chore(tradein): деактивировать N1.ru scraper — источник переехал в Cian

N1.ru перестал быть самостоятельным источником (переехал в Cian, подтверждено
владельцем 2026-06). Активный скрап прекращён; 389 исторических listings в БД
сохраняются без изменений.

Убрано:
- app/services/scrapers/n1.py (удалён)
- run_n1_city_sweep + N1CitySweepCounters + N1_SWEEP_MAX_CONSECUTIVE_FAILURES
  из scrape_pipeline.py
- trigger_n1_city_sweep_run + dispatch elif в scheduler.py
- POST /scrape/n1 + GET /scrape/n1/runs endpoints из admin.py
- N1Scraper из scrape_around dispatch dict и ScrapeRequest.multi_room_n1
- "n1": 5.0 из scraper_settings._DEFAULT_DELAY_BY_SOURCE
- tests/test_n1_city_sweep.py, tests/test_n1_floor_parse.py (удалены)

Добавлено:
- data/sql/114_disable_n1_sweep.sql: UPDATE scrape_schedules SET enabled=false
  WHERE source='n1_city_sweep' (idempotent, строка сохраняется для audit-истории)

Сохранено (историческое):
- schemas/search.py Literal[..."n1"] — API-фильтр исторических listings
- config.py scrape_allowed_hosts n1.ru / ekaterinburg.n1.ru
- exporters/trade_in_pdf.py — цвет/лейбл N1 для отображения старых записей
- geocoder.py и admin.py docstring N1-комментарии где не активный код

1802 passed, 2 deselected (test_search_cache_hit, test_cian_valuation.cache_hit)

tradein-mvp/backend/app/api/v1/admin.py

@@ -27,7 +27,6 @@ from app.services.scrape_pipeline import (
     run_avito_city_sweep,
     run_cian_city_sweep,
     run_cian_full_load,
-    run_n1_city_sweep,
     run_yandex_city_sweep,
     run_yandex_full_load,
 )
@@ -44,7 +43,6 @@ from app.services.scrapers.avito_imv import (
 from app.services.scrapers.base import save_listings
 from app.services.scrapers.browser_fetcher import BrowserFetcher
 from app.services.scrapers.cian import CianScraper
-from app.services.scrapers.n1 import N1Scraper
 from app.services.scrapers.yandex_detail import YandexDetailScraper
 from app.services.scrapers.yandex_newbuilding import YandexNewbuildingScraper
 from app.services.scrapers.yandex_realty import YandexRealtyScraper
@@ -67,14 +65,14 @@ def _assert_allowed_url(url: str) -> None:
         raise HTTPException(status_code=400, detail="host not allowed")
 
 
-_ALL_SOURCES = ["avito", "cian", "yandex", "n1"]
+_ALL_SOURCES = ["avito", "cian", "yandex"]
 
 
 class ScrapeRequest(BaseModel):
     lat: float
     lon: float
     radius_m: int = Field(default=1000, ge=100, le=20000)
-    sources: list[Literal["avito", "cian", "yandex", "n1"]] = Field(
+    sources: list[Literal["avito", "cian", "yandex"]] = Field(
         default_factory=lambda: list(_ALL_SOURCES)
     )
     # Если True — скрейп Yandex по 5 сегментам комнат, не одним общим запросом.
@@ -83,8 +81,6 @@ class ScrapeRequest(BaseModel):
     deep_yandex: bool = False
     # Если True — скрейп Cian по 4 сегментам комнат отдельно (~4× лотов).
     multi_room_cian: bool = False
-    # Если True — скрейп N1 по 4 сегментам комнат отдельно (~4× лотов).
-    multi_room_n1: bool = False
 
 
 class ScrapeResult(BaseModel):
@@ -120,7 +116,6 @@ async def scrape_around(
             "avito": AvitoScraper,
             "cian": CianScraper,
             "yandex": YandexRealtyScraper,
-            "n1": N1Scraper,
         }.get(source)
         if scraper_cls is None:
             continue
@@ -141,10 +136,6 @@ async def scrape_around(
                 lots = await scraper.fetch_around_multi_room(
                     payload.lat, payload.lon, payload.radius_m
                 )
-            elif source == "n1" and payload.multi_room_n1:
-                lots = await scraper.fetch_around_multi_room(
-                    payload.lat, payload.lon, payload.radius_m
-                )
             else:
                 lots = await scraper.fetch_around(payload.lat, payload.lon, payload.radius_m)
         inserted, updated = save_listings(db, lots)
@@ -792,72 +783,6 @@ def cancel_avito_city_sweep(
     return {"ok": True, "run_id": run_id, "cancelled": cancelled}
 
 
-# ── N1.ru city sweep (#575) — on-demand trigger + runs list ──────────────────
-
-
-@router.post("/scrape/n1", response_model=CitySweepStartResponse)
-async def start_n1_city_sweep(
-    payload: CitySweepStartRequest,
-    background_tasks: BackgroundTasks,
-    db: Annotated[Session, Depends(get_db)],
-) -> CitySweepStartResponse:
-    """Запустить N1.ru ЕКБ city sweep в background (#575). Returns run_id.
-
-    Зеркало avito-city-sweep, но N1 — простой SERP (search → save), без
-    houses/detail-enrich: поля enrich_houses/detail_top_n из payload игнорируются.
-    Помимо этого эндпоинта N1 крутится по расписанию (scrape_schedules
-    source='n1_city_sweep', seed 084).
-    """
-    run_id = runs_mod.create_run(db, source="n1_city_sweep", params=payload.model_dump())
-
-    async def _sweep_task() -> None:
-        sweep_db = SessionLocal()
-        try:
-            await run_n1_city_sweep(
-                sweep_db,
-                run_id=run_id,
-                radius_m=payload.radius_m,
-                pages_per_anchor=payload.pages_per_anchor,
-                request_delay_sec=payload.request_delay_sec,
-            )
-        except Exception:
-            logger.exception("n1-sweep background task run_id=%d crashed", run_id)
-        finally:
-            sweep_db.close()
-
-    background_tasks.add_task(_sweep_task)
-    logger.info("n1-sweep queued run_id=%d params=%s", run_id, payload.model_dump())
-    return CitySweepStartResponse(
-        run_id=run_id,
-        status="running",
-        pages_per_anchor=payload.pages_per_anchor,
-        detail_top_n=0,  # N1 sweep не имеет detail-шага
-    )
-
-
-@router.get("/scrape/n1/runs", response_model=list[ScrapeRunRow])
-def list_n1_city_sweep_runs(
-    db: Annotated[Session, Depends(get_db)],
-    limit: int = 10,
-) -> list[ScrapeRunRow]:
-    """Список последних N N1-sweep runs (для UI polling). Default limit=10."""
-    rows = runs_mod.list_recent(db, source="n1_city_sweep", limit=limit)
-    return [
-        ScrapeRunRow(
-            run_id=r["run_id"],
-            source=r["source"],
-            status=r["status"],
-            params=r.get("params"),
-            counters=r.get("counters"),
-            error=r.get("error"),
-            started_at=r["started_at"].isoformat() if r.get("started_at") else None,
-            finished_at=r["finished_at"].isoformat() if r.get("finished_at") else None,
-            heartbeat_at=r["heartbeat_at"].isoformat() if r.get("heartbeat_at") else None,
-        )
-        for r in rows
-    ]
-
-
 # ── Cian city sweep (#860) — on-demand trigger + runs list ───────────────────
 
 

tradein-mvp/backend/app/services/scheduler.py

@@ -10,7 +10,6 @@
 Sources:
   - avito_city_sweep        → run_avito_city_sweep (scrape_pipeline.py)
   - yandex_city_sweep       → run_yandex_city_sweep (scrape_pipeline.py, #561; shipped DORMANT)
-  - n1_city_sweep           → run_n1_city_sweep (scrape_pipeline.py, #575)
   - cian_history_backfill   → run_cian_backfill (this module, #560)
   - rosreestr_dkp_import    → import_rosreestr_dkp (this module, #563)
   - listing_source_snapshot → snapshot_listing_sources (tasks/listing_source_snapshot.py, #570;
@@ -81,7 +80,6 @@ from app.services import scrape_runs as runs_mod
 from app.services.scrape_pipeline import (
     run_avito_city_sweep,
     run_cian_city_sweep,
-    run_n1_city_sweep,
     run_yandex_city_sweep,
 )
 
@@ -333,39 +331,6 @@ async def trigger_yandex_city_sweep_run(db: Session, schedule_row: dict[str, Any
     return run_id
 
 
-async def trigger_n1_city_sweep_run(db: Session, schedule_row: dict[str, Any]) -> int | None:
-    """Создать новый scrape_runs + launch run_n1_city_sweep в asyncio.create_task (#575).
-
-    Зеркало trigger_yandex_city_sweep_run — N1 sweep простой (search → save, без
-    houses/detail). Returns run_id (или None если skip — есть running run).
-    """
-    run_id = _claim_run(db, schedule_row)
-    if run_id is None:
-        return None
-
-    params = schedule_row.get("default_params") or {}
-
-    async def _run() -> None:
-        run_db = SessionLocal()
-        try:
-            await run_n1_city_sweep(
-                run_db,
-                run_id=run_id,
-                pages_per_anchor=int(params.get("pages_per_anchor", 2)),
-                request_delay_sec=float(params.get("request_delay_sec", 7.0)),
-                radius_m=int(params.get("radius_m", 1500)),
-            )
-        except Exception:
-            logger.exception("scheduler: run_n1_city_sweep crashed run_id=%d", run_id)
-        finally:
-            run_db.close()
-
-    task = asyncio.create_task(_run())
-    task.add_done_callback(lambda t: t.exception() if not t.cancelled() else None)
-    logger.info("scheduler: triggered n1_city_sweep run_id=%d", run_id)
-    return run_id
-
-
 async def trigger_cian_backfill_run(db: Session, schedule_row: dict[str, Any]) -> int | None:
     """Создать scrape_runs + launch run_cian_backfill в asyncio.create_task.
 
@@ -1289,8 +1254,6 @@ async def scheduler_loop() -> None:
                         await trigger_avito_city_sweep_run(db, sch)
                     elif source == "yandex_city_sweep":
                         await trigger_yandex_city_sweep_run(db, sch)
-                    elif source == "n1_city_sweep":
-                        await trigger_n1_city_sweep_run(db, sch)
                     elif source == "cian_history_backfill":
                         await trigger_cian_backfill_run(db, sch)
                     elif source == "cian_city_sweep":

tradein-mvp/backend/app/services/scrape_pipeline.py

@@ -46,7 +46,6 @@ from app.services.scrapers.avito_exceptions import AvitoBlockedError, AvitoRateL
 from app.services.scrapers.avito_houses import fetch_house_catalog, save_house_catalog_enrichment
 from app.services.scrapers.base import ScrapedLot, save_listings
 from app.services.scrapers.browser_fetcher import BrowserFetcher
-from app.services.scrapers.n1 import N1Scraper
 from app.services.scrapers.yandex_realty import YandexRealtyScraper
 
 logger = logging.getLogger(__name__)
@@ -1037,155 +1036,6 @@ async def run_yandex_city_sweep(
         raise
 
 
-# ── N1.ru city sweep (#575) ─────────────────────────────────────────
-
-
-@dataclass
-class N1CitySweepCounters:
-    """Aggregate counters для N1.ru city sweep run.
-
-    Как Yandex: N1 SERP — простой search → save, без house-catalog / detail-enrich
-    шага, поэтому houses/detail-полей нет.
-    """
-
-    anchors_total: int = 0
-    anchors_done: int = 0
-    pages_fetched: int = 0
-    lots_fetched: int = 0
-    lots_inserted: int = 0
-    lots_updated: int = 0
-    errors_count: int = 0
-
-    def to_dict(self) -> dict[str, int]:
-        return {f.name: getattr(self, f.name) for f in fields(self)}
-
-
-# Defensive abort: N1Scraper.fetch_around глотает ошибки и возвращает [] (нет
-# block/rate exception-классов). Прерываем sweep после N подряд anchor'ов,
-# упавших с исключением — сигнал системного блока IP. Пустой результат (0 lots)
-# — НЕ ошибка, счётчик не растёт (зеркало yandex).
-N1_SWEEP_MAX_CONSECUTIVE_FAILURES = 3
-
-
-async def run_n1_city_sweep(
-    db: Session,
-    *,
-    run_id: int,
-    anchors: list[tuple[float, float, str]] | None = None,
-    radius_m: int = 1500,
-    pages_per_anchor: int = 2,
-    request_delay_sec: float | None = None,
-) -> N1CitySweepCounters:
-    """N1.ru city sweep: iterate anchors × pages → save (#575).
-
-    Зеркало run_yandex_city_sweep — N1 это региональный SERP-портал ЕКБ без
-    house-catalog / detail-enrich шага:
-      per anchor: N1Scraper().fetch_around(lat, lon, radius_m, page=p)
-      → save_listings(db, lots) → accumulate counters.
-
-    Anti-bot / cancel семантика (зеркало yandex/avito sweep):
-      - Cooperative cancel: is_cancelled(db, run_id) перед каждым anchor.
-      - N1Scraper не определяет block/rate exception-классов (fetch_around ловит
-        ошибки внутри → []). Per-anchor ловим broad Exception, логируем, continue;
-        защитно прерываем (early abort + mark_banned) при N подряд исключениях.
-      - request_delay_sec применяется как доп. asyncio.sleep МЕЖДУ anchor'ами
-        (N1Scraper сам спит self.request_delay_sec между страницами).
-    """
-    _anchors = anchors if anchors is not None else EKB_ANCHORS
-    counters = N1CitySweepCounters(anchors_total=len(_anchors))
-    inter_anchor_delay = request_delay_sec if request_delay_sec is not None else 7.0
-    consecutive_failures = 0
-
-    try:
-        for idx, (lat, lon, name) in enumerate(_anchors, start=1):
-            if scrape_runs.is_cancelled(db, run_id):
-                logger.info(
-                    "n1-sweep run_id=%d: cancelled at anchor #%d/%d (%s)",
-                    run_id,
-                    idx,
-                    len(_anchors),
-                    name,
-                )
-                scrape_runs.update_heartbeat(db, run_id, counters.to_dict())
-                return counters
-
-            logger.info(
-                "n1-sweep run_id=%d anchor #%d/%d (%s, %.4f, %.4f)",
-                run_id,
-                idx,
-                len(_anchors),
-                name,
-                lat,
-                lon,
-            )
-            try:
-                anchor_lots: list[ScrapedLot] = []
-                async with N1Scraper() as scraper:
-                    for page in range(1, pages_per_anchor + 1):
-                        page_lots = await scraper.fetch_around(lat, lon, radius_m, page=page)
-                        counters.pages_fetched += 1
-                        if not page_lots:
-                            break
-                        anchor_lots.extend(page_lots)
-
-                counters.lots_fetched += len(anchor_lots)
-                if anchor_lots:
-                    inserted, updated = save_listings(db, anchor_lots, run_id=run_id)
-                    counters.lots_inserted += inserted
-                    counters.lots_updated += updated
-                consecutive_failures = 0
-            except Exception as e:
-                logger.exception("n1-sweep run_id=%d: anchor %s failed", run_id, name)
-                counters.errors_count += 1
-                consecutive_failures += 1
-                if consecutive_failures >= N1_SWEEP_MAX_CONSECUTIVE_FAILURES:
-                    logger.error(
-                        "n1-sweep run_id=%d ABORT at anchor #%d/%d (%s) — "
-                        "%d consecutive failures (IP likely blocked): %s",
-                        run_id,
-                        idx,
-                        len(_anchors),
-                        name,
-                        consecutive_failures,
-                        e,
-                    )
-                    counters.anchors_done = idx
-                    scrape_runs.update_heartbeat(db, run_id, counters.to_dict())
-                    scrape_runs.mark_banned(
-                        db,
-                        run_id,
-                        f"n1 sweep aborted: {consecutive_failures} consecutive "
-                        f"anchor failures (last: {e})",
-                        counters.to_dict(),
-                    )
-                    return counters
-
-            counters.anchors_done = idx
-            scrape_runs.update_heartbeat(db, run_id, counters.to_dict())
-
-            if idx < len(_anchors):
-                await asyncio.sleep(inter_anchor_delay)
-
-        scrape_runs.mark_done(db, run_id, counters.to_dict())
-        logger.info(
-            "n1-sweep run_id=%d done: anchors=%d/%d pages=%d lots=%d (ins=%d/upd=%d) errors=%d",
-            run_id,
-            counters.anchors_done,
-            counters.anchors_total,
-            counters.pages_fetched,
-            counters.lots_fetched,
-            counters.lots_inserted,
-            counters.lots_updated,
-            counters.errors_count,
-        )
-        return counters
-
-    except Exception as exc:
-        logger.exception("n1-sweep run_id=%d: fatal error", run_id)
-        scrape_runs.mark_failed(db, run_id, str(exc), counters.to_dict())
-        raise
-
-
 # ── Cian city sweep (#860) ──────────────────────────────────────────────────
 
 

tradein-mvp/backend/app/services/scraper_settings.py

@@ -32,7 +32,6 @@ _DEFAULT_DELAY_BY_SOURCE: dict[str, float] = {
     "avito": 7.0,
     "cian": 5.0,
     "domklik": 8.0,
-    "n1": 5.0,
     "yandex": 5.0,
     "yandex_detail": 5.0,
     "yandex_newbuilding": 5.0,

tradein-mvp/backend/app/services/scrapers/n1.py

@@ -1,271 +0,0 @@
-"""N1.ru scraper — вторичка ЕКБ (ekaterinburg.n1.ru).
-
-N1.ru — региональный портал недвижимости (хорошо представлен на Урале).
-SSR HTML с DOM-карточками `data-test="offers-list-item"`.
-
-Использует curl_cffi (impersonate=chrome120) — как Cian/Avito.
-
-Карточка:
-  - a[href^="/view/NNN/"]  — ссылка на объявление + offer_id
-    текст ссылки = "1/11 2-к, Репина, 75/2 стр." (этаж/этажность + комнаты + адрес)
-    Этаж/этажность — leading-токен «N/M» перед сегментом комнат.
-  - [data-test="offers-list-item-param-total-area"]  — "79 м2"
-  - [data-test="offers-list-item-param-price"]       — "15 700 000"
-  - img.src — фото
-
-Координаты карточки не отдаёт → оставляем lat/lon = None,
-geocode-missing проставит реальные по адресу.
-"""
-
-from __future__ import annotations
-
-import logging
-import re
-from typing import Any
-
-from curl_cffi.requests import AsyncSession
-from selectolax.parser import HTMLParser
-
-from app.services.scraper_settings import get_scraper_delay
-from app.services.scrapers.base import BaseScraper, ScrapedLot
-
-logger = logging.getLogger(__name__)
-
-
-class N1Scraper(BaseScraper):
-    """N1.ru vtorichka parser. Источник = 'n1'."""
-
-    name = "n1"
-    base_url = "https://ekaterinburg.n1.ru"
-    # Класс-дефолт; реальное значение загружается из scraper_settings при создании экземпляра.
-    request_delay_sec = 5.0
-
-    def __init__(self) -> None:
-        super().__init__()
-        self.request_delay_sec = get_scraper_delay(self.name)
-        self._cffi: AsyncSession | None = None
-
-    async def __aenter__(self) -> N1Scraper:
-        await super().__aenter__()
-        self._cffi = AsyncSession(
-            impersonate="chrome120",
-            timeout=25,
-            headers={
-                "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
-                "Accept-Language": "ru-RU,ru;q=0.9,en;q=0.8",
-                "Sec-Fetch-Dest": "document",
-                "Sec-Fetch-Mode": "navigate",
-                "Sec-Fetch-Site": "none",
-                "Upgrade-Insecure-Requests": "1",
-            },
-        )
-        return self
-
-    async def __aexit__(self, *args: Any) -> None:
-        if self._cffi is not None:
-            await self._cffi.close()
-        await super().__aexit__(*args)
-
-    async def fetch_around(
-        self,
-        lat: float,
-        lon: float,
-        radius_m: int = 1000,
-        rooms: int | None = None,
-        page: int = 1,
-    ) -> list[ScrapedLot]:
-        self._anchor_lat = lat
-        self._anchor_lon = lon
-        self._anchor_radius_m = radius_m
-
-        url = self._build_url(rooms, page)
-        try:
-            assert self._cffi is not None
-            response = await self._cffi.get(url)
-        except Exception:
-            logger.exception("n1 fetch failed for %s", url)
-            return []
-        if response.status_code != 200:
-            logger.warning("n1 returned %d for %s", response.status_code, url)
-            return []
-
-        lots = self._parse_html(response.text)
-        logger.info(
-            "n1: %d lots around (%.4f, %.4f) rooms=%s page=%d",
-            len(lots),
-            lat,
-            lon,
-            rooms,
-            page,
-        )
-        await self.sleep_between_requests()
-        return lots
-
-    async def fetch_around_multi_room(
-        self,
-        lat: float,
-        lon: float,
-        radius_m: int = 1000,
-        pages: int = 8,
-    ) -> list[ScrapedLot]:
-        """Скрейп N1 по сегментам комнат 1/2/3/4 × N страниц — расширяет выборку."""
-        seen: dict[str, ScrapedLot] = {}
-        for rooms in (1, 2, 3, 4):
-            for page in range(1, pages + 1):
-                try:
-                    lots = await self.fetch_around(lat, lon, radius_m, rooms=rooms, page=page)
-                except Exception:
-                    logger.exception("n1 multi-room fetch failed rooms=%s page=%d", rooms, page)
-                    continue
-                if not lots:
-                    break  # пустая страница → дальше нет смысла
-                for lot in lots:
-                    key = lot.source_id or lot.source_url
-                    if key and key not in seen:
-                        seen[key] = lot
-        logger.info("n1 multi-room: %d unique lots around (%.4f, %.4f)", len(seen), lat, lon)
-        return list(seen.values())
-
-    def _build_url(self, rooms: int | None = None, page: int = 1) -> str:
-        """URL каталога вторички ЕКБ.
-
-        N1 фильтр комнат: /kupit/kvartiry/vtorichka/1-komnatnye/ итд.
-        Пагинация: ?page=N.
-        """
-        rooms_segment = {
-            1: "1-komnatnye/",
-            2: "2-komnatnye/",
-            3: "3-komnatnye/",
-            4: "4-komnatnye/",
-        }.get(rooms or 0, "")
-        url = f"{self.base_url}/kupit/kvartiry/vtorichka/{rooms_segment}"
-        if page > 1:
-            url += f"?page={page}"
-        return url
-
-    def _parse_html(self, html: str) -> list[ScrapedLot]:
-        tree = HTMLParser(html)
-        cards = tree.css('[data-test="offers-list-item"]')
-        return [lot for lot in (self._card_to_lot(c) for c in cards) if lot is not None]
-
-    def _card_to_lot(self, card: Any) -> ScrapedLot | None:
-        try:
-            # Ссылка на объявление /view/NNN/
-            link = card.css_first('a[href*="/view/"]')
-            if link is None:
-                return None
-            href = link.attributes.get("href", "")
-            m_id = re.search(r"/view/(\d+)/", href)
-            offer_id = m_id.group(1) if m_id else None
-            url = href if href.startswith("http") else f"{self.base_url}{href}"
-
-            # Заголовок ссылки: "1/11 2-к, Репина, 75/2 стр." (этаж/этажность + комнаты + адрес)
-            title = link.text(strip=True) or ""
-            floor, total_floors = _extract_floor_from_title(title)
-            rooms = _extract_rooms(title)
-            # Адрес = title без ведущего «N/M» и без префикса комнат
-            address = re.sub(
-                r"^\s*(\d+-к|студи[яи])\s*,?\s*", "", title, flags=re.IGNORECASE
-            ).strip()
-            if not address:
-                address = "Екатеринбург (N1)"
-
-            # Площадь
-            area_el = card.css_first('[data-test="offers-list-item-param-total-area"]')
-            area = _extract_area(area_el.text(strip=True) if area_el else "")
-
-            # Цена
-            price_el = card.css_first('[data-test="offers-list-item-param-price"]')
-            price = _extract_price(price_el.text(strip=True) if price_el else "")
-            if not price:
-                return None
-
-            # Координаты НЕ ставим — N1 card не отдаёт. Оставляем None,
-            # geocode-missing проставит реальные по address (N1 отдаёт всё ЕКБ,
-            # jitter вокруг якоря был бы фейком).
-            lat = lon = None
-
-            # Фото
-            photos: list[str] = []
-            for img in card.css("img")[:5]:
-                src = img.attributes.get("src", "")
-                if src and src.startswith("http") and src not in photos:
-                    photos.append(src)
-
-            return ScrapedLot(
-                source="n1",
-                source_url=url,
-                source_id=offer_id,
-                address=address,
-                lat=lat,
-                lon=lon,
-                rooms=rooms,
-                area_m2=area,
-                floor=floor,
-                total_floors=total_floors,
-                price_rub=price,
-                photo_urls=photos,
-                raw_payload={"title": title},
-            )
-        except Exception:
-            logger.exception("n1 card parse failed")
-            return None
-
-
-# ── Helpers ─────────────────────────────────────────────────────────────────
-
-_RE_ROOMS = re.compile(r"(\d+)\s*-?\s*к", re.IGNORECASE)
-_RE_STUDIO = re.compile(r"студи", re.IGNORECASE)
-_RE_AREA = re.compile(r"([\d]+(?:[.,]\d+)?)")
-_RE_DIGITS = re.compile(r"\d")
-
-# Ведущий токен «N/M» в тексте карточки N1: «1/11 2-к, Репина, 75/2 стр.»
-# Якорь ^ гарантирует: только если N/M стоит В НАЧАЛЕ строки (до сегмента комнат).
-# Дробь в адресе вида «75/2» в середине строки НЕ совпадёт.
-_RE_FLOOR_LEADING = re.compile(r"^\s*(\d{1,3})\s*/\s*(\d{1,3})\b")
-
-
-def _extract_rooms(text: str) -> int | None:
-    if _RE_STUDIO.search(text):
-        return 0
-    m = _RE_ROOMS.search(text)
-    return int(m.group(1)) if m else None
-
-
-def _extract_area(text: str) -> float | None:
-    m = _RE_AREA.search(text)
-    if m:
-        try:
-            return float(m.group(1).replace(",", "."))
-        except ValueError:
-            return None
-    return None
-
-
-def _extract_price(text: str) -> int | None:
-    digits = "".join(_RE_DIGITS.findall(text))
-    return int(digits) if digits else None
-
-
-def _extract_floor_from_title(text: str) -> tuple[int | None, int | None]:
-    """Извлекает этаж и этажность из ведущего токена «N/M» в тексте карточки N1.
-
-    N1 SERP отдаёт текст вида «1/11 2-к, Репина, 75/2 стр.» — этаж/этажность
-    идут первыми, до сегмента комнат. Регулярка привязана к началу строки (^),
-    поэтому дробь в номере дома («75/2») в середине строки не захватывается.
-
-    Sanity-проверки:
-      - floor >= 1
-      - total_floors >= floor
-      - total_floors <= 100
-
-    Возвращает (None, None) при отсутствии совпадения или нарушении диапазона.
-    """
-    m = _RE_FLOOR_LEADING.match(text)
-    if m is None:
-        return None, None
-    floor = int(m.group(1))
-    total_floors = int(m.group(2))
-    if floor < 1 or total_floors < floor or total_floors > 100:
-        return None, None
-    return floor, total_floors

tradein-mvp/backend/data/sql/114_disable_n1_sweep.sql

@@ -0,0 +1,18 @@
+-- Миграция 114: деактивировать N1.ru city sweep.
+--
+-- N1.ru — региональный портал ЕКБ — переехал в Cian (подтверждено владельцем 2026-06).
+-- Активный скрап прекращается. Строку НЕ удаляем — нужна для audit-истории run'ов.
+-- Исторические listings source='n1' (389 записей в БД) сохраняются без изменений.
+
+BEGIN;
+
+UPDATE scrape_schedules
+SET    enabled = false
+WHERE  source  = 'n1_city_sweep'
+  AND  enabled = true;  -- idempotent: повторный запуск не трогает уже-disabled
+
+COMMENT ON TABLE scrape_schedules IS
+    'Периодические задачи: avito/yandex/cian city sweep, rosreestr import, '
+    'listing_source_snapshot. N1 city sweep деактивирован (2026-06, переехал в cian).';
+
+COMMIT;

tradein-mvp/backend/tests/test_n1_city_sweep.py

@@ -1,228 +0,0 @@
-"""Offline unit tests for run_n1_city_sweep (#575).
-
-Pure & fast: NO live network, NO DB. Mirrors test_yandex_city_sweep — monkeypatch
-N1Scraper lifecycle + fetch_around, scrape_pipeline.save_listings, scrape_runs.*.
-Asserts orchestration semantics only (live N1 behavior verified by post-deploy QA).
-
-NB: run_n1_city_sweep пагинирует page=1..pages_per_anchor (1-based), в отличие от
-yandex (0-based) — тесты учитывают это.
-"""
-
-import os
-
-os.environ.setdefault("DATABASE_URL", "postgresql+psycopg://test:test@localhost:5432/test")
-
-from typing import Any
-
-import pytest
-
-from app.services import scrape_pipeline
-from app.services.scrapers.base import ScrapedLot
-from app.services.scrapers.n1 import N1Scraper
-
-TEST_ANCHORS: list[tuple[float, float, str]] = [
-    (56.8400, 60.6050, "A1"),
-    (56.7950, 60.5300, "A2"),
-    (56.8970, 60.6100, "A3"),
-]
-
-
-def _fake_lot(offer_id: str) -> ScrapedLot:
-    return ScrapedLot(
-        source="n1",
-        source_url=f"https://ekb.n1.ru/view/{offer_id}/",
-        source_id=offer_id,
-        address="Екатеринбург, улица Тестовая, 1",
-        price_rub=5_000_000,
-        area_m2=50.0,
-        rooms=2,
-    )
-
-
-class _FakeRuns:
-    def __init__(self, *, cancel_at_anchor: int | None = None) -> None:
-        self.cancel_at_anchor = cancel_at_anchor
-        self.is_cancelled_calls = 0
-        self.heartbeats: list[dict[str, int]] = []
-        self.done: dict[str, int] | None = None
-        self.failed: tuple[str, dict[str, int]] | None = None
-        self.banned: tuple[str, dict[str, int]] | None = None
-
-    def is_cancelled(self, _db: Any, _run_id: int) -> bool:
-        self.is_cancelled_calls += 1
-        if self.cancel_at_anchor is None:
-            return False
-        return self.is_cancelled_calls >= self.cancel_at_anchor
-
-    def update_heartbeat(self, _db: Any, _run_id: int, counters: dict[str, int]) -> None:
-        self.heartbeats.append(dict(counters))
-
-    def mark_done(self, _db: Any, _run_id: int, counters: dict[str, int]) -> None:
-        self.done = dict(counters)
-
-    def mark_failed(self, _db: Any, _run_id: int, error: str, counters: dict[str, int]) -> None:
-        self.failed = (error, dict(counters))
-
-    def mark_banned(self, _db: Any, _run_id: int, error: str, counters: dict[str, int]) -> None:
-        self.banned = (error, dict(counters))
-
-
-@pytest.fixture(autouse=True)
-def _no_sleep(monkeypatch: pytest.MonkeyPatch) -> None:
-    async def _instant(_secs: float) -> None:
-        return None
-
-    monkeypatch.setattr(scrape_pipeline.asyncio, "sleep", _instant)
-
-
-@pytest.fixture(autouse=True)
-def _stub_scraper_lifecycle(monkeypatch: pytest.MonkeyPatch) -> None:
-    """Neutralize __init__/__aenter__/__aexit__ — no get_scraper_delay DB, no curl_cffi."""
-
-    def _init(self: N1Scraper) -> None:
-        self.request_delay_sec = 0.0
-        self._cffi = None
-
-    async def _aenter(self: N1Scraper) -> N1Scraper:
-        return self
-
-    async def _aexit(self: N1Scraper, *_args: Any) -> None:
-        return None
-
-    monkeypatch.setattr(N1Scraper, "__init__", _init)
-    monkeypatch.setattr(N1Scraper, "__aenter__", _aenter)
-    monkeypatch.setattr(N1Scraper, "__aexit__", _aexit)
-
-
-def _install_runs(monkeypatch: pytest.MonkeyPatch, fake: _FakeRuns) -> None:
-    monkeypatch.setattr(scrape_pipeline.scrape_runs, "is_cancelled", fake.is_cancelled)
-    monkeypatch.setattr(scrape_pipeline.scrape_runs, "update_heartbeat", fake.update_heartbeat)
-    monkeypatch.setattr(scrape_pipeline.scrape_runs, "mark_done", fake.mark_done)
-    monkeypatch.setattr(scrape_pipeline.scrape_runs, "mark_failed", fake.mark_failed)
-    monkeypatch.setattr(scrape_pipeline.scrape_runs, "mark_banned", fake.mark_banned)
-
-
-async def test_sweep_iterates_all_anchors_and_aggregates(monkeypatch: pytest.MonkeyPatch) -> None:
-    fetch_calls: list[tuple[float, float, int]] = []
-    save_calls: list[int] = []
-
-    async def fake_fetch(self, lat, lon, radius_m=1000, rooms=None, page=1):
-        fetch_calls.append((lat, lon, page))
-        return [_fake_lot(f"{lat}-{page}-{i}") for i in range(2)]
-
-    def fake_save(_db, lots, *, run_id=None):
-        save_calls.append(len(lots))
-        return len(lots), 0
-
-    monkeypatch.setattr(N1Scraper, "fetch_around", fake_fetch)
-    monkeypatch.setattr(scrape_pipeline, "save_listings", fake_save)
-    fake = _FakeRuns()
-    _install_runs(monkeypatch, fake)
-
-    counters = await scrape_pipeline.run_n1_city_sweep(
-        db=object(), run_id=1, anchors=TEST_ANCHORS, pages_per_anchor=2, request_delay_sec=0.0
-    )
-
-    assert len(fetch_calls) == len(TEST_ANCHORS) * 2
-    assert [p for _, _, p in fetch_calls[:2]] == [1, 2]  # 1-based pagination
-    assert len(save_calls) == len(TEST_ANCHORS)
-    assert counters.anchors_done == len(TEST_ANCHORS)
-    assert counters.pages_fetched == len(TEST_ANCHORS) * 2
-    assert counters.lots_fetched == 12
-    assert counters.lots_inserted == 12
-    assert counters.errors_count == 0
-    assert fake.done is not None and fake.done["lots_fetched"] == 12
-    assert fake.failed is None and fake.banned is None
-
-
-async def test_save_listings_receives_run_id(monkeypatch: pytest.MonkeyPatch) -> None:
-    seen: list[int | None] = []
-
-    async def fake_fetch(self, lat, lon, radius_m=1000, rooms=None, page=1):
-        return [_fake_lot(f"{lat}-{page}")]
-
-    def fake_save(_db, lots, *, run_id=None):
-        seen.append(run_id)
-        return len(lots), 0
-
-    monkeypatch.setattr(N1Scraper, "fetch_around", fake_fetch)
-    monkeypatch.setattr(scrape_pipeline, "save_listings", fake_save)
-    _install_runs(monkeypatch, _FakeRuns())
-
-    await scrape_pipeline.run_n1_city_sweep(
-        db=object(), run_id=777, anchors=TEST_ANCHORS, pages_per_anchor=1
-    )
-    assert seen == [777, 777, 777]
-
-
-async def test_empty_page_stops_pagination(monkeypatch: pytest.MonkeyPatch) -> None:
-    pages_seen: list[int] = []
-
-    async def fake_fetch(self, lat, lon, radius_m=1000, rooms=None, page=1):
-        pages_seen.append(page)
-        if page == 1:
-            return [_fake_lot(f"{lat}-1")]
-        return []  # page 2 empty → stop
-
-    monkeypatch.setattr(N1Scraper, "fetch_around", fake_fetch)
-    monkeypatch.setattr(
-        scrape_pipeline, "save_listings", lambda _db, lots, *, run_id=None: (len(lots), 0)
-    )
-    fake = _FakeRuns()
-    _install_runs(monkeypatch, fake)
-
-    counters = await scrape_pipeline.run_n1_city_sweep(
-        db=object(), run_id=2, anchors=TEST_ANCHORS, pages_per_anchor=5
-    )
-    assert pages_seen == [1, 2, 1, 2, 1, 2]
-    assert counters.pages_fetched == 6
-    assert counters.lots_fetched == 3
-    assert counters.anchors_done == 3
-
-
-async def test_cancel_midway_stops_remaining(monkeypatch: pytest.MonkeyPatch) -> None:
-    fetch_calls: list[str] = []
-
-    async def fake_fetch(self, lat, lon, radius_m=1000, rooms=None, page=1):
-        fetch_calls.append(f"{lat}")
-        return [_fake_lot(f"{lat}-{page}")]
-
-    monkeypatch.setattr(N1Scraper, "fetch_around", fake_fetch)
-    monkeypatch.setattr(
-        scrape_pipeline, "save_listings", lambda _db, lots, *, run_id=None: (len(lots), 0)
-    )
-    fake = _FakeRuns(cancel_at_anchor=2)
-    _install_runs(monkeypatch, fake)
-
-    counters = await scrape_pipeline.run_n1_city_sweep(
-        db=object(), run_id=3, anchors=TEST_ANCHORS, pages_per_anchor=1
-    )
-    assert len(fetch_calls) == 1
-    assert counters.anchors_done == 1
-    assert fake.done is None and fake.banned is None
-
-
-async def test_consecutive_failures_abort_sweep(monkeypatch: pytest.MonkeyPatch) -> None:
-    attempts: list[str] = []
-
-    async def fake_fetch(self, lat, lon, radius_m=1000, rooms=None, page=1):
-        attempts.append(f"{lat}")
-        raise RuntimeError("n1 blocked")
-
-    monkeypatch.setattr(N1Scraper, "fetch_around", fake_fetch)
-    monkeypatch.setattr(
-        scrape_pipeline, "save_listings", lambda _db, lots, *, run_id=None: (len(lots), 0)
-    )
-    fake = _FakeRuns()
-    _install_runs(monkeypatch, fake)
-
-    anchors = [*TEST_ANCHORS, (56.77, 60.55, "A4"), (56.865, 60.62, "A5")]
-    counters = await scrape_pipeline.run_n1_city_sweep(
-        db=object(), run_id=5, anchors=anchors, pages_per_anchor=1
-    )
-    n = scrape_pipeline.N1_SWEEP_MAX_CONSECUTIVE_FAILURES
-    assert len(attempts) == n
-    assert counters.errors_count == n
-    assert counters.anchors_done == n
-    assert fake.banned is not None and "consecutive" in fake.banned[0]
-    assert fake.done is None

tradein-mvp/backend/tests/test_n1_floor_parse.py

@@ -1,169 +0,0 @@
-"""Unit tests for N1 floor/total_floors extraction (#795).
-
-Tests _extract_floor_from_title() directly (pure function, no network/DB).
-Also covers _card_to_lot integration via a minimal fake card HTML node.
-"""
-
-from __future__ import annotations
-
-import pytest
-
-from app.services.scrapers.n1 import _extract_floor_from_title
-
-# ── _extract_floor_from_title: direct unit tests ────────────────────────────
-
-
-@pytest.mark.parametrize(
-    "text, expected_floor, expected_total",
-    [
-        # Typical N1 card: floor 1 of 11
-        ("1/11 2-к, Репина, 75/2 стр.", 1, 11),
-        # Floor in middle of building
-        ("5/9 3-к, Ленина, 10 стр.", 5, 9),
-        # Top floor
-        ("17/17 1-к, Малышева, 12", 17, 17),
-        # Leading whitespace tolerated
-        ("  3/16 студия, Вайнера, 5", 3, 16),
-        # Three-digit total floors (edge of range)
-        ("2/100 2-к, Тверитина, 8", 2, 100),
-    ],
-)
-def test_extract_floor_happy_path(text: str, expected_floor: int, expected_total: int) -> None:
-    floor, total = _extract_floor_from_title(text)
-    assert floor == expected_floor
-    assert total == expected_total
-
-
-@pytest.mark.parametrize(
-    "text",
-    [
-        # No leading N/M at all — classic old format without floor token
-        ("2-к, Репина, 75/2 стр."),
-        # House number fraction in middle — must NOT be parsed as floor
-        ("Репина, 75/2 стр."),
-        # N/M not at string start (preceded by other text)
-        ("квартира 3/9, Ленина, 1"),
-        # floor=0 is invalid
-        ("0/9 1-к, Тестовая, 1"),
-        # total_floors < floor — sanity fail
-        ("9/5 2-к, Тестовая, 1"),
-        # total_floors > 100 — sanity fail
-        ("2/101 1-к, Тестовая, 1"),
-        # Empty string
-        (""),
-        # Only digits, no slash
-        ("5 2-к, Тестовая, 1"),
-    ],
-)
-def test_extract_floor_returns_none(text: str) -> None:
-    floor, total = _extract_floor_from_title(text)
-    assert floor is None
-    assert total is None
-
-
-def test_house_number_fraction_not_misparsed() -> None:
-    """«75/2 стр.» appearing after rooms segment must not match floor regex."""
-    # The fraction 75/2 comes AFTER the rooms prefix, not at string start.
-    floor, total = _extract_floor_from_title("3-к, Репина, 75/2 стр.")
-    assert floor is None
-    assert total is None
-
-
-def test_non_leading_fraction_after_comma() -> None:
-    """Fraction embedded mid-string via comma-separated address token."""
-    floor, total = _extract_floor_from_title("ул. 8 Марта, 51/3, 2-к")
-    assert floor is None
-    assert total is None
-
-
-# ── Integration: _card_to_lot via fake selectolax node ──────────────────────
-
-
-class _FakeAttr(dict):
-    """Minimal attributes dict for a fake HTML node."""
-
-
-class _FakeNode:
-    """Minimal selectolax node stub for _card_to_lot tests."""
-
-    def __init__(
-        self,
-        *,
-        href: str = "/view/12345/",
-        title_text: str = "1/11 2-к, Репина, 75/2 стр.",
-        area_text: str = "52 м2",
-        price_text: str = "8 500 000",
-        img_src: str = "https://img.n1.ru/photo.jpg",
-    ) -> None:
-        self._href = href
-        self._title_text = title_text
-        self._area_text = area_text
-        self._price_text = price_text
-        self._img_src = img_src
-
-    def css_first(self, selector: str) -> _FakeNode | None:
-        if 'a[href*="/view/"]' in selector:
-            return self
-        if "total-area" in selector:
-            return _FakeTextNode(self._area_text)
-        if "param-price" in selector:
-            return _FakeTextNode(self._price_text)
-        return None
-
-    def css(self, selector: str) -> list[_FakeNode]:
-        if selector == "img":
-            return [_FakeImgNode(self._img_src)]
-        return []
-
-    @property
-    def attributes(self) -> dict:
-        return {"href": self._href}
-
-    def text(self, strip: bool = False) -> str:
-        t = self._title_text
-        return t.strip() if strip else t
-
-
-class _FakeTextNode:
-    def __init__(self, text: str) -> None:
-        self._text = text
-
-    def text(self, strip: bool = False) -> str:
-        return self._text.strip() if strip else self._text
-
-
-class _FakeImgNode:
-    def __init__(self, src: str) -> None:
-        self.attributes = {"src": src}
-
-
-def _call_card_to_lot(card: _FakeNode):  # type: ignore[return]
-    from app.services.scrapers.n1 import N1Scraper
-
-    scraper = object.__new__(N1Scraper)
-    scraper.base_url = "https://ekaterinburg.n1.ru"  # type: ignore[attr-defined]
-    return scraper._card_to_lot(card)  # type: ignore[attr-defined]
-
-
-def test_card_to_lot_parses_floor() -> None:
-    card = _FakeNode(title_text="1/11 2-к, Репина, 75/2 стр.")
-    lot = _call_card_to_lot(card)
-    assert lot is not None
-    assert lot.floor == 1
-    assert lot.total_floors == 11
-
-
-def test_card_to_lot_floor_none_when_no_token() -> None:
-    card = _FakeNode(title_text="2-к, Репина, 75/2 стр.")
-    lot = _call_card_to_lot(card)
-    assert lot is not None
-    assert lot.floor is None
-    assert lot.total_floors is None
-
-
-def test_card_to_lot_listing_date_is_none() -> None:
-    """N1 SERP has no date — listing_date must remain None."""
-    card = _FakeNode(title_text="3/9 1-к, Малышева, 12")
-    lot = _call_card_to_lot(card)
-    assert lot is not None
-    assert lot.listing_date is None

tradein-mvp/backend/tests/test_scraper_settings.py

@@ -149,7 +149,7 @@ def test_invalidate_cache_specific_source():
     mock_open, _ = _mock_session_uniform(6.0)
     with patch(_PATCH_TARGET, mock_open):
         ss.get_scraper_delay("yandex")  # +2 (yandex + global)
-        ss.invalidate_cache("yandex")   # clears yandex key
+        ss.invalidate_cache("yandex")  # clears yandex key
         ss.get_scraper_delay("yandex")  # +1 (yandex stale; global still cached)
     assert mock_open.call_count == 3
 
@@ -159,7 +159,7 @@ def test_invalidate_cache_all():
     mock_open, _ = _mock_session_uniform(6.0)
     with patch(_PATCH_TARGET, mock_open):
         ss.get_scraper_delay("yandex")  # +2 (yandex + global)
-        ss.invalidate_cache()           # clears everything
+        ss.invalidate_cache()  # clears everything
         ss.get_scraper_delay("yandex")  # +2 (both stale)
     assert mock_open.call_count == 4
 
@@ -169,8 +169,8 @@ def test_invalidate_alias_invalidates_umbrella():
     mock_open, _ = _mock_session_uniform(6.0)
     with patch(_PATCH_TARGET, mock_open):
         ss.get_scraper_delay("yandex_detail")  # +2 (yandex + global)
-        ss.invalidate_cache("yandex_detail")    # clears yandex (via alias)
+        ss.invalidate_cache("yandex_detail")  # clears yandex (via alias)
-        ss.get_scraper_delay("yandex_detail")   # +1 (yandex stale; global fresh)
+        ss.get_scraper_delay("yandex_detail")  # +1 (yandex stale; global fresh)
     assert mock_open.call_count == 3
 
 
@@ -214,11 +214,11 @@ def test_missing_per_source_row_uses_class_default(monkeypatch):
     """Missing source row → _DEFAULT_DELAY_BY_SOURCE. global=0 → no floor."""
     invalidate_cache()
 
-    mock_db = _make_mock_session({"n1": None, "global": 0.0})
+    mock_db = _make_mock_session({"domklik": None, "global": 0.0})
     monkeypatch.setattr(ss_mod, "_open_session", lambda: mock_db)
 
-    delay = get_scraper_delay("n1")
+    delay = get_scraper_delay("domklik")
-    assert delay == ss_mod._DEFAULT_DELAY_BY_SOURCE["n1"]
+    assert delay == ss_mod._DEFAULT_DELAY_BY_SOURCE["domklik"]
 
 
 def test_global_missing_row_treated_as_zero(monkeypatch):