Многоагентный аудит + имплементация: один воркер на файл, точечные правки. Верификация: py_compile (47/47 .py) + tsc --noEmit (0 ошибок). Unit-тесты не прогонялись (окружение не поднято: rollup native dep / нет pytest-venv). Полностью исправлено (169): #1336, #1337, #1339, #1340, #1341, #1342, #1343, #1345, #1346, #1348, #1349, #1350, #1351, #1354, #1356, #1358, #1359, #1360, #1362, #1364, #1365, #1366, #1367, #1368, #1369, #1370, #1371, #1372, #1373, #1374, #1375, #1376, #1377, #1378, #1379, #1380, #1381, #1382, #1384, #1385, #1386, #1387, #1388, #1389, #1390, #1391, #1392, #1394, #1395, #1396, #1397, #1399, #1400, #1401, #1402, #1403, #1404, #1408, #1409, #1410, #1411, #1412, #1413, #1414, #1415, #1416, #1417, #1418, #1420, #1423, #1425, #1426, #1427, #1428, #1429, #1430, #1431, #1432, #1433, #1434, #1435, #1437, #1438, #1439, #1440, #1441, #1442, #1443, #1444, #1445, #1446, #1447, #1448, #1449, #1450, #1451, #1452, #1453, #1454, #1455, #1456, #1457, #1458, #1459, #1460, #1461, #1462, #1463, #1464, #1465, #1466, #1467, #1468, #1469, #1471, #1472, #1473, #1474, #1476, #1478, #1479, #1481, #1482, #1483, #1484, #1485, #1487, #1488, #1489, #1490, #1491, #1492, #1493, #1494, #1495, #1496, #1497, #1499, #1500, #1501, #1502, #1504, #1505, #1506, #1507, #1510, #1514, #1515, #1516, #1517, #1518, #1519, #1521, #1522, #1523, #1524, #1525, #1526, #1527, #1528, #1529, #1531, #1532, #1533, #1534, #1535, #1536, #1537, #1538 Частично (9, in-file часть, остаток cross-file): #1361, #1419, #1422, #1424, #1470, #1475, #1477, #1480, #1498 Требуют cross-file (3, не тронуты): #1338, #1363, #1421 Пропущено (1): #1539 Не входило в партию: 22 needs-Leha issue (нужны решения владельца). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
389 lines
15 KiB
Python
389 lines
15 KiB
Python
"""DaData clients — /clean/address (enrichment) + /suggest/address (autocomplete).
|
||
|
||
/clean/address — обогащение **целевого адреса** канонической формой,
|
||
kadastr_num, ФИАС, координаты, метро. Используется в estimator flow для
|
||
on-demand обогащения payload.address (PR Q1). Требует token + secret,
|
||
demo tier 100/день.
|
||
|
||
/suggest/address — автокомплит для /api/v1/geocode/suggest (PR Q2).
|
||
Заменяет заблокированный Yandex Suggest как второй tier после Cadastral FDW.
|
||
Token-only (X-Secret не нужен), 10k/день free. Возвращает список candidate'ов
|
||
с координатами + ФИАС + структурными полями.
|
||
|
||
ENV: DADATA_API_TOKEN, DADATA_API_SECRET.
|
||
- clean_address: требует оба.
|
||
- suggest_addresses: требует только DADATA_API_TOKEN.
|
||
Если требуемые credentials не заданы → graceful возврат None / [] (не break flow).
|
||
|
||
Docs:
|
||
- https://dadata.ru/api/clean/address/
|
||
- https://dadata.ru/api/suggest/address/
|
||
"""
|
||
|
||
from __future__ import annotations
|
||
|
||
import logging
|
||
from dataclasses import dataclass
|
||
from typing import Any
|
||
|
||
import httpx
|
||
|
||
from app.core.config import settings
|
||
|
||
logger = logging.getLogger(__name__)
|
||
|
||
DADATA_CLEAN_URL = "https://cleaner.dadata.ru/api/v1/clean/address"
|
||
DADATA_SUGGEST_URL = "https://suggestions.dadata.ru/suggestions/api/4_1/rs/suggest/address"
|
||
_DADATA_TIMEOUT_S = 8.0
|
||
_DADATA_SUGGEST_TIMEOUT_S = 5.0
|
||
|
||
|
||
@dataclass(frozen=True, slots=True)
|
||
class DadataAddressResult:
|
||
"""Parsed DaData enrichment payload — slots для compact representation.
|
||
|
||
canonical_address — поле `result` из ответа («г Екатеринбург, ул Малышева, д 125»).
|
||
house_cadnum — кадастровый номер ДОМА (не участка). Формат «66:41:0704045:350».
|
||
house_fias_id — UUID ФИАС дома.
|
||
lat / lon — координаты от DaData (precision до дома).
|
||
qc_geo — quality code геокодинга: 0=exact, 1=street, 2=settlement…
|
||
qc_house — quality code дома: 2=в ФИАС, 10=на картах…
|
||
kladr_id — старый ОКРС-классификатор (legacy, может быть полезен).
|
||
okato / oktmo — admin codes для отчётности.
|
||
metro — список ближайших станций метро (если есть): [{name,line,distance}…].
|
||
raw — полный ответ для debugging / future fields.
|
||
"""
|
||
|
||
canonical_address: str | None
|
||
house_cadnum: str | None
|
||
house_fias_id: str | None
|
||
lat: float | None
|
||
lon: float | None
|
||
qc_geo: int | None
|
||
qc_house: int | None
|
||
kladr_id: str | None
|
||
okato: str | None
|
||
oktmo: str | None
|
||
metro: list[dict[str, Any]]
|
||
raw: dict[str, Any]
|
||
|
||
|
||
def _credentials() -> tuple[str, str] | None:
|
||
"""Returns (token, secret) если оба заданы, иначе None."""
|
||
token = (settings.dadata_api_token or "").strip()
|
||
secret = (settings.dadata_api_secret or "").strip()
|
||
if not token or not secret:
|
||
return None
|
||
return token, secret
|
||
|
||
|
||
def _coerce_int(value: Any) -> int | None:
|
||
"""DaData отдаёт qc-коды строками иногда. Coerce → int безопасно."""
|
||
if value is None:
|
||
return None
|
||
try:
|
||
return int(value)
|
||
except (TypeError, ValueError):
|
||
return None
|
||
|
||
|
||
def _coerce_float(value: Any) -> float | None:
|
||
if value is None:
|
||
return None
|
||
try:
|
||
return float(value)
|
||
except (TypeError, ValueError):
|
||
return None
|
||
|
||
|
||
def _parse_response(item: dict[str, Any]) -> DadataAddressResult:
|
||
"""Парсит single DaData payload item в DadataAddressResult.
|
||
|
||
Не raise'ит — пропускает невалидные поля как None.
|
||
"""
|
||
metro_raw = item.get("metro") or []
|
||
if not isinstance(metro_raw, list):
|
||
metro_raw = []
|
||
|
||
return DadataAddressResult(
|
||
canonical_address=item.get("result"),
|
||
house_cadnum=item.get("house_cadnum"),
|
||
house_fias_id=item.get("house_fias_id"),
|
||
lat=_coerce_float(item.get("geo_lat")),
|
||
lon=_coerce_float(item.get("geo_lon")),
|
||
qc_geo=_coerce_int(item.get("qc_geo")),
|
||
qc_house=_coerce_int(item.get("qc_house")),
|
||
kladr_id=item.get("kladr_id"),
|
||
okato=item.get("okato"),
|
||
oktmo=item.get("oktmo"),
|
||
metro=metro_raw,
|
||
raw=item,
|
||
)
|
||
|
||
|
||
async def clean_address(address: str) -> DadataAddressResult | None:
|
||
"""Обогащает один адрес через DaData /clean/address.
|
||
|
||
Args:
|
||
address: свободный текст адреса (как пришёл от пользователя или из geocoder).
|
||
|
||
Returns:
|
||
DadataAddressResult или None если:
|
||
- credentials не заданы в ENV (graceful disable)
|
||
- адрес пустой / слишком короткий
|
||
- API вернул пустой массив / ошибку shape
|
||
- сетевая / HTTP ошибка (5xx, 429 quota, timeout)
|
||
- адрес не распознан (canonical_address пустой)
|
||
"""
|
||
if not address or len(address.strip()) < 3:
|
||
return None
|
||
|
||
creds = _credentials()
|
||
if creds is None:
|
||
logger.debug("dadata: credentials не заданы — skip enrichment")
|
||
return None
|
||
|
||
token, secret = creds
|
||
headers = {
|
||
"Authorization": f"Token {token}",
|
||
"X-Secret": secret,
|
||
"Content-Type": "application/json",
|
||
"Accept": "application/json",
|
||
}
|
||
body = [address.strip()]
|
||
|
||
try:
|
||
async with httpx.AsyncClient(timeout=_DADATA_TIMEOUT_S) as client:
|
||
response = await client.post(DADATA_CLEAN_URL, headers=headers, json=body)
|
||
except (httpx.TimeoutException, httpx.NetworkError) as exc:
|
||
logger.warning("dadata: network error для %r: %s", address[:60], exc)
|
||
return None
|
||
except Exception as exc: # pragma: no cover — defensive
|
||
logger.warning("dadata: unexpected client error для %r: %s", address[:60], exc)
|
||
return None
|
||
|
||
status = response.status_code
|
||
if status == 429:
|
||
logger.warning("dadata: HTTP 429 — quota exceeded (100/день demo limit?)")
|
||
return None
|
||
if status in (401, 403):
|
||
logger.error(
|
||
"dadata: HTTP %d — auth/secret rejected. Проверь DADATA_API_TOKEN/SECRET.",
|
||
status,
|
||
)
|
||
return None
|
||
if status >= 500:
|
||
logger.warning("dadata: HTTP %d — transient server error", status)
|
||
return None
|
||
if status >= 400:
|
||
body_preview = (response.text or "")[:200]
|
||
logger.warning("dadata: HTTP %d — bad request: %r", status, body_preview)
|
||
return None
|
||
|
||
try:
|
||
payload = response.json()
|
||
except ValueError as exc:
|
||
logger.warning("dadata: невалидный JSON в ответе: %s", exc)
|
||
return None
|
||
|
||
if not isinstance(payload, list) or not payload:
|
||
logger.info("dadata: пустой массив для %r", address[:60])
|
||
return None
|
||
|
||
item = payload[0]
|
||
if not isinstance(item, dict):
|
||
logger.warning("dadata: первый элемент не dict (%s)", type(item).__name__)
|
||
return None
|
||
|
||
result = _parse_response(item)
|
||
|
||
if not result.canonical_address:
|
||
# DaData вернул shape, но не распознал адрес — quality codes обычно qc_geo=5.
|
||
logger.info(
|
||
"dadata: адрес не распознан (qc_geo=%s qc_house=%s) для %r",
|
||
result.qc_geo,
|
||
result.qc_house,
|
||
address[:60],
|
||
)
|
||
return None
|
||
|
||
logger.info(
|
||
"dadata: enriched %r → canonical=%r cadnum=%s qc_geo=%s qc_house=%s",
|
||
address[:60],
|
||
(result.canonical_address or "")[:60],
|
||
result.house_cadnum,
|
||
result.qc_geo,
|
||
result.qc_house,
|
||
)
|
||
return result
|
||
|
||
|
||
# ── Suggest (autocomplete) ───────────────────────────────────────────────────
|
||
|
||
|
||
@dataclass(frozen=True, slots=True)
|
||
class DadataSuggestion:
|
||
"""Один candidate из ответа DaData /suggest/address.
|
||
|
||
value — короткая строка для отображения («ул Малышева, д 30»).
|
||
unrestricted_value — полная строка с регионом/страной («г Екатеринбург, ул …»).
|
||
lat / lon — координаты (если DaData их вернул; для some сегментов None).
|
||
fias_id — UUID ФИАС матчнутого объекта (дома/улицы/города).
|
||
house — номер дома если distinguished (строкой: «30» / «30к2»).
|
||
street — название улицы.
|
||
city — locality.
|
||
kind — `house` (fias_level=8/9) / `street` (=7) / `city` (<=6) /
|
||
`plot` (=65 planning, =75 plot).
|
||
Маппится на GeocodeSuggestion.kind в geocoder.py.
|
||
"""
|
||
|
||
value: str
|
||
unrestricted_value: str
|
||
lat: float | None
|
||
lon: float | None
|
||
fias_id: str | None
|
||
house: str | None
|
||
street: str | None
|
||
city: str | None
|
||
kind: str
|
||
|
||
|
||
def _classify_kind(fias_level: Any) -> str:
|
||
"""Маппит DaData `fias_level` → kind для GeocodeSuggestion.
|
||
|
||
DaData fias_level (https://dadata.ru/api/suggest/address/#fias_level):
|
||
0=country, 1=region, 3=area, 4=city, 5=settlement, 6=street_alt,
|
||
7=street, 8=house, 9=flat, 65=planning, 75=plot.
|
||
Для autocomplete нам важно: house / street / city.
|
||
"""
|
||
level = _coerce_int(fias_level)
|
||
if level is None:
|
||
return "city" # safe default
|
||
if level in (65, 75):
|
||
# 65=planning structure / 75=земельный участок — не дом.
|
||
# geocoder._DADATA_KIND_MAP не знает 'plot' → fallback на 'locality'.
|
||
return "plot"
|
||
if level in (8, 9):
|
||
# 8=house, 9=flat — оба привязаны к конкретному дому.
|
||
return "house"
|
||
if level == 7:
|
||
return "street"
|
||
return "city"
|
||
|
||
|
||
def _parse_suggestion(item: dict[str, Any]) -> DadataSuggestion | None:
|
||
"""Парсит один элемент `suggestions[]` → DadataSuggestion.
|
||
|
||
Возвращает None если shape невалидный (нет value / нет data).
|
||
"""
|
||
if not isinstance(item, dict):
|
||
return None
|
||
value = item.get("value")
|
||
if not value or not isinstance(value, str):
|
||
return None
|
||
data = item.get("data") or {}
|
||
if not isinstance(data, dict):
|
||
data = {}
|
||
|
||
return DadataSuggestion(
|
||
value=value,
|
||
unrestricted_value=str(item.get("unrestricted_value") or value),
|
||
lat=_coerce_float(data.get("geo_lat")),
|
||
lon=_coerce_float(data.get("geo_lon")),
|
||
fias_id=data.get("fias_id"),
|
||
house=data.get("house"),
|
||
street=data.get("street"),
|
||
city=data.get("city"),
|
||
kind=_classify_kind(data.get("fias_level")),
|
||
)
|
||
|
||
|
||
async def suggest_addresses(
|
||
query: str, limit: int = 8, city: str = "Екатеринбург"
|
||
) -> list[DadataSuggestion]:
|
||
"""Автокомплит адресов через DaData /suggest/address.
|
||
|
||
Args:
|
||
query: текст для подсказки (минимум 2 символа).
|
||
limit: сколько вариантов вернуть (DaData параметр `count`, max 20).
|
||
city: locality для constraint (передаётся в `locations`). По умолчанию ЕКБ.
|
||
|
||
Returns:
|
||
list[DadataSuggestion] — пустой список если:
|
||
- DADATA_API_TOKEN не задан (graceful disable)
|
||
- query короче 2 символов
|
||
- сетевая / HTTP ошибка / 429 quota / malformed response
|
||
|
||
Token-only auth (X-Secret для suggest не нужен — отличается от clean_address).
|
||
Quota: 10000/день на free tier (в 100x больше чем у clean_address).
|
||
"""
|
||
if not query or len(query.strip()) < 2:
|
||
return []
|
||
|
||
token = (settings.dadata_api_token or "").strip()
|
||
if not token:
|
||
logger.debug("dadata suggest: DADATA_API_TOKEN не задан — skip")
|
||
return []
|
||
|
||
headers = {
|
||
"Authorization": f"Token {token}",
|
||
"Content-Type": "application/json",
|
||
"Accept": "application/json",
|
||
}
|
||
body: dict[str, Any] = {
|
||
"query": query.strip(),
|
||
"count": max(1, min(int(limit), 20)),
|
||
}
|
||
if city:
|
||
body["locations"] = [{"city": city}]
|
||
|
||
try:
|
||
async with httpx.AsyncClient(timeout=_DADATA_SUGGEST_TIMEOUT_S) as client:
|
||
response = await client.post(DADATA_SUGGEST_URL, headers=headers, json=body)
|
||
except (httpx.TimeoutException, httpx.NetworkError) as exc:
|
||
logger.warning("dadata suggest: network error для %r: %s", query[:60], exc)
|
||
return []
|
||
except Exception as exc: # pragma: no cover — defensive
|
||
logger.warning("dadata suggest: unexpected client error для %r: %s", query[:60], exc)
|
||
return []
|
||
|
||
status = response.status_code
|
||
if status == 429:
|
||
logger.warning("dadata suggest: HTTP 429 — quota exceeded (10k/день free?)")
|
||
return []
|
||
if status in (401, 403):
|
||
logger.error(
|
||
"dadata suggest: HTTP %d — auth rejected. Проверь DADATA_API_TOKEN.", status
|
||
)
|
||
return []
|
||
if status >= 500:
|
||
logger.warning("dadata suggest: HTTP %d — transient server error", status)
|
||
return []
|
||
if status >= 400:
|
||
body_preview = (response.text or "")[:200]
|
||
logger.warning("dadata suggest: HTTP %d — bad request: %r", status, body_preview)
|
||
return []
|
||
|
||
try:
|
||
payload = response.json()
|
||
except ValueError as exc:
|
||
logger.warning("dadata suggest: невалидный JSON в ответе: %s", exc)
|
||
return []
|
||
|
||
if not isinstance(payload, dict):
|
||
logger.warning("dadata suggest: response не dict (%s)", type(payload).__name__)
|
||
return []
|
||
|
||
raw_suggestions = payload.get("suggestions")
|
||
if not isinstance(raw_suggestions, list):
|
||
logger.info("dadata suggest: нет ключа `suggestions` в ответе для %r", query[:60])
|
||
return []
|
||
|
||
out: list[DadataSuggestion] = []
|
||
for item in raw_suggestions:
|
||
parsed = _parse_suggestion(item)
|
||
if parsed is not None:
|
||
out.append(parsed)
|
||
|
||
logger.info("dadata suggest: %r → %d вариантов", query[:60], len(out))
|
||
return out
|