="naш.дом.рф kn-API scraper (CLI)")
+    parser.add_argument(
+        "--regions",
+        type=int,
+        nargs="+",
+        required=True,
+        help="Region codes (1..89). e.g. --regions 66 77",
+    )
+    parser.add_argument(
+        "--developers",
+        type=str,
+        nargs="*",
+        help="Optional developer IDs in '<id>_0' format. e.g. --developers 6208_0",
+    )
+    parser.add_argument("--headed", action="store_true", help="Show Chromium window during bootstrap")
+    parser.add_argument(
+        "--save-state",
+        type=str,
+        default=None,
+        help="Path to dump Playwright storage_state.json after bootstrap (commit to git for server reuse)",
+    )
+    parser.add_argument(
+        "--load-state",
+        type=str,
+        default=None,
+        help="Path to load Playwright storage_state.json (server uses fingerprint captured locally)",
+    )
+    parser.add_argument(
+        "--bootstrap-only",
+        action="store_true",
+        help="Just bootstrap a session and exit — useful for capturing storage_state without scraping",
+    )
+    parser.add_argument("--verbose", "-v", action="store_true")
+    args = parser.parse_args()
+
+    logging.basicConfig(
+        level=logging.DEBUG if args.verbose else logging.INFO,
+        format="%(asctime)s %(levelname)s %(name)s: %(message)s",
+    )
+
+    if not os.environ.get("DATABASE_URL"):
+        print("WARNING: DATABASE_URL not set; using default from app.core.config")
+
+    async def _run() -> None:
+        if args.bootstrap_only:
+            from app.services.scrapers.stealth import BrowserSession
+            async with BrowserSession(
+                region_code=args.regions[0],
+                headed=args.headed,
+                load_state=args.load_state,
+                save_state=args.save_state,
+            ):
+                print(f"bootstrap-only: state saved to {args.save_state}")
+            return
+
+        for region in args.regions:
+            summary = await run_region_sweep(
+                region_code=region,
+                developers=args.developers or None,
+                headed=args.headed,
+                load_state=args.load_state,
+                save_state=args.save_state if region == args.regions[0] else None,
+            )
+            print(f"region={region} -> {summary}")
+
+    asyncio.run(_run())
+
+
+if __name__ == "__main__":
+    main()

docker-compose.yml

@@ -54,6 +54,28 @@ services:
     depends_on:
       - backend
 
+  worker:
+    build: ./backend
+    env_file: ./backend/.env
+    depends_on:
+      postgres:
+        condition: service_healthy
+      redis:
+        condition: service_started
+    volumes:
+      - ./backend/app:/app/app
+    command: ["celery", "-A", "app.workers.celery_app", "worker", "--loglevel=info", "--concurrency=2"]
+
+  beat:
+    build: ./backend
+    env_file: ./backend/.env
+    depends_on:
+      redis:
+        condition: service_started
+    volumes:
+      - ./backend/app:/app/app
+    command: ["celery", "-A", "app.workers.celery_app", "beat", "--loglevel=info"]
+
 volumes:
   postgres_data:
   redis_data:

frontend/src/app/admin/scrape/page.tsx

@@ -0,0 +1,273 @@
+"use client";
+
+import { useState } from "react";
+import { useMutation, useQuery, useQueryClient } from "@tanstack/react-query";
+
+import { apiFetch } from "@/lib/api";
+
+interface RunRow {
+  run_id: number;
+  started_at: string | null;
+  finished_at: string | null;
+  region_codes: number[];
+  developer_ids: string[];
+  objects_count: number | null;
+  flats_count: number | null;
+  requests_count: number | null;
+  status: string;
+  error: string | null;
+  snapshot_date: string | null;
+}
+
+interface TriggerResp {
+  task_id: string;
+  region_code: number;
+  developers: string[] | null;
+}
+
+export default function ScrapeAdminPage() {
+  const [token, setToken] = useState<string>(() =>
+    typeof window === "undefined"
+      ? ""
+      : (localStorage.getItem("admin_token") ?? ""),
+  );
+  const [region, setRegion] = useState(66);
+  const [devs, setDevs] = useState("");
+  const [fetchFlats, setFetchFlats] = useState(true);
+  const queryClient = useQueryClient();
+
+  const runs = useQuery({
+    queryKey: ["scrape-runs", token],
+    queryFn: () =>
+      apiFetch<RunRow[]>("/api/v1/admin/scrape/runs?limit=20", {
+        headers: { "X-Admin-Token": token },
+      }),
+    enabled: !!token,
+    refetchInterval: 5000,
+  });
+
+  const trigger = useMutation({
+    mutationFn: () =>
+      apiFetch<TriggerResp>("/api/v1/admin/scrape/kn", {
+        method: "POST",
+        headers: { "X-Admin-Token": token },
+        body: JSON.stringify({
+          region_code: region,
+          developers: devs
+            ? devs
+                .split(",")
+                .map((s) => s.trim())
+                .filter(Boolean)
+            : null,
+          fetch_flats: fetchFlats,
+        }),
+      }),
+    onSuccess: () =>
+      queryClient.invalidateQueries({ queryKey: ["scrape-runs"] }),
+  });
+
+  const saveToken = (v: string) => {
+    setToken(v);
+    localStorage.setItem("admin_token", v);
+  };
+
+  return (
+    <main style={{ padding: 24, maxWidth: 1100, margin: "0 auto" }}>
+      <h1 style={{ marginTop: 0 }}>Скрапер наш.дом.рф — ручной запуск</h1>
+      <p style={{ color: "#5b6066", fontSize: 13 }}>
+        Запуск sweep-а вне расписания (расписание: см. SCRAPE_KN_CRON env, по
+        умолчанию пн 04:15 МСК с jitter 0–30мин).
+      </p>
+
+      <section style={cardStyle}>
+        <label style={{ display: "block", marginBottom: 12 }}>
+          <span style={labelStyle}>Admin Token</span>
+          <input
+            type="password"
+            value={token}
+            onChange={(e) => saveToken(e.target.value)}
+            placeholder="SCRAPE_ADMIN_TOKEN"
+            style={inputStyle}
+          />
+        </label>
+        <div style={{ display: "flex", gap: 16, flexWrap: "wrap" }}>
+          <label>
+            <span style={labelStyle}>Регион</span>
+            <input
+              type="number"
+              value={region}
+              min={1}
+              max={99}
+              onChange={(e) => setRegion(Number(e.target.value))}
+              style={{ ...inputStyle, width: 100 }}
+            />
+          </label>
+          <label style={{ flex: 1 }}>
+            <span style={labelStyle}>Девелоперы (csv, опц.)</span>
+            <input
+              value={devs}
+              onChange={(e) => setDevs(e.target.value)}
+              placeholder="6208_0,5791_0"
+              style={inputStyle}
+            />
+          </label>
+          <label
+            style={{
+              display: "flex",
+              alignItems: "center",
+              gap: 6,
+              marginTop: 22,
+            }}
+          >
+            <input
+              type="checkbox"
+              checked={fetchFlats}
+              onChange={(e) => setFetchFlats(e.target.checked)}
+            />
+            <span>Также собрать квартиры</span>
+          </label>
+        </div>
+        <div style={{ marginTop: 16 }}>
+          <button
+            onClick={() => trigger.mutate()}
+            disabled={!token || trigger.isPending}
+            style={primaryBtn}
+          >
+            {trigger.isPending ? "Запускаю…" : "Запустить сейчас"}
+          </button>
+          {trigger.isError ? (
+            <span style={{ color: "#b3261e", marginLeft: 12 }}>
+              Ошибка: {String(trigger.error)}
+            </span>
+          ) : null}
+          {trigger.isSuccess ? (
+            <span style={{ color: "#0a7a3a", marginLeft: 12 }}>
+              Поставлено в очередь: {trigger.data.task_id}
+            </span>
+          ) : null}
+        </div>
+      </section>
+
+      <section style={{ ...cardStyle, marginTop: 16 }}>
+        <h2 style={{ marginTop: 0, fontSize: 16 }}>Последние прогоны</h2>
+        {!token ? (
+          <p style={{ color: "#5b6066" }}>
+            Введите admin token чтобы увидеть журнал.
+          </p>
+        ) : runs.isLoading ? (
+          <p>Загрузка…</p>
+        ) : runs.isError ? (
+          <p style={{ color: "#b3261e" }}>Ошибка: {String(runs.error)}</p>
+        ) : (
+          <table
+            style={{ width: "100%", borderCollapse: "collapse", fontSize: 13 }}
+          >
+            <thead>
+              <tr style={{ background: "#f6f7f9" }}>
+                {[
+                  "#",
+                  "Старт",
+                  "Регион",
+                  "Dev",
+                  "Status",
+                  "Объектов",
+                  "Квартир",
+                  "Req",
+                  "Ошибка",
+                ].map((h) => (
+                  <th key={h} style={th}>
+                    {h}
+                  </th>
+                ))}
+              </tr>
+            </thead>
+            <tbody>
+              {(runs.data ?? []).map((r) => (
+                <tr
+                  key={r.run_id}
+                  style={{ borderBottom: "1px solid #eef0f3" }}
+                >
+                  <td style={td}>{r.run_id}</td>
+                  <td style={td}>
+                    {r.started_at?.slice(0, 19).replace("T", " ")}
+                  </td>
+                  <td style={td}>{r.region_codes.join(",")}</td>
+                  <td style={td}>{r.developer_ids?.join(",") || "—"}</td>
+                  <td
+                    style={{
+                      ...td,
+                      color:
+                        r.status === "done"
+                          ? "#0a7a3a"
+                          : r.status === "failed"
+                            ? "#b3261e"
+                            : "#9a6700",
+                      fontWeight: 600,
+                    }}
+                  >
+                    {r.status}
+                  </td>
+                  <td style={td}>{r.objects_count ?? "—"}</td>
+                  <td style={td}>{r.flats_count ?? "—"}</td>
+                  <td style={td}>{r.requests_count ?? "—"}</td>
+                  <td
+                    style={{
+                      ...td,
+                      color: "#b3261e",
+                      maxWidth: 240,
+                      overflow: "hidden",
+                      textOverflow: "ellipsis",
+                      whiteSpace: "nowrap",
+                    }}
+                  >
+                    {r.error ?? ""}
+                  </td>
+                </tr>
+              ))}
+            </tbody>
+          </table>
+        )}
+      </section>
+    </main>
+  );
+}
+
+const cardStyle = {
+  background: "#fff",
+  border: "1px solid #e6e8ec",
+  borderRadius: 12,
+  padding: 20,
+};
+const labelStyle = {
+  display: "block",
+  fontSize: 12,
+  color: "#5b6066",
+  marginBottom: 4,
+  textTransform: "uppercase" as const,
+  letterSpacing: 0.4,
+};
+const inputStyle = {
+  padding: "8px 10px",
+  border: "1px solid #d1d5db",
+  borderRadius: 6,
+  fontSize: 14,
+  width: "100%",
+  boxSizing: "border-box" as const,
+};
+const primaryBtn = {
+  padding: "10px 18px",
+  background: "#1d4ed8",
+  color: "#fff",
+  border: "none",
+  borderRadius: 6,
+  fontSize: 14,
+  fontWeight: 600,
+  cursor: "pointer",
+};
+const th = {
+  padding: "8px 10px",
+  textAlign: "left" as const,
+  fontWeight: 600,
+  borderBottom: "1px solid #e6e8ec",
+};
+const td = { padding: "8px 10px" };